Metal–organic frameworks with high working capacities and cyclic hydrothermal stabilities for fresh water production

Chemical Engineering Journal 286 (2016) 467–475

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Metal–organic frameworks with high working capacities and cyclic hydrothermal stabilities for fresh water production Seung-Ik Kim a, Tae-Ung Yoon a, Min-Bum Kim a, Seung-Joon Lee a, Young Kyu Hwang b,c, Jong-San Chang b,d, Hyun-Jong Kim e, Ho-Nyun Lee e, U-Hwang Lee b,c,⇑, Youn-Sang Bae a,⇑ a

Department of Chemical and Biomolecular Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Republic of Korea Center for Nanocatalysts, Green Chemistry and Engineering Division, Korea Research Institute of Chemical Technology (KRICT), Gajeong-ro 141, Yuseong-gu, Daejeon 305-600, Republic of Korea c Department of Green Chemistry, University of Science and Technology (UST), 217 Gajeong-Ro, Yuseong, Daejeon 305-350, Republic of Korea d Department of Chemistry, Sungkyunkwan University, Suwon 440-476, Republic of Korea e Surface Technology Group, Korea Institute of Industrial Technology (KITECH), Incheon 406-840, Republic of Korea b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 For medium humidity, MIL-100(M)

MOFs show high working capacities.  For high humidity, MIL-101(Cr)

exhibits a significantly high working capacity. III  All the MOFs consisted of only M sites show good cyclic water adsorption.  MIL-101(Cr) and MIL-100(M) MOFs are promising for fresh water production.

a r t i c l e

i n f o

Article history: Received 1 September 2015 Received in revised form 29 October 2015 Accepted 30 October 2015 Available online 4 November 2015 Keywords: Metal–organic frameworks (MOFs) Water adsorption Working capacity Cyclic adsorption Hydrothermal stability

a b s t r a c t In this work, we evaluated the working capacities of eight hydrothermally stable metal–organic frameworks (MOFs) for water adsorption under typical humidity conditions in three representative dry regions. Remarkably, three MIL-100(M) materials (M = Cr, Al, and Fe) and MIL-101(Cr) exhibited very high working capacities for medium and high humidity conditions due to their large surface areas. All of the MOFs consisting of only MIII sites (MIL-101(Cr), MIL-100(Cr), and MIL-100(Al)) showed good cyclic water adsorption/desorption performances and good hydrothermal stabilities. Due to the presence of FeII sites formed during activation at 250 °C, MIL-100(Fe) showed a considerable decrease in its water adsorption isotherm during the 2nd cycle although almost unchanged water uptakes were observed in the following cycles. When MIL-100(Fe) was activated at 150 °C (MIL-100(Fe)_150) to prevent formation of FeII sites, the sample showed good cyclic adsorption/desorption performance and good hydrothermal stability. Considering the high working capacities, cyclic adsorption/desorption behaviors, and good hydrothermal stabilities, MIL-101(Cr), MIL-100(Cr), MIL-100(Al) and MIL-100(Fe)_150 are promising adsorbents for producing drinking water in dry regions with medium or high humidity conditions during the night. Ó 2015 Elsevier B.V. All rights reserved.

⇑ Corresponding authors at: Center for Nanocatalysts, Green Chemistry and Engineering Division, Korea Research Institute of Chemical Technology (KRICT), Gajeong-ro 141, Yuseong-gu, Daejeon 305-600, Republic of Korea (U.-H. Lee), Department of Chemical and Biomolecular Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Republic of Korea. Tel.: +82 2 2123 2755; fax: +82 2 312 6401 (Y.-S. Bae). E-mail addresses: [email protected] (U-Hwang Lee), [email protected] (Y.-S. Bae). http://dx.doi.org/10.1016/j.cej.2015.10.098 1385-8947/Ó 2015 Elsevier B.V. All rights reserved.

468

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475

1. Introduction The water sorption behaviors of porous materials have been investigated for various applications, such as desiccant dehumidification [1–3], fresh water production in extremely dry regions [4–6], and heat-transformation processes [7–10]. For these applications, adsorbents with high water uptake capacities at desirable relative pressures and good cyclic adsorption/desorption performances are required. Until now, commercial adsorbents, such as activated alumina, silica gel, and zeolites have been used for these applications. However, those porous adsorbents have limited water uptake capacities and/or poor cyclic performances due to their moderate surface areas and strong interactions between water and the adsorbents. Recently, metal–organic frameworks (MOFs), which are synthesized by self-assembly of inorganic metal clusters and organic linkers, have emerged as a new class of porous materials [11–13]. MOFs have been vigorously studied as promising adsorbents for various applications, such as gas storage [14–18], separations [19–21], sensing [22,23] and catalysis [24] because of their extremely high surface areas, tunable pore sizes and adjustable internal surface properties [25]. As a result of these attractive features, MOFs may be considered as an alternative for water sorption applications. However, several issues related to the stability and reproducibility of MOFs remain because many MOFs have been found to be easily degradable under moist conditions. Because the water molecule has naturally strong nucleophilic sites, metal–ligand bonds in MOFs can be attacked by water molecules [26]. Fortunately, some types of MOFs, such as MIL-101(Cr) [27,28], MIL100(M) (M = Cr, Fe, and Al) [9,28,29] and UiO-66(Zr) [30–32] have been found to be remarkably stable under humid conditions due to their strong metal–ligand bonds and other structural features [26]. Recently, several MOFs with reasonable hydrothermal stabilities have been studied for dehumidification [28,33,34], water purification [35,36], adsorption heat-transformation [7,9,10] and fresh water production [28,37]. Yaghi and co-workers [37] evaluated the water adsorption performances of 23 materials, including 20 MOFs based on the following three criteria: (1) condensation pressure of water in the pores, (2) high water uptake capacity, and (3) good cycling performance and water stability. By considering all three criteria, they suggested that two zirconium-based MOFs, MOF-801-P and MOF-841, performed the best. Because MOF-801-P adsorbed a large amount of water at P/P0 = 0.1, they argued that this MOF was a good candidate for advanced thermal batteries. Because MOF-841 exhibited high water uptake at P/P0 = 0.3, they suggested that MOF-841 has the potential to be used to produce fresh water in remote desert areas. Most studies concerning water adsorption of MOFs have highlighted on the storage capacity at high humidity conditions, but for practical applications, the adsorbed amount at the discharge humidity condition should also be considered. The working capacity is defined by the difference in the adsorbed amount under adsorption conditions and the adsorbed amount under discharge conditions [38]. Therefore, the working capacity is more feasible for identifying the applicability of MOFs rather than the storage capacity. In the case of fresh water production in extremely dry areas, the concept of working capacity is well suited for the needed application. Based on the high working capacity of MOF-841 between P/P0 = 0.05–0.35, Yaghi and co-workers argued that MOF-841 is a good candidate to produce fresh water in desert areas, such as the city of Tabuk in Saudi Arabia where the typical RH conditions during the day and night are 5% and 35%, respectively [37]. In fact, RH conditions vary depending on the region and season of the year. For Pampas de La Joya located in southern Peru, the relative humidity (RH) during the day is approximately

