Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks

Article

Architectural Stabilization of a Gold(III)

Catalyst in Metal-Organic Frameworks

John S. Lee, Eugene A.

Kapustin, Xiaokun Pei,

Sebastia´n Llopis, Omar M.

Yaghi, F. Dean Toste

fdtoste@berkeley.edu

HIGHLIGHTS

A structurally well-deﬁned gold(III)

precatalyst was introduced into

two MOFs

Unimolecular decomposition of a

gold(III) catalyst in the MOFs was

suppressed

SXRD data revealed that the

geometry of the catalyst was

architecturally restrained

Incorporated gold(III) catalyst in

MOFs demonstrated excellent

recyclability

Toste and co-workers present a unique strategy for suppressing a unimolecular

decomposition pathway of a transition-metal catalyst via architectural

stabilization. The structural rigidity of metal-organic frameworks was utilized to

constrain the geometry of a gold(III) catalyst to suppress catalyst decomposition

by reductive elimination and, therefore, improve catalyst stability. Architectural

stabilization is anticipated to serve as a general strategy for the preservation of

ligand geometry in otherwise unstable systems.

Lee et al., Chem 6, 1–11

January 9, 2020 ª 2019 Elsevier Inc.

https://doi.org/10.1016/j.chempr.2019.10.022

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Article

Architectural Stabilization

of a Gold(III) Catalyst

in Metal-Organic Frameworks

John S. Lee,¹Eugene A. Kapustin,^1,2,3Xiaokun Pei,^1,2,3Sebastia´n Llopis,¹Omar M. Yaghi,^1,2,3,4,5

and F. Dean Toste^1,6,

*

SUMMARY

The Bigger Picture

Catalysis is one of the principles of

green chemistry, as catalysts have

the potential to promote a

Unimolecular decomposition pathways are challenging to address in transition-

metal catalysis and have previously not been suppressed via incorporation into a

solid support. Two robust metal-organic frameworks (IRMOF-10 and bio-

chemical reaction without

MOF-100) are used for the architectural stabilization of

a structurally

themselves being consumed.

However, in many cases, catalysts

can succumb to undesired

well-deﬁned gold(III) catalyst. The inherent rigidity of these materials is utilized

to preclude a unimolecular decomposition pathway—reductive elimination.

Through this architectural stabilization strategy, decomposition of the in-

corporated gold(III) catalyst in the metal-organic frameworks is not observed;

in contrast, the homogeneous analog is prone to decomposition in solution.

Stabilization of the catalyst in these metal-organic frameworks precludes leach-

ing and enables recyclability, which is crucial for productive heterogeneous

catalysis.

processes that limit the number of

times they can promote a reaction

(turnover number) or their reuse.

For example, homogeneous

transition-metal catalysts can

suffer unimolecular or bimolecular

decomposition reactions and can

be challenging to recycle. This

manuscript demonstrates that the

incorporation of a transition-metal

catalyst, based on gold(III), into

metal-organic frameworks (MOFs)

enabled both facile recovery and

recyclability compared with those

of its homogeneous analogs.

Moreover, by constraining the

geometry of the transition-metal

catalyst, the architectural rigidity

of MOFs suppressed a

INTRODUCTION

In mechanochemistry, tensile forces have traditionally been utilized to promote

various bond cleavage events,^1–3which can enable productive chemistry through

ring opening,^4–6rearrangement,⁷and catalyst activation.^8,9More recently, this

strategy has been applied toward preserving chemical bonds by suppressing an

undesired unimolecular decomposition pathway—a retro-Michael pathway of a

maleimide-thiol adduct (Scheme 1A).¹⁰Despite these advances in mechanochem-

istry, the static force provided by rigid materials has, to the best of our knowledge,

previously not been utilized toward the preservation of ligand geometry that are

sensitive to bending effects. In cases where reductive elimination is problematic in

transition-metal catalysis,^11–13rigidiﬁcation of ligands could potentially suppress

such unimolecular decomposition pathways. Traditionally, solid-state supports

have addressed bimolecular decomposition pathways of catalysts;^14–16however,

unimolecular decomposition pathways of homogeneous catalysts have previously

not been suppressed with solid-state supports. As a model system, we were inter-

ested in leveraging architectural stabilization to prevent a unimolecular decomposi-

tion pathway of IPrAu(III)(biphenyl)X (where IPr is [1,3-bis(2,6-diisopropylphenyl)

imidazole-2-ylidene] and X^–is a non-coordinating counteranion), which is known

to undergo reductive elimination to yield biphenylene and IPrAu(I)X (Scheme

1B).¹³We reasoned that a bifunctionalized IPrAu(III)(biphenyl)X catalyst could

unimolecular decomposition

pathway (reductive elimination).

