Optimized Immobilization Strategy for Dirhodium(II) Carboxylate Catalysts for C?H Functionalization and Their Implementation in a Packed Bed Flow Reactor

Article abstract of DOI:10.1002/anie.202005381

Herein we demonstrate a packed bed flow reactor capable of achieving highly regio- and stereoselective C?H functionalization reactions using a newly developed Rh₂(S-2-Cl-5-CF₃TPCP)₄ catalyst. To optimize the immobilized dirhodium catalyst employed in the flow reactor, we systematically study both (i) the effects of ligand immobilization position, demonstrating the critical factor that the catalyst-support attachment location can have on the catalyst performance, and (ii) silica support mesopore length, demonstrating that decreasing diffusional limitations leads to increased accessibility of the active site and higher catalyst turnover frequency. We employ the immobilized dirhodium catalyst in a simple packed bed flow reactor achieving comparable yields and levels of enantioselectivity to the homogeneous catalyst employed in batch and maintain this performance over ten catalyst recycles.

Full text of DOI:10.1002/anie.202005381

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Optimized immobilization strategy for dirhodium (II) carboxylate
catalysts for C–H functionalization and their implementation in a
packed bed flow reactor
Authors: Taylor A Hatridge, Wenbin Liu, Chun-Jae Yoo, Huw M L
Davies, and Christopher W Jones
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202005381
Link to VoR: https://doi.org/10.1002/anie.202005381

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
Optimized immobilization strategy for dirhodium (II) carboxylate
catalysts for C–H functionalization and their implementation in a
packed bed flow reactor
Taylor A. Hatridge,+[a] Wenbin Liu,+[b] Chun-Jae Yoo,+[a] Huw M. L. Davies,*[b] Christopher W. Jones*[a]
[a]
Ms. Taylor A. Hatridge, Dr. Chun-Jae Yoo, Prof. Dr. Christopher W. Jones
School of Chemical & Biomolecular Engineering
Georgia Institute of Technology
311 Ferst Dr, Atlanta, GA 30332, USA
E-mail: cjones@chbe.gatech.edu
[b]
Dr. Wenbin Liu, Prof. Dr. Huw M. L. Davies
Department of Chemistry
Emory University
1515 Dickey Drive, Atlanta, GA 30322, USA
E-mail: hmdavie@emory.edu
+These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of the document.
Abstract: Herein we demonstrate a packed bed flow reactor capable
of achieving highly regio- and stereoselective C–H functionalization
reactions using a newly developed Rh2(S-2-Cl-5-CF3TPCP)4 catalyst.
To optimize the immobilized dirhodium catalyst employed in the flow
reactor, we systematically study both (i) the effects of ligand
immobilization position, demonstrating the critical factor that the
catalyst-support attachment location can have on the catalyst
performance, and (ii) silica support mesopore length, demonstrating
that decreasing diffusional limitations leads to increased accessibility
of the active site and higher catalyst turnover frequency. We employ
the immobilized dirhodium catalyst in a simple packed bed flow
reactor achieving comparable yields and levels of enantioselectivity to
the homogeneous catalyst employed in batch and maintain this
performance over ten catalyst recycles.
Despite the promise of dirhodium catalyzed C–H functionalization
as an alternate methodology, certain obstacles have prevented
widespread industrial acceptance of rhodium carbene mediated
C–H functionalization. For example, a key challenge is the safety
concern associated with handling large quantities of toxic,
unstable diazo compounds at the manufacturing site;4 however,
strategic execution of flow chemistry may allow the upstream
synthesis of diazo compounds and their immediate downstream
consumption.5,6 Another barrier for industrial implementation is
the high cost of the noble metal and ligands, as well as the
intricate synthesis of the dirhodium tetracarboxylate catalyst
complex (Rh2L4). This challenge is best mitigated by maximizing
the total turnover number (TON) of the dirhodium catalyst, which
can be achieved via a range of techniques to prolong the catalyst
lifetime, including: catalyst design and optimization, increasing
the catalyst turnover frequency (TOF), or decreasing the rate of
catalyst deactivation.7 Additionally, if catalyst deactivation rates
are not too significant, catalyst TON can be elevated by catalyst
recycle, which can be facilitated by the development of
heterogeneous catalysts8 that are amenable to deployment in flow
reactors. Thus, various groups have studied the immobilization of
molecular catalysts based on numerous chiral ligand scaffolds.9
Davies et al. immobilized an array of chiral dirhodium catalysts on
Introduction
The use of noble metal catalysts to achieve selective C–H
functionalization reactions has enabled the development of
numerous new synthetic methodologies, unlocking novel routes
for total synthesis.1 In particular, the use of dirhodium catalysts
and diazo compounds has led to the ability to achieve highly
regio- and stereoselective reactions accessed through metal
carbene intermediates.2 The application of this rhodium carbene
chemistry to the pharmaceutical industry could help transform the
large-scale manufacture of medicinal drugs. Because
conventional organic synthesis of biologically active molecules
requires many steps and subsequent purifications, drug synthesis
often generates extensive amounts of waste and results in a
relatively low yield of the respective target molecules.3 Thus, there
is both environmental and economic pressure to implement more
efficient methods to achieve the synthesis of pharmaceutical
compounds via innovative engineering and chemistry solutions.