0%, while the typical RH at night is 45% [39]. For the Mojave Desert in California, typical RHs in the afternoon and night are 30% and 55%, respectively, during the winter [40]. Therefore, investigating the correlation between the working capacities of MOFs and various RH conditions that resemble possible climatic conditions is necessary. In this work, we synthesized eight representative MOFs that are known to be hydrothermally stable [26]: UiO-66(Zr) [30–32] and its three derivatives (UiO-66(Zr)-NH2 [32,41], UiO-66(Zr)-OH [32], and UiO-66(Zr)-(OH)2 [32]), MIL-101(Cr) [27,28], and MIL100-M (M = Fe, Cr, Al) [9,28,29]. Although some of these MOFs have been extensively studied for water adsorption, the water working capacities of these MOFs in terms of reliance on varied humidity conditions have not been fully unveiled so far. Therefore, we measured the water adsorption isotherms of the eight MOFs and evaluated their working capacities under three different RH conditions that were all representative of real humidity conditions. For MOFs with high working capacities, we measured the cyclic water adsorption/desorption isotherms to examine if the working capacities were maintained even after repeated uses. Especially for MIL-100(Fe), we investigated the effect of the activation temperature on the cyclic water adsorption behaviors. Moreover, we compared the powder X-ray diffraction (PXRD) patterns of the as-synthesized samples and the used samples after cyclic adsorption/desorption to investigate the hydrothermal stabilities. For the comparisons, MOF-801-P and MOF-841, which have been reported to have good water adsorption properties, were also employed in this study [37]. 2. Experimental UiO-66(Zr) and its derivatives (UiO-66(Zr)-NH2, UiO-66(Zr)-OH, and UiO-66(Zr)-(OH)2) [32], MIL-101(Cr) [42], MIL-100(Fe) [43], MIL-100(Cr) [44], and MIL-100(Al) [28] were synthesized according to previous protocols in the literature. PXRD (powder X-ray diffraction) patterns were recorded with a Rigaku Miniflex (Rigaku Co., Japan) using nickel-filtered Cu Ka radiation (k = 1.5418 Å) from 2° < 2h < 40° in 0.02° steps at 1 s per step. PXRD patterns of the as-synthesized samples were measured to verify the purity of the crystalline phases. For selected MOFs, the PXRD patterns of the water-exposed samples were obtained to investigate changes in the crystalline structures after cyclic water adsorption. The BET surface areas of MOFs were calculated from N2 adsorption isotherms at 77 K, which were obtained by the Autosorb iQ system (Quantachrome Instruments, USA). For each measurement, approximately 50 mg of sample was used. Before determining the isotherms, as-synthesized UiO-66 and its derivatives were degassed under vacuum (<10 4 mbar) at 90 °C for 1 h followed by 150 °C for 3 h. MIL-101(Cr) and a series of MIL-100(M) (M = Fe, Cr, Al) were degassed under vacuum (<10 4 mbar) at 250 °C for 6 h. After measuring the N2 isotherms, BET surface areas were calculated in the linear range determined using the consistency criteria [45–47]: UiO-66(Zr): 0.003 < P/P0 < 0.068; UiO-66NH2: 0.003 < P/P0 < 0.070; UiO-66-OH: 0.002 < P/P0 < 0.050; UiO66-(OH)2: 0.003 < P/P0 < 0.071; MIL-101(Cr): 0.007 < P/P0 < 0.198; MIL-100(Fe): 0.007 < P/P0 < 0.196; MIL-100(Cr): 0.007 < P/ P0 < 0.202; MIL-100(Al): 0.005 < P/P0 < 0.097. The adsorption isotherms of water in the MOFs at 298 K were measured using an Autosorb iQ system (Quantachrome Instruments) with a specially designed water circulation system to maintain constant temperature. The water adsorption isotherms were measured up to a relative humidity of 0.8. For the bestperforming MOFs, the repeated water adsorption/desorption isotherms at 298 K were obtained for five cycles. Between each cycle,

