A general and facile strategy for precisely controlling the crystal size of monodispersed metal–organic frameworks via separating the nucleation and growth

Xiaocheng Lan , Ning Huang , Jinfu Wang and Tiefeng Wang *
Beijing Key Laboratory of Green Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China. E-mail: wangtf@tsinghua.edu.cn; Tel: +86-10-62794132

Received 25th October 2017 , Accepted 1st December 2017

First published on 1st December 2017

A novel and facile strategy was proposed for precisely controlling the crystal size of ZIF-8, ZnCo bimetallic ZIFs and also NPs@ZIF-8 by separating the nucleation and growth. The fundamentals of crystallization of this strategy were elaborately investigated, and the size effect on the internal diffusion was explored for a series of Pt@ZIF-8 catalysts.

Metal–organic frameworks (MOFs), a class of porous materials with modifiable functions, have drawn great interest in many applications such as storage, gas separation and catalysis.1–3 Manipulation of the attributes of MOFs, such as the pore size, crystal shape and crystal size, is an important factor in optimizing the properties of MOFs for specific applications.4–6 In particular, the synthesis of monodispersed MOFs with a precisely tunable crystal size is important for studying the sorption kinetics or molecular diffusion,7–9 understanding the crystallization fundamentals,10–12 and investigating the size effect on the properties of MOFs and MOF-derived materials.13–16 Moreover, tuning the MOF crystal to nanosize plays a crucial role in applications such as sorption, membranes, drug delivery and catalysis, where the internal diffusion limitation is reduced due to the decreased diffusion length.17–20

Various methods have been reported for manipulating the crystal size of MOFs and they can be divided into two groups.4,5 The first group includes adjusting the precursor concentration or synthesis conditions. Nano-MOFs were prepared through ultrasonic treatment or by diluting the precursor solutions, but these methods significantly decreased the yield of MOFs and were difficult for precisely controlling the crystal size.21,22 The second group includes the use of additives to modulate the crystallization.11,23 However, this approach needed the combination of several methods to precisely control the crystal size of some MOFs.24 In addition, it is unsuitable for tuning the crystal size of multifunctional MOFs, such as NPs@MOFs, because the additives were unfavorable for encapsulating the NPs into the MOF matrix.25 Although some nanoMOFs have been prepared with the above methods, the precise control of the crystal size of MOFs, especially multifunctional MOFs, still remains a challenge. Recently, bimetallic MOFs (BMOFs) have shown enhanced properties and promising applications,26–29 but the crystal size of BMOFs depends on the bimetallic proportion.29,30 Therefore, it is of great importance to develop a general and facile method for precisely controlling the crystal size of MOFs at a specified bimetallic proportion.

It has been well documented that MOF crystallization involves nucleation and growth processes that determine the crystal size.5,31,32 The nucleation always proceeds parallel to the growth, thus fast nucleation results in small crystals.5,11,31 Based on the mechanism of crystallization, it is promising to control the crystal size via separating the nucleation and growth so as to individually modulate these two processes. Herein we propose a general and facile strategy for precisely tuning the crystal size of MOFs. As illustrated in Scheme 1, compared to the conventional synthetic method of MOFs, a small amount of metal node was pre-added to the organic ligand solvent before the total metal nodes were introduced. This small amount of metal node interacted with the organic ligand to form clusters, which acted as the seeds for later crystal growth. The crystal size can be precisely controlled by changing the amount of pre-added metal node. Such a unique strategy is efficient in fine-tuning the crystal size of ZIF-8, ZnCo-BMOF and also multifunctional NPs@ZIF-8. The fundamentals of crystallization of this unique strategy were elaborated using Pt@ZIF-8 as an example. The prepared Pt@ZIF-8 catalysts of different sizes were used to investigate the internal diffusion in ZIF-8, and the activity was found to increase by 11 fold when the crystal size was reduced from 440 nm to 45 nm.

image file: c7cc08244d-s1.tif
Scheme 1 Illustration of MOF crystallization for the conventional method and the nucleation and growth separated method.