a
pyridine functionalized, crosslinked resin; however, the
noncovalent coordination between pyridine and rhodium may
result in catalyst leaching during reaction.10 As an alternative
immobilization technique, Takeda et al. modified a chiral ligand of
several dirhodium catalysts that were then copolymerized with
styrene and a flexible crosslinker to provide polymeric catalyst
materials that offered relatively stable catalytic activity over
multiple recycles.11 Our group previously designed and
synthesized a polymer/silica composite hollow fiber reactor, in
which the embedded silica particles were functionalized with
immobilized dirhodium carboxylate catalysts, including (p-
dodecylphenyl-sulfonyl)prolinato (DOSP) and (4-bromophenyl)-
2,2-diphenyl-cyclopropanecarboxylato (p-BrTPCP) ligands.6b,12
However, the use of the unusual polymeric hollow fiber reactor
and the limited knowledge of the impact of different immobilization
strategies for the target carboxylate ligands are two aspects of our
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
prior work that could limit rapid deployment of C–H
functionalization in a practical flow process.
B
A
In recent years, Davies and coworkers have achieved the C–H
functionalization of unactivated C–H bonds using the sterically
hindered
dirhodium
tetrakis(1,2,2-triarylcyclopropane
carboxylate) catalysts, Rh2(TPCP)4. With substituents at different
positions of the C1-aryl rings, Rh2(TPCP)4 adopted defined high-
symmetry conformations and was capable of catalyzing highly
site- and stereoselective C–H functionalization reactions.
Modifying the identity and location of functional groups on the aryl
rings altered the symmetry and selectivity of the Rh2(TPCP)4
catalyst,13 which has inspired us to conduct a systematic
exploration on the effect of the location of linker installation for
catalyst immobilization in the ortho-Cl Rh2(TPCP)4 catalysts. In
this work, we evaluate the effect of ligand immobilization position
and employ a conventional, rapidly deployable fixed bed flow
reactor using a newly developed dirhodium catalyst, Rh2(S-2-Cl-
5-CF3TPCP)4, supported on porous silica powders for selective
functionalization of unactivated secondary (C2) sp3 C–H bonds.
Top
B
A
B
C
C
Bottom
Side
Figure 1. Chemical and crystal structures of Rh2(S-o-ClTPCP)4 catalyst (Teal:
Rh, Red: O, Green: Cl, Grey: C, White: H). Aryl rings are labeled A, B, and C to
show the possible locations for the introduction of a covalently bound linker to
tether the catalyst to a heterogeneous silica support.
Results and Discussion
Optimizing Immobilization Position
Therefore, we hypothesized that the installation of the linker would
be preferred on either ring B or C, as placement on ring A might
interfere with the active site and attenuate the catalyst’s activity
and selectivity. To test this hypothesis, we synthesized three
Rh2(S-o-ClTPCP)4 derivatives with an ethynyl substituent on rings
A, B, and C, respectively. The synthesis of the complexes
When considering the appropriate immobilization strategy for any
chiral catalyst, one must account for the catalyst’s inherent
symmetry and the nature of the catalyst’s binding site so as to
preserve its intrinsic performance. For the class of Rh2(TPCP)4
catalysts containing a triarylcyclopropane scaffold, the linker
could potentially be installed on any of the three phenyl rings;
however, these aryl rings are critical components of the catalyst’s
chiral pocket. To rigorously study the effect of linker location on
each of these rings, we chose Rh2(S-o-ClTPCP)4 as our model
catalyst, as its unique C4 symmetric configuration consists of the
three aryl rings oriented in three different directions (as shown in
Figure 1).13b The C1 o-Cl-aryl ring A points towards the active
rhodium site of the catalyst and constitutes the chiral reaction
pocket (top view); the C2-phenyl ring B, which is cis to the o-Cl-
aryl group, points to the equatorial side of the Rh–O disk and is
furthest away from the Rh center (side view); while the other C2-
phenyl ring C, which is trans to the o-Cl-aryl group, is at the
“closed” rhodium face, where four phenyl rings tilt toward each
other and block the second potential active rhodium site (bottom
view).