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475

the samples were regenerated under vacuum at ambient temperature for 2 h. 3. Results and discussion 3.1. Material characterizations Powder X-ray diffraction (PXRD) patterns of the as-synthesized samples were consistent with those calculated from single crystal structures (Fig. 1) [28,32]. For UiO-66(Zr) and its derivatives, the PXRD patterns of the samples coincided with one another, which confirmed the isostructural feature. Additionally, the PXRD patterns of the isostructural MIL-100(M) materials were also consistent with each other. Fig. 2 shows the N2 adsorption/desorption isotherms for the eight MOFs at 77 K. As shown in Table 1, the obtained BET surface areas of the eight MOFs were comparable to the literature values [9,32,48–51]. 3.2. Water adsorption isotherms The water isotherms for UiO-66(Zr) and its isostructural series MOFs are compared in Fig. 3a. The water isotherms of MIL-101 (Cr) and a series of MIL-100(M) materials are shown in Fig. 3b. For the comparisons, Fig. 3a also presents the isotherms of MOF801-P and MOF-841, which were taken from the literature [37]. From a rigorous study of 23 adsorbents, Yaghi and co-workers [37] had concluded that MOF-801-P and MOF-841 were the best performers based on the working capacities (P/P0 = 0.05–0.3), recyclabilities and water stabilities of the materials. UiO-66(Zr) was synthesized by ZrCl4 and 1,4benzenedicarboxylate (BDC), which forms 12-coordinated zirconium oxo clusters in its defect-free form [26]. UiO-66(Zr) is a microporous material with a pore size of approximately 6 Å in diameter. In Fig. 3a, UiO-66(Zr) displays a type IV water isotherm, which indicates that this material was weakly hydrophobic at low relative humidity conditions. The micropore-filling step began at P/ P0 = 0.2 with very little adsorption below this point, and the uptake amount increased steeply to P/P0 = 0.45 followed by a modest increase to P/P0 = 0.8. The obtained water uptake at P/P0 = 0.8 was 588 cm3/g, which agreed with the value reported (525 cm3/ g) under similar conditions [37]. To investigate the effect of hydrophilic functionalization of UiO66(Zr) on water adsorption, a series of isostructural MOFs were synthesized. To prepare this series of MOFs, hydrophilic moieties, such as amine (–NH2) or hydroxyl groups (–OH or –(OH)2) were employed to functionalize the BDC linker. As presented in Fig. 3a,

469

all of the functionalized UiO-66(Zr) materials show enhanced water adsorption in the low RH region due to the augmented interactions with water molecules, leading to isotherms that approached type I shapes [26]. Therefore, hydrophilic functionalization of MOFs would be useful for on-board vehicle thermal batteries where water capture is desirable in the low RH region (P/ P0 < 0.1) [37]. However, water adsorption of the functionalized UiO-66(Zr) materials in the high RH region decreased significantly compared with the original UiO-66(Zr) material. This can be explained by the decreased surface area due to the insertion of bulky groups in the pores (Table 1). UiO-66(Zr) and its functionalized materials show some hysteresis in water adsorption and desorption isotherms, especially at low relative humidity conditions. This may be elucidated by rehydroxylation of Zr-clusters during the water adsorption, which was already observed in previous studies [52,53]. MIL-101(Cr) is a mesoporous MOF with two types of cages of 29 and 34 Å diameters. The mesoporous cages are formed by microporous supertetrahedral units that consist of three chromium trimers connected by terephthalate linkers [27]. As presented in Fig. 3b, MIL-101(Cr) displays a type IV water isotherm, indicating a moderate interaction energy between water and the MOF [26]. Remarkably, MIL-101(Cr) showed a significant step at P/ P0 = 0.45–0.55, possibly from capillary condensation in the mesopores. Although two steps were expected to be observed due to the presence of the two types of mesopores with diameters of 34 and 29 Å [50], only a single step was seen in our isotherm. Kusgens et al. (2009) reported the same observation and attributed it to the similar hydrophilicities of the pores [54]. Due to its large surface area (3124 m2/g), the water uptake of MIL-101(Cr) at P/P0 = 0.8 was significantly high (1825 cm3/g). To see the effect of unsaturated metal sites on water adsorption, a series of isostructural MIL-100(M) materials (M = Cr, Fe and Al) were synthesized. The MIL-100(M) MOFs were comprised of two types of mesopores (25 and 29 Å diameters) formed by an octahedral of M3+ trimers connected with a 1,3,5-benzenetricarboxylate (BTC) ligand [55]. As displayed in Fig. 3b, the water adsorption isotherms of MIL-100(Cr), MIL-100(Fe) and MIL-100(Al) almost coincided with one another. The isotherm types were similar to type IV, but two steps at which the water loadings noticeably increased were clearly observed at P/P0 = 0.25 and 0.4 for all three MOFs. These two steps revealed that capillary condensations occur in two types of mesopores step by step. The water uptakes of the three MIL-100(M) materials at P/P0 = 0.8 were approximately 820–900 cm3/g, which were approximately 45–50% compared with MIL-101(Cr). The water uptakes of MIL-101(Cr) and

Fig. 1. Comparisons of measured PXRD patterns for as-synthesized MOFs with the corresponding simulated PXRD patterns: (a) UiO-66(Zr), UiO-66(Zr)–NH2, UiO-66(Zr)–OH and UiO-66(Zr)–(OH)2; (b) MIL-101(Cr), MIL-100(Fe), MIL-100(Cr) and MIL-100(Al).

470

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475

Fig. 2. N2 adsorption/desorption isotherms of the eight MOFs at 77 K: (a) UiO-66(Zr), UiO-66(Zr)–NH2, UiO-66(Zr)–OH and UiO-66(Zr)–(OH)2; (b) MIL-101(Cr), MIL-100(Fe), MIL-100(Cr) and MIL-100(Al). Filled and empty symbols indicate adsorption and desorption isotherms, respectively.

Table 1 BET surface areas of the eight MOFs. BET surface area (m2/g) MOF

This study

Reported

UiO-66(Zr) UiO-66(Zr)–NH2 UiO-66(Zr)–OH UiO-66(Zr)–(OH)2 MIL-101(Cr) MIL-100(Fe) MIL-100(Cr) MIL-100(Al)

1473 1286 858 510 3250 1948 1842 1786

1400 [48] 1200 [32] 946 [49] 560 [32] 3124 [50] 1917 [9] 1900 [51] 1814 [9]

