Synteza sieci metalo-organicznych Zn-MOF i Cd-MOF o mieszanych łącznikach z grupy acylohydrazonów i dikarboksylanów Kornel Roztocki Rozprawa doktorska wykonana na Wydziale Chemii Uniwersytetu Jagiellońskiego pod kierunkiem dra hab. Dariusza Matogi Kraków, 2019 Podziękowania Po pierwsze pragnę podziękować promotorowi Dariuszowi Matodze, który powierzył mi fascynujący temat oraz wspierał na każdym etapie pracy naukowej. Dziękuję Rodzicom, za to że wiele lat znosili moje „ekstrawaganckie” zachowania podczas powstawania doktoratu jak i w trakcie żmudnego procesu wychowywania. Również dziękuję pozostałym członkom naszej małej MOFowej komórki społecznej : Monice Szufli, Magdalenie Lupie i Damianowi Jędrzejowskiemu za wieloletnią współpracę i pomoc. Dziękuję również Wszystkim tym, których mnogość i obfitość jest tak duża, że gdybym ich i ich zasługi wobec mnie zaczął wymieniać to liczba zużytych arkuszy przewyższyłaby liczbę poświęconych rozprawie. Dlatego pozwolę sobie wymienić tylko Wita Rogala-Rogalskiego i Piotra Goszczyckiego Panowie wiedzą za co... „Kiedy Rüzum śpi budzą się koszmary I 3 Spis treści Podziękowania...................................................................................2 Spis treści.....................................................................................3 Streszczenie....................................................................................4 Lista publikacji wchodzących w skład rozprawy doktorskiej.......................................5 Dodatkowy dorobek naukowy.......................................................................6 Używane skróty..................................................................................7 1. Wstęp teoretyczny...........................................................................8 1.1 Sieci MOF..............................................................................8 1.2 Acylohydrazony........................................................................12 1.3 Stabilność sieci MOF..................................................................13 2. Cel pracy .................................................................................16 3. Metodyka badań.............................................................................17 4. Przewodnik po publikacjach ................................................................18 Ad. I Synteza, określenie struktury i właściwości pierwszych acylohydrazonowo-karboksylanowych sieci MOF opartych na jonach cynku, izonikotynoilo-hydrazonie aldehydu 4-pirydynowego (pcih) i kwasie tereftalowym (H2bdc) lub kwasie 4,4'-bisfenylodikarboksylowym (H2bpdc).............19 Ad. II Optymalizacja syntezy mechanochemicznej izoftalowo-hydrazonowgo Zn-MOFu wykazującego dynamiczną i selektywną adsorpcję CO2........................................20 Ad. III Badanie wpływu podstawnika na ułożenie warstw, stabilność oraz właściwości sorpcyjne serii Zn-MOFów zawierających podstawione izoftalany (Xiso2-)..............................21 Ad. IV Teoretyczny oraz eksperymentalny wgląd w proces adsorpcji oraz hydrotermiczną stabilność wodoodpornego Cd-MOFu opartego na 4,4'-sulfonobenzenodikarboksylowym kwasie (H2sdb)...................................................................................22 Ad. V Kontrola budowy węzłów sieci izoftalanowych Cd-MOF-ów za pomocą podstawnika oraz warunków syntezy..........................................................................23 Ad. VI Wykorzystanie krótszego oraz dłuższego łącznika acylohydrazonowego do otrzymania kadmowych sieci metalo-organicznych.......................................................23 5. Podsumowanie ..............................................................................25 6. Bibliografia...............................................................................26 7. Załączniki.................................................................................28 I 4 Streszczenie Głównym celem badań prowadzonych w ramach pracy doktorskiej była synteza, określenie struktury i charakterystyka właściwości fizykochemicznych porowatych acylohydrazonowo- karboksylanowych polimerów koordynacyjnych. Podczas preparatyki zastosowano klasyczną metodę syntezy w roztworze, oraz nowatorskie podejście z użyciem wpisującej się kanon zielonej chemii mechanosyntezy, gdzie produkt powstawał w wyniku ucierania stałych reagentów z dodatkiem niewielkiej ilości cieczy. Otrzymane połączenia charakteryzowano za pomocą analizy elementarnej, spektroskopii absorpcyjnej w podczerwieni FTIR, spektroskopii UV-Vis, analizy termicznej sprzężonej z detekcją mas TGA-MS, dyfrakcji promieniowania rentgenowskiego na próbkach polikrystalicznych (PXRD w stałej temperaturze oraz w zakresie 25-500 oC (TD-PXRD), pomiarów izoterm adsorpcji gazów (CO2, N2, H2 i H2O), przy czym dodatkowo monitorowano in situ proces adsorpcji CO2 za pomocą spektroskopii IR. Właściwości otrzymanych sieci metalo-organicznych skorelowano ze strukturą krystaliczną rozwiązaną w oparciu o pomiar dyfrakcji promieniowania rentgenowskiego na monokryształach (SC-XRD). Do głębszego zrozumienia obserwowanych zjawisk wykorzystano programy Zeo++ i PoreBlazer oraz symulacje komputerowe (Grand Canonical Monte-Carlo). Praca badawcza skupiała się na syntezie i scharakteryzowaniu nowych materiałów porowatych za pomocą technik eksperymentalnych i teoretycznych. Pozwala to na głębokie zrozumienie subtelnych zależności pomiędzy strukturą wewnętrzną, a właściwościami materiałów, takich jak hydrolityczna i termiczna stabilność, właściwości sorpcyjne czy dynamika strukturalna. Cele badawcze prowadzonych badań częściowo były realizowane w ramach stażu zagranicznego (w ramach programu NCN Etiuda 6 „Synteza sieci metalo-organicznych Zn-MOF i Cd-MOF o mieszanych łącznikach z grupy acylohydrazonów i dikarboksylanów”) oraz projektu NCN OPUS9 "Poszukiwanie przewodników protonowych MOF z podejściem mechanochemicznym” i zakończonego projektu NCN OPUS4 "Sieci metalo-organiczne MOF oparte na polifunkcyjnych ligandach hydrazonowych: synteza, struktura i właściwości fizykochemiczne”. I 5 Lista publikacji wchodzących w skład rozprawy doktorskiej I. K. Roztocki, I. Senkovska, S. Kaskel, D. Matoga "Carboxylate-Hydrazone Mixed-Linker Metal-Organic Frameworks: Synthesis, Structure and Selective Gas Adsorption", Eur. J. Inorg. Chem., 2016, 4450-4456. — invited article from a cluster issue on "Metal-Organic Frameworks Heading Towards Application". Doktorant po raz pierwszy otrzymał hydrazonowo-karboksylanowe sieci metalo-organiczne o mieszanych łącznikach. Zbadał ich właściwości oraz przygotował wersję roboczą znaczącej części artykułu (Results and Discussion, Experimental, Supplementary Information). Jest to praca przełomowa, ponieważ dała podwalinę bogatej chemii rodziny hydrazonowo-karboksylanowych polimerów koordynacyjnym. II. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga, "Isophtalate-hydrazone 2D zinc-organic framework: crystal structure, selective adsorption and tuning of mechanochemical synthetic conditions", Inorg. Chem., 2016, 55, 9663-9670. Doktorant otrzymał warstwowy układ dynamiczny za pomocą metody klasycznej z roztworu, oraz wraz z pomocą studenta I roku dokonał optymalizacji syntezy na drodze mechanochemicznej. Przeprowadził podstawową charakterystykę i przygotował wersję roboczą znaczącej części artykułu (Results and Discussion, Materials and Methods, Supplementary Information). Praca ta stanowiła podstawę rozwoju mechanochemii układów hydrazonowo-karboksylanowych. III. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga "Effect of Linker Substituent on Layers Arrangement, Stability, and Sorption of Zn-Isophthalate/Acylhydrazone Frameworks" Cryst. Growth Des., 2018, 18, 488-497. Doktorant wraz licencjantem opiekuna naukowego otrzymał serię warstwowych układów opartych na 5-podstawionych isoftalanach, sprawdził ich stabilność względem wody oraz wytrzymałość mechaniczną. Dokonał optymalizacji syntezy na drodze mechanochemicznej i przygotował wersję roboczą znaczącej części artykułu (Introduction, Results and Discussion, Supplementary Information). IV. K. Roztocki, M. Lupa, A. Sławek, W. Makowski, I. Senkovska, S. Kaskel, D. Matoga, "Water- stable Metal-organic Framework with Three Hydrogen-bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. Chem., 2018, 57, 3287-3296. Doktorant na przykładzie syntezowanej przez niego sieci metalo-organicznej z pomocą licencjantki opiekuna naukowego dokonał ewaluacji stabilności względem wielu destruktorów (wysoka temperatura, woda, stabilność hydrotermiczna, zachowanie porowatości przy wielokrotnych cyklach adsorpcji/desorpcji). Zaobserwowane zjawiska zostały również wytłumaczone za pomocą obliczeń Monte-Carlo (GCMC). Doktorant przygotował wersję roboczą znaczącej części artykułu (Introduction, Results and Discussion, Supplementary Information). V. K. Roztocki, M. Lupa, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga "Bulky substituent and solvent-induced alternative nodes for layered Cd-isophthalate/acylhydrazone frameworks" CrystEngComm, 2018, 20, 2841-2849. W zależności od warunków syntezy oraz podstawnika doktorant z magistrantką opiekuna naukowego uzyskał różne MOF-y z tych samych materiałów startowych. W oparciu o strukturę krystaliczną, pomiary PXRD w funkcji temperatury, analizę termograwimetryczną oraz parametry porowatości obliczone za pomocą programów komputerowych wytłumaczył wpływ ułożenia oraz podstawników na właściwości sorpcyjne otrzymanych polimerów jak również ich stabilność termiczną. Doktorant przygotował wersję roboczą znaczącej części artykułu (Results and Discussion, Supplementary Information). I 6 VI. K. Roztocki, M. Szufla, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga "Introducing a longer versus shorter acylohydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties". W zależności od użytego liganda acylohydrazonowego doktorant z licencjatką opiekuna naukowego uzyskał w syntezie z roztworu oraz mechanochemicznie dwa nieizostrukturalne MOF-y. W oparciu o strukturę krystaliczną, pomiary PXRD w funkcji temperatury, analizę termograwimetryczną oraz parametry porowatości obliczenie za pomocą programów komputerowych doktorant zbadał właściwości otrzymanych polimerów ( w tym sorpcyjne) i powiązał je z budową wewnętrzną. Doktorant również zbadał ich stabilność termiczną i hydrotermiczną jak również przygotował wersję roboczą znaczącej części artykułu (Introduction, Results and Discussion, Supplementary Information). Na użycie artykułów stanowiących podstawę tej pracy uzyskano zgodę wydawców, tj. Royal Society of Chemistry (RSC), American Chemical Society (ACS) i Wiley-VCH. Również na użycie rysunków zawartych we wstępie teoretycznym jak i przewodniku po publikacjach uzyskano zgodę wydawców. Dodatkowy dorobek naukowy Ponadto, wyniki badań uzyskane w doktoracie prezentowano na poniższych konferencjach naukowych oraz stały się podstawą międzynarodowego zgłoszenia patentowego (postępowanie w toku): Zgłoszenia patentowe w USPTO i EPO: „Metal-organic frameworks (MOFs), method for their preparation and their application", PCT/IB2015/054558 z dnia 16.06.2015. Zgłoszenie w USPTO: US 15/318,678 z dnia 14.12.2016. Zgłoszenie w EPO: EP 15750111.5 z dnia 15.01.2017. Twórcy: D. Matoga, K. Roztocki. Wspomniane wyżej zgłoszenie w trybie PCT poprzedzone zostało zgłoszeniem w UP RP: „Sieci metalo- organiczne typu MOF o wzorze [M2(dcx)2L2], zawierające dikarboksylany (dcx) oraz hydrazony (L), sposób ich otrzymywania i zastosowanie", P408565 z dnia 16.06.2014.Twórcy: D. Matoga, K. Roztocki Konferencje międzynarodowe: • “docMOF-2018: A PhD-run Symposium” • "European Conference on Metal Organic Frameworks and Porous Polymers" (EUROMOF2017) • "XIX th International Winter School on Coordination Chemistry" • "2nd International Conference on Functional Molecular Materials (FUNMAT2015)” • "European Conference on Metal Organic Frameworks and Porous Polymers” (EUROMOF 2015) • "XX th International Winter School on Coordination Chemistry" • "International Symposium on Nanostructured Functional Materials" Konferencje krajowe: • "Kopernikańskie Seminarium Doktoranckie" • "Zjazd Zimowy Sekcji Studenckiej PTChem" • "Horyzonty Nauki: Forum Prac Dyplomowych" Publikacje • K. Roztocki, D. Matoga, J. Szklarzewicz, Inorg. Chem. Commun. 2015, 57, 22-25. • K. Roztocki, D. Matoga, W. Nitek, Inorg. Chim. Acta, 2016, 448, 86-92. • D. Matoga, K. Roztocki, M. Wilke, F. Emmerling, M. Oszajca, M. Fitta and M. Balanda, CrystEngComm, 2017, 19, 2987-2995. I 7 Używane skróty bpy 4,4’-bipirydyl COF (ang. Covalent-Organic Framework) kowalencyjna sieć organiczna DMA N,N ’ -dimetyloacetamid DMF N,N ’ -dimetyloformamid EtOH Etanol GCMC Grand Canonical Monte Carlo H2bdc kwas tereftalowy H2bpdc kwas 4,4'-bisfenylodikarboksylowy H2oba kwas 4,4' - oksy-bis(benzenokarboksylowy) H2sba kwas 4,4'-sulfonobenzenodikarboksylowy H2Xiso (X = H lub CH3, lub tBu, lub OH, lub NH2 ) 5-podstawiony kwas izoftalowy IUPAC (ang. International Union of Pure and Applied Chemistry) Międzynarodowa Unia Chemii Czystej i Stosowanej MOF (ang. Metal-Organic Framework) sieć metalo-organiczna PXRD (ang. powder X-ray diffraction) proszkowa dyfrakcja rentgenowska pcih izonikotynoilo-hydrazon aldehydu 4-pirydynowego tBu podstawnik tertbutylowy (C4H9) tdih diizonikotynoilo - hydrazon al.dehydu tereftalowego SC-XRD (ang. single-crystal X-ray diffraction) dyfrakcja rentgenowska na monokrysztale TGA (ang. thermogravimetric analysis) analiza termograwimetryczna I 8 1. Wstęp teoretyczny 1.1 Sieci MOF Sieci metalo-organiczne (z angielskiego Metal-Organic Frameworks - MOFs) są porowatymi polimerami koordynacyjnymi, najczęściej krystalicznymi, zbudowanymi z kationu metalu lub klastera metalicznego pełniącego rolę węzła sieci oraz organicznych ligandów (tzw. łączników) łączących ze sobą węzły. W wyniku składania się komponentów powstają wolne przestrzenie (pory) zajmowane przez cząsteczki gości, które można usunąć za pomocą zmiany warunków fizycznych, np. temperatury, ciśnienia czy promieniowaniem elektromagnetycznym1. Na Rysunku 1 przedstawiono przykład archetypowych MOF-ów - MOF-52 oraz HKUST-13. Rysunek 1 Archetypowe MOFy: MOF-5 zbudowany z cynku oraz anionów kwasu tereftalowego, w którym każdy z węzłów [ZmO] jest mostkowany przez sześć tereftalanów (lewa część)2; HKUST-1 oparty na klastrach miedzi (II) typu „paddle-wheel” mostkowanych za pomocą anionu kwasu 1,2,3-benzenotrikarboksylowego (prawa część)3. Rysunek został użyty za zgodą redakcji The Scientific Journal of IUP AC1. Sieci MOF zostały pogrupowane ze względu na właściwości lub budowę. Jednym z archetypowych, ale dalej używanych przez badaczy, podziałów sieci metalo-organicznych, jest oparty na zachowaniu się materiału pod wpływem ewakuacji cząsteczek gości4 (Rysunek 2). I 9 Rysunek 2 Podział sieci MOF ze względu na zachowanie struktury krystalicznej pod wpływem aktywacji materiału. Rysunek został użyty za zgodą redakcji Springer Nature4. Sieci metalo-organiczne, które podczas usunięcia cząsteczek gości z porów tracą uporządkowanie dalekozasięgowe a tym samym krystaliczność oraz porowatość są zaliczane do I-generacji materiałów MOF. Gdy ewakuacja molekuł z porów nie powoduje zmian struktury krystalicznej materiału (długości wiązań, kątów pomiędzy nimi, czy tworzenia i zrywania wiązań chemicznych) takie sieci nazywane są MOFami II generacji. Do trzeciej generacji zalicza się tzw. „miękkie porowate kryształy” czyli sieci metalo-organiczne, które pod wpływem bodźca zewnętrznego zmieniają swoją strukturę krystaliczną. Intensywnie rozwijaną podgrupą sieci metalo-organicznych są materiały trzeciej generacji, czyli wykazujące dynamiczne zachowanie pod wpływem bodźca, np. wymiany lub usunięcia cząsteczek gości5 (Rysunek 3). Dynamiczne MOFy łączą w sobie krystaliczną strukturę wynikającą z natury wiązań koordynacyjnych z elastycznym zachowaniem prowadzącym do strukturalnych transformacji. Materiały tego typu zmieniają swoją strukturę w odpowiedzi na fizyczny lub chemiczny bodziec6-8. Odpowiedź materiału może być "dostrajana" poprzez odpowiednie projektowanie, cechy tego typu są unikatowe i nie występują w innych materiałach porowatych. | 10 Rysunek 3 Klasyfikacja MOF-ów ze względu na rodzaj dynamicznego zachowania po zmianie lub usunięciu cząsteczek gości. Rysunek został zaadaptowany za zgodą redakcji Royal Society of Chemistry5. Dynamikę strukturalną w porowatych polimerach koordynacyjnych można zaobserwować podczas procesu adsorpcji/desorpcji cząsteczek gości, ogrzewania, naświetlania czy użycia siły mechanicznej. Bardzo ciekawym przypadkiem przejścia fazowego jest oddychanie sieci czy ich puchnięcie pod wpływem adsorbenta9,10. Przykładowo sieć [Cu(bpy)2(BF4)2] ELM-11 (ELM = elastic layer material; gdzie bpy = 4,4’-bipirydyl), początkowo nie adsorbuje CO2, jednakże po przekroczeniu pewnego ciśnienia dochodzi do „otwarcia”, rozsunięcie dwuwymiarowych warstw i adsorpcji CO2. Warto zwrócić uwagę, że praca laboratoryjna z dynamicznymi układami jest wymagająca, a ich właściwości zależą od wcześniejszego traktowania próbki. Na przykład dla ELM-11 ciśnienie CO2, które inicjuje ekspansj ę warstw, jest zależne od tego w j akim alkoholu preparat przed aktywacj ą był kondycj onowany (Rysunek 4)11. Rysunek 4 Izoterma adsorpcji CO2 w temperaturze 273 K na materiale ELM-11 (niebieski), oraz materiale ELM-11 wcześniej traktowanym metanolem, etanolem oraz propanolem (kolejno fioletowy, czerwony i zielony). Próbki były przed aktywacją namaczane w 298 K. Adsorpcja - zamknięte symbole; desorpcja - otwarte symbole. Rysunek został użyty za zgodą redakcji American Chemical Society11. Bardzo ciekawym i niedawno odkrytym przykładem dynamicznego zachowania jest ujemna adsorpcja gazu zaobserwowana dla sieci DUT-49 czyli Cu2(BBCDC) [BBCDC = 9,9'-([1,1'-bifenyl]-4,4'- diyl)bis(9#-karbazolo-3,6-dikarboksylan)], materiału posiadającego jedną z najwyższych grawimetrycznych pojemności sorpcyjnych względem metanu wśród znanych sorbentów (308 mg/g, w 298 K)12. Podczas wysokociśnieniowej adsorpcji metanu w 111 K materiał w pierwszym etapie | 11 zachowuje się jak sztywna sieć, następnie po przekroczeniu ciśnienia ~ 9 kPa w wyniku interakcji z adsorbatem dochodzi do zmiany strukturalnej, kontrakcji sieci metalo-organicznej i „wyciśnięcia” z porów cząsteczek gazu13. W wyniku przejścia fazowego podczas zwiększania ciśnienia dochodzi do nieintuicyjnego spadku ilości zaadsorbowanego gazu, co jest demonstrowane przez nagły spadek na krzywej adsorpcji (Rysunek 5). Rysunek 5 Pomiary in situ PXRD podczas adsorpcji metanu w 111 K. a) Izoterma adsorpcji (niebieski) i desorpcji (czerwony). Pełnymi punktami zaznaczono ciśnienia, przy których wykonano pomiar dyfrakcji rentgenowskiej na materiale proszkowym. b) Korespondujący wykres zmian w rejestrowanych dyfraktogramach podczas procesu adsorpcji (niebieski) i desorpcji (czerwony). c) Zmiany długości krawędzi a komórki elementarnej podczas adsorpcji (niebieski) i desorpcji metanu (czerwony). Rysunek został użyty za zgodą redakcji Springer Nature13. Szerokie zainteresowanie środowiska akademickiego i przemysłu materiałami typu MOF jest spowodowane praktycznie nieskończonymi możliwościami syntetycznymi pozwalającymi na otrzymanie materiału o ściśle zadanej wielkości i otoczeniu chemicznym wolnych przestrzeni14 , dzięki czemu można dostosowywać materiał do aplikacji15. MOFy mogą przyczynić się do rozwiązania globalnych problemów ludzkości takich jak: wyłapywanie wody na terenach pustynnych16, oczyszczanie wody17, budowa wydajnych pomp ciepła18, wyłapywanie gazów19, nowoczesna i wydajna kataliza20-22, czy rozwój elektroniki molekularnej23. W szczególności poszukiwane są materiały wykazujące wielofunkcyjność24, którą można przykładowo wprowadzić za pomocą wbudowania w sieć łączników tego samego typu posiadających różne grupy funkcyjne lub/i łączników różnego rodzaju (Rysunek 6). I 12 Rysunek 6 Przykład MOFa posiadającego różne łączniki organiczne a) i różne metale b). Obie sieci metalo-organiczne zachowują strukturę krystaliczną odpowiednio sieci MOF-5 (a) i MOF-74 (b), jednakże typ funkcjonalności jest różny. Rysunek został użyty za zgodą redakcji John Wiley and Sons24. Innym sposobem wprowadzenia multifunkcjonalności jest jednoczesne wbudowanie kilku metali do sieci MOF na drodze preparatyki lub poprzez użycie post-syntetycznej metalacji. 1.2 Acylohydrazony Acylohydrazony są otrzymywane w reakcji kondensacji związków karbonylowych z hydrazydami. Posiadają one "kieszeń" acylohydrazonową, przez co wykazują właściwości chelatujące oraz mogą ulegać przemianie tautomerycznej z formy ketonowej w enolową i deprotonacji (Schemat 1). Schemat 1 Hydrazyd z grupą aromatyczną a) acylohydrazon w formie "keto" b); acylohydrazon w formie enolanowej c); zdeprotowany acylohydrazon d). AR - grupa arylowa (zawierająca heteroatomy, grupy funkcyjne). R1, R2 -niezależne alkilowe lub arylowe grupy; lub cykloalkilowe albo/i arylowe gupy. Schemat został użyty za zgodą redakcji John Wiley and Sons25. Kompleksy chelatowe oparte na acylohydrazonach, które posiadają dodatkowe wolne miejsca koordynacyjne, mogą być potencjalnymi metalo-ligandami, przydatnymi w dalszej preparatyce do konstruowania sieci metalo-organicznych o mieszanych metalach26. Podczas pracy laboratoryjnej otrzymano i scharakteryzowano połączenia posiadające cechy metalo-liganda, ale niestety preparatyka porowatej sieci metalo-organicznej opartej o te bloki budulcowe nie powiodła się27,28. I 13 Schemat 2 Ilustracja działania foto-przełączników w odpowiedzi na bodziec zewnętrzny Schemat został użyty za zgodą redakcji American Chemical Society 29. Acylohydrazony mogą tworzyć z kationami metali kompleksy przy udziale wiązania z ugrupowaniem C(O)N(H)N. W literaturze znanych jest wiele dyskretnych lub jednowymiarowych połączeń opartych na hydrazonach. Zostały one obszernie opisane w artykułach związanych z (i) hydrazonowymi przełącznikami30, (ii) bimetalicznymi kompleksami31 czy (iii) chelatowymi związkami żelaza w terapii przeciwnowotworowej32. Należy jednak zauważyć, że liczba połączeń posiadających wyższą wymiarowość (2D i 3D) jest niewielka33-39. Niedawno hydrazony z powodzeniem zostały użyte do konstrukcji dwuwymiarowych sieci organicznych (COFs)40,41. Natomiast w niniejszej pracy po raz pierwszy syntezowano grupę porowatych polimerów koordynacyjnych o mieszanych ligandach na bazie aroilohydrazonów oraz kwasów dikarboksylowych25. 1.3 Stabilność sieci MOF Do tej pory otrzymano wiele materiałów typu MOF, ale tylko te, które cechuje wysoka odporność chemiczna oraz stabilność hydrotermiczna mogą spowodować przełom w nowoczesnej gospodarce i z powodzeniem zostać użyte w katalizie, medycynie, wyłapywaniu i magazynowaniu gazów oraz elektronice. Dlatego należy poznać zależność pomiędzy budową wewnętrzną, właściwościami fizykochemicznymi oraz stabilnością materiału w różnych warunkach10,42-44. Przykładowo archetypowe MOFy (np. MOF-5 i izostrukturalna seria, HKUST-1, DUT-4), które wywarły duży wpływ na środowisko naukowe oraz rozwój chemii MOF, są nietrwałe w obecności wody45-47. Dla wielu MOF-ów najsłabszym punktem podczas kontaktu z wodą jest natura wiązania koordynacyjnego metal-ligand i degradacja tych sieci w warunkach hydrotermalnych jest zazwyczaj związana z substytucją liganda przez cząsteczkę wody (Rysunek 7). Taki mechanizm degradacji sieci metalo-organicznej jest bardzo często obserwowany dla karboksylano-cynkowych materiałów MOF46. Ważną i istotną gałęzią nauki związanej z acylohydrazonami jest chemia foto- i termo- przełączników, które ulegają zmianie pod wpływem interakcji ze światłem lub pod wpływem zmiany temperatury (Schemat 2)29. Podczas absorpcji promieniowania dochodzi do izomeryzacji geometrycznej i zmiany izomeru E na izomer Z lub na odwrót. I 14 Rysunek 7 Zmiana strukturalna sieci MOF-5 w obecności 2,3 % wagowego wody (góra) oraz zastępowanie atomu tlenu łączącego cztery atomy Zn przez cząsteczkę wody (dół). Zn-fioletowy, O-czerwony, C- szary, H-biały; tetraedryczną geometrię atomów Zn w klastrach ZmO obrazują niebieskie wielościany. Rysunek został użyty za zgodą redakcji American Chemical Society46. Dla przemysłowego zastosowania MOF-ów w technologiach związanych z obecnością wody, stabilność hydrotermiczna jest kluczowym czynnikiem. Przykładowo komercyjnie dostępne MOFy Basolite-300C (HKUST-1) i Basolite F300 (FeBTC) podczas wielokrotnej adsorpcji i desorpcji wody tracą porowatość. Po 20 cyklach maksymalna pojemność dla par wody spada do 53% (HKUST-1) i 74% (Fe-BTC) wartości pierwotnej48. Istotnym również zagadnieniem jest wytrzymałość III generacji sieci MOF na powtarzającą się adsorpcję i desorpcję gazów istotnych dla gospodarki. Wiele materiałów, które posiadają interesujące dynamiczne zachowanie nie zachowuje swoich właściwości pod wpływem wielokrotnej adsorpcji i desorpcji gazu. Przykładowo Bon i współautorzy wykonali po sto cykli adsorpcji/desorpcji butanu i azotu49, kolejno w temperaturze pokojowej i 77 K na materiałach DUT-8(Ni), ELM-11, MIL-53(Al), SNU-9. W przypadku DUT-8(Ni) zaobserwowano spadek pojemności sorpcyjnej (blisko o 80%) oraz zanik dynamicznego zachowania względem bodźca. Wielokrotna adsorpcja i desorpcja dla materiału SNU-9 spowodowała zmianę kształtu izotermy adsorpcji oraz zmniejszenie pojemności sorpcyjnej. Natomiast MOFy MIL-53(Al) i ELM-11 w wyniku cyklicznej adsorpcji i desorpcji butanu w temperaturze pokojowej nie zmieniły maksymalnej pojemności sorpcyjnej oraz zachowały swój dynamiczny charakter manifestujący się skokowym przebiegiem izotermy adsorpcji (Rysunek 8). I 15 Rysunek 8 a) Zmiany strukturalne pomiędzy strukturą o dużych porach i małych porach dla materiału MIL-53(Al) pod wpływem adsorpcji n-butanu. b) Izoterma adsorpcji n-butanu w 298 K (lewa część) i azotu w 77 K (prawa część) uzyskana dla materiału MIL-53 (czerwone koła), po 10 cyklach adsorpcji/desorpcji (zielone trójkąty) i po 100 cyklach adsorpcji/desorpcji (niebieskie kwadraty). Rysunek został użyty za zgodą redakcji American Chemical Society49. Dodatkowy przegląd literaturowy, który może być pomocny podczas analizy przewodnika po publikacjach został sporządzony do każdej z publikacji i można go znaleźć w sekcji „Introduction” w dołączonych artykułach naukowych. I 16 2. Cel pracy Celem pracy było otrzymanie nowej rodziny sieci metalo-organicznych o mieszanych łącznikach na bazie acylohydrazonów, kwasów dikarboksylowych oraz cynku(II) i kadmu(II). Wybór acylohydrazonów na bloki budulcowe był związany z posiadaniem przez nie grupy - CONHN-, która może tworzyć wewnątrz porów specyficzne oddziaływania z cząsteczkami gości. Istotnym jest poznanie korelacji pomiędzy strukturą wewnętrzną, stabilnością a właściwościami materiału, co bezpośrednio może przyczynić się do projektowania sieci o zadanych cechach dla potrzeb nowoczesnej gospodarki. Większość materiałów MOF nie jest stabilna względem destruktorów takich jak woda, powtarzające się cykle desorpcji i adsorpcji, czy wysoka temperatura. Dlatego jednym z szczegółowych celów pracy było zbadanie wpływu podstawników oraz ugrupowania acylohydrazonowego, będącego zarazem donorem i akceptorem wiązań wodorowych, na stabilność sieci MOF. Synteza sieci metalo-organicznych zazwyczaj konsumuje duże ilości drogich, organicznych rozpuszczalników i energii. Biorąc to pod uwagę celem było także opracowanie i zrozumienie alternatywnych ścieżek syntezy acylohydrazonowo-karboksylanowych sieci metalo- organicznych z użyciem mechanosyntezy, wpisującej się w kanon zielonej chemii. | 17 3. Metodyka badań Rozwiązanie struktur krystalicznych było prowadzone w oparciu o rentgenostrukturalne badania na materiałach monokrystalicznych (SC-XRD) przy użyciu dyfraktometru monokrystalicznego Oxford Diffraction SuperNova Dual (dr Maciej Hodorowicz). Natomiast do wizualizacji struktur były używane programy Mercury, Diamond oraz Topos (doktorant). Analiza wymiarowości, rozmiaru oraz powierzchni właściwej porów była przeprowadzana w oparciu o pomiary sorpcyjne w Dreźnie (N2, CO2, H2, H2O) na aparatach Autosorb IQ, BELSORP-max, Hydrosorb (pomiary przeprowadzał doktorant lub dr Irena Senkovska) oraz konfrontowana z wynikami obliczeń programów Zeo++ i Poreblazer (doktorant) lub uzyskanych przy użyciu metod GCMC (Grand Canonical Monte Carlo; mgr Andrzej Sławek). Monitorowanie in situ procesu adsorpcji CO2 za pomocą spektroskopii IR było wykonane na aparacie Bruker Tensor 27. Do pomiarów dyfrakcji rentgenowskiej na materiałach polikrystalicznych używano dyfraktometr PXRD (Rigaku Miniflex 600) z przystawką temperaturową Anton-Paar BTS500 pozwalającą na ogrzewanie próbki do 500 oC. Do pomiarów widm w podczerwieni wykorzystywano spektrometry IR Nicolet iS7 lub iS5 z przystawką do ciał stałych (ATR) iD7 lub iD5. Powyższe pomiary (zarówno PXRD jak i IR) przeprowadzał głównie doktorant, przy czym w niektórych przypadkach licencjanci lub magistrantka opiekuna naukowego. Synteza mechanochemiczna w pierwszym etapie powstawania doktoratu była przeprowadzana w moździerzu (doktorant, licencjant Damian Jędrzejowski), a następnie w młynie kulowym Retsch MM 200 (doktorant i licencjanci lub magistrantka opiekuna naukowego). Do pomiarów widm w zakresie UV-Vis używano spektrometru Shimadzu UV-3600 z przystawką do ciał stałych (pomiary przeprowadzał doktorant). Pomiary termograwimetryczne (aparat Mettler Toledo TGA/SDTA 851e), analiza elementarna pierwiastków (N, C, H, S; analizator Vario MICRO Cube) oraz spektroskopia magnetycznego rezonansu jądrowego ('H NMR; aparat Bruker Avance III 600, temperatura 300 K) były wykonywane w pracowniach wydziałowych. Dodatkowe szczegóły eksperymentalne można znaleźć w sekcjach: „Experimental Section" lub "Materials and Methods” w każdej z dołączonych publikacji. Wszystkie syntezy opracował doktorant oraz wykonał większość z nich, przy czym w niektórych przypadkach przeprowadzali je licencjanci (Damian Jędrzejowski, Monika Szufla) lub magistrantka (Magdalena Lupa) promotora, ściśle współpracując z doktorantem. | 18 4. Przewodnik po publikacjach Dla czytelności zrealizowane zadania badawcze w obrębie pracy doktorskiej zostały usystematyzowane w formie listy zgodnie z numeracją przyjętą dla publikacji. Dodatkowo na Rysunku 9 zestawiono wzory strukturalne organicznych prekursorów użytych do konstrukcji sieci MOF. I. Synteza, określenie struktury i właściwości pierwszych acylohydrazonowo-karboksylanowych sieci MOF opartych na jonach cynku, izonikotynoilo-hydrazonie aldehydu 4-pirydynowego (pcih) i kwasie tereftalowym (H2bdc) lub kwasie 4,4'-bisfenylodikarboksylowym (H2bpdc). II. Optymalizacja syntezy mechanochemicznej izoftalowo-hydrazonowgo Zn-MOFu wykazującego dynamiczną i selektywną adsorpcję CO2. III. Badanie wpływu podstawnika na ułożenie warstw, stabilność oraz właściwości sorpcyjne serii Zn-MOFów zawierających podstawione izoftalany (Xiso2-). IV. Teoretyczny oraz eksperymentalny wgląd w proces adsorpcji i hydrotermiczną stabilność wodoodpornego Cd-MOFu opartego na kwasie 4,4'-sulfonobenzenodikarboksylowym (H2sdb). V. Kontrola budowy węzłów sieci izoftalanowo-hydrazonowych Cd-MOFów za pomocą podstawnika oraz warunków syntezy. VI. Wykorzystanie liniowego krótszego oraz dłuższego łącznika acylohydrazonwego do otrzymania kadmowych sieci metalo-organicznych. Rysunek 9 Wzory strukturalne organicznych prekursorów użytych w preparatyce sieci MOF. | 19 Ad. I Synteza, określenie struktury i właściwości pierwszych acylohydrazonowo- karboksylanowych sieci MOF opartych na jonach cynku, izonikotynoilo-hydrazonie aldehydu 4- pirydynowego (pcih) i kwasie tereftalowym (H2bdc) lub kwasie 4,4'-bisfenylodikarboksylowym (H2bpdc). Cynkowe sieci metalo-organiczne o ogólnym wzorze {[Zn2(dcx)2(pcih)2]goście}n (dcx = liniowy dikarboksylowy anion; pcih = hydrazon) zostały syntezowane poprzez ogrzewanie azotanu cynku, uprzednio otrzymanego izonikotynoilohydrazonu aldehydu 4-pirydynowego (pcih) oraz odpowiedniego kwasu dikarboksylowego; dcx = bdc2-, anion 1,4-benzenodikarboksylowy; lub bpdc2-, anion 4,4'- bisfenylodikarboksylowy (Schemat 3). Schemat 3 Ścieżka syntezy mikroporowatych sieci {[Zn2(dcx)2(pcih)2]goście}n opartych na kwasie a) 1,4-benzenodikarboksylowym i b) 4,4'-bisfenylodikarboksylowym. Dodatkowo przedstawiono fragmenty strukturalne [Zn2(dcx)2(pcih)2]. Sieci 1-2 są trójwymiarowymi polimerami koordynacyjnymi typu podpórkowanych warstw. Karboksylany koordynują ekwatorialnie do atomów cynku tworząc warstwy [Zn2(dcx)2]n, które są połączone ze sobą przez aksjalny obojętny łącznik pcih, tworząc trójwymiarowy szkielet. W wyniku interpenetracji, dwukrotnej dla MOFa 1 i trójkrotnej dla sieci 2 dochodzi do zmniejszenia się wolnych przestrzeni (Rysunek 10). Rysunek 10 Struktura krystaliczna oraz izotermy adsorpcji (pełne symbole) i desorpcja (puste symbole) CO2 (195 K) i N2 (77 K) dla sieci: a) [Zn2(bdc)2(pcih)2]n (1) oraz b) [Zn2(bpdc)2(pcih)2]n (2). Analiza termograwimetryczna, ex situ IR oraz PXRD pozwoliły na określenie warunków aktywacji oraz zbadanie stabilności sieci metalo-organicznych 1-2. Materiały są termicznie stabilne do 330 oC (1) i 340 oC (2), a wyniku termicznej aktywacji MOF 1 traci jedną cząsteczkę N,N'-dimetyloformamidu DMF oraz dwie H2O (ubytek masy: znaleziony 11,8 %; obliczony 10,7 %), natomiast MOF 2 dwie cząsteczki DMF i jedną H2O (ubytek masy: znaleziony 12,5 %; obliczony 13,3 %). Analiza sorpcyjna sieci 1-2 wykazała, że materiały selektywnie adsorbują CO2 i są praktycznie nieporowate względem N2 (Rysunek 10). W oparciu o izotermy adsorpcji/desorpcji dla CO2 I 20 wyznaczono powierzchnie właściwe równe 203 m2-g-1 (1) i 293 m2-g-1 (2), natomiast geometryczna powierzchnia właściwa BET obliczona za pomocą programu Poreblazer wyniosła 408 m2-g-1 dla 1 i 383 m2-g-1 dla 2. Obliczenia wykonane za pomocą programu50 Zeo++ wskazują, że "okno" porów dla materiału 1 wynosi 3,14 Â. Większa cząsteczka N2 (średnica kinetyczna 3,64 Â) nie ulega adsorpcji z powodu ograniczeń geometrycznych. Natomiast w wyniku drgań sieci oraz oddziaływania z grupami acylohydrazonowymi dochodzi do adsorpcji CO2 (średnica kinetyczna 3,30 Â), która jest limitowana przez dyfuzję, co jest demonstrowane przez nachylenie krzywej adsorpcji wynikającej z utrudnionego dostępu do wolnych przestrzeni. W przypadku sieci 2 okno poru ma średnicę 3,52 Â, dlatego dla CO2 nie obserwuje się efektu limitującego i izoterma adsorpcji jest I typu (fizysorpcja). Ad. II Optymalizacja syntezy mechanochemicznej izoftalowo-hydrazonowgo Zn-MOFu wykazującego dynamiczną i selektywną adsorpcję CO2. Zastępując liniowe dikarboksylany kwasem izoftalowym (H2iso) uzyskano dwuwymiarową sieć metalo-organiczną {[Zn2(iso)2(pcih)2]-2DMF}n (3) zbudowaną z jednowymiarowych łańcuchów cynkowo-karboksylanowych połączonych aksjalnie w polimer dwuwymiarowy (2D) przez ligand pcih (Rysunek 11 a). Do syntezy wykorzystano klasyczną metodę w roztworze (96 h; 60 oC; wydajność 72,3%) oraz mechanosyntezę z dodatkiem niewielkiej ilości cieczy tzw. LAG (ang. liquid-assisted grinding). Stałe reagenty były ucierane w agatowym moździerzu, synteza trwała 8-20 minut; otrzymano czysty polimer koordynacyjny z wydajnością bliską 100%, co zostało potwierdzone przez analizę elementarną, IR, oraz PXRD (Rysunek 11 b,c). Rysunek 11 {[Zn2(iso)2(pcih)2]-2DMF}n (3): a) struktura krystaliczna; b) możliwe ścieżki syntezy mechanochemicznej oraz c) dyfraktogramy proszkowe preparatów otrzymanych z różnych substratów cynkowych, porównane z dyfraktogramem obliczonym za pomocą programu Mercury. Zbadano czynniki wpływające na przebieg reakcji mechanochemicznej, między innymi wpływ soli używanych w preparatyce, czy stosunek objętości rozpuszczalnika (^L) do sumy masy reagentów (mg), tak zwany współczynnik n51. W przypadku ZnCO3, Zn(OH)2 i Zn(CH3COO)2 otrzymano ilościowo krystaliczne produkty przy stosunku DMF do reagentów wynoszącym n = 0,49-0,63 ^L/mg, natomiast dla ZnO należało zwiększyć n do 1,58 ^L/mg (150 ^L DMF na 94,7 mg substratów) lub użyć ok. 1 % wag. dodatku soli NH4NO3 lub (NH4)2SO4 w celu całkowitego przeragowania reagentów. Natomiast synteza w wersji bez dodatku cieczy lub z użyciem MeOH, EtOH, H2O jak i/lub z ZnNO3 /ZnCh nie prowadzi do powstania 3. Generalnie niższe energie sieciowe wyjściowych reagnetów stałych sprzyjają tworzeniu się produktu. W prawdzie energie sieciowe ZnO (3971 kJ-mol"1), ZnCO3 (3273 kJ-mol"1) i Zn(OH)2 (3158 kJ-mol-1) są wyższe niż energie sieciowe Zn(NO3)2 (2649 kJ-mol-1) i ZnCh (2748 kJ-mol-1), jednak rozważając entalpię reakcji jako główną siłę napędową rekacji należy też wziąźć pod uwagę tworzenie się stabilnych produktów ubocznych. W przypadku stosowania soli lub tlenków o zasadowym charakterze powstające produkty uboczne (CO2, H2O i CH3COOH) charakteryzują się stosunkowo wysokimi ujemnymi standardowymi entalpiami tworzenia. I 21 Oprócz odpowiedniego reagenta wyjściowego dodatek DMF który w produkcie końcowym zajmuje pory jest konieczny do formowania sieci 3. MOF może być również otrzymywany na drodze mechanosyntezy z przejściowego polimeru koordynacyjnego [Zn(iso)(H2O)]n (Rysunek 11 b).W pierwszym kroku w wyniku ucierania ZnO i H2iso w obecności wody (60 ^L rozpuszczalnika i 49,5 mg reagentów) otrzymano znany w literaturze dwuwymiarowy polimer koordynacyjny [Zn(iso)(H2O)]n zidentyfikowany za pomocą PXRD. W kolejnym kroku produkt przejściowy został utarty z ligandem pcih w obecności DMF, co prowadziło do uzyskania sieci MOF {[Zn2(iso)2(pcih)2p2DMF}n (3). Reakcja w ciele stałym pomiędzy produktem przejściowym oraz ligandem pcih wymaga usunięcia cząsteczki wody ze sfery koordynacyjnej cynku jak również zastąpienie warstw [Zn(iso)]n przez jednowymiarowe podwójne łańcuchy [Zn2(iso)2]n i ich połączenie za pomocą N,N-donorowego liganda acylohydrazonowego (pcih). Aktywowany materiał 3 wykazuje dynamiczne właściwości sorpcyjne względem CO2, w pierwszej kolejności przy ciśnieniu pp = 0,17 (odpowiadającym objętości porów Vp = 0,087 cm3-g_1) dochodzi do wypełnienia porów 0D (pierwsze plateau na krzywej; Rysunek 12). Następnie ulegają rozrywaniu wiązania wodorowe i dochodzi do separacji warstw. Prowadzi to do powstania dodatkowej przestrzeni i dalszej adsorpcji CO2 (p/p0 = 0,99; Vp = 0,144 cm3-g_1). Obserwacje te są spójne z obliczeniami w programach Zeo++ i Mercury. Teoretyczna objętość porów po usunięciu rozpuszczalnika wynosi 0,090 cm3-g_1, co odpowiada pierwszemu plateau. Rysunek 12 a) Izoterma adsorpcji N2 i CO2 dla {[Zn2(iso)2(pcih)2]-2DMF}n (3). b) Wizualizacja wolnych przestrzeni w strukturze {[Zn2(iso)2(pcih)2]n (promień sondy 1.2 Â; program Mercury). c) Wiązania wodorowe łączące naprzemienne warstwy. Ad. III Badanie wpływu podstawnika na ułożenie warstw, stabilność oraz właściwości sorpcyjne serii Zn-MOFów zawierających podstawione izoftalany (Xiso2-). Otrzymano w roztworze i mechanochemicznie rodzinę sieci metalo-organicznych {[Zn2(Xiso)2(pcih)2]-goście}n zawierających 5-podstawione izoftalany [Xiso2-, X = H (3,4), lub CH3 (5), lub OH (6), lub NH2 (7)] oraz izonikotynoilo-hydrazon aldehydu 4-pirydynowego (pcih). Za pomocą badań rentgenostrukturalnych określono ich wewnętrzną budowę, wykazując, że materiały 4- 7 posiadają warstwową strukturę analogiczną do materiału 3, natomiast za wzajemne ułożenie warstw są odpowiedzialne supramolekularne oddziaływania: wiązania wodorowe lub oddziaływania CH—rc pomiędzy przyległymi warstwami lub wiązania wodorowe między warstwami oraz cząsteczkami gości w sieci 4 (Rysunek 13). | 22 Rysunek 13 a) Supramolekularne oddziaływania pomiędzy przyległymi warstwami i/lub cząsteczkami gości (góra) oraz wzajemne ułożenie warstw (dół) dla sieci 3-6 idąc od lewej. b) Izotermy adsorpcji CO2 (195 K) dla rodziny {[Zn2(Xiso)2(pcih)2]}n (w skali logarytmicznej) wskazujące na różnice w zakresie niskich ciśnień. c) Wycinek widm IR rejestrowanych in situ podczas adsorpcji CO2 (zakres 2310-2360 cm-1). DMA = N,N-dimetyloacetamid. Oddziaływania te są również odpowiedzialne za kształt oraz objętość międzywarstwowych porów. Wszystkie materiały z tej rodziny selektywnie adsorbują CO2 względem N2. Wykazano, że polarne podstawniki (NH2 i OH) zwiększają chemiczne powinowactwo materiału do CO2 oraz stabilność termiczną jak i hydrolityczną. Nie udało się wyznaczyć struktury dla materiału opartego na NH2iso2-, natomiast analiza elementarna, IR, oraz PXRD potwierdzają jego izostrukturalność z siecią {[Zn2(OHiso)2(pcih)2]}n. Ad. IV Teoretyczny oraz eksperymentalny wgląd w proces adsorpcji oraz hydrotermiczną stabilność wodoodpornego Cd-MOFu opartego na 4,4'-sulfonobenzenodikarboksylowym kwasie (H2sdb). Wykorzystując sfunkcjonalizowany kątowy kwas 4,4'-sulfonobenzenodikarboksylowy oraz ligand pcih otrzymano mikroporowatą, pierwszą w rodzinie nieinterpenetrowaną trójwymiarową sieć o wzorze {[Cd2(sdb)2(pcih)2]-2DMF-H2O}n (8). Materiał ten wykazuje wysoką stabilność termiczną do 300 oC, wysoką odporność hydrotermalną (24h we wrzącej wodzie, 260 oC w parach H2O, lekko kwaśne środowisko do pH = 3). Aktywowany materiał selektywnie adsorbuje CO2 (195 K) względem N2 (77 K), jak również