These ﬁndings enumerate a

strategy for the design of stable

and reusable transition-metal

catalysts that are otherwise prone

to unimolecular decomposition

pathways.

be incorporated into

a robust porous material to architecturally lock the

geometry of the catalyst. Contrary to common solid-state supports, metal-organic

frameworks (MOFs) allow for the precise placement of molecules in a well-ordered

Chem 6, 1–11, January 9, 2020 ª 2019 Elsevier Inc.

1

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

fashion,^17–23which can constrain the geometry of incorporated guests within the

framework. Here, we demonstrate that a unimolecular decomposition pathway of

IPrAu(III)(biphenyl)X catalyst is prohibited due to architectural stabilization in

MOFs by preserving the geometry of the gold(III) complex such that the linear geom-

etry of the biphenyl ligand is maintained (Scheme 1B). Deviation of its linear geom-

etry is architecturally forbidden because reductive elimination in a rigid MOF would

either necessitate the formation of a defect site or result in strain throughout the

material. In particular, we demonstrate two strategies for introducing a gold(III)

catalyst into MOFs with two distinct biphenyldicarboxylate (BPDC) binding modes,

which resulted in no observation of the reductive elimination products in contrast to

its homogeneous analogs.

RESULTS AND DISCUSSION

In order to probe the architectural stabilization of a gold(III) catalyst in MOFs, we

ﬁrst incorporated

a gold(III) precatalyst, IPrAu(III)(BPDC)Cl, into two MOFs

through a mixed-linker synthesis and a solvent-assisted linker exchange (see

Experimental Procedures for details). IPrAu(III)(BPDC)Cl was encompassed into a

microporous IRMOF-10 system through a mixed-linker synthesis strategy with

5%–16% occupancy of gold(III) precatalyst to yield IPrAu(III)Cl-IRMOF-10 (Fig-

ure 1A). A partial structure of IPrAu(III)Cl-IRMOF-10 (5% gold occupancy) with

primitive cubic (pcu) topology was conﬁrmed via single-crystal X-ray diffraction

(SXRD) and powder X-ray diffraction (PXRD). However, the electron density of

the gold, chloride, and IPr moieties could not be assigned with the IPrAu(III)Cl-

IRMOF-10 SXRD data. The chemical composition of IPrAu(III)Cl-IRMOF-10 was

determined by ¹H nuclear magnetic resonance (NMR) analysis and inductively

coupled plasma-atomic emission spectroscopy (ICP-AES) of the digested samples

with an observed IPrAu(III)(BPDC)Cl to BPDC ratio of 16:84 and zinc to gold ratio

of 89:11. Mixed-linker syntheses with a targeted H₂IPrAu(III)(BPDC)Cl/H₂BPDC

ratio > 5:95 in IRMOF-10 yielded crystals that were not suitable for SXRD charac-

terization due to an increase in defects, which affected the overall crystallinity;

an increase in defects was attributed to steric clash in the framework between

gold(III) complexes bearing bulky IPr ligands. Low gold(III) loading and high

symmetry in IRMOF-10 with 5% occupancy of gold(III) complex precluded the

possibility of obtaining a full crystal structure of the precatalyst in IPrAu(III)Cl-

IRMOF-10 via SXRD to ensure that the catalytic moiety is indeed geometrically

constrained between secondary building units (SBUs).

