Chelation-directed C-H activation/C-C bond forming reactions catalyzed by Pd(II) nanoparticles supported on multiwalled carbon nanotubes

Article abstract of DOI:10.1039/c7cc02122d

Chelation-directed C-H activation/C-C bond forming reactions utilizing homogeneous palladium(II) and the Pd(II)/Pd(iv) catalytic cycle have been previously reported. Here we report the first use of a solid-supported Pd(II) catalyst [Pd(II) nanoparticles on multiwalled carbon nanotubes, Pd(II)/MWCNT] to carry out C-H activation/C-C bond forming reactions. The results presented demonstrate that the solid-supported Pd(II)/MWCNT catalyst can effectively catalyze these arylation reactions using the Pd(II)/Pd(iv) catalytic cycle. We also show that the solid-supported catalyst is recyclable, has turnover frequencies up to 2.9-fold higher than the homogeneous catalyst, and results in low levels of residual palladium contamination in the products.

Full text of DOI:10.1039/c7cc02122d

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Chelation-directed C–H activation/C–C bond
forming reactions catalyzed by Pd(II) nanoparticles
supported on multiwalled carbon nanotubes†
Cite this: Chem. Commun., 2017,
53, 7022
Received 20th March 2017,
Accepted 5th June 2017
Sudha Korwar,a Michael Burkholder,b Stanley E. Gilliland III,b Kendra Brinkley,b
acd
B. Frank Guptonb and Keith C. Ellis
*
DOI: 10.1039/c7cc02122d
rsc.li/chemcomm
Chelation-directed C–H activation/C–C bond forming reactions
utilizing homogeneous palladium(II) and the Pd(II)/Pd(IV) catalytic
cycle have been previously reported. Here we report the first use of
a solid-supported Pd(II) catalyst [Pd(II) nanoparticles on multiwalled
carbon nanotubes, Pd(II)/MWCNT] to carry out C–H activation/C–C
bond forming reactions. The results presented demonstrate that
the solid-supported Pd(II)/MWCNT catalyst can effectively catalyze
these arylation reactions using the Pd(II)/Pd(IV) catalytic cycle.
We also show that the solid-supported catalyst is recyclable, has
turnover frequencies up to 2.9-fold higher than the homogeneous
catalyst, and results in low levels of residual palladium contamina-
tion in the products.
Fig. 1 Chelation-directed C–H activation/C–C bond forming reactions
catalyzed by Pd(II).
Development of methodologies for the formation of carbon– been reported by Sanford2 (Fig. 1) and others.3 In contrast to
carbon bonds has been an active area of investigation in the traditional cross-coupling reactions that utilize Pd(0)/Pd(II)
organic chemistry for decades. Many of the latest innovations cycle, these C–H activation reactions utilize Pd(II)/Pd(IV) catalytic
in carbon–carbon bond forming reactions currently come from cycle (Scheme 1).2b
the C–H activation field, which seeks to selectively and directly
While there are examples of chelation-directed C–H activation/
functionalize unreactive C–H bonds. C–H activation/C–C bond C–C bond forming reactions using homogeneous Pd(II) sources, to
forming reactions are highly sought for their atom economy, our knowledge, there are no examples of this reaction that utilize
synthetic utility, and cost savings because they reduce the need a solid-supported Pd(II) catalyst. To date, there are only a few
for multi-step pre-functionalization of substrates.1
examples of the use of solid-supported catalysts in direct arylation
One methodology for selective C–H activation/C–C bond or C–H activation/C–C bond forming reactions of any type.4
forming reactions is the use of palladium(II) catalysts in combi- Solid-supported catalysts have several advantages including:
nation with intramolecular directing groups such as basic, (1) ease of catalyst recovery, (2) reducing or eliminating palladium
heteroaromatic nitrogens or sp2 hybridized oxygens in amides. contamination in the products following the reaction, and (3) the
