Supporting Information (SI) for Cryst. Growth Des. DOI: 10.1021/acs.cgd.7b00892 Modulator effect in UiO-66-NDC (1,4-naphthalenedicarboxilic acid) synthesis and comparison with UiO-67-NDC isoreticular MOFs Vera V. Butova a,*, Andriy P. Budnyk a, Alexander A. Guda a, Kirill A. Lomachenko a,b, Aram L. Bugaev a,c, Alexander V. Soldatov a, Sachin M. Chavan d, Sigurd Øien-Ødegaard d, Unni Olsbye d, Karl Petter Lillerud d, Cesare Atzori c, Silvia Bordiga, c,d Carlo Lamberti a,e,* a International research center Smart Materials, Southern Federal University, 5 Zorge str., Rostov-on- Don, 344090, Russia. b European Synchrotron Radiation Facility, 71 avenue des Martyrs, CS 40220, Grenoble Cedex 9, 38043 France. c Department of Chemistry, NIS Interdepartmental Center, INSTM Reference center, University of Turin, Via P. Quarello 15, I-10135 Turin, Italy. d Department of Chemistry, University of Oslo, N-0371 Oslo, Norway e Department of Chemistry, CrisDi Interdepartmental Center, INSTM Reference center, University of Turin, Via P. Giuria 7, I-10125 Turin, Italy. *E-mail: vbutova@sfedu.ru; carlo.lamberti@unito.it S1
S1. Additional Rietveld refinement with fully occupancy of linker atoms In the main text, the refinements of the XRPD pattern of sample UiO-66-NDC sample were performed fixing the occupancy of the SBU atoms (Zr, O1 and O2) to unit and by fixing to 0.87 the occupancy of the carbon atoms of the NDC linker, according to TGA data. To further validate our refinment strategy we repeated the last refinment (see Figure 2b and Table 3) by fixing the occupancy factors of the carbon atoms of the NDC linker to unit; the results are reported in Table S1. Table S1. Atomic parameters resulting from the Rietveld Refinement of the UiO-66-NDC-0BC MOF in the Fm-3m space group (N. 225): fractional coordinates (x, y, z); isotropic atomic displacement parameters (Uiso), occupancy factors, site degeneration and number of atoms in the unit cell. Values of Rwp and Rp parameters are 0.80 and 0.85, respectively, and reduced 2 = 1.63 for 45 variables. The refined cell parameter is a = 20.850(1) Å [V = 9064(2) Å 3 ]. The occupancy factors of all atoms of have been fixed to unit atom x y z Uiso (Å 2 ) Occupancy factor site Atoms/unit cell Zr 0.1177(1) 0 0 0.0117(8) 1.00 b 24e 24 O1 0.1698(3) 0 0.0894(4) 0.021(3) 1.00 b 96j 96 O2 0.0647(4) х х 0.051(5) 1.00 b 32f 32 C11 0.1501(2) 0 х 0.003(4) 1.00 b 48h 48 C12 0.2009(2) 0 x 0.067(8) 1.00 b 48h 48 C13 0.2600(3) 0.0280(5) 0.1886(3) 0.042(7) 0.5 b 192l 96 C14 0.2721(5) 0.056(1) 0.1292(4) 0.24(3) a 0.25 b 192l 48 C15 0.3314(7) 0.084(2) 0.1169(6) 0.36(4) a 0.25 b 192l 48 a An unique U iso parameter for C14 and C15 atoms was set proportional to the distance from the rotational axis. b All occupancy factors were set to unit. Within less than 2 estimated standard deviations (esd), the atomic coordinates for all atoms are equivalent in both refinements; the same holds for the lattice parameter and the corresponding cell volume. The Uiso parameters of the SBU atoms (Zr, O1 and O2) are equivalent within less than one esd. As expected by an increase of the occupancy factor (from 0.87 to 1.00), all Uiso parameters of the carbon atoms of the linker increased; this is evident in table S1 for C12 and C13, while the changes in the Uiso for atoms C11, C14 and C15 were smaller that the corresponding error bars (e.g. for C11 Uiso moved from 0.0026 to 0.0033). We consequently decided to consider as final refinement that obtained by fixing the occupancies of the atoms of the NDC linkers to 0.87. S2
Intensity (a.u.) Intensity (a.u.) Intensity (a.u.) S2. Water stability tests In a typical test procedure, the sample powder was filled into a glass beaker with 15 ml of water and held under magnetic stirring at room temperature for 24 h. After that, precipitate was collected by centrifugation and dried at 40 C under vacuum. After XRPD measurement the same sample was activated at 200 C for 12 h and XRPD was measured again. The series of three XRD patterns (the initial, after contact with water and after successive activation) are reported in Figure S1 for all samples. There were no appreciable changes in the structure of the phase found in all the patterns, which undoubtedly means that samples preserved their crystal structure after each step of the test. (a) 0BC (3) (2) (b) (1) 10BC (3) (2) (c) (1) 60BC (3) (2) (1) 10 20 30 40 50 60 2 (deg) Figure S1. Water stability test of 0BC, 10BC, 60BC samples: (1) as synthesized, (2) after immersion in water for 24 h and successive drying at 40 C under vacuum, and (3) successive activation at 200 C for 12 h. = 1.5406 Å. S3