In order to mitigate steric clash and increase the occupancy of the gold(III) species in

MOFs, bio-MOF-100,²⁴a mesoporous MOF featuring larger pores (ꢀ4.5 nm), was

chosen as a support for SXRD characterization of the gold(III) precatalyst and cata-

lyst. Interestingly, bio-MOF-100 features two symmetrically distinct BPDC moieties

in its asymmetric unit. This feature offers the opportunity to selectively displace one

type of BPDC in the framework and thus make it amenable to low guest loading with

high occupancy and, as a consequence, enable SXRD characterization (see Supple-

mental Information). Additionally, bio-MOF-100 possesses a unique binding mode

of BPDC to the SBU compared to that of IRMOF-10, which offers an alternative strat-

egy of rigidifying the biphenyl ligand. IPrAu(III)(BPDC)Cl was incorporated into mes-

oporous bio-MOF-100 through a solvent-assisted linker exchange with up to 40%

occupancy of gold(III) precatalyst per substitutable site to yield IPrAu(III)Cl-bio-

MOF-100 (Figure 1B), where the occupancy of gold(III) precatalyst was assigned

via SXRD. With 40% occupancy of gold(III) precatalyst, gold, chloride, and 7 atoms

of the IPr ligand could be assigned in the IPrAu(III)Cl-bio-MOF-100 SXRD crystal

¹Department of Chemistry, University of

California, Berkeley, Berkeley, CA, USA

²Materials Science Division, Lawrence Berkeley

National Laboratory, Berkeley, CA, USA

³Kavli Energy NanoSciences Institute at Berkeley,

Berkeley, CA, USA

⁴Berkeley Global Science Institute, Berkeley, CA,

USA

⁵UC Berkeley-KACST Joint Center of Excellence

for Nanomaterials for Clean Energy Applications,

King Abdulaziz City for Science and Technology,

Riyadh, Saudi Arabia

⁶Lead Contact

*Correspondence: fdtoste@berkeley.edu

https://doi.org/10.1016/j.chempr.2019.10.022

2

Chem 6, 1–11, January 9, 2020

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

A previous work — mechanochemical stabilization of a C—S bond

without force

with force

F

N

O

NH

HS

O

+H₂O

OH

+

N

O

S

F

retro-Michael products

stable C—S bond

B this work — architectural stabilization of Au—C bonds

without architectural stabilization

with architectural stabilization

+

dipp

–

N

X

OH₂

N

dipp = 2,6-diisopropylphenyl

+

Au

dipp

–

X

N

dipp

+

Au

OH₂

120º

180º

stable Au—C bonds

reductive elimination products

Scheme 1. Mechanochemical and Architectural Stabilization of Chemical Bonds

structure (Figure S7). The chemical composition of IPrAu(III)Cl-bio-MOF-100 was

further conﬁrmed by ¹H NMR and ICP-AES analysis of digested samples.

The catalytically active, cationic gold(III) species in the IRMOF-10 system were ac-

cessed by the treatment of IPrAu(III)Cl-IRMOF-10 with 1 equiv of AgSbF₆(relative

to gold) to access IPrAu(III)SbF₆-IRMOF-10, which is analogous to the conditions

for activation of the homogeneous complex.²⁵Chloride abstraction was evidenced

by the reactivity observed in the IPrAu(III)SbF₆-IRMOF-10-catalyzed cycloisomeriza-

tion reaction of 1,5-enyne substrate 1 to yield the corresponding bicyclohexene

product 2 (Table 1, entry 1). In contrast, the addition of substrate 1 to IPrAu(III)Cl-

IRMOF-10 without AgSbF₆treatment, under otherwise equivalent conditions,

resulted in no background reactivity (entry 2). Additionally, subjecting pristine

IRMOF, with or without AgSbF₆treatment, to substrate 1 did not yield any product

(entries 3 and 4). These observations support the conclusion that the zinc-based

SBUs and silver salt are not responsible for the reactivity observed with IPrAu(III)

Chem 6, 1–11, January 9, 2020

3

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Figure 1. Structures of IPrAu(III)Cl-IRMOF-10 and IPrAu(III)Cl-bio-MOF

Structure of IPrAu(III)Cl-IRMOF-10 (A) obtained from modeling. Partial structure of IPrAu(III)Cl-bio-MOF-100 (B) was identiﬁed from single-crystal X-ray

diffraction; the remainder of the structure (dipp groups on IPr) was obtained from modeling. Only one gold complex is shown in the structures, while the

other symmetrically equivalent positions are omitted for clarity.