Based on the transmission electron microscopy (TEM) results (Fig. S2–S5, ESI), the particle size was plotted as a function of the mole ratio between pre-added Zn2+ and the organic ligand (R), as shown in Fig. 1. The samples were denoted as MOF_X, where X was the mean crystal size determined by TEM. Generally, the crystal size of the MOF prepared by the conventional method (R = 0) is affected by the total proportion of the metal node and ligand,4,5 and these MOFs had different initial crystal sizes (ZIF-8: 744 nm; Zn50Co50-BMOF: 128 nm; Zn20Co80-BMOF 231 nm; Pt@ZIF-8: 440 nm). For ZnCo-BMOF, it has been reported that the crystal size depends on the proportion of Co/Zn and the decrease of the Zn2+ metal node always increases the crystal size,29,30 which is consistent with the present results. To the best of our knowledge, the precise control of the ZnCo-BMOF size at a specific Co/Zn ratio has not been reported.

image file: c7cc08244d-f1.tif
Fig. 1 Dependence of the particle size on the mole ratio of the pre-added Zn2+/organic ligand (R).

After pre-adding a small amount of Zn2+ (R < 0.02), the crystal size of the MOFs decreased from the initial size and showed a strong dependence on the value of R. By increasing the value of R, the crystal size of these MOFs can be tuned to the nanoscale (<50 nm), which is important for enhancing the properties of MOFs and MOF-derived materials.6 Cravillon et al.12 prepared ZIF-8 of 18–50 nm in the presence of an n-butylamine additive. Herein, monodispersed ZIF-8 of 27 nm was obtained by optimizing the R value (0.018) without any additives. It has been reported that the PVP protected NPs were beneficial for the nucleation of the seeds in the crystallization of ZIFs.25 The nucleation threshold of ZIFs was decreased by the addition of Pt NPs. Thus, Pt@ZIF-8 showed a quite different dependence of particle size on the R value compared with ZIF-8. The crystal structure of each series of MOFs was confirmed by X-ray diffraction (Fig. S6–S9, ESI). All MOFs with decreased sizes showed identical reflections of the MOFs prepared by the conventional method. The uniform element distribution of Co and Zn and the Co/Zn mole proportion of ZnCo-BMOFs were determined by elemental mapping and EDS, respectively (Fig. S10 and S11, ESI). The Zn and Co elements coexisted and were highly dispersed, indicating that the self-assembly of Zn2+ and Co2+ with the ligand was not affected by the pre-added metal nodes. Note that the Co/Zn ratio is slightly increased in Zn50Co50-BMOF (Fig. S10, ESI) and Zn20Co80-BMOF (Fig. S11, ESI) due to the increased amount of pre-added Zn2+ ions. For the multifunctional MOF, the N2 adsorption isotherm (Fig. S12, ESI) and thermogravimetric analysis (Fig. S13, ESI) showed that the Pt@ZIF-8 catalyst with a decreased crystal size preserved the micro-porosity and good thermal stability. The distribution of the Pt NPs in Pt@ZIF-8 was determined using TEM images (Fig. 2) and the HRTEM images of the fringes of the catalysts (Fig. S14, ESI). Most of the Pt NPs were encapsulated with good dispersion in the inner space of ZIF-8. Pt@ZIF-8 with R = 0.0044 was reproduced three times, and the results are listed in Table S2 (ESI). The samples prepared using different batches showed a very similar particle size, indicating good reproducibility.

image file: c7cc08244d-f2.tif
Fig. 2 TEM and SEM images of the Pt@ZIF-8 crystal. (a) Pt@ZIF-8_440, (b) Pt@ZIF-8_204, (c) Pt@ZIF-8_111, (d) Pt@ZIF-8_60, (e) Pt@ZIF-8_45, and (f) Pt NPs. (The scale bars in the SEM images are 2 μm).