wykazuje odwracalną (35 cykle) adsorpcję H2O (298 K) potwierdzoną przez badania w warunkach izotermicznych oraz izobarycznych. Rysunek 14 a) Izoterma adsorpcji (pełne symbole) i desorpcji (puste symbole) H2O w 298 K (niebieski) i 308 K (czarny) oraz symulowana w 298 K (żółte symbole). b) Hydrotermiczna stabilność sorpcyjna podczas powtarzanych izobarycznych pomiarów desorpcji/adsorpcji par wody: zależność pojemności sorpcyjnej (czarny) i temperatury (czerwony) od liczby cykli. c) reprezentacja wiązań wodorowych między cząsteczkami wody oraz trzema akceptorami z sieci. Eksperymentalne izosteryczne ciepło adsorpcji dla wody (48,9 kJ/mol) wskazuje na niespecyficzne oddziaływania wody z siecią. Powolna wymiana w komorze klimatycznej cząsteczek gości na wodę umożliwiła przejście kryształ-kryształ dzięki czemu rozwiązano i porównano struktury zsyntezowanej | 23 sieci oraz po wymianie cząsteczek gości. Badania SC-XRD pozwoliły na wizualizację orientacji cząsteczek wody w strukturze, tworzą one pięcioczłonowy klaster stabilizowany przez trzy typy silnych oddziaływań wodorowych z udziałem grup dekorujących kanały (Rysunek 14 c). Wnikliwy wgląd w mechanizm adsorpcji wody oraz wyjaśnienie roli grup hydrazonowych w selektywnej adsorpcji CO2 podparto obliczeniami Monte Carlo (GCMC) jak również badaniami in situ adsorpcji CO2. Ad. V Kontrola budowy węzłów sieci izoftalanowych Cd-MOF-ów za pomocą podstawnika oraz warunków syntezy. Otrzymano serię trzech kadmowych hydrazonowo-kraboksylanowych MOF-ów (9-11) opartych na 5-tertbutylo podstawionym izoftalanie (tBu-iso2-), oraz niepodstawionym kwasie izoftalowym (iso2-). Wykazano, że dla sieci z dużym tertbutylowym podstawnikiem w zależności od warunków syntezy tworzą się dwuwymiarowe polimery koordynacyjne zbudowane z łańcuchów [Cd(tBu-iso)]n (9) i [Cd2(tBu-iso)2]n (10) połączonych za pomocą łącznika pcih. Natomiast w przypadku niepodstawionego izoftalanu (iso2-), niezależnie od warunków syntezy, zawsze otrzymuje się ten sam produkt izostrukturalny do sieci 10 (Rysunek 15 a). Rysunek 15 a) Fragment struktury krystalicznej MOFów 9-11. b) Izotermy adsorpcji dla materiałów 9-11. Materiały te również selektywnie adsorbują CO2 względem N2. W oparciu o strukturę krystaliczną, pomiary PXRD w funkcji temperatury, analizę termograwimetryczną oraz parametry porowatości wyznaczone za pomocą programów komputerowych (Zeo++, Mercury) wytłumaczono wpływ budowy węzła oraz podstawników na właściwości sorpcyj ne otrzymanych polimerów, j ak również ich stabilność termiczną. Ad. VI Wykorzystanie krótszego oraz dłuższego łącznika acylohydrazonowego do otrzymania kadmowych sieci metalo-organicznych. Jedną z możliwych ścieżek syntezy dynamicznych polimerów koordynacyjnych jest dobór łączników, które posiadają ugrupowania atomów mogące ulegać izomeryzacji lub rotacji. Zgodnie z tą zasadą oraz celem zwiększenia objętości porów w stosunku do rodziny opartej na łączniku pcih otrzymano i po raz pierwszy użyto do preparatyki sieci MOF łącznik diacylohydrazonowy (tdih). (Rysunek 16 a). Ugrupowania acylohydrazonowe łącznika tdih i tetraedryczny atom tlenu z H2oba łączący pierścienie fenylowe zapewniają możliwość rotacji fragmentów łączników wokół pojedynczych wiązań C-C i C-O, co może być sposobem na wprowadzenie dynamiki strukturalnej w sieć MOF. I 24 Rysunek 16 a) Komponenty wykorzystane w syntezie sieci {[Cd2(oba)2(tdih)2]-6DMF-7H2O}n (12). b) Możliwe metody syntezy sieci 12. c) Struktura krystaliczna sieci [Cd2(oba)2(tdih)2]n ukazująca potencjalnie dostępne przestrzenie oraz dwa typy kanałów (usunięto cząsteczki gości). W oparciu o H2oba, tdih oraz jony Cd2+ otrzymano w roztworze oraz na drodze mechanochemicznej (wariant bezpośredni i dwuetapowy) porowatą trójwymiarową sieć {[Cd2(oba)2(tdih)2]-6DMF-7H2O}n (12). Natomiast zastąpienie łącznika tdih (~20 Â), krótszym hydrazonem pcih (~11 Â) prowadziło do uzyskania nieizostrukturalnej sieci MOF o budowie warstwowej (13). Rysunek 17 Struktura krystaliczna sieci MOF 12 (góra) i 13 (dół): a) sfera koordynacyjna wraz z ligandami [Cd2(oba)2(tdih)2] (12) i [Cd2(oba)2(pcih)2] (13). b) Fragment [Cd2(oba)2]n (warstwa (2D) w 12; 1D podwójny łańcuch w 13), c) wiązania wodorowe występujące w sieciach 12 i 13 , d) reprezentacja wolnych przestrzeni (program Mercury, promień sondy 1,2 Â). W podpunktach a), b) i d) atomy wodoru zostały ominięte dla czytelności rysunku. Za pomocą SC-XRD określono budowę obu MOF-ów, wykazując, że w sieci 12 i 13 znajdują się dwa typy kanałów (Rysunek 17). TGA wykazała ubytek masy o 27 % (12) i o 16 % (13; ubytek związany z ewakuacją cząsteczek gości), natomiast obliczenia wykazują, że wolne przestrzenie (okupowane przez gości) zajmują 43,5 % objętości komórki elementarnej. Po namoczeniu w H2O, MOF 12 ulega przemianie fazowej, co potwierdzają badania PXRD, niestety nie udało się rozwiązać struktury po przemianie fazowej. Warto zwrócić uwagę, że w strukturze po wymianie DMFu na wodę znajduje się tylko 13 cząsteczek H2O {[Cd2(oba)2(tdih)2]-13H2O}n (13 % mas.; TGA, analiza elementarna), co świadczy o tym, że część struktury pierwotnej nie jest dla nich dostępna. W wyniku oddziaływania sieci z cząsteczkami gości, lub przy ich usuwaniu zachodzą zmiany strukturalne. W przypadku sieci 13 wymiana cząsteczek gości z DMF na H2O jest w pełni odwracalna. I 25 5. Podsumowanie Wykorzystując dwa łączniki acylohydrazonowe pełniące funkcję N-donorowych ligandów, wybrane kwasy dikarboksylowe oraz jony cynku(II) i kadmu(II) otrzymano rodzinę acylohydrazonowo- karboksylanowych MOF-ów. W wyniku pracy laboratoryjnej określono strukturę i dokonano charakterystyki właściwości fizykochemicznych trzynastu nowych sieci MOF opartych na mieszanych łącznikach. Dwa pierwsze połączenia (1,2), które otrzymano poprzez użycie liniowych dikarboksylanów (H2bdc i H2bpdc) oraz krótszego łącznika acylohydrazonowego (pcih) to trójwymiarowe interpenetrowane polimery koordynacyjne typu podpórkowanych warstw. Zmieniając liniowy ligand na kątowy kwas izoftalowy oraz jego pochodne (H2Xiso; X = H lub CH3, lub tBu, lub OH, lub NH2) uzyskano serię dwuwymiarowych sieci MOF (3-7, oraz 9-11). Dla sieci 3 zbadano czynniki wpływające na przebieg reakcji mechanochemicznej wykazując, że MOF 3 może być otrzymany w reakcji mechanosyntezy pomiędzy wszystkimi komponetami z dodatkiem niewielkiej ilości cieczy lub otrzymywany z produktu przejściowego - polimeru koordynacyjnego [Zn(iso)(H2O)]n. Ponadto wykazano, że do syntezy jest konieczne użycie odpowiedniego materiału wyjściowego, co zostało wyjaśnione w oparciu o analizę energii sieciowych substratów i standardowych entalpii tworzenia produktów ubocznych. Z kolei w przypadku jonów kadmu i kwasu 5-tertbutyloizoftalowego w zależności od warunków syntezy w roztworze wykazano, że tworzą się polimery koordynacyjne zbudowane z jednowymiarowych łańcuchów [Cd(tBu-iso)]n (9) i [Cd2(tBu-iso)2]n (10) połączonych w sieci 2D za pomocą liganda pcih, natomiast dla MOFa 11 opartego na kadmie(II) i niepodstawionym izoftalanie takiej zależności się nie obserwuje. Wykorzystując sfunkcjonalizowany kątowy kwas 4,'4-sulfonobenzenodikarboksylowy oraz ligand pcih otrzymano mikroporowatą, pierwszą w rodzinie nieinterpenetrowaną trójwymiarową sieć o wzorze {[Cd2(sdb)2(pcih)2]-2DMF-H2O}n (8). W celu zwiększenia objętości porów w stosunku do rodziny opartej na łączniku pcih otrzymano i wykorzystano do preparatyki łącznik diacylohydrazonowy (tdih), oraz kwas 4,4'-okso-bis(benzenodikarboksylowy) H2oba. W oparciu o te łączniki otrzymano w roztworze oraz na drodze mechanochemicznej (bezpośredniej i przez produkt pośredni) porowatą trójwymiarową sieć {[Cd2(oba)2(tdih)2p6DMF-7H2O}n (12). Natomiast zastąpienie łącznika tdih (~20 Â), krótszym acylohydrazonem pcih (~11 Â) prowadziło do uzyskania nieizostrukturalnej dwuwymiarowej sieci MOF (13). Dla materiałów 1-13 zbadano właściwości sorpcyjne a dla sieci 3-13 przeprowadzono również badania stabilności względem wielu destruktorów. Materiały 1-13 wykazują różne właściwości względem adsorbowanych gazów w zależności od budowy wewnętrznej. Przykładowo podczas adsorpcji CO2 w 195 K w przypadku sieci MOF 1 obserwuje się limitowanie adsorpcji przez dyfuzję spowodowane rozmiarem okna porów. Natomiast dla sieci 2 nie obserwuje się tego efektu. Interesujące zachowanie podczas adsorpcji CO2 (195 K) wykazuje dwuwymiarowy MOF 3. W pierwszej kolejności podczas adsorpcji dochodzi do wypełnienia 0D porów. Następnie ulegają rozrywaniu wiązania wodorowe i dochodzi do rozsunięcia warstw. Prowadzi to do powstania dodatkowej przestrzeni i dalszej adsorpcji CO2. W przypadku układów warstwowych 3-7 wykazano, że polarne podstawniki (NH2 i OH) zwiększają chemiczne powinowactwo materiału do CO2 oraz stabilność termiczną jak i hydrolityczną. Materiał 12 oparty na dłuższym łączniku tdih w wyniku interakcji z H2O ulega nieodwracalnej przemianie strukturalnej. Natomiast w przypadku sieci 13 wymiana cząsteczek gości z DMF na H2O jest w pełni odwracalna. Trójwymiarowa mikroporowata sieć 8 wykazuje wysoką stabilność termiczną do 300 oC i wysoką odporność hydrotermalną. Dodatkowo aktywowany materiał selektywnie adsorbuje CO2 (195 K) względem N2 (77 K), jak również wykazuje odwracalną (35 cykle) adsorpcję H2O (298 K) potwierdzoną przez badania w warunkach izotermicznych oraz izobarycznych. Wnikliwy wgląd w mechanizm adsorpcji wody oraz wyjaśnienie roli grup acylohydrazonowych w selektywnej adsorpcji CO2 podparto obliczeniami Monte Carlo (GCMC) jak również badaniami in situ adsorpcji CO2. I 26 6. Bibliografia 1 S. R. Batten, N. R. Champness, X.-M. Chen, J. Garcia-Martinez, S. Kitagawa, L. Öhrström, M. O’Keeffe, S. M. Paik and J. Reedijk, Pure Appl. Chem., 2013, 85, 1715-1724. 2 H. Li, M. Eddaoudi, M. O’Keeffe and O. M. Yaghi, Nature, 1999, 402, 276-279. 3 S. S.-Y. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen and I. D. Williams, Science, 1999, 283, 1148-1150. 4 S. Horike, S. Shimomura and S. Kitagawa, Nat. Chem., 2009, 1, 695-704. 5 A. Schneemann, V. Bon, I. Schwedler, I. Senkovska, S. Kaskel and R. A. Fischer, Chem. Soc. Rev., 2014, 43, 6062-6096. 6 D. Matoga, K. Roztocki, M. Wilke, F. Emmerling, M. Oszajca, M. Fitta and M. Balanda, CrystEngComm, 2017, 19, 2987-2995. 7 D. Matoga, M. Oszajca and M. Molenda, Chem. Commun., 2015, 51, 7637-7640. 8 D. Matoga, B. Gil, W. Nitek, A. D. Todd and C. W. Bielawski, CrystEngComm, 2014, 16, 4959- 4962. 9 K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, Inorg. Chem., 2016, 55, 9663-9670. 10 K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, Cryst. Growth Des., 2017, 18, 488-497. 11 Y. Cheng, H. Kajiro, H. Noguchi, A. Kondo, T. Ohba, Y. Hattori, K. Kaneko and H. Kanoh, Langmuir, 2011, 27, 6905-6909. 12 U. Stoeck, S. Krause, V. Bon, I. Senkovska and S. Kaskel, Chem. Commun., 2012, 48, 10841-10843. 13 S. Krause, V. Bon, I. Senkovska, U. Stoeck, D. Wallacher, D. M. Többens, S. Zander, R. S. Pillai, G. Maurin, F.-X. Coudert and S. Kaskel, Nature, 2016, 532, 348. 14 D. Farrusseng (Ed.), Metal-Organic Frameworks, Wiley- VCH Verlag & Co. KGaA, Weinheim, 2011. 15 R.-B. Lin, S. Xiang, B. Li, Y. Cui, W. Zhou, G. Qian and B. Chen, Isr. J. Chem, 2018, 58, 949-961. 16 H. Kim, S. Yang, S. R. Rao, S. Narayanan, E. A. Kapustin, H. Furukawa, A. S. Umans, O. M. Yaghi and E. N. Wang, Science, 2017, 356, 430-434. 17 J. A. Prince, S. Bhuvana, V. Anbharasi, N. Ayyanar, K. V. K. Boodhoo and G. Singh, Sci. Rep., 2014, 4, 6555. 18 M. F. de Lange, K. J. F. M. Verouden, T. J. H. Vlugt, J. Gascon and F. Kapteijn, Chem. Rev., 2015, 115, 12205-12250. 19 M. Hefti, L. Joss, Z. Bjelobrk and M. Mazzotti, Faraday Discuss., 2016, 192, 153-179. 20 S. Kaskel (Ed.), The Chemistry of Metal-Organic Frameworks: Synthesis, Characterization, and Applications, Wiley- VCH Verlag & Co. KGaA, Weinheim, 2016. 21 Y.-B. Huang, J. Liang, X.-S. Wang and R. Cao, Chem. Soc. Rev., 2017, 46, 126-157. 22 J. W. Maina, C. Pozo-Gonzalo, L. Kong, J. Schutz, M. Hill and L. F. Dumee, Mater. Horiz., 2017, 4, 345-361. 23 W. P. Lustig, S. Mukherjee, N. D. Rudd, A. V. Desai, J. Li and S. K. Ghosh, Chem. Soc. Rev., 2017, 46,3242-3285. 24 H. Furukawa, U. Müller and O. M. Yaghi, Angew. Chem., Int. Ed., 2015, 54, 3417-3430. 25 K. Roztocki, I. Senkovska, S. Kaskel and D. Matoga, Eur. J. Inorg. Chem., 2016, 2016, 4450-4456. 26 M. C. Das, S. Xiang, Z. Zhang and B. Chen, Angew. Chem., Int. Ed. 2011, 50, 10510-10520. 27 K. Roztocki, D. Matoga and J. Szklarzewicz, Inorg. Chem. Commun., 2015, 57, 22-25. 28 K. Roztocki, D. Matoga and W. Nitek, Inorg. Chim. Acta, 2016, 448, 86-92. 29 D. J. van Dijken, P. Kovancek, S. P. Ihrig and S. Hecht, J. Am. Chem. Soc. 2015, 137, 14982-14991. 30 C. Lu, B. Htan, C. Ma, R.-Z. Liao and Q. Gan, Eur. J. Org. Chem., 2018, 2018, 7046-7050. 31 S. Mameri, D. Specklin and R. Welter, C. R. Chim., 2015, 18, 1370-1384. 32 D. R. Richardson, D. S. Kalinowski, S. Lau, P. J. Jansson and D. B. Lovejoy, Biochim Biophys. Acta, Gen. Subj., 2009, 1790, 702-717. 33 W.-X. Ni, M. Li, X.-P. Zhou, Z. Li, X.-C. Huang and D. Li, Chem. Commun., 2007, 3479-3481. 34 W.-X. Ni, M. Li, S.-Z. Zhan, J.-Z. Hou and D. Li, Inorg. Chem., 2009, 48, 1433-1441. 35 H. Hosseini-Monfared, R. Bikas, J. Sanchiz, T. Lis, M. Siczek, J. Tucek, R. Zboril and P. Mayer, Polyhedron, 2013, 61, 45-55. I 27 36 H. Hosseini-Monfared, R. Bikas, R. Szymczak, P. Aleshkevych, A. M. Owczarzak and M. Kubicki, Polyhedron, 2013, 63, 74-82. 37 C.-L. Guo, X.-Z. Li, X.-M. Zhang, L. Wang and L.-N. Zhu, CrystEngComm, 2014, 16, 4095-4099. 38 R. Bikas, P. Aleshkevych, H. Hosseini-Monfared, J. Sanchiz, R. Szymczak and T. Lis, Dalton Trans., 2015, 44, 1782-1789. 39 R. Bikas, H. Hosseini-Monfared, M. Siczek, A. Gutiérrez, M. S. Krawczyk and T. Lis, Polyhedron, 2014, 67, 396-404. 40 D. N. Bunck and W. R. Dichtel, J. Am. Chem. Soc., 2013, 135, 14952-14955. 41 L. Stegbauer, K. Schwinghammer and B. V. Lotsch, Chem. Sci., 2014, 5, 2789-2793. 42 K. Roztocki, M. Lupa, A. Sławek, W. Makowski, I. Senkovska, S. Kaskel and D. Matoga, Inorg. Chem., 2018, 57, 3287-3296. 43 N. C. Burtch, H. Jasuja and K. S. Walton, Chem. Rev, 2014, 114, 10575-10612. 44 A. J. Howarth, Y. Liu, P. Li, Z. Li, T. C. Wang, J. T. Hupp and O. K. Farha, Nat. Rev. Mater, 2016, 1, 15018. 45 J. J. Low, A. I. Benin, P. Jakubczak, J. F. Abrahamian, S. A. Faheem and R. R. Willis, J. Am. Chem. Soc., 2009, 131, 15834-15842. 46 J. A. Greathouse and M. D. Allendorf, J. Am. Chem. Soc., 2006, 128, 10678-10679. 47 P. M. Schoenecker, C. G. Carson, H. Jasuja, C. J. J. Flemming and K. S. Walton, Ind. Eng. Chem. Res., 2012, 51, 6513-6519. 48 S. K. Henninger, F. Jeremias, H. Kummer and C. Janiak, Eur. J. Inorg., 2012, 2012, 2625-2634. 49 V. Bon, N. Kavoosi, I. Senkovska and S. Kaskel, ACS Appl. Mater. Interfaces, 2015, 7, 22292- 22300. 50 T. F. Willems, C. H. Rycroft, M. Kazi, J. C. Meza and M. Haranczyk, Micropor. Mesopor. Mat., 2012, 149, 134-141. 51 T. Friscić, S. L. Childs, S. A. A. Rizvi and W. Jones, CrystEngComm, 2009, 11, 418-426. I 28 7. Załączniki I 29 Carboxylate-Hydrazone Mixed-Linker Metal- Organic Frameworks: Synthesis, Structure, and Selective Gas Adsorption Autorzy: Kornel Roztocki, Irena Senkovska, Stefan Kaskel, Dariusz Matoga Opublikowano w: Eur. J. Inorg. Chem. 2016, 4450-4456 Publikacja I I 30 DOI: 10.1002/ejic.201600134 I Porous MOFs Carboxylate-Hydrazone Mixed-Linker Metal-Organic Frameworks: Synthesis, Structure, and Selective Gas Adsorption Kornel Roztocki,[al Irena Senkovska,Ibl Stefan Kaskel,[bl and Dariusz Matoga*tal Abstract: New mixed-linker metal-organic framework (MOF) materials incorporating both carboxylate and hydrazone linkers were prepared. The zinc-based 3D MOFs were obtained by util- izing a presynthesized a roy I hydrazone [4-pyridinecarbaldehyde isonicotinoyl hydrazone (PCIH)] and para-dicarboxylic acids [1,4- benzenedicarboxylic acid (H2BDC) and 4,4'-biphenyldicarbox- ylic acid (H2BPDC)]. Single-crystal X-ray diffraction revealed the interpenetrated, pillar-layered structures of the MOFs, including layers formed by dicarboxylate-bridged Zn2 nodes as well as hydrazone pillars. The microporous [Zn2(dcx)2(hdz)2]*guests frameworks (dcx = linear dicarboxylate dianion; hdz = hydraz- one) were found to adsorb C02 selectively over N2 upon ther- mal activation. The frameworks represent rare MOFs of the type [Zn2(dcx)2(pil)2] (pil = neutral pillar), in which two pillars at each metallic site of the Zn2 node connect adjacent layers. Introduction Throughout approximately 20 years of intensive research on metal-organic frameworks (MOFs), two general strategies for their construction are well established: (1) de novo approach involving self-assembly of non-polymeric reactants with the op- tional use of templates,41-41 (2) postsynthetic modification of a presynthesized framework involving heterogeneous chemical reactions within nodes or linkers,t4-7] or even the replacement of entire building blocks.181 In spite of fundamental scientific interest, exploration of these strategies often aims at introduc- ing or improving various functionalities in the synthesized frameworks. Desirable MOF properties with associated applica- tions (e.g., gas sorption, catalysis, luminescence, photoactivity, electronic and ionic conductivity, etc.) often arise from the pres- ence of specific metal centers within the nodes, organic linkers and their functional groups, as well as guests in the pores/cavi- ties.191 In particular, combining various functionalities within a single metal-organic framework, recently reviewed by Fu- rukawa et al.,[101 may lead to highly sought multifunctional ma- terials. From the point of view of the linker, introducing "hetero- geneity within order" in MOFs can be realized by utilizing two or more organic ligands for the synthesis of a MOF and/or by mixing functional groups along the backbone of the MOF through two or more organic linkers of the same type, each having unique functionality.1101 In this context, we became in- terested in the synthesis of mixed-linker frameworks incorporat- ing aroylhydrazones, that is, a subclass of acylhydrazones that are organic compounds bearing C=0 and N-H functional groups, potentially transferrable onto pore walls. Aroylhydrazones are obtainable through hydrazide-ketone/ aldehyde condensation, and they exhibit flexible metal-chelat- ing capabilities through their keto-enol tautomerism and possi- ble reversible deprotonation (Scheme 1). The empty N,O-donor Scheme 1. General formulae: (a) aromatic acid hydrazide, (b) aroylhydrazone in its keto form, (c) aroylhydrazone in its enol form, and (d) aroylhydrazone in its deprotonated enolate form. AR: aryl group (including substituents, heteroatoms, fused rings); R1, R2: independent alkyl or aryl groups or together cycloalkyl or aryl group. [a] Faculty of Chemistry, Jagiellonian University Ingardena 3, 30-060 Kraków, Poland E-mail: dariusz. matoga@uj. edu.pl http-y/www2.chemia.uj.edu.pl/~matoga [b] Department of Inorganic Chemistry, Technische Universität Dresden Bergstrasse 66, 01062 Dresden, Germany S Supporting information for this article is available on the WWW under h ttpj/dx. doi.org/10.1002/ejic.201600134. chelating "pockets" of aroylhydrazones that are incorporated into frameworks can potentially make them amenable to post- synthetic metalation.171 On the other hand, previously reported aroylhydrazone- based chelate complexes may also be utilized as building blocks (metalloligands) in a stepwise approach towards mixed-metal Eur. J. Inorg. Chem. 2016, 4450-4456 Wiley Online Library © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4450 I 31 organic frameworks.111,12] Finally, the propensity of acylhydra- zones to undergo hydrolysis to carboxylic acids in situ in the presence of metal ions may also be used to advantage for the construction of MOFs.[13] A large number of discrete metal com- plexes or 1D coordination polymers involving aroylhydrazones have been reported, some of which were included in recent thematic reviews on, for example, hydrazone-based switches,1141 bis-aroylhydrazone ligands,1151 selected dinuclear complexes,1161 and iron chelators for cancer treatment.1171 Very recently, a one- dimensional coordination polymer of copper(ll) with both acet- ate and tridentate aroylhydrazone bridges was published.1181 However, to the best of our knowledge, only a few aroyl- hydrazone-bridged coordination polymers or MOFs with higher 2D or 3D dimensionalities have been reported.119-291 Similarly, aroylhydrazones were only very recently successfully implemented as linkers in 2D covalent organic frameworks (COFs).130"321 Herein, we report new mixed-linker MOFs incorporating both carboxylate and hydrazone linkers. Two zinc(ll)-based 3D MOFs were obtained by utilizing a presynthesized aroylhydrazone [4- pyridinecarbaldehyde isonicotinoyl hydrazone (PCIH)] and para- dicarboxylic acids [1,4-benzenedicarboxylic acid (H2BDC) or 4,4/-biphenyldicarboxylic acid (H2BPDC)]. Single-crystal X-ray diffraction revealed the interpenetrated, pillar-layered struc- tures of the products, including layers formed by dicarboxylate- bridged Zn2 nodes and hydrazones acting as pillars. The [Zn2(dcx)2(hdz)2]*guests (dcx = linear dicarboxylate dianion; hdz = hydrazone) microporous frameworks were found to ad- sorb C02 selectively over N2 upon thermal activation. These frameworks represent very rare MOFs[33-351 of the type [Zn2(dcx)2(pil)2] (pil = neutral pillar), in which two pillars at each metallic site of the Zn2 node connect dicarboxylate-bridged lay- ers, unlike the well-known family of de novo obtained MOFs, [Zn2(dcx)2(pil)], with a single neutral pillar per dinuclear node with either a paddle-wheel135-551 or non-paddle-wheel frame- work.t56'571 Results and Discussion General Remarks Pale-yellow mixed-linker zinc(ll) MOFs of the type {[Zn2(dcx)2- (hdz)2]-guests}n (1 and 2) were prepared by heating N,N-6\- methylformamide (DMF)/H20 solutions of Zn(N03)2*6H20, the corresponding para-dicarboxylic acid (H2dcx = H2BDC or H2BPDC), and 4-pyridinecarbaldehyde isonicotinoyl hydrazone (hdz = PCIH) in sealed vials (Scheme 2). In both frameworks, the in situ deprotonated dicarboxylic acids act as linkers that compensate the positive charge of the metal atoms and that form zinc-carboxylate layers. These layers are pillared by neu- tral PCIH, as confirmed by single-crystal X-ray diffraction (XRD). The neutral hydrazone in its keto form bears both the N-H and C=0 functional groups, which are proton-donating and proton- accepting sites, respectively. To the best of our knowledge, frameworks 1 and 2 represent the first zinc-based coordination polymers incorporating PCIH (in either neutral or deprotonated form). In the literature, this hydrazone has only been shown to form mixed-valence homometallic copper(l/ll) coordination polymers119,201 as well as 1D or 2D coordination polymers based on Mn(NCS)2 units.1271 Scheme 2. Synthetic route to {[Zr^dcxyhdz^hguests},, MOFs based on (a) 1,4-benzenedicarboxylic acid (H2BDC) and (b) biphenyl-4,4'-dicarboxylic acid (H2BPDC). Crystal Structures Single-crystal X-ray diffraction revealed that 1 crystallizes in the orthorhombic system, space group Ibca, with one zinc(ll) ion, one terephthalate ligand, and one coordinating PCIH molecule in the asymmetric unit. The overall coordination geometry of the zinc ion in 1 can be described as a distorted pentagonal bipyramid consisting of three equatorial 0 carboxylate donors from three different BDC2- ligands and two N pyridyl atoms from two PCIH ligands occupying axial positions (Figure 1). All of the Zn-0 bonds (Znl-028 1.990, Znl-032 1.992, Znl-031 2.001 Â) as well as the Zn-N bonds (Zn1-N1 2.224, Zn1-N15 2.224 Â) have almost identical lengths. This indicates axial elon- gation of the zinc coordination polyhedra (Figure 1, see also Table S1 in the Supporting Information). The benzenedicarboxylate ions in 1 function as |i3 linkers between the carboxylate-bridged Zn2 clusters and as linkers connecting dinuclear secondary building unit (SBU) nodes form the layers of a (4,4) topology lying in the ab plane. The axial sites of the nodes are occupied by |j2-PCIH hydrazone pillars that extend the layers along the c axis into a 3D pillar-layered framework (Figure 1). The framework voids are occupied by H20 and DMF guests as well as by atoms of a neighboring network, which thus leads to a doubly interpenetrated structure. This twofold interpenetration is enabled by strong N-H—O hydrogen bonds [d(N9***028) = 2.804 Â, Z = 158°] between the uncoordinated oxygen atoms of the BDC2- linkers and the hydrogen atoms of the NH groups located on the PCIH pillars in an adjacent network (Figure 1). In spite of the interpenetra- tion, the structure contains one-dimensional channels running along the [100] direction, potentially accessible to guest mol- ecules (Figure 2). Replacement of BDC2- in 1 with longer para-dicarboxylate (BPDC2-) in 2 leads to a few significant changes in the structure, Eur. J. Inorg. Chem. 2016, 4450-4456 www.eurjic.org © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4451 I 32 which thus changes its coordination geometry to a distorted octahedron (Figure 3). This involvement in hydrogen bonding leads to elongation of the Znl-012 bond (2.398 Â) relative to the rest of the Zn-0 bonds (Zn1-011 2.103, Zn1-021 2.011, Znl-022 2.023 Â) (Table S3). The axial Zn-N bonds for the octa- hedral metal centers of 2 are slightly shorter (Zn1-N41 2.177, Zn1-N55 2.139 Â) than the corresponding bonds in framework 1 with pentacoordinate zinc centers. The structure contains one-dimensional channels running along the [001] direction (Figure 2). Figure 1. X-ray crystal structure of {[Zn2(BDC)2(PCIH)2]*2H20*DMF}n (1). (a) [Zn2(BDC)2(PCIH)2] unit with 30 % displacement ellipsoids, (b) Coordina- tion environment within the Zn2 cluster with atom-labeling scheme and 30 % displacement ellipsoids, (c) Twofold network interpenetration (space-filled representation along the a axis); guest molecules are omitted, (d) Schematic representation of hydrogen bonds between the two interpenetrated net- works. The H atoms in panels (a)-(c) are omitted for clarity. Figure 2. Solvent-accessible voids (brown regions) in the X-ray crystal struc- tures of 1 (left) and 2 (right) calculated by using Mercury software by using a probe molecule with a diameter of 2.4 Â. The unit cell of 1 is not indicated as it is larger than the shown structural part. however, with preservation of a 3D doubly pillar-layered net- work. The main structural changes include an increase in the number of interpenetrating networks in 2 to three as well as a change in the dicarboxylate coordination mode within the Zn2 carboxylate layers (Figure 3). The threefold interpenetration in 2 originates from strong N-H—O hydrogen bonds [c/(N49—012) = 2.889 Â, Z = 169°] between the hydrazone NH groups and the carboxylate oxygen atom of an adjacent network, similarly to that observed in the structure of 1. However, unlike in 1, the carboxylate proton acceptor 012 is coordinated to a zinc center, Figure 3. X-ray crystal structure of {[Zn2(BPDC)2(PCIH)2]-H202DMF}n (2). (a) [Zn2(BPDC)2(PCIH)2] unit with 50 % displacement ellipsoids, (b) Coordina- tion environment within the Zn2 cluster with atom-labeling scheme and 50 % displacement ellipsoids, (c) Threefold network interpenetration (space-filled representation); guest molecules are omitted, (d) Schematic representation of hydrogen bonds between the three interpenetrated networks. H atoms are omitted for clarity. Guest Removal Thermogravimetric analysis (TGA) of 1 and 2 revealed a step- wise weight loss (Figure S3) with an approximate plateau in the range of 200-300 °C upon heating. The first step corresponds to the loss of two H20 molecules and one DMF molecule per formula unit of 1 (calcd. weight loss 10.5 %; found 11.8 %) and one H20 molecule and two DMF molecules per formula unit of 2 (calcd. weight loss 13.3 %; found 12.5 %). The next observed step with DTG minima at approximately 330 °C for 1 and 340 °C for 2 is associated with decomposition of the frameworks. Guest removal experiments prior to ex situ IR spectroscopy and pow- der XRD measurements were performed in the temperature re- gion for which the TGA data (Figure S3) indicated that only water and DMF desorption occur. Eur. J. Inorg. Chem. 2016, 4450-4456 www.eurjic.org 4452 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim I 33 The IR spectra of the as-synthesized MOFs exhibit several characteristic absorption bands of 4-pyridinecarbaldehyde iso- nicotinoyl hydrazone, the corresponding dicarboxylate anion, and DMF (Figure 4).[581 The appearance of strong bands at 1580 and 1392 cm-1 for 1 and at 1581 and 1405 cm-1 for 2 can be ascribed to the v(COO)as and v(COO)s vibrations, respectively, and confirms the presence of carboxylate groups. The amide N-H and C=0 bands of PCIH are observed at 3222 (m) and 1680 cm-1 (s) for 1 and at 3217 (m) and 1694 cm-1 (s) for 2. In the free PCIH ligand, these bands are found at 3448 and 1688 cm'1, respectively.1191 The considerable redshift of the N- H bands in the spectra of the MOFs can be explained by the formation of hydrogen bonds between the N-H groups and the dicarboxylate oxygen atoms. The amide C=0 band of the PCIH ligand in both frameworks, present during thermal activation and resolvation, is indicated in Figure 3 as v(CO-N). The appear- ance of an additional strong band in the characteristic range for carbonyl groups (at 1661 cm-1 for 1 and 1679 cm-1 for 2), which disappears after thermal activation of the frameworks at 200 °C, confirms the presence of guest DMF molecules in both MOFs (Figure 4). bonds with the DMF and H20 guest molecules. This is in agree- ment with the X-ray crystal structures of 1 and 2, which show that hydrogen bonds exist between the interpenetrating sub- networks only. The complimentary powder XRD patters re- corded for the as-synthesized, heated, and resolvated materials show retention of the crystallinity and the unit-cell parameters for both MOFs. The aforementioned experiments indicate that both frameworks 1 and 2 are robust under the studied condi- tions and that removal of the guest molecules is reversible with retention of the framework structures. Gas Adsorption (C02 vs. N2) Sorption analysis revealed that both activated frameworks (after removal of the guest molecules) selectively adsorb C02 over N2 (Figure 5). The very low N2 uptake observed for activated 1 and 2 is in strong contrast to a family of singly pillared paddle- wheel frameworks of the type [Zn2(dcx)2(pil)], many of which show considerable N2 uptake, sometimes even despite frame- work interpénétrations.135-441 On the other hand, one analogue of our frameworks belonging to a family of doubly pillared [Zn2(dcx)2(pil)2] MOFs [with dcx = benzenedicarboxylate and pil = 1,4-bis(4-pyridylethynyl)benzene] was reported not to ad- sorb nitrogen gas.[3S1 The authors mentioned no accessible voids in its structure as the explanation for this behavior, and the structure of the as-synthesized compound also did not con- tain guests. The sorption behavior for other analogues was not discussed.133,341 The selectivity of C02 over N2 among pillar-lay- Figure 4. (a and c) IR spectra and (b and d) powder XRD patterns of {[Zn2(BDC)2(PCIH)2]-2H2O.DMF}n (1) (left) and {[Zn2(BPDC)2(PCIH)2]-H20- 2DMF}„ (2) (right) under various conditions: calculated from the single-crystal structures (calc), as synthesized (as), dried at 200 °C and 10.0 kPa for 1 h (a), and resolvated at room temperature in DMF/H20 solution for 1 h (re). For enlarged views of the IR spectra see Figures S1 and S2. To explore the behavior of the frameworks upon desorption of the guests, a series of guest removal and resolvation experi- ments as monitored by ex situ IR spectroscopy and powder XRD measurements were performed (Figure 4). Analysis of the MOFs by IR spectroscopy revealed that the vibrations of the skeletal atoms remained intact during the aforementioned process, which confirmed retention of the framework structures. In this cycle, disappearance and reappearance of bands attributed to the DMF molecules (and corresponding to C=0 stretching vi- brations) were observed. Additionally, the lack of distinct changes occurring within the v(COO) region may be explained in terms of carboxylate non-involvement in the hydrogen Figure 5. C02 and N2 adsorption (solid symbols) and desorption (open sym- bols) isotherms for activated (a) {[Zn2(BDC)2(PCIH)2>2H2ODMF}„ (1) and (b) {[Zn2(BPDC)2(PCIH)2].H20-2DMF}n (2) recorded at 195 K (C02) and 77 K (N2). Eur. J. Inorg. Chem. 2016, 4450-4456 www.eurjic.org © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4453 I 34 ered frameworks was reported by Fischer et al. for a series of [Zn2(dcx)2(pil)] MOFs with alkoxy-functionalized dicarboxyl- ates.[50,52,531 The authors pointed out two general factors that influence such behavior: polar pore surface and narrow pores, which may change dynamically. In our case, both as-synthetized MOFs 1 and 2 have potential accessible voids (Figure 2) after removal of the solvent mol- ecules. The geometrical surface areas calculated by using Pore- blazer software1591 amount to 408 m2 g-1 for 1 and 383 m2 g-1 for 2. The apparent BET surface area calculated from the C02 adsorption isotherm of 2 measured at 195 K (Figure 5b) was 293 m2 g-1. Calculation of the surface area for compound 1 was hindered owing to the unusual slope of the C02 adsorption branch (Figure 5a); therefore, the desorption branch was used for the calculations. In this case, the apparent BET surface area amounted to 203 m2 g-1. The pore volumes calculated at a rela- tive pressure of 0.99 were 0.07 and 0.15 cm3 g-1 for 1 and 2, respectively. Comparison of the powder XRD patterns of the as- synthesized and activated compounds points to no pronounced flexibility of the frameworks. Therefore, the selective adsorption of C02 over N2 can be explained by a molecular-sieving effect. Indeed, according to Zeo++ calculations/601 the pore window size and maximum pore diameter are 3.14 and 5.07  for 1 and 3.52 and 4.96  for 2, respectively. The real size of the pore windows can vary slightly owing to rotational freedom of the phenyl rings of the linkers. For geometrical calculations, only one position could be taken into account. The kinetic diameters of C02 and N2 are 3.30 and 3.64 Â, respectively. Given that the size of a nitrogen molecule (3.64 Â) is much larger than the pore window size of 1 (3.14 Â), no nitrogen uptake is observed at 77 K. Smaller C02 molecules (3.30 Â) can clearly enter the pores of 1 at 195 K, but limiting diffusion and poor equilibration during adsorption is confirmed by the slope of the adsorption curve and broad hysteresis between the adsorption and de- sorption branches. The C02 desorption branch follows a type I isotherm. A similar situation is observed for adsorption of nitro- gen at 77 K on 2, for which the pore window size (3.52 Â) and adsorptive size (3.64 À) are close to each other. In contrast, the C02 adsorption isotherm at 195 K is of type I, which points to the accessibility of the framework for this gas. The functionalization of frameworks with high-affinity ad- sorption sites such as unsaturated metal centers or Lewis basic sites has been reported to be an important strategy to improve the selective adsorption of C02.[61"651 Here, both MOFs 1 and 2 have basic NH groups on their pore walls and are thus potential candidates for C02/N2 separation processes. However, separa- tion of these gases purely on the basis of pore size is challeng- ing because of similar kinetic diameters, and only a couple of MOFs are reported to show such separation behavior.[66_69] Conclusion We synthesized new mixed-linker MOFs incorporating both hydrazone and carboxylate linkers. The two microporous frame- works belong to a very rare group of doubly pillar-layered 3D MOFs of the type [Zn2(dcx)2(pil)2]. Functional groups introduced with the linkers (carboxylate and amide) are responsible for their doubly and triply interpenetrated structures. The com- pounds show the ability to adsorb C02 selectively over N2 ow- ing to small pores and framework functionality. Assuming that these frameworks or their derivatives may be obtained as non- interpenetrated analogues, this should open the way to post- synthetic modifications by utilizing the chelating pockets of the hydrazone linkers. Further work on carboxylate-hydrazone mixed-linker MOFs is currently underway. Experimental Section Materials and Methods: 4-Pyridinecarbaldehyde isonicotinoyl hydrazone (PCIH) was prepared according to a published method.1191 All other reagents and solvents were of analytical grade (Sigma-Aldrich, POCH, Polmos) and were used without further puri- fication. Carbon, hydrogen, and nitrogen contents were determined by conventional microanalysis by using an Elementar Vario MICRO Cube elemental analyzer. IR spectra were recorded with a Thermo Scientific Nicolet iS5 FTIR spectrophotometer equipped with an iD5 diamond ATR attachment. Thermogravimetric analysis (TGA) was performed with a Mettler-Toledo TGA/SDTA 85Ie instrument at a heating rate of 5 °C min-1 in the temperature range of 25 to 600 °C (approximate sample weight 15 mg). The measurements were per- formed at atmospheric pressure under flowing argon. Powder X-ray diffraction (XRD) patterns were recorded at room temperature (295 K) with a Rigaku Miniflex 600 diffractometer with Cu-Ka radia- tion (A = 1.5418 Â) in the 26 range from 3 to 50° with a 0.05° step at a scan speed of 2 ° min-1. Gas adsorption studies were performed with a BELSORP-max adsorption apparatus (MicrotracBEL Corp.); 77 K was achieved by a liquid nitrogen bath, and 195 K was achieved by a dry ice/2-propanol bath. Prior to the adsorption ex- periments the samples were evacuated at 260 °C for 16 h. Synthesis of {[Zn2(BDC)2(PCIH)2]-2H20-DMF}n (1): PCIH (45.2 mg, 0.200 mmol), Zn(N03)2-6H20 (59.5 mg, 0.200 mmol), and H2BDC (33.0 mg, 0.200 mmol) were dissolved in DMF (16.2 mL) and H20 (1.8 mL) by sonication (60 s), and the mixture was heated at 50 °C for 72 h. Pale-yellow crystals of 1 were filtered off, washed with DMF, and dried in an oven at 60 °C and 50.0 kPa for 30 min. Yield: 49 mg (48 %). FTIR (ATR): v = 1580 (s) [v(COO)as], 1392 (s) [v(COO)s], 1661 (s) [v(C=0)DMF], 1680 (s) [v(C=0)PCIH], 3222 cm-1 (m) [v(NH)]. C43H39N9013Zn2 (1020.6): calcd. C 50.60, H 3.85, N 12.35; found C 49.80, H 3.91, N 12.58. Synthesis of {[Zn2(BPDC)2(PCIH)2]-H20-2DMF}„ (2): PCIH (158 mg; 0.700 mmol), H2BPDC (145 mg; 0.600 mmol), and Zn(N03)2*6H20 (178.5 mg; 0.600 mmol) were dissolved in DMF (86 mL) and water (7 mL). The mixture was heated at 80 °C for 144 h to yield a fine- crystalline yellow product. The pale-yellow product 2 was washed with DMF and dried in a vacuum oven (30 min, 60 °C, 50.0 kPa). Yield: 150 mg (21 %). FTIR (ATR): v = 1581 (s) [v(COO)as], 1405 (s) [v(COO)s], 1679 (s) [v(C=0)DMF], 1694 (s) [v(C=0)PCIH], 3217 cm-1 (m) [v(NH)]. C58H52N10013Zn2 (1227.9): calcd. N 11.41, C 56.73, H 4.27; found N 11.59, C 57.12, H 4.01. Crystallographic Data Collection and Structure Refinement: Sin- gle crystals of 1 and 2 suitable for X-ray analysis were selected from bulk materials prepared as described above. Intensity data for both compounds were collected with an Oxford Diffraction SuperNova four-circle diffractometer, equipped with a Mo-Ku (0.71069 Â) radia- tion source, a graphite monochromator, and Oxford CryoJet system for measurements at low temperatures. Measurements for 1 were performed under ambient conditions (298 K) and for 2 at low tem- perature (120 K). The positions of all non-hydrogen atoms were Eur. J. Inorg. Chem. 2016, 4450-4456 www.eurjic.org © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4454 I 35 Table 1. Crystal data and structure refinement parameters for frameworks 1 and 2. Compound 1 2 Empirical formula Zn2C43H28N90]2 ZnC29H25N506 Formula mass 993.48 604.91 Crystal size [mm] 0.357x0.262x0.215 0.440x0.393x0.107 Crystal system orthorhombic triclinic Space group Ibca PÎ 0 [Ä] 15.1123(3) 9.003(5) b [Â] 9.9069(3) 12.118(5) c[A] 31.2591(6) 12.865(5) '« n 90 98.740(5) ßn 90 90.251(5) r n 90 93.325(5) V[k3] 9404.0(3) 1384.8(11) T[ K] 298(2) 120(2) Z 8 2 Dca led. [g Cm-3] 1.403 1.451 fj [mm-1] 1.082 0.939 Reflections measured 61329 19255 Reflections unique [fl(int)] 5813 (0.0539) 6446 (0.0342) Reflections observed [/ > 2cr(/)] 4390 5549 ft, [/ > 2o(Dl 0.0724 0.0499 wR2 [/ > 2o(l)] 0.1918 0.1111 determined by direct methods by using SIR-97.I70] Alt non-hydrogen atoms were refined anisotropically by using weighted full-matrix least squares on F2. Refinement and further calculations were per- formed by using SHELXL-97.1711 All hydrogen atoms joined to carbon atoms were positioned with an idealized geometry and were refined by using a riding model with tVj50(H} fixed at 1.2 L/eq of C (for 1) and with (7iso(H) fixed at 1.2 Ueq of C and 1.5 Ueq for methyl groups {for 2). Hydrogen atoms joined to nitrogen atoms were found from the difference Fourier map and were refined with- out restraints. The hydrogen atoms of water and DMF molecules for 1 could not be located. Table 1 contains the crystal data and struc- ture refinement parameters for frameworks 1 and 2. CCDC 1008301 (for 1) and 1433069 (for 2) contain the supplementary crystallo- graphy data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre. Acknowledgments The National Science Centre in Poland (grant no. 2012/07/B/ ST5/00904) is gratefully acknowledged for financial support of this research. The research was performed partially with the equipment purchased as a result of the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (contract no. POIG.02.01.00-12-023/08). We thank Dr. W. Nitek (Jagiellonian University) for assistance with single-crystal XRD measurements and structure refinements. Keywords: Adsorption • Carboxylate ligands • Crystal engineering ■ Hydrazones - Metal-organic frameworks • Zinc [1] S. Kitagawa, R. Kitaura, S.