SbF₆-IRMOF-10. Additionally, another IPrAu(III)SbF₆-IRMOF-10-catalyzed cycloiso-

merization reaction yielded a product distribution that was consistent with that of

the homogeneous gold(III) analog, which further supports the conclusion that

cationic gold(III) species are responsible for the observed reactivity (Table S1).

Efforts toward accessing cationic gold(III) species in bio-MOF-100 through AgSbF₆

treatment resulted in a decrease in crystallinity. We posited that this MOF degrada-

tion might be attributed to protonation of the BPDC linkers by HSbF₆generated

from the hydrolysis of AgSbF₆in the presence of adventitious water. Addition of Na-

BAr^F₄, a common halide-abstracting agent that is less prone to hydrolysis,²⁶also

yielded poorly crystalline frameworks, presumably due to hard acid-hard base inter-

actions between sodium cations and the carboxylate-based linkers. Thus, TlPF₆was

chosen as the halide-abstracting agent, as it is less sensitive toward hydrolysis and

features a soft thallium cation. After TlPF₆treatment of IPrAu(III)Cl-bio-MOF-100,

the sample retained crystallinity, yielding the desired IPrAu(III)PF₆-bio-MOF-100.

SXRD measurement revealed the preservation of 40% gold(III) occupancy in the

framework, which is consistent with that of the precatalyst structure. In the crystal

4

Chem 6, 1–11, January 9, 2020

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Table 1. Control Experiments with IRMOF

Entry

MOF

Conversion (%)

1

2

3

4

IPrAu(III)SbF₆-IRMOF-10 (3 mol % Au)

IPrAu(III)Cl-IRMOF-10 (3 mol % Au)

IRMOF-9-AgSbF₆

44

<1

IRMOF-9

See Experimental Procedures for general reaction conditions.

structure of IPrAu(III)PF₆-bio-MOF-100, the chloride ligand was no longer observed,

which indicates successful halide abstraction from the precatalyst to form the

desired cationic catalyst (Figure S8).

IPrAu(III)PF₆-bio-MOF-100 had very low reactivity toward the cycloisomerization re-

action of substrate 1 to product 2; this low reactivity was attributed to a potential

decrease in the rate of diffusion of nonpolar substrates through the intrinsically

anionic bio-MOF-100 framework. Raising the temperature to increase the rate of

diffusion of 1 was, however, not compatible with this cycloisomerization reaction

due to the thermal instability of 1,5-enynes. Thus, alkynyl cylcoheptatriene substrate

3 was chosen as a model substrate, as it has higher thermal stability than 1 and is

known to undergo a gold-catalyzed cycloisomerization reaction to yield the corre-

sponding indene products 4 and 5.²⁷Addition of 3 to IPrAu(III)PF₆-bio-MOF-100

resulted in formation of desired products 4 and 5 at elevated temperatures with

consistent product selectivity with that of the homogeneous gold(III) analog (Table 2,

entry 1; Table S1). Similar to the IRMOF-based gold(III) reactivity, no product was

observed in the corresponding control experiments with bio-MOF-100 (entries

2–4). These data further demonstrate that chloride abstraction from the gold(III) pre-

catalyst was successful by TlPF₆treatment of IPrAu(III)Cl-bio-MOF-100.

The chemical stability of this cationic gold(III) catalyst in IRMOF-10 and bio-MOF-

100 was compared with the stability of their homogeneous counterparts. Reductive

elimination of IPrAu(III)(biphenyl)SbF₆is known to be exacerbated in the presence of

a trap for cationic gold(I) species, 1,3,5-trimethoxybenzene (TMB), to yield a Au(I)-

TMB adduct (Table 3, entry 3).¹³In contrast, no 2,7-biphenylene dicarboxylic acid

or Au(I)-TMB adduct was observed with the MOF analogs under equivalent

conditions in the supernatant by digestion ¹H NMR analysis or ICP-AES (entries 1

and 2). Additionally, we observed 78% decomposition of a homogeneous gold(III)

analog to the corresponding reductive elimination products when heated to 55^ꢁC

(entry 5). In bio-MOF-100, the gold(III) occupancy remained unperturbed under

these conditions, and no reductive elimination products were observed in the

supernatant by ¹H NMR analysis or ICP-AES (entry 4). These results are consistent

with the hypothesis that the architectural stabilization afforded by IRMOF-10 and

bio-MOF-100 is robust enough to prevent this unimolecular decomposition.