involved
a six-step sequence (Figure 2, see Supporting
Information for detailed synthesis). The first step is a rhodium-
catalyzed asymmetric cyclopropanation of an aryldiazoacetate
with the 1,1-diarylethylene. The derivative 8a was generated
using the symmetric 1,1-diphenylethylene, while the other two
derivatives (8b and 8c) were obtained as diastereomers using 1-
phenyl,1’-(p-bromophenyl)ethylene. The TIPS-protected ethynyl
group was then installed by a palladium-catalyzed Sonogashira
coupling, followed by hydrolysis to reveal the desired carboxylic
acid ligand. Then, a controlled mono-ligand exchange, followed
by deprotection gave the Rh2(S-o-ClTPCP)4 derivative with
terminal ethynyl group at the desired location (9a-c). Further
modification
via
copper-catalyzed
[3+2]
alkyne–azide
cycloaddition creates two families of modified catalysts: an
analogous homogeneous catalyst linked to a “dummy” silane (2a,
2b, 2c, Figure 2) and a heterogeneous catalyst linked to a solid
silica support (Syloid®, Grace) (3a, 3b, 3c, Figure 2).
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 2. The synthesized homogeneous Rh2(S-o-ClTPCP)4 derivative catalysts with a) dummy silane groups at A, B, and C positions, and b) the immobilized
Rh2(S-o-ClTPCP)4 derivative catalysts via A, B, and C positions on a silica support
In this way, we were able to decouple the potential effects of the
additional triazolepropyl trimethylsilane group covalently bound to
the catalyst’s chiral ligand and a possible steric hindrance
between the catalytic active site and the silica support surface.
The catalyst derivatives 2 and 3 were then subjected to a
reference reaction between 4-bromopentylbenzene (10) and
trifluoroethyl 4-bromophenyldiazoacetate (11). The substrate, 10,
has two possible C–H bonds that can be functionalized: the
electronically favored benzylic C–H bond and the sterically
favored terminal methylene (C2) C–H bond. Due to the sterically
limited active site constructed by the four o-Cl-aryl rings in the
Rh2(S-o-ClTPCP)4 catalyst, the C2 C–H bonds with less steric
bulk are preferred, resulting in exceptionally high regioselectivity
toward C2 C–H functionalization (91:9 rr, entry 1, Figure 3).13e
The “dummy” silane-linked homogeneous catalysts (2a, 2b, 2c)
stereoselective manner; however, the site selectivity and
enantiomeric performance were slightly influenced by the ligand
modification (Figure 3). Previous computational studies on the
ortho-Cl-TPCP catalysts have revealed that this class of catalysts
is quite rigid13e and consequently, the linker is reasonably well
acommodated at all three positions; however, placing the linker at
certain positions may alter the orientation of the chiral ligands and
introduce steric hindrance at the catalytic active site. For catalyst
2a, the placement of the “dummy” silane on ring A resulted in a
yield (72%), regioselectivity (88:12), and enantiomeric excess
(72%) that were markedly lower than those of the unmodified
catalyst and the other derivatives 2b and 2c. Thus, we may
conclude that the presence of a linker at position A affects the
active site structure and may sterically hinder the substrate’s
approach to the reactive carbene, supporting our earlier
hypothesis. Additionally, the heterogeneous silica-supported
all conducted the desired C–H functionalization in
a
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
catalysts (3a, 3b, 3c) showed similar reaction profiles to their
homogeneous “dummy” linked counterparts (2a, 2b, 2c),
demonstrating that changing the nature of the catalyst (from
soluble to silica-supported) did not significantly impede its
reactivity. However, the results confirmed that the location of the
linker does affect catalyst performance. While the derivatives with
linkers at rings B and C showed overall similar reactivity, placing
the linker on ring A resulted in further decreased selectivity and
yield. After immobilization on the silica surface, the active site of
the catalyst 3a would open toward the silica support (entry d,
Figure 2), where the surface species might interact with the active
site or hinder the substrate accessibility. Thus, we decided to
move forward with the linker located on ring B, as its equatorial
position is conserved in other catalyst geometries in the
tetracarboxylate family, making it the optimal anchor site for all
known TPCP-based dirhodium catalysts.13, 14
(0.20 mmolRh2/g) supports. The prepared regular and platelet
SBA-15 supports have similar surface area (573 m2/g vs 576
m2/g), pore volume (1.3 cm3/g vs 1.1 cm3/g), and pore diameter
(12 nm vs 13 nm); the only distinguishable difference is pore
length, which can be verified by SEM and TEM images (Figure
4). While the regular SBA-15 has a straight or slightly bent rod-
shaped structure (pore length >1 µm), the platelet SBA-15 has
short (< 400 nm), straight pores.
