Author Manuscript: Published in Chemical Communications, 2012, 48, 11868-11870,

DOI, 10.1039/C2CC36651G

Solution Mediated Phase Transformation (RHO to SOD) in Porous Co-Imidazolate based Zeolitic Framework with High Water Stability

Bishnu P. Biswal, Tamas Panda and Rahul Banerjee*†

Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XXDOI: 10.1039/b000000x

Here we report a highly porous, water stable Co based ZIF[CoNIm (RHO)] and its solution mediated phasetransformation to a less porous and water unstable ZIF[CoNIm (SOD)]. CoNIm (RHO) has high Langmuir surfacearea [2087 m2g-1] as well as high water adsorption [200cm3(STP)g-1] capacity.

Zeolitic Imidazolate Frameworks (ZIFs) are one sub class ofMetal Organic Frameworks (MOFs) where tetrahedral transitionmetals (Zn/Co) are connected via imidazolate (Im) type linkerswith different functionalities.1 Since discovery, ZIFs have shownnumerous potential applications like gas sorption, selectiveseparation, drug delivery, catalysis and sensing.2 Almost 30varieties of diverse net structures with several interesting zeolitictopologies (e.g. LTA, ANA, GME, SOD, RHO, DFT, MER, POZ,MOZ etc.) have been reported so far.1b, 2a, 3 However, most of theZIFs reported till date contain Zn(II) as metal centre with onlylimited report on Co(II) and Cd(II) based zeolitic frameworks.1b, 4

As a result, the chemical (especially water) stability of othertransition metal [Co(II), Cd(II) etc.] based ZIFs are stillunexplored. Here for the first time, we report a highly porous,stable Cobalt nitroimidazolate based ZIF [CoNIm (RHO)] thatfurther completely transfers to a known less porous CoNIm(SOD) phase (previously reported as ZIF-65).1b The unusual factwe report here is the high stability of CoNIm (RHO) in watercompare to CoNIm (SOD) [as well as ZnNIm (SOD)], althoughboth structures have similar chemical compositions [Co(NIm)2]as well as similar no of Co–N bonds[CoN4] per Co atom. CoNIm(RHO) ZIF shows high water vapour uptake at STP over all ZIFsreported so far.

CoNIm (SOD) have been synthesized by reacting 0.2 M 2-nitroimidazole (2-NIm) and Co(NO3)2.6H2O solution (2:1) inDMF at 120 C for 24 hours (h) (Figure 1 and Section 2, ESI†).However, when the same experiment was attempted in DEF, weobserved small crystals at early stages of synthesis 10-12 h withslightly different external morphology (rhombic dodecahedron)than CoNIm (SOD) (cube like). PXRD patterns indicate adifferent phase than CoNIm (SOD). In order to get the crystalstructure, we performed single crystal X-ray diffraction, whichrevealed a RHO net [CoNIm (RHO)], with large cage (Figure1b). However, when the same reaction mixture in DEF was left tocontinue for 24 h, we could obtain pure CoNIm (SOD) crystals(Figure 1d). The above observations motivated us to study thisphase transformation process in detail.

Figure 1. Scheme of synthesis for CoNIm (RHO) and CoNIm (SOD)ZIFs including digital photographs (a, d); nets (yellow ball indicates thefree space inside the framework) (b, e); CoNIm (RHO) and CoNIm(SOD) ZIF cages from single crystal XRD structure (c, f), with CoN4 pinktetrahedra [both sided arrow indicates the pore diameter (dp) and poreaperture (da) in each case]. H atoms have been omitted for clarity. C, grey;N, blue; O, red; Co, pink.

We investigated the entire phase transformation process till 24h by collecting solid samples (crystals) at each 3 h interval. After12 h of aging good quality crystals suitable for single crystal X-ray diffraction could be obtained. As the process goes on to 15 h,the sizes of CoNIm (RHO) crystals start decreasing. At 18 h, wefound a mixture of RHO and SOD phase. We continuedmonitoring the phase transformation and at 21 h, characteristicPXRD peaks of CoNIm (RHO) starts disappearing and relativepeak intensity of SOD phase increases (Figure 2a and Figure S1,ESI†). After 24 h of reaction pure CoNIm (SOD) crystals remainin the reaction media. We kept the reaction ongoing for 36 andsubsequently for 48 h. But no further change from CoNIm (SOD)phase has been noticed. In order to check the stability of theCoNIm (RHO) crystals; we added fresh 2-NIm (0.2 M, in DEF)solution to the dry CoNIm (RHO) crystals and kept it at 120 Cfor 24 h. This experiment does not result in any phase change.From these observations it is clear that during the period of 15 to18 h CoNIm (RHO) phase starts transforming to CoNIm (SOD)phase in the same reaction vial. However, CoNIm (RHO) is quitestable once the crystals are isolated from the reaction mixture(Section S3, ESI† for the detail PXRD study).