S3. Zr K-edge X-ray absorption study Zirconium K-edge XANES and EXAFS spectra were measured using Rigaku R-XAS spectrometer installed in the Southern Federal University, Russia. Bremsstrahlung was produced by an X-ray tube with fixed water-cooled W anode operating at 30 kv and 70 ma. White beam was monochromatized by Johansson-type Si(620) crystal with energy resolution of around 15 ev at 18 kev. The samples in powder form were pressed into pellets of 18 mm in diameter. The mass the samples was optimized to obtain the best signal-to-noise ratio in EXAFS spectra. Measurements were performed at room temperature in transmission mode using a single scintillation counter as a detector. In order to collect I0 data, the sample was removed from the beam and acquisition was repeated using the same parameters. The choice of this method was justified by an outstanding stability of the X-ray source. Total acquisition time per sample was around 3.5 hours. Subsequent data treatment was carried out in Athena code of Demeter package. 1 Figure S2. Zr K-edge XAFS spectra of UiO-66 (orange), 0BA (red) and 60BA (green). Part (a): Raw (not normalized) XANES spectra, where the edge jump reflects the amount of Zr in the sample. Part (b): k 2 -weighted (k) data. Part (c): k 2 -weighted, phase uncorrected, modulus of the Fourier transform of the (k) data reported in part (b) extracted in the range 3.8 10 Å -1 (b). Part (d): as part (c) for the imaginary part. Raw (i.e. not normalized) XANES spectra reported in Figure S2a reflect the lower Zr concentration in 60BA sample with respect to 0BA one. Indeed, if some cornerstones had not been missing there, the edge jump of 60BA sample would have been higher than the one of 0BA, due to in average lower amount of atoms in the linker molecule and therefore higher theoretical Zr concentration. However, experimental XANES edge jumps of 0BA and 60BA samples are 0.82 and 0.7, respectively, which proves the reverse, given the equal total mass of the samples as well as very similar structure and chemical composition. The data reported in Figure S7a prove the absence of S4
Weight loss, % SBU as schematized in Figure S5c (and Figure 2b of the main text) of about 15 %. EXAFS data (Figure S2b-d) confirm the similarity of the SBU in the 0BA and 60BA materials compared to a standard UiO-66 material. This testifies that the local environment of Zr is the same in all samples and corresponds to that of the standard UiO-66 sample. 2 3 S4. TGA Figure S3 reports the raw TGA data for whole set of UiO-66-NDC samples (see Table 1 of the main text): 0BA (red), 10BA (blue) and 60BA (green). The inset reports a magnification of the high temperature region, where ZrO2 is the only phase present in the sample. From this datum it emerges that the amount of Zr atoms present in the samples (i.e. the fractions of SBUs) increases in the order 0BA > 10BA > 60BA. This evidence supports the thesis of a higher fraction of missing SBUs in the UiO-66-NDC 60BA material, according to the scheme reported in Figure 6c. The same data are reported in the main text (Figure 8) rescaled defining 100 % as the remaining ZrO2 fraction after full combustion of the organic part. 100 0BC 10BC 90 60BC 80 70 60 34 32 50 40 550 600 30 0 100 200 300 400 500 600 700 Temperature ( C) Figure S3. Raw TGA data for 0BA (red), 10BA (blue) and 60BA (green) samples. The inset reports a magnification of the high temperature region, where ZrO 2 is the only phase present in the sample. S5
S5. Raman The comparison of the spectra reported in Figure S4 with the ones reported in the main text, Figure 10b, is showing the conversion of the linker in its deprotonated form, NDC 2, during the MOF synthesis. This is demonstrated by the presence of two bands, fingerprints for carboxylic acids, in the Raman spectra of H2NDC in DMSO solution (red curve in Figure S4) assigned to (C=O) and (C O) stretching modes falling respectively at 1706 cm 1 and 1260 cm 1. In solid state their position is redshifted for the (C=O) and blueshifted for the (C O) (at 1642 cm 1 and 1263 cm 1, respectively) due to strong hydrogen bonding. 4 These two bands are killed during the conversion to the conjugated base NDC 2 while a new band, falling at 1401 cm 1, can be tentatively assigned to the in-phase O C O symmetric stretching of carboxylates. The comparison of this data with the MOFs spectra reported in Figure 10b is confirming that all the linker used during the synthesis is converted to its carboxylate counterpart. Figure S4. Raman spectra of the linker used in the present work (1,4-Naphthalenedicarboxylic or 1,4-H 2NDC) in DMSO saturated solution (red), solid (blue) and its conjugated base, NDC 2 obtained titrating the powdered solid with 1M aqueous KOH (magenta). The spectrum of pure DMSO (black) is reported for the sake of comparison. References (1) Ravel, B.; Newville, M., J. Synchrot. Radiat. 2005, 12, 537-541. (2) Cavka, J. H.; Jakobsen, S.; Olsbye, U.; Guillou, N.; Lamberti, C.; Bordiga, S.; Lillerud, K. P., J. Am. Chem. Soc. 2008, 130, 13850-13851. (3) Valenzano, L.; Civalleri, B.; Chavan, S.; Bordiga, S.; Nilsen, M. H.; Jakobsen, S.; Lillerud, K. P.; Lamberti, C., Chem. Mat. 2011, 23, 1700-1718. (4) Colthup, N. B.; Daly, L. H.; Wiberley, S. E., Introduction to infrared and Raman spectroscopy (3 rd Ed.). Academic Press: San Diego, 1990. S6