Hierarchically Porous Co-MOF-74 Hollow Nanorods for Enhanced

Subscriber access provided by UNIV OF LOUISIANA

Letter

Hierarchically Porous Co-MOF-74 Hollow Nanorods for Enhanced Dynamic CO2 Separation Xiahui Zhang, Chong Yang Chuah, Panpan Dong, Younghwan Cha, Tae-Hyun Bae, and Min-Kyu Song ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b17180 • Publication Date (Web): 27 Nov 2018 Downloaded from http://pubs.acs.org on November 29, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Hierarchically Porous Co-MOF-74 Hollow Nanorods for Enhanced Dynamic CO 2 Separation Xiahui Zhang,a† Chong Yang Chuah,b† Panpan Dong, a Young-Hwan Cha,a Tae-Hyun Bae,b* and Min-Kyu Song*a a School of Mechanical and Materials Engineering, Washington State University, Pullman, Washington 99164, United States b School of Chemical and Biomedical Engineering, Nanyang Technological University Singapore, 637459, Singapore † These authors contributed equally to this work.

*Corresponding Author Prof. T.-H. Bae Email: [email protected]

Prof. M. -K. Song E-mail: [email protected]

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Metal–organic frameworks (MOFs) with coordinatively unsaturated (open) metal sites have been intensively investigated in gas separations because their active sites can selectively interact with targeted molecules such as CO 2 . Although such MOFs have shown to exhibit exceptional CO 2 uptake capacity at equilibrium, the dynamic separation behavior is often not satisfactory to be considered in practical applications. Herein we report a facile and efficient self-sacrifice template strategy based on the nanoscale Kirkendall effect to form novel Co-MOF-74 hollow nanorods enabling adsorption/desorption of gas molecules in a facilitated manner. The time-dependent microscopic and diffraction examinations were performed to elucidate the formation mechanism of Co-MOF-74 hollow nanorods and to obtain insights into the factors critical to maintaining the rod-like morphology. Such nanostructured MOF exhibited much sharper CO 2 molecular separation behavior than conventional MOF bulk crystals under a dynamic flow condition, owing to its enhanced adsorption kinetics through the shortened diffusion distance. Such enhanced dynamic molecular separation behavior was further confirmed by chromatographic separations where a significant peak narrowing was demonstrated.

KEYWORDS: Metal-organic frameworks, template method, Nanostructures, Kirkendall effect, Gas separation


Page 2 of 18

Page 3 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Concerns on global warming and climate change have triggered extensive research efforts on the capture, storage, and utilization of CO 2 emitted from anthropogenic sources.1,2 Effective systems for CO 2 capture should demonstrate a large adsorption capacity under dynamic condition, a decent selectivity for CO 2 to secure a high purity of the product as well as a low energy penalty for regeneration.3 Adsorptive gas separation using porous materials such as zeolites, porous carbons and metal-organic frameworks (MOFs) has been considered as a promising process for this application as it has a potential to meet the requirements aforementioned. MOFs are a class of microporous crystalline materials built with metal nodes and organic ligands that are held together via coordination bonding. In comparison to other porous materials, MOFs hold a great promise for a wide range of applications, such as gas storage and separation owing to their high accessible surface area and tunable functionality.4-8 Indeed, some MOFs possessing functional groups and/or open metal sites that can interact with CO 2 have shown an outstanding CO 2 uptake capacity along with a good selectivity at equilibrium condition which is ideally controlled. For example, isoreticular M-MOF-74 materials (M = Mn, Co, Mg, Ni, also known as M-CPO-27) with high density of open metal sites (ca. 3.3 sites per 1 nm2) and onedimensional large pore channels have been reported to exhibit high adsorption capacities for guest molecules and good selectivity for gas separation.3,6,9 It is worth noting that among various isoreticular M-MOF-74, Co-MOF-74 generally exhibits higher retention of CO 2 uptake capacity and better regenerability than those of Mg-MOF-74 in the presence of moisture, although MgMOF-74 has been reported to possess the largest CO 2 adsorption capacity at low pressure under dry condition.6,9 This suggests that Co-MOF-74 may have better applicability in practical operations than Mg-MOF-74. However, such outstanding performance cannot guarantee the success in the realistic operation as the gas separation is always carried out under a dynamic