To understand the fundamentals of crystallization of this unique strategy, we carefully investigated the crystallization processes using Pt@ZIF-8 as an example. The crystallization of Pt@ZIF-8 was similar to that of other MOFs, except that the Pt NPs were pre-mixed with organic ligands. The samples in the solution after pre-adding a small amount of Zn2+ were characterized by HR-TEM (Fig. S15, ESI). Pt NPs can interact with 2-methylimidazole through hydrophobic interactions between the apolar group of PVP and the organic linkers,25 but cannot further form amorphous clusters of Pt, metal node and ligand composites due to the lack of the metal node (Zn2+). Thus, no amorphous clusters were formed in the conventional MOF synthesis method (Fig. S15a, ESI). In contrast, some amorphous nanoclusters were formed after pre-adding a small amount of Zn2+ (Fig. S15b–e, ESI). These surface-activated clusters acted as crystallization seeds and grew fast and simultaneously into Pt@ZIF-8 crystals after the total Zn2+ precursors were added. With the increase in the R value, more amorphous clusters were formed as crystallization seeds, which consumed the same amount of metal nodes and ligand precursors, thus resulting in a decreased crystal size.

Fig. S16 (ESI) shows the photographs at different times during the growth process of Pt@ZIF-8_X. Generally, after the total Zn2+ precursor is added, the brown transparent solution becomes turbid due to the formation of Pt@ZIF-8,25 followed by the settling of the particles when the precursors are depleted. For Pt@ZIF-8_440 (Fig. S16a, ESI), the nucleation process required a longer time and the solution became obviously turbid after 10 min. For Pt@ZIF-8_X (X = 204, 111, 60 and 45) (Fig. S16b–e, ESI), the seeds were pre-formed and the crystal growth was faster than the nucleation, thus the solutions of Pt@ZIF-8_X (X = 204, 111, 60 and 45) became turbid faster compared to the solution of Pt@ZIF-8_440. Upon increasing the R value, the time at which the solution became turbid decreased, indicating that more amorphous clusters were nucleated. The products were washed with methanol and collected by centrifugation. Interestingly, the weight of the obtained product was the same and the Pt metal loading detected by ICP analysis was about 1.0% for all the samples (Table S2, ESI). It was noted that the added Pt NPs in the synthesis of Pt@ZIF-8_X were the same, and the colorless and transparent upper phase obtained after 24 h (Fig. S16, ESI) indicated that most of the Pt NPs were incorporated into the framework of ZIF-8. These results showed that the yield of Pt@ZIF-8 was also good via this new synthesis strategy. The effect of the NPs was investigated by Pt@ZIF-8 with different metal loadings, and the results (Fig. S17, ESI) indicated that the NPs were beneficial for the nucleation. This approach was also successfully extended to synthesize other M@ZIF-8 catalysts (M = Ru, Pd, and Rh) (Fig. S18, ESI).

The successful precise control of the crystal size using our strategy depended on two key factors. One factor was the relatively slow nucleation in this process. For the synthesis of ZIFs, the nucleation process was the controlling step of crystallization12,33 and was slower than the growth.4,31 Thus, the pre-nucleated seeds can grow fast and simultaneously into crystals with a narrow size distribution. The other factor is the separation of the nucleation and growth processes. In this conventional synthetic approach, the nucleation proceeds parallel to the growth. Thus, with sufficient metal nodes and ligand precursors, the seeds usually grow to a normal size that is determined by the total proportion of the metal node and ligand. In our strategy, a small amount of metal node, which was sufficient for the threshold of nucleation but was not enough for further crystal growth, was pre-added to facilitate the formation of clusters as seeds, and the number density of the seeds could be controlled by the R value.

Recently, encapsulating NPs into the inner space of MOFs resulted in outstanding performance in size-selective reactions, hydrogenation and oxidation reactions.25,34–36 However, the incorporation of NPs inside MOFs inherently limits the reactant diffusion rate due to the small pore size and thick shell of the MOFs, resulting in diffusion limitation and low catalytic activity.37,38 The precisely controlled crystal size of Pt@ZIF-8 with the same metal loading encouraged us to explore the molecular internal diffusion in MOFs.