-l. Noro, Angew. Chem. Int. Ed. 2004, 43, 2334- 2375; Angew. Chem, 2004, 116, 2388-2430. [2] C. Dey, T. Kundu, B. P. Biswal, A. Malliek, R. Banerjee, Acta Crystallogr., Sect. B 2014, 70, 3-10. [3] Z. Zhang, M. J. Zaworotko, Chem. Soc. Rev. 2014, 43, 5444-5455. [4] Y. Sun, H.-C. Zhou, Sei. Technol. Adv. Mater. 2015, 16, 054202-054212. E5] S. M. Cohen, Chem. Rev. 2012, 112, 970-1000. [6] C. K. Brożek, M. Dinca, Chem. Soc. Rev. 2014, 43, 5456-5467. [7] J. D. Evans, C. J. Sumby, C. J. Doonan, Chem. Soc Rev. 2014, 43, 5933- 5951. [8] P. Deria, J. E. Mondloch, O. Karagiaridi, W. Bury, J. T. Hupp, O. K. Farha, Chem. Soc. Rev. 2014, 43, 5896-5912. [9] M. D. Allendorf, V. Stavila, CrystEngComm 2015, 77, 229-246. [10] H. Furukawa, U. Müller, O. Yaghi, Angew. Chem. Int. Ed. 2015, 54, 3417- 3430; Angew. Chem. 2015, /27, 3480-3494 . [11] M. C. Das, S. Xiang, Z. Zhang, B. Chen, Angew. Chem. Int. Ed. 2011, 50, 10510-10520; Angew. Chem. 2011, 123, 10696-10707. [12] S. Kitagawa, S.-L Noro, T. Nakamura, Chem. Commun. 2006, 701-707. [13] D. Matoga, B. Gil, W. Nitek, A. D. Todd, C. W. Bielawski, CrystEngComm 2014, 16, 4959-4962. [14] X. Su, I. Aprahamian, Chem. Soc. Rev. 2014, 43, 1963-1981. [15] A.-M. Stadler, J. Harrowfield, Inorg. Chim. Acta 2009, 362, 4298-4314. [16] S. Mameri, D. Specklin, R. Welter, C. R. Chim. 2015, 18, 1370-1384. [17] D. R. Richardson, D. S. Kalinowski, S. Lau, P. J. Jansson, D. B. Lovejoy, Biochim. Biophys. Acta, Gen. Sub}. 2009, 1790, 702-717. [18] R. Bikas, P. Aleshkevych, H. Hosseini-Monfared, J. Sanchiz, R. Szymczak, T. Us, Dalton Trans. 2015, 44, 1782-1789. [19] W.-X. Ni, M. Li, X.-P. Zhou, Z. Li, X.-C. Huang, D. Li, Chem. Commun. 2007, 3479-3481. [20] W.-X. Ni, M. Li, S.-Z. Zhan, J.-Z. Hou, D. Li, Inorg. Chem. 2009, 48, 1433- 1441. [21] H. Hosseini-Monfared, R. Bikas, J. Sanchiz, T. Lis, M. Siczek, J. Tucek, R. Zboril, P. Mayer, Polyhedron 2013, 61, 45-55. [22] H. Hosseini-Monfared, R. Bikas, R. Szymczak, P. Aleshkevych, A. M. Owczarzak, M. Kubicki, Polyhedron 2013, 63, 74-82. [23] C.-L. Guo, X.-Z. Li, X.-M. Zhang, L. Wang, L.-N. Zhu, CrystEngComm 2014, 16, 4095-4099. [24] R. Bikas, H, Hosseini-Monfared, M. Siczek, A. Gutiérrez, M. S, Krawczyk, T. Lis, Polyhedron 2014, 67, 396-404. [25] R. Bikas, H. Hosseini-Monfared, V. Vasylyeva, J. Sanchiz, J. Alonso, J. M. Barandiaran, C. Janiak, Dalton Trans. 2014, 43, 11925-11935. [26] K. Roztocki, D. Matoga, J. Szklarzewicz, Inorg. Chem. Commun. 2015, 57, 22-25. [27] M. Servati-Gargari, G. Mahmoudi, S. R. Batten, V. Stilinovic, D. Butler, L. Beauvais, W. S. Kassel, W. G. Dougherty, D. VanDeerver, Cryst. Growth Des. 2015, 15, 1336-1343. [28] D. Specklin, S. Mameri, E. Loukopoulos, G. E. Kostakis, R. Welter, Polyhe- dron 2015, 100, 359-372. [29] Y.-F. Liu, Y.-P. Liu, K.-K. Zhang, Q.-L, Ren, J, Qin, Acta Crystallogr., Sect. C 2015, 71, 116-121. Eur. J. Inorg. Chem. 2016, 4450-4456 www.eurjic.org 4455 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim I 36 [30] F. J. Uribe-Romo, C. J. Doonan, H. Furukawa, K. Oisaki, O. M. Yaghi, J. Am. Chem. Soc. 2011, 133, 11478-11481. [31] D. N. Bunck, W. R. Dichtei, J. Am. Chem. Soc. 2013, 135, 14952-14955. [32] L. Stegbauer, K. Schwinghammer, B. V. Lotsch, Chem. Sei. 2014, 5, 2789- 2793. [33] J. Tao, M.-L. Tong, X.-M. Chen, J. Chem. Soc., Dalton Trans. 2000, 3669- 3674. [34] T. Yamada, H. Kitagawa, J. Am. Chem. Soc. 2009, 131, 6312-6313. [35] T. Yamada, S. Iwakiri, T. Hara, K. Kanaizuka, M. Kurmoo, H. Kitagawa, Cryst. Growth Des. 2011, 11, 1798-1806. [36] D. N. Dybtsev, H. Chun, K. Kim, Angew. Chem. Int. Ed. 2004, 43, 5033- 5036; Angew. Chem. 2004, 116, 5143-5146. [37] H. Chun, D. N. Dybtsev, H. Kim, K. Kim, Chem. Eur. J. 2005, 11, 3521- 3529. [38] B.-Q. Ma, K. L. Mülfort, J. Hupp, Inorg. Chem. 2005, 44, 4912-4914. [39] B. Chen, C. Liang, J. Yang, D. S. Contreras, Y. L. Clancy, E. B. Lobkovsky, O. M. Yaghi, S. Dai, Angew. Chem. Int. Ed. 2006, 45, 1390-1393; Angew. Chem. 2006, 118, 1418-1421. [40] S.-H. Cho, B. Ma, S. T. Nguyen, J. T. Hupp, T. E. Albrecht-Schmitt, Chem. Commun. 2006, 2563-2565. [41] B. Chen, S. Ma, F. Zapata, E. B. Lobkovsky, J. Yang, Inorg. Chem. 2006, 45, 5718-5720. [42] M. Xue, S. Ma, Z. Jin, R. M. Schaffino, G.-S. Zhu, E. B. Lobkovsky, S.-L. Qiu, B. Chen, Inorg. Chem. 2008, 47, 6825-6828. [43] Z. Wang, K. K. Tanabe, S. M. Cohen, Inorg. Chem. 2009, 48, 296-306. [44] P. V. Dau, M. Kim, S. J. Garibay, F. H. L. Münch, C. E. Moore, S. M. Cohen, Inorg. Chem. 2012, 51, 5671-5676. [45] J. Y. Lee, D. H. Olson, L. Pan, T. J. Emge, J. Li, Adv. Funct. Mater. 2007, 17, 1255-1262. [46] Y. Takashima, V. M. Martinez, S. Furukawa, M. Kondo, S. Shimomura, H. Uehara, M. Nakahama, K. Sugimoto, S. Kitagawa, Nat. Commun. 2011, 2, 168-175. [47] l.-H. Park, K. Kim, S. S. Lee, J. J. Vittal, Cryst. Growth Des. 2012, 12, 3397- 3401. [48] R. S. Forgan, R. J. Marshall, M. Struckman, A. B. Bleine, D.-L. Long, M. C. Bernini, D. Fairen-Jimenez, CrystEngComm 2015, 17, 299-306. [49] Y.-P. Diao, K. Li, S.-S. Huang, X.-H. Shu, K.-X. Liu, X.-M. Deng, Acta Crystal- logr., Sect. C 2009, 65, 82-85. [50] S. Henke, R. Schmid, J.-D. Grunwaldt, R. A. Fischer, Chem. Eur. J. 2010, 16, 14296-14306. [51] Q. Gao, Y.-B. Xie, J.-R. Li, D.-Q. Yuan, A. A. Yakovenko, J.-H. Sun, H.-C. Zhou, Cryst. Growth Des. 2012, 12, 281-288. [52] S. Henke, A. Schneemann, S. Kapoor, R. Winter, R. A. Fischer, J. Mater. Chem. 2012, 22, 909-918. [53] S. Henke, A. Schneemann, A. Wütscher, R. A. Fischer, J. Am. Chem. Soc. 2012, 134, 9464-9474. [54] K. Zhu, V. N. Vukotic, C. A. O'Keefe, R. W. Schurko, S. J. Loeb, J. Am. Chem. Soc. 2014, 136, 7403-7409. [55] V. Safarifard, A. Morsali, CrystEngComm 2014, 16, 8660-8663. [56] A. Lan, K. Li, H. Wu, D. H. Olson, T. J. Emge, W. Ki, M. Hong, J. Li, Angew. Chem. Int. Ed. 2009, 48, 2334-2338; Angew. Chem. 2009, 121, 2370-2374. [57] S. Pramanik, Z. Hu, X. Zhang, C. Zheng, S. Kelly, J. Li, Chem. Eur. J. 2013, 19, 15964-15971. [58] G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and Charts, 3rd ed., John Wiley & Sons, Chichester, UK, 2001. [59] L. Sarkisov, A. Harrison, Mol. Simul. 2011, 37, 1248-1257. [60] T. F. Willems, C. H. Rycroft, M. Kazi, J. C. Meza, M. Haranczyk, Microporous Mesoporous Mater. 2012, 149, 134-141. [61] Z. Zhang, Y. Zhao, Q. Gong, Z. Li, J. Li, Chem. Commun. 2013, 49, 653- 661. [62] K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T.-H. Bae, J. R. Long, Chem. Rev. 2012, 112, 724-781. [63] J. Liu, P. K. Thallapally, B. P. McGrail, D. R. Brown, J. Liu, Chem. Soc. Rev. 2012, 41, 2308-2322. [64] J.-R. Li, Y. Ma, M. C. McCarthy, J. Sculley, J. Yu, H.-K. Jeong, P. B. Balbuena, H.-C. Zhou, Coord. Chem. Rev. 2011, 255, 1791-1823. [65] D. M. D'Alessandro, B. Smit, J. R. Long, Angew. Chem. Int. Ed. 2010, 49, 6058-6082; Angew. Chem. 2010, 122, 6194-6219. [66] G. Sastre, J. van den Bergh, F. Kapteijn, D. Denysenko, D. Volkmer, Dalton Trans. 2014, 43, 9612-9619. [67] L. An-Hui, S. Dai (Eds.), Green Chemistry and Sustainable Technology Series: Porous Materials for Carbon Dioxide Capture, Springer, Berlin, Germany, 2014. [68] Y. S. Bae, 0. K. Farha, J. T. Hupp, R. Q. Snurr, J. Mater. Chem. 2009, 19, 2131-2134. [69] G. Orcajo, G. Calleja, J. A. Botas, L. Wojtas, M. H. Alkordi, M. Sanchez- Sanchez, Cryst. Growth Des. 2014, 14, 739-746. [70] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 1999, 32, 115-119. [71] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112-122. Received: February 10, 2016 Published Online: April 21, 2016 Eur. J. Inorg. Chem. 2016, 4450-4456 www.eurjic.org © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4456 I 37 Isophthalate-Hydrazone 2D Zinc-Organic Framework: Crystal Structure, Selective Adsorption, and Tuning of Mechanochemical Synthetic Conditions Autorzy: Kornel Roztocki, Damian Jędrzejowski, Maciej Hodorowicz, Irena Senkovska, Stefan Kaskel, Dariusz Matoga Opublikowano w: Inorg. Chem. 2016, 55, 9663-9670 Publikacja II I 38 Inorçanictaistry pubs.acs.org/IC Isophthalate-Hydrazone 2D Zinc-Organic Framework: Crystal Structure, Selective Adsorption, and Tuning of Mechanochemical Synthetic Conditions Kornel Roztocki, Damian Jędrzejowski, Maciej Hodorowicz, ' Irena Senkovska,1 Stefan Kaskel, and Dariusz Matoga*’’ ^Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland ^Department of Inorganic Chemistry, Technische Universität Dresden, Bergstrasse 66, 01062 Dresden, Germany Q Supporting Information ABSTRACT: A new layered mixed-linker metal—organic framework [Zn2( iso)2(pcih)2]„ (MOF) built from isophthalate ions (iso2 ) and 4- pyridinecarbaldehyde isonicotinoyl hydrazone (pcih) was prepared using both solution and mechanochemical methods. By use of the latter, the 2D MOF is obtained either in a one-mortar three-component grinding or on the way of a two-step mechanosynthesis. Tuning of mechanochemical synthetic conditions allowed us to identify both necessary and favorable factors for the solid-state formation of the MOF. Single-crystal X-ray diffraction reveals the presence of interdigitated layers in the ABAB arrangement and interlayer 0D cavities filled with guest molecules. Upon thermal activation, the dynamic framework exhibits stepwise and selective adsorption of C02 over N2 as well as high- pressure H2 adsorption reaching maximum excess of 1.15 wt% at 77 K. The mechanochemical synthetic protocol is expanded to a few other interdigitated structures. ■ INTRODUCTION Intensive scientific investigations of metal—organic frameworks (MOFs), i.e., porous coordination polymers (PCPs), are largely driven by their extraordinary porosity as well as flexible and modular structures.1’2 These features allow for rational tailoring of these materials for a plethora of possible applications such as, e.g., in catalysis,3 gas storage and separations,4 drug delivery,'^ and sensing. The empty cavities of MOFs along with their flexibility are essential attributes to create guest-induced properties and structural transformations.^ In the case of two-dimensional MOFs, voids are mostly represented by interlayer spaces that may become accessible for either guests or molecules enabling 2D to 3D reactions.9 A change in the amount or chemical nature of interlayer guests may lead to layers displacement (e.g., increased separation, delamination) and accompanying change of physicochemical properties. For instance, such dynamic behaviors of layers are responsible for adsorption-induced gate-opening effects in MOF materials.10 Development of synthetic routes is essential in the advancement of the chemistry of MOFs and their potential applications. The majority of conventional MOF preparation routes require both costly solvents and relatively long heating times.11 However, since the first successful synthesis of a MOF through facile milling of reactants,1- mechanochemical methods have emerged as a powerful tool for ecological and cost- effective preparations of extended frameworks.13 These techniques enable syntheses without bulk solvents, either as neat or as liquid-assisted grinding (LAG), and thus also from poorly soluble substrates. The possibility to vary liquid additive or its amount in LAG as well as to vary initial metal-containing reactant provides a good opportunity to optimize and to direct mechanochemical reaction. For instance, dependent on a solvent and its amount used in LAG, different coordination polymers can be obtained from the same reactants.14 Similarly, using appropriate metal-containing precursors can be a key factor to obtain desirable MOFs. Recent applications of such a strategy, with preassembled benzoate metal clusters, have led to successful mechanochemical syntheses of MOF-5 and UiO- 66.1 ^ Moreover, metal—organic frameworks themselves have been found to be reactive under grinding conditions. In particular, both interconversions of MOF structures and synthesis of mixed-ligand materials from single-linker MOFs have been reported.16 Similarly, very recent postsynthetic grinding of JUK-1 with an ionic solid resulted in a remarkable reversible bilayer unzipping and formation of a single-layer JUK-2.17 Herein, we present the synthesis, crystal structure, and adsorption properties of a new layered mixed-linker MOF [Zn2(iso)2(pcih)2]„ constructed from isophthalate ions (iso2-) and N,N-donor bridging ligand: 4-pyridinecarbaldehyde isonicotinoyl hydrazone (pcih), belonging to a class of Received: June 10, 2016 Published: September 22, 2016 r ACS Publications ® 2016 American Chemical Society DOI: 10.1021 /acs.inorgchem.6b01405 Inorg. Chem. 2016, 55, 9663-9670 9663 I 39 Figure I. X-ray crystal structure of {[Zn2(iso)2(pcih)2]-2DMF}n (l): (a) arrangement of adjacent layers (guest molecules are omitted); (b) schematic representation of hydrogen bonds between the two adjacent layers; (c) solvent-accessible voids calculated with Mercury software by using a probe molecule with a radius of 1.2 A; (d) arrangement of hydrazone linkers and carboxylate—zinc chains in a single layer; (e) ID double chain of [Zn2(iso)2]„ in a single layer; (f) coordination environment within Zn2 cluster with atom-labeling scheme and 30% displacement ellipsoids. H atoms in a, c, and e were omitted for clarity. acylhydrazones. The replacement of recently used linear the [Zn2(pdcx)2(pcih)2]„ MOF for CO^ N2, and H2 gases are dicarboxylic acids with their angular analogue led to the discussed along with its structural features, structurally different material of a layered topology as opposed to the previously reported interpenetrated 3D MOFs ® RESULTS AND DISCUSSION [Zn2(pdcx)2(pcih)2]„ (pdcx = 1,4-benzenedicarboxylate or Crystal Structure. Single crystals of {[Zn2(iso)2(pcih)2]- 4,4'-biphenyldicarboxylate).ls Moreover, the 2D MOF exhibits 2DMF}„ (l) (pcih = 4-pyridinecarbaldehyde isonicotinoyl different sorption behavior and synthetic availability. Here, the hydrazone; H2iso = isophthalic acid) suitable for X-ray layered MOF [Zn2(iso)2(pcih)2]„, unlike [Zn2(pdcx)2(pcih)2]łi; diffraction (XRD) were grown from DMF/H20 solution (see is easily obtained by mechanochemical methods, either in a Materials and Methods). The XRD experiment reveals that 1 one-mortar three-component grinding or on the way of a two- crystallizes in triclinic system, space group PI, and contains two step mechanosynthesis. Such one-step multicomponent and zinc(n) ions' tw0 Pcih' and two inhalate linkers as well as i.. . . .. i i i . « . r .. two DMF molecules in an asymmetric unit. The framework 1 is multistep one-pot sequential mechanochemical transformations 7 L.„ , Lii , r .. i. j . , a 2D mixed-linker coordination polymer with adjacent layers are still very rare however not only observed for the solid-state . . , . _ . _ , . . i • r » w i i u i i i arranged in the AdAB sequence. 1 he layers interact via strong synthes.s of MOFs. Very recently, these paral el mechanochem- N—H -O hydrogen bonds [d(N39-071 ) = 2.830(3) À, angle ical strategies have also been successfully used for the synthesis 172(3)„. d(N19_088#) = 2 858(3) A, angle 172(4)°] between of orgaxiometalhc compounds and semiconducting nano- the hydrazone NH group and the carboxylate oxygen (Figure crystals." Herein, various mechanochemical synthetic con- 1). Furthermore, the adjacent layers propagating in the ac plane ditions that allow to identify necessary and favorable factors for are interdigitated and create 0D pores in the structure that are the solid-state formation of the [Zn2(pdcx)2(pcih)2]„ MOF, are occupied by DMF molecules (Figure lc). The calculations described. A few other previously reported analogous performed by Mercury 3.5.1 software (with a probe radius = interdigitated layered frameworks are also successfully synthe- 1.20 À) reveal that these cavities occupy 272.6 Â3, i.e., 12.5% of sized by this method. Furthermore, the adsorption properties of the unit cell volume. Each 2D sheet is assembled from ID 9664 DOI: 10.1021/acs.inorgchem.6b01405 Inorg. Chem. 2016, 55, 9663-9670 Inorganic Chemistry I 40 Inorganic Chemistry [Zn2(iso)2]„ double chains running along the [100] direction and connected by pcih along the c axis (Figure Id and le). The isophthalate dianion acts as a fiyic,K ,k linker forming carboxylate-bridged Zn2 clusters and linking them into ID double chains. The coordination geometry of zinc atoms in the cluster can be described as disordered octahedral in which equatorial positions are occupied by oxygen atoms from three different iso2- ligands. Two N pyridyl atoms of //2-pcih ligands coordinate axially and complete the coordination environment of zinc(ll) ions (Figure If). The Zn—N bonds (Zn2—N31 = 2.185(2) À and Zn2—N45 = 2.186(2) Â) are of almost identical lengths, unlike the Zn—O bonds where Zn2—088 = 2.401(2) À is longer in comparison to the rest of the zinc oxygen bonds (Zn2—089 = 2.085(2) À; Zn2-091 = 2.013(2) A; Zn2-092 = 2.043(2) A) (Figure Id). As revealed by single-crystal X-ray diffraction, this is caused by the involvement of the 088 atoms in the hydrogen bonds with NH groups of pcih ligands from adjacent layers. The infrared spectrum of {[Zn2(iso)2(pcih)2]-2DMF}„ (l), shown in Figure SI, contains several characteristic absorption bands corresponding to pcih ligands, DMF molecules, and 1,3- benzenedicarboxylate ions iso2-. The strong bands observed at 1391 and 1557 cm-1 can be attributed to symmetric v{COO)s and asymmetric ^(COO)^ vibrations of carboxylates. The amide N—H and C=0 bands as well as those of imine N=C group of pcih hydrazone are found at 3448, 1687, and 1611 cm-1, respectively. The formation of hydrogen bonds between N—H groups and carboxylate oxygen atoms causes a significant blue shift of the amide N—H band in 1 as compared to free pcih ligand (3207 cm-1).21 Analogous assignments have recently been reported for two interpenetrated 3D frameworks {[Zn2(pdcx)2(pcih)2]guests}N based on p-dicarboxylic acids (pdcx = 1,4-benzenedicarboxylate or 4,4'-biphenyldicarboxy- late).18 The appearance of an additional strong band, which was observed in the characteristic range for carbonyl groups at 1667 cm-1 and disappeared after the framework was thermally activated at 200 °C, confirms the presence of guest DMF molecules in the investigated MOF (Figures SI and S2). Synthetic Routes. A microporous, mixed-linker metal- organic framework is representative of rare coordination polymers of the type [Zn2(iso)2(lig)2]22 (lig = neutral ligand). It could be successfully prepared with the use of both wet method as well as mechanosynthesis from various sources of zinc(ll) ions (Scheme l). Scheme 1. Two General Synthetic Routes to {[Zn2(iso)2(pcih)2]-2DMF}„ (l) The reaction between Zn(N03)2-6H20, H2iso, and pcih in DMF/H20 solution was conducted in a sealed vessel at elevated temperature for nearly 3 days. This synthesis leads to the formation of single crystals of 1 suitable for XRD, allowing crystal structure elucidation. However, this procedure requires a considerable amount of solvent and energy and thus cannot be considered as environmentally friendly. On the other hand, framework 1 can also be prepared quantitatively by grinding a variety of zinc compounds (including ZnO, [Zn(C03)]2- [Zn(OH)2]3, Zn(C03), and Zn(CH3COO)2) together with benzene-1,3-dicarboxylic acid and 4-pyridinecarbaldehyde isonicotinoyl hydrazone in a ratio of 1:1:1 (Figure 2). All Figure 2. PXRD patterns of {[Zn2(iso)2(pcih)2]'2DMF}„ (l) obtained in LAG from various zinc precursors compared to the pattern simulated from single-crystal XRD (calc). mechanochemical reactions of various zinc precursors (LAG with DMF) were accompanied by a release of CO^ H20, or CH3COOH evaporating from the reaction mixture. In all cases, grinding was conducted in the presence of a small amount of DMF (LAG23 = liquid-assisted grinding) which ascertains the suitable environment as well as fills the interlayer regions of the 2D MOF product. The framework 1 was not obtained when the substrates were either neatly ground or when H20, MeOH, or EtOH was used instead of DMF (Figure S3). These findings indicate that DMF molecules act as a template and enable mechanochemical template-directed synthesis of the frame- work.24 The powder X-ray diffraction revealed that regardless of the used zinc precursor, the highly crystalline product of 1 was formed. These experiments, combined with elemental analyses and IR spectra of solids after grinding, also confirm that the framework is formed quantitatively under appropriate LAG conditions. In the case of all precursors except ZnO, the ratio of liquid volume to weight of reactants (//) ' within the range from 0.49 to 0.63 fih/mg is sufficient for exclusive formation of the solid 1. However, in the case of LAG with ZnO and r] = 0.63 ^L/mg (e.g., 60 //L DMF/94.7 mg reactants) unreacted ZnO was detected in the PXRD pattern (Figure S4). Increasing the amount of DMF to t] = 158 //L/mg (e.g., 150 /^L/94.7 mg reactants) or grinding with small addition of salts [NH4N03 or (NH4)zS04; mass fraction w = 1%; ILAG26] caused complete DOI: 10.1021 /acs.i norgchem.6b01405 Inorg. Chem. 2016, 55, 9663-9670 9665 I 41 Inorganic Chemistry disappearance of ZnO from the final product (Figure S4). The lower lattice energies of ZnC03 (3273 kj-mol-1) and Zn(OH)2 (3158 kj-mol-1) compared to that of ZnO (3971 kj-mol-1)27 are favorable for the formation of 1. However, it is noteworthy that attempts of grinding Zn(N03)2 or ZnCl?, whose lattice energies are even smaller (2649 and 2748 kj-mol-1, respectively),28 together with pcih and H2iso were unsuccessful and gave unidentified, predominantly amorphous products (Figure S3). Considering the reaction enthalpy as driving the mechanochemical reaction, the formation of molecules released from these salts during MOF formation should also be taken into account. For instance, standard enthalpies of formation of C02(g), H20(g), and CH3COOH(g) are relatively high (approximately —393, —242, and —434 kj mol-1, respectively) as compared to that of HCl(g) (—92 kjmol- ). Therefore, salts or oxides of basic character, preferably with low lattice energy and releasing stable byproducts, are best reactants for mechanochemical synthesis of framework 1. It has also been found that {[Zn2(iso)2(pcih)2]-2DMF}„ (l) can be synthesized mechanochemically in a stepwise manner. In the first step, water-assisted grinding (60 yL/49.5 mg reactants) between ZnO and H2iso gave a product previously reported in the literature which is a 2D coordination polymer [Zn(iso)(H20)]„ (2).2J The PXRD pattern of the intermediate 2 is identical with that calculated from the data in the Cambridge Structural Database (CSD code QOGJIF). In the next step, 2 and pcih were ground in the presence of 60 fiL of DMF/95.7 mg of reactants (Figure S5) which led to the final product 1. Mechanistically, the reaction between 2 and pcih requires removal of one HzO molecule per zinc ion as well as zinc— carboxylate bond rearrangement to replace [Zn(iso)]„ layers with [Zn2(iso)2]f) double chains containing dinuclear zinc clusters pillared by N,N-donor bridges (pcih). Various mechanochemical routes to framework 1 have been summar- ized in Scheme 2. Scheme 2. Various Mechanochemical Routes to {[Znj(iso)2(pcIh)j]-2DMF}„ (1) Additionally, a few more grinding trials to synthesize analogous interdigitated structures of the type [M2(ang_dcx)2(lin_lig)2] (M = Zn, Cd; ang_dcx = angular dicarboxylate; lin lig = linear neutral bridging ligand) have been carried out. By use of the same environmentally friendly synthetic protocol we successfully obtained previously reported and structurally characterized interdigitated 2D frameworks: {[Zn2(iso)2(bpy),]DMF]}„ {[Cd2(bpndc)2(bpy)2]-(H20)- (DMF)}„, and {[Cd,(iso)2(bpy)2]-guests]}„ (bpy = 4,4'- bipyridine; bpndc = benzophenone-4,4'-dicarboxylate) (Sup- porting Information, Figures S6—S9) 22aj30'31 Sorption Properties. Thermogravimetric analysis of frame- work 1 shows a stepwise weight loss with a stable plateau in the range 200—320 °C (Figure S10). The first pronounced step is associated with the loss of two N,N'-dimethylformamide molecules per formula unit (found 13.6%; calcd weight loss 13.9%). The second distinct step with a dTG minimum at approximately 350 ÔC corresponds to decomposition of the framework. The PXRD pattern of the as-synthesized 1 very well matches that calculated from single-crystal data; it differs however from the pattern of the desolvated phase 1' (Figure 3). These PXRD patterns indicate that 1 shows the retention of crystallinity and structural flexibility after guest removal. Figure 3. PXRD patterns of {[Zn2(iso)2(pcih)2]-2DMF}fl (l): (a) simulated from single-crystal data, (b) as synthesized, and (c) desolvated (!'). Sorption analysis reveals that framework 1' selectively adsorbs C02 over N2 after thermal activation in vacuum at 180 °C (Figure 4). In the literature, the group of similar coordination polymers with interdigitated layered structures, belonging to the family of [Zn2(dcx)2(lig)2] frameworks6 (dcx = dicarboxylate), has been reported to show very interesting sorption properties due to their structural flexibility. In most cases these frameworks do not adsorb N2 or exhibit the gate- opening phenomenon. Herein, we report the first mechanosyn- thesis of such layered frameworks on the example of a new {[Zn2( iso)2(pcih)2]‘2DMF}„. Generally, selective adsorption of such interdigitated frameworks can be explained either as an effect of pores size or polarity or as a kinetic issue.22’"'0 DOI: 10.1021 /acs.inorgchem.6b01405 tnorg. Chem. 2016, 55, 9663-9670 9666 I 42 Inorganic Chemistry Figure 4. Adsorption isotherms of C02 at 195 K and N2 at 77 K on [Zn2(iso)2(pcili)2]„ (l')- Solid and open symbols represent adsorption and desorption branches, respectively. However, according to the Zeo++ calculations,32 the pore window size and maximum pore diameter are 1.80 and 4.15 À for the as-made 1. The kinetic diameters of investigated adsorptives are much larger than the pore window size in 1 (3.64 À for 3.30  for CO^ and 2.89 À for H2); therefore, no of N2, C02, or H2 should be able to enter the pores. On the other hand, the PXRD pattern of 1' indicates some structural changes of 1 upon desolvation, most likely also leading to interlayer pore window size change. Regardless of the window size, however, carbon dioxide is a favored adsorptive because of decoration of the framework pores by polar NH and CO groups of pcih that interact with the high quadrupole moment of C02.33 Therefore, the selective C02 over N2 adsorption behavior can be explained by the C02 affinity toward the structurally flexible framework rather than by the size of the adsorbent. The adsorption curve for C02 has a double step profile with a wide hysteresis caused by the guest-induced framework response (Figure 4).34 The first distinct step within the p/p0 range of 10-5—0.173 ends with C02 adsorption of 97 mg-g_1 (49 cm3-g_1) and can be associated with filling the voids volume in 1'. The theoretical pore volume of 1 calculated geometrically using Zeo-H- (probe radius 0.8 A) or Mercury (probe radius 1.2 A) for the single-crystal structure of as-made material after excluding the solvent is 0.090 cm3-g_1. The pore volume of 1' derived from the C02 adsorption isotherm at p/po = 0.17 (first plateau) is 0.087 cm3-g-1, matching well the theoretical value. Further increase of the pressure leads to the steep increase in C02 uptake, which can be explained by the framework layers displacement. Obviously, the expansion of the interlayer distance generates additional accessible space, and the total pore volume calculated at p/p0 = 0.99 amounts to 0.144 cm3-g_1, surpassing nearly twice the theoretical pore volume of I. High pore volume is reflected in high adsorbed C02 amount of 91 cm3-g_1 (181 mgg-1). In order to investigate the adsorption behavior of 1 using smaller adsorptive at low temperature, the adsorption measure- ment up to 100 bar for H2 was carried out at 77 K (Figure Sll). The isotherm shows a maximum excess adsorption capacity of II.47 mg/g (I.I5 wt%) at 18 bar and a wide range hysteresis loop, indicating possibly some structural transformation of the network during the adsorption. ■ CONCLUSION In summary, we report a new two-dimensional MOF built from a linear bridging hydrazone and bent dicarboxylate linkers. Dynamic interdigitated layers containing amide groups are responsible for selective, stepwise adsorption of C02 over N2. Moreover, the activated framework is also able to adsorb H2 at 77 K. The results reveal that the MOF is obtainable by both solution and one- or two-step mechanochemical syntheses. Within the latter, metal-containing precursors of a basic character, preferably with low lattice energy and releasing stable byproducts, favor the template-directed MOF formation. The findings improve our insight into possibilities of grinding- induced transformations and behavior of layered materials. ■ MATERIALS AND METHODS 4-Pyridinecarbaldehyde isonicotinoyl hydrazone (pcih)21 and zinc carbonate'^ were prepared according to the published methods. All other reagents and solvents were of analytical grade (Sigma-Aldrich, POCH, Polmos) and used without further purification. Carbon, hydrogen, and nitrogen were determined by conventional microanalysis with the use of an Elementar Vario MICRO Cube elemental analyzer. IR spectra were recorded on a Thermo Scientific Nicolet iS5 FT-IR spectrophotometer equipped with an iD5 diamond ATR attachment. Thermogravimetric analyses (TGA) were performed on a Mettler- Toledo TGA/SDTA 851° instrument at a heating rate of 5 °C min-1 in a temperature range of 25—600 °C (approximate sample weight of 50 mg). The measurement was performed at atmospheric pressure under flowing argon. Powder X-ray diffraction (PXRD) patterns were recorded at room temperature (295 K) on a Rigaku Miniflex 600 diffractometer with Cu Table 1. Various Mechanosynthetic Conditions Used in Attempts of Preparation of {[Zn2(iso)2(pcih)2]*2DMF}„ (l) DOI: 10.1021 /acs.inorgchem.6b01405 Inorg, Chem. 2016, 55, 9663-9670 9667 zinc source LAG time [min] additional salt n [//L-mg *] ZnO 20 ZnO 60 juL of MeOH 20 0.63 ZnO 60 fxL of EtOH 20 0.63 ZnO 60 fiL of H20 20 0.63 ZnO 150 fń, of DMF 25 1.58 ZnO 60 fiL of DMF 10 1 mg of NH4NO3 0.63 ZnO 60 ji/L of DMF 10 1 mg of (NH4)2(S04) 0.63 Zn(0Ac)2-*H20 60 fjL of DMF 20 0.49 ZnCOj 60 fiL of DMF 8 0.58 [ZnC O3 ] 2 [Zn{ OH )2] 3 60 //L of DMF 12 0.60 ZnCl2 60 fiL of DMF 20 0.57 Zn(N0j)26H:0 60 fiL of DMF 17 0.43 I 43 Inorganic Chemistry Stepwise Mechanosynthesis of {[Zn2(iso)2(pcih)2]-2DMF}n (1). ZnO (16.3 mg, 0.20 mmol) and 1,3-benzenedicarboxylic acid (H2iso) (33.2 mg, 0.20 mmol) were ground together with 60 /jL of H20 (// = 1.21 //L-mg-1) in an agate mortar for 7 min. After that 4- pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (45.2 mg, 0.20 mmol) and 60 //L of DMF (// = 0.63 ;/L-mg-1) were added to the mortar, and grinding for 11 min was carried out. Crystallographic Data Collection and Structure Refinement. A single crystal of 1 suitable for X-ray analysis was selected from bulk material (see the syntheses section). Diffraction data for compounds were collected on an Oxford Diffraction SuperNova four circle diffractometer equipped with the Mo Ka (0.71073 À) radiation source, graphite monochromator, and Oxford Cryojet system for measure- ments at 120 K. The position of all non-hydrogen atoms was determined by direct methods using SIR-97.'6 All non-hydrogen atoms were refined anisotropically using weighted full-matrix least- squares on F2. Refinement and further calculations were carried out using SHELXL 2014/6/ All hydrogen atoms joined to carbon atoms were positioned with idealized geometries and refined using a riding model with 17^(H) fixed at 1.5 Ucq of C for methyl groups and 1.2 Ueq of C for other groups. The H atoms attached to the N atoms were found in the difference-Fourier map and refined with an isotropic thermal parameter. The C atoms of one of the two DMF solvent molecules were refined as equally disordered over two sets of sites using DFIX and DANG instructions. Since some anisotropic displacement ellipsoids were rather elongated, DELU/S1MU restraints were also applied. The structure refinement was handicapped by the pseudosymmetry (pseudocentering I) of the structure and by disorder of solvent molecules. Use of PLATON’s ADDSYM algorithm clearly reveals that the correct space group choice has been used for the structure solution (model) and refinement. Crystal data and structure refinement parameters have been collected in Table 2. ■ ASSOCIATED CONTENT O Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorg- chem.6b01405. IR spectrum, PXRD patterns, TGA data, and X-ray crystal data (CCDC1451728 for l) (PDF) (CIF) Kcr radiation (A = 1.5418 A) in a 20 range from 3° to 45° with a 0.05° step at a scan speed of 2.5° min-1. Nitrogen and carbon dioxide adsorption studies were performed on a BELSORP-max adsorption apparatus (MicrotracBEL Corp.); 77 K was achieved by a liquid nitrogen bath, and 195 K was achieved by a dry ice/acetone bath. Samples were evacuated at 180 °C for 16 h prior to adsorption measurements. Syntheses. Synthesis of {[Zn2(iso)2(pcih)2]-2DMF}n in Solution (1). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Zn(N03)2-6H20 (88.0 mg, 0.295 mmol), and 1,3- benzenedicarboxylic acid (H2iso) (49.8 mg, 0.300 mmol) were dissolved in DMF (16 mL) and H20 (1.5 mL) by sonification (60 s) and heated at 70 °C for 96 h. Yellow crystals of I were filtered off, washed with DMF, and dried in oven at 60 °C and 500 mbar for 0.5 h. Yield: 114.8 mg (72.3%). Anal. Calcd for C^H^N^O^Zn^ C, 52.24; H, 4.00; N, 13.24. Found: C, 52.06; H, 4.19; Ń, 13.21. FTIR (ATR, cm-1): COOX* 1557s, COO)s 1391s, ^(C=0)DMF 1667s, v{C= N)pcih 1611s, ^(C=0)pcih 1687s, i'(NH) 3207m. Mechanosynthesis of {[Zn2(iso)2(pcih)2]-2DMF}n (1). 4-Pyridine- carbaldehyde isonicotinoyl hydrazone (pcih) (45.2 mg, 0.200 mmol), ZnO (16.3 mg, 0.200 mmol), and 1,3-benzenedicarboxylic acid (H2iso) (33.2 mg, 0.200 mmol) were ground together with 150 nL of DMF (// = 1.58 /vL-mg-1) in a mortar for 25 min. All other mechanosyntheses were carried out in the same way except that various LAG solvents and zinc sources were used instead of DMF and ZnO (Table l). ■ AUTHOR INFORMATION Corresponding Author *E-mail: dariusz.matoga(a)uj.edu.pl. Author Contributions K.R. and D.J. carried out wet and mechanochemical syntheses as well as related experiments. M.H. refined the crystal structure. I.S. and S.K. performed and discussed adsorption measurements. D.M. conceived and led the project overall. K.R. and D.M. wrote the manuscript with feedback from all coauthors. Notes The authors declare no competing financial interest. ■ ACKNOWLEDGMENTS The National Science Centre (NCN, Poland) is gratefully acknowledged for the financial support (Grant no. 2015/17/B/ ST5/01190) of this research. The research was carried out partially with the equipment purchased thanks to the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (contract no. POIG.02.01.00-12-023/08). DOI: 10.1021/acs.inorgchem.6b01405 Inorg. Chem. 2016, 55, 9663-9670 Table 2. Crystal Data and Structure Refinement Parameters for {[Zn2(iso)2(pcih)2]*2DMF}„ (l) 9668 compound 1 empirical formula Zn2C46H4|N10O12 fw 1056.63 cryst size (mm) 0.400 X 0.360 X 0.170 cryst syst triclinic space group Pi unit cell dimens »(A) 10.0578(3) b(A) 15.0382(5) c (A) 15.8120(5) a (deg) 101.385(3) ß (deg) 94.968(2) Y (deg) 109.331(3) vol (A3) 2182.30(13) temp (K) 120(2) Z 2 density (calcd) (g/cm3) 1.608 abs coeff (mm-1) 1.179 F(000) 1086 theta range for data 2.937-28.725 collection (deg) index ranges — 13 < h < 13, -18 2ff(/)] completeness to theta = 99.8% 25.242° abs corr Gaussian max and min 0.842 and 0.699 transmission refinement method full-matrix least-squares on F2 data/restraints/params 10 262/10/655 goodness-of-fit on F2 1.057 final R indices R, = 0.0440, wR2 = 0.1088 [J > 2»(I)] R indices (all data) R, = 0.0616, wR2 = 0.1211 I 44 Inorganic Chemistry ■ REFERENCES (1) Books on MOFs: (a) In Metal-Organic Frameworks; Farrusseng, D., Ed.; Wiley-VCH Verlag & Co. KGaA: Weinheim, 2011. (b) In Metal-Organic Frameworks-, MacGillivray, L. R., Ed.; John Wiley & Sons Inc.: Hoboken, 2010. (2) Recent general reviews on MOFs: (a) Aliendorf, M. D.; Stavila, V. Crystal engineering, structure-function relationships, and the future of metal-organic frameworks. CrystEngComm 2015, 17, 229—246. (b) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. The chemistry and applications of metal-organic frameworks. Science 2013, 341, 1230444. (3) (a) Gascon, J.; Corma, A; Kapteijn, F.; LLabres i Xamena, F. X. Metal Organic Framework Catalysis: Quo vadis? ACS Catal. 2014, 4, 361—378. (b) Zhang, T.; Lin, W. Metal-organic frameworks for artificial photosynthesis and photocatalysis. Chem. Soc. Rev. 2014, 43, 5982—5993. (c) Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H. Commercial metal-organic frameworks as heterogeneous catalysts. Chem. Commun. 2012, 48, 11275—11288. (4) Li, B.; Wen, H.-M.; Zhou, W.; Chen, B. J. Porous Metal—Organic Frameworks for Gas Storage and Separation: What, How, and Why? J. Phys. Chem. Lett. 2014, 5, 3468-3479. (5) (a) Horcajada, P.; Gref, R.; Baati, T.; Allan, P. K.; Maurin, G.; Couvreur, P.; Férey, G.; Morris, R. E.; Serre, C. Metal—Organic Frameworks in Biomedicine. Chem. Rev. 2012, 112, 1232—1268. (b) Sun, C.-Y.; Qin, C.; Wang, X.-L.; Su, Z.-M. Metal-organic frameworks as potential drug delivery systems. Expert Opin. Drug Delivery 2013, 10, 89-101. (6) (a) Kreno, L. E.; Leong, K; Farha, O. K; Aliendorf, M.; Van Duyne, R. P.; Hupp, J. T. Metal—Organic Framework Materials as Chemical Sensors. Chem. Rev. 2012, 112, 1105—1125. (b) Hu, Z.; Deibert, B. J.; Li, J. Luminescent metal-organic frameworks for chemical sensing and explosive detection. Chem. Soc. Rev. 2014, 43, 5815-5840. (7) Allendorf, M. D.; Foster, M. E.; Léonard, F.; Stavila, V.; Feng, P. L.; Doty, F. P.; Leong, K; Ma, E. Y.; Johnston, S. R.; Talin, A A J. Guest-Induced Emergent Properties in Metal—Organic Frameworks. J. Phys. Chem. Lett. 2015, 6, 1182-1195. (8) Schneemann, A; Bon, V.; Schwedler, L; Senkovska, I.; Kaskel, S.; Fischer, R. A. Flexible metal-organic frameworks. Chem. Soc. Rev. 2014, 43, 6062-6096. (9) (a) Roth, W. J.; Gil, B.; Makowski, W.; Marszalek, B.; Eliasova, P. Layer like porous materials with hierarchical structure. Chem. Soc. Rev. 2016, 45, 3400. (b) Guillerm, V.; Kim, D.; Eubank, J. F.; Luebke, R.; Liu, X.; Adil, K.; Lah, M. S.; Eddaoudi, M. A supermolecular building approach for the design and construction of metal-organic frameworks. Chem. Soc. Rev. 2014, 43, 6141-6172. (10) Horike, S.; Shimomura, S.; Kitagawa, S. Soft porous crystals. Nat. Chem. 2009, 1, 695-704. (11) Stock, N.; Biswas, S. Synthesis of Metal-Organic Frameworks (MOFs): Routes to Various MOF Topologies, Morphologies, and Composites. Chem. Rev. 2012, 112, 933—969. (12) Pichon, A; Lazuen-Garay, A.; James, S. L. Solvent-free synthesis of a microporous metal-organic framework. CrystEngComm 2006, 8, 211-214. (13) (a) James, S. L.; Adams, C. J.; Bolm, C.; Braga, D.; Collier, P.; Friscić, T.; Grepioni, F.; Harris, K D. M.; Hyett, G.; Jones, W.; Krebs, A; Mack, J.; Maini, L.; Orpen, A G.; Parkin, I. P.; Shearouse, W. C.; Steed, J. W.; Waddell, D. C. Mechanochemistry: opportunities for new and cleaner synthesis. Chem. Soc. Rev. 2012, 41, 413—447. (b) Friscić, T. Supramolecular concepts and new techniques in mechanochemis- try: cocrystals, cages, rotaxanes, open metal-organic frameworks. Chem. Soc. Rev. 2012, 41, 3493—3510. (c) Boldyreva, E. Mechanochemistry of inorganic and organic systems: what is similar, what is different? Chem. Soc. Rev. 2013, 42, 7719-7738. (14) Friscić, T.; Fabian, L. Mechanochemical conversion of a metal oxide into coordination polymers and porous frameworks using liquid- assisted grinding (LAG). CrystEngComm 2009, 11, 743—745. (15) (a) Prochowicz, D.; Sokołowski, K; Justyniak, I.; Komowicz, A; Fairen-Jimenez, D.; Friscić, T.; Lewiński, J. A mechanochemical strategy for IRMOF assembly based on pre-designed oxo-zinc precursors. Chem. Commun. 