condition where the diffusion of adsorbate within the adsorbent also plays a critical role. Hence, the key to their successful application in practical gas capture technologies is to enhance the dynamic capacity by engineering the structure of MOFs including morphology and nanoarchitecture. Recently, Dinca’s group reported that isoreticular MOFs with large pore size (23 Å) exhibited faster ammonia sorption kinetics than that with small pore size (13 Å).10 In addition to isoreticular MOFs with the design of pore size, the hollow-structured nanomaterials is another desirable platform to facilitate adsorption-desorption cycling by shortening the diffusion distance.11-13 There are a few publications in which hollow MOFs were synthesized either by hard or soft template methods14,15 or by post-synthetic modification.16,17 These methods require the removal of templates or additional treatments, resulting in the increased complexity of processing steps and amounts of chemicals used. Recently, the synthesis of hollow ZIF-67 nanorods by using self-sacrificing template has been reported.12 However, it remains difficult to develop a facile method to fabricate hierarchically porous, hollow nanocrystalline MOFs with open metal sites. Herein, we report a self-sacrifice template strategy based on the nanoscale Kirkendall effect to fabricate novel Co-MOF-74 hollow nanorods with granular shell, which enables a rapid gas adsorption-desorption under a dynamic condition (Fig. 1). When tested under a dynamic flow condition, such nanostructured MOF-74 exhibited much sharper CO 2 molecular separation behavior than conventional MOF bulk crystals owing to its enhanced adsorption kinetics through the shortened diffusion distance. Such enhanced dynamic molecular separation behavior was further confirmed by chromatographic separations where a significant peak narrowing was achieved. Fig. 1 shows the schematic diagram of the formation of Co-MOF-74 hollow nanorod and its architecture. The Co precursor nanorods of Co 3 (OH)(CH 3 COO) 5 were first synthesized with a


Page 4 of 18

Page 5 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


polyvinylpyrrolidone (PVP)-assisted hydrolysis method in ethanol via a modified procedure described elsewhere.12 The as-prepared Co precursor nanorods then were transformed to Co-MOF74 hollow nanorods in dimethylformamide (DMF) solution of 2,5-dihydroxyterephthalic acid (H 4 DOBDC) (see Experimental for details). The reaction of chemical conversion is described by the following Equation (1). 2Co 3 (OH)(CH 3 COO) 5 + 3H 4 DOBDC → 3Co 2 (DOBDC) + 10CH 3 COOH + 2H 2 O

Conversion

: CO2

: Co-MOF-74 nanorods

(1) Fig. 1 Schematic of formation and unique architecture of Co-MOF-74 hollow nanorods.

The Co precursors serve as not only the self-sacrificing templates but also the deprotonation agents. The mechanism behind the formation of hollow nanorods can be explained by the nanoscale Kirkendall effect as a consequence of the difference in diffusion rates of two ion species, which has been widely employed to create hollow structure during the chemical transformation.12,18,19 From the PXRD pattern, the phase of Co precursors has been identified as cobalt hydroxide acetate of Co 3 (OH)(CH 3 COO) 5 (Fig. 2a). Fig. 2b and 2d show the field-emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM) images of Co precursor nanorods, respectively, indicating a solid core with a diameter of ca. 100 nm and a length 5 ACS Paragon Plus Environment


Page 6 of 18

of ca. 500 nm. After 2-hour chemical transformation, Co-MOF-74 hollow nanorods were obtained without noticeable peaks of Co precursors in PXRD patterns, revealing the full conversion (Fig. 2a). The average crystallite size of Co-MOF-74 hollow nanorods, calculated from PXRD result using Scherrer’s equation, is ca. 13 nm.20,21 The corresponding FESEM images (Fig. 2c) showed that the overall rod-like shape is well maintained after transformation and the surface is constructed by numerous interconnected rod-like subunits (nanorods with a diameter of ca. 5 nm and a length of ca. 20 nm), which is in good agreement with the crystallite size estimated from PXRD pattern. TEM images (Fig. 2e) clearly revealed the hollow nature of the Co-MOF-74 nanorods by the sharp contrast between the granular shell (ca. 40 nm) and the central void space. The interparticle pores formed by rod-like subunits can provide fast diffusion pathway for incoming guest molecules in and out of MOF crystals. In addition, the thin shells could further significantly reduce the gas diffusion distance to facilitate adsorption/desorption of gas molecules, compared to the bulk (a)

(b)

(c)

500 nm

200 nm

(d)

(e)

200 nm

100 nm

crystals.


Page 7 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 2 (a) PXRD patterns of Co precursor nanorods and Co-MOF-74 hollow nanorods. SEM images of (b) Co precursor nanorods and (c) Co-MOF-74 hollow nanorods. TEM images of (d) Co precursor nanorods and (e) Co-MOF-74 hollow nanorods.

To gain more insights into the formation process of hollow structure based on the Kirkendall effect, we carried out time-dependent experiments and the results are shown in Fig. 3. Since the deprotonated ligand anion have the bigger ionic size than Co2+ cation, it is expected that the diffusion rate of Co2+ cations is faster than ligand anions. It’s worth noting that the sequence of mixing linker solution and precursor suspension is the key to maintaining the morphology. After a reaction time of 2 min, a solid and robust Co-MOF-74 shell was in-situ formed on the surface of Co precursors to establish a core-shell structure (Fig. 3b) when adding Co precursors into linker solution. Such a robust shell is functioned as not only a template to maintain the rod-like morphology but also a self-limited barrier against the inward diffusion rate of ligand anions. When adding linker solution into Co precursors, however, no robust shell was formed, resulting in the severe dissolution of Co precursors in the linker solution and then the recrystallization of Co-MOF74, and therefore Co-MOF-74 nanoparticles were obtained instead as shown in Fig. S1. After 5 min of transformation reaction, some voids appeared in the center of Co precursors, clearly revealing that the outward diffusion of smaller ions (i.e., Co2+ cations) from the Co solid precursors dominates the growth of Co-MOF-74 shell. To determine where Co-MOF-74 subunits are, the asprepared Co-MOF-74/Co precursor nanorods after 5-min reaction were washed by methanol to remove Co precursors as shown in Fig. S2. Comparing Fig. 3b with Fig. S2, it is apparent that the shell is constructed by Co-MOF-74. In this regard, it is noted that the thickness of Co-MOF-74 shell can be easily tailored by adjusting the conversion time and subsequently etching away the unconverted Co precursor in the core using methanol. After 30-min conversion, TEM image of the product (Fig. 3d) shows that the size of voids is enlarged whilst the unconverted Co precursors 7 ACS Paragon Plus Environment


Page 8 of 18

still exists in the core, leading to a yolk-shell structure. Corresponding PXRD pattern (Fig. 3g) further confirms the existence of Co precursors. After 2 hours, the conversion reaction was completed as shown in Fig. 3f and Fig. 2e, which is consistent with the PXRD results (Fig. 2a). It is worth noting that the diameter of Co-MOF-74 nanorods is ~150% of that of Co precursor nanorods as revealed in Fig. 3a and 3f, most likely due to the decreased density of highly porous MOF-74 (i.e., increased volume). Based on the above time-dependent analysis, the formation mechanism of such hollow nanorods is schematically summarized in Fig. 3h. (a)

0 min (b)

200 nm

200 nm

30 min (e)

(d)

200 nm

200 nm

2 min

(c)

(g)

5 min

200 nm

60 min (f)

120 min

200 nm 60 min 74

e

Co-MOF-74 nanorods

Porous Hollow Structure

Fig. 3 TEM images showing the evolution of Co precursor nanorods to Co-MOF-74 hollow nanorods at (a–f) 0 min, 2 min, 5 min, 30 min, 60 min, and 120 min, respectively. (g) PXRD pattern of the product after 30-min conversion. (h) Schematic of formation process of Co-MOF-74 hollow nanorods.

Co-MOF-74 bulk rods were prepared as a control sample with a conventional solvothermal reaction via a modified procedure described elsewhere.8 The corresponding SEM image shows that the Co-MOF-74 bulk rods have a diameter of ca. 10 μm and a length of ca. 40 μm (Fig. S3).


Page 9 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The thermal stability of Co-MOF-74 hollow nanorods and bulk rods at 180 °C was confirmed by PXRD patterns as shown in Fig. S4. Thermogravimetric analysis (TGA) was also performed to study the thermal stability of as-prepared two MOFs as shown in Fig. S5. From Fig. S5, Co-MOF74 bulk rods start to decompose at ca. 300 °C, exhibiting better thermal stability than Co-MOF-74 hollow nanorods (ca. 250 °C). It is reasonable that bulk materials have better thermal stability than nanomaterials due to low surface energy and defects. It is worth noting that the regeneration temperature for CO 2 capture is lower than 180 °C, indicating both Co-MOF-74 crystals can be regenerated without any decomposition issue. We carried out N 2 physisorption analysis at 77 K to investigate pore characteristics of the synthesized Co-MOF-74. As shown in Fig. 4a, both samples exhibit a high N 2 uptake at the lowpressure region, indicating the presence of large amount of micropores. However, Co-MOF-74 hollow nanorods show a hysteresis loop between adsorption and desorption branches, revealing the presence of mesopores contributed by the interparticle pores of granular shell (Fig. 2c and 2e). The corresponding pore size distribution curve of Co-MOF-74 hollow nanorods reveals the sizes of these pores to be ranged in 1–10 nm (Fig. 4b), indicating the co-existence of hierarchical micropores and mesopores. On the contrary, only well-defined micropores with maxima at ~1.1 nm were observed in Co-MOF-74 bulk crystals, which is consistent with literature.22 As shown in Table S1, both Co-MOF-74 bulk rods and hollow nanorods possess a high Brunauer–Emmett– Teller (BET) surface area of 1049 m2 g−1 and 804 m2/g, respectively, further confirming the high crystallinity of as-prepared Co-MOF-74 as revealed by PXRD in Fig. 2a. We infer that Co-MOF74 hollow nanorods prepared at lower temperature within shorter reaction time may have slightly lower crystallinity than Co-MOF-74 bulk rods solvothermally synthesized, giving rise to slightly lower surface area than bulk rods.



Subsequently, CO 2 and N 2 adsorption properties of Co-MOF-74 crystals at the static equilibrium condition were investigated. The pure component isotherms which were measured at 25 °C are displayed in Fig. 4c. In general, both Co-MOF-74 samples demonstrated strong affinity for CO 2 over N 2 , which is attributed to the presence of coordinatively unsaturated open metal sites.23 It is well known that the open metal sites allows a reversible interaction between the framework and CO 2 which has a higher polarizability (26.3 × 10−25 cm3 vs. 17.6 × 10−25 cm3) and a quadrupole moment (13.4 × 10−40 C m-2 vs. 4.7 × 10−40 C m−2)24 than N 2 . Nonetheless, the CoMOF-74 hollow nanorods exhibited a decreased CO 2 uptake capacity at equilibrium as compared to that of the bulk crystals (ca. 4.10 mmol g−1 vs ca. 6.50 mmol g−1 at 25 °C and 1 bar) presumably due to the slightly decreased specific surface area (Table S1).25 However, in practical operations, gas separations are conducted under a dynamic flow condition in which the diffusions of gases within the adsorbents can limit the overall performance.25,26 Thus, to study the behavior of adsorbents under a dynamic flow condition, breakthrough measurements were conducted using a binary mixture of CO 2 /N 2 at 25 °C and the results are depicted in Fig. 4d. In contrast to the equilibrium measurement, the CO 2 breakthrough occurs slightly earlier for the bulk Co-MOF-74 as compared to the hollow nanorods. More importantly, the shape of breakthrough curve has become significantly sharper by changing the morphology of MOF-74 crystal from the bulk rods to the hollow nanorods. These results imply that the dynamic separation of gas molecules is no more limited by the slow mass transfer within the adsorbent for the case of Co-MOF-74 hollow nanorods. Such improved performance of Co-MOF-74 hollow nanorods under a dynamic condition is attributed to the shortened diffusion distance for gas molecules within the adsorbent by reducing the particle size and removing the core area that requires the elongated diffusion time to be saturated with adsorbate. It should be noted that the diffusion path can also be shortened by


Page 10 of 18

Page 11 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


downsizing the crystals. However, the downsized crystals reduce the sizes of interstitial spaces in the packed column, resulting in an increased pressure drop during the operation at a flow condition. Such high pressure drop often makes the operation difficult or impossible in practical applications. To verify the stability and the regenerability of Co-MOF-74 hollow nanorods, the adsorptiondesorption cycling was repeated 10 times under a dynamic flow condition. As shown in Fig. S8, Co-MOF-74 nanorods demonstrated a stable performance during this cycling measurement implying that our MOF crystals which can be readily regenerated under a dynamic flow condition have a good stability.

(c)

(e)

(d)

bulk rods

(f)

hollow nanorods



Fig. 4 (a) N 2 adsorption and desorption isotherms of Co-MOF-74 bulk rods and hollow nanorods. (b) Corresponding pore distributions calculated using BJH method (and DFT method: inset). It should be noted that only micropores smaller than 2 nm were observed for the bulk Co-MOF 74 crystals. In contrast, both mesopores (>2 nm) and micropores (

Hierarchically Porous Co-MOF-74 Hollow Nanorods for Enhanced

Recommend Documents