Examples of selective, oxidative chelation-directed C–H activa- ability to recycle the catalyst and reuse it multiple times.5
tion/C–C bond forming reactions catalyzed by Pd(OAc)2 have
We have previously reported the use of a solid-supported
Pd(II)/MWCNT catalyst in N-chelation directed C–H activation
reactions, including alkoxylation (C–OAc, C–OMe) and halo-
genation (C–Cl, C–Br) reactions.6 Our previous findings demon-
strated that these reactions occur with higher rates compared
to the homogenous Pd(II) catalyst and that our Pd(II)/MWCNT
catalyst also had the advantages of recyclability and negligible
Pd contamination in the products.6
a Department of Medicinal Chemistry, School of Pharmacy, Virginia Commonwealth
University, Richmond, Virginia 23298-0540, USA. E-mail: kcellis@vcu.edu
b Department of Chemical and Life Sciences Engineering, School of Engineering,
Virginia Commonwealth University, Richmond, Virginia 23284, USA
c The Institute for Structural Biology, Drug Discovery, and Development,
Virginia Commonwealth University, Richmond, Virginia 23219, USA
d Massey Cancer Center, Virginia Commonwealth University, Richmond,
Virginia 23298-0037, USA
Encouraged by our previous results, the goal of this current work
is to test the hypothesis that the solid-supported Pd(II)/MWCNT
nanoparticle catalyst can be used to catalyze chelation-directed
† Electronic supplementary information (ESI) available: Experimental details and
spectal data. See DOI: 10.1039/c7cc02122d
7022 | Chem. Commun., 2017, 53, 7022--7025
This journal is ©The Royal Society of Chemistry 2017

View Article Online
Communication
ChemComm
Table 1 Exploration of substrates
Entry Substrate
Product
Pd(II)/MWCNT Pd(OAc)2
1
90%, 3 h
88%, 3 h
Scheme 1 Pd(II)/Pd(IV) catalytic cycle for chelation-directed C–H activation/
C–C bond forming reactions.
2
80%, 24 h
80%, 24 h
C–H activation/C–C bond forming reactions that undergo the
Pd(II)/Pd(IV) catalytic cycle. Here we report our initial study on
these C–H arylation reactions directed by chelating groups.
To begin our work, we explored the reaction of known
substrates for chelation-directed C–H activation/C–C bond form-
ing reactions with the symmetrical [Ph2I]BF4 arylation reagent
under previously reported conditions. As the catalyst, we used
solid-supported Pd(II)/MWCNT nanoparticles that we have pre-
viously reported.6 The results for these reactions are shown in
Table 1, along with the results using homogenous Pd(II) catalyst
(Pd(OAc)2).2a Treatment of 2-phenyl-3-methylpyridine (1) with
Pd(II)/MWCNT (5 mol%) and [Ph2I]BF4 (1.2 eq.) in acetic acid
at 100 1C for 3 hours (Table 1, entry 1) afforded the desired
2-([1,10-biphenyl]-2-yl)-3-methylpyridine (1a) in 90% yield, com-
parable to the yield of 88% with the homogeneous catalyst.
Treatment of 1-(3-(pyridin-2-yl)phenyl)ethanone (2) with Pd(II)/
MWCNT and [Ph2I]BF4 at 100 1C for 24 hours (Table 1, entry 2)
3
4
27%, 12 h
32%, 24 h
19%, 12 h
49%, 12 ha
75%, 24 ha
b
5
—
a
Previously reported results with homogeneous Pd(OAc)2; see ref. 2a.
Yield with homogeneous Pd catalyst has not been reported.
b
afforded the desired 1-(2-(pyridin-2-yl)-[1,10-biphenyl]-4-yl)ethanone in 19% yield. Benzo[h]quinoline (5) is sterically quite large as a
(2a) in 80% yield, again comparable to the yield of 80% with the substrate for this arylation reaction and chelation-directed C–H
homogeneous catalyst. Notably, only the single isomeric product 2a activation/C–C bond forming reactions with [Ph2I]BF4 have not
was formed in this reaction, suggesting that the regioselectivity of been previously reported on this compound. We therefore
the C–H activation step is dictated by steric factors rather than by expected not to form any of the product 5a in this reaction.