2015, 51, 4032—4035. (b) Uzarevic, K; Wang, T. C.; Moon, S.-Y.; Fidelii, A M.; Hupp, J. T.; Farha, O. K; Friscić, T. Mechanochemical and solvent-free assembly of zirconium- based metal-organic frameworks. Chem. Commun. 2016, 52, 2133— 2136. (16) Yuan, W.; Friscić, T.; Apperley, D.; James, S. L. High Reactivity of Metal—Organic Frameworks under Grinding Conditions: Parallels with Organic Molecular Materials. Angew. Chem., Int. Ed. 2010, 49, 3916-3919. (17) Matoga, D.; Oszajca, M.; Molenda, M. Ground to conduct: mechanochemical synthesis of a metal-organic framework with high proton conductivity. Chem. Commun. 2015, 51, 7637—7640. (18) Roztocki, K; Senkovska, I.; Kaskel, S.; Matoga, D. Carboxylate— Hydrazone Mixed-Linker Metal—Organic Frameworks: Synthesis, Structure, and Selective Gas Adsorption. Eur. J. Inorg. Chem. 2016, DOI: 10.1002/ejic.201600134. (19) Hernandez, J. G.; Butler, I. S.; Friscić, T. Multi-step and multi- component organometallic synthesis in one pot using orthogonal mechanochemical reactions. Chem. Sei. 2014, 5, 3576—3582. (20) Krupiński, P.; Kornowicz, A.; Sokołowski, K.; Cieślak, A. M.; Lewiński, J. Applying Mechanochemistry for Bottom-Up Synthesis and Host—Guest Surface Modification of Semiconducting Nanocrystals: A Case of Water-Soluble /7-Cyclodextrin-Coated Zinc Oxide. Chem. - Eur. J. 2016, 22, 7817-7823. (21) Ni, W.-X.; Li, M.; Zhou, X.-P.; Li, Z.; Huang, X-C; Li, D. pH- Induced formation of metalloligand: increasing structure dimension- ality by tuning number of ligand functional sites. Chem. Commun. 2007, 3479-3481. (22) (a) Horike, S.; Tanaka, D.; Nakagawa, K; Kitagawa, S. Selective guest sorption in an interdigitated porous framework with hydro- phobic pore surfaces. Chem. Commun. 2007, 3395—3397. (b) Fukush- ima, T.; Horike, S.; Inubushi, Y.; Nakagawa, K; Kubota, Y.; Takata, M.; Kitagawa, S. Solid Solutions of Soft Porous Coordination Polymers: Fine-Tuning of Gas Adsorption Properties. Angew. Chem. 2010, 122, 4930—4934. (c) Takashima, Y.; Furukawa, S.; Kitagawa, S. Control of the charge-transfer interaction between a flexible porous coordination host and aromatic guests by framework isomerism. CrystEngComm 2011, 13, 3360—3363. (d) Hijikata, Y.; Horike, S.; Sugimoto, M.; Sato, H.; Matsuda, R.; Kitagawa, S. Relationship between Channel and Sorption Properties in Coordination Polymers with Interdigitated Structures. Chem. - Eur. J. 2011, 17, 5138—5144. (23) (a) Shan, N.; Toda, F.; Jones, W. Mechanochemistry and co- crystal formation: effect of solvent on reaction kinetics. Chem. Commun. 2002, 2372—2373. (b) Friscić, T.; Jones, W. Recent Advances in Understanding the Mechanism of Cocrystal Formation via Grinding. Cryst. Growth Des. 2009, 9, 1621—1637. (24) Zhang, Z.; Zaworotko, M. J. Template-directed synthesis of metal-organic materials. Chem. Soc. Rev. 2014, 43, 5444—5455. (25) Friscić, T.; Childs, S. L.; Rizvi, S. A A; Jones, W. The role of solvent in mechanochemical and sonochemical cocrystal formation: a solubility-based approach for predicting cocrystallisation outcome. CrystEngComm 2009, 11, 418—426. (26) Friscić, T.; Reid, D. G.; Halasz, I.; Stein, R. S.; Dinnebier, R. E.; Duer, M. J. Ion- and Liquid-Assisted Grinding: Improved Mechanochemical Synthesis of Metal—Organic Frameworks Reveals Salt Inclusion and Anion Templating. Angew. Chem., Int. Ed. 2010, 49, 712-715. (27) Yoder, C. H.; Flora, N. Geochemical applications of the simple salt approximation to the lattice energies of complex materials. Am. Mineral. 2005, 90, 488-496. (28) Lide, D. R. (Editor-in-Chief) CRC Handbook of Chemistry and Physics, 88th ed.; CRC Press, 2007—2008. (29) Otto, T. J.; Wheeler, K. A Aqua(benzene-l,3-dicarboxyIato)- zinc(Il). Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 2001, 57, 704-705. (30) Tanaka, D.; Nakagawa, K.; Higuchi, M.; Horike, S.; Kubota, Y.; Kobayashi, T. C.; Takata, M.; Kitagawa, S. Kinetic Gate-Opening DOI: 10.1021/acs.inorgchem.6b01405 Inorg. Chem. 2016, 55, 9663-9670 9669 Inorganic Chemistry_________________________________________________________________________________________________Q Process in a Flexible Porous Coordination Polymer, Angew. Chem., Int. Ed 2008, 47, 3914-3918. (31) (a) Tao, J.j Chen, X M.j Huang, R. B.; Zheng, L. S. Hydrothermal syntheses and crystal structures of two rectangular grid coordination polymers based on unique prismatic [M8(ip)s(4,4'- bipy)8] building blocks [M = Ni(n) or Cd(ll), ip = isophthalate, bipy = bipyridine]. J. Solid State Chem. 2003, 170, 130—134* (b) Tian, G.; Zhu, G. S.; Fang, Q. R.; Guo, X. D.; Xue, M.; Sun, J. Y.; Qiu, S. L. Design, synthesis and fluorescence of two-dimensional pillared layers by connecting infinite one-dimensional chains via 4,4'-bipyridine. /. Mol. Struct. 2006, 787, 45-49. (32) Willems, T. F.; Rycroft, C. H.j Kazi, M.; Meza, J. C.; Harańczyk, M. Algorithms and tools for high-throughput geometry-based analysis of crystalline porous materials. Microporous Mesoporous Mater. 2012, 149, 134-141. (33) Prochowicz, D.j Justyniak, I.; Kornowicz, A; Kaczorowski, T.; Kaszkur, Z.; Lewiński, J. Construction of a Porous Homochiral Coordination Polymer with Two Types of Cunln Alternating Units Linked by Quinine: A Solvothermal and a Mechanochemical Approach. Chem. - Eur. J. 2012, 18, 7367-7371. (34) (a) Llewellyn, P. L.; Bourrelly, S.j Serre, C.; Filinchuk, Y.; Ferey, G. How hydration drastically improves adsorption selectivity for CO2 over CH4 in the flexible chromium terephthalate MIL-53. Angew. Chem., Int. Ed. 2006, 45, 7751-7754. (b) Kubota, Y.; Takata, M.; Matsuda, R.; Kitaura, R; Kitagawa, S.; Kobayashi, T. C. Metastable Sorption State of a Metal—Organic Porous Material Determined by In Situ Synchrotron Powder Diffraction. Angew. Chem. 2006, 118, 5054— 5058. (c) Kondo, A; Noguchi, H.; Carlucci, L.; Proserpio, D. M.; Ciani, G.; Kajiro, H.j Ohba, T.; Kanoh, H.; Kaneko, K Double—Step Gas Sorption of a Two-Dimensional Metal—Organic Framework. J. Am. Chem. Soc. 2007, 129, 12362. (35) Zhang, R.; Zhang, F.; Feng, J.; Qian, Y. Green and facile synthesis of porous ZnC03 as a novel anode material for advanced lithium-ion batteries. Mater. Lett. 2014, 118, 5—7. (36) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A; Moliterni, A G.; Poiidori, G.; Spagna, R. SIR97. J. Appl. Crystallogr. 1999, 32, 115-119. (37) (a) Sheldrick, G. M. SHELXS. Acta CrystallogrSect. A: Found. Crystallogr. 2008, 64, 112—122. (b) Sheldrick, G. M. SHELXLj 2014/ 6; University of Gottingen, Germany, 2014. I 45 9670 DOI: 10.1021 /acs.inorgchem.6b01405 Inorg. Chem. 2016, 55, 9663-9670 I 46 I 47 Effect of linker substituent on layers arrangement, stability and sorption of Zn isophthalate/acylhydrazone-frameworks Autorzy: Kornel Roztocki, Damian Jędrzejowski, Maciej Hodorowicz, Irena Senkovska, Stefan Kaskel, Dariusz Matoga Opublikowano w: Cryst. Growth Des., 2018, 18 (1), pp 488-497 Publikacja III I 48 ® Cite This: Cryst. Growth Des. 2018, 18, 488-497 pubs.acs.org/crystal Effect of Linker Substituent on Layers Arrangement, Stability, and Sorption of Zn-lsophthalate/Acylhydrazone Frameworks Kornel Roztocki, Damian Jędrzejowski, Maciej Hodorowicz/ [rena Senkovska, Stefan Kaskel, and Dariusz Matoga* ^Faculty of Chemistry, Jagiellonian University, Gronostajowa 2, 30-387 Kraków, Poland "Department of Inorganic Chemistry, Technische Universität Dresden, Bergstrasse 66, 01062 Dresden, Germany Q Supporting Information ABSTRACT: A series of mixed-linker metal—organic frame- works [Zn2(Xiso)2(pcih)2]„ containing substituted isophtha- late linkers (Xiso2-; X = OH or CH3 or NH2 or H) and 4-pyridinecarbaldehyde isonicotinoyl hydrazone pillars (pcih) have been prepared by using both solution and mechano- chemical methods. Single-crystal X-ray diffraction reveals their interdigitated two-dimensional structures with different arrangements of layers, dependent on hydrogen bonding and CH-;r interactions involving the substituents and/or linkers. These supramolecular interactions are responsible for the for- mation of interlayer pores of various volumes, shapes, and dimensionality. All materials exhibit selective gas adsorption of C02 over N2 with diverse profiles. Polar groups (OH and NH2) of the isophthalate linkers increase chemical affinity to carbon dioxide as well as hydrolytic and thermal stability of the frameworks. ■ INTRODUCTION In recent years, metal—organic frameworks (MOFs) or porous coordination polymers (PCPs) have been attracting consid- erable scientific interest due to their potential use in various fields of modern industry,1’2 e.^., gas storage and separation, - catalysis,7-10 drug delivery,11'2 electronic devices,1-16 and others.118 Such applications are closely related to their remark- able properties such as high internal surface areas, nearly limitless crystal engineering possibilities, as well as unique and versatile structural flexibility and responsivity.11,-22 The latter behavior of MOFs can be triggered by various stimuli and is enabled by labile intraframework metal-coordination bonds or by weaker interframework interactions (e.g., hydrogen bonds, van der Waals forces). Selected recent reports on unprece- dented MOF dynamics include negative gas adsorption pheno- menon,2 ' self-accelerating sorption,24 and reversible three-step reaction cycle for a family of layered MOFs.2>-~ There is a wide range of well developed tools for the design, engineering, and synthesis of MOF materials which enable the control of pore size, chemical affinity, stability, and other pro- perties of porous coordination polymers.28 The desirable MOF properties including stimuli-responsive behaviors can be achieved not only through a careful choice of metal centers and organic struts but also by more sophisticated methods such as chemical reactions within a metal center or linker,“9-'1 replacement of guests molecules or even whole building blocks, " as well as controlling electronic properties utilizing structural defects33 '4 and downsizing effects/''' Although the methods of synthesis of MOFs are well-known, the majority of them require an organic solvent, relatively long reaction times, and heating. However, since the influential work of James et al.,36 who showed that an extended MOF can be facilely prepared by 10 min of grinding dry components in a mortar, solvent-free methods have been introduced as an ecological and cost-effective alternative to conventional synthesis of porous coordination polymers.2'37-39 Sustaining the worldwide environment requires appropriate porous materials for adsorption and separation of C02 from exhaust gases.40,41 MOFs, with their high external surface, fre- quently combined with selective adsorption of carbon dioxide, have emerged as a reasonable class of materials for this urgent environmental problem. Along with the chemical affinity toward the greenhouse gas, there are additional requirements imposed on MOFs. Apart from the creation of specific interactions between a framework and CO^ the material must also be chemically (e.g., hydrolytically) and thermally stable.42-44 With regard to the former, very stable MOFs even in strongly alkaline solutions were reported recently.45’46 However, both academic and industrial societies need to put efforts into deep understanding of MOF modi operandi including stability, selec- tivity, and structure as well as the subde interrelationship between these features. Herein, we present the synthesis, crystal structure, sorption properties, and stability studies for a family of new mixed- linker MOFs {[Zn2(Xiso)2(pcih)2]-guest},, constructed from Received: October 20, 2017 Revised: November 16, 2017 Published: November 20, 2017 r ACS Publications ® 2017 American Chemical Society DOI: 10.1021 /acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 488 I 49 Crystal Growth & Design Scheme 1. Two General Synthetic Routes Leading to the Frameworks {[Zn2(Xiso)2(pcih)2] *guest}„rt aH2Xiso: 5-substituted 1,3-benzenedicarboxylic acid (X = OH (l); CH3 (2); NH2 (3) H (4-4b)); pcih: 4-pyridinecarbaldehyde isonicotinoyl hydrazone. Guest molecules: 1 (2DMF-H20); 2 (2DMF-2H20); 3 (3DMF); 4 (2DMF); 4b (2DMA-4H20). " 4-pyridinecarbaldehyde isonicotinoyl hydrazone (pcih) and 5-substituted isophthalate ions [Xiso2-, where X = OH (l); CH3 (2); NH2 (3); H (4-4b)]. All studied frameworks consist of the same neutral pcih linker belonging to a class of acylhy- drazones, which possess C=0 and N—H functional groups. In response to environmental expectations, we show that zinc MOFs can be easily prepared by grinding solid reactants with a small amount of a liquid, as an alternative to syntheses in solution. Single-crystal X-ray diffraction experiments reveal that the frameworks 1—4b are interdigitated layered coordination polymers built from metal-carboxylate clusters containing double pcih struts, similar as in the CID family,4 ~M) but with the key difference being structure-directing functional groups on N-donor linkers. The obtained new MOFs differ in a sub- stituent on the isophthalate linker, which provides an excellent opportunity to study the influence of pendant side groups on their structures as well as on the relationship between the structure and properties, including gas adsorption and stability. In the literature, changes of linker pendant groups have been shown to influence MOF topology,'"'1 structural flexibility/"'2 and stability.'1 In this work we present the effect of intermolecular interactions on mutual orientation of MOF layers, their C02 Table 1. Crystal Data and Structure Refinement Parameters for {[Zn2(Xiso)2(pcih)2]*guest}„ (l—4b) 489 DOI: 10.1021/acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 this work this work ref 57 this work compound 1 2 4 4b empirical formula ZnC20HI4N4O6 ZnC2IH16N4Os ZnaC+ÄiNuAj Zn2C48HSoNloO i5 formula weight 471.72 469.75 1056.63 1137.72 crystal size (mm) 0.300 X 0.100 X 0.050 0.300 X 0.300 X 0.300 0.400 X 0.360 X 0.170 0.400 X 0.200 X 0.100 crystal system triclinic triclinic triclinic monoclinic space group Unit cell dimensions Pi Pi Pi C2/c a (A) 9.6070(3) 8.7652(5) 10.0578(3) 16.5542(5) MA) 10.0320(4) 10.0829(6) 15.0382(5) 15.6748(4) c (A) 12.4170(5) 15.7010(8 15.8120(5) 19.6992(5) a (deg) 106.406(3) 89.527(4) 101.385(3) 90 P (deg) 90.745(3) 80.123(4) 94.968(2) 100.010(2) r (deg) 102.261(2) 77.805(3) 109.331(3) 90° volume (A3) 1118.40(7) 1335.65(13) 2182.30(13) 5033.8 temperature (K) 100(2) 100(2) 120(2) 100(2) Z 2 2 2 4 density (calculated) (g/cm3) 1.401 1.168 1.608 1.501 absorption coefficient (mm *) 1.140 0.951 1.179 1.032 F(000) 480 480 1086 2352 theta range for data collection 3.137—27.484 (deg) 2.484-29.108 2.937-28.725 3.342-27.464 index ranges —12 < h < 12, -13 < fc < 13, -16 < I < 16 -11 < h < 11, -13 < fc < 13, -19 < / < 21 -13 2o(I)] 3985 5912 7821 4521 completeness to theta = 25.242° (%) 99.5 95.4 99.8 99.7 absorption correction none none Gaussian none refinement method full-matrix least-squares on F2 full-matrix least-squares on F2 full-matrix least-squares on F2 full-matrix least-squares on F2 data/restraints/parameters 5095/4/288 6745/2/286 10262/10/655 5747/4/351 goodness-of-fit on F2 1.032 1.039 1.057 1.069 final R indices [/ > 2a(J)] R, = 0.0396, wR2 = 0.0906 R, = 0.0427, wR2 = 0.1149 R, = 0.0440, wR2 = 0.1088 R, = 0.0590, wR2 = 0.1476 R indices (all data) R, = 0.0580, wR2 = 0.0978 R, = 0.0498, wR2 = 0.1192 R, = 0.0616, jvR2 = 0.1211 R, = 0.0785, u/R2 = 0.1586 I 50 Crystal Growth & Design__________________________________________ and N2 sorption properties, as well as thermal and hydrolytic stability. ■ RESULTS AND DISCUSSION Synthesis. A family of microporous mixed-ligand two- dimensional (2D) metal-organic frameworks {[Zn2(Xiso)2- (pcih)2]-guest}„ has been prepared using 4-pyridinecarbaldehyde isonicotinoyl hydrazone (pcih), 5-substituted isophthalic acids [H2Xiso, where X= OH (l); CH3 (2); NH, (3); H (4, 4b)] and diverse sources of zinc(Il) cations. In all reactions the in situ deprotonated dicarboxylic acids compensate positive charge of zinc(ll) ions and link them into one-dimensional double chains. The zinc-carboxylate chains are further connected by neutral pcih linkers as evidenced by single-crystal X-ray dif- fraction (SCXRD). Pale yellow crystals suitable for SCXRD (except 3) were grown in a sealed vessel by heating the solution containing N,N-dimethylformamide (DMF)/water (1—4) or N,N-dimethylacetamide (DMA)/water (4b), appropriate ligands and Zn(N03)2-6H20 (Scheme 1 and Experimental Section). Alternatively, all coordination polymers (l—4b) can be obtained quantitatively in a mechanochemical reaction where Zn(CH3C00)2-2H20, pcih and H2Xiso are ground together with a small amount of DMF or DMA (known as LAG - liquid assisted grinding^4,55). Apart from acting as a diffusion facilitat- ing medium, the additive molecules fill the voids of frameworks 1—4b upon their formation. This procedure requires a small ratio of liquid volume to weight of reactants (rj)s6 within the range of 0.80—0.82 fjL/mg and is completed in a short period of time (approximately 15 min). Tuning of mechanochemical synthetic conditions in the case of 4 was described earlier in detail.57 The comparison of the two synthetic approaches clearly shows that mechanosynthesis meets the requirements of green chemistry,58-60 but does not lead to the formation of single crystals of 1—4b suitable for SCXRD. The purity of all microporous mixed-linker MOFs, representing rare coordina- tion polymers of the type [Zn2(Xiso)2(lig)2]„ -S0'61-64 (fog = neutral ligand), was confirmed by elemental analysis (see Experimental Section), infrared spectroscopy (FT-IR), as well as powder X-ray diffraction (PXRD) (see Figure Si). Structure. A series of metal-organic frameworks [Zn2(Xiso)2(pcih)2]„ (1—4b) were prepared using a neutral acylhydrazone ligand (pcih) bearing both N—H and C=0 functional groups, appropriate isophthalic acid H2Xiso and initial zinc(II) reactant. Single-crystals of 1, 2, 4, and 4b suitable for X-ray diffraction were selected from the bulk. The crystal- lography data and structure refinement parameters are shown in Table 1. The frameworks 1, 2, and 4 crystallize in a triclinic PI space group, wherein mere exchange of DMF guest molec- ules in the latter (4) with DMA leads to monoclinic C2/c system (4b). All of the mixed-linker frameworks have the same layered topology. In the case of 3, its isostructurality with the rest of MOFs was confirmed by PXRD (Figure Slg). In each of the frameworks, the //3-/c2,#c1,#c1-X-iso2- anions links neighboring nodes of Zn2 clusters into one-dimensional double chains, and the chains are further pillared by pcih ligands which increases the coordination network dimensionality to 2D (Figure l). The layers are arranged in an interdigitated ABAB manner and form porous structures with interlayer voids occupied by solvent molecules. The coordination geometry of zinc atoms in each cluster is disordered octahedral wherein equatorial posi- tions are occupied by oxygen atoms from three different Xiso2- ligands and two N pyridyl atoms of /v2-pcih ligands completing the coordination sphere through axial coordination (Figure la). Figure 1. Crystal structure of the {[Zn2(Xiso)2(pcih)2] -guest},, family (l, 2, 4, 4b). (a) Coordination environment within Zn2 cluster with atom labeling scheme and 30% displacement ellipsoids exemplified for 1. (b) Zinc-isophthalate ribbon chains, (c) Solvent accessible voids were calculated with Mercury software by using a probe molecule with a radius of 1.2 Â: zero-dimensional pores exemplified for 1, also observed for 4, 4b (top); and one-dimensional channels for 2 (bottom). H atoms were omitted for clarity. The Zn(l)—N(ll) and Zn(l)—N(l2) in all MOFs are of almost identical lengths (Table Si), unlike Zn—O bonds where Zn(l)—0(71) is elongated as compared to the other zinc—oxygen bonds. However, the Zn(l)—0(71) bond in 2 is considerably shorter than in other frameworks (Table Si) which is strictly associated with its noninvolvement in hydrogen bonding, observed for all other MOF structures (Figure 2). The arrangement of alternating AB layers in each framework is governed by supramolecular interactions between the adjacent layers as well as between a layer and interlayer solvent molecules. These interactions are responsible both for out-of- plane displacements in the zinc-isophthalate chains (Figure l) as well as for the relative arrangement of their planes of pro- pagation between adjacent layers (Figure 2b). The separations of these planes range from 3.4/11.6  for 2 through 6.7/9.0 À in 1 and 4 to equidistant 7.8/7.8  for 4b. In the latter there are no direct interlayer interactions and the layers are posi- tioned through strong framework-guest (N—H)pdh-"Oguest and (O—H) t-0110[lhthlllK hydrogen bonds [d(N19-099) = 2.855(4) A, angle 173(4)°; d(099-071#7) = 2.788(4) A, angle 175(5)°]. Contrarily, in the rest of frameworks direct interlayer inter- actions are observed. In 1 and 4 adjacent sheets interact via strong (N-H)pcih-Oisophthalatc hydrogen bonds [d(N39-07l) = 2.820(3) À, angle 172(4)° for 1; and d(N39-07l) = 2.830(3) À, angle 172(4)°, d(N19-088#7) = 2.858(3) Â, angle 172(4)° for 4] between acylhydrazone NH group and carboxylate oxygen (Figure 2). Additionally in 1 there is a strong (O—H)Xiso- -Opcih hydrogen bond [d(067—018#7) = 2.797(4) À, angle 139(3)°] involving polar OH substituent of isophthalate linker and acylhydrazone oxygen atom. The largest deviation from the equidistant separation between the planes of propagation of Zn-Xiso chains is observed in the structure of 2 based on 5-methylisophthalate ions (3.4/11.6 À). It can be explained by both steric effect as well as polarity of isophthalate substituent. The bigger nonpolar methyl group separates layers, interacting only by weak CH3—n forces with a pyridine ring of pcih from adjacent layer {d[C62-centroid(NllC12C13C14C15Cl6)] = 3.88 Ä}, whereas polar OH groups (in l) are engaged in for- mation of strong hydrogen bonds as donors to an oxygen atom DOI: 10.1021 /acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 490 I 51 Figure 2. Supramolecular interactions involving adjacent layers in the {[Zn2(Xiso)2(pcih)2]-guest}n family (l, 2, 4, 4b): (a) schematic representation of intermolecular interactions involving two adjacent layers and solvent molecules; (b) arrangement of adjacent layers with separation distances between the planes including zinc-isophthalate chains (H atoms and guest molecules are omitted). Table 2. Selected FTIR Absorption Bands for 1—4brt ‘’The data for pcih are given for comparison. Given as wavenumbers in cm \ of an acylhydrazone group, mentioned above. This effect is also (Figure S2). The broad signal in the range 3572—3345 cm-1 responsible for different dimensionality of pores in the struc- with a maximum at 3462 cm-1 for 4b is associated with vibra- tures: MOF 2 has one-dimensional bottled-neck type channels tions of disordered water molecules occupying interlayer voids, propagating along the a axis, as opposed to zero-dimensional Similar signal is observed for 2 (3572—3345 cm-1 with a maxi- cage-like pores observed in other structures (Figure l). mum at 3503 cm-1). The appearance of an additional strong The infrared spectra of all MOFs {[Zn(Xiso)2(pcih)2]- band, which was observed in the characteristic range for car- guest}„ (l—4b) (Figure S2) show several characteristic absorp- bonyl groups at 1667—1670 cm-1 and disappeared after the tion bands corresponding to 5X-l,3-benzenodicarboxylate ions frameworks were thermally activated, confirms the presence of Xiso2-, pcih ligands and guest molecules (H20, DMA, DMF). guest DMF molecules in the investigated MOFs (Figure 6a). The signals in the range 1558—1557 cm-1 and 1385—1399 cm-1 Analogous assignments have recently been reported for two are associated with asymmetric and symmetric vibration of car- interpenetrated 3D frameworks {[Zn2(pdcx)2(pcih)2]-guests}n boxylates, respectively (Table 2). The C=N absorption band, based on para-dicarboxylic acids (pdcx = 1,4-benzenedicarbox- observed at 1605 cm-1 in the free pcih molecule, is slightly ylate or 4,4'-biphenyldicarboxylate).6> changed for 1—4b (1611 — 1615 cm-1), as a result of hydrogen Stability. Thermogravimetric analysis carried out for 1—4 bonding interactions. Similar observation is found for carbonyl revealed for each MOF a stepwise weight loss with a distinct L'(CO)pcih stretching vibrations. The most significant difference plateau in the range dependent on the substituent of iso- (above 200 cm-1) between free (3445 cm-1) and MOF incor- phthalate linker: OH 230—350; CH3 220—290; NH2 280—340; porated acylhydrazone (in the range 2195—3204 cm-1) is and H 200—320 °C (Figure 3). The first step leading to the observed for the position of amide N—H bands. These redshifts plateau corresponds to the loss of guest molecules: two DMF are caused by formation of hydrogen bonds between N—H and one H20 for 1 (found: 15.6%, calcd weight loss: 14.8%); groups and isophthalate oxygen (l, 4, 4b) or guest molecules (2). two DMF and two H20 for 2 (found: 16.1%, calcd weight loss: For MOFs 1, 2, and 3 additional bands of medium inten- 16.2%); three DMF for 3 (found: 18.7%, calcd weight loss: sity are observed at 3195, 3064, and 3421, 3344 cm-1 which are 18.9%); and two DMF for 4 (found: 13.9%, calcd weight loss: ascribed to OH, CH3, and NH, tinker substituents, respectively 13.8%); all molecules given per one Zn2 node unit. The next 491 DOI: 10.1021/acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 Crystal Growth & Design V (NH) * (co)pdh V (CO)DMF/DMA. * (CNO,** V (COO)B V (COO)^ 1 OH 3195 m 1683 s 1668 s 1614 m 1558 m 1385 s 2 CH3 3202 m 1684 m 1670 s 1615 s 1557 s 1386 s 3 NH2 3191 m 1682 m 1667 s 1614 m 1557 s 1385 s 4 H 3207 m 1687 s 1667 s 1611 s 1557 s 1391 s 4b 3204 m 1691 m *1647 m 1611 s 1558 s 1394 s pcih 3445 m 1687 s 1605 w I 52 Crystal Growth & Design Figure 3. Thermal stability of {[Zn2(Xiso)2(pcih)2]-guests},, (1—4). (a) TG and dTG curves, (b) temperature-dependent PXRD patterns. step with a dTG minimum at 370 (l), 323 (2), 364 (3), and unsubstituted isophthalate linkers (4) through extra involvement 348 °C (4) is associated with the frameworks decomposition. in interlayer hydrogen bonding. In contrast, adjacent layers of 2 Main essential factors related to thermal stability of the studied are connected only by weak CH3-;r interaction, which results in MOFs, i.e., intralayer node-linker bond strengths and the num- the lowest decomposition temperature of all frameworks, ber of linkers connected to each Zn2 node, are the same for all Variable-temperature PXRD patterns recorded for 1—4 frameworks. However, the degradation temperatures correlate (Figure 3) indicate that frameworks with polar substituents with the type and number of intermolecular forces involving (l, 3) retain their crystallinity after removing guest molecules adjacent layers (Figure 2). Polar substituents (OH, NH2) increase before degradation process starts, whereas 2 and 4 lose long degradation temperatures as compared to the framework with distance ordering below the decomposition temperature as 492 DOI: 10.1021/acs.cgd.7b01468 Cryst Growth Des, 2013, 18, 488-497 I 53 Figure 5. Adsorption isotherms of C02 at 195 K for {[Zn2(Xiso)2(pcih)2]}„ (1—4). (a) Solid and open symbols represent adsorption and desorption branches, respectively, (b) Semilog plot indicating the differences in low pressure region. 1 - blue, 2 - red, 3 - yellow, 4 - black. 493 DOI: 10.1021/acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 indicated by TG measurements (approximately at 140 (2) and at temperatures much below the onset of thermal degradation 290 °C (4), respectively). On the other hand, guest removal of the framework, demonstrated by thermogravimetric studies, experiments, carried out for 2 and 4 at the temperatures where Since the latter, along with hydrolytic stability, is mostly TGA data (Figure 3) indicated solvent removal only, followed governed by the strength of metal-linker coordination bonds, by cooling to room temperature, lead to crystalline phases, dif- the framework 2 exhibits the lowest degradation temperature as ferent from the initial ones (Figure 4). This indicates structural the one with the longest Zn—O bonds (except the bond with ------------------------------------------------------------------ oxygen atom that is involved in hydrogen bonding in the rest of frameworks). Similarly, the comparison of Zn—O metal-linker . 2des bonds (excluding carboxylate oxygen atoms involved as H-bond —11 a ------------------------— --------------- acceptors) clearly shows the shortest lengths for 1 which is in agreement with its highest thermal and hydrolytic stability. I I J,i I aJI A A a 238 Sorption Properties. Figure 5 shows C02 equilibrium —^ adsorption and desorption isotherms measured at 195 K after thermal activation of the materials 1—4 in a vacuum at 180 °C. • Sorption analysis reveals that all activated MOFs (l—4) selec- It a . a „ /L . tively adsorb smaller C02 (kinetic diameter = 3.30 Â) over larger N2 (kinetic diameter = 3.64 Â). No significant uptake of i I i I J I . 4as nitrogen could be measured at 77 K up to 1 bar. In general, 1 1 * 1 * 1--------------------------------- selective adsorption for such interdieitated frameworks can be 5 15 25 35 1 • 1 • C • I ■ f 1 • ■ 2e[°] explained in terms or size or polarity or pores or as a kinetic issue.39'61-64,66_69 According to the Zeo++ calculations70 Figure 4. PXRD patterns for 2 (above) and 4 (below). Black: / ^ 11 .» \ « , 1 . > . , 0 . . , . , \ ,,,,,,, , (see lable 3), both pore window size and maximum pore as-synthesized materials (2as and 4as; and red: thermally desolvated .. - . , - „ - . , . * r 1 1 j . nT j j ai \ diameter tor 2 are the largest or all frameworks (4.12 and 7.20 frameworks cooled to RT (2des and 4des). . . 0 v ------------------------------------------------------------------ A, respectively). This approximate computation carried out for flexibility of both frameworks, however, with retention of the as-synthesized framework 2 indicates that N2 molecules layered topologies which was confirmed by IR spectra showing should be able to enter its pores, at least in the as-made structure. intact skeletal vibrations during thermal activation and cooling The rest of the frameworks, exhibiting the same selectivity, have (see Figure 6a below). considerably smaller pore window sizes than 2 (Table 3), and Additional chemical stability tests for the frameworks 1—4 these are even smaller than kinetic diameter of C02. Therefore, were performed in water where the materials remained immersed it seems that in these cases contact with C02 adsorptive for 4 days at ambient conditions. The PXRD patterns of all molecules triggers the increase of interlayer pore sizes allowing MOFs after such treatment clearly demonstrate that the frame- for internal penetration. In the case of frameworks 2 and 4 works with polar substituents on isophthalate linkers (l and 3) structural changes have been observed upon thermal activation remain unchanged, whereas 2 and 4 undergo structural as well as upon immersion in water. Thus, in general, selec- changes to new crystalline phases (Figure S3 in Supporting tive C02 versus N2 adsorption behavior in the family of Information). Complementary IR spectra for 2 and 4 show only {[Zn(Xiso)2(pcih)2]}„ MOFs (1—4) can be explained by subde band shifts within the region of skeletal vibrations, which carbon dioxide affinity toward acylhydrazone groups decorating confirms the retention of 2D framework structures during condi- the framework pores. tioning in water. Thus, similarly as in the case of thermal stability, As we reported in our previous paper,s the C02 adsorption polar substituents are also responsible for the higher stability of curve for 4 has a double step profile with a wide hysteresis the layered Zn-isophthalate/acylhydrazone frameworks, through caused by the guest-induced framework response (Figure 5; their involvement in strong interlayer hydrogen bonding. black line). The first distinct plateau is associated with filling The weakest interlayer interactions, found for 2, are responsible the pore volume of 4 (theoretical/experimental pore volume for the loss of long-range ordering (shown by TD-PXRD) 0.090/0.087 cm3 g_1, Table 3). The crystal structures of 1 and 3 Crystal Growth & Design I 54 Crystal Growth & Design____________________________________________________________________________________________________________________________Q Table 3. Porosity Parameters for {[Zn2(Xiso)2(pcih)2]}„ ( 1 —4) Compared with in Situ IR Spectroscopy during C02 Adsorption" MOF pm [A] mpd [A] Vpt [cm3-g_l] Vpt [cm3-g_1] v (C-O) [cm-1] fwhm [cm-1] 1 2.07 4.17 0.090 0.170 2335 4.6 2 4.12 7.20 0.145 0.108 2334 5.0 3 0.168 2333 4.5 4 1.80 4.15 0.090 0.087'’/0.163 2337 3.3 aPvn> Pore windows size; mpdl maximum pore diameter; Vpv theoretical pore volume; VpC, experimental pore volume (based on limiting uptake in the adsorption isotherm). pws> wpd and Vpt were calculated in Zeo+2 software. C—O) - IR absorption bands recorded during C02 adsorption at 213 K: I/ (C—O) wavenumbers in the C—O vibrations region. FWHM- full width at half maximum of the v{Q—O) bands. bPore volume derived from the first plateau at p/p0 = 0.17 (gas uptake 49 cm3g_1). Figure 6. In situ IR spectra for 1—4: thermally activated material (black line), during C02 adsorption at 213 K (red line) and difference curve (green line), (a) In the 400—4000 cm-1 range (b) in the region of C—O vibration. (isostructurality based on PXRD data) bear a close resemblance larger (4.12 Â) than the kinetic diameter of C02 (3.30 À). to that of 4 (Figures 1 and 2): with zero-dimensional pores Simultaneously it has the highest theoretical pore volume between hydrogen-bonded layers in the same arrangement, and (0.145 cm3g-1) of all MOFs in the group (Table 3). Surpris- the same value of theoretical pore volume (0.090 cm3-g-1, ingly though, the experimental pore volume calculated for Table 3), therefore should have similar adsorption properties. {[Zn2(CH3iso)2(pcih)2]}„ (2) based on the maximum amount However, as shown in Figure 5, the adsorption behavior differs of adsorbed CO^ Vpe = 0.108 cm3-g-1, is considerably lower, for all compounds, reflecting the differences in interaction of It suggests that the structural change the framework 2 undergoes C02 with linker substituents. The most polar groups in com- upon thermal activation and cooling, prior to adsorption, is pounds 1 and 3 cause the filling of the pore at lower relative associated with a decrease of available voids (Figures 3 and 4). pressure in comparison to unsubstituted compound 4, resulting The ability of 2 to undergo a continuous structural trans- in type I like isotherms. The stronger affinity to NH2 in compar- formation is also manifested in a slope of adsorption branch of ison to OH is also postulated in the step in the isotherm of 1 the isotherm as well as hysteresis (Figure 5), in contrast to the within the p/p0 range of 0.052—0.172, absent in the isotherm of 3. rest of {[Zn2(Xiso)2(pcih)2]}„ frameworks. This behavior indi- The amount of C02 adsorbed by materials 1, 3, and 4 cates that C02 difiusion is the limiting step during the adsorption reaching nearly the same value at p/p0 = 0.99 (l: 94.1 cm3-g_1; on activated 2, which also results in a lower uptake of C02. 3: 95.5 cm3-g-1; 4: 92.4 cm3 g-1) indicates the presence of To get a deeper insight into the adsorption process, the additional accessible space for each framework {V?c, Table 3) as in situ IR spectra before and during adsorption of C02 on compared to pore volumes based on crystal structures of activated samples of 1—4 (after water and DMF removal) were as-made materials (Vpt, Table 3). However, the energy required measured. New bands appearing in the IR spectra (Figure 6 for opening the pores in 4 is much higher than in 1 and 3, and Table 2) of 1—4 upon C02 addition, at 2335, 2334, 2333, which is manifested in a distinct intermediate plateau on the and 2337 cm-1, respectively, clearly prove that MOFs after adsorption curve of 4. evacuation of guest molecules adsorb C02 at —60 °C. Free carbon On the other hand, of all studied MOFs, only the framework dioxide exhibits IR maximum at a higher wavenumber 2 possesses one-dimensional pores, and their window size is (2345 cm-1), which indicates a slight weakening of C—O 494 DOI: 10.1021/acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 I 55 Crystal Growth & Design_________________________________________ bonds upon adsorption, probably by interaction with acylhydra- zone groups as well as with substituents of isophthalate ions. The strength of interaction is dependent on the substituent. As shown in Figure 6b and Table 2, the lowest deviation of v{C—O) (8 cm-1) compared to free C02 is observed for the framework based on unsubstituted isophthalate, whereas the highest (12 cm-1) is observed for the one with NH2 sub- stituents. Furthermore, the frill width at half-maximum values (fwhm) of the i/(C—O) band in the spectra of 1—3 also indicate that the isophthalate substituents are engaged in interactions with C02 adsorptive. Significantly wider signals observed for 1—3 (fwhm = 4.6; 5; 4.5 cm-1, respectively) compared to 4 (fwhm = 3.3 cm-1) can be explained by the increase of the number of intermolecular forces between the framework and adsorbed C02. The lack of distinct l3C02 signal, obscured by the main wide band of indicates the presence of nonspecific adsorbate-adsorptive interactions.-6 In addition, the IR spectra of materials 1—4 after evacuation of guest molec- ules as well as upon C02 adsorption do not change intralayer coordination bonds, which is confirmed by retention of char- acteristic bands related to their skeletal vibrations. Collectively, all above-mentioned observations support the hypothesis that the C02 adsorption occurs in free interlayer spaces of the frameworks, and the substituents are involved in the interac- tions with the adsorptive. ■ CONCLUSION We have synthesized a series of new 2D mixed-linker coordina- tion polymers with an acylhydrazone linker and with either substituted or nonsubstitued isophthalates. The results reveal that all of the MOFs are obtainable by both solution and mechanochemical synthetic methods. Computational data and X-ray structural analysis explained the role of amide groups in selective adsorption of C02 over N2. In this study we have also demonstrated that exchange of linker substituents lining the pore walls of the frameworks considerably changes interlayer supramolecular interactions and thus layers arrangement, which further entails change of pore volumes, shape, dimensionality, and chemical affinity. As a result the linker substituent affects both thermal and chemical stability of the frameworks as well as adsorption properties. Introducing polar groups on the framework walls increases their stability and affinity to carbon dioxide. The findings provide further instructions for the design of layered, interdigitated frameworks demonstrating selective adsorption and desirable stability. ■ EXPERIMENTAL SECTION 4-Pyridinecarbaldehyde isonicotinoyl hydrazone (pcih) and {[Zn2(iso)2- (pcih),]-2DMF}n were prepared according to the published meth- ods.^ ’ 1 All other reagents and solvents were of analytical grade (Sigma- Aldrich, POCH, Polmos) and were used without further purification. Carbon, hydrogen, and nitrogen were determined by conventional microanalysis with the use of an Elementar Vario MICRO Cube elemental analyzer. IR spectra were recorded on a Thermo Scientific Nicolet iSlO FT-IR spectrophotometer equipped with an iD7 diamond ATR attachment Thermogravimetric analyses (TGA) were performed on a Mettler- Toledo TGA/SDTA 85 Ie instrument at a heating rate of 5 °C min-1 in the temperature range of 25—600 °C (approximately sample weight of 50 mg). The measurement was performed at atmospheric pressure under flowing argon. In situ IR spectra were recorded on a Bruker Tensor 27 spec- trometer equipped with an MCT detector and working with the spectral resolution of 2 cm-1. The samples were activated in the form of self-supporting wafers for 1 h at 180 °C prior to the adsorption of probe molecules at —60 °C for C02 (Linde Gas Polska, 99.95% used without further purification). PXRD patterns were recorded at room temperature (295 K) on a Rigaku Miniflex 600 diffractometer with Cu—Ka radiation {X = 1.5418 A) in a 20 range from 3° to 45° with a 0.05° step at a scan speed of 2.5° min-1. Temperature-dependent powder X-ray diffraction experi- ments were performed using Anton Paar BTS 500 heating stage from 30 to 390 °C with a 10 °C step. At each temperature samples were conditioned for 10 min prior to the measurement. Nitrogen and carbon dioxide adsorption studies were performed on Autosorb iQ_ adsorption apparatus (Quantachrome); 77 K was achieved by liquid nitrogen bath, and 195 K was achieved by dry ice/ isopropanol bath. Crystallographic Data Collection and Structure Refinement. Diffraction intensity data for single crystal of three new compounds (l, 2, and 4b) were collected at 100 K on a KappaCCD (Nonius) diffractometer with graphite-monochromated Mo Ka radiation {X = 0.71073 A). Cell refinement and data reduction were performed using firmware. ' Positions of all of non-hydrogen atoms were determined by direct methods using SIR-97. 