MIL-100(M) at P/P0 = 0.8 correlated well with the BET surface areas. These results indicated that the type of unsaturated metal in MIL-100(M) does not significantly affect water adsorption. However, in a later section, we will see that some interesting cyclic water adsorption/desorption behaviors depend on the type of unsaturated metal. 3.3. Working capacity For real applications, such as fresh water production the working capacity is more important than the storage capacity. The working capacity can be calculated from the difference in the water uptakes under adsorption and desorption conditions. To

investigate the applicability of the eight MOFs for different climate conditions, the working capacities were calculated from the water adsorption/desorption isotherms in Fig. 3 based on three different relative humidity (RH) conditions: (1) P/P0 = 0.05–0.35 (Case 1), (2) P/P0 = 0–0.45 (Case 2), and (3) P/P0 = 0.30–0.55 (Case 3). Cases 1–3 were based on the typical RH conditions in the city of Tabuk in Saudi Arabia [37], Pampas de La Joya, Atacama Desert in southern Peru [39], and Mojave Desert in California during the winter [40], respectively. Table 2 summarizes the obtained working capacities of the eight MOFs for the three different cases. For the comparisons, the working capacities of MOF-801-P and MOF-841 were also calculated from water isotherms in the literature [37]. For Case 1 (P/P0 = 0.05–0.35), a series of MIL-100(M) materials exhibited relatively high working capacities (277–358 cm3/g) because the first capillary condensation related to filling smaller pores began at P/P0 = 0.25 and reached plateaus with water uptakes of 413–480 cm3/g at P/P0 = 0.35. UiO-66(Zr) showed a modest working capacity (217 cm3/g) because a step in the isotherm began at P/P0 = 0.25 and was not close to saturation at P/ P0 = 0.35. Despite its high surface area, MIL-101(Cr) exhibited a minor working capacity because capillary condensation had not even begun at P/P0 = 0.35. For Case 1, MIL-100(Fe) and MIL-100 (Al) exhibited higher working capacities than MOF-801-P, but none of the eight MOFs studied showed higher working capacities than MOF-841. This is because MOF-841 had a negligible water uptake at P/P0 = 0.05 but a significant water uptake at P/P0 = 0.35.

Fig. 3. Water adsorption/desorption isotherms of the eight MOFs at 298 K: (a) UiO-66(Zr), UiO-66(Zr)–NH2, UiO-66(Zr)–OH and UiO-66(Zr)–(OH)2; (b) MIL-101(Cr), MIL-100 (Fe), MIL-100(Cr) and MIL-100(Al). For the comparisons, the water adsorption/desorption isotherms of MOF-801-P and MOF-841 were extracted from the literature [37]. Vertical dashed, dashed-dot and solid lines represent the RH conditions for Case 1, Case 2 and Case 3, respectively.

471

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475 Table 2 Working capacities of MOFs calculated from water adsorption/desorption isotherms based on three different RH region cases.a MOF

Case 1 (Tabuk) 0.05 6 P/P0 6 0.35

Case 2 (Pampas) 0.00 6 P/P0 6 0.45b

Case 3 (Mojave) 0.30 6 P/P0 6 0.55

This study UiO-66(Zr) UiO-66(Zr)–NH2 UiO-66(Zr)-OH UiO-66(Zr)–(OH)2 MIL-101(Cr) MIL-100(Fe) MIL-100(Cr) MIL-100(Al)

217 210 78 41 88 333 277 358

475 328 192 191 312 747 500 526

281 37 16 3 1522 296 354 405

Literature MOF-801-P [37] MOF-841 [37]

350c 540

411 558

21 44

a For the working capacity calculations, water uptakes at high RH conditions (P/P0 = 0.35, 0.45 and 0.55) were obtained from adsorption isotherms, but water uptakes at low RH conditions (P/P0 = 0.05 and 0.30) were obtained from the desorption isotherms. b For Case 2, all of the adsorbed water molecules were assumed to be desorbed at P/P0 = 0. c The working capacity of MOF-801-P for Case 1 was obtained from the water adsorption isotherm at RH conditions of P/P0 = 0.35 and 0.05 because the desorption isotherm of MOF-801-P was not provided below P/P0 = 0.1.

For Case 2 (P/P0 = 0.00–0.45), MIL-100(Fe) exhibited a significantly higher working capacity (747 cm3/g) than MOF-801-P and MOF-841. This is because the secondary capillary condensation was almost completed at P/P0 = 0.45. In contrast, MIL-100(Cr) and MIL-100(Al) showed modest working capacities (500–526 cm3/g) because the plateaus of the secondary capillary condensations were observed under the higher RH conditions (approximately P/ P0 = 0.55). UiO-66(Zr) showed a significant increase in working capacity when the RH condition was changed from Case 1 (P/ P0 = 0.05–0.35) to Case 2 (P/P0 = 0.00–0.45). This was because the isotherm increased steeply at P/P0 = 0.25–0.45, possibly due to the micropore filling. As a result, UiO-66(Zr) displayed a higher working capacity (475 cm3/g) than MOF-801-P. MIL-101(Cr) still showed a modest working capacity because a capillary condensation did not begin until P/P0 = 0.45. For Case 3 (P/P0 = 0.30–0.55), MIL-101(Cr) exhibited an extremely high working capacity (1522 cm3/g) because the isotherm increased sharply from 312 (at P/P0 = 0.45) to 1755 cm3/g (at P/ P0 = 0.55). The large water uptake at P/P0 = 0.55 correlated well with the high surface area (3250 m2/g) of MIL-101(Cr). The minor water uptake at low RH conditions may be explained by the hydrophobic nature of mesopores. Therefore, the high working capacity of MIL-101(Cr) was attributed to proper mesopore sizes with hydrophobic surfaces and high surface areas. The series of MIL-100(M) (M = Cr, Fe, and Al) and UiO-66(Zr) displayed modest working capacities (281–405 cm3/g), although they exhibited high water uptakes (530–790 cm3/g) at P/P0 = 0.55. This was because these MOFs still had considerable amounts of water molecules when the pressure was reduced to P/P0 = 0.30. However, MOF801-P and MOF-841 showed trivial working capacities because their isotherms were close to saturations even at P/P0 = 0.30. For all three cases, UiO-66(Zr)–NH2, UiO-66(Zr)–OH and UiO-66 (Zr)–(OH)2 showed only modest or minor working capacities. This resulted from the combined effect of the following two factors: (1) suppressed water uptakes at high RH conditions due to low surface areas of the functionalized MOFs and (2) increased water uptakes at low RH conditions due to hydrophilic functionalization. Consequently, we concluded that hydrophilic functionalization negatively affected the working capacity of UiO-66(Zr). In summary, in terms of working capacity, MOF-841 was the best adsorbent under the low humidity condition (Case 1) due to the sharp increase in the isotherm at P/P0 = 0.25. For the medium humidity condition (Case 2), MIL-100(Fe) showed the highest working capacity due to completion of two consecutive steps in the isotherm at P/P0 = 0.45. For the high humidity condition (Case