Previous works have shown that the uniform pores of MNPs@ZIF-8 had molecular sieving ability, which was applied to the size-selective hydrogenation of alkenes.25,39,40 In the current work, the liquid-phase hydrogenation of 1-hexene, cyclohexene and cyclooctene were carried out to investigate the size-selective properties of Pt@ZIF-8 (Table S3, ESI). In these reactions, the reactants with a small molecular size, such as 1-hexene, can diffuse through the pore apertures of ZIF-8, while the reactants with a molecular size larger than the aperture size, such as cyclooctene, are prevented from adsorbing on the Pt NPs encapsulated in ZIF-8. The comparison of the catalysts was based on a comparable conversion of 1-hexene (∼80%). For Pt/SiO2 without size selectivity, the conversion of cyclohexene and cyclooctene were 11.6% and 3.8%, respectively. For the Pt@ZIF-8_X catalysts, no reaction of cyclooctene was detected, indicating that the Pt NPs were well encapsulated in the inner space of ZIF-8 for all sizes of Pt@ZIF-8. It was noted that although cyclohexene had a larger molecular size (0.42 nm) than the aperture size of ZIF-8 (0.34 nm), some cyclohexene was hydrogenated. This was ascribed to the flexibility of the ZIF-8 framework.34 However, because the molecular size of cyclohexene was similar to the aperture size of ZIF-8, the internal diffusion of cyclohexene was significantly limited, and as a result the conversion of cyclohexene over Pt@ZIF-8 (0.2–2.4%) was much lower than that over Pt/SiO2 (11.6%).

The hydrogenation of 1-hexene was carried out as an example to investigate the internal diffusion limitation over Pt@ZIF-8_X (Fig. 3a). Obviously, Pt@ZIF-8_440 with the largest catalyst size (440 nm) displayed the lowest activity, with a conversion of only 63.2% after 7 h. The Pt@ZIF-8_45 catalyst with a catalyst size of 45 nm exhibited the highest catalytic activity, and the reactant was completely hydrogenated within 5 h. For further quantitative discussion, the reaction rates (r) based on the total weight of Pt at a conversion of ∼50% were calculated (Table S4, ESI). The reaction rate increased by 11 folds, from 0.04 s−1 to 0.43 s−1, when the crystal size decreased from 440 nm to 45 nm. Based on the results in Table S4 (ESI), r/r0 is plotted as a function of D0/D in Fig. 3b, indicating that the reaction rate increased linearly with decreasing crystal size. These results indicated that the reaction activity over Pt@ZIF-8 was controlled by internal diffusion, therefore it was efficiently enhanced by decreasing the crystal size.

image file: c7cc08244d-f3.tif
Fig. 3 Hydrogenation reaction of 1-hexene: (a) conversion versus reaction time, (b) correlation between D0/D and r/r0 calculated from Table S4 (ESI), where D0 and r0 are the crystal size and reaction rate of Pt@ZIF-8_440, respectively.

In summary, we have developed a facile and efficient strategy for precisely controlling the crystal size of MOFs. The nucleation and growth processes were separated and individually modulated by pre-adding a small amount of metal node in the ligand precursor to form amorphous clusters. These clusters acted as crystallization seeds and grew fast and simultaneously into MOF crystals after the addition of the total metal node precursors. The number density of the seeds was adjusted by changing the R value so as to precisely control the MOF crystal size. The hydrogenation over Pt@ZIF-8 catalysts of different sizes showed that the catalytic activity was controlled by internal diffusion, and the hydrogenation activity was enhanced by 11 folds after reducing the crystal size from 440 nm to 45 nm. This study opens new opportunities for understanding the fundamentals of crystallization, investigating the size effects of MOFs and MOF-derived materials and determining the high activities of multifunctional MOFs.

We acknowledge financial support by the National Natural Science Foundation of China (No. 21676155).