4 All non-hydrogen atoms were refined anisotropically using weighted full-matrix least-squares on F2. Refinement and further calculations were carried out using SHELXL 2014/7. 75, 6 All hydrogen atoms joined to carbon atoms were posi- tioned with idealized geometries and refined using a riding model with L7iso(H) fixed at 1.5 l/eq of C for methyl groups and 1.2 Ueq of C for other groups. The hydrogen atoms of one water (099) molecules of 4b (for the second water molecule (098) hydrogen atoms are indeterminate) and H atoms attached to the N atoms were found in the difference-Fourier map and refined with an isotropic thermal parameter. However, the crystal structure data of 1 and 2 show that the DMF solvents molecules are heavily disordered and were removed using the SQUEEZE procedure implemented in the PLATON package. ' In the case of 4b the atoms of one of the two H20 sol- vent molecules were refined using DFIX and DANG instructions. The figures were made using Mercury software. CCDC 1580226, 1580227, 1451728, and 1580228 contain the supplementary crystallo- graphic data for 1, 2, 4, and 4b. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. Syntheses. Synthesis of {[Zn2(OHiso)2(pcih)2]-2DMFH20}n in Solution (1). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Zn(N03)2-6H20 (89.2 mg, 0.300 mmol), and 5-hydroxy-l,3-benzenedicarboxylic acid (H2OHiso) (54.6 mg, 0.300 mmol) were dissolved in DMF (15 mL) and H20 (2.0 mL) by sonification (60 s) and heated at 70 °C for 72 h. Yellow crystals of 1 were filtered off, washed with DMF, and dried in oven at 60 °C and 500 mbar for 0.5 h. Yield: 105.3 mg (63.4%). Anal. Calc, for C^H^NnAsZnz: C 49.88, H 4.00, N 12.64%. Found: C 50.76, H 4.00, N 13.24%. FTIR (ATR, cm'1): ^(OH) 3358m, i/(NH) 3195m, i/(C=0)pcih 1683s, */(C=0)DMF 1668s, *(C=N)pcih 1614m, 1/(000),, 1558s, */(COO)s 1385s. Synthesis of {[Zn2(Meiso)2(pcih)2]-2DMF-2H20}n in Solution (2). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Zn(N03)2-6H20 (89.2 mg, 0.300 mmol), and 5-methyl- 1.3-benzenedicarboxylic acid (H2Meiso) (54.0 mg, 0.300 mmol) were dissolved in DMF (16 mL) and H20 (1.8 mL) by sonification (60 s) and heated at 70 °C for 96 h. Yellow crystals of 2 were filtered off, washed with DMF, and dried in oven at 60 °C for 0.5 h. Yield: 65.0 mg (38.6%). Anal. Calc, for C48HS0N10O14Zn2: Anal. Calc, for C48H50N10O14Zn2: C 51.39, H 4.49, N 12.49%. Found: C 50.66, H 4.19, N 11.91%. FTIR (ATR, cm"1): i/(NH) 3202m, */(C=0)pcih 1684m, i/(C=0)DMF 1670s, v(C=N)pcih 1615s, 4/(000)^ 1557s, i/(COO)s 1386s. Synthesis of {[Zn2(NH2iso)2(pcih)2]-3DMF}n in Solution (3). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Zn(N03)2-6H20 (89.2 mg, 0.300 mmol), and 5-hydroxy- 1.3-benzenedicarboxylic acid (H2NH2iso) (54.3 mg, 0.300 mmol) were dissolved in DMF (15 mL) and H20 (2.0 mL) by sonification (60 s) and heated at 70 °C for 60 h. Yellow-brown crystals of 3 were filtered DOI: 10.1021 /acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 495 I 56 Crystal Growth & Design off, washed with DMF, and dried in oven at 60 °C and 500 mbar for 0.5 h. Yield: 126.3 mg (72.5%). Anal. Calc, for C49HslN13013Zn2: C 50.70, H 4.43, N 15.69%. Found: C 49.97, H 3.97, N 15.28%. FTIR (ATR, cm-1): i/(NH) 3409w, i/(NH) 3341w, v{NH) 3191m, v(C=0)pcih 1682m, i/(C=0)DMF 1667s, i/(C=N)pcih 1614m, i/(COO)as 1557s, */(COO)s 1385s. Synthesis of {[Zn2(iso)2(pcih)2]-2DMA-4H20}n in Solution (4b). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Zn(N03)2-6H20 (89.2 mg, 0.300 mmol), and 1,3-benzenedicarboxylic acid (H2iso) (49.8 mg, 0.300 mmol) were dissolved in DMA (16 mL) and H20 (1.2 mL) by sonification (60 s) and heated at 50 °C for 7 days Yellow crystals of 4b were filtered off, washed with DMA, and dried in oven at 60 °C and 500 mbar for 0.5 h. Yield: 57.0 mg (32.8%). Anal. Calc, for C^HjoN^O^Zn^ C 51.39, H 4.49, N 12.49%. Found: C 52.34, H 4.06, N 12.67%. FTIR (ATR, cm-1): t-(NH) 3204m, ^(C=0)pdh 1691m, i'(C=0)DMA 1647m, i/(C=N) dh 1611s, ^COO^ 1558s, ^(COO)s 1394s. Mechanosyntnesis of {[Zn2(Xiso)2(pcih)2]-(Guests)} n (l-4b). 4-Pyridinecarbaldehyde isonicotinoyl hydrazone (pcih) (45.2 mg, 0.200 mmol), Zn(CH3C00)2-2H20 (43.9 mg, 0.200 mmol), and 5-X-1,3-benzenedicarboxylic add (H2Xiso) (0.200 mmol, weights given in Table 4) were ground together with 100 //L DMF in a vibrating mill (4 agate balls, 9.5 mm diameter) for 20 min with 15 Hz frequency. The same experiment was repeated in an agate mortar by manual grinding until liquid additive evaporated (grinding times given in Table l). Table 4. Mechanosynthetic Conditions Used in Preparations of 1—4b (Manual Grinding) MOF "imxiso [mg] time [min] t] [//L-mg-1] 1 36.4 13 0.80 2 36.0 12 0.80 3 36.2 10 0.80 4b 33.2 11 0.82 ■ ASSOCIATED CONTENT G Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.cgd.7b01468. IR spectrum, PXRD patterns, TGA data (PDF) Accession Codes CCDC 1580226—1580228 and 1451728 contain the supple- mentary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/ cif, or by emailing data_request(S)ccdc.cam.ac.uk, or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033. ■ AUTHOR INFORMATION Corresponding Author *E-mail: dariusz.matoga(5)uj.edu.pl. ORCID ® Irena Senkovska: 0000-0001-7052-1029 Stefan Kaskel: 0000-0003-4572-0303 Dariusz Matoga: 0000-0002-0064-5541 Notes The authors declare no competing financial interest. ■ ACKNOWLEDGMENTS The National Science Centre (NCN, Poland) is gratefully acknowledged for the financial support (Grant No. 2015/17/ B/ST5/01190) of this research. The research was carried out partially with the equipment purchased thanks to the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (Contract No. POIG.02.01.00-12-023/08). ■ REFERENCES (1) Silva, P.; Vilela, S. M. F.; Tome, J. P. C.; Almeida Paz, F. A. Chem. Soc. Rev. 2015, 44, 6774-6803. (2) Ren, J.; Dyosiba, X.; Musyoka, N. M.; Langmi, H. W.; Mathe, M.; Liao, S. Coord. Chem. Rev. 2017, 352, 187—219. (3) Kang, Z.; Fan, L.; Sun, D. /. Mater. Chem. A 2017, 5, 10073— 10091. (4) Zhang, Z.; Yao, Z.-Z.; Xiang, S.; Chen, B. Energy Environ. Sei. 2014, 7, 2868-2899. (5) Hefti, M.; Joss, L.; Bjelobrk, Z.; Mazzotti, M. Faraday Discuss. 2016, 192, 153-179. (6) Yu, J.; Xie, L.-H.; Li, J.-R; Ma, Y.; Seminario, J. M.; Balbuena, P. B. Chem. Rev. 2017, 117, 9674-9754. (7) Chughtai, A H.; Ahmad, N.; Younus, H. A; Laypkov, A; Verpoort, F. Chem. Soc. Rev. 2015, 44, 6804—6849. (8) Nath, I.; Chakraborty, J.; Verpoort, F. Chem. Soc. Rev. 2016, 45, 4127-4170. (9) Dhakshinamoorthy, A; Opanasenko, M.; Cejka, J.; Garcia, H. Catal. Sei. Technol 2013, 3, 2509-2540. (10) Gascon, J.; Corma, A; Kapteijn, F.; Llabrés i Xamena, F. X. ACS Catal. 2014, 4, 361-378. (11) Wyszogrodzka, G.; Marszalek, B.; Gil, B.; Dorożyński, P. Drug Discovery Today 2016, 21, 1009—1018. (12) Horcajada, P.; Gref, R.; Baati, T.; Allan, P. K.; Maurin, G.; Couvreur, P.; Férey, G.; Morris, R. E.; Serre, C. Chem. Rev. 2012, 112, 1232-1268. (13) Falcaro, P.; Ricco, R; Doherty, C. M.; Liang, K; Hill, A J.; Styles, M. J. Chem. Soc. Rev. 2014, 43, 5513—5560. (14) Morozan, A; Jaouen, F. Energy Environ. Sei. 2012, 5, 9269— 9290. (15) Silva, C. G.; Corma, A; Garcia, H. J. Mater. Chem. 2010, 20, 3141-3156. (16) Li, S.-L.; Xu, Q. Energy Environ. Sci. 2013, 6, 1656—1683. (17) DeCoste, J. B.; Peterson, G. W. Chem. Rev. 2014, 114, 5695— 5727. (18) Lustig, W. P.; Mukherjee, S.; Rudd, N. D.; Desai, A V.; Li, J.; Ghosh, S. K. Chem. Soc. Rev. 2017, 46, 3242—3285. (19) Morris, R E.; Brammer, L. Chem. Soc. Rev. 2017, 46, 5444— 5462. (20) Coudert, F.-X. Chem. Mater. 2015, 27, 1905-1916. (21) Schneemann, A; Bon, V.; Schwedler, I.; Senkovska, I.; Kaskel, S.; Fischer, R A Chem. Soc. Rev. 2014, 43, 6062—6096. (22) Yang, F.; Xu, G.; Dou, Y.; Wang, B.; Zhang, H.; Wu, H.; Zhou, W.; Li, J.-R; Chen, B. Nature Energy 2017, 2, 877. (23) Krause, S.; Bon, V.; Senkovska, I.; Stoeck, U.; Wallacher, D.; Toebbens, D. M.; Zander, S.; Pillai, R S.; Maurin, G.; Coudert, F.-X.; Kaskel, S. Nature 2016, 532, 348-352. (24) Sato, H.; Kosaka, W.; Matsuda, R; Hori, A; Hijikata, Y.; Belosludov, R V.; Sakaki, S.; Takata, M.; Kitagawa, S. Science 2014, 343, 167-170. (25) Matoga, D.; Oszajca, M.; Molenda, M. Chem. Commun. 2015, 51, 7637-7640. (26) Matoga, D.; Gil, B.; Nitek, W.; Todd, A. D.; Bielawski, C. W. CrystEngComm 2014, 16, 4959—4962. (27) Matoga, D.; Roztocki, K; Wilke, M.; Emmerling, F.; Oszajca, M.; Fitta, M.; Balanda, M. CrystEngComm 2017, 19, 2987—2995. (28) Stock, N.; Biswas, S. Chem. Rev. 2012, 112, 933—969. (29) Tanabe, K K; Cohen, S. M. Chem. Soc. Rev. 2011, 40, 498—519. (30) Evans, J. D.; Sumby, C. J.; Doonan, C. J. Chem. Soc. Rev. 2014, 43, 5933-5951. (31) Cohen, S. M. Chem. Rev. 2012, 112, 970-1000. (32) Deria, P.; Mondloch, J. E.; Karagiaridi, O.; Bury, W.; Hupp, J. T.; Farha, O. K Chem. Soc. Rev. 2014, 43, 5896—5912. DOI: 10.1021 /acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 496 I 57 Crystal Growth & Design (33) Sholl, D. S.; Lively, R. P. J. Phys. Chem. Lett. 2015, 6, 3437— 3444. (34) Wu, H.; Chua, Y. S.; Krungleviciute, V.; Tyagi, M.; Chen, P.; Yildirim, T.; Zhou, W. J. Am. Chem. Soc. 2013, J35, 10525-10532. (35) Linder-Patton, O. M.; Bloch, W. M.; Coghlan, C. J.; Sumida, K.; Kitagawa, S.; Furukawa, S.; Doonan, C. J.; Sumby, C. J. CrystEngComm 2016, 18, 4172-4179. (36) Pichon, A; Lazuen-Garay, A.; James, S. L. CrystEngComm 2006, 8, 211-214. (37) Mottillo, C.; Friscić, T. Molecules 2017, 22, 144. (38) Julien, P. A; Mottillo, C.; Friscić, T. Green Chem. 2017, 19, 2729-2747. (39) Rubio-Martinez, M.; Avci-Camur, C.; Thornton, A W.; Imaz, I.; Maspoch, D.; Hill, M. R. Chem. Soc. Rev. 2017, 46, 3453-3480. (40) Seoane, B.; Coronas, J.; Gascon, L; Benavides, M. E.; Karvan, O.; Caro, J.; Kapteijn, F.; Gascon, J. Chem. Soc. Rev. 2015, 44, 2421 — 2454. (41) Maina, J. W.; Pozo-Gonzalo, C.; Kong, L.; Schutz, J.; Hill, M.; Dumee, L. F. Mater. Horiz. 2017, 4, 345—361. (42) Wang, C.; Liu, X.; Keser Demir, N.; Chen, J. P.; Li, K. Chem. Soc. Rev. 2016, 45, 5107-5134. (43) Howarth, A J.; Liu, Y.; Li, P.; Li, Z.; Wang, T. C.; Hupp, J. T.; Farha, O. K. Nat. Rev. Mater. 2016, J, 15018. (44) Burtch, N. C.; Jasuja, H.; Walton, K. S. Chem. Rev. 2014, 114, 10575-10612. (45) Lv, X.-L.; Wang, K.; Wang, B.; Su, J.; Zou, X; Xie, Y.; Li, J.-R; Zhou, H.-C. J. Am. Chem. Soc. 2017, 139, 211-217. (46) Wang, K; Lv, X.-L.; Feng, D.; Li, J.; Chen, S.; Sun, J.; Song, L.; Xie, Y.; Li, J.-R.; Zhou, H.-C. /. Am. Chem. Soc. 2016, 138, 914-919. (47) Fukushima, T.; Horike, S.; Inubushi, Y.; Nakagawa, K.; Kubota, Y.; Takata, M.; Kitagawa, S. Angew. Chem., Int. Ed. 2010, 49, 4820— 4824. (48) Horike, S.; Tanaka, D.; Nakagawa, K.; Kitagawa, S. Chem. Commun. 2007, 32, 3395-3397. (49) Tanaka, D.; Nakagawa, K.; Higuchi, M.; Horike, S.; Kubota, Y.j Kobayashi, T. C.; Takata, M.; Kitagawa, S. Angew. Chem., Int. Ed. 2008, 47, 3914-3918. (50) Nakagawa, K.; Tanaka, D.; Horike, S.; Shimomura, S.; Higuchi, M.; Kitagawa, S. Chem. Commun. 2010, 46, 4258—4260. (51) Schneemann, A; Rudolf, R.; Henke, S.; Takahashi, Y.; Banh, H.; Hante, I.; Schneider, C.; Noro, S.; Fischer, R. A Dalton Trans. 2017, 46, 8198-8203. (52) Henke, S.; Schneemann, A; Wütscher, A.; Fischer, R A /. Am. Chem. Soc. 2012, 134, 9464-9474. (53) Makal, T. A; Wang, X.; Zhou, H.-C. Cryst. Growth Des. 2013, 13, 4760-4768. (54) Shan, N.; Toda, F.; Jones, W. Chem. Commun. 2002, 2372— 2373. (55) Friscić, T.; Jones, W. Cryst. Growth Des. 2009, 9, 1621—1637. (56) Friscić, T.; Childs, S. L.; Rizvi, S. A. A.; Jones, W. CrystEngComm 2009, 11, 418—426. (57) Roztocki, K.; Jędrzejowski, D.; Hodorowicz, M.; Senkovska, I.j Kaskel, S.; Matoga, D. Inorg. Chem. 2016, 55, 9663-9670. (58) Anastas, P. T.; Warner, J. C. Green Chemistry, Theory and Practice, Oxford University Press: New York, 1998. (59) Clark, J. H. Green chemistry: challenges and opportunities. Green Chem. 1999, 1, 1—8. (60) Cemansky, R. Nature 2015, 519, 379-380. (61) Horike, S.; Tanaka, D.; Nakagawa, K.; Kitagawa, S. Chem. Commun. 2007, 3395-3397. (62) Fukushima, T.; Horike, S.; Inubushi, Y.; Nakagawa, K.; Kubota, Y.; Takata, M.; Kitagawa, S. Angew. Chem. 2010, J22, 4930—4934. (63) Takashima, Y.; Furukawa, S.; Kitagawa, S. CrystEngComm 2011, 13, 3360-3363. (64) Hijikata, Y.; Horike, S.; Sugimoto, M.; Sato, H.; Matsuda, R.; Kitagawa, S. Chem. - Eur. J. 2011, 17, 5138—5144. (65) Roztocki, K; Senkovska, I.; Kaskel, S.; Matoga, D. Eur. J. Inorg. Chem. 2016, 2016, 4450-4456. (66) Zhang, Z.; Zaworotko, M. J. Chem. Soc. Rev. 2014, 43, 5444— 5455. (67) Prochowicz, D.; Justyniak, I.; Kornowicz, A; Kaczorowski, T.; Kaszkur, Z.; Lewiński, J. Chem. - Eur. J. 2012, 18, 7367—7371. (68) Matsuda, R; Tsujino, T.; Sato, H.; Kubota, Y.; Morishige, K.; Takata, M.; Kitagawa, S. Chem. Sei. 2010, 1, 315—321. (69) Kim, H.; Samsonenko, D. G.; Yoon, M.; Yoon, J. W.; Hwang, Y. K; Chang, J. S.; Kim, K. Chem. Commun. 2008, 39, 4697—4699. (70) Willems, T. F.; Rycroft, C. H.; Kazi, M.; Meza, J. C.; Harańczyk, M. Microporous Mesoporous Mater. 2012, 149, 134—141. (71) Ni, W.-X.; Li, M.; Zhou, X-P.; Li, Z.; Huang, X.-C.; Li, D. Chem. Commun. 2007, 33, 3479-3481. (72) COLLECT; Nonius B.V.: Delft, The Netherlands, 1998. (73) Otwinowski, Z.; Minor, W., Eds. Methods in Enzymology, Vol. 276, Macromolecular Crystallography, Part A; Academic Press: New York, 1997, pp 307-326. (74) Altomare, A; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A; Molitemi, A. G. G.; Polidori, G.; Spagna, R. /. Appl. Crystallogr. 1999, 32, 115. (75) Spek, A. L. Acta Crystallogr. 2015, C71, 9—18. (76) Sheldrick, G. M. SHELXS-97: Program for the Crystal Structure Refinement-, University of Göttingen: Göttingen, Germany, 2008. (77) Sheldrick, G. M. SHELXL-2014/7: Program for the Solution of Crystal Structures; University of Göttingen: Göttingen, Germany, 2014. 497 DOI: 10.1021/acs.cgd.7b01468 Cryst. Growth Des. 2018, 18, 488-497 I 58 I 59 Water-Stable Metal-Organic Framework with Three Hydrogen-Bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-Hydrolytic Stability Autorzy: Kornel Roztocki, Magdalena Lupa, Andrzej Sławek, Wacław Makowski, Irena Senkovska, Stefan Kaskel, Dariusz Matoga Opublikowano w: Inorg. Chem. 2018, 57, 3287-3296 Publikacja IV Downloaded via UNIWERSYTETU JAGIELLOŃSKIEGO on November 9, 2018 at 10:16:00 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. I 60 Inorganic Chemistry © Cite This: Inorg. Chem. 2018, 57, 3287-3296 pubs.acs.org/IC Water-Stable Metal-Organic Framework with Three Hydrogen-Bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-Hydrolytic Stability Kornel Roztocki, Magdalena Lupa, Andrzej Sławek, ' Wadaw Makowski, Irena Senkovska, Stefan Kaskel, and Dariusz Matoga*'+® ^Faculty of Chemistry, Jagiellonian University, Gronostajowa 2, 30-387 Kraków, Poland ^Department of Inorganic Chemistry, Technische Universität Dresden, Bergstrasse 66, 01062 Dresden, Germany Q Supporting Information ABSTRACT : A new microporous cadmium metal—organic framework was synthesized both mechanochemically and in solution by using a sulfonyl- functionalized dicarboxylate linker and an acylhydrazone colinker. The three-dimensional framework is highly stable upon heating to 300 °C as well as in aqueous solutions at elevated temperatures or acidic conditions. The thermally activated material exhibits steep water vapor uptake at low relative pressures at 298 K and excellent recyclability up to 260 °C as confirmed by both quasi-equilibrated temperature-programmed desorption and adsorp- tion (QE-TPDA) method as well as adsorption isotherm measurements. Reversible isotherms and hysteretic isobars recorded for the desorption— adsorption cycles indicate the maximum uptake of 0.19 g/g (at 298 K, up to p/p0 = l) or 0.18 g/g (at 1 bar, within 295—375 K range), respectively. The experimental isosteric heat of adsorption (48.9 kj/mol) indicates non- coordinative interactions of water molecules with the framework. Exchange of the solvent molecules in the as-made material with water, performed in the single-crystal to single-crystal manner, allows direct comparison of both X-ray crystal structures. The single-crystal X-ray diffraction for the water-loaded framework demonstrates the orientation of water clusters in the framework cavities and reveals their strong hydrogen bonding with sulfonyl, acyl, and carboxylate groups of the two linkers. The grand canonical Monte Carlo (GCMC) simulations of H20 adsorption corroborate the experimental findings and reveal preferable locations of guest molecules in the framework voids at various pressures. Additionally, both experimental and GCMC simulation insights into the adsorption of C02 (at 195 K) on the activated framework are presented. ■ INTRODUCTION In recent years a considerable scientific interest has been raised on porous materials, and especially on metal—organic frame- works (MOFs). In particular, this is due to their properties promising for the potential use in various technologies associated with omnipresent water, for example, heat pumps,1-4 water purification/ proton conductivity,6’7 and even water harvesting from air. The remarkable attention toward MOFs is also closely related to their compelling properties such as nearly infinite crystal engineering possibilities combined with high internal surfaces as well as with functionality and responsivity.9’10 Among numerous useful features of MOFs, stability toward moisture is a crucial factor for materials that are candidates for industrial applications. For instance, the archetypical frameworks that have had a big impact on the development of MOFs proved to be unstable toward water (e.g., MOF-5 and its isoreticular series, HKUST- 1, DUT-4).1,-13 For many MOFs, the weakest point is the nature of metal—ligand coordination bonds, and the degrada- tion of these frameworks in hydrothermal conditions usually starts by substitution of a ligand by water molecule, especially among zinc-carboxylate MOFs.1- In the literature, the MOFs with reported behavior upon contact with either liquid or gaseous water are still few and far between. Therefore, there is a need to holistically investigate the correlation between water adsorption properties, such as water loading, heat of adsorption, sorption mechanism, hydrolytic stability, and a framework structure. Taking into consideration the aforementioned factors, in this work we present a versatile approach toward understanding of water adsorption process, thermo-hydrolytic stability, as well as interactions between them and a framework on the example of new highly water-stable mixed-linker Cd-MOF based on 4,4'- sulfonyldibenzoic carboxylate (sdb2-) and 4-pyridinecarbox- aldehyde hydrazone (pcih). This moderately hydrophilic framework {[Cd2(sdb)2(pcih)2]-2DMFH20}„ (l) (DMF = dimethylformamide) can withstand heating to 300 °C as well as Received: January 9, 2018 Published: March 2, 2018 r ACS Publications ® 2018 American Chemical Society DOI: 10.1021 /acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 3287 I 61 Inorganic Chemistry__________________________________________________________________________________________________________ soaking in aqueous solutions at elevated temperatures or acidic suitable for SC-XRD. The framework 1 is the first 3D non- conditions. From the industrial point of view, hydrothermal interpenetrated example of MOFs incorporating both carbox- stability is the indispensable feature for potential applications ylate and acylhydrazone linkers as contrasted to the previously associated with water. For example, commercially available reported layered materials [Zn2(Xiso)2(pcih)2]„18'19 as well as a Basolite 300C (HKUST-l) and Basolite F300 (Fe-BTC) group of 3D interpenetrated MOFs [M2(pcdx)2(pcih)2]„ (M = dramatically lose their capacity after 20 cycles, with loading Cd2+, Zn2+; pcdx = 1,4-benzenedicarboxylate, 2-amino-1,4- reduced to 53% and 74% from initial capacity, respectively.1 In benzenedicarboxylate acid, or 4,4'-biphenyldicarboxylate).20,21 contrast, the water loading for 1 does not change after 35 By use of a V-shaped dicarboxylic acid (H2sdb) the inter- cycles, and the MOF is hydrolytically stable up to 260 °C, penetration could be prevented leading to the largest voids which was confirmed by the quasi-equilibrated temperature- among all carboxylate/acylhydrazone MOFs reported. Simulta- programmed desorption and adsorption (QE-TPDA) method. neously, introducing sulfonyl groups on the pore walls rendered By using X-ray diffraction (XRD) and grand canonical Monte the resulting Cd-MOF moderately hydrophilic through adding Carlo (GCMC) simulations, subtle interactions between the hydrogen-bond acceptors between hydrophobic phenyl rings, framework and water molecules were thoroughly investigated. In this way, together with carboxylate oxygen atoms and the Furthermore, the adsorption properties of [Cd2(sdb)2(pcih)2]„ acylhydrazone carbonyl group, the framework bears three toward C02 (at 195 K) and N2 (at 77 K) gases are discussed various functional groups that are capable of forming hydrogen along with the framework structural features. bonds as acceptors. Structure. SC-XRD revealed that Cd-MOF crystallizes in H RESULTS AND DISCUSSION the monoclinic system (space group P2/c), with one sdb2- General Remarks. A new non-interpenetrated cadmium- anion, one pcih ligand, and one Cd2+ ion in the asymmetric based three-dimensional (3D) metal-organic framework unit. Compound 1 forms an ^ 8-c uninodal net, where {[Cd2(sdb)2(pcih)2]-2DMF-H20}„ (l), composed of two coordination geometry of the Cd2+ion can be described as a various linkers (4,4,-sulfonyldibenzoic carboxylate (sdb2-) and disordered octahedral consisting of four equatorial O 4-pyridinecarboxaldehyde hydrazone (pcih)) was successfully carboxylate donors from three different sdb2' linkers and two synthesized by solvent-based as well as solvent-free methods N-pyridyl atoms from two pcih ligands occupying axial (Scheme 1). In the first approach the carboxylic acid (H2sdb), positions (Figure 1, Figure S2). Symmetrically distributed Scheme 1. Two General Synthetic Routes to the Framework {[Cd2(sdb)2(pcih)2]-2DMF-H20}„ (l) the acylhydrazone (pcih), and cadmium nitrate were heated in a sealed vessel for 4 days in a dimethylformamide (DMF)/water mixture, which led to formation of single crystals suitable for single-crystal (SC) XRD-based crystal structure elucidation. This method, however, cannot be regarded as environmental- friendly, since it requires considerable amount of solvent and energy input. To address this issue, the Cd-MOF 1 can alternatively be prepared quantitatively by one-pot three- component grinding of Cd(OH)2, H2sdb, and pcih in a 1:1:1 ratio in the presence of a small amount of DMF as a liquid additive (Figure Si). Both the added DMF as well as water produced in the course of reaction provide suitable environment and fill the one- dimensional (ID) channels of the assembled Cd-MOF. The comparison of the two synthetic approaches clearly shows that mechanosynthesis meets the requirements of green chem- istry1-1 but does not lead to formation of single crystals of 1 Figure 1. X-ray crystal structure of {[Cd2(sdb)2(pcih)2]-2DMF-H20}„ (l): (a) [Cd2(sdb)2(pcih)2] unit; (b) simplified representation of 2D [Cd2(sdb)2]„ layers (red) connected by pcih ligands (black), (c) ID channels propagating through layers of [Cd2(sdb)2]„; (d) intraframe- work hydrogen bonds (in blue), (a—c) Hydrogen atoms were omitted for clarity. sdb2- ligands and cadmium ions create dinuclear secondary building blocks [Cd2(COO)2] (SBUs) with Cd—Cd distance of 3.960 A in each cluster. The sdb2- ions in Cd-MOF function as //3-/c2/c1/cI angular linkers between Cd2 clusters forming layers of a (4,4) topology. The axial positions of each cadmium atom of a dinuclear node are occupied by N-donor //2-ac1/c1 pcih bridging ligands that link the layers into 3D non-interpenetrated pillared-layered frame- work. Free voids in the structure are ID channels that propagate along the a direction and are decorated by sulfone 3288 DOI: 10.1021/acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 I 62 Figure 3. (a) H20 physisorption isotherms at 298 K (blue) and 308 K (black) for {[Cd2(sdb)2(prih)2]}n; simulated adsorption branch (orange) is given for comparison; solid and open symbols represent adsorption and desorption branches, respectively; vertical dashed line represents the pressure value corresponding to vapor—liquid equilibrium obtained from the Antoine equation at 298 (b) Adsorption isobars of H20 at ca. 2250 Pa. (c) Schematic representation of hydrogen bonds (cyan lines) involving water molecules and three various acceptors located on the framework, (d) Water molecules positions (oxygen atoms represented by red spheres): located in the crystal structure from SC-XRD (big spheres) are overlapped with the most probable positions of guest molecules in the structure at the saturation conditions obtained from GCMC calculations (small spheres). S02 groups of the sdb2- linkers. Coordination sphere of shift of the amide N—H band in IR spectrum to 3217 cm-1 in 1 cadmium centers is a distorted octahedron. The axial Cd—N as compared to free pcih ligand (3190 cm-1). The appearance bonds, COI—N15 and CdOl—N6, are of almost identical of an additional strong band, which was observed in the lengths (2.312(2) and 2.290(2) Â, respectively), unlike characteristic range for carbonyl groups at 1673 cm-1 and equatorial Cd—O bonds where d(Cd01—035) = 2.497(2)  disappeared after the framework was thermally activated at 200 is considerably longer in comparison with the rest of cadmium °C and 50 mbar, confirms the presence of guest DMF oxygen bonds: d(Cd01—027) = 2.325(3) À; d(Cd01—037) = molecules in the investigated MOF (Figures S3 and S4). Other 2.259(2) Â; d(Cd01—035) = 2.225(3) Â. As revealed by SC- characteristic absorption bands corresponding to linkers as well XRD, this is caused by the involvement of the carboxylate as guest molecule are described in Figure S3, oxygen atoms 0004 in strong hydrogen bonds with NH groups C02 and N2 Physisorption. To evaluate the porosity of 1, of the pcih ligands (d(N9 - 035)) = 2.86 À, angle = 164°). The the sample was desolvated, and adsorption of nitrogen (kinetic formation of these hydrogen bonds causes a significant blue diameter = 3.64 A) at 77 K was investigated. According to 3289 DOI: 10.1021/acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 Inorganic Chemistry_________________________________________________________________________________________________Q Figure 2. (a) C02 physisorption isotherms at 195 K (semilog plot) for {[Cd2(sdb)2(pcih)2]}„: experimental (black/red) and calculated (orange); vertical dashed line represents the pressure value corresponding to vapor—liquid equilibrium obtained from the Antoine equation at 195 K.22 (b) Average occupation profiles for C02 at 100 kPa (top) and 1 kPa (bottom), (c) Temperature-dependent in situ IR spectra for {[Cd2(sdb)2(pcih)2]}„. I 63 Inorganic Chemistry_____________________________________________ theoretical pore size distribution calculations (see Molecular Simulations section and Figure S5), the pore diameter in this highly robust MOF (see hydrolytic and thermal stability section, and Figure S4) is ~4.4 A, which explicitly indicates that N2 should be able to enter the ID channels. However, no significant uptake of N2 at 77 K could be measured (up to 30 cm3/g at 77 K and p/po = l). Interestingly, when C02 was applied (kinetic diameter = 3.30 Â) as adsorptive in adsorption experiment at 195 K, significant uptake was observed (Figure 2a, Figure S6). The isotherm is of type la without a hysteretic loop. The uptake at p/p0 = 0.99 amounts to 124 cm3/g corresponding to the total pore volume of 0.22 cm3/g. The C02 molecule, which possesses high quadruple moment value, interacts with the framework through polar —C(0)= N—NH— groups of the pcih ligands as well as moderately with S02 groups from sdb2- anions, as evidenced by GCMC calculations (Figure S7). To get a deeper insight into the process of C02 adsorption, the isotherm at 195 K was calculated by Monte Carlo methods, and sufficient agreement with experiment was obtained. On the one hand, the discrepancy between experimental and calculated isotherm indicates that the chosen model does not perfectly reflect packing of C02 molecules in the studied structure. On the other hand, both isotherms exhibit a steep slope at similar pressure, which indicates that exploited force field correctly mimics guest—host interactions. The average occupation profiles confirm that the most probable location of C02 is in the proximity of polar —C(0)=N—NH— and S02 groups lining the framework walls (red regions in Figure 2b) and thus form the evidence for the key role of the acylhydrazone linker in the observed higher C02 uptake. Analogous behavior with regard to C02 and N2 adsorption was observed for recently reported Zn-acylhydrazone/dicarboxylate-frameworks.18-~1 To get a deeper insight into the process, the in situ IR spectra during adsorption of C02 on the activated material 1 at different temperatures were measured (Figure 2c, Figure S8). In low-temperature region upon addition of CO^ a new band appears in the IR spectra of 1 at 2339 cm-1, which becomes extinct at room temperature. The gradual decrease of the C02 signal with temperature shows that the interactions between carbon dioxide and the framework are rather weak. Free carbon dioxide exhibits IR band at a higher wavenumber (2345 cm-1), which indicates a slight weakening of C—O bonds upon adsorption and interaction with acylhydrazone groups. The lack of distinct 13C02 signal, obscured by the main wide band of 12CO^ indicates the presence of nonspecific adsorbate— adsorptive interactions. The electron density slightly shifts from C02 toward the -C=N—NH- group of the pcih linker, which causes blue shift of the C=N band energy (u = 1614 cm-1) after C02 adsorption as compared to the empty framework (i/(CN) = 1611 cm-1). This is in agreement with calculated C02 occupation profiles (Figure 2b, Figure S7). In addition, neither evacuation of guest molecules nor C02 adsorption change intralayer coordination bonds in the MOF, which is confirmed by retention of characteristic IR bands related to their skeletal vibrations. Collectively, all above- mentioned observations show that higher C02 (at 195 K) over N2 (at 77 K) adsorption in the family of mixed-linker acylhydrazone/carboxylate MOFs is based on the presence of the acylhydrazone -C=N—NH- group. Water Adsorption. The water adsorption was investigated at 298 and 308 K (Figure 3a, Figure S6). The isotherms have a sigmoidal shape, and significant adsorption of water vapor takes place at relative pressures higher than 0.15, postulated as a steeply rising isotherm in the relative pressure range of 0.15— 0.34. Starting from p/p0 > 0.34 a linear adsorption is observed with the maximum water loading reaching 0.19 wt % (232 cm3/ g) at p/po = 0.95. The theoretical pore volume of 1 calculated from helium void fraction (HVF) for the single-crystal structure of the as-made material is 0.192 cm3 g-1 (excluding the solvent) and matches well the experimental value of 0.19 cm3 g-1 observed in the H20 isotherm (298 K) at p/p0 of 0.95. To better understand the adsorption behavior, the isobar in the QE-TPDA experiment was measured (Figure 3b). In isobaric conditions (at ca. 2250 Pa), the maximum amount of water uptake (0.18 wt % equivalent to 220 cm3/g) is in agreement with the isothermal experiment. The observed hysteresis in the isobar within the studied range of temperatures (295—375 K) indicates a different mechanism of adsorption and desorption (Figure 3b). Desorption may deviate from quasi equilibrium condition due to diffusion limitations related to presence of relatively stable H20 clusters within the framework. The adsorption branch of the isobar shows a low increase of water loading upon cooling to T = 325 K, which is followed by an abrupt water uptake in the temperature range of 325—295 K. This observation is in agreement with GCMC indicating that the limiting stage of adsorption process is the formation of small clusters of water molecules within the pores (Figure S9). When the amount of the adsorptive further increases, these clusters start to attract more water molecules, which leads to a considerable filling of the pores. Thus, in both the adsorption isotherms and the isobars, the water loading increases very slowly at the beginning and is followed by an abrupt uptake at either higher pressures or lower temperatures, respectively. To get a deeper insight in the interactions between the framework and the adsorbate, water molecules were localized by SC-XRD for the water-exchanged framework 1H20. The suitable crystal was obtained by slow leaching of DMF guest molecules in the as-made material 1, which was conditioned for 48 h in an environmental chamber at relative humidity (RH) = 90%. Such conditions enabled single-crystal to single-crystal (SCSC) guest exchange and prevented crystal cracking. SC- XRD performed for 1H20 revealed the presence of a cluster of five water molecules strongly hydrogen-bonded to three oxygen atoms of various functional groups of the linkers: cadmium- coordinated carboxylate, carbonyl group of the acylhydrazone pcih linker, and sulfonyl group of the 4,4'-sulfonodicarboxylate linker (Figure 3c, Figure S10, Table Si). However, in the refinement process an additional electron density, which could not be ascribed, was observed on the Fourier maps. This remains in agreement with additional experiments (elemental analysis and thermogravimetric analysis) performed for the ih2o sample that clearly demonstrated the presence of extra four disordered water molecules per Cd2 formula unit (Figure SI 1, Table S2). The total number of nine water molecules in the formula of lHzO corresponds to a water vapor uptake of 202 cm3/g observed for the activated MOF 1 in the adsorption isotherm measurements at p/p0 = 0.60. Additional GCMCs were employed to find the most probable positions of the crystallographically disordered water molecules. On the one hand, the Cd-MOF 1 possesses ID bottle-like channels, and, as shown in Figure 3d, the water, located in the structure of 1H20 by SC-XRD, does not occupy narrow parts of the channels that are decorated only by sulfonyl groups and benzene rings. On the other hand, however, the Monte Carlo calculations show there is a very high probability of finding water molecules in DOI: 10.1021 /acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 3290 I 64 Inorganic Chemistry this part of the channel (Figure S9). The lack of strong interactions with this quite hydrophobic region of the framework explains why these molecules could not be located by the SC-XRD method. Collectively, all above observations demonstrate that, for each Cd2 cluster in lHzO, there are (a) five ordered water molecules positioned through forming strong hydrogen-bonds with three various framework acceptor groups in the wider parts of the ID channels and (b) four unordered (without long-range ordering) water guests occupying the narrow regions of the channels. The average isosteric heat of adsorption (Figure S12) calculated from the isotherms at 298 and 308 K is 48.9 kj- mol- , which is by 8.1 kj-mol-1 higher than the latent heat of evaporation of water (40.7 kj-mol-1). The value obtained by GCMC calculations is similar and is equal to 52 kj-mol“1. It indicates noncoordinative interactions of water molecules with the framework. In addition, multiple adsorption—desorption cycles performed for {[Cd2(sdb)2(pcih)2]}„ clearly demon- strate that the water loading does not change after 35 steps (Figure 4, Figure SI3). Combining it with relatively low heat of adsorption makes this MOF a potentially good candidate for the cooling applications. Hydrolytic and Thermal Stability. In a few recently published reviews/perspectives on stability of MOFs, the authors emphasize necessity and importance of investigations under harsh hydrolytic conditions (such as high temperature and exposure to pure water and aqueous solutions of varied pH) as relevant to numerous applications of these materi- als.23-2^ In this work we studied both thermal and hydrolytic stability of 1 under diversified conditions. To test hydrolytic stability, the framework 1 was immersed in water and aqueous solutions at different pH values as well as conditioned in an environmental chamber (RH = 90%; T = 60 °C; see Table l). Powder X-ray diffraction (PXRD), IR spectra, and QE-TPDA analysis reveal that Cd-MOF 1 is stable in water even at high temperatures and in a weakly acidic media (approximately down to pH = 3), whereas it undergoes degradation under basic conditions (Table 1 and Figure 5). The analysis of the ex situ PXRD patterns of all tested samples: soaked in water, in aqueous HC1 solution (pH = 5 and pH = 3), and conditioned at high RH at elevated temperature (RH = 90%, T = 60 °C) indicated retention of crystallinity (Figure 5). On the other hand, the soaking of MOF 1 for 24 h in more acidic media (HC1 solution, pH = l) led to the complete degradation of the framework due to protonation of sdb2- linkers. This effect is demonstrated both by vanishing of reflection peaks in the PXRD pattern as well as by appearance Table 1. Conditions and Results of Stability Tests Lasting 24 ha T [°C] pH outcome h2o 70 ~7 stable 95 ~7 stable RH 90% 60 stable HC1 RT 5 stable 3 stable 1 unstable NaOH RT 9 unstable 11 unstable 14 unstable ‘‘RH—relative humidity; RT—room temperature. of several characteristic IR stretching O—H bands from noncoordinated neutral H2sdb in the 2500—3000 cm-1 range (Figure 5, Figure S14). Similarly, in basic solutions (NaOH, pH = 9), hydrolysis of Schiff base linkers and partial degradation of the coordination polymer was manifested by a reduction of intensity of PXRD reflection peaks (Figure 5). The conditioning of the as-synthesized 1 in the humidity chamber causes DMF leaching, which is demonstrated by the disappearance of the characteristic IR band at 1673 cm-1, corresponding to DMF carbonyl groups (Figure 5). Temper- ature-dependent powder X-ray diffraction (TD-PXRD) for 1 revealed that the Cd-MOF belongs to the robust family of MOFs and that it is stable up to ~300 °C. After this temperature is exceeded, the intensity of signals gradually diminishes, which is associated with breaking of coordination bonds and degradation of the material (Figure 6a). Thermogravimetric analysis (TGA) for Cd-MOF 1 shows stepwise loss of mass with a plateau in the range of 280—310 °C (Figure 6b, Figure Sll). The first step, before the plateau is reached, is associated with the loss of two DMF and one H20 molecules per Cd2 formula unit (found 11.9%; calcd 11.3%). The second pronounced step with a dTG minimum at ~330 °C is ascribed to breaking of coordination bonds between cadmium ions and linkers and decomposition of the framework. To get a deeper insight into the framework thermal stability, additional TG measurements were performed for the sample of 1 that had been soaked in water for 24 h (1H20). This material retains its structure and reversibly exchanges DMF molecule for H20, which was confirmed by both PXRD and IR spectroscopy (Figure S4). The first distinct step in the TGA curve for 1H20, with a dTGA minimum at 93 °C, corresponds to the loss of nine water guest molecules per Cd2 cluster (found 11.2%, calcd DOI: 10.1021 /acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 Figure 4. QE-TPDA profiles of water vapor desorption—adsorption cycles at 200 °C (left) and stability of porosity upon repeated water vapor adsorption—desorption cycles (right). 3291 I 65 Figure 5. PXRD patterns (top) and. corresponding IR spectra (bottom) of Cd-MOF 1 immersed in aqueous solutions for 24 h under various pH and temperature conditions. 