As a further evaluation that catalysis was occurring in the pores of the framework

rather than at the surface or in bulk solution, the impact of substrate size was evalu-

ated in IRMOF-10, as it features smaller pore dimensions than bio-MOF-100

Chem 6, 1–11, January 9, 2020

5

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Table 2. Control Experiments with bio-MOF

Entry

MOF

Conversion (%)

4:5

1

2

3

4

IPrAu(III)PF₆-bio-MOF-100 (7 mol % Au)

IPrAu(III)Cl-bioMOF-100 (7 mol % Au)

bio-MOF-100-TlPF₆

96

<1

60:40

–

bio-MOF-100

See Experimental Procedures for general reaction conditions.

(Figure 2). A Au(III)-IRMOF-10-catalyzed reaction of 1,5-enyne substrate 6, which is

slightly larger along one dimension, did not show a substantial decrease in reactivity

compared to substrate 1. On the other hand, when the steric bulk was extended

along two dimensions with substrate 7, a decrease in reactivity to 2% conversion af-

ter 22 h was observed. Further extension of steric bulk along these two dimensions

with substrate 8 resulted in no observed product formation by ¹H NMR. In contrast,

full conversion was observed with substrates 1, 6, 7, and 8 with 4 mol % homoge-

neous IPrAu(biphenyl)SbF₆after 22 h (see Supplemental Information). These data

are consistent with the hypothesis that the catalysis observed with IPrAu(III)SbF₆-

IRMOF-10 occurs within the pores, and the leaching of catalytically active species

into solution is unlikely. The lack of catalytically active species in solution further

highlights the effectiveness of architectural stabilization to prohibit the formation

of undesired gold(I) species in bulk solution.

After evaluating the stability of both Au(III)-MOF systems toward reductive elimina-

tion, the reuse of both systems was evaluated to further assess the impact of the

architectural stability of these frameworks on the catalyst recyclability and longevity.

To this end, employing IPrAu(III)SbF₆-IRMOF-10 with 3 mol % gold loading as a cata-

lyst, 44% and 46% conversion of enyne 1 to bicyclohexene 2 was observed in cycles 1

and 2, respectively (Figure 3A). Reactivity toward the cycloisomerization of 1 per-

sisted in cycles 3–5, albeit at lower conversions. We hypothesized that this decrease

in reactivity might be attributed to the slow trapping of Au(III)Cl species in the pres-

ence of AgCl within the pores, which is a phenomenon that has previously been

observed with solid-supported cationic gold species.²⁸Indeed, we observed a

rebound in reactivity upon treatment of IPrAu(III)SbF₆-IRMOF-10 with 1 equiv of

AgSbF₆(relative to gold) in cycle 6 with continued reactivity over the subsequent cy-

cle. Recyclable reactivity over 7 cycles was also observed with IPrAu(III)PF₆-bio-

MOF-100-catalyzed cycloisomerization reaction of 3 to yield 4 and 5 (Figure 3B),

and recyclable reactivity was also observed at shorter reaction durations (Table

S3). Additionally, no loss in the reactivity of IPrAu(III)PF₆-bio-MOF-100 was observed

after storing the catalyst for 29 days, demonstrating that catalyst deactivation does

not occur even after long-term storage. Recyclability reactivity in both IRMOF-10

and bio-MOF-100 further demonstrates the robustness of these systems engen-

dered by architectural stabilization. This architectural stabilization strategy should

prove general to access other immobilized transition-metal catalysts that are other-

wise prone to unimolecular decomposition pathways and are consequently unstable

or inaccessible in solution.