(a)
(b)
1000
Regular SBA-15
Platelet SBA-15
Regular SBA-15
Platelet SBA-15
25
800
600
400
200
0
20
15
10
5
0
0.0
0.2
0.4
0.6
0.8
1.0
1
10
Pore size (nm)
100
Relative pressure (P/P0)
(c)
(d)
(e)
(f)
Figure 3. Investigation of the effect of immobilization location (A, B, or C) and
catalyst nature (homogeneous as 2a-c or immobilized on a silica support as 3a-
c) on Rh2(S-o-ClTPCP)4 catalyst activity and selectivity.
200 nm
200 nm
Figure 4. (a) Cryogenic nitrogen adsorption isotherms and (b) pore size
distributions of regular and platelet SBA-15. SEM and TEM images of (c, e)
regular SBA-15 (d, f) platelet SBA-15. The white lines represent pore directions.
Tuning the Solid Silica Support
Next, we sought to modify the solid support as a means to probe
whether active site accessibility affects the reactivity of the silica-
supported Rh2(S-o-ClTPCP)3(ethynyl@B) catalysts. SBA-15 is a
model, ordered mesoporous silica support material containing
one dimensional cylindrical mesopores.15 While the well-defined
nature of the support material and uniform pore size and structure
allows the synthesis of well-defined supported catalysts, the
presence of long straight mesopores surrounding the active
rhodium sites can lead to substrate or product transport limitations,
depending on the reaction conditions and the diameter and length
of the mesopores. Given that the dirhodium catalyzed C–H
functionalization reactions are generally quite fast,16 we
postulated that transport through the channels could have a non-
negligible effect on the overall reaction rate. Moreover, the
decomposition of diazo compounds generates nitrogen gas,
which may result in significant transport limitations in the porous
support structure of our heterogeneous system. Thus, we
hypothesized that shortening the pore length of the SBA-15 silica
support would result in a faster diffusion of the reactants and
products to and from the catalyst immobilized within the SBA-15
mesopores, while allowing the evolved nitrogen gas to more
rapidly escape from the catalyst. To test this hypothesis, the
Rh2(S-o-Cl-TPCP)3(ethynyl@B) catalyst was immobilized on
regular SBA-15 (0.18 mmolRh2/g) and quasi-2D, platelet SBA-15
The two SBA-15 immobilized catalysts were subjected to a batch
cyclopropanation reaction between trichloroethyl diazoacetate
and styrene, which was monitored via in situ FTIR. The rate of the
dirhodium-catalyzed cyclopropanation reaction was evaluated
based on the disappearance of the peak at 2100 cm-1, which
corresponds to the C=N2 stretch of the diazo compound (Figure
5). The 0.13 M solution of 13 was consumed by the Rh2(S-o-Cl-
TPCP)4 catalyst analogs at different rates: the homogeneous
catalyst completed the reaction in 9 min, the catalyst immobilized
on the quasi-2D support completed the reaction in 55 min, and
the catalyst immobilized on the regular SBA-15 support
completed the reaction in 125 min. The TOF of initial diazo
consumption followed a similar trend. The homogenous Rh2(S-o-
Cl-TPCP)4 catalyst gave a TOF of 45100 h-1,which was noticeably
faster than that of the immobilized catalysts, and the TOF for the
platelet SBA-15 immobilized catalyst (13650 h-1) was higher than
that of the regular SBA-15 immobilized catalyst (7460 h-1),
demonstrating that shortening the support channels results in an
increased reaction rate.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
0.15
Homogeneous Rh catalyst
Heterogeneous Rh/Platelet SBA-15
Heterogeneous Rh/Regular SBA-15
Figure 6. Comparison of catalytic activity and selectivity between the Rh2(S-2-
Cl-5-BrTPCP)4 and newly developed Rh2(S-2-Cl-5-CF3TPCP)4 catalysts.
0.10
0.05
0.00
7460 h-1
Implementation in Packed Bed Flow Reactor
13650 h-1
Next, our optimized immobilized catalyst was implemented in flow.
To mitigate pressure drop within the packed bed reactor and
improve catalyst performance, the immobilized Rh2(S-2-Cl-5-
CF3TPCP)3(platelet SBA-15@B) was diluted with commercial
silica (SiliaFlash P60). However, the abundance of silanols on the
hydrophilic silica surface resulted in the introduction of O–H
insertion reactions, which competed with the desired C–H
functionalization. Therefore, we passivated the silanol groups with
hexamethyldisilazane (HMDS) to create a hydrophobic silica
surface that would attenuate O–H insertion reactions. The
preparation of the packed bed and the flow system configuration
are summarized in the Supporting Information.