11.2%). The number of guest water molecules for 1H20 was also confirmed by elemental analysis (Table S2). The comparison of TG analyses for both samples clearly indicates weaker interactions of water molecules in the framework 1H20 and thus softer activation conditions required for 1H20 compared to the as-made MOF 1 (containing more strongly interacting DMF molecules). In general, the exchange of MOF guest molecules can be described as the simplest postsynthetic modification. In case of Cd-MOF 1 both evacuation as well as replacement of initial guests have impact on their UV—vis spectra. The as-synthesized material 1 is colorless and, after evacuation of DMF/H20 (lact) or after replacing them with water (1H20), becomes yellow, which is confirmed by diffuse reflectance spectra of these samples (Figure SIS). For useful materials, there is an urgent necessity of investigation of their stability under versatile conditions. Commonly used TGA combined with TD-PXRD are not sufficient to describe all essential factors that have an impact on decomposition of MOFs. Such materials often lose their long- range ordering before decomposition temperature (indicated by TGA) is reached, and they partially lose their most exciting porosity feature before reflection peaks in TD-PXRD are vanished. Both these techniques do not give any information on the stability under harsh thermo-hydrolytic conditions, for example, water stream at high temperatures, which often appear in many industrial processes. To fill this gap we propose a new approach, which takes these factors into consideration. To examine the stability of 1 versus water vapor at elevated temperatures, the sample of the MOF was investigated by the QE-TPDA method (quasi-equilibrated temperature-pro- grammed desorption and adsorption), in which the activated material was heated at a rate of 5 °C*min-1 in the flow of He saturated with water vapor at room temperature (RT). Initial temperature was always ambient, while maximum temperature was stepwise increased from 200 to 320 °C with a 20 °C interval. For each temperature maximum, three desorption— adsorption cycles were recorded. Quantitative results of the stability of porosity analysis obtained with the QE-TPDA experimental technique are presented in Figure 6. The amount of adsorbed water was obtained by integrating the profiles recorded for each cycle (Figure 6d). Clearly, the material 1 maintains both crystallinity as well as sorption capacity approximately to 260 °C in the flow of water vapor. Within the first 11 cycles, water loading (222—215 cm3-g-1) is nearly constant, and in the twelfth cycle the porosity drops profoundly to 203 cm3-g_1, which is correlated with the onset of degradation of the framework under thermo-hydrolytic conditions. ■ CONCLUSION The first highly water-stable, robust non-interpenetrated 3D mixed-linker coordination polymer with a sulfonyl-function- alized dicarboxylate linker and an acylhydrazone colinker has been obtained in solution and by mechanochemical synthetic method. Sorption analysis, TD-PXRD, QE-TPDA, as well as TGA reveal that this cadmium-based MOF exhibits steep water vapor uptake at low pressures, excellent recyclability up to 260 °C, as well as is high stability upon heating to 300 °C. Single- crystal diffraction combined with GCMC simulations enabled the localization of adsorbed water molecules in the pores, as DOI: 10.1021 /acs.inorgchem.8b00078 Inorg- Chem. 2018, 57, 3287-3296 Inorganic Chemistry________________________________________________________________________________________________Q 3292 I 66 Inorganic Chemistry_________________________________________________________________________________________________Q Figure 6. Stability tests for {[Cd2(sdb)2(pcih)j] 2DMF'H20}„ (l): (a) TD-PXRD patterns, (b) TG and dTG curves for water-exchanged sample (1H20). (c) Thermo-hydrolytic stability of porosity upon repeated water vapor desorption—adsorption cycles at increasing temperatures: sorption capacity (black) and temperature ramp pattern (red) vs cycle number, (d) QE-TPDA profiles of water vapor desorption-adsorption cycles in the 200—320 °C range. well as contributed to the understanding of the interactions from 3° to 45° with a 0.05° step at a scan speed of 2.5° min-1. TD- between water molecules and the framework. This deep insight PXRD experiments were performed using Anton Paar BTS 500 into both sorption properties as well as methodology of testing heating stage from 30 to 350 C. At each temperature samples were hydrothermal conditions provides farther hints for the design conditioned for 10 min prior to the measurement. f Ł i « , rri « r 3 . .• , . r Electronic diffuse reflectance spectra were measured in BaSOa of water-stable MOFs and for understating mechanisms of „ .. _ __ , r. . Î , ° pellets with BaSU4 as a reference using Shimadzu 2101PC equipped a sorption. with ISR-260 attachment. Nitrogen and carbon dioxide adsorption/desorption studies were I MATERIALS AND METHODS performed on a BELSORP-max adsorption apparatus (MicrotracBEL «iii j . . ^ i , j / \ r.-n Corp.): 77 K was achieved by liquid nitrogen bath, 195 K was achieved 4-Pyndinecarbaldehyde isonicotinoyl hydrazone (pcih) [20 J was if., ,. , ' r , , , , ... i r-i i .i i *ii , i by dry ice/Isopropanol bath. Water vapor isotherms were recorded on prepared according to the published method. All other reagents and ' ' , * r . . 11. • 1 . r » .. t j te- au • L t» 1 \ j Hydrosorb (Quantachrome). Prior to the physisorption measurements solvents were of analytical grade (Sigma-Aldnch, POCH, Polmos; and ' r 7 r , 1 ... . r .1 -c . • the samples were soaked with dichloromethane (DCM) for 5—7 d. were used without further purification. r ' ' Carbon, hydrogen, sulfur, and nitrogen were determined by ^ thf the samPIes were ^acuated at RT for 16 h and additionally conventional microanalysis with the use of an Elementar Vario at ^ '' 101 ^ 1 MICRO Cube elemental analyzer. To determine stability of porosity of 1 upon adsorption of water IR spectra were recorded on a Thermo Scientific Nicolet iSlO FT- molecules we used a novel QE-TPDA technique. ' Prior to the IR spectrophotometer equipped with an iD7 diamond ATR experiment, a sample of 4.S mg of hydrated MOF was activated by attachment heating in a flow (6.75 cm /min) of pure helium (purity 5.0, Air TGA was performed on a Mettler-Toledo TGA/SDTA 851= Products) to 200 "C at S “C/min and then cooled. After activation the instrument at a heating rate of 10 °C min“1 in a temperature range flow was switched to helium-containing steam saturated at 25 °C, and of 25-600 °C (approximate sample weight of 50 mg). The RT sorption began. After stabilization of the thermal conductivity measurement was performed at atmospheric pressure under flowing detector (TCD) signal, indicating the end of the adsorption process, argon, the QE-TPDA experiment was performed by heating and cooling the In situ IR spectra were recorded on a Bruker Tensor 27 sample in the flow of He/H20 mixture according to the linear spectrometer equipped with a mercury cadmium telluride (MCT) temperature program with 5 °C/min ramps. Analogous experiment, detector and working with the spectral resolution of 2 cm-1. The however, with lower 1 °C/min ramps was also performed (Figure samples were activated in the form of self-supporting wafers for 1 h at S16). Initial temperature was always ambient, while maximum 220 °C prior to the adsorption of probe molecules at different temperature was stepwise increased from 200 to 320 °C with 20 °C temperatures for C02 (Linde Gas Polska, 99.95% used without further interval. For each maximum temperature three desorption-adsorption purification). cycles were recorded. After each cycle the sample was kept in RT for PXRD patterns were recorded at RT (295 K) on a Rigaku Miniflex 1.5 h before starting the next cycle to make sure the adsorption 600 diffractometer with Cu Ka radiation (/I = 1.5418 A) in a 26 range process was fully completed. Obtained in experiment maxima 3293 DOI: 10.1021/acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 I 67 Inorganic Chemistry_____________________________________________ correspond to desorption and minima to adsorption, while together they form the QE-TPDA profiles (Figure 6d). Sorption capacities were determined by integrating desorption maxima over the range from 25 to 120 °C and recalculating the obtained areas by adequate calibration constant. “N Desorption and adsorption isobars were calculated from the averaged first three desorption/adsorption cycles (Figure 3).29 Since the temperature required for desorption of water from 1 is much lower than the temperature of degradation of this material it was possible to study its resistance to multiple desorption—adsorption cycles. A sample of 5.9 mg of the studied MOF (1 H20) was activated in pure helium by heating to 200 °C with 5 °C ramp. Then helium flow was switched to water/He mixture. After stabilization of the TCD signal 35 desorption—adsorption cycles were measured with linear temperature program (5 °C/min rate). Maximum temperature was equal to 200 °C for each cycle. To make sure that adsorption process fully ended, 90 min of RT isothermal sorption was applied after each finished desorption—adsorption cycle. To obtain sorption capacity, desorption cycles were integrated over the range from 25 to 120 °C. Syntheses. Synthesis of {[Cd2(sdb)2(pcih)2]-2DMFH20}n (1) in solution. 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Cd(N03)2-4H20 (92.5 mg, 0.295 mmol), and 4,4'-sulfonyldibenzoic acid (H2sdb) (91.6 mg, 0.300 mmol) were dissolved in DMF (16 mL) and H20 (1.8 mL) by sonification (60 s) and heated at 70 °C for 4 d. Colorless crystals of 1 were filtered off, washed with DMF, and dried in oven at 60 °C and 500 mbar for 0.5 h. Yield: 135.9 mg (62.5%). Anal. Calcd for CS8H52N 10O17S2Cd2: C 48.04, H 3.61, N 9.66, S 4.42%. Found: C 47.93^ H 3.72, N 9.80 S 4.60%. FTIR (ATR, cm"1): 1/(000),, 1557s, v{COO)s 1401s, v(C= 0)DMF 1673 s, i/(C=N)pcih 1609s, i/(C=0)pdh 1694 s, i/(NH) 3218m. Mechanosynthesis of {[Cd2(sdb)2(pcih)2]-2DMFH20}n (1). 4- Pyridinecarbaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Cd(OH)2 (43.6 mg, 0.300 mmol), and 4,4'- sulfonyldibenzoic acid (H2sdb) (91.6 mg, 0.300 mmol) were ground togedier with 200 //L of DMF (rj = 1.58 //L-mg-1) in a oscillator mill Retsch MM200 for 90 min (agate jar; frequency 15 Hz). Synthesis of Cd(OH)2. KOH (1.23 g, 0.022 mol) and Cd(N03)- 4H,0 (3.08 g, 0.010 mol) were dissolved separately in 5 mL of water. The KOH solution was slowly dropped to the solution containing Cd(N03)2. Colorless product was filtered off, washed several times with H20, and dried in oven at 60 °C and 500 mbar for 2 h. Yield: 1.21 g (82.9%). Guest Exchange. Synthesis of {[Cd2(sdb)2(pcih)2]-9H20}n (1H20). {[Cd2(sdb)2(pcih)2]-2DMF-H20}n (l) (100 mg) was immersed in 5 mL of water for 24 h at room temperature. Yellow crystals were filtered off, washed with H20, and dried in ambient conditions in air for 0.5 h. Anal. Calcd for C52H54N8023S2Cd2: C 43.13, H 3.57, N 7.74, S 4.43%. Found: C 44.23, H 3.39, N 7.39, S 4.90%. Synthesis of {[Cd2(sdb)2(pcih)2]-9H20}n (1H20). {Cd2(sdb)2(pcih)2]-2DMF-H20}n (l) (100 mg) was conditioned in an environmental chamber (RH = 90%; T = 50 °C). Anal. Calcd for Cc2HS4N 802,S2Cd2: C 43.13, H 3.57, N 7.74, S 4.43%. Found: C 43.82, H 3.69, N 7.77, S 4.40%. Stability Test. {Cd2(sdb)2(pcih)2]-2DMFH20}„ (l) (100 mg) was immersed in water and aqueous solutions at different pH values as well as conditioned in an environmental chamber (RH = 90%; T = 60 °C; see Table l), Crystallographic Data Collection and Structure Refinement. A single crystal of 1 suitable for X-ray analysis was selected from bulk material (see Syntheses section). Diffraction data for 1 and 1H20 were collected on an Oxford Diffraction SuperNova Dual four-circle diffractometer equipped with an Atlas detector using mirror- monochromatized Mo Ka radiation (A = 0.71073 Â) and Oxford Cryojet system for measurements at 120 and 130 K, respectively. '0 The data were processed using (CrysAlisPro Oxford Diffraction Ltd. (2009)). The structures were solved by direct methods implemented in SHELXS 2014/6 and refined by a full-matrix least-squares procedure based on F2 with SHELXL 2014/6.31 All hydrogen atoms joined to carbon atoms were positioned with idealized geometries and refined using a riding model with [/^(H) fixed at 1.5 L/eq of C for methyl groups and 1.2 L/eq of C for other groups. The H atoms attached to the N atoms were found in the difference Fourier map and refined with an isotropic thermal parameter. The hydrogen atoms of guest water molecules in the structure of 1H20 could not be located in the difference Fourier map (alert B). Because of severe disorder of DMF and H20 solvent molecules in 1, the SQUEEZE procedure32 implemented via the WinGX suite3 ’ was applied to account for regions of diffuse electron density that could not be satisfactorily modeled (alert A). The squeezed void volume was 858.3 À3 (Vunit cey = 3237.2 Â3), equivalent to 26.5% of the unit cell. Molecular Simulations. GCMC molecular simulations were used to reproduce experimental adsorption isotherms and provide insight into adsorption on the molecular level.31 Such calculations are considered as an efficient computational tool being used for modeling static (equilibrium) properties related to adsorption and were successfully exploited for numerous adsorbent—adsorbate sys- 35-40 tems. To describe water molecules we used three-site SPC/E41 model, where point charges are placed on each atom, and only oxygen atom is van der Waals (vdW) interactions center. The flexibility of the molecule was described via bonded intramolecular interactions, that is, harmonic bond stretching and harmonic bond bending. Carbon dioxide molecules were also described by full atom model; however, they were assumed rigid. To reproduce quadrupole moment of the C02 molecule, charges were placed on each atom. Simulation box consisted of 2 X 3 X 1 unit cells of {[Cd2(sdb)2(pcih)2]]}, which is larger than twice the cutoff distance (>24 Â). Periodic boundary conditions were adopted in the three directions. The framework was assumed rigid, as the experimental results indicate that the structure of this MOF does not change significantly upon adsorption (Figure 4, Figure S4, and Table 2). Positions of the atoms were taken from cif files that can be found in the Supporting Information. To calculate Coulombic interactions, point charges—determined with the use of an extended charge equilibration method42—were placed on each framework atom. On the one hand, for calculations of C02 adsorption we used the values given in Table S3. On the other hand, for the adsorption of water the guest—guest interactions dominate in the system, because the studied material is moderate hydrophilic. For this reason it is difficult to initiate a stable adsorbent— adsorbate state when the structure is empty, because only one molecule can be inserted in a single Monte Carlo cycle. This problem was bypassed by increasing the point charges used for the MOF framework, which enforced adsorption by strengthening the Coulombic interactions. To initiate adsorption at the studied experimental conditions, that is, to insert the first molecules, the charges needed to be multiplied by at least 1.5. To obtain the adsorption isotherm and isobar presented in this work we performed the calculations with a set of charges with scaling parameters between 1.0 and 2.0 (0.1 interval) starting from the initial configurations of water molecules calculated with the values of charges multiplied by 1.5. The best agreement between experimental and calculated data was obtained when charges of framework atoms were scaled by 1.2. However, it is not possible to obtain any significant loading with the use of this model at pressures lower than ca. 1000 Pa when starting simulation from an empty structure. Data obtained for C02 and H20 in GCMC calculations are presented in Table S4. To describe the effective vdW forces between guest—guest and guest—host atoms we used Lennard-Jones (L-J) potentials with parameters obtained with Lorentz—Berthelot combining rules. The values of the parameters for framework atoms of {[Cd2(sdb)2(pcih)2]} were taken from DREIDING43 force field, except those for cadmium taken from UFF force field. A full set of parameters for L-J potentials and point charges can be found in Table S3. Theoretical pore volume was calculated from helium void fraction, which was determined using a simulation of a single helium molecule at the reference conditions (298 K).4'-' Pore volume (expressed in cm3/ g) was computed by multiplying helium void fraction by 1 X 103 and dividing the result by framework density (expressed in kg/m3). DOI: 10.1021/acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 3294 I 68 Inorganic Chemistry Table 2. Crystal Data and Structure Refinement Parameters for {[Cd2(sdb)2(pcih)2]*2DMF*H20}M (l) and {[Cd2(sdb)2(pcih)2]-9H20}„ ( 1H20) compound 1 ih2o empirical formula C26H18CdN407S C26H,gCdN40967S formula weight 642.90 685.67 crystal size (mm) 0.400 X 0.360 X 0.170 0.200 X 0.200 X 0.100 crystal system monoclinic monoclinic space group P2/c P2/c unit cell dimensions a (A) 12.7710(4) 12.7840(2) fc (A3 9.5275(4) 9.5272(2) c(A) 26.7963(18) 26.6342(6) a, r (deg) 90 90 ß (dug) 96.845(4) 97.133(2) volume (A3) 3237.2(3) 3218.83(11) temperature (K) 120(2) 130(1) Z 4 4 density (calculated) (g/cm ) 1.319 1.415 absorption 0.782 0.797 coefficient (mm-1) F(000) 1288 1374 0 range for data 3.138 to 30.427 2.985 to 29.598 collection (deg) index ranges —18 < h < 9, —6 < k < -16 2,7(1)] completeness to 99.5% 99.6% d = 25.242° refinement method full-matrix least-squares full-matrix least-squares on on F2 F2 data/restraints/ 8784/1/357 8150/31/407 parameters goodness-of-fit on F2 0.922 1.067 final R indices [J > R, = 0.0474, wRi = 0.0817 R, = 0.0520, wR7 = 0.1301 urn R indices (all data) R, = 0.0854, wR2 = 0.0926 R, = 0.0713, wR2 = 0.1407 Pore size distribution was calculated geometrically using the method of Gelb and Gubbins.46,4/ All above-described calculations were performed with the RASPA simulation code.48’49 ■ ASSOCIATED CONTENT Q Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorg- chem.8b00078. IR and UV—vis spectra, PXRD patterns, TGA data, crystal structure drawings, theoretical C02 and H20 occupation profiles, pore size distribution, GCMC simulations data, isosteric heat of adsorption plot, additional sorption data, and X-ray crystal data. (CCDC1480164 for 1 CCDC1812596 for 1H20) (PDF) Accession Codes CCDC 1480164 and 1812596 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_request^)ccdc.cam.ac.uk, or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033. ■ AUTHOR INFORMATION Corresponding Author *E-mail: dariusz.matoga(S)uj.edu.pl. ORCID Irena Senkovska: 0000-0001-7052-1029 Stefan Kaskel: 0000-0003-4572-0303 Dariusz Matoga: 0000-0002-0064-5541 Notes The authors declare no competing financial interest. ■ ACKNOWLEDGMENTS The National Science Centre (NCN, Poland) is gratefully acknowledged for the financial support (Grant No. 2015/17/ B/ST5/01190) of this research. We thank Dr. M. Hodorowicz (Jagiellonian Univ.) for SCXRD measurements and structure refinements. The research was performed partially with the equipment purchased thanks to the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (Contract No. POIG.02.01.00-12-023/08). ■ REFERENCES (1) Henninger, S. K.; Jeremias, F.; Kummer, H.; Janiak, C. MOFs for Use in Adsorption Heat Pump Processes. Eur. J. Inorg. Chem. 2012, 2012, 2625-2634. (2) de Lange, M. F.; van Velzen, B. L.; Ottevanger, C. P.; Verouden, K. J. F. M.; Lin, L.-C.; Vlugt, T. J. H.; Gascon, J.; Kapteijn, F. Metal— Organie Frameworks in Adsorption-Driven Heat Pumps: The Potential of Alcohols as Working Fluids. Langmuir 2015, 31, 12783-12796. (3) Canivet, J.j Fateeva, A; Guo, Y.; Coasne, B.; Farrusseng, D. Water Adsorption in MOFs: Fundamentals and Applications. Chem. Soc. Rev. 2014, 43, 5594-5617. (4) de Lange, M. F.; Verouden, K. J. F. M.; Vlugt, T. J. H.; Gascon, J.; Kapteijn, F. Adsorption-Driven Heat Pumps: The Potential of Metal— Organie Frameworks. Chem. Rev. 2015, 115, 12205—12250. (5) Prince, J. A; Bhuvana, S.; Anbharasi, V.; Ayyanar, N.; Boodhoo, K V. K; Singh, G. Self-Cleaning Metal Organie Framework (MOF) Based Ultra Filtration Membranes - A Solution to Bio-Fouling in Membrane Separation Processes. Sci. Rep. 2015, 4, 6555. (6) Yang, F.; Xu, G.; Dou, Y.; Wang, B.; Zhang, H.; Wu, H.; Zhou, W.; Li, J.-R.; Chen, B. A Flexible Metal—organie Framework with a High Density of Sulfonic Add Sites for Proton Conduction. Nat. Energy 2017, 2, 877-883. (7) Matoga, D.; Oszajca, M.; Molenda, M. Ground to Conduct: Mechanochemical Synthesis of a Metal-Organic Framework with High Proton Conductivity. Chem. Commun. 2015, 51, 7637—7640. (8) Kim, H.; Yang, S.; Rao, S. R.; Narayanan, S.; Kapustin, E. A; Furukawa, H.j Umans, A. S.; Yaghi, O. M.; Wang, E. N. Water Harvesting from Air with Metal-Organic Frameworks Powered by Natural Sunlight. Science 2017, 356, 430—434. (9) Coudert, F.-X. Responsive Metal—Organic Frameworks and Framework Materials: Under Pressure, Taking the Heat, in the Spotlight, with Friends. Chem. Mater. 2015, 27, 1905—1916. (10) Matoga, D.; Roztocki, K; Wilke, M.; Emmerling, F.; Oszajca, M.; Fitta, M.; Balanda, M. Crystalline Bilayers Unzipped and Rezipped: Solid-State Reaction Cycle of a Metal-Organic Framework with Triple Rearrangement of Intralayer Bonds. CrystEngComm 2017, 19, 2987-2995. (11) Low, J. J.; Benin, A I.; Jakubczak, P.; Abrahamian, J. F.; Faheem, S. A; Willis, R R. Virtual High Throughput Screening Confirmed DOI: 10.1021/acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 3295 I 69 Inorganic Chemistry Experimentally: Porous Coordination Polymer Hydration. /. Am. Chem. Soc. 2009, 131, 15834-15842. (12) Greathouse, J. A; Aliendorf, M. D. The Interaction of Water with MOF-5 Simulated by Molecular Dynamics. J. Am. Chem. Soc. 2006, 128, 10678-10679. (13) Schoenecker, P. M.; Carson, C. G.; Jasuja, H.; Flemming, C. J. J.; Walton, K. S. Effect of Water Adsorption on Retention of Structure and Surface Area of Metal—Organic Frameworks. Ind. Eng. Chem. Res. 2012, SI, 6513-6519. ( 14) Furukawa, H.; Gândara, F.; Zhang, Y.-B.; Jiang, J.; Queen, W. L.; Hudson, M. R.; Yaghi, O. M. Water Adsorption in Porous Metal- Organic Frameworks and Related Materials. /. Am. Chem. Soc. 2014, 136, 4369-4381. (15) Anastas, P. T.; Warner, J. C. Green Chemistry: Theory and Practice; Oxford University Press: New York, 1998; p 30. (16) Clark, J. H. Green chemistry: challenges and opportunities. Green Chem. 1999, 1, 1—8. (17) Cemansky, R Chemistry: Green refill. Nature 2015, 519, 379— 380. (18) Roztocki, K.; Jędrzejowski, D.; Hodorowicz, M.; Senkovska, L; Kaskel, S.; Matoga, D. Isophthalate—Hydrazone 2D Zinc—Organic Framework: Crystal Structure, Selective Adsorption, and Tuning of Mechanochemical Synthetic Conditions. Inorg. Chem. 2016, 55, 9663— 9670. (19) Roztocki, K.; Jędrzejowski, D.; Hodorowicz, M.; Senkovska, I.; Kaskel, S.; Matoga, D. Effect of Linker Substituent on Layers Arrangement, Stability, and Sorption of Zn Isophthalate/Acylhydra- zone Frameworks. Cryst. Growth Des. 2018, 18, 488—497. (20) Roztocki, K; Senkovska, I.; Kaskel, S.; Matoga, D. Carboxylate— Hydrazone Mixed-Linker Metal—Organic Frameworks: Synthesis, Structure, and Selective Gas Adsorption. Eur. J. Inorg. Chem. 2016, 2016, 4450-4456. (21) Parmar, B.; Rachuri, Y.; Bisht, K. K.; Suresh, E. Mixed-Ligand LMOF Fluorosensors for Detection of Cr(VI) Oxyanions and Fe3+/ Pd2+ Cations in Aqueous Media. Inorg. Chem. 2017, 56, 10939—10949. (22) NIST Chemistry WebBook; http://webbook.nist.gov/ chemistry; access date: 2018/02/21. (23) Howarth, A J.; Liu, Y.; Li, P.; Li, Z.; Wang, T. C.; Hupp, J. T.; Farha, O. K Chemical, Thermal and Mechanical Stabilities of Metal- organic Frameworks. Nature Rev. Mater. 2016, 1, 15018. (24) Burtch, N. C.; Jasuja, H.; Walton, K S. Water Stability and Adsorption in Metal—Organic Frameworks. Chem. Rev. 2014, 114, 10575-10612. (25) Wang, C.; Liu, X.; Keser Demir, N.; Chen, J. P.; Li, K Applications of Water Stable Metal-Organic Frameworks. Chem. Soc. Rev. 2016, 45, 5107-5134. (26) Makowski, W. Quasi-Equilibrated Temperature Programmed Desorption and Adsorption: A New Method for Determination of the Isosteric Adsorption Heat. Thermochim. Acta 2007, 454, 26—32. (27) Makowski, W.; Ogorzałek, L. Determination of the Adsorption Heat of N-Hexane and n-Heptane on Zeolites Beta, L, 5A, 13X, Y and ZSM-5 by Means of Quasi-Equilibrated Temperature-Programmed Desorption and Adsorption (QE-TPDA). Thermochim. Acta 2007, 465, 30-39. (28) Slawek, A; Vicent-Luna, J. M.; Marszałek, B.j Makowski, W.; Calero, S. Ordering of N-Alkanes Adsorbed in the Micropores of A1P04—5: A Combined Molecular Simulations and Quasi-Equili- brated Thermodesorption Study. /. Phys. Chem. C 2017, 121, 25292— 25302. (29) Slawek, A; Vicent-Luna, J. M.; Marszałek, B.; Balestra, S. R. G.; Makowski, W.; Calero, S. Adsorption of N-Alkanes in MFI and MEL: Quasi-Equilibrated Thermodesorption Combined with Molecular Simulations. /. Phys. Chem. C 2016, 120, 25338-25350. (30) Crysalis Pro; Oxford Diffraction Ltd: Yamton, England, 2010. (31) Sheldrick, G. M. SHELXL-2014/7: Program for the Solution of Crystal Structures; University of Göttingen: Göttingen, Germany, 2014. (32) Sheldrick, G. M. Crystal Structure Refinement with SHELXL. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3—8. (33) Farrugia, L. J. It WinGX and It ORTEP for Windows: An Update. /. Appl. Crystallogr. 2012, 45, 849—854. (34) Frenkel, D.; Smit, B. Understanding Molecular Simulation; Academic Press: San Diego, CA, 1996. (35) Liu, B.; Smit, B.; Rey, F.; Valencia, S.; Calero, S. A New United Atom Force Field for Adsorption of Alkenes in Zeolites. /. Phys. Chem. C 2008, 112, 2492-2498. (36) Dubbeldam, D.; Krishna, R.; Calero, S.; Yazaydin, A Ö. Computer-Assisted Screening of Ordered Crystalline Nanoporous Adsorbents for Separation of Alkane Isomers. Angew. Chem., Int. Ed. 2012, 51, 11867-11871. (37) Calero, S.; Dubbeldam, D.; Krishna, R.; Smit, B.; Vlugt, T. J. H.; Denayer, J. F. M.; Martens, J. A; Maesen, T. L. M. Understanding the Role of Sodium during Adsorption: A Force Field for Alkanes in Sodium-Exchanged Faujasites. J. Am. Chem. Soc. 2004, 126, 11377— 11386. (38) Karra, J. R.; Walton, K S. Molecular Simulations and Experimental Studies of C02, CO, and N2 Adsorption in Metal- Organic Frameworks. J. Phys. Chem. C 2010, 114, 15735—15740. (39) Paranthaman, S.; Coudert, F.-X.; Fuchs, A. H. Water Adsorption in Hydrophobic MOF Channels. Phys. Chem. Chem. Phys. 2010, 12, 8124-8130. (40) Tylianakis, E.; Froudakis, G. E. Grand Canonical Monte Carlo Method for Gas Adsorption and Separation. J. Comput. Theor. Nanosci. 2009, 6, 335-348. (41) Berendsen, H. J. C.; Grigera, J. R.; Straatsma, T. P. The Missing Term in Effective Pair Potentials. J. Phys. Chem. 1987, 91, 6269—6271. (42) Wilmer, C. E.; Kim, K C.; Snurr, R. Q. An Extended Charge Equilibration Method. J. Phys. Chem. Lett. 2012, 3, 2506—2511. (43) Mayo, S. L.; Olafson, B. D.; Goddard, W. A DREIDING: A Generic Force Field for Molecular Simulations. ). Phys. Chem. 1990, 94, 8897-8909. (44) Rappe, A. K; Casewit, C. J.; Colwell, K. S.; Goddard, W. A; Skiff, W. M. a Full Periodic Table Force Field for Molecular Mechanics and Molecular Dynamics Simulations. /. Am. Chem. Soc. 1992, 114, 10024-10035. (45) Talu, O.; Myers, A L. Molecular Simulation of Adsorption: Gibbs Dividing Surface and Comparison with Experiment. AIChE J. 2001, 47, 1160-1168. (46) Gelb, L. D.; Gubbins, K E. Pore Size Distributions in Porous Glasses: A Computer Simulation Study. Langmuir 1999, 15, 305—308. (47) Sarkisov, L.; Harrison, A Computational Structure Character- isation Tools in Application to Ordered and Disordered Porous Materials. Mol. Simul. 2011, 37, 1248-1257. (48) Dubbeldam, D.; Calero, S.; Ellis, D. E.; Snurr, R Q. RASPA: Molecular Simulation Software for Adsorption and Diffusion in Flexible Nanoporous Materials. Mol. Simul. 2016, 42, 81—101. (49) Dubbeldam, D.; Torres-Knoop, A; Walton, K S. On the Inner Workings of Monte Carlo Codes. Mol. Simul. 2013, 39, 1253—1292. DOI: 10.1021 /acs.inorgchem.8b00078 Inorg. Chem. 2018, 57, 3287-3296 3296 I 70 I 71 Bulky substituent and solvent induce alternative nodes for layered Cd-isophthalate/acylhydrazone frameworks Autorzy: Kornel Roztocki, Magdalena Lupa, Maciej Hodorowicz , Irena Senkovska, Stefan Kaskel, Dariusz Matoga Opublikowano w: CrystEngComm, 2018, 20, 2841-2849 Publikacja V Published on 01 May 2018. Downloaded by UNIWERSYTET JAGIELLOŃSKI on 24/05/2018 08:55:12. I 72 CrystEngComm Kornel Roztocki,a Magdalena Lupa,3 Maciej Hodorowicz, Stefan Kaskel#b and Dariusz Matoga©*3 1 Irena Senkovska, ©b Received 19th February 2018, Accepted 21st April 2018 DOI: 10.1039/c8ce00269j rsc.li/crystengcomm A series of three cadmium-based two-dimensional (2D) metal-organic frameworks, featuring 4-pyridinecarboxaldehyde isonicotinoyl hydrazone bridging pillars (pcih) and either 5-tert-butyl-substituted or non-substituted isophthalate linkers, have been prepared and characterized. Isophthalates with bulky tert-butyl substituents (tBu-iso2-) form two-dimensional (2D) frameworks with various metal node nuclearities: [Cd(fBu-iso)(pcih)(DMF)]n or [Cd2(tBu-iso)2(pcih)2]n. dependent on synthetic conditions; whereas, in contrast, unsubstituted isophthalates (iso2-) yield only the framework with dinuclear nodes [Cd2(iso)2(pcih)2ln. regardless of the conditions. Diverse nodes of interdigitated layers have a significant in- fluence on the thermal stability and adsorption behavior of the materials. Introduction Metal-organic frameworks (MOFs) are coordination polymers built from metal ions or clusters (nodes) and organic ligands which link the nodes. The broad interest in the field of MOFs is propelled by their common ciystallinity combined with po- rosity and the possibility to design both the environment and size of their voids in a molecular fashion.1-4 These features make MOFs suitable materials for addressing global chal- lenges in domains such as water harvesting and purifica- tion,5,6 catalysis,7-9 gas sequestration,10"12 molecular electronics13-18 and heat pumps.19 The de novo synthesis of coordination polymers (including MOFs) is based on a self-assembly process20 involving linkers and metal node precursors as building blocks.21 It is a part of crystal engineering that encompasses analysis of crystal structures in terms of intermolecular interactions as well as construction of crystals with desired topologies and proper- ties.22 Crystal engineering of coordination polymers with as- sembly control, however, is still challenging, since even using initially the same building blocks cannot guarantee the same final product. The obtained coordination polymers may have (i) different network structures and different chemical com- ° Faculty of Chemistry, Jagiellonian University, Gronostajowa 2, 30-387 Kraków, Poland. E-mail: dariusz.matoga@uj.edu.pl b Department of Inorganic Chemistry, Technische Universität Dresden, Bergstrasse 66, 01062 Dresden, Germany t Electronic supplementary information (ESI) available: IR spectra, PXRD pat- terns, N2 adsorption isotherms, and additional structure drawings. CCDC 1813611, 1813612 and 1816234 for 1-3. For ESI and ciystallographic data in GIF or other electronic format see DOI: 10.1039/c8ce00269j positions or (ii) different network structures of the same chemical composition (sometimes even whole crystals with identical compositions).23 The latter phenomenon is called supramolecular isomerism and it is commonly categorized into four types: (i) structural (different structures with the same components of the network), (ii) conformational (differ- ent conformations of ligands), (iii) catenane (different ways of interpenetration), and (iv) optical (different chiralities).24 Structural supramolecular isomers in the MOF family may have different topologies, i.e. connectivity of the adjacent framework nodes.25 The topology of MOFs can be tuned by merely changing a pendant group on a linker. For instance, functionalization of terephthalic acid (H2tfa) at the 2- and 5-positions leads to zinc-terephthalate/4,4'-bipyridine frame- works with a honeycomb-like topology,26 whereas the use of the unsubstituted acid yields a square-grid pillared-layered structure of [Zn2(tfa)2(bipy)]„ (bipy = 4,4'-bipyridine).27 Simi- larly, the change of linker functionalization has an effect on the topology of zinc-isophthalate/4,4'-bipyridine frameworks through various lateral arrangements of the functionalized isophthalates (fu-iso) within [Zn2(fu-iso)4] entities.28 In general, a framework architecture, which is correlated with its chemical properties, can be adjusted by a made-to- measure procedure that includes not only selection of ligands and metals but also tuning of reaction conditions as well. This includes for instance how energy is introduced into a system, physical conditions such as temperature and pres- sure, the presence of additional molecules or ions as tem- plates, and the solvents used.29-31 Selected recent examples where precise control of synthetic conditions is crucial for obtaining a targeted MOF include (i) physicochemical control This journal is © The Royal Society of Chemistry 2018 CrystEngComm. 2018, 20, 2841-2849 | 2841 PAPER View Article Online View Journal | View Issue H) Check for updates Cite this: CrystEngComm. 2018, 20, 2841 Bulky substituent and solvent-induced alternative nodes for layered Cd-isophthalate/acylhydrazone frameworks! Published on 01 May 2018. Downloaded by UNIWERSYTET JAGIELLOŃSKI on 24/05/2018 08:55:12. I 73 Paper View Article Online CrystEngComm (e.g. exclusive mechanochemical routes to proton-conductive materials32,33 and liquid-additive-dependent complete recon- struction by grinding"34); (ii) physical control (e.g. the effect of reaction time on successive crystallizations35 and the effect of temperature change on the formation of supramolecular stereoisomers36); and (iii) chemical control (e.g. the influence of pH on various MOF topologies37 and the influence of sol- vent on MOF dimensionality and isomerism38). Herein, we demonstrate that by using different reaction conditions and/or linker functionalization, we change the metal node nuclearity in resulting cadmium-acylhydrazone/ isophthalate frameworks. The interplay between the solvent ratio used for the synthesis and isophthalate linker functionalization at the 5-position leads to a series of 2D lay- ered frameworks: {[Cd(£Bu-iso)(pcih)(DMF)]*2DMF}„ (1), {[Cd2(fBu-iso)2(pcih)2]-2DMF-4//20}„ (2), and {[Cd2(iso)2(pcih)2] •2DMF}„ (3) (where tBu-iso = 5-£erf-butyl-isophthalate; iso = isophthalate; pcih = 4-pyridinecarboxaldehyde isonicotinoyl hydrazone), differing in their metal node nuclearities. In gen- eral, controlling both the formation of node clusters and the framework topology is a key aspect in MOF syntheses, since they directly influence the materials’ properties. Here, diverse nodes of interdigitated layers have a significant influence on the thermal stabilities and C02 adsorption isotherms of the materials, which is discussed in this paper. Results and discussion General remarks Three new mixed-linker 2D metal-organic frameworks based on 4-pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih), cadmium(u) and deprotonated 5-fert-butylisophthalic acid (H2ćBu-iso) or isophthalic acid (H2iso) have been prepared (Scheme 1). Single crystals of the frameworks 1-3 suitable for structure determinations were selected from the bulk materials. As revealed by X-ray diffraction and further corroborated by both elemental analysis and TGA, frameworks 1 and 2 are based on the same two linkers but have different metal node nuclearities (mononuclear and dinuclear nodes, respectively), dependent on the water-to-DMF solvent ratio used for their syn- theses (Scheme 1). This is the first example of the influence of synthetic conditions on the formation of node clusters ob- served in a family of mixed-linker carboxylate-hydrazone MOFs reported to date.39-43 On the other hand, the node nuclearity in the related framework 3, incorporating unsubstituted iso- phthalate instead of the one with the tert-butyl group, is not governed by the H20/DMF ratio and is the same as that in the dinuclear-node network 2. Regardless of the initial solvent ra- tios used in the synthesis, the same product 3 is obtained. The purity of the microporous mixed-linker metal-organic frame- works 1-3 was confirmed by elemental analysis (see the Experi- mental section), infrared spectroscopy (ATR-FTIR) and powder X-ray diffraction (PXRD) (Fig. Slf). Single-crystal X-ray diffraction revealed that all the porous co- ordination polymers 1-3 are 2D mixed-linker MOFs with adja- cent interdigitated layers arranged in the ABAB sequence (Fig. 1). All three frameworks have an identical underlying uninodal sql topology determined by TOPOSPro software.44 Mutual positions of alternating AB layers in each frame- work are governed by supramolecular interactions between the adjacent layers as well as between a layer and interlayer solvent molecules. The layers in framework 1 interact via the hydrogen bond N-H-O between the pcih NH group and car- boxylate oxygen atom [d = 2.859(5) Â; angle 157(5)°]. On the other hand, in compound 2 there are no direct hydrogen bonds and the layers are positioned through strong frame- work-guest (N-H)pt.ih - 0H2o and (O-H^o'-Omuiso hydrogen bonds [c/(N25-098) = 2.791(3) A, angle 164(3)°; 2) used for the assembly of sterically-hindered MOFs (1 and 2) leads to the kinetically controlled framework 1 with single metal nodes. This high DMF/H20 ratio favors the formation of complexes and aggre- gates with coordinated DMF molecules at the nucléation stage. When bulky hydrophobic substituents (fert-butyl) are additionally present on the isophthalate ions at this stage, it leads to favorable arrangements of ‘bi-chains’ with the DMF methyl groups pointing inside and the tert-butyl groups pointing outside the ‘bi-chains'. The coordinated DMF mole- cules prevent the formation of dinuclear nodes, and as a re- sult the ciystal growth of 1 begins with ‘bi-chain’ arrange- ments containing both types of non-polar groups (Fig. 2). Thermal stability Thermogravimetric analysis carried out for 1-3 revealed a stepwise weight loss for each MOF with a distinct plateau in the range of: 1, 230-310; 2, 250-300; and 3, 190-340 °C (Fig. 3). The first step leading to the plateau corresponds to the loss of guest molecules as well as labile monodentate li- gands (in the case of 1): three DMF molecules (one is a li- gand) for 1 (found: 27.0%, calcd. weight loss: 28.1%); two DMF and four H20 molecules for 2 (found: 15.5%, calcd. weight loss: 16.2%); and two DMF molecules for 3 (found: 12.9%, calcd. weight loss: 12.6%); all molecules given per Cd (1) and Cd2 (2 and 3) node unit. The next step with dTG min- ima at 336 (1), 320 (2) and 365 °C (3) is associated with the framework decomposition. The temperature-dependent PXRD patterns, recorded for 1-3 (Fig. 3), indicate that the frameworks undergo transfor- mations into distinct new ciystalline phases below the de- composition temperature determined by the TG measure- ments [approx. at 180 (1) and 160 °C (2, 3), respectively]. In the case of 1, additional partial amorphization of the sample upon thermal activation (200 °C) can be observed (Fig. S4f). The ex situ IR spectra of all compounds show that the main skeletal vibrations along with the characteristic carboxylate and pcih (C=0 and C=N) stretching bands remain intact af- ter thermal activation at 200 °C. These observations indicate that structural changes occurring at higher temperatures (in- dicated by TD-PXRD) do not involve intralayer rearrangements and are associated with interlayer flexibility. The factors that influence the thermal stability of the lay- ered metal-organic frameworks and the retention/change of This journal is © The Royal Society of Chemistry 2018 View Article Online Paper ciystalline phases after guest removal are: a) the presence or absence of hydrogen bonds (that are stronger than vdW inter- actions) between alternating layers and b) the nature of the node (e.g. the presence or absence of: unsaturated metal cen- ters, labile ligands, metal clusters, etc.). Compound 3 does not have unsaturated metal centers or labile ligands and its layers are stabilized by strong hydrogen bonds; therefore, it is the most thermally stable of all the frameworks. In con- trast, the metal centers in 1 contain labile monodentate DMF ligands, and in the case of 2, its adjacent layers are connected only by weak CH3-7t interactions. This results in lower decomposition temperatures of frameworks 1 (dTG = 336 °C) and 2 (dTG = 320 °C) as compared to 3 (dTG = 365 °C). Adsorption properties Prior to the physisorption experiments, the samples were soaked in dichloromethane for 7 days. During this time, the supernatant solvent was replaced by fresh solvent 3 times. To remove the solvent, the samples were evacuated at room tem- perature for 16 h and additionally at 150 °C for 16 h. Adsorp- tion analysis reveals that all the activated CdMOFs 1-3 are porous towards C02 at 195 K and nonporous for N2 at 77 K (Fig. 4 and S5|). Veiy low N2 adsorption is a common prop- erty for the family of mixed-linker acylhydrazone/carboxylate metal-organic frameworks.39-43 This C02 over N2 selectivity of 1-3 could be attributed to the presence of Lewis basic NH groups (of acylhydrazone moieties) that decorate the pore walls of the MOFs.43 In general, introducing basic sites (or otherwise introducing unsaturated metal centers) is a well- known strategy to induce C02 adsorption selectivity of MOFs.45,46 The C02 physisorption isotherm for framework 1 after synthesis and desolvation has a maximum uptake of 96 cm3 g-1 at plpo = 0.99, which corresponds to a total pore volume of 0.171 cm3 g_1 (Table 1). This value is in agree- ment with the pore volume of 0.179 cm3 g-1 calculated by Mercuiy software for [Cd(fBu-iso)(pcih)(DMF)]„ (without guest molecules) with a probe radius of 1.4 A (Table 1). The presence of coordinated DMF in the activated sample prior to adsorption measurements has been confirmed by IR spectroscopy and is also in agreement with the TG analysis (Fig. S6f). After storage for ca. 2 years in an argon atmo- sphere, the sample shows a slightly lower uptake and broad- ening of the hysteresis, suggesting that the structure suffers from possible DMF decomposition. To obtain a deeper in- sight into the C02 adsorption, temperature-dependent in situ IR spectra have been recorded for 1 at low temperatures (Fig. 5). The C02 molecule does not directly interact with the unsaturated metal centers, which is demonstrated by (i) the lack of a distinct 13C02 signal obscured by the main wide band of 12C02 (ref. 47), as well as by (ii) the slight weakening of the C-O bonds upon C02 adsorption (the band at 2335 cm-1 compared to that of free C02 at 2345 cm-1) and (iii) the desorption of C02 at higher temperatures CrystEngComm. 