3), MIL-101(Cr) exhibited a significantly high working capacity due to the combined effect of the high surface area and hydrophobic nature of the mesopores. These results indicated that proper adsorbents should be chosen depending on the climate conditions. 3.4. Cyclic water adsorption and hydrothermal stability A general scheme for a solar-driven technology of fresh water production from the atmosphere has been proposed to be effective in extreme dry areas [4,5,28]. During the night time, water is adsorbed in an adsorbent when ambient air with a high relative humidity passes through a packed bed filled with the adsorbent. During the day time, water is desorbed from the adsorbent when ambient dry air (50–90 °C) heated by solar radiation passes through the packed bed. The desorbed water vapor is then condensed on the heat exchanger surface and collected in a water tank. For the actual applications of the adsorbents, cyclic water adsorption and hydrothermal stability need to be tested [26]. Because MIL-101(Cr) and the three MIL-100(M) materials exhibited superior working capacities, we investigated these MOFs for cyclic water adsorption and hydrothermal stability. Fig. 4 displays the five cycles of water adsorption/desorption isotherms for the four selected MOFs. Between each cycle, all of the samples were evacuated at room temperature for 2 h to demonstrate recyclability without thermal treatment. In an actual water production system, an adsorption bed may be regenerated by flowing a dry air at elevated temperature (50–90 °C) for several hours [4,28]. However, in this study, we regenerated the adsorbents under vacuum at ambient temperature for 2 h because this generally shows a similar degree of regeneration, compared with a purging by an inert gas at elevated temperature for several hours [37]. The working capacities for all five cycles were calculated based on the P/P0 range from Case 3 for MIL-101(Cr) and the P/P0 range from Case 2 for the three MIL-100(M) materials (Fig. 5). MIL-101(Cr) showed almost unchanged water adsorption/desorption isotherms throughout the five cycles (Fig. 4a). The obtained working capacities were almost the same in all five cycles (Fig. 5). This is remarkable because the regenerations were performed under very mild conditions. For MIL-100(Cr) and MIL-100(Al), the water uptakes decrease slightly in the 2nd cycle but remained almost constant in the following cycles (Fig. 4b and c). The measured working capacities showed similar trends (Fig. 5). However, the water adsorption isotherm and working capacity of MIL-100(Fe) decreased

472

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475

Fig. 4. Cyclic water adsorption/desorption isotherms for five cycles: (a) MIL-101(Cr), (b) MIL-100(Cr), (c) MIL-100(Al), and (d) MIL-100(Fe). Black curves represent adsorption isotherms, while gray ones indicate desorption isotherms. All samples were activated initially at 250 °C for 6 h under vacuum. Between each cycle, the samples were regenerated at very mild conditions (at ambient temperature for 2 h).

Fig. 5. Cyclic working capacities calculated from the water adsorption/desorption isotherms in Fig. 4. The working capacities of MIL-101(Cr) were calculated based on the RH range of Case 3 while the working capacities of MIL-100(M) were based on the RH range of Case 2.

‘significantly’ in the 2nd cycle (Fig. 4d). Nevertheless, in the following cycles, the water adsorption isotherms and working capacities were almost unchanged. Next, we focused on the observed decreases in water adsorption isotherms during the 2nd cycle. For MIL-100(Cr) and MIL-100(Al), similar degrees of decreases were observed throughout the entire RH region (Fig. 4b and c). Because the water uptake at a certain

RH is the accumulation of water uptakes before that RH, the decreases in the water isotherms can be ascribed to decreases in the water uptakes in the low RH regions. This may be because the water molecules that strongly adsorbed on the coordinatively unsaturated sites (CUS) were not fully desorbed by evacuation at ambient temperature for 2 h. Compared with MIL-100(Cr) and MIL-100(Al), MIL-100(Fe) showed a significant decrease in its water adsorption isotherm during the 2nd cycle. Similarly to the other two MOFs, MIL-100(Fe) showed a decrease in water uptake in the low RH region, which may have resulted from strong water adsorption on the CUS. However, for MIL-100(Fe), the significant decrease in the water adsorption isotherm was observed at the second condensation step (P/P0 > 0.4) that may occur on larger mesopores (29 Å). We can speculate that these larger mesopores were partially collapsed during the first water adsorption/desorption cycle after being activated at 250 °C. This may be connected to a partial reduction of MIL-100(Fe) during activation at 250 °C due to the loss of anionic ligands (F and OH ). From a rigorous in situ spectroscopic studies, Chang and co-workers found that approximately one out of the three FeIII CUSs was reduced to the FeII state after being activated at 250 °C [43]. This partial reduction has not been reported for MIL-100(Cr) or MIL-100(Al). To investigate the hydrothermal stabilities of the four selected MOFs, we compared the PXRD patterns between the assynthesized samples and the used samples after 5 water adsorption/desorption cycle measurements. As displayed in Fig. 6, no noticeable changes were observed in the peak positions or the intensities for MIL-101(Cr), MIL-100(Cr) and MIL-100(Al). This showed that these three MOFs were hydrothermally stable. However, for MIL-100(Fe), the peak intensities clearly decreased after

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475

473

Fig. 6. Comparisons of PXRD patterns between the as-synthesized samples (red) and the samples used after cyclic water adsorption/desorption measurements (blue): (a) MIL-101(Cr), (b) MIL-100(Fe), (c) MIL-100(Cr), and (d) MIL-100(Al). All samples were activated initially at 250 °C for 6 h under vacuum. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7. (a) Cyclic water adsorption/desorption isotherms of MIL-100(Fe)_150. The MIL-100(Fe) sample was activated initially at 150 °C for 6 h under vacuum. After the 1st and 2nd cycles, the samples were regenerated at ambient temperature for 2 h. After the 3rd cycle, the sample was regenerated at 150 °C for 2 h under vacuum. (b) Comparisons of the PXRD patterns of MIL-100(Fe)_150: as-synthesized samples (red) and used samples after cyclic water adsorption/desorption measurements (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

repeated water adsorption, although the peak positions were unchanged. This supported our hypothesis that larger mesopores in MIL-100(Fe) were partially collapsed during the first water adsorption/desorption cycle after being activated at 250 °C.