dual chelation effects of the pyridinyl nitrogen and the ketone to
We next wished to explore the scope of aryl rings that could
the Pd(II) metal center. Treatment of 1-acetyl-indoline (3) with Pd(II)/ be transferred in the C–H activation/C–C bond forming reaction
MWCNT and [Ph2I]BF4 at 100 1C for 12 hours (Table 1, entry 3) using the solid-supported Pd(II)/MWCNT catalyst. In order to
afforded the desired 1-acetyl-7-phenyl-indoline (3a) in only 27% selectively transfer a specific aryl ring, we used unsymmetrical
yield, far lower than the previously reported yield of 49% with the arylating reagents [Mes-I-Ar], consisting of a mesitylene group
homogeneous catalyst. Similarly, treatment of 1-phenylpyrrolidin-2- (which cannot be transferred due to its large steric bulk) and
one (4) with Pd(II)/MWCNT and [Ph2I]BF4 at 100 1C for 24 hours a substituted aryl group (which is transferred selectively to
(Table 1, entry 4) afforded the desired 1-([1,10-biphenyl]-2- the substrate). For this scope study, we utilized 2-phenyl-3-
yl)pyrrolidin-2-one (4a) in only 32% yield, lower than the methylpyridine (1) as the substrate. Aryl rings with both elec-
previously reported yield of 75% with the homogeneous tronically neutral (4-Me, 1b) and electronically withdrawing
catalyst. Together, entries 3 and 4 demonstrate that while the groups (4-F, 1c; 4-NO2, 1f; and 4-CHO, 1i) in the para position
Pd(II)/MWCNT catalyst can produce product in these reactions, were effectively transferred in the reaction (Table 2, entries 1, 2,
substrates with amide-type chelation groups are not as effective and 5) with yields of 87% (1b), 79% (1c), 61% (1f), and 77% (1i),
substrates with the solid-supported catalyst as they are with homo- respectively. However, the electronically rich 4-methoxy-phenyl
geneous Pd(II). This could be because the lone pair of electrons on group (Table 2, entry 3) did not react very well, affording a yield
the amide oxygen are less basic, making Pd(II) chelation the rate of only 33% of 1d. This is in contrast to the reported yield of
limiting step. We concluded our exploration of scaffolds with 81% for 1d using a homogenous Pd(II) catalyst.2a Increasing
the treatment of benzo[h]quinoline (5) with Pd(II)/MWCNT and the reaction time did not improve the yield. This effect was not
[Ph2I]BF4 at 100 1C for 12 hours (Table 1, entry 5), which to our seen with the 4-ethoxy-phenyl group (Table 2, entry 4), which
surprise afforded the desired 10-phenylbenzo[h]quinoline (5a) afforded a yield of 79% of 1e. We also explored the steric
This journal is ©The Royal Society of Chemistry 2017
Chem. Commun., 2017, 53, 7022--7025 | 7023

View Article Online
ChemComm
Communication
Table 2 Scope of arylation reagents with 2-phenyl-3-methylpyridine
weakly electron withdrawing ortho–CHO-phenyl group was not
effectively transferred and gave a poor yield of 18%, compared to the
77% yield with the para-CHO-phenyl group (Table 2, entry 8 vs. 9).
The difference in the reactivities of ortho-nitro (Table 2, entry 7) and
ortho-CHO (Table 2, entry 9) could be explained by the difference in
electronic nature of the substituents. The strong electron withdrawing
effect of the nitro group makes the [Mes-I-o-NO2-Ph] reagent highly
reactive towards oxidative addition to the Pd(II)-substrate complex to
afford the Pd(IV) reaction intermediate (Scheme 1). Though it is in the
ortho position and will cause some steric hindrance, the strong
electronics might be dictating the overall course of the reaction,
which explains the higher yield. The ortho-CHO, being sterically bulky
and only weakly electron withdrawing, is less reactive for the oxidative
addition and therefore affords a lower yield.
We also investigated whether the rigid and bulky benzo[h]-
quinoline substrate 5 could undergo C–H activation/C–C
bond forming reactions using the solid-supported Pd(II)/MWCNT
catalyst and unsymmetrical [Mes-I-Ar] arylating reagents.