6

Chem 6, 1–11, January 9, 2020

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Table 3. Stability of Homogeneous Gold(III) Complexes versus MOF Analogs

Entry

Catalyst

T (^ꢁC)

25

Additive

Decomposition (%)

1

2

3

4

IPrAu(III)SbF₆-IRMOF-10

IPrAu(III)PF₆-bio-MOF-100

IPrAu(III)(biphenyl)SbF₆

IPrAu(III)PF₆-bio-MOF-100^a

TMB (10 equiv)

none

<5

68

<5

78

25

55

a

IPrAu(III)(Me₂BPDC)SbF₆

55

none

See Supplemental Information for general reaction conditions.

^aCDCl₃was used instead of CD₂Cl₂.

EXPERIMENTAL PROCEDURES

Synthesis of Au(III)-MOFs for Catalysis

IPrAu(III)Cl-IRMOF-10

To a 2 dram vial was added IPrAu(H₂BPDC)Cl (8.6 mg, 0.010 mmol, 0.25 equiv),

H₂BPDC (7.3 mg, 0.030 mmol, 0.75 equiv), Zn(NO₃)₂$4 H₂O (42 mg, 0.16 mmol,

4.0 equiv), and diethyl formamide (2.14 mL). The reaction mixture was capped, son-

icated for 5 min, then heated at 90^ꢁC in an oven for 24 h. This yielded yellow crystals,

which were washed with DMF (6 mL 3 5), DCM (6 mL 3 15), then MeNO₂(6 mL 3 5).

Due to a loss of crystallinity of IPrAu(III)Cl-IRMOF-10 in the absence of solvent, the

crystals were immersed in solvent prior to AgSbF₆treatment. 16% loading

IPrAu(BPDC)Cl versus BPDC was observed by digestion ¹H NMR analysis. Zn:Au ra-

tio of 88:12 observed by ICP-AES (expected Zn:Au ratio: 89:11).

IPrAu(III)SbF₆-IRMOF-10

To a 2 mL vial was added IPrAu(III)Cl-IRMOF-10 (16% IPrAu(BPDC)Cl loading, 2 mg,

1 equiv) immersed in MeNO₂(0.1 mL), followed by the addition of a solution of

AgSbF₆(7 mM in MeNO₂, 0.13 mL, 1 equiv). After 48 h, the crystals were washed

with MeNO₂(2 mL 3 3) then DCM (2 mL 35). Due to a loss of crystallinity of

IPrAu(III)SbF₆-IRMOF-10 in the absence of solvent, the crystals were left immersed

in solvent prior to catalysis.

IPrAu(III)Cl-bio-MOF-100

To a 1 dram vial was added bio-MOF-100 (20 mg), IPrAu(H₂BPDC)Cl (20 mg), and

DMF (0.9 mL). The reaction mixture was capped, then heated at 50^ꢁC in an oven for

96 h. This yielded colorless crystals, which were washed with DMF (3 mL 3 7), DCM

(3 mL 3 15), then MeNO₂(3 mL 3 5). Due to a loss of crystallinity of IPrAu(III)Cl-bio-

MOF-100 in the absence of solvent, the crystals were immersed in solvent prior to

TlPF₆treatment. 15% loading IPrAu(BPDC)Cl versus BPDC was observed by digestion

¹H NMR. Zn:Au ratio of 89:11 observed by ICP-AES (expected Zn:Au ratio: 90:10).

IPrAu(III)PF₆-bio-MOF-100

To a 2 mL vial was added IPrAu(III)Cl-bio-MOF-100 (15% IPrAu(BPDC)Cl loading,

5 mg, 1 equiv) immersed in MeNO₂(0.1 mL), followed by the addition of a solution

of TlPF₆(15 mM in MeNO₂, 0.10 mL, 1 equiv). After 48 h, the crystals were washed

Chem 6, 1–11, January 9, 2020

7

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Figure 2. Impact of Substrate Size on Catalysis with IPrAu(III)SbF₆-IRMOF-10

See Experimental Procedures for general reaction conditions.

with MeNO₂(2 mL 3 3) then CHCl₃(2 mL 3 5). Due to a loss of crystallinity of

IPrAu(III)PF₆-bio-MOF-100 in the absence of solvent, the crystals were left immersed

in solvent prior to catalysis.