To strengthen this hypothesis, we also activated MIL-100(Fe) at 150 °C for 6 h under vacuum (MIL-100(Fe)_150) to prevent formation of FeII CUS. Chang and co-workers confirmed that the FeII CUS content is negligible after degassing at 150 °C [43]. As displayed in

474

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475

Fig. 7a, the water uptakes in the 1st cycle were slightly lower than those for the sample activated at 250 °C. This was an expected result because FeIII CUS would adsorb less water molecules than FeII CUS. After the 1st and 2nd cycles, the sample was evacuated at room temperature for 2 h. Remarkably, for MIL-100(Fe)_150, the water adsorption isotherms were almost unchanged in the 2nd and 3rd cycles, although trivial decreases in water uptakes were observed in the 2nd cycle. The trivial decreases in the water uptakes can be explained by strong water adsorption on FeIII CUS, which was not fully regenerated by evacuation at ambient temperature. Remarkably, even this minor decrease in the water uptake was fully recovered after regenerating the sample at 150 °C for 2 h (4th cycle). When we measured the PXRD pattern after the 4th cycle of water adsorption/desorption measurements, there were no noticeable changes in the peak positions or intensities (Fig. 7b). These results confirmed that MIL-100(Fe)_150 was hydrothermally stable unlike the sample activated at 250 °C. This can be explained by the absence of FeII CUS in MIL-100(Fe)_150. From these results, we conclude that the significant decrease in water adsorption isotherm of MIL-100(Fe) during the 2nd cycle was attributed to the presence of FeII CUS, which was formed during activation at 250 °C. These results agree with a previous finding that MIII-containing MOFs are more resistant to water reactions than MII-containing MOFs [56].

4. Conclusions In this work, we investigated the potential use of eight hydrothermally stable MOFs for water adsorption applications that focused on fresh water production in extremely dry regions. For the evaluations, we obtained the water adsorption isotherms and working capacities for typical humidity conditions in three representative dry regions. For the low humidity condition (Case 1), none of the eight MOFs showed high working capacities. However, for the medium and high humidity conditions (Cases 2 and 3), three MIL-100(M) materials (M = Cr, Al, and Fe) and MIL-101(Cr) displayed very high working capacities due to their large surface areas. For these four MOFs with high working capacities, we measured the cyclic water adsorption/desorption isotherms and PXRD patterns of the used samples after repeated water adsorption and desorption cycles. Interestingly, we found that all of the MOFs consisting of only MIII CUS (MIL-101(Cr), MIL-100(Cr), and MIL100(Al)) exhibited good cyclic water adsorption performances and good hydrothermal stabilities. However, MIL-100(Fe) showed a considerable decrease in its water adsorption isotherm during the 2nd cycle and poor hydrothermal stability, which was attributed to the presence of FeII CUS that was formed during activation at 250 °C. When we activated MIL-100(Fe) at 150 °C (MIL-100(Fe) _150) to prevent formation of FeII CUS, the sample showed good cyclic adsorption/desorption performance and good hydrothermal stability. Remarkably, all of the MOFs consisting of only MIII CUS exhibited reasonably good cyclic adsorption/desorption properties even if the samples were regenerated at room temperature for 2 h between each cycle. Considering the high working capacities, cyclic adsorption/desorption behaviors, and good hydrothermal stabilities, MIL-101(Cr), MIL-100(Cr), MIL-100(Al) and MIL-100(Fe)_150 are promising adsorbents for producing drinking water in dry regions with medium or high humidity conditions during the night. Among these MOFs, MIL-101(Cr) and MIL-100(Fe) are currently produced on large scales at reasonable prices.

Acknowledgments This work was supported by the Technology Innovation Program (10048649) funded by the Ministry of Trade, Industry & Energy (MI, Korea). This research was also supported by the Pioneer Research Center Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning (NRF-2010-0019531).