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. H. Furukawa, K. E. Cordova, M. O’Keeffe and O. M. Yaghi, Science, 2013, 341, 1230444 CrossRef PubMed.
  2. A. H. Chughtai, N. Ahmad, H. A. Younus, A. Laypkov and F. Verpoort, Chem. Soc. Rev., 2015, 44, 6804–6849 RSC.
  3. Y. Cui, B. Li, H. He, W. Zhou, B. Chen and G. Qian, Acc. Chem. Res., 2016, 49, 483–493 CrossRef CAS PubMed.
  4. V. Valtchev and L. Tosheva, Chem. Rev., 2013, 113, 6734–6760 CrossRef CAS PubMed.
  5. N. Stock and S. Biswas, Chem. Rev., 2011, 112, 933–969 CrossRef PubMed.
  6. M. Sindoro, N. Yanai, A.-Y. Jee and S. Granick, Acc. Chem. Res., 2013, 47, 459–469 CrossRef PubMed.
  7. H. Uehara, S. P. Diring, S. Furukawa, Z. Kalay, M. Tsotsalas, M. Nakahama, K. Hirai, M. Kondo, O. Sakata and S. Kitagawa, J. Am. Chem. Soc., 2011, 133, 11932–11935 CrossRef CAS PubMed.
  8. S. Furukawa, J. Reboul, S. Diring, K. Sumida and S. Kitagawa, Chem. Soc. Rev., 2014, 43, 5700–5734 RSC.
  9. Q. Liu, L.-N. Jin and W.-Y. Sun, Chem. Commun., 2012, 48, 8814–8816 RSC.
  10. B. Seoane, S. Castellanos, A. Dikhtiarenko, F. Kapteijn and J. Gascon, Coord. Chem. Rev., 2016, 307, 147–187 CrossRef CAS.
  11. S. Diring, S. Furukawa, Y. Takashima, T. Tsuruoka and S. Kitagawa, Chem. Mater., 2010, 22, 4531–4538 CrossRef CAS.
  12. J. Cravillon, R. Nayuk, S. Springer, A. Feldhoff, K. Huber and M. Wiebcke, Chem. Mater., 2011, 23, 2130–2141 CrossRef CAS.
  13. A. Laybourn, J. Katrib, R. S. Ferrari-John, C. G. Morris, S. Yang, O. Udoudo, T. L. Easun, C. Dodds, N. R. Champness and S. W. Kingman, J. Mater. Chem. A, 2017, 5, 7333–7338 CAS.
  14. T. Kiyonaga, M. Higuchi, T. Kajiwara, Y. Takashima, J. Duan, K. Nagashima and S. Kitagawa, Chem. Commun., 2015, 51, 2728–2730 RSC.
  15. N. L. Torad, M. Hu, Y. Kamachi, K. Takai, M. Imura, M. Naito and Y. Yamauchi, Chem. Commun., 2013, 49, 2521–2523 RSC.
  16. X. Cao, C. Tan, M. Sindoro and H. Zhang, Chem. Soc. Rev., 2017, 46, 2660–2677 RSC.
  17. A. Zirehpour, A. Rahimpour and M. Ulbricht, J. Membr. Sci., 2017, 531, 59–67 CrossRef CAS.
  18. C. Wang, J. Tuninetti, Z. Wang, C. Zhang, R. Ciganda, L. Salmon, S. Moya, J. Ruiz and D. Astruc, J. Am. Chem. Soc., 2017, 139, 11610–11615 CrossRef CAS PubMed.
  19. X. Lan, N. Huang, J. Wang and T. Wang, Catal. Sci. Technol., 2017, 7, 2601–2608 CAS.
  20. K. M. Taylor-Pashow, J. Della Rocca, R. C. Huxford and W. Lin, Chem. Commun., 2010, 46, 5832–5849 RSC.
  21. S. H. Jhung, J. H. Lee, J. W. Yoon, C. Serre, G. Férey and J. S. Chang, Adv. Mater., 2007, 19, 121–124 CrossRef CAS.