General Procedures for Catalysis with Au(III)-MOFs

IPrAu(III)SbF₆-IRMOF-10-Catalyzed 1,5-Enyne Cycloisomerization Reaction

To a 2 mL vial was added IPrAu(III)SbF₆-IRMOF-10 (16% IPrAu(BPDC)Cl loading, 0.03

equiv) immersed in DCM (0.3 mL), followed by the addition of 1,5-enyne (1 equiv).

After 22 h, the organic supernatant was removed with DCM (2 mL 3 5). The washed

IPrAu(III)SbF₆-IRMOF-10 crystals were resubjected to the same conditions for

recyclability studies. Conversions were determined by ¹H NMR spectroscopy. No

additional conversion was detected in the supernatant upon removal of IPrAu(III)

SbF₆-IRMOF-10. ¹H NMR spectra of products 2 and 6 match those previously

reported.²⁹For control experiments, an equivalent amount of IPrAu(III)Cl-IRMOF-

10 or IRMOF-9 was used instead of IPrAu(III)SbF₆-IRMOF-10. 14% loading

IPrAu(BPDC)Cl versus BPDC was observed by digestion ¹H NMR analysis for

IPrAu(III)SbF₆-IRMOF-10 after catalysis. Zn:Au ratio of 88:12 observed for IPrAu(III)

SbF₆-IRMOF-10 by ICP-AES after catalysis.

IPrAu(III)PF₆-bio-MOF-100-Catalyzed Alkynyl Cycloheptatriene

Cycloisomerization Reaction

To a 2 mL vial was added IPrAu(III)PF₆-bio-MOF-100 (15% IPrAu(BPDC)Cl loading,

5 mg, 0.0015 mmol, 0.07 equiv) immersed in CHCl₃(0.1 mL), followed by the addi-

tion of substrate 3 (5 mg, 0.022 mmol, 1 equiv). After heating the reaction mixture at

55^ꢁC for 46 h, the organic supernatant was removed with CHCl₃(2 mL 3 5). The

washed IPrAu(III)PF₆-bio-MOF-100 crystals were resubjected to the same conditions

for recyclability studies. Conversions were determined by ¹H NMR spectroscopy.

¹H NMR spectra of products 4 and 5 match those previously reported.²⁷For control

experiments, an equivalent amount of IPrAu(III)Cl-bio-MOF-100 or bio-MOF-100

was used instead of IPrAu(III)PF₆-bio-MOF-100. 14% loading IPrAu(BPDC)Cl versus

8

Chem 6, 1–11, January 9, 2020

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

Figure 3. Recyclability of IPrAu(III)SbF₆-IRMOF-10 and Au(III)PF₆-bio-MOF-100

IPrAu(III)SbF₆-IRMOF-10 (A) and Au(III)PF₆-bio-MOF-100 (B). See Experimental Procedures for

general reaction conditions.

BPDC was observed by digestion ¹H NMR analysis for IPrAu(III)PF₆-bio-MOF-100

after catalysis. Zn:Au ratio of 88:12 observed for IPrAu(III)PF₆-bio-MOF-100 by

ICP-AES after catalysis.

Other experimental details, procedures, and characterization data (Figures S1–S40

and Tables S1–S7) are provided in the Supplemental Information.

Chem 6, 1–11, January 9, 2020

9

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

DATA AND CODE AVAILABILITY

The accession numbers for the crystallographic data associated with the reported

structures in this paper are CCDC: 1955738 [IPrAu(III)(H2BPDC)Cl], CCDC:

1955739 [IPrAu(III)(Me2BPDC)Cl], CCDC: 1955737 [IPrAu(III)Cl-bio-MOF-100], and

CCDC: 1955736 [IPrAu(III)PF6-bio-MOF-100].

SUPPLEMENTAL INFORMATION

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

2019.10.022.

ACKNOWLEDGMENTS

F.D.T. thanks the NIHGMS (R35 GM118190), J.S.L. thanks the NSF-GRFP (DGE

1106400 and 1752814), and S.L. thanks predoctoral fellowships from MINECO

(BES-2015-072627) for ﬁnancial support. Financial support for MOF SXRD and

PXRD studies in the O.M.Y. laboratory was provided by King Abdulaziz City for Sci-

ence and Technology as part of a joint KACSTÀUC Berkeley collaboration (Center of

Excellence for Nanomaterials and Clean Energy Applications). We thank Dr. Suhong

Kim, Dr. Christian S. Diercks, Banruo Huang, Dr. Cynthia M. Hong, Dr. Patrick T. Bo-

han, Edward Miller, Danny Q. Thach, and Jeffrey S. Derrick for helpful discussions.