References [1] D. La, Y.J. Dai, Y. Li, R.Z. Wang, T.S. Ge, Technical development of rotary desiccant dehumidification and air conditioning: a review, Renewable Sustainable Energy Rev. 14 (2010) 130–147. [2] S. Nakabayashi, K. Nagano, M. Nakamura, J. Togawa, A. Kurokawa, Improvement of water vapor adsorption ability of natural mesoporous material by impregnating with chloride salts for development of a new desiccant filter, Adsorpt. J. Int. Adsorpt. Soc. 17 (2011) 675–686. [3] C.Y. Tso, C.Y.H. Chao, Activated carbon, silica-gel and calcium chloride composite adsorbents for energy efficient solar adsorption cooling and dehumidification systems, Int. J. Refrig. Rev. Int. Froid 35 (2012) 1626–1638. [4] Y.I. Aristov, M.M. Tokarev, L.G. Gordeeva, V.N. Snytnikov, V.N. Parmon, New composite sorbents for solar-driven technology of fresh water production from the atmosphere, Solar Energy 66 (1999) 165–168. [5] J.G. Ji, R.Z. Wang, L.X. Li, New composite adsorbent for solar-driven fresh water production from the atmosphere, Desalination 212 (2007) 176–182. [6] H. Yang, H. Zhu, M.M.R.M. Hendrix, N.J.H.G.M. Lousberg, G. de With, A.C.C. Esteves, J.H. Xin, Temperature-triggered collection and release of water from fogs by a sponge-like cotton fabric, Adv. Mater. 25 (2013) 1150–1154. [7] J. Ehrenmann, S.K. Henninger, C. Janiak, Water adsorption characteristics of MIL-101 for heat-transformation applications of MOFs, Eur. J. Inorg. Chem. 471–474 (2011). [8] S.K. Henninger, F. Jeremias, H. Kummer, C. Janiak, MOFs for use in adsorption heat pump processes, Eur. J. Inorg. Chem. 2625–2634 (2012). [9] F. Jeremias, A. Khutia, S.K. Henninger, C. Janiak, MIL-100(Al, Fe) as water adsorbents for heat transformation purposes-a promising application, J. Mater. Chem. 22 (2012) 10148–10151. [10] C. Janiak, S.K. Henninger, Porous coordination polymers as novel sorption materials for heat transformation processes, Chimia 67 (2013) 419–424. [11] J.L.C. Rowsell, O.M. Yaghi, Metal–organic frameworks: a new class of porous materials, Microporous Mesoporous Mater. 73 (2004) 3–14. [12] K.M. Thomas, Adsorption and desorption of hydrogen on metal–organic framework materials for storage applications: comparison with other nanoporous materials, Dalton Trans. 1487 (2009). [13] G. Ferey, Hybrid porous solids: past, present, future, Chem. Soc. Rev. 37 (2008) 191–214. [14] R.E. Morris, P.S. Wheatley, Gas storage in nanoporous materials, Angew. Chem. Int. Ed. 47 (2008) 4966–4981. [15] J.L.C. Rowsell, O.M. Yaghi, Strategies for hydrogen storage in metal–organic frameworks, Angew. Chem. Int. Ed. 44 (2005) 4670–4679. [16] L.J. Murray, M. Dinca, J.R. Long, Hydrogen storage in metal–organic frameworks, Chem. Soc. Rev. 38 (2009) 1294–1314. [17] D.J. Collins, H.-C. Zhou, Hydrogen storage in metal–organic frameworks, J. Mater. Chem. 17 (2007) 3154–3160. [18] H. Furukawa, N. Ko, Y.B. Go, N. Aratani, S.B. Choi, E. Choi, A.O. Yazaydin, R.Q. Snurr, M. O’Keeffe, J. Kim, O.M. Yaghi, Ultrahigh porosity in metal–organic frameworks, Science 329 (2010) 424–428. [19] U. Mueller, M. Schubert, F. Teich, H. Puetter, K. Schierle-Arndt, J. Pastre, Metal– organic frameworks – prospective industrial applications, J. Mater. Chem. 16 (2006) 626–636. [20] R.J. Kuppler, D.J. Timmons, Q.R. Fang, J.R. Li, T.A. Makal, M.D. Young, D.Q. Yuan, D. Zhao, W.J. Zhuang, H.C. Zhou, Potential applications of metal–organic frameworks, Coord. Chem. Rev. 253 (2009) 3042–3066. [21] J.R. Li, R.J. Kuppler, H.C. Zhou, Selective gas adsorption and separation in metal–organic frameworks, Chem. Soc. Rev. 38 (2009) 1477–1504. [22] S.T. Meek, J.A. Greathouse, M.D. Allendorf, Metal–organic frameworks: a rapidly growing class of versatile nanoporous materials, Adv. Mater. 23 (2011) 249–267. [23] L.E. Kreno, K. Leong, O.K. Farha, M. Allendorf, R.P. Van Duyne, J.T. Hupp, Metal– organic framework materials as chemical sensors, Chem. Rev. 112 (2012) 1105–1125. [24] J. Lee, O.K. Farha, J. Roberts, K.A. Scheidt, S.T. Nguyen, J.T. Hupp, Metal–organic framework materials as catalysts, Chem. Soc. Rev. 38 (2009) 1450–1459. [25] H.-C. Zhou, J.R. Long, O.M. Yaghi, Introduction to metal–organic frameworks, Chem. Rev. 112 (2012) 673–674. [26] N.C. Burtch, H. Jasuja, K.S. Walton, Water stability and adsorption in metal– organic frameworks, Chem. Rev. 114 (2014) 10575–10612. [27] G. Ferey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surble, I. Margiolaki, A chromium terephthalate-based solid with unusually large pore volumes and surface area, Science 309 (2005) 2040–2042.

S.-I. Kim et al. / Chemical Engineering Journal 286 (2016) 467–475 [28] Y.-K. Seo, J.W. Yoon, J.S. Lee, Y.K. Hwang, C.-H. Jun, J.-S. Chang, S. Wuttke, P. Bazin, A. Vimont, M. Daturi, S. Bourrelly, P.L. Llewellyn, P. Horcajada, C. Serre, G. Ferey, Energy-efficient dehumidification over hierachically porous metal– organic frameworks as advanced water adsorbents, Adv. Mater. 24 (2012) 806–810. [29] E. Soubeyrand-Lenoir, C. Vagner, J.W. Yoon, P. Bazin, F. Ragon, Y.K. Hwang, C. Serre, J.-S. Chang, P.L. Llewellyn, How water fosters a remarkable 5-fold increase in low-pressure CO2 uptake within mesoporous MIL-100(Fe), J. Am. Chem. Soc. 134 (2012) 10174–10181. [30] P.S. Barcia, D. Guimaraes, P.A.P. Mendes, J.A.C. Silva, V. Guillerm, H. Chevreau, C. Serre, A.E. Rodrigues, Reverse shape selectivity in the adsorption of hexane and xylene isomers in MOF UiO-66, Microporous Mesoporous Mater. 139 (2011) 67–73. [31] J.H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga, K.P. Lillerud, A new zirconium inorganic building brick forming metal organic frameworks with exceptional stability, J. Am. Chem. Soc. 130 (2008) 13850– 13851. [32] M.J. Katz, Z.J. Brown, Y.J. Colon, P.W. Siu, K.A. Scheidt, R.Q. Snurr, J.T. Hupp, O.K. Farha, A facile synthesis of UiO-66, UiO-67 and their derivatives, Chem. Commun. 49 (2013) 9449–9451. [33] P. Guo, A.G. Wong-Foy, A.J. Matzger, Microporous coordination polymers as efficient sorbents for air dehumidification, Langmuir 30 (2014) 1921–1925. [34] R. Plessius, R. Kromhout, A.L.D. Ramos, M. Ferbinteanu, M.C. MittelmeijerHazeleger, R. Krishna, G. Rothenberg, S. Tanase, Highly selective water adsorption in a lanthanum metal–organic framework, Chem. Eur. J. 20 (2014) 7922–7925. [35] L. Xie, D. Liu, H. Huang, Q. Yang, C. Zhong, Efficient capture of nitrobenzene from waste water using metal–organic frameworks, Chem. Eng. J. 246 (2014) 142–149. [36] N.A. Khan, B.K. Jung, Z. Hasan, S.H. Jhung, Adsorption and removal of phthalic acid and diethyl phthalate from water with zeolitic imidazolate and metal– organic frameworks, J. Hazard. Mater. 282 (2015) 194–200. [37] H. Furukawa, F. Gandara, Y.-B. Zhang, J. Jiang, W.L. Queen, M.R. Hudson, O.M. Yaghi, Water adsorption in porous metal–organic frameworks and related materials, J. Am. Chem. Soc. 136 (2014) 4369–4381. [38] Y.-S. Bae, R.Q. Snurr, Optimal isosteric heat of adsorption for hydrogen storage and delivery using metal–organic frameworks, Microporous Mesoporous Mater. 132 (2010) 300–303. [39] J.E. Valdivia-Silva, R. Navarro-Gonzalez, L. Fletcher, S. Perez-Montano, R. Condori-Apaza, F. Ortega-Gutierrez, C. McKay, Climatological characteristics in the extreme hyper-arid region of Pampas de La Joya, Peru. Astrobiological approach in four years of observation: 2004–2008, Int. J. Astrobiol. 11 (2012) 25–35. [40] A. Libal, What is the humidity of the Mojave desert?, , 2015. [41] G.E. Cmarik, M. Kim, S.M. Cohen, K.S. Walton, Tuning the adsorption properties of UiO-66 via ligand functionalization, Langmuir 28 (2012) 15606–15613. [42] Y.K. Hwang, D.-Y. Hong, J.-S. Chang, S.H. Jhung, Y.-K. Seo, J. Kim, A. Vimont, M. Daturi, C. Serre, G. Ferey, Amine grafting on coordinatively unsaturated metal