45100 h-1
0
10 20 30 40 50 60 70 80 90 100 110 120 130 140
Reaction time (min)
Figure 5. Effect of the length of mesopores in SBA-15 silica support on reaction
time and catalyst TOF. Immobilization of the catalyst within a quasi-2D, platelet
SBA-15 support mitigates some diffusional limitations of reactant and product
to and from the dirhodium active site, increasing catalyst TOF.18
Development of Rh2(S-2-Cl-5-CF3TPCP)4 Catalyst
The performance of the packed bed reactor, along with a
comparison of performance to that of the homogeneous Rh2(S-2-
Cl-5-CF3TPCP) catalyst employed in batch, are shown in Figure
7. With two exceptions (20c, 20d), the C–H insertion products
were synthesized with high yields (75-93%), as well as regio- and
stereoselectivities (86-95% ee) commensurate with those
obtained via the homogeneous/batch analog (20a-b, 20e-i).
Although substrates 20c and 20d were obtained with lower yields
(32% and 25%, respectively), a relatively high enantioselectivity
(82% ee) was maintained in both reactions. These low yields are
presumably caused by the slower rate of diazo decomposition due
to the presence of the strongly electron-withdrawing nitro group
(20c) and the coordination between the rhodium active site and
the pyridine nitrogen (20d) for these two substrates. The similarity
in performance between the heterogeneous catalyst implemented
in the packed bed flow reactor and the homogeneous catalyst
employed in batch supports our previous hypothesis that the
immobilization via the equatorial ring B is robust and may be
broadly applied to other Rh2(TPCP)4-derived catalysts.
Additionally, the weight hourly space velocity (WHSV) of our
system is large (570 hr-1 for substrates 20a-b, 20e-i; 285 hr-1 for
substrates 20c,d) compared to some other reported supported
catalysts for site- and enantioselective organic reactions.17 This
high WHSV and the low residence time (53 s for substrates 20a-
b, 20e-i; 106 s for substrates 20c,d) show that our system has an
efficient use of catalyst and may be capable of achieving high
productivity to enable the scale-up of this system.
After optimizing the catalyst immobilization position and silica
support material, we sought to employ the optimal catalyst in the
Rh2(S-o-ClTPCP)4 series in a packed bed flow reactor for C–H
insertion of C2 C–H bonds. The Rh2(S-2-Cl-5-BrTPCP)4 catalyst
(16) consistently gives enantio- and regioselectivities that
outperform those of the Rh2(S-o-ClTPCP)4 catalyst.13e However,
the installation of the ethynyl group, which is necessary for
immobilization, requires a Sonogashira coupling with an aryl
bromide leaving group. The presence of two aryl bromides on the
Rh2(S-2-Cl-5-BrTPCP)3(Br@B) catalyst presents a selectivity
issue that may result in a loss of control over the site of catalyst
immobilization. Thus, we decided to develop a new Rh2(S-2-Cl-5-
BrTPCP)4 analog with similar or better performance. To do so, the
bromide moiety on the o-Cl-aryl ring would need to be replaced
by a substituent with similar steric and electronic properties, but
with no lability to Sonogashira coupling; hence, a trifluoromethyl
group was proposed as a replacement functional group. The new
Rh2(S-2-Cl-5-CF3TPCP)4 catalyst (17) was synthesized (See
Supporting Information for detailed synthesis) and subjected to
the reference reaction with 1-bromo-4-pentylbenzene. The
performance of the new catalyst was found to be comparable in
site-, diastereo-, enantioselectivity, and yield to that of the Rh2(S-
2-Cl-5-BrTPCP)4 catalyst (Figure 6). Thus, the Rh2(S-2-Cl-5-
CF3TPCP)4 was chosen as the optimal dirhodium catalyst for our
flow reaction purposes and was subjected to immobilization on
the platelet SBA-15 support via ring B.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
location is removed from the vicinity of the active site. To increase
the immobilized catalyst’s TOF, the dirhodium carboxylate
catalysts were immobilized on a quasi-2D, mesoporous platelet
SBA-15 support via the equatorial aryl ring. Using this optimized
immobilization strategy, the newly developed Rh2(S-2-Cl-5-
CF3TPCP)4 catalyst was deployed as a heterogeneous catalyst in
a packed bed flow reactor to achieve enantioselective C–H
functionalization reactions with similar reactivity and selectivity to
the homogeneous catalyst utilized in batch. This catalytic
performance was maintained over multiple recycles; therefore, we
expect that this work may contribute to larger-scale synthesis of
pharmaceutical compounds via the coupling of flow synthesis of
diazo compounds6 combined with flow synthesis using optimal
immobilized dirhodium catalysts.