In order to understand this solution mediated RHO to SODphase transformation we collected Scanning Electron Microscopy(SEM) images of all the crystalline samples (Section S5, ESI†).These images indicate that, CoNIm (RHO) crystals have well

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

defined homogeneous rhombic dodecahedron shaped morphology(~10±5 µm) at the beginning (3 h). This phenomenon continuesup to 12 h (size 30±5 µm), but after that the crystal sizes startdecreasing and the shape changes from rhombic dodecahedron tosemi spherical (size 0.5–1 µm). At 21 h, small crystallitesagglomerate to form crystals with rhombic dodecahedron shapedmorphology (size 10±2 µm). As mentioned earlier, samplescollected after 18 h and 21 h of reaction contain mixture of RHOas well as SOD phase. After 24 h of reaction, these sphere shapedcrystallites take a cube like shape and size increases to 180±10µm.

CoNIm (RHO) crystallizes in Im-3m (cubic) space group with48 Co(II) ions within a unit cell [a, b, c ~ 29.0302(5) and , β, γ ~90] volume of 24465.3(7) Å3 (Section S8, ESI†). In the completestructure, each Co(II) ion is tetrahedrally-coordinated to fournitrogen atoms of four 2-NIm moiety to form an extended 3Dzeolitic (RHO) network (Figure 1c). The Co–NIm–Co anglevaries from 135.7 to 139.2 (reported RHO ZIFs withstands avariation from 141.8 to 151.3).The framework is composed ofcages with 8, 6, and 4 member ring windows. Major constituentof the framework are α cages comprised of 6 octagons, 8hexagons and 12 squares (Figure 1a). Each α cage is connected tosix other α cages by sharing double 8 member ring units. It isnoteworthy that CoNIm (RHO) adopts Im-3m symmetry,1c, 4, 6 asopposed to the smaller pore Pm-3m forms of RHO ZIFs [ZIF-11(Zn), ZIF-12(Co)] that have been reported previously.1a CoNIm(SOD) also crystallizes in a cubic space group (I-43m) [a, b, c~17.2 Å and , β, γ ~ 90] with unit cell volume of 5152.2(2) Å3.CoNIm (SOD) framework contains β cages (24 Co atoms),composed of 6 and 4 member ring windows [46.68] connected toneighbouring pores. In ZIFs, we noticed that imidazolefunctionalized at 4 and 5 positions (–CH3, –Cl, –CHO, –NH2, –CN, –C4H4, –C2N2 and –CH2OH)5 tend to produce RHO topologyin which these functional groups are directed towards the porewindows. As a consequence, it constricts the pore aperature(da~3-4 Å, dd~14-16 Å). On the other hand small and rigidfunctional groups (–CH3, –Cl, –Br, and –CHO)1d at 2 positionprefer to adopt the SOD topology (Figure S14, ESI†). CoNIm(RHO) has small 2-nitroimidazolate (2-NIm) as the linker, whererigid –NO2 group occupy the 2 position. Yet it forms both RHO aswell as SOD net. All nitro groups of 2-NIm in CoNIm (RHO) areaway from the pore windows of 8 and 6 member rings, whichresults large pore aperture (7.1 Å) and pore diameter (22.3 Å)(Figure 1c and Figure S13 in ESI†).

Several ZIFs (ZIF-8, ZIF-11 etc.)1a have shown significantchemical stability. However, as mentioned earlier all these ZIFscontain Zn(II), as the metal centre. It is noteworthy that waterstability of Co(II) based ZIFs has never been reported. Thereforewe investigated the water stability of CoNIm (RHO) in deionisedwater as well as in hot water. PXRD patterns of water treatedCoNIm (RHO) samples confirm that these crystals retain theircrystallinity in deionised as well as in hot water (85 C) for 7days.

Figure 2. (a) PXRD patterns representing the phase change at differentstages. (b) PXRD patterns showing water stability of CoNIm (RHO). (c)Water vapour adsorption isotherm of CoNIm (RHO) at STP (293K andP/Po=0.9).