  22. Y. Pan, Y. Liu, G. Zeng, L. Zhao and Z. Lai, Chem. Commun., 2011, 47, 2071–2073 RSC.
  23. Y. Zhao, Q. Zhang, Y. Li, R. Zhang and G. Lu, ACS Appl. Mater. Interfaces, 2017, 9, 15079–15085 CAS.
  24. J. Zhou, X. Yu, X. Fan, X. Wang, H. Li, Y. Zhang, W. Li, J. Zheng, B. Wang and X. Li, J. Mater. Chem. A, 2015, 3, 8272–8275 CAS.
  25. G. Lu, S. Li, Z. Guo, O. K. Farha, B. G. Hauser, X. Qi, Y. Wang, X. Wang, S. Han and X. Liu, Nat. Chem., 2012, 4, 310–316 CrossRef CAS PubMed.
  26. K. Zhou, B. Mousavi, Z. Luo, S. Phatanasri, S. Chaemchuen and F. Verpoort, J. Mater. Chem. A, 2017, 5, 952–957 CAS.
  27. P. Zhang, B. Y. Guan, L. Yu and X. W. D. Lou, Angew. Chem., Int. Ed., 2017, 56, 7141–7145 CrossRef CAS PubMed.
  28. M. Ding, W. Shi, L. Guo, Z. Y. Leong, A. Baji and H. Y. Yang, J. Mater. Chem. A, 2017, 5, 6113–6121 CAS.
  29. Y. Z. Chen, C. Wang, Z. Y. Wu, Y. Xiong, Q. Xu, S. H. Yu and H. L. Jiang, Adv. Mater., 2015, 27, 5010–5016 CrossRef CAS PubMed.
  30. F. Hillman, J. M. Zimmerman, S.-M. Paek, M. R. Hamid, W. T. Lim and H.-K. Jeong, J. Mater. Chem. A, 2017, 5, 6090–6099 CAS.
  31. J. Cravillon, C. A. Schröder, R. Nayuk, J. Gummel, K. Huber and M. Wiebcke, Angew. Chem., Int. Ed., 2011, 123, 8217–8221 CrossRef.
  32. S. Hermes, T. Witte, T. Hikov, D. Zacher, S. Bahnmüller, G. Langstein, K. Huber and R. A. Fischer, J. Am. Chem. Soc., 2007, 129, 5324–5325 CrossRef CAS PubMed.
  33. S. R. Venna, J. B. Jasinski and M. A. Carreon, J. Am. Chem. Soc., 2010, 132, 18030–18033 CrossRef CAS PubMed.
  34. C.-H. Kuo, Y. Tang, L.-Y. Chou, B. T. Sneed, C. N. Brodsky, Z. Zhao and C.-K. Tsung, J. Am. Chem. Soc., 2012, 134, 14345–14348 CrossRef CAS PubMed.
  35. Y. Huang, Y. Zhang, X. Chen, D. Wu, Z. Yi and R. Cao, Chem. Commun., 2014, 50, 10115–10117 RSC.
  36. M. Zhang, Y. Yang, C. Li, Q. Liu, C. T. Williams and C. Liang, Catal. Sci. Technol., 2014, 4, 329–332 CAS.
  37. W. Zhang, Y. Liu, G. Lu, Y. Wang, S. Li, C. Cui, J. Wu, Z. Xu, D. Tian and W. Huang, Adv. Mater., 2015, 27, 2923–2929 CrossRef CAS PubMed.
  38. M. Zhao, K. Deng, L. He, Y. Liu, G. Li, H. Zhao and Z. Tang, J. Am. Chem. Soc., 2014, 136, 1738–1741 CrossRef CAS PubMed.
  39. P. Hu, J. V. Morabito and C.-K. Tsung, ACS Catal., 2014, 4, 4409–4419 CrossRef CAS.
  40. P. Wang, J. Zhao, X. Li, Y. Yang, Q. Yang and C. Li, Chem. Commun., 2013, 49, 3330–3332 RSC.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c7cc08244d

This journal is © The Royal Society of Chemistry 2018