Though the yield of the 10-phenylbenzo[h]quinoline (5a) is low
(19%; Table 3, entry 1), this is an attractive transformation due to
its simplicity and use of lower cost palladium catalyst. Synthesis
of products such as 5a–c (Table 3) have been reported, but
typically requires more expensive catalysts such as rhodium,
ruthenium, or iridium.7 Treatment of benzo[h]quinoline (5) with
Pd(II)/MWCNT and [Mes-I-p-Me-Ph]BF4 at 100 1C for 12 hours
resulted in arylation with the electronically neutral p-Me-Ph
group to afford the desired 10-(p-tolyl)benzo[h]quinoline (5b)
in 15% yield (Table 3, entry 2), similar to the results using the
symmetrical [Ph2I]BF4 arylation reagent to afford 10-phenyl-
benzo[h]quinoline (5a) (Table 3, entry 1). Similar treatment with
[Mes-I-p-F-Ph]BF4 resulted in arylation with the electronically
withdrawing p-F-Ph group to afford the desired 10-(4-florophenyl)-
benzo[h]quinoline (5c) in 35% yield (Table 3, entry 3). The
increase in yield for this reaction could be due to the electron
withdrawing nature of 4-F-phenyl group, making the [Mes-I-p-F-Ph]
Entry
1
Arylation reagent
Product
Yield (%)
87
[Mes-I-p-Me-Ph]BF4
2
3
4
5
[Mes-I-p-F-Ph]BF4
79
33
79
61
[Mes-I-p-OMe-Ph]BF4
[Mes-I-p-OEt-Ph]BF4
[Mes-I-p-NO2-Ph]BF4
6
7
8
9
[Mes-I-m-NO2-Ph]BF4
[Mes-I-o-NO2-Ph]BF4
[Mes-I-p-CHO-Ph]BF4
[Mes-I-o-CHO-Ph]BF4
69
58
77
18
Table 3 Scope of arylation reagents with benzo[h]quinoline
Entry
1
Arylating agent
[Ph2I]BF4
Product
Yield (%)
19
2
3
[Mes-I-p-Me-Ph]BF4
[Mes-I-p-F-Ph]BF4
15
35
tolerance of the reaction for transfer of aryl groups with
substitutions at the para, meta, or ortho positions. With the
strongly electron withdrawing nitro-phenyl group, para- (1f), meta-
(1g), and ortho-substituted (1h) phenyl rings were all effectively
transferred (Table 2, entries 5–7), with even the sterically hindered
ortho-NO2-phenyl group affording a 58% yield. However, the
7024 | Chem. Commun., 2017, 53, 7022--7025
This journal is ©The Royal Society of Chemistry 2017

View Article Online
Communication
ChemComm
Table 4 Turnover frequencies for C–H activation/C–C bond forming
Table 5 Recycling of the Pd(II)/MWCNT catalyst
reactions
Turnover frequency (hꢀ1
)
Entry Product
Pd(II)/MWCNT
Pd(OAc)2
Fold increase
2.9
Run
% Conversiona
1
16.48
5.74
Initial reaction
Recycle 1
Recycle 2
100
100
100
94
Recycle 3
Recycle 4
7
a
% Conversion measured by GC-MS.
2
3
1.62
3.63
0.64
1.41
2.5
2.6
reaction medium. Combined with the ease of removing the catalyst
from the reaction mixtures, this low level of palladium in the reaction
mixture is an improvement on the existing homogeneous catalyst for
this type of C–H activation reactions.
To demonstrate that the trace palladium in the reaction
mixture is not the source of catalytic activity, a hot filtration
experiment was performed with 2-phenyl-3-methylpyridine (1).
After 12 h at 100 1C, the Pd(II)/MWCNT catalyst was removed by
hot filtration over Celite. Fresh substrate and oxidant were
added to the filtrate, which was reheated to 100 1C. No further
conversion to product or catalytic activity was observed in
the filtrate in the absence of Pd(II)/MWCNT, showing that the
o40 ppm of residual Pd that remains in solution is not adequate
to catalyze the C–H activation/C–C bond forming reaction.
In conclusion, we have demonstrated that solid-supported Pd(II)/
MWCNT can catalyze C–H activation/C–C bond forming reactions of
heteroatom-chelating substrate with both symmetrical ([Ph2I]BF4) and
unsymmetrical ([Mes-I-Ar]) arylating reagents. The catalyst can be
recovered and recycled up to three times in these reactions and
turnover frequencies are up to 2.9-fold higher with the solid-supported
catalyst. The solid-supported Pd(II)/MWCNT also offers the advan-
tages of ease of removal by filtration and low levels of residual
palladium metal contamination (o40 ppm) in the products.