In boiling water, however CoNIm (RHO) remain stable only for24 h, as beyond that the crystals disintegrate into a pinkcrystalline solid with different PXRD pattern (Figure S2, ESI†).On the other hand, CoNIm (SOD) crystals are unstable in waterand transfer into a different crystalline phase within 24 h (FigureS3, ESI†). This phenomenon is unusual as CoNIm (RHO) haslarger solvent accessible void (60.7%) and should, let the watermolecules react to Co centres, disintegrate much easily than theCoNIm (SOD) [solvent accessible void (51.5%)]. A possiblereason could be the hydrophobic nature of the 8 member rings inCoNIm (RHO), as all the –NO2 groups of 2-NIm remain awayfrom the pore windows. This could resist the easy approach ofwater molecule towards the pore (Figure S13, ESI†). All the cagewindows in CoNIm (SOD) are hydrophilic as the nitro groups aredirected towards the pore aperture and facilitate the hydrolysis ofCo(II) centre. There have been several reports of water adsorptionin MOFs,7a with only two reports on water sorption in Zn basedZIF [ZIF-71 and ZIF-8].7b, c CoNIm (RHO) shows high watervapour uptake of 200 cm3(STP)g-1, at a relative pressure (P/Po) of0.9 (Figure 2c). This water uptake is substantially higher than thewater uptake of ZIF-71 (~10 cm3g-1) and ZIF-8 (~150 cm3g-1)under similar experimental condition. Since CoNIm (SOD) [aswell as ZnNIm (SOD)] is unstable in water we were unable tocollect the water adsorption data. TGA analyses of as-synthesizedCoNIm (RHO), indicates an initial weight loss (32-35%) till 150C due to the escape of amide guest (DEF) from the pores,followed by a long plateau in the range 150-310 C, whichindicates the stability of the empty framework. The frameworkdecomposes suddenly after 310 C, within a narrow temperaturerange (5 C). In case of CoNIm (SOD), there is a continuousweight loss (20-22%) after 100 to 150 C; that could probably bedue to the loss of entrapped DEF/DMF molecules. Theframework starts decomposing after 310 C with further

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

Figure 3. (a) N2 adsorption isotherms of activated samples at 77 K temperature (space fill models indicate the pore opening for gas access); (b) H2

adsorption isotherms at 77 K temperature; (c) CO2 adsorption isotherms at 273 K (indicated by spheres) and 298 K (indicated by squares) temperatures. Afilled and open circle represents adsorption and desorption respectively.

weight loss of 30%. Variable temperature powder X-raydiffraction (VT-PXRD, 25-300 C) for both CoNIm (RHO) andCoNIm (SOD) indicate that CoNIm (RHO) framework is stableup to 300 C, without phase change. Whereas for CoNIm (SOD)framework after 150 C, an extra peak appears at 7 degree (2θ),which indicates some structural change (Section S4, ESI†).

The activated sample of CoNIm (RHO) exhibits type-I N2

adsorption isotherm with Brunauer-Emmet-Teller (BET) andLangmuir surface area of 1858 and 2087 m2g-1 respectively(Figure 3a). To the best of our knowledge these values are highestamong all ZIFs reported so far, except CdIF-1, 4 and 9.4 Similarlythe BET and Langmuir surface area of CoNIm (SOD) are 1097and 1235 m2g-1 respectively. High surface area of CoNIm (RHO)results due to the large cage [22.3 Å] with high pore aperture [(da)of 7.1 Å], which allow more N2 gas molecules to diffuse in to thecage. But in case of CoNIm (SOD) due to narrow pore aperture(3.4 Å) with pore diameter (10.4 Å); less amount of N2 gasmolecules can be absorbed into the cage, which impose lesssurface area. Since CoNIm (RHO) and CoNIm (SOD) has highsurface area, large pores and –NO2 functional groups withimidazolate linkers, we decided to collect H2 and CO2 adsorptionisotherms for both ZIFs. CoNIm (RHO) showed H2 uptake of 1.5wt% at 77 K and 1 bar pressure (Figure 3b). On the other handCoNIm (SOD) shows lower H2 uptake of 1.1 wt% at identicalconditions. However the initial uptake (P/Po<0.24) of CoNIm(SOD) is slightly higher than that of CoNIm (RHO). As thepressure approaches 1 bar, CoNIm (RHO) facilitates more H2 gasuptake. CoNIm (RHO) adsorbed 2.99 mmol g-1 (273 K) and 1.92mmol g-1 (298 K) of CO2 at 1 bar pressure (Figure 3c), whichoutperforms all other ZIFs except ZIF-96 (2.16 mmol g-1 at 298K),5 ZIF-78 (3.48 mmol g-1 at 273 K) and ZIF-69 (3.03 mmol g-1

at 273 K).2a The CO2 uptake for CoNIm (SOD) ZIF was 2.0 mmolg-1 (273 K) and 1.6 mmol g-1 (298 K) respectively at 1 barpressure.