[43]

[44]

[45] [46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

475

centers of MOFs: consequences for catalysis and metal encapsulation, Angew. Chem. Int. Ed. 47 (2008) 4144–4148. J.W. Yoon, Y.-K. Seo, Y.K. Hwang, J.-S. Chang, H. Leclerc, S. Wuttke, P. Bazin, A. Vimont, M. Daturi, E. Bloch, P.L. Llewellyn, C. Serre, P. Horcajada, J.-M. Greneche, A.E. Rodrigues, G. Ferey, Controlled reducibility of a metal–organic framework with coordinatively unsaturated sites for preferential gas sorption, Angew. Chem. Int. Ed. 49 (2010) 5949–5952. G. Ferey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surble, J. Dutour, I. Margiolaki, A hybrid solid with giant pores prepared by a combination of targeted chemistry, simulation, and powder diffraction, Angew. Chem. Int. Ed. 43 (2004) 6296–6301. J. Rouquerol, P. Llewellyn, F. Rouquerol, Is the BET equation applicable to microporous adsorbents?, Studies Surf Sci. Catal. 160 (2007) 49–56. K.S. Walton, R.Q. Snurr, Applicability of the BET method for determining surface areas of microporous metal–organic frameworks, J. Am. Chem. Soc. 129 (2007) 8552–8556. Y.-S. Bae, A.O. Yazaydin, R.Q. Snurr, Evaluation of the BET method for determining surface areas of MOFs and zeolites that contain ultramicropores, Langmuir 26 (2010) 5475–5483. A. Schaate, P. Roy, A. Godt, J. Lippke, F. Waltz, M. Wiebcke, P. Behrens, Modulated synthesis of Zr-based metal–organic frameworks: from nano to single crystals, Chem. Eur. J. 17 (2011) 6643–6651. H. Jasuja, G.W. Peterson, J.B. Decoste, M.A. Browe, K.S. Walton, Evaluation of MOFs for air purification and air quality control applications: ammonia removal from air, Chem. Eng. Sci. 124 (2015) 118–124. G. Akiyama, R. Matsuda, H. Sato, A. Hori, M. Takata, S. Kitagawa, Effect of functional groups in MIL-101 on water sorption behavior, Microporous Mesoporous Mater. 157 (2012) 89–93. P.L. Llewellyn, S. Bourrelly, C. Serre, A. Vimont, M. Daturi, L. Hamon, G. De Weireld, J.S. Chang, D.Y. Hong, Y.K. Hwang, S.H. Jhung, G. Ferey, High uptakes of CO2 and CH4 in mesoporous metal–organic frameworks MIL-100 and MIL101, Langmuir 24 (2008) 7245–7250. A.D. Wiersum, E. Soubeyrand-Lenoir, Q.Y. Yang, B. Moulin, V. Guillerm, M. Ben Yahia, S. Bourrelly, A. Vimont, S. Miller, C. Vagner, M. Daturi, G. Clet, C. Serre, G. Maurin, P.L. Llewellyn, An evaluation of UiO-66 for gas-based applications, Chem. Asian J. 6 (2011) 3270–3280. P.M. Schoenecker, C.G. Carson, H. Jasuja, C.J.J. Flemming, K.S. Walton, Effect of water adsorption on retention of structure and surface area of metal–organic frameworks, Ind. Eng. Chem. Res. 51 (2012) 6513–6519. P. Kusgens, M. Rose, I. Senkovska, H. Frode, A. Henschel, S. Siegle, S. Kaskel, Characterization of metal–organic frameworks by water adsorption, Microporous Mesoporous Mater. 120 (2009) 325–330. D. Cunha, M. Ben Yahia, S. Hall, S.R. Miller, H. Chevreau, E. Elkaim, G. Maurin, P. Horcajada, C. Serre, Rationale of drug encapsulation and release from biocompatible porous metal–organic frameworks, Chem. Mater. 25 (2013) 2767–2776. J.J. Low, A.I. Benin, P. Jakubczak, J.F. Abrahamian, S.A. Faheem, R.R. Willis, Virtual high throughput screening confirmed experimentally: porous coordination polymer hydration, J. Am. Chem. Soc. 131 (2009) 15834–15842.