We acknowledge financial support for this work by the
National Science Foundation (CeRCaS NSF I/UCRC), the Virginia
Center for Innovation Technology (MF16-013-LS), and the Virginia
Commonwealth University Quest for Innovation fund.
reagent more reactive towards oxidative addition to the Pd(II)-
substrate complex to afford the Pd(IV) reaction intermediate
(Scheme 1). Attempts to carry out the C–H activation/C–C bond
forming reaction on benzo[h]quinoline with electronically
donating [Mes-I-Ar] arylating reagents failed to demonstrate
any conversion to the desired products.
We observed in our previous work that C–H activation reactions
catalyzed by the solid-supported Pd(II)/MWCNT catalyst have faster
reaction kinetics and higher turnover frequencies (TOFs) compared
to the homogeneous catalyst. To quantify this observation for the
C–H activation/C–C bond forming reactions, we calculated turnover
frequencies for several substrates (Table 4) with both the solid-
supported Pd(II)/MWCNT and homogeneous palladium catalysts.
Turnover frequencies were calculated as the moles of product per
moles of Pd(II) per hour. For these C–H activation/C–C bond forming
reactions, turnover frequencies with the solid-supported catalyst
were 2.5- to 2.9-fold higher than for the homogeneous catalyst. This
increase in the turnover frequencies, while modest, is an advantage
of using the Pd(II)/MWCNT catalyst for these types of C–H activation
or direct arylation type reactions.
To demonstrate the ability of the Pd(II)/MWCNT catalyst to be
recycled, we ran the C–H activation/C–C bond forming reaction on
2-phenyl-3-methylpyridine (1), recovered the catalyst by centrifuga-
Notes and references
tion, and iteratively repeated the reaction with the same batch of 1 X. Chen, K. M. Engle, D.-H. Wang and J.-Q. Yu, Angew. Chem., Int. Ed.,
2009, 48, 5094–5115.
2 (a) D. Kalyani, N. R. Deprez, L. V. Desai and M. S. Sanford, J. Am.
Chem. Soc., 2005, 127, 7330–7331; (b) N. R. Deprez and M. S. Sanford,
catalyst. We were able to recycle the catalyst 3 times with minimal
reduction in yield and no catalyst deactivation (Table 5, details in the
ESI†). In the fourth recycle reaction, the percent conversion dropped
below 10% demonstrating that the catalyst was deactivated.
J. Am. Chem. Soc., 2009, 131, 11234–11241.
3 R. Khan, R. Felix, P. D. Kemmitt, S. J. Coles, I. J. Day, G. J. Tizzard and
J. Spencer, Adv. Synth. Catal., 2016, 358, 98–109.
To determine whether the palladium metal from the solid sup-
ported Pd(II)/MWCNT catalyst leached into the C–H activation/
C–C bond forming reactions and contaminated the products, we
removed the catalyst by filtration over Celite from the arylation
reaction of 2-phenyl-3-methylpyridine (1) and measured the palladium
content in solution by ICP-MS. The palladium content of the reaction
mixture was found to be 38.4 ppm, demonstrating that very little
(5.6%) of the Pd metal on the solid supported surface leached into the
4 (a) A. J. Reay and I. J. S. Fairlamb, Chem. Commun., 2015, 51,
16289–16307; (b) R. Cano, A. F. Schmidt and G. P. McGlacken, Chem.
Sci., 2015, 6, 5338–5346.
5 L. Djakovitch and F.-X. Felpin, ChemCatChem, 2014, 6, 2175–2187.
6 S. Korwar, K. Brinkley, A. R. Siamaki, B. F. Gupton and K. C. Ellis, Org.
Lett., 2015, 17, 1782–1785.
7 (a) P. Gao, W. Guo, J. Xue, Y. Zhao, Y. Yuan, Y. Xia and Z. Shi, J. Am.
Chem. Soc., 2015, 137, 12231–12240; (b) X. Yu, J. Wang, W. Guo,
Y. Tian and J. Wang, Organometallics, 2016, 35, 1876–1884;
(c) S. Gonell and E. Peris, ACS Catal., 2014, 4, 2811–2817.
This journal is ©The Royal Society of Chemistry 2017
Chem. Commun., 2017, 53, 7022--7025 | 7025