2018, 20. 2841-2849 | 2845 Published on 01 May 2018. Downloaded by UNIWERSYTET JAGIELLOŃSKI on 24/05/2018 08:55:12. I 77 View Article Online Paper CrystEngComm Fig. 4 Physisorption isotherms of C02 at 195 K for investigated compounds: (a) isotherm of compound 1 measured after synthesis and subsequent activation (black circles), isotherms of the first run measured after 2 years of storage in a glove box (red triangles) and the second run performed after run 1 (blue squares) (the sample was evacuated between the 2 runs at room temperature); (b) isotherm of compound 2 measured after synthesis and activation (black circles), isotherm of the first run measured after 2 years of storage in a glove box (red triangles); c) isotherm of compound 3 measured after synthesis and activation (black circles), isotherms of the first run measured after 2 years of storage in a glove box (red triangles), the second run performed after run 1 (green diamonds, the sample was evacuated at room temperature between run 1 and run 2), and the third run (blue squares, the sample was evacuated at 150 °C between run 2 and run 3); (d) semi-log plot of adsorption isotherms indicating the differences in the low pressure region: 1 - green squares, 2 - red triangles, 3 - black circles. Adsorption - solid symbols, desorption - empty symbols. Table 1 Comparison of experimental and calculated pore volumes for 1-3 “ Calculated based on the Gurvich rule and C02 physisorption data. b Derived from the first plateau at pfp0 = 0.07 (gas uptake of 45 cm3 g-1).c Calculated after excluding DMF guest molecules (2DMF). above -30 °C. These factors indicate nonspecific adsorbate- adsorptive interactions and the inaccessibility of cadmium centers even after the removal of coordinated DMF mole- cules with a decrease of the metal coordination number from seven to six. In framework 2 which is constructed from ID [Cd2(fBu- iso)2]„ ladder chains connected by the pcih acylhydrazone (Fig. 1 and 2), each side of a 2D sheet is decorated by bulky nonpolar rerr-butyl groups with high rotational degrees of freedom. These groups do not interact with C02 molecules and hinder their diffusion which is demonstrated by the lin- ear adsorption curve (poor equilibration) as well as the broad hysteresis between the adsorption and desorption branches (Fig. 4). A similar behavior, however, of lower magnitude, was observed for the analogous MOF [Zn2(Meiso)2(pcih)2]rt based on the isophthalate linker substituted with smaller methyl groups (Meiso2-).42 The uptake of C02 for framework 2 at a plpo of 0.99 (66 cm3 g"1; Vp0re = 0.118 cm3 g-1) is associated with filling of the voids (0.126 cm3 g \ calculated by Mercury software with a 1.4 A probe radius) after excluding the guests. After storage for ca. 2 years in an argon atmosphere, the sam- ple of compound 2 showed comparable adsorption behavior to that observed on the freshly synthesized and activated sample. It is worth mentioning that the sample adsorbs helium even at room temperature, pointing out the presence of ultra-micropores in the compound, therefore, the measure- ment of the helium dead volume has led to unreasonable ef- fects on the C02 adsorption isotherm. Therefore, the isotherm was collected in a helium-free mode (NO Void Analysis™ (NOVA) mode of the Quantachrome Autosorb-IQ instrument) using pre-calibrated sample cell. 2846 I CrystEngComm, 2018, 20. 2841-2849 This journal is © The Royal Society of Chemistry 2018 1 2 3 V [cm1 g~'] Experimental“ 0.171 0.118 0.080fc 0.157 Mercury 0.179f 0.126 0.084 (probe radius 1.4 Â) Published on 01 May 2018. Downloaded by UNIWERSYTET JAGIELLOŃSKI on 24/05/2018 08:55:12. I 78 Fig. 5 Temperature-dependent in situ IR spectra of [Cd(tBu- iso)(pcih)(DMF)]n (lact) during C02 adsorption. The C02 adsorption isotherm for framework 3 has a simi- lar shape to that observed for the analogous zinc-based framework [Zn2(iso)2(pcih)2].40 Both adsorption and desorp- tion branches have double step profiles with a hysteresis caused by the guest-induced framework response (Fig. 4).48’49 The first distinct step within the plp0 range of 10_5-0.075 ends with a C02 uptake of 45 cm3 g_1 (Vpore = 0.080 cm3 g"1) and can be associated with filling of the voids in the thermally activated framework (3act). Further increase in the pressure leads to the steep increase in the C02 uptake which can be explained by the framework layers displacement. This ‘gate-opening’ for 3 occurs at a considerably lower pressure {p/po ~0.075) than for the isostructural zinc analogue (p/p0 ~0.173).40 Obviously, the expansion of the interlayer distance generates additional accessible space, and framework 3 fi- nally adsorbs C02 with an amount of 88 cm3 g1 at plp0 = 0.99 (Vpore = 0.157 cm3 g"1). The aforementioned observa- tions are in agreement with theoretical pore volumes calcu- lated by the Mercury software with a 1.4  probe radius: 0.084 cm3 g"1. Storage of sample 3 under argon does not sig- nificantly affect the adsorption capacity, and a comparable uptake saturation is reached (Fig. 4c). However, as already reported in the literature for flexible porous coordination polymers,50 the adsorption behavior is affected by the re- peated adsorption/desorption, which is reflected in the change in the adsorbed amount and the shape of the adsorp- tion isotherm (Fig. 4c). More detailed investigations are needed to gain deeper insights into the origin of such phenomena. Conclusion Three new cadmium 2D metal-organic frameworks based on an acylhydrazone bridging ligand and either 5-te/t-butyl- substituted or non-substituted isophthalate linkers have been obtained. We have shown that the node nuclearity in the co- ordination polymers with the substituted isophthalates is controlled by the reaction conditions. In contrast, the node This journal is © The Royal Society of Chemistry 2018 View Article Online Paper nuclearity of the framework built from the non-substituted isophthalates remains the same regardless of synthetic condi- tions, which demonstrates an important role of steric hin- drance in the process of MOF self-assembly. An increase of the concentration of non-polar groups in the presence of other non-polar groups at the nucléation stage may lead to the formation of kinetically-controlled frameworks. Addition- ally, decorating the framework walls with bulky non-polar substituents decreases their thermal stability. All three MOFs exhibit different C02 adsorption properties which are corre- lated with their crystal structures. The results provide an in- sight into the role of synthetic conditions and linker substitu- ents in the construction of layered interdigitated porous coordination polymers. Experimental section Materials and methods 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) was prepared according to a published method.39 All other re- agents and solvents were of analytical grade (Sigma-Aldrich, POCH, Polmos) and were used without further purification. Carbon, hydrogen and nitrogen were determined by con- ventional microanalysis with the use of an Elementar Vario MICRO Cube elemental analyzer. IR spectra were recorded on a Thermo Scientific Nicolet iSlO FT-IR spectrophotometer equipped with an iD7 diamond ATR attachment. Thermogravimetric analyses (TGA) were performed on a Mettler-Toledo TGA/SDTA 85Ie instrument at a heating rate of 10 °C min-1 in a temperature range of 25-600 °C (approx. sample weight of 50 mg). The measurement was performed at atmospheric pressure under flowing argon. In situ IR spectra were recorded on a Bruker Tensor 27 spectrometer equipped with an MCT detector and working with a spectral resolution of 2 cm'1. The samples were acti- vated in the form of self-supporting wafers for 1 hour at 220 °C prior to the adsorption of probe molecules at different temperatures for C02 (Linde Gas Polska, 99.95% used with- out further purification). Powder X-ray diffraction (PXRD) patterns were recorded at room temperature (295 K) on a Rigaku Miniflex 600 diffrac- tometer with Cu-Ka radiation [X = 1.5418 A) in the 29 range from 3° to 45° with 0.05° steps at a scan speed of 2.5° min-1. Temperature-dependent powder X-ray diffraction experiments were performed using an Anton Paar BTS 500 heating stage from 30 to 370 °C. At each temperature, the samples were conditioned for 10 minutes prior to the measurement at a scan speed of 4.0° min-1. Nitrogen and carbon dioxide adsorption studies were performed on a BELSORP-max (MicrotracBEL Corp.) or Autosorb-IQ (Quantachrome Instruments) adsorption appa- ratus; 77 K was achieved with a liquid nitrogen bath and 195 K was achieved with a dry ice/isopropanol bath. Prior to the physisorption experiments, the samples were soaked in dichloromethane for 7 days. During this time, the CrystEngComm. 2018. 20. 2841-2849 | 2847 Published on OL May 2018. Downloaded by UNIWERSYTET JAGIELLOŃSKI on 24/05/20L8 08:55:12. I 79 Paper supernatant solvent was replaced by fresh solvent 3 times. To remove the solvent, the samples were evacuated at room temperature for 16 h and additionally at 150 °C for 16 h. Syntheses Synthesis of {[Cd(£Bu-iso)(pcih)(DMF)]-2DMF}„ (1). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Cd(N03)2-4H20 (92.5 mg, 0.295 mmol) and 5-iert-butylisophthalic acid (H2iBu-iso) (66.7 mg, 0.300 mmol) were dissolved in DMF (16.2 mL) and H20 (1.8 mL) by sonica- tion (60 s) and heated at 70 °C for 3 days. Colorless crystals of 1 were filtered off, washed with DMF and dried in an oven at 60°C and 500 mbar for 0.5 hours. Yield: 100.1 mg (42.7%). Anal. calc, for C33H43N708Cd: C 50.49, H 5.57, N 12.60%. Found: C 50.56, H 5.54, N 12.56%. FTIR (ATR. cm-1): v(COO)as 1557s, v(COO)s 1417s, v(fBu) 2963 m, v(C=0)DMF 1645s, v(C=N)pcih 1609s, v(C=0)pcih 1673s, v(NH) 3206 m. Synthesis of {[Cd2(£Bu-iso)2(pcih)2]-2DMF-4H20}n (2). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Cd(N03)2-4H20 (92.5 mg, 0.295 mmol) and 5-te/t-butylisophthalic acid (H2£Bu-iso) (66.7 mg, 0.300 mmol) were dissolved in DMF (9 mL) and H20 (9 mL) by sonication (60 s) and heated at 70 °C for 3 days. Colorless crystals of 2 were filtered off, washed with DMF and dried in an oven at 60 °C and 500 mbar for 0.5 hours. Yield: 80.5 mg (31.1%). Anal. calc, for C54H66Ni0Osl6Cd2: C 48.19, H 4.95, N 10.31%. Found: C 48.55, H 4.98, N 10.48%. FTIR (ATR. cm-1): v(COO)as 1557s, v(COO)s 1418s, 1430s, v(fBu) 2966 m, v(C=0)dmf 1661s, v(C=N)Pcih 1608 m, v(C=0)pcih 1673s, v(NH) 3208 m. Synthesis of {[Cd2(iso)2(pcih)2]-2DMF}„ (3). 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (67.9 mg, 0.300 mmol), Cd(N03)2-4H20 (92.5 mg, 0.295 mmol) and 1,3-benzenedicarboxylic acid (H2iso) (49.8 mg, 0.300 mmol) were dissolved in DMF (16 mL) and H20 (1.5 mL) by sonica- tion (60 s) and heated at 100 °C for 12 hours. Yellow crystals of 3 were filtered off, washed with DMF and dried in an oven at 60 °C and 500 mbar for 0.5 hours. Yield: 115 mg (68.9%). Anal. calc, for C46H42N10O12Cd2: C 47.96, H 3.68, N 12.16%. Found: C 47.34, H 3.56, N 12.00%. FTIR (ATR, cm"1): v(COO)as 1557s, v(COO)s 1384s, v(C=0)DMF 1668s, v(C=N)pcih 1603s, v(C=0)pcih 1687s, v(NH) 3207 m. Synthesis with the 1:1 (v/v) solvent ratio. 4-Pyridinecarboxaldehyde isonicotinoyl hydrazone (pcih) (203.4 mg, 0.900 mmol), Cd(N03)2-4H20 (154.8 mg, 0.900 mmol) and 1,3-benzenedicarboxylic acid (H2iso) (149.4 mg, 0.900 mmol) were dissolved in DMF (81.0 mL) and H20 (81.0 mL) by sonication (60 s) and heated at 100 °C for 24 hours. Yellow crystals of 3 were filtered off, washed with DMF and dried in air. Yield: 106 mg (10.2%). Anal. calc, for C46H42N10O12Cd2: C 47.96, H 3.68, N 12.16%. Found: C 47.69, H 3.59, N 12.04%. FTIR (ATR, cm"1): v(COO)as 1557s, t{COO)s 1384s, v(C=0)dmf 1668s, v(C=N)pcih 1603s, v(C=0)pcjh 1687S, i{NH) 3208 m. 2848 I CrystEngComm. 2018, 20, 2841-2849 View Article Online CrystEngComm Ciystallographic data collection and structure refinement Crystals of 1, 2 and 3 suitable for X-ray analysis were selected from the materials prepared as described in the syntheses above. The details of data collection and structure refinement parameters are summarized in Table Sl.f The intensity data for all structures were collected on a Super Nova diffractome- ter using monochromatic MoKa radiation, X = 0.71073 A. The positions of all atoms were determined by direct methods. All non-hydrogen atoms were refined anisotropically using weighted full-matrix least-squares on F2. In the structures, the hydrogen atoms connected to carbon atoms were in- cluded in calculated positions from the geometry of the mole- cules, whereas the hydrogen atoms of the water molecules and the nitrogen atoms were included from the difference maps and were refined with isotropic thermal parameters. The structures were solved using SIR-97 and refined by the SHELXL-2014 program.51’52 In the case of 3, the PLATON SQUEEZE program (Spek. A.L. PLATON SQUEEZE), a tool for the calculation of the disordered solvent contribution to the calculated structure factors, was used to treat regions with disordered solvent molecules which could not be sensibly modelled in terms of atomic sites. CCDC 1813611, 1813612 and 1816234 contain the supplementary crystallographic data for 1, 2 and 3, respectively. Conflicts of interest There are no conflicts to declare. Acknowledgements The National Science Centre (NCN, Poland) is gratefully ac- knowledged for the financial support (Grant no. 2015/17/B/ ST5/01190) of this research. This research was carried out par- tially with the equipment purchased thanks to the financial support from the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (contract no. POIG.02.01.00-12-023/08). The DFG is gratefully acknowledged for the financial support in the frame of FOR2433. References 1 Metal-Organic Frameworks, ed. D. Farrusseng, Wiley- VCH Verlag 8i Co. KGaA, Weinheim, 2011. 2 Metal-Organic Frameworks, ed. L. R. MacGillivray, John Wiley 8i Sons Inc, Hoboken, 2010. 3 H. Furukawa, K. E. Cordova, M. O'Keeffe and O. M. Yaghi, Science, 2003, 34,123044. 4 M. D. Allendorf and V. Stavila, CrystEngComm, 2015,17, 229-246. 5 J. A. Prince, S. Bhuvana, V. Anbharasi, N. Ayyanar, K. V. K. Boodhoo and G. Singh, Set Rep., 2014, 4, 6555. 6 H. Kim, S. Yang, S. R. Rao, S. Narayanan, E. A. Kapustin, H. Furukawa, A. S. Umans, O. M. Yaghi and E. N. Wang, Science, 2017, 356, 430. This journal is © The Royal Society of Chemistry 2018 Published on 01 May 2018. Downloaded by UNIWERSYTET JAGIELLOŃSKI on 24/05/2018 08:55:12. I 80 CrystEngComm 7 J. Liu, L. Chen, H. Cui, J. Zhang, L. Zhang and C.-Y. Su, Chem. Soc. Rev., 2014, 43, 6011-6061. 8 J. W. Maina, C. Pozo-Gonzalo, L. Kong, J. Schutz, M. Hill and L. F. Dumee, Mater. Horiz., 2017, 4, 345-361. 9 I. Nath, J. Chakraborty and F. Verpoort, Chem. Soc. Rev., 2016, 45, 4127-4170. 10 J. Yu, L.-H. Xie, J.-R. Li, Y. Ma, J. M. Seminario and P. B. Balbuena, Chem. Rev., 2017, 117, 9674-9754. 11 Z. Zhang, Z.-Z. Yao, S. Xiang and B. Chen, Energy Environ. Sei., 2014, 7, 2868-2899. 12 E. Barea, C. Montoro and J. A. R. Navarro, Chem. Soc. Rev., 2014, 43, 5419-5430. 13 W. P. Lustig, S. Mukheijee, N. D. Rudd, A. V. Desai, J. Li and S. K. Ghosh, Chem. Soc. Rev., 2017, 46, 3242-3285. 14 N. T. T. Nguyen, H. Furukawa, F. Gândara, C. A. Trickett, H. M. Jeong, K. E. Cordova and O. M. Yaghi, J. Am. Chem. Soc., 2015, 137, 15394-15397. 15 C. G. Silva, A. Corma and H. Garcia, J. Mater. Chem., 2010, 20, 3141-3156. 16 A. Morozan and F. Jaouen, Energy Environ. Sei., 2012, 5, 9269-9290. 17 L. Sun, M. G. Campbell and M. Dincä, Angew. Chem., Int. Ed., 2016, 55, 3566-3579. 18 I. Stassen, N. Burtch, A. Talin, P. Falcaro, M. Allendorf and R. Ameloot, Chem. Soc. Rev., 2017, 46, 3185-3241. 19 M. F. de Lange, K. J. F. M. Verouden, T. J. H. Vlugt, J. Gascon and F. Kapteijn, Chem. Rev., 2015, 115, 12205-12250. 20 G. M. Whitesides and B. Grzybowski, Science, 2002, 295, 2418-2421. 21 S. Kitagawa, R. Kitaura and S.-I. Noro, Angew. Chem,. Int. Ed., 2004, 43, 2334-2375. 22 G. Desiraju, IUCrJ, 2017, 4, 710-711. 23 J.-P. Zhang, X.-C. Huang and X.-M. Chen, Chem. Soc. Rev., 2009, 38, 2385-2396. 24 B. Moulton and M. J. Zaworotko, Chem. Rev., 2001, 101, 1629-1658. 25 M. O'Keeffe and O. M. Yaghi, Chem. Rev., 2012, 112, 675-702. 26 S. Henke and R. A. Fischer, J. Am. Chem. Soc., 2011, 133, 2064-2067. 27 B. Chen, C. Liang, J. Yang, D. S. Contreras, Y. L. Clancy, E. B. Lobkovsky, O. M. Yaghi and S. Dai, Angew. Chem., Int. Ed., 2006, 45, 1390-1393. 28 A. Schneemann, R. Rudolf, S. Henke, Y. Takahashi, H. Banh, I. Hante, C. Schneider, S. Noro and R. A. Fischer, Dalton Trans., 2017, 46, 8198-8203. 29 N. Stock and S. Biswas, Chem. Rev., 2012,112, 933-969. 30 C. Dey, T. Kundu, B. P. Biswal, A. Mallick and R. Banerjee, Acta Crystallogr., Sect. B: Struct. Sei., Cryst. Eng. Mater., 2014, 70, 3-10. 31 Z. Zhang and M. J. Zaworotko, Chem. Soc. Rev., 2014, 43, 5444-5455. This journal is © The Royal Society of Chemistry 2018 View Article Online Paper 32 D. Matoga, M. Oszajca and M. Molenda, Chem. Commun., 2015, 51, 7637-7640. 33 D. Matoga, K. Roztocki, M. Wilke, F. Emmerling, M. Oszajca, M. Fitta and M. Balanda, Crystalline bilayers unzipped and rezipped: solid-state reaction cycle of a metal-organic frame- work with triple rearrangement of intralayer bonds, CrystEngComm, 2017, 19, 2987-2995. 34 W. Yuan, T. Friscić, D. Apperley and S. L. James, Angew. Chem., 2010, 122, 4008-4011. 35 H. H. M. Yeung, Y. Wu, S. Henke, A. K. Cheetham, D. O’Hare and R. I. Walton, Angew. Chem., Int. Ed., 2016, 55, 2012-2016. 36 D. F. Sun, Y. X. Ke, T. M. Mattox, B. A. Ooro and H. C. Zhou, Chem. Commun., 2005, 5447-5449. 37 C. Volkringer, T. Loiseau, N. Guillou, G. Férey, M. Haouas, F. Taulelle, E. Elkaim and N. Stock, Inorg. Chem., 2010, 49, 9852-9862. 38 Y. Takashima, S. Furukawa and S. Kitagawa, CrystEngComm, 2011, 13, 3360-3363. 39 K. Roztocki, I. Senkovska, S. Kaskel and D. Matoga, Eur. J. Inorg. Chem., 2016, 27, 4450-4456. 40 K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, Inorg. Chem., 2016, 55, 9663-9670. 41 B. Parmar, Y. Rachuri, K. K. Bisht and E. Suresh, Inorg. Chem., 2017, 56, 10939-10949. 42 K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, Cryst. Growth Des., 2018, 18, 488-497. 43 K. Roztocki, M. Lupa, A. Slawek, W. Makowski, I. Senkovska, S. Kaskel and D. Matoga, Inorg. Chem., 2018, 57, 3287-3296. 44 V. A. Blatov, A. P. Shevchenko and D. M. Proserpio, Cryst. Growth Des., 2014, 14, 3576-3586. 45 K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T.-H. Bae and J. R. Long, Chem. Rev., 2012, 112, 724-781. 46 J. Liu, P. K. Thallapally, B. P. McGrail, D. R. Brown and J. Liu, Chem. Soc. Rev., 2012, 41, 2308-2322. 47 D. Matoga, B. Gil, W. Nitek, A. D. Todd and C. W. Bielawski, CrystEngComm, 2014, 16, 4959-4962. 48 A. Kondo, H. Noguchi, L. Carlucci, D. M. Proserpio, G. Ciani, H. Kajiro, T. Ohba, H. Kanoh and K. Kaneko, J. Am. Chem. Soc., 2007, 129, 12362-12363. 49 Y. Kubota, M. Takata, R. Matsuda, R. Kitaura, S. Kitagawa and T. C. Kobayashi, Angew. Chem., Int. Ed., 2006, 45, 4932-4936. 50 V. Bon, N. Kavoosi, I. Senkovska and S. Kaskel, ACS Appl. Mater. Interfaces, 2015, 7, 22292-22300. 51 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R. Spagna,/. Appl. Crystallogr., 1999, 32,115-119. 52 G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3-8. CrystEngComm. 2018, 20. 2841-2849 | 2849 I 81 Introducing a longer versus shorter acylhydrazone linker to a metal-organic framework: non- isoreticular structures, diverse stability and adsorption properties Autorzy: Kornel Roztocki, Monika Szufla, Maciej Hodorowicz , Irena Senkovska, Stefan Kaskel, Dariusz Matoga W przygotowaniu. Publikacja VI I 82 Introducing a longer versus shorter acylhydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties Kornel Roztocki,[a]Monika Szufla,[a]Maciej Hodorowicz/a] Irena Senkovska,[b] Stefan Kaskel,[b Dariusz Matoga[a] * [a]Faculty of Chemistry, Jagiellonian University, Gronostajowa 2, 30-387 Kraków, Poland Department of Inorganic Chemistry, Technische Universität Dresden, Bergstrasse 66, 01062 Dresden, Germany ABSTRACT: A 20-Â-long diacylhydrazone linker was used for the first time as a building block for the construction of a metal-organic framework (MOF). The terephthalaldehyde di-isonicotinoylhydrazone (tdih) is considerably longer than its 11-Â-long mono-acylhydrazone counterpart, 4-pyridinecarboxaldehyde isonicotinoylhydrazone (pcih), used so far as a MOF linker. By utilizing the two ditopic hydrazone linkers providing linear connectivity, two non-isoreticular cadmium-organic frameworks, {[Cd2(oba)2(hydrazone)2]}n (with oba = 4,4’-oxybis(benzenedicarboxylate) co-linker) were obtained by solution and mechanochemical methods. In spite of the same secondary building units, the frameworks differ in dimensionality as well as exhibit various stability and adsorption behavior, that are compared and discussed. The insights into structure-property relationships are provided by single-crystal X-ray diffraction (SC-XRD), temperature-dependent powder XRD, thermogravimetric analysis, CO2 (at 195 K), N2 (at 77 K) and H2O (at 298 K) gas adsorption measurements, as well as quasi-equilibrated temperature-programmed desorption and adsorption (QE-TPDA). The work opens possibilities to synthesize new functional materials based on the longer linker. Introduction The considerable interest of academia and industry in the metal-organic frameworks (MOFs) chemistry1-3 is driven to a large extent by their well-ordered porosity connected with structural flexibility and modularity.4,5 These features are a result of the amalgamation of organic and inorganic chemistry inside MOFs, which allows for almost infinite tunability, beyond conventional solid-state materials, to address plethora of possible and urgent applications such as drug delivery6,7, electronic devices,8-10 catalysis11,12 and photocatalysis13, gas storage and separation,14-17 water purification18 and others. Additionally, the properties of numerous MOF materials can be adjusted by controlling particles size19, defects20 as well as post-synthetic modifications.21-23 According to number of researchers, synthetic MOF chemistry is coming into a stage where scientific effort should be allocated in synthesis of MOFs with new topologies24 and high stability,25 as well as in searching the way to shape and fabricate already existing MOFs.26,27 Besides, there is still an urgent demand to design cost-effective synthetic routes that minimise the use of hazardous substances and energy,28 as well as to create (multi)functional and/or switchable MOFs, e.g. through simultaneous incorporation of various types of linkers or metals.29 Apart from current experimental challenges, the computational methods are also in the center of interest; for example, by using high throughput computational screening, approximately 125000 MOF structures (including 5000 experimental and hypothetical ones) were recently screened for selective Xe over Kr adsorption, and the most selective MOF for this sophisticated task was identified.30 In general, however, there is a necessity to include new materials in a database for future screening for other applications; and thus a necessity both to search for facile and non-expensive methods of their synthesis as well as to I 83 work out methods of evaluation of diverse stability of these materials against different destructors31 which may hinder their shaping and utilization for several commercial processes. Following the synthetic challenge to build multifunctional MOFs, we have been recently exploring the chemistry of metal-organic frameworks based on mixed carboxylate and acylhydrazone linkers, the latter introducing both hydrogen-bond donor (NH) and acceptor (C=O) groups as well as their coordination versatility towards a diverse range of metals.32-35 In 2016 we reported the first mixed- linker MOFs based on an acylhydrazone and a dicarboxylate.36 Since then several articles reporting new frameworks from this family and their properties, have been published. In particular, the carboxylic- acylohydrazone MOFs have been reported to exhibit numerous interesting behaviors, such as flexibility,37 high mechanical and hydrothermal stability,31,38 gas adsorption selectivity,36-39 structural diversity,39 multifunctional catalytic activity40 and sensing potential.41 However, all of the aforementioned frameworks are based on the same relatively short (~11 Â) 4-pyridinecarboxaldehyde isonicotinoylhydrazone (pcih) linkers and versatile dicarboxylic co-linkers. Therefore, this work was undertaken with the following motivation: (i) to construct a MOF with a longer acylhydrazone linker than 4-pyridinecarboxaldehyde isonicotinoylhydrazone used so far; (ii) to analyse its crystal structure along with related physicochemical properties including sorption behavior and various types of stability; and (iii) to compare the structural and physicochemical features with analogous pcih-based material. Herein, we present two new cadmium metal-organic frameworks based on acylhydrazone linkers providing linear connectivity: a shorter pcih or a longer terephthalaldehyde di- isonicotinoylhydrazone (tdih), each with the same additional dicarboxylate co-linker. In spite of the same secondary building units, the frameworks have non-isoreticular structures with different dimensionalities. We present and compare various alternative methods of synthesis of the two materials including deposition from solutions and optimized grinding of solid reactants. Both Cd-MOFs exhibit adsorption behaviors that are untypical of the so far known mixed-linker carboxylate-acylhydrazone MOFs, with CO2/N2 non-selectivity being the major difference. The adsorption properties towards CO2 (at 195 K) and N2 (at 77 K) gases are discussed along with structural features of the frameworks. Furthermore, by using of single-crystal X-ray diffraction (SC-XRD), temperature-dependent powder XRD, thermogravimetric analysis and quasi-equilibrated temperature-programmed desorption and adsorption (QE-TPDA) method, versatile types of the frameworks' stability are also investigated and discussed. Our study is the first example of utilization of terephthalaldehyde di-isonicotinoylhydrazone molecule as a MOF linker. This opens new synthetic possibilities with other metals and/or co-linkers which may provide interesting materials for a variety of applications. Results and Discussion General remarks Two mixed-linker cadmium(II) metal-organic frameworks of various dimensionalities: 2D {[Cd2(oba)2(pcih)2]-3DMF}n (1) and 3D {[Cd2(oba)2(tdih)2]-7H2O-6DMF}n (2), based on 4,4’- oxybis(benzenedicarboxylate) (oba2-) and either 4-pyridinecarboxaldehyde isonicotinoylhydrazone (pcih) or terephthalaldehyde di-isonicotinoylhydrazone (tdih) (Figure 1), were synthesized by solvent- based and solvent-free methods. In the solvent-based one, relatively high amounts of both organic solvent and energy (heating at 70 oC for approx. 4 days) were consumed to yield single crystals of compounds 1 and 2 suitable for X-ray diffraction measurements. Alternatively, the two materials were prepared by the solvent-free mechanochemical method, where solid reagents were ground together in a ball-mill in either a one-pot or a two-pot variant (Figures S1-S4, Supplementary Information). The latter approach meets the expectations of green chemistry, i.e. a reaction is fast and no organic solvent is used. The purity of products 1 and 2 obtained by the two methods was confirmed by elemental analysis, IR spectroscopy and powder X-ray diffraction (Table S1; Figures S2-S4, Supplementary Information). I 84 Figure 1. Linkers/linker precursors used for syntheses of compounds 1 and 2. H2oba: 4,4’-oxybis(benzenedicarboxylic) acid; pcih: 4-pyridinecarboxaldehyde isonicotinoylhydrazone (in 1); tdih: terephthalaldehyde di-isonicotinoylhydrazone (in 2). In the mechanochemical one-pot approach, CdO or Cd(OH)2, H2oba and pcih or tdih were ground together in the 1:1:1 molar ratio with the addition of a small amount of DMF (150 ^L), and in case of the CdO substrate, with the addition of a catalytic amount of NH4NO3 (4 wt%). In contrast, the two-pot reaction which proceeds via an intermediate coordination polymer of [Cd(^-H2O)(oba)(H2O)]n (CCDC1476612),42 does not require addition of the catalyst. Both products 1 and 2 are obtainable in the two-pot variant from a pre-assembled intermediate which is built of Cd2+ ions that are linked by the oba2- dicarboxylates and aqua bridges into a 3D framework, wherein each metal centre has a labile terminal aqua ligand. Analogous behaviour was shown for example for archetypical UiO-66 and its NH2 analogue which were easily obtained in a gram-scale milling by utilization of pre-assembled benzoate or methacrylate precursors.43 In our experiments, cadmium hydroxide was found as the preferable metal substrate. Firstly, when it was used as the substrate for grinding then independently on the variant used in the mechanochemical approach, no catalyst was required for the formation of 1 and 2 (see Figure S2-S3, Supplementary Information and Materials and Method section). Secondly, in case of the materials obtained from CdO, larger discrepancies were found in their elemental analyses (Table S1, Supplementary Information), which may indicate the presence of unreacted CdO as an impurity. The main factor responsible for reactivity is the chemical character of Cd(OH)2 which is more basic than CdO and thus favours deprotonation of H2oba. Crystal structures Single crystals of compounds {[Cd2(oba)2(pcih)2]-3DMF}n (1) and {[Cd2(oba)2(tdih)2]-7H2O-6DMF}n (2) used for X-ray diffraction (XRD) were grown in a DMF/H2O mixture (see Materials and Methods section). Despite using the same type of building blocks for the construction of the two materials, i.e. the H2oba linker precursor and a pyridyl-functionalized acylhydrazone building block of a linear connectivity (pcih or tdih differing in length), resulting products 1 and 2 do not form an isoreticular series. The MOFs crystallize in monoclinic and triclinic crystal systems with the space groups PI and C2/c, respectively. As revealed by XRD experiments, the Cd2+ ions in both frameworks have disordered pentagonal bipyramidal geometry where central ion is coordinated equatorially by five oxygen atoms from three different dicarboxylate acids (oba2-) (Figure S5, Supplementary Information). The coordination sphere is completed by axial N-pyridyl atoms from two neutral N,N-donor linkers; terephthalaldehyde di-isonicotinoylhydrazone (tdih) and 4-pyridinecarboxaldehyde isonicotinoylhydrazone (pcih), respectively for 1 and 2 (Figure 2, Figure S5, Supplementary Information). In both materials the oba2- ligands and Cd2+ ions form the [Cd2(COO)2] secondary building units (SBUs) with Cd-Cd distances of 3.836  (1) and 3.868  (2) in each cluster. The oba2- carboxylates I 85 act as ^3-k2k2k1 angular ligands connecting SBU units into the [Cd2(oba)2]n double chains (1) and layers (2). The double chains in 1 are further connected by the pcih ligands into the 2D framework of sql topology (determined by ToposPro software).44 The alternating layers in 1 are arranged in the ABAB manner and interact via strong N-H---O hydrogen bonds between NH group of the pcih acylhydrazone and carboxylate oxygen atom [d(N9-H(N9)--O47) = 2.85 À, angle = 171°] forming a two-dimensional pore structure (Figure 2; Figure S6, Supplementary Information). In contrast, the layers of [Cd2(oba)2]n in the framework 2 are further linked by the tdih ligand to form a three-dimensional non-interpenetrated pillar-layered structure. Solvent-accessible voids in the structure, occupied by DMF and H2O molecules in the as-synthesized material, consist of two intersecting 1D channels that form a 2D channel system propagating along the (001) plane. Two types of hydrogen bonds can be distinguished in the as- synthesized compound 2: intraframework and between the framework and guest molecules. The former includes carboxylate oxygen atoms interacting with NH groups of the tdih ligands (d(N20- H(N20)- --O57)) = 2.89 À, angle = 170°), whereas the latter involves NH group of the tdih ligands and oxygen atoms of DMF molecules d((N9-H(N9)---O64)) = 2.84 À, angle = 164°). The formation of hydrogen bonds in both MOFs is corroborated by their IR spectra. The spectra show the presence of the v(NH) amide bands that are shifted as compared to pure ligands [3218 (1) as compared to 3190 cm-1 (pcih); 3235 (2) as compared to 3249 cm-1 (tdih)] In spite of identical acylhydrazone linker topology, the use of linear acylhydrazone linkers of various length leads to non-isoreticular cadmium-organic frameworks 1 and 2. Replacing shorter pcih with longer tdih linker strongly affects conformation of the oba2- co-linker that is capable of twisting about the C-O and the C-C single bonds. As a result, dihedral angles between the CdOOC planes of the oba2- co-linker are significantly different in the two compounds (20.0o and 78.3o for 1 and 2, respectively), and the Cd- ••Cd distance (via the oba2- linker) is shorter for 2 (15.09 Â) than for 1 (15.20 Â) [Figure S7, Supplementary Information]. The main consequence, however, is significantly different angles between axial axes of neighboring coordination centres, i.e. parallel versus nearly perpendicular arrangement, for 1 and 2, respectively. This considerable difference leads to different network dimensionalities and topologies: three-dimensional 3,5-c network of unknown topology for 2 as opposed to two-dimensional sql topology for 1 (according to ToposPro). Figure 2. X-ray crystal structures of MOFs 1 (top) and 2 (bottom): (a) cluster coordination spheres of [Cd2(oba)2(pcih)2] (in 1) and [Cd2(oba)2(tdih)2] (in 2), (b) [Cd2(oba)2]n fragments (1D double chains in 1 and 2D layers in 2), (c) interlayer (in 1) and intraframework (in 2) hydrogen bonds, (d) solvent accessible voids (calculated with Mercury software by using a probe molecule with a radius of 1.2Â. In a), b) ,d) hydrogens atoms are omitted for clarity. Thermal stability and adsorption properties I 86 In order to assess thermal stability as well as to determine activation conditions for porous coordination polymers 1-2, both thermogravimetric (TG) and in situ temperature-dependent powder X-ray diffraction analyses (TD-PXRD) have been utilized (Figure 3). The TG curves indicate that the materials are thermally stable approx. up to 300 oC with DTG minima of 362 oC (1) and 342 oC (2), wherein upon heating to 280 oC weight losses of 15.7% and 29.1% are observed, respectively. These weight losses observed for both compounds are associated with removal of guests molecules from channels of the frameworks, and are in agreement with theoretical values (calculated for 3DMF: 15.6 % (1); for 7H2O and 6DMF: 27.6 % (2)). Figure 3. Thermal stability of compounds 1-2. TG and DTG curves (left), and TD-PXRD patterns (right). Temperature-dependent powder X-ray diffraction patterns confirm thermal stability of compound 1 determined by thermogravimetric analysis. When the sample is conditioned for 10 min at 330 oC, first clear changes in the XRD pattem, including the decrease of the distinguished peak at 20 = 6.5o, are observed (Figure 3). In contrast, variable temperature PXRD patterns recorded for compound 2, indicate that the framework undergoes two phase transitions in the range of 80-160 oC. The phase at higher temperatures and below 310 oC is characterized by two broad reflections which indicates loss of long-range ordering within the MOF and its predominant amorphous character above 160 oC. In order to avoid potential decompositions/phase transitions of the frameworks 1-2 indicated by TD-PXRD analysis (Figure 3), further confirmed by adsorption measurement upon thermal activation of 1 at 180 oC (Figure S8, Supplementary Information), gentle activation procedures have been adopted for both compounds before adsorption measurements. The samples were soaked for a few days in dichloromethane (DCM), with several replacement of the supernatant by fresh portions of DCM (such samples are referred to as 1DCM and 2DCM, respectively) , followed by room-temperature activation for approx. 