Acknowledgements
Financial support was provided by NSF under the CCI Center for
Selective C−H Functionalization (CHE‐1700982). This material
is based upon work supported by the National Science
Foundation Graduate Research Fellowship Program under
Grant No. DGE–1650044. Any opinions, findings, and
conclusions or recommendations expressed in this material are
those of the author(s) and do not necessarily reflect the views of
the National Science Foundation.
Figure 7. Substrate scope showing Rh2(S-2-Cl-5-CF3TPCP)4 catalyst
performance in flow (top, blue) and batch (bottom, green) reactions. The flow
reactions were conducted at a flow rate of 0.02 mL/min, with the exception of
substrates 20c and 20d, which were run at a flow rate of 0.01 mL/min to
accommodate decreased reactivity.
Keywords: C–H activation • supported catalyst • carbene •
One of the advantages of our catalyst immobilization strategy is
that it enables the recycling of the dirhodium catalyst. To conduct
our recyclability study, the same batch of catalyst was subjected
to ten consecutive reactions without repacking the column
between runs. As shown in Figure 8, the yields were consistent,
mesoporous silica • diazonium compound
[1]
a) K. Godula, D. Sames, Science 2006, 312 (5770), 67-72; b) J. Wencel-
Delord, F. Glorius, Nat. Chem. 2013, 5 (5), 369-375; c) J. Yamaguchi, A.
D. Yamaguchi, K. Itami, Angew. Chem. Int. Ed. 2012, 51 (36), 8960-
9009; d) D. A. Colby, R. G. Bergman, J. A. Ellman, Chem Rev 2010, 110,
624-655; e) D. A. Coby, A. S. Tsai, R. G. Bergman, J. A. Ellman,
Accounts of Chem. Res. 2012, 45(6), 814-825; f) W. R. Gutekunst, P. S.
Baran, Chem. Soc. Rev. 2011, 40(4), 1845-2040.
with only
a slight decrease from 75% to 67%, and the
enantioselectivities were quite stable, decreasing only slightly
from 93% to 89%. As we have previously shown,4b the rhodium
catalyst is not prone to leaching from the silica support and the
slight decrease in activity can be attributed to a slow deactivation
of the catalyst over time.
[2]
[3]
a) M. P. Doyle, D. C. Forbes, Chem. Rev. 1998, 98, 911-936; b) M. P.
Doyle, R. Duffy, M. Ratnikov, L. Zhou, Chem. Rev. 2010, 110, 704-724;
c) A. Ford, H. Miel, A. Ring, C. N. Slattery, A. R. Maguire, M. A. McKervey,
Chem. Rev. 2015, 115, 9981-10080.
a) C. Jimenez-Gonzalez, D. J. Constable, C. S. Ponder, Chem. Soc. Rev.
2012, 41 (4), 1485-1498; b) C. Jimenez-Gonzalez, C. S. Ponder, Q. B.
Broxterman, J. B. Manley, Org. Process Res. Dev. 2011, 15 (4), 912-
917; c) R. A. Sheldon, Green Chem. 2007, 9 (12) 1273–1283; d) L.
Vaccaro, M. Curini, F. Ferlin, D. Lanari, A. Marrocchi, O. Piermatti, V.
Trombettoni, Pure Appl. Chem. 2018, 90 (1), 21-33.
[4]
[5]
a) J. H. Simpson, A. S. Kotnis, R. P. Deshpande, D. J. Kacsur, J. Hamm,
G. Kodersha, W. Merkl, D. Domina, S. Y. Wang in Managing Hazardous
Reactions and Compounds in Process Chemistry, (Eds.: Pesti, J. A.;
Abdel-Magid, A. F.), American Chemical Society: Washington DC, 2014,
pp. 235−244; b) S. P. Green, K. M. Wheelhouse, A. D. Payne, J. P. Hallet,
P. W. Miller, J. A. Bull, Org. Process Res. Dev. 2020, 24(1), 67-84.
a) T. Ye, M. A. McKervey, Chem. Rev. 1994, 94, 1091−1160; b) M. A.
McKervey, Chem. Rev. 2015, 115, 9981−10080; c) S. T. R. Muller, A.
Murat, D. Maillos, P. Lesimple, P. Hellier, T. Wirth, Chem. Eur. J. 2015,
21, 7016-7020; d) J. S. Poh, S. Makai, T. von Keutz, D. N. Tran, C.