In conclusion, we have isolated highly porous, water stableCo(II) based ZIF [CoNIm (RHO)] and observed solutionmediated phase transformation to a less porous SOD ZIF. CoNIm(RHO) showed BET and Langmuir surface area as high as 1858and 2087 m2g-1 respectively. These numbers stand very highamong all ZIFs (Zn/Co based) and zeolitic type materials reportedtill date. CoNIm (RHO) shows unusually high water stabilitycompared to CoNIm (SOD). To our knowledge CoNIm (RHO) isthe first example of porous Co based ZIF with high porosity(~2000 m2g-1) as well as water adsorption [200 cm3(STP)g-1] atP/Po=0.9. Also CoNIm (RHO) has H2 uptake capacity of 1.5 wt%at (77 K and 1 bar) along with CO2 uptake capacity of 2.99 mmolg-1 (273 K and 1 bar). We believe that these observations, on

understanding of ZIF to ZIF phase transformation, will provide apathway towards the synthesis of novel MOFs/ZIFs.

Notes and references†Physical/Materials Chemistry Division, CSIR-National ChemicalLaboratory, Dr. Homi Bhabha Road, Pune 411008, India; Tel:+912025902535; E-mail: r.banerjee@ncl.res.in.†Electronic Supplementary Information (ESI) available: Experimentalprocedures, crystallographic data (CIF) and additional supporting data.CCDC 886561. DOI: 10.1039/b000000x/

1. (a) K. S. Park, Z. Ni, A. P. Côté, J. Y. Choi, R. D. Huang, F. J. Uribe-Romo, H. K. Chae, M. O’Keeffe and O. M. Yaghi, Proc. Natl. Acad.Sci. U. S. A, 2006, 103, 10186; (b) R. Banerjee, A. Phan, B. Wang, C.Knobler, H. Furukawa, M. O’Keeffe and O. M. Yaghi, Science, 2008,319, 939; (c) J. Kahr, J. P. S. Mowat, A. M. Z. Slawin, R. E. Morris,D. Fairen-Jimenez and P. A. Wright, Chem. Commun., 2012, 48,6690; (d) K. Li, D. H. Olson, J. Seidel, T. J. Emge, H. Gong, H.Zeng, and J. Li, J. Am. Chem. Soc., 2009, 131, 10368; (e) F. Wang, Z.S. Liu, H. Yang, Y. X. Tan and J. Zhang, Angew. Chem. Int. Ed.,2011, 50, 450.

2. (a) R. Banerjee, H. Furukawa, D. Britt, C. Knobler, M. O’Keeffe andO. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 3875; (b) G. Lu, J. T.Hupp, J. Am. Chem. Soc., 2010, 132, 7832; (c) B. Zheng, J. Bai, J.Duan, L. Wojtas, M. J. Zaworotko, J. Am. Chem. Soc., 2011, 133,748.

3. A. Phan, C. J. Doonan, F. J. Uribe-Romo, C. B. Knobler, M.O’Keeffe and O. M. Yaghi, Acc. Chem. Res., 2010, 43, 58.

4. Y. Q. Tian, S. Y. Yao, D. Gu, K. H. Cui, D. W. Guo, G. Zhang, Z. X.Chen and D. Y. Zhao, Chem. Eur. J., 2010, 16, 1137.

5. W. Morris, B. Leung, H. Furukawa, O. K. Yaghi, N. He, H. Hayashi,Y. Houndonougbo, M. Asta, B. B. Laird, and O. M. Yaghi, J. Am.Chem. Soc., 2010, 132, 11006.

6. X. C. Huang, Y. Y. Lin, J. P. Jhang and X. M. Chen, Angew. Chem.Int. Ed., 2006, 45, 1557.

7. (a) A. Shigematsu, T. Yamada, and H. Kitagawa, J. Am. Chem. Soc.,2012, 134, 13145; (b) R. P. Lively, M. E. Dose, J. A. Thompson, B.A. McCool, R. R. Chanceab and W. J. Korosb, Chem. Commun.,2011, 47, 8667; (c) P. Küsgens, M. Rose, I. Senkovska, H. Fröde, A.Henschel, S. Siegle, S. Kaskel, Micropor. Mesopor. Mater., 2009,120, 325.

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90