16 hours under vacuum. The adsorption experiments reveal that the two MOFs 1 and 2 are porous towards both N2 at 77 K and CO2 at 195 K (Figure 4). This observation is in contrast with previous adsorption behaviour in the family of mixed-linker acylhydrazone/carboxylate metal-organic frameworks36,37,37-39 {[M2(dca)2(pcih)2]guest}n (M = Zn2+, Cd2+; dca2- - dicarboxylate), where the adsorption selectivity towards CO2 versus N2 was explained by the presence of Lewis basic NH groups from acylhydrazone moieties that decorated pores walls31. However, unlike compounds 1-2, the representatives of this family so far possess only one-dimensional channels or zero-dimensional voids. Even three-dimensional highly robust MOF [Cd2(sba)2(pcih)2]n (sda2- - 4,4’-sulfonyldibenzoic carboxylate), which has bigger window size (4.4 Â) than kinetic diameter of nitrogen (3.64 Â), I 87 demonstrates selective adsorption of CO2.31 As shown in Figure 2, two-dimensional voids in MOFs 1-2 are constructed by two intersecting channels: i) a bigger one, decorated by polar acylhydrazone - C(O)=N-NH- groups of the pcih or tdih linkers, and ii) a smaller one, propagating in close proximity of the oba2- linkers. The latter, more nonpolar channels, provide extra diffusion pathways for N2 into the frameworks 1-2. Figure 4. Physisorption isotherms of N2 (77 K) and CO2 (195 K) for CdMOFs 1-2. The CO2 (195 K) and N2 (77 K) physisorption isotherms for framework 1 after activation show maximum loading of 128 cm3 g-1 at p/po = 0.99 and 148 cm3 g-1 at p/p0 = 0.98, respectively. These values, according to the Gurvich rule45 (for both gases) correspond to a total pore volume of 0.228 cm3g- 1 (Table 1). Table 1. Porosity parameters for 1-2 calculated by Zeo++ software and obtained from CO2 (195 K) adsorption isotherms. MOF pws [Â] * mpd [Â] * Vpt [cm^g-1 ]* Vpe [cm3-g_1 ] 1 6.01 4.05 0.2 3 0.23# 2 6.94 9.07 0.32 0.32## * probe radius = 1.6 Â, pws-pore window size, mpd- maximum pore diameter, Vpt-theoretical pore volume, Vpe- experimental pore volume (based on limiting loading in adsorption isotherms for CO2) # p/po ~1.0, uptake 128 cm3-g_1 ## p/po ~ 1.0, uptake 181 cm3-g_1 This is in agreement with the theoretical pore volume of 0.23 cm3 g-1 calculated for the [Cd2(oba)2(pcih)2] framework by the Zeo++ software46 using the probe with a radius of 1.65  (corresponding to carbon dioxide diameter of 3.30 Â). The slight hysteresis for CO2 between adsorption and desorption branches is caused by weak interactions of the adsorptive with polar acylhydrazone - C(O)=N-NH- groups of the pcih linkers. The framework 1 belongs to a rare family of interdigitated 2D coordination polymers of general formula [M2(adc)2(lig)2]n (adc - angular dicarboxylate, lig - neutral N-donor ligand), whose members show selective CO2 (195K) adsorption over N2 (77 K). The utilization of the longer bent dicarboxylate, as compared e.g. to isophthalates, is responsible for the second highest CO2 uptake as well as for the loss of CO2 over N2 selectivity in this family (Table 2). In order to verify reversibility of adsorption, the second run of N2 (77 K) has been carried out on the sample evacuated at room temperature after the first run (Figure S9, Supplementary Information). The total uptake of N2 is almost unchanged in the second run (147 cm3g-1) as compared to the first (148 cm3 g-1), however in the first stage of the adsorption branch, at p/po ~ 0.05, the capacity drops significantly from 141 to 130 cm3g-1. These phenomena can be explained by increasing disorder of interdigitated layers of 1 upon subsequent activation process, which as a result causes more steric hindrances and limits gas diffusion into pores.39 I 88 Table 2. Examples of mixed-linker (dicarboxylate/N,N-donor) coordination polymers with layered interdigitated structures. MOF CO2 uptake at 195 K [cm3-g-1] Voids volume [%] Ref. Cd2(oba)2(pcih)2 128 31.4 This work Cd2(iso)2(pcih)2 88 17.0 39 Cd2(tBu-iso)2(pcih)2 66 21.9 39 Zn2(iso)2(pcih)2 91 12.5 37 Zn2(OH-iso)2(pcih)2 94 18.2 38 Zn2(CH3-iso)2(pcih)2 96 28 38 Zn2(iso)2(dsptztz)2 138 25.8 47 Zn2(iso)2(bpy)2 CID-1 58 12.2 48 Zn2(azdc)2(bpy)2 CID-13 77 22.3 49 Zn2(iso)2(bpb)2 CID-21 44 26.8 50 Zn2(iso)2(bpt)2 CID-22 33 27.4 50 Zn2(iso)2(bpa)2 CID-23 89 27.5 50 Zn2(ndc)2(bpy)2 CID-3 77 25.6 51 Zn2(NO2-iso)2(bpy)2 CID-5 57 -* 14 Zn2(CH3O-iso)2(bpy)2 CID-6 58 15.8 14 Xiso2-= 5-substituted isophthalate (X = H, CH3, OCH3, OH, NH2, NO2, tBu) bpy = 4,4'-bipyridyl bpndc2- = benzophenone-4,4’ -dicarboxylate bpb = 1,4- bis(4-pyridyl)benzene bpt = 3,6-bis(4-pyridyl)-1,2,4,5-tetrazine bpa = 1,4-bis(4-pyridyl)acetylene) dptztz = 2,5-di(4-pyridyl)thiazolo[5,4-d]thiazole azdc2- = 1,6-azulenedicarboxylate ndc2- = 2,6-naphthalenedicarboxylate *there is no guest-accessible void volume Replacement of the pcih linker with the longer tdih diacylhydrazone ligand, utilized as a linker in MOFs for the first time, leads to 3D acylhydrazone-carboxylate framework 2 which has the highest theoretical pore volume of 0.32 cm3 g-1, the highest CO2 (195 K) uptake of 181 cm3 g-1 as well as the highest maximum pore diameter (9.1 Â) and window size (6.94 Â) in the family. The experimental pore volume calculated from the CO2 adsorption branch at p/p0 = 0.99 equals to 0.32 cm3g-1 (Vpe)œ2 and is in a agreement with the value calculated geometrically (Table 1). However, the analogous value of (Vpe)N2 = 0.496 cm3g-1, calculated at p/p0 = 0.99 (320 cm3g-1) from the N2 (77 K) adsorption isotherm, surpasses (Vpe)CO2 by nearly 50%. To verify whether it is not an experimental error, a few adsorption-desorption cycles were performed as well as different batches of compounds were verified (Figure S10, Supplementary Information). The result is repeatable and this observation will be subject of further more detailed investigations. In order to get a fuller insight into adsorption processes, water vapor adsorption isotherms and isobars have been measured for activated MOFs 1 and 2 (Figures 5 and 6). The isotherms recorded at 298 K for both MOFs have shown their instability towards this small adsorptive demonstrating by large untypical hysteresis, long time of measurement (even 5 days) as well as the highest H2O uptake for lower p/p0 (105 cm3g-1 at p/p0 = 0.94 for 1; and 225 cm3g-1 at p/p0 = 0.94 for 2) than for maximal partial I 89 pressure (103 cm3g-1 for 1; and 199 cm3g-1 for 2). To investigate the stability in liquid H2O, frameworks 1-2 have been immersed in water at ambient temperature for 1 -6 days, and after this treatment the water- loaded CdMOFs were soaked in DMF or DMF/H2O (9:1 v/v) solution, to study their guest exchange reversibility. The framework 1 undergoes fully reversible phase transition (Figure S11, Supplementary Information), which according to our previous study of interdigitated frameworks, is caused by a reversible change of relative arrangement of adjacent layers, caused by exchange of guest molecules38. However, this MOF is not stable during repeatable cycles of water adsorption-desorption, which is demonstrated by loss of porosity (Figure S12, Supplementary Information). Figure 5. Physisorption isotherms of H2O (298 K) for MOFs 1-2. Closed symbols - adsorption; empty symbols -desorption. In contrast, CdMOF-2 after soaking in water, has undergone irreversible transformation to a highly water stable phase {[Cd2(oba)2(tdih)2]T3H2O}n (2H2O) that has characteristic narrow band at ~ 3500 cm-1 in IR spectrum, ascribed to the presence of highly ordered water molecules positioned by strong hydrogen- bonds with acylhydrazone groups (Figure S13 and S14, Supplementary Information).31 Figure 6 MOF-2: QE-TPDA profiles of water vapor desorption-adsorption cycles at 200 oC (left), desorption-adsorption isotherms (middle) and stability of porosity upon repeated water vapor desorption-adsorption cycle (right). The stability of the [Cd2(oba)2(tdih)2] phase as well as its porosity towards H2O have been elucidated by QE-TPDA (quasi-equilibrated temperature-programmed desorption and adsorption), which clearly demonstrated that H2O loading of the [Cd2(oba)2(tdih)2] framework does not change after 22 cycles. The maximal amount of water adsorbed in isobaric conditions (~150 cm3 g-1; 12.1 % by mass) is in agreement with the data obtained from thermogravimetric and elemental analyses for the water-loaded CdMOF-2 (Figure S14, Supplementary Information and Experimental section). Conclusion By utilizing two ditopic acylhydrazone linkers providing linear connectivity: shorter 4- pyridinecarboxaldehyde isonicotinoylhydrazone or longer terephthalaldehyde di- isonicotinoylhydrazone, two non-isoreticular cadmium-organic frameworks have been obtained. The frameworks possess the same secondary building units including Cd2 cluster but differ in dimensionality. The Cd-MOFs exhibit sorption behavior untypical of the so far known mixed-linker carboxylate- I 90 acylhydrazone MOFs, without CO2/N2 selectivity, commonly observed in this framework family. This phenomenon is explained by the coexistence of two types of channels in their structures, not observed for previous MOFs, that provide extra diffusion pathways for adsorptives. The frameworks immersed in water undergo either reversible or irreversible phase transitions, for shorter- or longer-linker MOF respectively; whereas the latter gives the longer-linker framework ability to reversibly absorb water vapor. The first use of a diacylhydrazone building block for the construction of metal-organic frameworks opens new synthetic possibilities e.g. with other metals and co-linkers and this work is currently in progress. Experimental Materials and Methods 4-pyridinecarbaldehyde isonicotinoylhydrazone (pcih) was prepared according to the published method.36 All other reagents and solvents were of analytical grade (Sigma Aldrich, POCH, Polmos) and were used without further purification. Carbon, hydrogen and nitrogen were determined by conventional microanalysis with the use of an Elementar Vario MICRO Cube elemental analyzer. IR spectra were recorded on a Thermo Scientific Nicolet iS10 FT-IR spectrophotometer equipped with an iD7 diamond ATR attachment. Thermogravimetric analyses (TGA) were performed on a Mettler-Toledo TGA/SDTA 851e instrument at a heating rate of 10 °C min-1 in a temperature range of 25 - 600 °C (approx. sample weight of 50 mg). The measurement was performed at atmospheric pressure under argon flow. Powder X-ray diffraction (PXRD) patterns were recorded at room temperature (295 K) on a Rigaku Miniflex 600 diffractometer with Cu-Ka radiation (A = 1.5418 Â) in a 20 range from 3° to 45° with a 0.05° step at a scan speed of 2.5° min-1. Temperature-dependent powder X-ray diffraction experiments were performed using Anton Paar BTS 500 heating stage from 30 to 350 oC. At each temperature samples were conditioned for 10 minutes prior to the measurement. Nitrogen and carbon dioxide adsorption/desorption studies were performed on a BELSORP- max adsorption apparatus (MicrotracBEL Corp.); 77 K was achieved by liquid nitrogen bath (195 K) was achieved by dry ice/isopropanol bath. Water vapor isotherms were recorded on Hydrosorb (Quantachrome). Prior to the sorption measurements the sample 1 was soaked with DCM for 5 - 7 days. After that the samples were evacuated at 80 °C (1) or RT (2) for 16 h. For some experiments, indicated in the manuscript, sample 2 was evacuated at 180 °C for 20 h. The 1H NMR spectrum was recorded with a Bruker Avance III 600 at 300 K. The chemical shifts (S) are reported in parts per million (ppm) on a scale downfield from tetramethylsilane. The 1H NMR spectra were referenced internally to the residual proton resonance in DMSO-d6 (S 2.49 ppm). In order to determine stability of porosity of 1 and 2 upon adsorption of water molecules we used a quasi-equilibrated temperature-programmed desorption and adsorption (QE-TPDA) technique.52,53 Prior to the experiment a sample of 6.1 mg (1DCM) and 5.9 mg (2H2O) were activated by heating in a flow (6.75 cm3/min) of pure helium (purity 5.0, Air Products) to 120 °C (1DCM) and 200 oC (2H2O) at 5 °C/min and then cooled down. After activation the flow was switched to helium containing steam saturated at 25 °C and room temperature sorption began. After stabilization of the TCD signal, 9 (1) and 21 (2) desorption-adsorption cycles were measured with linear temperature program (5 °C /min rate). Maximum temperature was equal to 120 oC (1) and 200 °C (2) for each cycle. To make sure that adsorption process fully ended, 90 minutes of RT isothermal sorption was applied after each finished desorption-adsorption cycle. The maxima obtained in experiment correspond to desorption and the minima to adsorption, while together they form the QE-TPDA profiles (Figure 6 and Figure S12, Supplementary Information). Sorption capacities were determined by integrating desorption maxima over the range from 25 °C to 120 °C and recalculating the obtained areas by adequate calibration constant.54 Desorption and adsorption isobars were calculated from the first desorption/adsorption cycle (Figure 6 and Figure S12, Supplementary Information).55 I 91 Syntheses Synthesis of {[Cd2(oba)2(pcih)2]3DMF}n (1) in solution: 4-Pyridinecarboxaldehyde isonicotinoylhydrazone (pcih) (67.9 mg, 0.300 mmol), Cd(NO3)2-4H2O (92.5 mg, 0.295 mmol) and 4,4’-oxybis(benzenedicarboxylic) acid (H2oba) (77.4 mg, 0.300 mmol) were dissolved in DMF (16 mL) and H2O (1.8 mL) by sonification (60 s) and heated at 70 °C for 4 days. Colorless crystals of 1 were filtered off, washed with DMF and dried in oven at 60 °C for 0.5 hour. Yield: 159.9 mg (73.7 %). Anal. Calc. for C^HstNuO^^: C 51.92, H 4.21, N 10.92. Found: C 51.49, H 03.90, N 10.72. FTIR (ATR, cm'1): v(COO)as 1402s, v(COO)s 1559m/1532s, v(C=0)dmf 1675s, v(C=N)pcih 1594s, v(CH)pcih 3071w, v(NH) 3218m. Synthesis of single crystals suitable for SC-XRD measurement: Pcih (22.6 mg, 0.100 mmol), Cd(NO3)2-4H2O (31.3mg, 0.100 mmol) and H2oba (25.8 mg, 0.1 mmol) were dissolved in DMF (9.0 mL) by sonification (60 s), next H2O (9.0 mL) was added and the mixture was heated at 70 °C for 4 days. Synthesis of {[Cd2(oba)2(pcih)2]3H2O}n (1H2O): {[Cd2(oba)2(pcih)2]-3DMF }n (1) (100 mg) was immersed in 15 mL water for 2 days at room temperature. Cream-colored powder was filtered off, washed with H2O and dried in air at ambient conditions for 0.5 hour. Anal. Calc. For {[Cd2(oba)2(pcih)2]-3H2O}n :C 50.22, H 3.40, N 9.01. Found: C 50.31, H 3.11, N 8.95%. Synthesis of terephthalaldehyde di-isonicotinoylhydrazone (tdih) Isonicotinic acid hydrazide (1.37 g, 10.0 mmol) was added to ethanol (20 mL), and the mixture was heated to boiling. Terephthalaldehyde (0.67 g, 5.0 mmol) suspended in ethanol (20 mL) was added. The reaction mixture was refluxed for 1.5 hours, then allowed to cool and stand overnight, producing a white crystalline solid, which was filtered off, washed with 10 mL of ethanol and dried in air. Yield: 1.26 g (94.6 %). Anal. Calc. for C20H16N 6O2: C 64.51, H 4.33, N 22.57 %. Found: C 64.16, H 4.37, N 22.11%. v(C=N) 1544m, v(C=O) 1653s, v(CH) 3070w, v(NH) 3249m. 'H-NMR and IR spectra are attached in the SI file (Figures S15-S16, Supplementary Information). Synthesis of {[Cd2(oba)2(tdih)2]7H2O6DMF}n (2) in solution Tdih (55.9 mg, 0.150 mmol), Cd(NO3)2-4H2O (46.3 mg, 0.150 mmol) and H2oba (38.6 mg, 0.150 mmol) were dissolved in DMF (16.2 mL) and H2O (1.8 mL) by sonification (60 s) and heated at 70 °C for 96 hours. Yellow crystals of 2 were filtered off, washed with DMF and dried in oven at 60°C and 500 mbar for 0.5 hour. Yield: 127.8 mg (85.5 %). Anal. Calc. for C86H104N18O27Cd2: C 50.47, H 5.12, N 12.32 %. Found: C 51.51, H 4.77, N 12.71 %. FTIR (ATR, cm"1): v(COO)s 1395shm, 1407m; v(COO)as 1534m, 1558m; v(C=O)tdih 1658s; v(C=0)dmf 1674s, v(CH)tdih 3070w, v(NH) 3235w. Synthesis of {[Cd2(oba)2(tdih)2]13H2O}n (2H2O): {[Cd2(oba)2(tdih)2]7-H2O-6DMF }n (2) (200 mg) was immersed in 18 mL water for 1-3 days at room temperature. Cream-colored powder was filtered off, washed with H2O and dried in air at ambient conditions for 0.5 hour. Anal. Calc. For {[Cd2(oba)2(tdih)2]-13H2O}n : C 47.59, H 4.35, N 9.79,. Found: C 47.76, H 4.24, N 9.76%. One-pot mechanosynthesis of {[Cd2(oba)2(pcih)2]3DMF}n (1) and {[Cd2(oba)2(tdih)2]7H2O6DMF}n (2) pcih (67.8 mg, 0.300 mmol) or tdih (111.7 mg, 0.300 mmol), CdO (38.5 mg, 0.300 mmol), H2oba (77.5 mg, 0.300 mmol) and NH4NO3 (7.4 mg, 0. 093 mol, 4 wt% for 1 and 9.1 mg, 114 mol, 4 wt% for 2) were ground together with an addition of DMF (120 ^L, n = 0.65 ^L-mg-1 for 1; 150 ^L, n = 0.66 ^L-mg- 1 for 2 ) in an agate milling ball (40 min, 15 Hz). After that the product was washed several times with I 92 DMF. The same approach was used to synthesize 1 and 2 from Cd(OH)2, but reaction was carried out without NH4NO3 and Cd(OH)2 (43.9 mg , 0.300 mmol) was used instead of CdO. Stepwise mechanosynthesis of {[Cd2(oba)2(pcih)2]3DMF}n (1) and {[Cd2(oba)2(tdih)2]7H2O6DMF}n (2) CdO (38.5 mg, 0.300 mmol) and H2oba (77.5 mg, 0.300 mmol) were ground together with an addition of 150 ^L H2O (n = 1.30 ^l-mg-1) an agate milling ball (40 min, 15 Hz). After that pcih (67.8 mg, 0.300 mmol) (1) or tdih (111.7 mg, 0.300 mmol) (2) and 156 ^L DMF (n = 0.85 ^L-mg-1) for 1; 190 ^L DMF (n = 0.83 ^L-mg-1) for 2 were added to the intermediate products and grinding was carried out for 40 min (15 Hz). The same approach was used to synthesize 1 and 2 from Cd(OH)2, but Cd(OH)2 (43.9 mg, 0.300 mmol) was used instead of CdO. Crystallographic data collection and structure refinement Diffraction intensity data for single crystals of compounds 1 and 2 were collected at 100 K on a KappaCCD (Nonius) diffractometer with graphite-monochromated Mo Ka radiation (X = 0.71073 Â). Cell refinement and data reduction were performed using firmware.56,57 Positions of all of non-hydrogen atoms were determined by direct methods using SIR-97.58 All non-hydrogen atoms were refined anisotropically using weighted full-matrix least-squares on F2. Refinement and further calculations were carried out using SHELXL 2014/7.59,60 All hydrogen atoms joined to carbon atoms were positioned with an idealized geometries and refined using a riding model with Uiso(H) fixed at 1.5 Ueq of C for methyl groups and 1.2 Ueq of C for other groups. The hydrogen atoms of the water (O66) molecule in 2 are indeterminate, H atoms attached to the N atoms were found in the difference-Fourier map and refined with an isotropic thermal parameter. Additionally, the crystal structure data shows that one DMF solvent molecule is heavily disordered and was removed using the SQUEEZE procedure implemented in the PLATON package.59 In case of other two DMF solvent molecules atoms were refined using DFIX and DANG instructions.60,61 The SQUEZZE procedure was also applied for 1 due to the presence of disordered guest molecules. The figures were made using CCDC1852965 (1) and CCDC1581727 (2) cif files that contain the supplementary crystallographic data. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. I 93 Table 3. Crystal data and structure refinement parameters for {[Cd2(oba)2(tdihhp3DMF}n (1) and {[Cd2(oba)2(tdih)2] •7H2Ü-6DMF}n (2). Compound 1 2 Empirical formula CdC26H18N4Ü6 CdC43H45N9O11 Formula weight 594.84 976.28 Crystal size (mm) 0.400 X 0.300 X 0.200 0.300 x 0.200 x 0.030 Crystal system Triclinic Monoclinic Space group PI C 2/c Unit cell dimensions a (Â) 9.6614(2) 40.7755(7) b (Â) 12.2246(3) 8.8797(2) c (Â) 15.1946(3) 29.8769(5) « (°) 110.712(2) 90 ß (°) 102.063(2) 109.734(1) y (°) 93.664(2) 90 Volume (Â3) 1622.79(7) 10182.3(3) Temperature (K) 100(2) 100(2) Z 2 8 Density (calculated) (g/cm3) 1.217 1.274 Absorption coefficient (mm-1) 0.710 0.490 F(000) 596 4016 Theta range for data collection (°) 2.947 to 29.431 2.354 to 27.485 Index ranges -13 2sigma (T)] 7659 8927 Completeness 94.7% to theta = 29.431° 99.5% to theta = 27.485° Absorption correction MULTI-SCAN none Max. and min. transmission 1.000, 0.788 0.985, 0.889 Refinement method Full-matrix least-squares on F2 Full-matrix least-squares on F2 Data / restraints / parameters 9086 / 1 /338 11632 / 3 / 543 Goodness-of-fit on F2 1.076 1.061 Final R indices [I>2sigma(I)] R1 = 0.0293, wR2 = 0.0622 R1 = 0.0673, wR2 = 0.1726 R indices (all data) R1 = 0.0416, wR2 = 0.0680 R1 = 0.0887, wR2 = 0.1838 I 94 Associated Contents The Supporting Information is available free of charge on the RSC Publication website at DOI:... IR spectra, PXRD patterns, N2 adsorption isotherms, TG curves additional structure drawings and X-ray crystal data (CCDC 1852965 and 1581727) for 1 and 2, respectively. Acknowledgements The National Science Centre (NCN, Poland) is gratefully acknowledged for the financial support (Grant no. 2015/17/B/ST5/01190) of this research. KR additionally thanks the National Science Centre (NCN, Poland) for a doctoral scholarship within ETIUDA funding scheme (Grant no. 2018/28/T/ST5/00333). The research was carried out partially with the equipment purchased thanks to the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (contract no. POIG.02.01.00-12-023/08). Corresponding Author *E-mail: dariusz.matoga@uj .edu.pl References 1 D. Farrusseng (Ed.), Metal-Organic Frameworks, Wiley- VCH Verlag & Co. KGaA, Weinheim, 2011, Metal-Organic Frameworks. 2 MacGillivray, Metal-Organic Frameworks, John Wiley & Sons Inc, Hoboken, 2010, Metal-Organic Frameworks. 3 S. Kaskel (Ed.), The Chemistry of Metal-Organic Frameworks: Synthesis, Characterization, and Applications, Wiley- VCH Verlag & Co. KGaA, Weinheim, 2016. 4 J.-P. Zhang, H.-L. Zhou, D.-D. Zhou, P.-Q. Liao and X.-M. Chen, Natl. Sci. Rev., 2017,. 5 A. Schneemann, V. Bon, I. Schwedler, I. Senkovska, S. Kaskel and R. A. Fischer, Chem. Soc. Rev., 2014, 43, 6062-6096. 6 C.-Y. Sun, C. Qin, X.-L. Wang and Z.-M. Su, Expert Opin. Drug Deliv., 2013, 10, 89-101. 7 M.-X. Wu and Y.-W. Yang, Adv. Mat., 29, 1606134. 8 P. Falcaro, R. Ricco, C. M. Doherty, K. Liang, A. J. Hill and M. J. Styles, Chem. Soc. Rev., 2014, 43, 5513-5560. 9 I. Stassen, N. Burtch, A. Talin, P. Falcaro, M. Allendorf and R. Ameloot, Chem. Soc. Rev., 2017, 46, 3185-3241. 10 W.-J. Li, J. Liu, Z.-H. Sun, T.-F. Liu, J. Lü, S.-Y. Gao, C. He, R. Cao and J.-H. Luo, Nat. Comm., 2016, 7, 11830. 11 A. H. Chughtai, N. Ahmad, H. A. Younus, A. Laypkov and F. Verpoort, Chem. Soc. Rev., 2015, 44, 6804-6849. 12 Y.-B. Huang, J. Liang, X.-S. Wang and R. Cao, Chem. Soc. Rev., 2017, 46, 126-157 13 A. Dhakshinamoorthy, Z. Li and H. Garcia, Chem. Soc. Rev., 2018. 14 T. Fukushima, S. Horike, Y. Inubushi, K. Nakagawa, Y. Kubota, M. Takata and S. Kitagawa, Angew. Chem. Int. Ed., 2010, 49, 4820-4824. 15 J. A. Mason, M. Veenstra and J. R. Long, Chem. Sci., 2014, 5, 32-51. 16 J. Yu, L.-H. Xie, J.-R. Li, Y. Ma, J. M. Seminario and P. B. Balbuena, Chem. Rev., 2017, 117, 9674- 9754. 17 Z. Zhang, Z.-Z. Yao, S. Xiang and B. Chen, Energy Environ. Sci., 2014, 7, 2868-2899. 18 J. A. Prince, S. Bhuvana, V. Anbharasi, N. Ayyanar, K. V. K. Boodhoo and G. Singh, Sci. Rep., 2014, 4, 6555. 19 S. Tanaka, K. Fujita, Y. Miyake, M. Miyamoto, Y. Hasegawa, T. Makino, S. Van der Perre, J. Cousin Saint Remi, T. Van Assche, G. V. Baron and J. F. M. Denayer, J. Phys. Chem. C, 2015, 119, 28430- 28439. 20 A. K. Cheetham, T. D. Bennett, F.-X. Coudert and A. L. Goodwin, Dalton Trans., 2016, 45, 4113- 4126. 21 J. D. Evans, C. J. Sumby and C. J. Doonan, Chem. Soc. Rev., 2014, 43, 5933-5951. I 95 22 P. Deria, J. E. Mondloch, O. Karagiaridi, W. Bury, J. T. Hupp and O. K. Farha, Chem. Soc. Rev., 2014, 43, 5896-5912. 23 S. M. Cohen, Chem. Rev., 2012, 112, 970-1000. 24 C. H. Hendon, A. J. Rieth, M. D. Korzyński and M. Dincä, ACS Cent. Sci., 2017, 3, 554-563. 25 A. J. Howarth, Y. Liu, P. Li, Z. Li, T. C. Wang, J. T. Hupp and O. K. Farha, Nat. Rev. Mat., 2016, 1, 15018. 26 M. S. Denny Jr., J. C. Moreton, L. Benz and S. M. Cohen, Nat. Rev. Mat., 2016, 1, 16078. 27 Z. Kang, L. Fan and D. Sun, J. Mater. Chem. A, 2017, 5, 10073-10091. 28 P. A. Julien, C. Mottillo and T. Friscic, Green Chem., 2017, 19, 2729-2747. 29 H. Furukawa, U. Müller and O. M. Yaghi, Angew. Chem. Int. Ed., 2015, 54, 3417-3430. 30 D. Banerjee, C. M. Simon, A. M. Plonka, R. K. Motkuri, J. Liu, X. Chen, B. Smit, J. B. Parise, M. Haranczyk and P. K. Thallapally, Nat. Comm., 2016, 7, 11831. 31 K. Roztocki, M. Lupa, A. Sławek, W. Makowski, I. Senkovska, S. Kaskel and D. Matoga, Inorg. Chem., 2018, 57, 3287-3296. 32 K. Roztocki, D. Matoga and J. Szklarzewicz, Inorg. Chem. Comm., 2015, 57, 22-25. 33 D. Matoga, J. Szklarzewicz, K. Stadnicka and M. S. Shongwe, Inorg. Chem., 2007, 46, 9042-9044. 34 D. R. Richardson and P. V. Bernhardt, J. Biol. Inorg. Chem., 1999, 4, 266-273. 35 W. Henderson, L. L. Koh, J. D. Ranford, W. T. Robinson, J. O. Svensson, J. J. Vittal, Y. M. Wang and Y. Xu, J. Chem. Soc., Dalton Trans., 1999, 3341-3343. 36 K. Roztocki, I. Senkovska, S. Kaskel and D. Matoga, Eur. J. Inorg. Chem., 2016, 2016, 4450-4456. 37 K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, Inorg. Chem., 2016, 55, 9663-9670. 38 K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, Cryst. Growth Des., 2017, 18, 488-497. 39 K. Roztocki, M. Lupa, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga, CrystEngComm, 2018, 20, 2841-2849. 40 P. Patel, B. Parmar, R. I. Kureshy, N. H. Khan and E. Suresh, Dalton Trans., 2018, 47, 8041-8051. 41 B. Parmar, Y. Rachuri, K. K. Bisht and E. Suresh, Inorg. Chem., 2017, 56, 10939-10949. 42 S. Singh and A. Bhim, J. Solid State Chem., 2016, 244, 151-159. 43 K. Uzarevic, T. C. Wang, S.-Y. Moon, A. M. Fidelli, J. T. Hupp, O. K. Farha and T. Friscic, Chem. Commun., 2016, 52, 2133-2136. 44 V. A. Blatov, A. P. Shevchenko and D. M. Proserpio, Cryst. Growth Des., 2014, 14, 3576-3586. 45 L. Gurvich, J. Phys. Chem. Soc. Russ. 1915, 47, 805. 46 T. F. Willems, C. H. Rycroft, M. Kazi, J. C. Meza and M. Haranczyk, Micropor. Mesopor. Mat., 2012, 149, 134-141. 47 S. Millan, G. Makhloufi and C. Janiak, Crystals, 2018, 8, 30. 48 S. Horike, D. Tanaka, K. Nakagawa and S. Kitagawa, Chem. Commun., 2007, 3395-3397. 49 K. Kishida, S. Horike, K. Nakagawa and S. Kitagawa, Chem. Lett., 2012, 41, 425-426. 50 Y. Hijikata, S. Horike, M. Sugimoto, H. Sato, R. Matsuda and S. Kitagawa, Chem. Eur. J., 2011, 17, 5138-5144. 51 K. Nakagawa, D. Tanaka, S. Horike, S. Shimomura, M. Higuchi and S. Kitagawa, Chem. Commun., 2010, 46, 4258-4260. 52 W. Makowski, Thermochim. Acta., 2007, 454, 26-32. 53 W. Makowski and Ł. Ogorzałek, Thermochim. Acta., 2007, 465, 30-39. 54 A. Sławek, J. M. Vicent-Luna, B. Marszałek, S. R. G. Balestra, W. Makowski and S. Calero, J. Phys. Chem. C, 2016, 120, 25338-25350. 55 A. Sławek, J. M. Vicent-Luna, B. Marszałek, W. Makowski and S. Calero, J. Phys. Chem. C, 2017, 121, 25292-25302. 56 Nonius (1998). COLLECT. Nonius BV, Delft, The Netherlands. 57 Z. Otwinowski; W. Minor, (1997). Methods in Enzymology, Vol. 276, Macromolecular Crystallography, Part A, edited by C. W. Carter Jr & R. M. Sweet, pp. 307-326. New York: Academic Press, Method in enzymology. 58 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R. Spagna, J. Appl. Crystallogr, 1999, 32, 115-119. 59 A. L. Spek, Acta Crystallogr. A, 2015, 71, 9-18. I 96 60 G. M. Sheldrick, Program for the Crystal Structure Refinement, University of Göttingen, Göttingen, Germany, 2008, SHELXS-97. 61 G. M. Sheldrick, Program for the Solution of Crystal Structures, University of Göttingen, Göttingen, Germany, 2014, SHELXL-2014/7. I 97 Supplementary material accompanying Introducing a longer versus shorter acylohydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties Kornel Roztocki,[a]Monika Szufla,[a]Maciej Hodorowicz,[a Irena Senkovska,[b Stefan Kaskel,[b Dariusz Matoga[a] * [a]Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland [b]Department of Inorganic Chemistry, Technische Universität Dresden, Bergstrasse 66, 01062 Dresden, Germany Figure S1. Various synthetic routes to {[Cd2(oba)2(tdih)2]-7H2Ü-6DMF}n (2)............................98 Figure S2. IR spectra and PXRD patterns of 1 (left) and 2 (right) synthesized via liquid-assisted grinding from CdO as a one-step (red) or a two-step reaction (green), compared with the pattern calculated from SCXRD data (black) and with materials obtained in solution (blue).....................98 Figure S3. IR spectra and PXRD patterns of 1 (left) and 2 (right) synthesized via liquid-assisted grinding from Cd(OH)2 as a one-step (red) or a two-step reaction (green), compared with the pattern calculated from SCXRD data (black) and with materials obtained in solution (blue).....................99 Figure S4. PXRD pattern of the intermediate product [Cd(^-H2O)(oba)(H2O)]n: calculated (bottom), obtained from CdO (middle) and Cd(OH)2 (top)..........................................................99 Figure S5. X-ray crystal structures of 1 (left) and 2 (right): coordination environment within Cd2 cluster with atom-labelling scheme and bond lengths..................................................100 Figure S6. Arrangement of adjacent layers in the X-ray crystal structure of 1;. view along the c axis. Hydrogen atoms were omitted for clarity..............................................................100 Figure S7. The oba linker conformation in 1 (top) and 2 (below). Hydrogen atoms were omitted for clarity..............................................................................................101 Figure S8. N2 (77 K) adsorption isotherms for the samples of 1 activated in different ways.......101 Figure S9. Two cycles of N2 (77 K) adsorption isotherm for 1, before each cycle the sample was activated at room temperature for 16 hours...........................................................102 Figure S10. Three cycles of N2 (77 K) adsorption isotherm for 2, before each cycle the sample was activated at room temperature for 16 hours...........................................................102 Figure S11. PXRD patterns of material 1 soaked in water, and after that (as 1H2O) in DMF.............103 Figure S12. Figure 6 CdMOF-1: QE-TPDA profiles of water vapor desorption-adsorption cycles at 200 oC (left), desorption-adsorption isotherms (middle) and stability of porosity upon repeated water vapor desorption-adsorption cycle (right)............................................................103 Figure S13. PXRD patterns (left) and IR spectra (right) for 2 soaked in water and in DMF-H2O (9:1 v/v) mixture. The band at ~ 3500 cm-1, indicated by * (IR spectrum), is ascribed to the presence of highly ordered water molecules positioned by strong hydrogen-bonds with acylhydrazone groups. . 103 Figure S14. TG curves for 1H2O ({[Cd2(oba)2(pcih)2]- 3H2O}n) [left] and 2H2O ({[Cd2(oba)2(tdih)2]-13H2O}n) [right]................................................................104 I 98 Figure S15. 'H NMR spectrum of terephthalaldehyde di-isonicotinoylhydrazone (tdih)...........104 Figure S16. IR spectrum of terephthalaldehyde di-isonicotinoylhydrazone (tdih)...................105 Table S1. Experimental and calculated elemental analysis for compounds 1 and 2 obtained by different mchanochemical variants (indicated as 'stepwise' or 'one step')..................................100 Figure S1. Various synthetic routes to {[Cd2(oba)2(tdih)2]-7H2O-6DMF}n (2). Figure S2. IR spectra and PXRD patterns of 1 (left) and 2 (right) synthesized via liquid-assisted grinding from CdO as a one-step (red) or a two-step reaction (green), compared with the pattern calculated from SCXRD data (black) and with materials obtained in solution (blue). I 99 Figure S3. IR spectra and PXRD patterns of 1 (left) and 2 (right) synthesized via liquid-assisted grinding from Cd(OH)2 as a one-step (red) or a two-step reaction (green), compared with the pattern calculated from SCXRD data (black) and with materials obtained in solution (blue). Figure S4. PXRD pattern of the intermediate product [Cd(^-H2O)(oba)(H2O)]n: calculated (bottom), obtained from CdO (middle) and Cd(OH)2 (top). I 100 Table S1. Experimental and calculated elemental analysis for compounds 1 and 2 obtained by different mchanochemical variants (indicated as 'stepwise' or 'one step'). Figure S5. X-ray crystal structures of 1 (left) and 2 (right): coordination environment within Cd2 cluster with atom-labelling scheme and bond lengths. Figure S6. Arrangement of adjacent layers in the X-ray crystal structure of 1;. view along the c axis. Hydrogen atoms were omitted for clarity. Sample N [%] C [%] H [%] 1 calculated 10.92 51.92 4.21 1 (stepwise) from Cd(OH)2 10.47 51.23 3.648 1 (one step) 10.68 51.62 3.835 1 (one step) from CdO 12.18 47.05 5.353 1 (stepwise) 11.39 49.2 5.029 2 calculated 12.71 51.51 4.77 2 (stepwise) from Cd(OH)2 12.65 50.97 4.898 2 (one step) 13.42 49.96 5.758 2 (one step) from CdO 12.44 49.67 5.151 2 (stepwise) 12.26 50.19 5.156 I 101 Figure S7. The oba linker conformation in 1 (top) and 2 (below). Hydrogen atoms were omitted for clarity. Figure S8. N2 (77 K) adsorption isotherms for the samples of 1 activated in different ways I 102 Figure S9. Two cycles of N2 (77 K) adsorption isotherm for 1, before each cycle the sample was activated at room temperature for 16 hours. Figure S10. Three cycles of N2 (77 K) adsorption isotherm for 2, before each cycle the sample was activated at room temperature for 16 hours. I 103 Figure S11. PXRD patterns of material 1 soaked in water, and after that (as 1H2O) in DMF. Figure S12. MOF-1: QE-TPDA profiles of water vapor desorption-adsorption cycles at 200 oC (left), desorption-adsorption isotherms (middle) and stability of porosity upon repeated water vapor desorption-adsorption cycle (right). Figure S13. PXRD patterns (left) and IR spectra (right) for 2 soaked in water and in DMF-H2O (9:1 v/v) mixture. The band at ~ 3500 cm-1, indicated by * (IR spectrum), is ascribed to the presence of highly ordered water molecules positioned by strong hydrogen-bonds with acylhydrazone groups. I 104 Figure S14. TG curves for 1H2O ({[Cd2(oba)2(pcih)2]-3H2O}n) [left] and 2H2O ({[Cd2(oba)2(tdih)2]- 13H20}n) [right]. Figure S15. 'H NMR spectrum of terephthalaldehyde di-isonicotinoylhydrazone (tdih). I 105 Figure S16. IR spectrum of terephthalaldehyde di-isonicotinoylhydrazone (tdih). I 106 Oświadczenie Ja, Dariusz Matoga, oświadczam, że moim wkładem w prace: I. K. Roztocki, I. Senkovska, S. Kaskel, D. Matoga "Carboxylate-Hydrazone Mixed-Linker Metal- Organic Frameworks: Synthesis, Structure and Selective Gas Adsorption", Eur. J. Inorg. Chem.. 2016. 4450-4456. — invited article from a cluster issue on "Metal-Organic Frameworks Heading Towards Application". II. K. Roztocki. D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga, "Isophtalate- hydrazone 2D zinc-organic framework: crystal structure, selective adsorption and tuning of mechanochemical synthetic conditions", Inorg. Chem.. 2016. 55. 9663-9670. III. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga "Effect of Linker Substituent on Layers Arrangement, Stability, and Sorption of Zn-Isophthalate/Acylhydrazone Frameworks” Cryst. Growth Des.. 2018. 18 488^197. IV. K. Roztocki, M. Lupa, A. Sławek, W. Makowski, I. Senkovska, S. Kaskel, D. Matoga, "Water- stable Metal-organic Framework with Three Hydrogen-bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. Chem.,_2018. 57, 3287-3296. V. K. Roztocki, M. Lupa, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga "Bulky substituent and solvent-induced alternative nodes for layered Cd-isophthalate/acylhydrazone frameworks" CrvstEneComm. 2018, 20. 2841-2849. VI. K. Roztocki, M. Szufla, M. Hodorowicz, 1. Senkovska, S. Kaskel and D. Matoga " Introducing a longer versus shorter acylohydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties" Było: konsultacja koncepcji pracy i dyskusja wyników badań, nadzór nad poprawnością merytoryczną i jakością pracy, redagowanie ostatecznej wersji artykułów, korespondencja ze współpracownikami z Drezna i z redakcjami czasopism. I 107 Oświadczenie Ja, Maciej Hodorowicz, oświadczam, że moim wkładem w prace: I. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga, "Isophtalate- hydrazone 2D zinc-organic framework: crystal structure, selective adsorption and tuning of mcchanochemical synthetic conditions", Inorg. Chem.. 2016. 55. 9663-9670. ü. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga "Effect of Linker Substituent on Layers Arrangement, Stability', and Sorption of Zn-Isophthaiaie/Acylhydrazone Frameworks" Cryst Growth Des., 2018. 18 488-497. III. K. Roztocki, M. Lupa, M. Hodorowicz, 1. Senkovska, S. Kaskel and D. Matoga "Bulky substituent and solvent-induced alternative nodes for layered Cd-isophthalate/acylhydrazone frameworks" CrystEngComm, 2018. 20.2841-2849. ’ IV. K. Roztocki, M. Szufla, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga " Introducing a longer versus shorter acylohydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties" Dalton Wykonanie pomiarów dyfrakcji rentgenostrukturalnej na materiałach monokrystalicznych (SC-XRD) oraz w oparciu o zebrane dane rozwiązanie i udokładnienie struktury krystalicznej opublikowanych sieci metalo-organicznych. Było: I 108 Oświadczenie Ja, Damian Jędrzejowski, oświadczam, że moim wkładem w prace: I. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga, "Isophtalale- hydrazone 2D zinc-organic framework: crystal structure, selective adsorption and tuning of mechanochemical synthetic conditions", Inorg. Chem.. 2016. 55. 9663-9670. II. K. Roztocki, D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga "Effect of Linker Substituent on Layers Arrangement, Stability, and Sorption of Zn-Isophthalate/Acylhydrazone Frameworks" Cryst. Growth Des.. 2018. 18 488-497. Było: Opracowanie warunków mechanosyntezy i charakterystyka produktów (I), zaprojektowanie, opracowane syntezy mokrej i mechanochemicznej oraz charakterystyka produktów (II) a także pomoc w opracowaniu manuskryptów, zwłaszcza w kontekście przygotowania rysunków, wykonywania przeglądu literaturowego i opisu wykonanych eksperymentów (I i II). I 109 Oświadczenie Ja, Andrzej Sławek, oświadczam, że moim wkładem w prace: I. K. Roztocki, M. Lupa, A. Sławek, W. Makowski, I. Senkovska, S. Kaskel, D. Matoga, "Water-stable Metal-organic Framework with Three Hydrogen-bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. Chem.. 2018. 57. 3287-3296. Były: • pomiary adsorpcji wody w badanym materiale oraz testy jego stabilności metodą QE- TPDA. Opracowanie wyników oraz pomoc w ich interpretacji. • symulacje molekularne adsorpcji wody oraz dwutlenku węgla w badanym materiale. Opracowanie wyników oraz pomoc ich interpretacji. • pomoc w przygotowaniu manuskryptu w części dotyczącej eksperymentów QE-TPDA oraz symulacji molekularnych adsorpcji I 110 Oświadczenie Ja, Monika Szufla, oświadczam, że moim wkładem w prace: I. K. Roztocki, M. Szufla, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga Introducing a longer versus shorter acylohydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties" Było: • Pomoc w wykonaniu części eksperymentów w bezpośredniej współpracy z Kornelem Roztockim, • Pomoc w przygotowaniu suplementu do publikacji. I 111 Oświadczenie Ja, Wacław Makowski, oświadczam, że moim wkładem w pracę: I. K. Roztocki, M. Lupa, A. Sławek, W. Makowski, I. Senkovska, S. Kaskel, D. Matoga, "Water-stable Metal-organic Framework with Three Hydrogen-bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. Chem.. 2018. 57. 3287-3296. był udział w zaplanowaniu badań adsorpcji wody oraz trwałości hydrotermalnej tytułowego preparatu metodą kwazi-równowagowej termodesorpcji oraz w opracowaniu ich wyników. Uczestniczyłem także w pisaniu i redagowaniu odpowiednich fragmentów manuskryptu pracy. Kraków, 4 marca 2019 Drhah Wading Makowski I 112 C 'o-Author Statement I Irena Senkovska. hereby declare that my contribution to the papers I K. Kn/tocki. I Senkovska, S Kaskel, D Maloga "Carboxylate-Hydrazone Mixed-l.inker Metal-Organic Frameworks Synthesis, Structure and Selective Gas Adsorption", lur. J. Inorg. ( hem.. 2016.44SO 4456, — invited article from a cluster issue on "Metal-Organic Frameworks Heading Towards Application" II k. Ko/loeki. I) Jędrzejowski, M Hodorowicz, I Senkovska, S Kaskel, D Matoga, "Isophtalate-hydrazone 2D zinc-organic framework crystal structure, se ective adsorption and tuning of mechanochemical synthetic conditions , norg. ( hem.. 2016. 55. 9663 -9670 III K. Koztocki, I) Jędrzejowski, M Hodorowicz, I Senkovska, S. Kaskel D Matoga "Effect of I.inker Substituent on l-ayers Arrangement, Stability, and Sorption o Zn-Isophthalate/Acyl hydrazone Frameworks" Cryst. (irowth l)es. Ł 2018 ,J 8 _48Sr4— IV K. Roztocki, M l.upa, A. Sławek, W Makowski, I. Senkovska, S. Kaskel, D Matoga, "Water-stable Mctal-orgamc Framework with Three Hydrogen-bond Acceptore Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. ('hem.J20\8, 57, 3287-3296 V K. Roztocki. M Lupa, M Hodorowicz, I Senkovska, S Kaskel and D Matoga "Bulky substituent and solvent-induced alterative no£s for lay«ed Cd- isophthalate/acylhydrazone frameworks ( rysthngÇ omfWŁ20l8, 20,2841 2849 VI K Ko/tocki M Szufla, M Hodorowicz, I Senkovska, S. Kaskel and D_ Matoga " Inuoducmg a longer versus shorter acylohydrazone linker to a metal-orgamc framework “reticular structures, diverse stability and adsorption properties was . performing th« p. »d»rp.,o» m——» d-V> ««I d— of the ,«»1« (.-VI, • manuscript proofreading (l-VI) 26 02 2019 Dr I Senkovska I 113 Co-Author Statement I. Stefan Kaskel. hereby declare that my contribution to paper: I. k. Roztocki. I. Senkovska. S. Kaskel. D. Matoga "Carboxylate-Hydrazone Mixed-Linker Metal-Organic Frameworks: Synthesis, Structure and Selective Gas Adsorption", Eur. J. Inorg. Chem., 2016. 4450-4456. — invited article from a cluster issue on "Metal-Organic Frameworks Heading Towards Application". II. K. Roztocki. D. Jędrzejowski, M. Hodorowicz, I. Senkovska, S. Kaskel, D. Matoga, "lsophtalate-hydrazone 2D zinc-organic framework: crystal structure, selective adsorption and tuning of mechanochemical synthetic conditions", Inorg. ChemJIO 16, 55, 9663-9670. III. k. Roztocki. D. Jędrzejowski, M. Hodorowicz, I. Senkovska. S. Kaskel, D. Matoga "Effect of Linker Substituent on Layers Arrangement, Stability, and Sorption of Zn-Isophthalate/Acylhydrazone Frameworks" Cryst. Growth Des.. 2018. 18 488-497. IV. k. Roztocki. M. Lupa. A. Sławek. W. Makowski. 1. Senkovska. S. Kaskel, D. Matoga, "Water-stable Metal-organic Framework with Three Hydrogen-bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. Chem.. 2018. 57. 3287-3296. V. k. Roztocki. M. Lupa. M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga "Bulky substituent and solvent-induced alternative nodes for layered Cd- isophthalate/acylhydrazone frameworks" CrystEngComm^2018. 20. 2841-2849. VI. k. Roztocki. M. Szufla, M. Hodorowicz, I. Senkovska, S. Kaskel and D. Matoga " Introducing a longer versus shorter acylohydrazone linker to a metal-organic framework: non-isoreticular structures, diverse stability and adsorption properties" Manuscript proofreading and consultation of results of gas adsorption measurements. Technische Universität Dresden Profetsur für Anocflamicbe Ch*inle I Prof. Dr. Stefan Kaskel BefQStraße 66 01069 Dresden I 114 Oświadczenie Ja, Magdalena Lupa. oświadczam, że moim wkładem w prace: 1 k. Roztocki. M. Lupa, M. Hodorowicz, 1 Senkovska, S. Kaskel and D. Matoga "Bulky substituent and solvent-induced alternative nodes for layered Cd- isophthalate/acylhydrazone frameworks" CrystEngComm. 2018.20. 2841-2849. II. k. Roztocki. M. Lupa, A. Slawek, W Makowski, I. Senkovska, S. Kaskel, D. Matoga, "Water-stable Metal-organic Framework with Three Hydrogen-bond Acceptors: Versatile Theoretical and Experimental Insights into Adsorption Ability and Thermo-hydrolytic Stability", Inorg. Chem.. 2018. 57. 3287-3296. • Pomoc w wykonaniu części eksperymentów w bezpośredniej współpracy z Kornelem Roztockim. • Pomoc w przygotowaniu suplementu do publikacji. Było: