Article

Lattice Expansion and Contraction in Metal-Organic Frameworks by Sequential LinkerReinstallation

Liang Feng, Shuai Yuan,

Jun-Sheng Qin, ..., Lin Cheng,

Sherzod T. Madrahimov,

Hong-Cai Zhou

shuai.yuan@chem.tamu.edu (S.Y.)

zhou@chem.tamu.edu (H.-C.Z.)

HIGHLIGHTS

Post-synthetic linker replacement

within an inert MOF is achieved

MOF expansion and contraction

are accomplished via labilization

and reinstallation

Control over MOF flexibility via

defect creation enables precise

pore engineering

Dynamic covalent and

coordination chemistry are

combined to engineer MOF pores

Linker reinstallation has been successfully introduced into inert frameworks,

allowing post-synthetic linker replacement, framework expansion, and

contraction. Benefiting from the framework flexibility via defect creation,

labilization of the initial linkers enables sequential installation of longer or shorter

linkers within a robust MOF.

Feng et al., Matter 1, 1–12

July 10, 2019 ª 2019 Elsevier Inc.

https://doi.org/10.1016/j.matt.2019.02.002

Article

Lattice Expansion and Contractionin Metal-Organic Frameworksby Sequential Linker ReinstallationLiang Feng,1 Shuai Yuan,1,* Jun-Sheng Qin,1 Ying Wang,1,2,5 Angelo Kirchon,1 Di Qiu,2 Lin Cheng,2

Sherzod T. Madrahimov,4 and Hong-Cai Zhou1,3,6,*

Progress and Potential

Linker exchange has become an

emerging theme in the synthesis

of metal-organic frameworks

(MOFs), but its utility is mainly

restricted to inapplicable MOFs

based on labile, low-valent

metals. In this report, post-

synthetic linker replacement

within an inert framework, usually

deemed difficult or impossible to

accomplish, has been achieved by

linker reinstallation. Labilization of

the initial linker enables

SUMMARY

Isoreticular expansion of metal-organic frameworks (MOFs) by linker elongation

often leads to interpenetration or other undesired structures. Here we report a

sequential linker labilization and reinstallation method to expand the unit cell di-

mensions of MOFs while manipulating the framework structure and interpenetra-

tion. A stable Zr-basedMOF was initially synthesized as a template. Subsequently,

labile linkerswith imine bondswerepost-synthetically introduced into the structure

to destabilize the Zr-MOF. Eventually, gradual dissociation of the imine-based

linkers and reinstallation of longer linkers into the defective spaces lead to the for-

mation of non-interpenetrated isoreticular Zr-MOFs with progressively increased

pore sizes. Similarly, lattice contraction can also be realized by incorporating

shorter linkers. In addition to providing a powerful tool that yields control over

the structures and functions of MOFs, this work also highlights new opportunities

by combining thedynamic covalent chemistrywith coordinationchemistry inMOFs.

sequential insertion of longer or

shorter linkers within a robust

MOF, leading to convenient

framework expansion and

contraction. The enhancement of

framework flexibility via defect

creation results in a powerful tool

for pore engineering, enabling

the state-of-the-art design of

robust frameworks with enhanced

tunability through a combination

of the dynamic covalent approach

and tuning of the coordination

kinetics. The resulting non-

interpenetrated robust MOFs with

large pore sizes epitomize ideal

platforms for drug delivery,

biomolecule immobilization, and

large-molecule catalysis.

INTRODUCTION

Design and synthesis of highly porous metal-organic frameworks (MOFs) with large

pore sizes have attracted significant interest during the past two decades.1–8 The

nanoscale voids inside porous materials enable small molecules such as gases and

even large molecules such as proteins and enzymes to be trapped and trans-

ported.9,10 To expand the pore sizes of MOFs several techniques have been devel-

oped, but none more applicable than isoreticular expansion.2,11 For example, the

isoreticular MOF-74 series was obtained by liner elongation, leading to the expan-

sion of pore sizes from 1.4 to 9.8 nm.12 However, large-pore MOFs are not always

preferred when longer linkers are applied in a solvothermal synthesis. In many

cases, the reaction between clusters and elongated linkers often results in inter-

penetrated structures or undesired topologies, which in turn reduces the pore

sizes of the obtained MOFs. Over the years, many different methods were attemp-

ted in order to control the structure and topology of MOFs by designing non-inter-

penetrated networks or using templates.13 One successful example is a stepwise

linker-exchange strategy, whereby the short linkers of small-pore MOFs are re-

placed by progressively longer ones via post-synthetic linker exchange. Eventually

the cavity of MOFs was expanded while the topology and non-interpenetrated

network were maintained. This method was proved to be effective for Zn-based

MOFs including bio-MOF-100 and a series of layer-pillared structures.14–16

Although successful, the application of stepwise linker exchange is limited to low-

valent metal-based MOFs, in which the M(II)-carboxylate coordination bonds are

Matter 1, 1–12, July 10, 2019 ª 2019 Elsevier Inc. 1

relatively labile. However, the high-valent metal-based MOFs are not able to un-

dergo the exchange process of longer linkers due to the strong M(III/IV)-carboxylate

bonds and rigid framework structures.17 To date, controlling interpenetration of

high-valent metal-based MOFs, such as Zr(IV), has yet to be explored. For instance,

linker elongation of Zr-MOFs with the UiO (University of Oslo) structure often results

in interpenetrated structures, known as porous interpenetrated zirconium-organic

frameworks (PIZOFs).18–20 The interpenetrated Zr-MOFs yield smaller cavities and

more densely packed framework fragments, which are expected to be thermody-

namically more favorable than the non-interpenetrated ones. Therefore, a one-pot

synthesis typically ends up with interpenetrated framework structures as the thermo-

dynamic products, while post-synthetic methods are considered to be favorable

over one-pot syntheses when preparing non-interpenetrated structures. Ideally,

non-interpenetrated Zr-MOFs with short linkers could be used as templates for linker

exchange, with longer linkers allowing for highly porous Zr-MOFs as kinetic products

to be isolated by circumventing the undesirable thermodynamic sink in a one-pot

reaction.

However, direct linker exchange of UiO-type MOFs by slightly longer linkers was

proved to be unsuccessful due to extremely low linker-exchange rates. The slow

linker-exchange rate within UiO-typeMOFs is attributed to the kinetic barrier caused

by the stable Zr(IV)-carboxylate bonds and the rigid framework.21 We proposed that

the kinetic barrier of linker exchange can be overcome by destabilizing the parent Zr-

MOFs. By labilizing and removing the original linkers, secondary linkers with

different lengths can be subsequently reinstalled into the unsaturated coordination

sites. Here, we demonstrate that the exchange of Zr-MOFs with linkers of different

lengths can be achieved by sequential linker labilization and reinstallation. A stable,

non-interpenetrated Zr-MOF was initially synthesized as a template. Subsequently,

labile linkers with imine bonds were post-synthetically introduced into the structure

to destabilize the Zr-MOF, followed by the progressive dissociation and removal of

imine-based linker, which allows for the continuous incorporation of secondary

linkers with different lengths. This method enables pore expansion and contraction

by reinstalling longer or shorter linkers, giving rise to a series of non-interpenetrated

isoreticular Zr-MOFs.

1Department of Chemistry, Texas A&MUniversity, College Station, TX 77843-3255, USA

2College of Chemistry, Tianjin Normal University,Tianjin 300387, China

3Department of Materials Science andEngineering, Texas A&M University, CollegeStation, TX 77843-3003, USA

4Department of Chemistry, Texas A&MUniversityat Qatar, P.O. Box 23874, Doha, Qatar

5Key Laboratory of Advanced Energy MaterialsChemistry (Ministry of Education), NankaiUniversity, Tianjin 300071, China

6Lead Contact

*Correspondence:shuai.yuan@chem.tamu.edu (S.Y.),zhou@chem.tamu.edu (H.-C.Z.)

https://doi.org/10.1016/j.matt.2019.02.002

RESULTS AND DISCUSSION

Overcoming the Kinetic Barrier of Linker Exchange

Although the linker exchange of Zr-MOFs has been reported, linkers with the same

lengths are often required. Linker exchange of Zr-MOFs with longer linkers has

been attempted but to no avail. Considering the strong Zr(IV)-carboxylate bonds

and the rigid framework, the linker-exchange process needs to overcome a high

kinetic barrier. To accelerate the linker-exchange process, the parent Zr-MOF

should be destabilized. As shown in Scheme 1, a stable Zr-MOF is labilized by intro-

ducing organic linkers that contain labile imine bonds. The creation of missing-

linker defects by the subsequent removal of imine linkers introduces flexibility to

the rigid framework, which in turn allows for the installation of longer linkers into

the defect sites. In other words, the labilized MOF will act as a reaction intermedi-

ate, which allows for mild insertion of longer linkers to complete the linker reinstal-

lation process.

To prove our hypothesis, we chose a previously reported Zr-MOF, namely Zr6-

AZDC or UiO-67.5, as the parent MOF.22–26 UiO-67.5 is constructed by 12 con-

nected Zr6 clusters and azobenzene-based linkers (L1, azobenzene-4,40-dicarboxylic

2 Matter 1, 1–12, July 10, 2019

Scheme 1. Overcoming the Kinetic Barrier of Linker Exchange by Linker Labilization and

Reinstallation

acid). The orange color and distinctive ultraviolet-visible (UV-vis) absorption of azo-

benzenemoieties allow for easymonitoring of the linker-exchange process by UV-vis

spectroscopy. Initially, direct linker exchange for longer linkers was attempted for

UiO-67.5 by incubating the crystal of UiO-67.5 in an N,N0-dimethylformamide

(DMF) solution containing L2 (terphenyl-4,400-dicarboxylic acid, 50 mM) at 85�C.The color of the UiO-67.5 crystals as well as the solution was unchanged after

24 h. The powder X-ray diffraction (PXRD) of the sample was also maintained, indi-

cating a failed linker-exchange process.

Instead, UiO-67.5 can be labilized by replacing the stable L1 linker with an imine-

based linker of identical length (L10, 4-carboxybenzylidene-4-aminobenzoic acid,

Scheme 2). The labilized analog of UiO-67.5 is denoted as PCN-161. According

to the literature, L10 can be dissociated into 4-aminobenzoic acid and 4-formylben-

zoic acid through hydrolysis by breaking the imine bond to create missing-linker

defects (Scheme 1).27 As expected, the labile imine linker in PCN-161 can be easily

and successfully exchanged by L2 when incubated in the solution of L2/DMF at

85�C. The linker-exchange process was monitored by 1H-nuclear magnetic reso-

nance (NMR) spectra (Figure S8). 1H-NMR spectra of digested samples revealed

that the L10 were completely replaced by L2 after 24 h (Figure S11). The final prod-

uct was known as UiO-68 and the corresponding PXRD data matched well with

the simulations based on single structure of UiO-68 (Figure S6A). Furthermore,

the transformation occurred in a single-crystal to single-crystal manner, so

that the structure of product can be clearly characterized by single-crystal X-ray

diffraction (SCXRD, Table S1). The structures and lattice parameters of the product

are consistent with the reported values for UiO-68. The morphology of crystals was

unchanged after the linker reinstallation process was performed, although an

obvious color change from yellow to colorless was observed (Figures S18 and

S19). Additionally, the supernatant was analyzed by inductively coupled plasma

mass spectrometry (ICP-MS), which showed no Zr leaching during the linker rein-

stallation process. All of the evidence clearly shows that a single-crystal to sin-

gle-crystal transformation process occurs without dissolving or destructing the

parent framework.

Matter 1, 1–12, July 10, 2019 3

Scheme 2. Stable (L0–L4) and Labile Linkers (L10–L40) Used in This Work

Sequential Linker Labilization and Reinstallation

To further examine the possibility of continuously expanding the unit cell dimensions

of non-interpenetrated Zr-MOFs, we carried out stepwise linker labilization and re-

installation by iteratively applying linker labilization and reinstallation (Figure 2).

A series of stable linkers was selected (L1 to L4) and their imine-based labile analogs

with identical lengths were also designed and synthesized (L10 to L40, Scheme 2).

Sequential linker reinstallation was realized by replacing the short imine linker with

progressively longer ones (Schemes S3–S5). For example, PCN-162 was obtained

by incubating the crystal of PCN-161 in the solution of L20. Sequential exchangeof PCN-162 with L30 and L40 further gave rise to PCN-163 and PCN-164. The com-

plete exchange of each step was characterized by 1H-NMR spectra of digested

4 Matter 1, 1–12, July 10, 2019

Figure 1. Lattice Expansion Monitored by PXRD Analysis

(A) Simulated and experimental PXRD of non-interpenetrated PCN-16X (X = 1–4) obtained by

sequential linker labilization and reinstallation showing the expansion of unit cell.

(B) Experimental and simulated PXRD of interpenetrated PCN-163 and PCN-164 obtained by

one-pot synthesis.

samples (Figures S9–S14). The gradually enlarged unit cell dimensions were clearly

reflected by the peak shift in PXRD patterns (Figure 1A). The unit cell parameters

were determined by PXRD and are shown in Table S2.

Single-crystal structures of PCN-161 and PCN-162 were successfully obtained,

providing direct evidence of linker exchange (Table S1). SCXRD experiments re-

vealed that the PCN-161 crystallizes in cubic space group Fm-3m. Each Zr6 cluster

is connected to 12 L10 linkers, giving rise to a non-interpenetrated framework with

fcu topology. The L10 was 4-fold disordered due to the high symmetry. PCN-162

shows the same network structure as PCN-161. However, it crystallizes in the lower

symmetry space group Pn-3. As a result of the reduced symmetry, the conformation

of L20 can be clearly determined. The replacement of L10 by L20 was clearly observedin single-crystal structures. The cell-edge length of PCN-162 increased by 2.65 A

compared with that of PCN-161. The unit cell parameters determined by SCXRD

matched well with the PXRD results. The diffraction of PCN-163 and PCN-164 crys-

tals were too weak for structure refinement, due to their large unit cells and the

multi-step modifications. Their structural models were built in Materials Studio by

isoreticular expansion of PCN-161 and PCN-162 structures. The simulated PXRD

patterns of PCN-161, PCN-162 (based on single-crystal structures), PCN-163, and

PCN-164 (based on simulated structures) were in good agreement with the experi-

mental data (Figures 1 and S2; Table S2).

It is noteworthy that PCN-163 and PCN-164 were obtained with non-interpenetrated

structures, which would be extremely difficult to obtain from a one-pot synthesis. In

contrast, one-pot synthesis of Zr-MOFs from L30, L40, or linkers with similar lengths

Matter 1, 1–12, July 10, 2019 5

Figure 2. Illustration of Continuous Lattice Expansion and Contraction in MOFs by Sequential Linker Labilization and Reinstallation, which Can Be

Utilized to Generate a Series of Non-interpenetrated Isoreticular MOFs

always generate interpenetrated structures, as indicated by SCXRD and PXRD (Fig-

ure 1B). A single-crystal structure of interpenetrated PCN-164 was obtained, de-

noted as PCN-164-inter (Table S1). PCN-164-inter crystallizes in the cubic space

group Fd-3m. It consists of two sets of independent and mutually interpenetrating

UiO-type frameworks, and similar structures, known as PIZOFs, have been previously

reported.18–20 Due to the low solubility of L30, only microcrystalline powders, de-

noted as PCN-163-inter, were obtained from one-pot synthesis, and the resulting

structural model was built in Materials Studio by isoreticular contraction of PCN-

164-inter. Experimental PXRD patterns of PCN-163-inter and PCN-164-inter indi-

cate the absence of the second diffraction peaks, corresponding to the extinction

of (200) plane diffraction, and this matches well with the simulated PXRD patterns

(Figure S3). The sequential linker labilization and reinstallation allows for the isola-

tion of non-interpenetrated structures that are thermodynamically unfavorable dur-

ing one-pot synthesis.

To further explore the scope of the linker labilization and reinstallation method, we

treated the labilized MOFs with imine linkers (i.e., PCN-16X series) with the solutions

of linear linkers of different lengths. The products were digested and analyzed by1H-NMR to verify the linker reinstallation (Figures S9–S14). The results are summa-

rized in Figure 2. Generally the labilized MOFs can tolerate a linker difference within

approximately 2.50 A. This distance represents the maximum flexibility of the frame-

work (Scheme S6). For example, the imine linker in PCN-161 (L10, 13.12 A) can be re-

placed by L0 (biphenyl-4,40-dicarboxylic acid, 11.00 A), L1 (13.24 A), L2 (15.70 A),

and L20 (15.74 A) through single-crystal to single-crystal transformation. However,

longer linkers such as L30 (17.90 A) or L40 (19.40 A) cannot be accommodated by

PCN-161 (Figure S5). For comparison, the exchange of stable MOFs (i.e., UiO-6X se-

ries) requires the linker length to be identical (Figure S7). This indicates that the labile

linkers enhance the flexibility of the framework so that linkers with different lengths

can be easily incorporated.

Stability and Porosity

The incorporation of labile imine bonds into the PCN-16X series reduced the chem-

ical and mechanical stability of the framework (Figure S4). Although the PCN-16X

6 Matter 1, 1–12, July 10, 2019

Figure 3. Porosity Analysis of Non-interpenetrated UiO Series Prepared from Linker

Reinstallation

(A) N2 sorption of UiO-6X at 77 K, 1 bar.

(B) Pore size distribution of UiO-6X derived from N2 adsorption isotherms.

series is stable in common organic solvents including DMF, methanol, ethanol, and

acetone, the frameworks are sensitive toward water. PXRD patterns show that PCN-

16X partially lost its crystallinity after immersion in water for 24 h. The mechanical

stability of PCN-16X series is also compromised, as indicated by the collapsed

framework after solvent removal (Figure S17). Indeed, direct activation of PCN-

162 from acetone leads to framework collapse as indicated by N2 sorption iso-

therms. For PCN-162 to retain its porosity, the as-synthesized sample needs to be

exchanged with low-surface-tension solvents before activation. Following the litera-

ture,28 PCN-162 was exchanged with CH2Cl2 and hexane, leading to successful acti-

vation (Figures S15–S17). Additionally we failed to activate PCN-163 and PCN-164,

possibly due to the large pore sizes and labile imine linkers.

The stability of PCN-16X series can be enhanced by replacing the labile imine linkers

with stable ones. The resulting materials, known as the UiO series, display good

chemical stability in water as indicated by PXRD patterns. Furthermore, the UiO se-

ries can be directly activated without losing any crystallinity or porosity (Figure 3). N2

sorption isotherms at 77 K show progressively elevated N2 total uptakes, which is in

line with the change of unit cell dimensions (Figure 3A). The pore size distribution

derived from adsorption isotherms further indicates the gradual increase of pore

sizes from 1.5 to 2.2 nm (Figure 3B). These data further verified the non-interpene-

trated structure obtained by sequential linker labilization and reinstallation.

The density change of the samples was also studied during the linker-exchange pro-

cess. As shown in Figure S15, the height of the samples inside an NMR tube

increased from 0.51 to 0.68 cm (PCN-162, pale yellow) after linker reinstallation.

Furthermore, the density difference between PCN-161 and PCN-162 can be directly

observed during the exchange process, as the lighter PCN-162 layer lies above the

heavier PCN-161.

Dynamic Dissociation/Association of Imine Bonds

The conventional linker-exchange process relies on the reversible dissociation/asso-

ciation of coordination bonds. In contrast, sequential linker labilization and reinstal-

lation takes advantage of the dynamic covalent bonds as well. The dynamic imine

bond in the PCN-16X series not only facilitates the linker-exchange process but

also provides additional opportunities to introduce new functional groups into

MOFs. Reactions based on imine bonds have been well developed in organic chem-

istry, which is an essential component in dynamic covalent chemistry.29–31 We

Matter 1, 1–12, July 10, 2019 7

Figure 4. Fragment Exchange Showing the Dynamics of Covalent Imine Bond

showed that the reversible dissociation/association of the imine bonds in PCN-164

enables the substitution of phenylenediamine fragments (Figures 4 and S21). The

linker-fragment exchange was carried out by immersing PCN-164 crystals in the so-

lution of substitute diamines at 85�C for 24 h. PXRD measurements showed that the

crystallinity was maintained after fragment exchange (Figure S21C). 1H-NMR of the

digested samples further verified the exchange with the diamine fragment (Figures

S21B and S21D). The substituents on the diamine fragment can affect the basicity of

the resulting imine moieties, which in turn alters the pore environment of PCN-164.

As a proof of concept, the catalytic activity of functionalized PCN-164 was tested in

Knoevenagel condensation reactions (Figure 4). The Knoevenagel condensation re-

action is a model of C–C bond-forming reactions catalyzed using basic conditions. It

also plays an important role in the synthesis of organic intermediate such as couma-

rins and their derivatives.32–34 Among functionalized PCN-164 samples, the diami-

nocyclohexane-modified sample shows the highest activity due to the enhanced ba-

sicity of the imine sites. Substrates with various functional groups (-H, -Me, -NO2,

and -OH) were converted to their corresponding products in substantial conversion

percentages. Furthermore, the catalytic activity of the MOF catalyst was well main-

tained after three catalytic cycles (Table S5).

In fact the imine bond, as a representative example of dynamic covalent bonds, has

been commonly used as a connection in covalent-organic frameworks (COFs).

Notably, Yaghi and coworkers previously reported an in situ generation of Ti-

MOFs combining the chemistry of MOFs and COFs based on a preformed cluster

assembly process.35 However, this method shows limitations in obtaining large sin-

gle-crystalline MOFs. Our post-synthetic approach brings new opportunities into

the preparation of high-crystalline M(IV)-based MOFs combining the reversible for-

mation of imine bonds during post-synthetic process. Considering the diversity of

both MOFs and dynamic covalent chemistry, a large family of MOFs with various

metal clusters and dynamic linkers can be expected. This work also provides a route

toward activating the inert framework using dynamic covalent chemistry, which can

be potentially applied to a variety of materials including supramolecular cages and

COFs. Research along this line is now under way in our laboratory.

Conclusion

In conclusion, we report the sequential linker labilization and reinstallation as a

powerful tool to tailor pore environments and control interpenetration of isoreticular

8 Matter 1, 1–12, July 10, 2019

MOFs. Continuous expansion (or contraction) of unit cell dimensions of Zr-MOFs was

realized by installing progressively longer (or shorter) linkers into the labilized parent

framework. This method not only allows the construction of non-interpenetrated

Zr-MOFs with large pore sizes, but also facilities the functionalization of pore envi-

ronments by modifying the imine bonds. Furthermore, the idea of utilizing dynamic

covalent chemistry to manipulate the properties of MOFs will lead to new discov-

eries in the field of MOFs and beyond.

EXPERIMENTAL PROCEDURES

Synthesis of UiO-67.5 or PCN-160

ZrCl4 (200 mg), L1 (100 mg), trifluoroacetic acid (1.0 mL), and DMF (20 mL) were

charged in a Pyrex vial. The mixture was heated in an oven at 120�C for 72 h. After

cooling down to room temperature, the crystals were harvested.

Synthesis of PCN-161

PCN-161 was synthesized by repeated linker exchange of PCN-160 with L10 DMF so-

lution. Generally, crystals of PCN-160 (100 mg) were incubated with the solution of

L10 in DMF (80 mM, 20 mL) at 85�C for 24 h. The supernatant was replaced with fresh

solution of L10 in DMF (80 mM, 20 mL) every 6 h.

Synthesis of UiO-68

UiO-68 was synthesized by repeated linker reinstallation of PCN-161 with L2 DMF

solution. Generally, crystals of PCN-161 (100 mg) were incubated with the solution

of L2 in DMF (80 mM, 20 mL) at 85�C for 48 h. The supernatant was replaced with

fresh solution of L2 in DMF (80 mM, 20 mL) every 12 h.

Synthesis of UiO-67

UiO-67 was synthesized by repeated linker reinstallation of PCN-161 with L0 DMF

solution. Generally, crystals of PCN-161 (100 mg) were incubated with the solution

of L0 in DMF (80 mM, 20 mL) at 85�C for 48 h. The supernatant was replaced with

fresh solution of L0 in DMF (80 mM, 20 mL) every 12 h.

Synthesis of PCN-162

PCN-162 could be synthesized through two routes. (1) PCN-162 was synthesized by

repeated linker exchange of UiO-68 with L20 DMF solution. Generally, crystals of

UiO-68 (100 mg) were incubated with the solution of L20 in DMF (80 mM, 20 mL)

at 85�C for 24 h. The supernatant was replaced with fresh solution of L20 in DMF

(80 mM, 20 mL) every 6 h. (2) PCN-162 could also be obtained by direct linker rein-

stallation with PCN-161. Generally, crystals of PCN-161 (100 mg) were incubated

with the solution of L20 in DMF (80 mM, 20 mL) at 85�C for 48 h. The supernatant

was replaced with fresh solution of L20 in DMF (80 mM, 20 mL) every 12 h.

Synthesis of PCN-163

PCN-163 was synthesized by repeated linker reinstallation of PCN-162 with L30

DMSO solution. Generally, crystals of PCN-162 (100mg) were incubated with the so-

lution of L30 in DMSO (80 mM, 20 mL) at 85�C for 72 h. The supernatant was replaced

with fresh solution of L30 in DMSO (80 mM, 20 mL) every 12 h. DMSO was chosen as

the reinstallation solution because L30 shows better solubility in DMSO than in DMF

or DEF.

Synthesis of PCN-164

PCN-164 was synthesized by repeated linker reinstallation of PCN-163 with L40 DMF

solution. Generally, crystals of PCN-163 (20 mg) were incubated with the solution of

Matter 1, 1–12, July 10, 2019 9

L40 in DMF (80 mM, 20 mL) at 85�C for 72 h. The supernatant was replaced with fresh

solution of L40 in DMF (80 mM, 20 mL) every 12 h.

Synthesis of UiO-69

UiO-69 was synthesized by repeated linker exchange of PCN-164 with L4 DMF solu-

tion. Generally, crystals of PCN-164 (100 mg) were incubated with the solution of L4

in DMF (80 mM, 20 mL) at 85�C for 72 h. The supernatant was replaced with fresh

solution of L4 in DMF (80 mM, 20 mL) every 12 h.

Synthesis of PCN-163-Inter

PCN-163-inter was synthesized through one-pot reaction. ZrCl4 (20 mg), L30

(200 mg), benzoic acid (100 mg), and DMF (2 mL) were charged in a Pyrex vial.

The mixture was heated in an oven at 120�C for 48 h. After cooling down to room

temperature, the microcrystalline crystals were harvested.

Synthesis of PCN-164-Inter

PCN-163-inter was synthesized through one-pot reaction. ZrCl4 (20 mg), L40

(200 mg), benzoic acid (100 mg), and DMF (2 mL) were charged in a Pyrex vial.

The mixture was heated in an oven at 120�C for 48 h. After cooling down to room

temperature, the microcrystalline crystals were harvested.

Synthesis of UiO-69-Inter

UiO-69-inter was synthesized through one-pot reaction. ZrCl4 (20 mg), L4 (20 mg),

benzoic acid (200 mg), and DMF (2 mL) were charged in a Pyrex vial. The mixture

was heated in an oven at 120�C for 48 h. After cooling down to room temperature,

the microcrystalline crystals were harvested.

Data Availability

Experimental procedures for the syntheses of the linker, PXRD, N2 adsorption iso-

therms, 1H-NMR spectrum, and crystallographic data of the structure (CIF) are pro-

vided in Supplemental Information. All relevant data supporting the findings of this

study are available from the corresponding authors on request. The X-ray crystallo-

graphic coordinates for structures reported in this study have been deposited at the

Cambridge Crystallographic Data Centre (CCDC) under deposition number CCDC

1900119-1900123.

SUPPLEMENTAL INFORMATION

Supplemental Information can be found online at https://doi.org/10.1016/j.matt.

2019.02.002.

ACKNOWLEDGMENTS

The gas adsorption-desorption studies of this research were supported as part of

the Center for Gas Separations, an Energy Frontier Research Center funded by the

US Department of Energy, Office of Science, Basic Energy Sciences under award

number DE-SC0001015. The SCXRD and PXRD characterization and analysis

were funded by the Robert A. Welch Foundation through a Welch Endowed Chair

to H.-C.Z. (A-0030). The catalysis work was funded by the Qatar National Research

Fund under award number NPRP9-377-1-080, and US Department of Energy Office

of Fossil Energy, National Energy Technology Laboratory (DE-FE0026472). The

authors acknowledge the experimental help and discussion from Mr. Kunyu

Wang, Dr. Liang-Liang Zhang, and Dr. Peng Zhang. Use of the Advanced Photon

Source, an Office of Science User Facility operated for the US Department of En-

ergy (DOE) Office of Science by Argonne National Laboratory, was supported by

10 Matter 1, 1–12, July 10, 2019

the US DOE under contract no. DE-AC02-06CH11357. Parts of chemicals in this

work are provided by CSN Pharm. The authors also acknowledge the financial

support of the National Natural Science Foundation of China (21471113), Training

Program of Outstanding Youth Innovation Team of Tianjin Normal University, Na-

tional Top-Level Talent Training Program of Tianjin Normal University, Training

Program of Young and Middle-Aged Innovative Talent and Backbone, Tianjin

‘‘131’’ Innovative Talent, 111 project (B12015), and Opening Fund of Key Labora-

tory of Advanced Energy Materials Chemistry (Ministry of Education), Nankai

University.

AUTHOR CONTRIBUTIONS

Conceptualization, L.F., S.Y., and H.-C.Z.; Methodology, L.F., and S.Y.; Investiga-

tion, L.F., S.Y., J.-S.Q., and Y.W.; Writing – Original Draft, L.F. and S.Y.; Writing – Re-

view & Editing, A.K., D.Q., L.C., S.T.M., and H.-C.Z.; Funding Acquisition, S.T.M.

and H.-C.Z.; Resources, H.-C.Z.; Supervision, S.Y. and H.-C.Z.

DECLARATION OF INTERESTS

The authors declare no competing financial interests.

Received: December 7, 2018

Revised: January 12, 2019

Accepted: February 22, 2019

Published: March 27, 2019

REFERENCES

1. Cohen, S.M. (2012). Postsynthetic methods forthe functionalization of metal-organicframeworks. Chem. Rev. 112, 970–1000.

2. Eddaoudi, M., Kim, J., Rosi, N., Vodak, D.,Wachter, J., O’Keeffe, M., and Yaghi, O.M.(2002). Systematic design of pore size andfunctionality in isoreticular MOFs and theirapplication in methane storage. Science 295,469–472.

3. Li, J.R., Kuppler, R.J., and Zhou, H.C. (2009).Selective gas adsorption and separation inmetal-organic frameworks. Chem. Soc. Rev. 38,1477–1504.

4. Li, J.R., Sculley, J., and Zhou, H.C. (2012).Metal-organic frameworks for separations.Chem. Rev. 112, 869–932.

5. Lu, W.G., Wei, Z.W., Gu, Z.Y., Liu, T.F., Park, J.,Park, J., Tian, J., Zhang, M.W., Zhang, Q.,Gentle, T., et al. (2014). Tuning the structureand function of metal-organic frameworks vialinker design. Chem. Soc. Rev. 43, 5561–5593.

6. Zhou, H.C., Long, J.R., and Yaghi, O.M. (2012).Introduction to metal-organic frameworks.Chem. Rev. 112, 673–674.

7. Feng, L., Yuan, S., Zhang, L.L., Tan, K., Li, J.L.,Kirchon, A., Liu, L.M., Zhang, P., Han, Y.,Chabal, Y.J., and Zhou, H.C. (2018). Creatinghierarchical pores by controlled linkerthermolysis in multivariate metal-organicframeworks. J. Am. Chem. Soc. 140, 2363–2372.

8. Yuan, S.A., Zou, L.F., Qin, J.S., Li, J.L., Huang,L., Feng, L.A., Wang, X.A., Bosch, M., Alsalme,A., Cagin, T., and Zhou, H.C. (2017).

Construction of hierarchically porous metal-organic frameworks through linker labilization.Nat. Commun. 8, 15356.

9. Lee, J., Farha, O.K., Roberts, J., Scheidt, K.A.,Nguyen, S.T., and Hupp, J.T. (2009). Metal-organic framework materials as catalysts.Chem. Soc. Rev. 38, 1450–1459.

10. Lian, X.Z., Fang, Y., Joseph, E., Wang, Q., Li,J.L., Banerjee, S., Lollar, C., Wang, X., andZhou, H.C. (2017). Enzyme-MOF (metal-organicframework) composites. Chem. Soc. Rev. 46,3386–3401.

11. Kirchon, A., Feng, L., Drake, H.F., Joseph, E.A.,and Zhou, H.C. (2018). From fundamentals toapplications: a toolbox for robust andmultifunctional MOF materials. Chem. Soc.Rev. 47, 8611–8638.

12. Deng, H.X., Grunder, S., Cordova, K.E.,Valente, C., Furukawa, H., Hmadeh, M.,Gandara, F., Whalley, A.C., Liu, Z., Asahina, S.,et al. (2012). Large-pore apertures in a series ofmetal-organic frameworks. Science 336, 1018–1023.

13. Shekhah, O., Wang, H., Paradinas, M., Ocal, C.,Schupbach, B., Terfort, A., Zacher, D., Fischer,R.A., and Woll, C. (2009). Controllinginterpenetration in metal-organic frameworksby liquid-phase epitaxy. Nat. Mater. 8,481–484.

14. Li, T., Kozlowski, M.T., Doud, E.A., Blakely,M.N., and Rosi, N.L. (2013). Stepwise ligandexchange for the preparation of a family ofmesoporous MOFs. J. Am. Chem. Soc. 135,11688–11691.

15. Karagiaridi, O., Bury, W., Tylianakis, E.,Sarjeant, A.A., Hupp, J.T., and Farha, O.K.(2013). Opening metal-organic frameworksvol. 2: inserting longer pillars into pillared-paddlewheel structures through solvent-assisted linker exchange. Chem. Mater. 25,3499–3503.

16. Liu, C., Zeng, C.J., Luo, T.Y., Merg, A.D., Jin,R.C., and Rosi, N.L. (2016). Establishingporosity gradients within metal-organicframeworks using partial postsynthetic ligandexchange. J. Am. Chem. Soc. 138, 12045–12048.

17. Yuan, S., Feng, L., Wang, K., Pang, J., Bosch,M., Lollar, C., Sun, Y., Qin, J., Yang, X., Zhang,P., et al. (2018). Stable metal-organicframeworks: design, synthesis, andapplications. Adv. Mater. 30, e1704303.

18. Lippke, J., Brosent, B., von Zons, T., Virmani, E.,Lilienthal, S., Preusse, T., Hulsmann, M.,Schneider, A.M., Wuttke, S., Behrens, P., andGodt, A. (2017). Expanding the group ofporous interpenetrated Zr-organic frameworks(PIZOFs) with linkers of different lengths. Inorg.Chem. 56, 748–761.

19. Schaate, A., Roy, P., Preusse, T., Lohmeier, S.J.,Godt, A., and Behrens, P. (2011). Porousinterpenetrated zirconium-organic frameworks(PIZOFs): a chemically versatile family of metal-organic frameworks. Chem. Eur. J. 17, 9320–9325.

20. Cavka, J.H., Jakobsen, S., Olsbye, U., Guillou,N., Lamberti, C., Bordiga, S., and Lillerud, K.P.(2008). A new zirconium inorganic buildingbrick forming metal organic frameworks with

Matter 1, 1–12, July 10, 2019 11

exceptional stability. J. Am. Chem. Soc. 130,13850–13851.

21. Bai, Y., Dou, Y.B., Xie, L.H., Rutledge, W., Li,J.R., and Zhou, H.C. (2016). Zr-based metal-organic frameworks: design, synthesis,structure, and applications. Chem. Soc. Rev.45, 2327–2367.

22. Yang, Q., Guillerm, V., Ragon, F., Wiersum,A.D., Llewellyn, P.L., Zhong, C., Devic, T., Serre,C., andMaurin, G. (2012). CH4 storage and CO2

capture in highly porous zirconium oxide basedmetal-organic frameworks. Chem. Commun.48, 9831–9833.

23. Hoang, L.T.M., Ngo, L.H., Nguyen, H.L.,Nguyen, H.T.H., Nguyen, C.K., Nguyen, B.T.,Ton, Q.T., Nguyen, H.K.D., Cordova, K.E., andTruong, T. (2015). An azobenzene-containingmetal-organic framework as an efficientheterogeneous catalyst for direct amidation ofbenzoic acids: synthesis of bioactivecompounds. Chem. Commun. 51, 17132–17135.

24. Marshall, R.J., Hobday, C.L., Murphie, C.F.,Griffin, S.L., Morrison, C.A., Moggach, S.A., andForgan, R.S. (2016). Amino acids ashighly efficient modulators for singlecrystals of zirconium and hafnium metal-organic frameworks. J. Mater. Chem. A 4,6955–6963.

12 Matter 1, 1–12, July 10, 2019

25. Gao, W.-Y., Thiounn, T., Wojtas, L., Chen, Y.-S.,and Ma, S. (2016). Two highly poroussingle-crystalline zirconium-based metal-organic frameworks. Sci. China Chem. 59,980–983.

26. Epley, C.C., Love, M.D., and Morris, A.J. (2017).Characterizing defects in a UiO-AZB metal-organic framework. Inorg. Chem. 56, 13777–13784.

27. Yuan, S., Zou, L.F., Li, H.X., Chen, Y.P., Qin, J.S.,Zhang, Q., Lu, W.G., Hall, M.B., and Zhou, H.C.(2016). Flexible zirconium metal-organicframeworks as bioinspired switchable catalysts.Angew. Chem. Int. Ed. 55, 10776–10780.

28. Ma, J., Kalenak, A.P., Wong-Foy, A.G., andMatzger, A.J. (2017). Rapid guestexchange and ultra-low surface tensionsolvents optimize metal-organic frameworkactivation. Angew. Chem. Int. Ed. 56, 14618–14621.

29. Rowan, S.J., Cantrill, S.J., Cousins, G.R.L.,Sanders, J.K.M., and Stoddart, J.F. (2002).Dynamic covalent chemistry. Angew. Chem.Int. Ed. 41, 898–952.

30. Jin, Y.H., Yu, C., Denman, R.J., and Zhang, W.(2013). Recent advances in dynamiccovalent chemistry. Chem. Soc. Rev. 42,6634–6654.

31. Zhang, G., Tsujimoto, M., Packwood, D.,Duong, N.T., Nishiyama, Y., Kadota, K.,Kitagawa, S., and Horike, S. (2018).Construction of a hierarchical architecture ofcovalent organic frameworks via apostsynthetic approach. J. Am. Chem. Soc.140, 2602–2609.

32. Shen, K., Zhang, L., Chen, X.D., Liu, L.M.,Zhang, D.L., Han, Y., Chen, J.Y., Long, J.L.,Luque, R., Li, Y.W., and Chen, B.L. (2018).Ordered macro-microporous metal-organicframework single crystals. Science 359,206–210.

33. Dhakshinamoorthy, A., Heidenreich, N.,Lenzen, D., and Stock, N. (2017). Knoevenagelcondensation reaction catalysed by Al-MOFswith CAU-1 and CAU-10-type structures.Crystengcomm 19, 4187–4193.

34. Fang, Q.R., Gu, S., Zheng, J., Zhuang, Z.B., Qiu,S.L., and Yan, Y.S. (2014). 3Dmicroporous base-functionalized covalent organic frameworks forsize-selective catalysis. Angew. Chem. Int. Ed.53, 2878–2882.

35. Nguyen, H.L., Gandara, F., Furukawa, H., Doan,T.L.H., Cordova, K.E., and Yaghi, O.M. (2016).A titanium-organic framework as an exemplarof combining the chemistry of metal- andcovalent-organic frameworks. J. Am. Chem.Soc. 138, 4330–4333.