Battilocchio, P. Pasau, S. V. Ley, Angew. Chem. Int. Ed. 2015, 54, 7920-
7923; e) N. M. Roda, D. N. Tran, C. Battilocchio, R. Labes, R. J. Ingham,
J. M. Hawkins, S. V. Ley, Org. Biomol. Chem. 2015, 13, 2550-2554; f) D.
N. Tran, C. Battilocchio, S. B. Lou, J. M. Hawkins, S. V. Ley, Chem. Sci.
2015, 6, 1120-1125; g) B. Pieber, C. O. Kappe, Org. Lett. 2016, 18,
1076-1079; h) K.A. Mix, M. R. Aronoff, R. T. Raines, ACS Chem. Biol.
2016, 11, 3233−3244; i) H. Wang, B. Martin, B. Schenkel, Org. Process
Res. Dev. 2018, 22, 446-456; j) R. J. Sullivan, G. P. R. Freure, S. G.
Newman, ACS Catal. 2019, 9, 5623-5630; k) E. M. D. Allouche, A. B.
Charette, Chem. Sci., 2019, 10, 3802-3806.
Figure 8. Recyclability results for the immobilized Rh2(S-2-Cl-5-
CF3TPCP)3(platelet SBA-15@B) catalyst. Yield and selectivity are largely
maintained over the ten reaction cycles.
Conclusion
To optimize the immobilization of dirhodium tetracarboxylate
catalysts, we systematically designed and synthesized an array
of dirhodum catalysts with an inert linker covalently tethered to
different aryl ring sites. The catalyst immobilized via the aryl ring
at the equatorial position showed catalytic activity and selectivity
most similar to those of the unmodified catalyst, as the linker
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
[6]
a) D. Rackl, C. J. Yoo, C. W. Jones, H. M. L. Davies, Org. Lett. 2017, 19
(12), 3055-3058; b) C. J. Yoo, D. Rackl, W. Liu, C. B. Hoyt, B. Pimentel,
R. P. Liveley, H. M. L. Davies, C. W. Jones, Angew. Chem. Int. Ed. 2018,
57 (34), 10923-10927.
[7]
[8]
a) C. W. Jones Top. Catal. 2010, 53 (13-14), 942-952; b) S. L. Scott ACS
Catal. 2018, 8, 8597-8599.
a) D. E. De Vos, M. Dams, B. F. Sels, P. A. Jacobs, Chem. Rev., 2002,
102, 3615-3640 ; b) X. S. Zhao, X. Y. Bao, W. Guo, F. Y. Lee, Mater.
Today, 2006, 9, 32-39; c) S. Shylesh, V. Schünemann, W. R. Thiel,
Angew. Chem. Int. Ed., 2010, 49(20), 3428-3459.
[9]
a) C. E. Song, S. Lee, Chem. Rev. 2002, 102, 3495−3524; b) A. Hu, H.
L. Ngo, W. Lin, J. Am. Chem. Soc. 2003, 125, 38, 11490–11491; c) P.
McMorn, G. J. Hutchings, Chem. Soc. Rev., 2004, 33, 108-122; d) C. Li,
Catal. Rev., 2004, 46, 419-492 ; e) Q. H. Xia, H. Q. Ge, C. P. Ye, Z. M.
Liu, K. X. Su, Chem. Rev., 2005, 105(5), 1603–1662; f) A. Corma, H.
Garcia, Adv. Synth. Catal., 2006, 348, 1391–1412; g) M. Heitbaum, F.
Glorius, I. Escher, Angew. Chem. Int. Ed., 2006, 45(29), 4732-4762; h)
J. M. Fraile, J. I. Garcia, J. A. Mayoral, Chem. Rev., 2009, 109(2), 360–
417.
[10] a) H. M. L. Davies, A. M. Walji, Org. Lett. 2003, 5 (4), 479-482; b) H. M.
L. Davies, A. M. Walji, Org. Lett. 2005, 7 (14), 2941-2944; c) H. M. L.
Davies, A. M. Walji, T. Nagashima, J. Am. Chem. Soc. 2004, 126, 4271-
4280.
[11] a) K. Takeda, T. Oohara, M. Anada, H. Nambu, S. Hashimoto, Angew.
Chem. Int. Ed. 2010, 49 (39), 6979-6983; b) K. Takeda, T. Oohara, N.
Shimada, H. Nambu, S. Hashimoto, Chem. Eur. J. 2011, 17 (50), 13992-
13998.
[12] a) K. M. Chepiga, Y. Feng, N. A. Brunelli, C. W. Jones, H. M. L. Davies,
Org. Lett. 2013, 15 (24), 6136-6139; b) E. G. Moschetta, S. Negretti, K.