O.M.Y. acknowledges the collaboration, input, and support of Prince Turki bin

Saud bin Mohammed Al-Saud (president of KACST). This research used resources

of beamlines 11.3.1 and 12.2.1 at Advanced Light Source, which is a DOE Ofﬁce

of Science User Facility under contract no. DE-AC02-05-CH11231.

AUTHOR CONTRIBUTIONS

J.S.L. conceptualized the experiments. J.S.L. synthesized the organic and organo-

metallic compounds. J.S.L. and E.A.K. synthesized the MOFs. E.A.K. and X.P.

collected and analyzed the SXRD and PXRD data. J.S.L. performed the catalytic, re-

cyclability, and stability studies. X.P. collected and analyzed the ICP-AES data. S.L.

performed the recyclability and solvent screening studies. F.D.T. supervised the

project. J.S.L. wrote the original manuscript. All authors proofread, commented

on, and approved the ﬁnal manuscript for submission.

DECLARATION OF INTERESTS

The authors declare no competing interests.

Received: April 18, 2019

Revised: May 16, 2019

Accepted: October 29, 2019

Published: November 25, 2019

REFERENCES AND NOTES

1. Caruso, M.M., Davis, D.A., Shen, Q., Odom,

S.A., Sottos, N.R., White, S.R., and Moore, J.S.

(2009). Mechanically-induced chemical

changes in polymeric materials. Chem. Rev.

109 , 5755–5798.

activation of covalent bonds. Chem. Rev. 105 ,

2921–2948.

6. Kean, Z.S., Niu, Z., Hewage, G.B., Rheingold,

A.L., and Craig, S.L. (2013). Stress-

responsive polymers containing cyclobutane

core mechanophores: reactivity and

mechanistic insights. J. Am. Chem. Soc. 135 ,

13598–13604.

4. Hickenboth, C.R., Moore, J.S., White, S.R.,

Sottos, N.R., Baudry, J., and Wilson, S.R. (2007).

Biasing reaction pathways with mechanical

force. Nature 446 , 423–427.

2. May, P.A., and Moore, J.S. (2013). Polymer

mechanochemistry: techniques to generate

molecular force via elongational ﬂows. Chem.

Soc. Rev. 42 , 7497–7506.

5. Chen, Z., Mercer, J.A.M., Zhu, X., Romaniuk,

J.A.H., Pfattner, R., Cegelski, L., Martinez, T.J.,

Burns, N.Z., and Xia, Y. (2017).

Mechanochemical unzipping of insulating

polyladderene to semiconducting

7. Diesendruck, C.E., Steinberg, B.D., Sugai, N.,

Silberstein, M.N., Sottos, N.R., White, S.R.,

Braun, P.V., and Moore, J.S. (2012). Proton-

coupled mechanochemical transduction: a

mechanogenerated acid. J. Am. Chem. Soc.

134 , 12446–12449.

3. Beyer, M.K., and Clausen-Schaumann, H.

(2005). Mechanochemistry: the mechanical

polyacetylene. Science 357 , 475–479.

10

Chem 6, 1–11, January 9, 2020

Please cite this article in press as: Lee et al., Architectural Stabilization of a Gold(III) Catalyst in Metal-Organic Frameworks, Chem (2019), https://

doi.org/10.1016/j.chempr.2019.10.022

8. Piermattei, A., Karthikeyan, S., and Sijbesma,

R.P. (2009). Activating catalysts with mechanical

force. Nat. Chem. 1 , 133–137.

chemistry. J. Organomet. Chem. 691 , 5448–

5455.

22. Zhang, T., and Lin, W. (2014). Metal–organic

frameworks for artiﬁcial photosynthesis and

photocatalysis. Chem. Soc. Rev. 43 , 5982–5993.

15. Zhang, T., Manna, K., and Lin, W. (2016). Metal–

organic frameworks stabilize solution-

inaccessible cobalt catalysts for highly efﬁcient