M. Chepiga, N. A. Brunelli, Y. Lebreche, Y. Feng, F. Rezaei, R. P. Lively,
W. J. Koros, H. M. L. Davies, C. W. Jones, Angew. Chem. Int. Ed. 2015,
54 (22), 6470-6474.
[13] a) K. Liao, S. Negretti, D. G. Musaev, J. Bacsa, H. M. L. Davies, Nature
2016, 533 (7602), 230-234; b) K. Liao, T. C. Pickel, V. Boyarskikh, J.
Bacsa, D. G. Musaev, H. M. L. Davies, Nature 2017, 551 (7682), 609-
613; c) K. Liao, Y. F. Yang, Y. Li, J. N. Sanders, K. N. Houk, D. G. Musaev,
H. M. L. Davies, Nat. Chem. 2018, 10 (10), 1048-1055; d) K. Liao, W. Liu,
Z. L. Niemeyer, Z. Ren, J. Bacsa, D. G. Musaev, M. S. Sigman, H. M. L.
Davies, ACS Catal. 2017, 8 (1), 678-682; e) W. Liu, Z. Ren, A. T. Bosse,
K. Liao, E. L. Goldstein, J. Bacsa, D. G. Musaev, B. M. Stoltz, H. M. L.
Davies, J. Am. Chem. Soc. 2018, 140 (38), 12247-12255.
[14] a) C. Qin, V. Boyarskikh, J. H. Hansen, K. I. Hardcastle, D. G. Musaev,
H. M. L. Davies, J. Am. Chem. Soc. 2011, 133 (47), 19198-19204; b) C.
Qin, H. M. L. Davies, J. Am. Chem. Soc. 2014, 136 (27), 9792-6.
[15] D. Zhao, J. Sun, Q. Li, G. D. Stucky, Chem. Mater. 2000, 12, 275-279.
[16] B. Wei, J. C. Sharland, P. Lin, S. M. Wilkerson-Hill, F. A. Fullilove, S.
McKinnon, D. G. Blackmond, H. M. L. Davies, ACS. Catal. 2019, 10 (2),
1161-1170.
[17] a) K. Tanabe, W. F. Hoelderich, Appl. Catal. A-Gen., 1999, 181, 399-
434; b) T. B. Lin, D. L. Chung, J. R. Chang, Ind. Eng. Chem. Res., 1999,
38(4), 1271–1276; c) A. Sakthivel, S. K. Badamali, P. Selvam, Micropor.
Mesopor. Mat., 2000, 39, 457-463; d) B. Li, R. Yan, L. Wang, Y. Diao, Z.
Li, S. Zhang, Ind. Eng. Chem. Res., 2014, 53(4), 1386–1394; e) Q. N.
Wang, L. Shi, A. H. Lu, ChemCatChem, 2015, 7, 2846 –2852; f) Z. Li, A.
W. Peters, V. Bernales, M.l A. Ortuño, N. M. Schweitzer, M. R.
DeStefano, L. C. Gallington, A. E. Platero-Prats, K. W. Chapman, C. J.
Cramer, L. Gagliardi, J. T. Hupp, O. K. Farha, ACS Cent. Sci., 2017, 3,
31−38; g) J. M. Carceller, M. Mifsud, M. J. Climent, S. Iborra, A. Corma,
Green Chem., 2020, 22, 2767-2777.
[18] The TOF was determined with a 0.13 molar solution of the diazo
compound with 0.05 mol.% catalyst and 5 equiv. of styrene. Moisture and
excess styrene are known to retard the catalyst, and even higher TOF
can be obtained with Rh2(S-o-ClTPCP)4 under more stringent dry
conditions using only 2.3 equiv. of styrene. See ref 12 for details.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202005381
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
O
Rh
Rh
O
Si
O
O
N
N
B
O
O
N
O
Cl
Cl
CF3
OH
3
CF3
Supported Rh2(S-2-Cl-5-CF3TPCP)4 catalysts ancored to mesoporous silica in three different locations demonstrates optimal
tethering location, allowing for deployment in a fixed bed flow reactor for enanti- and regioselective C-H functionalization. Tuning
silica particle size/shape enhances reactivity, with optimized catalyst enabling multiple catalysts recycles with comparable yields and
selectivities to batch reactions. The catalyst immobilization methodology is widely applicable to a family of dirhodium carbene
catalysts offering varying C-Hregioselectivity.
Institute and/or researcher Twitter usernames: @NSFCCHF; #NSFFunded
8
This article is protected by copyright. All rights reserved.