Tandem one-pot CO2 reduction by PMHS and silyloxycarbonylation of aryl/vinyl halides to access carboxylic acids

Article abstract of DOI:10.1039/c8cc06820h

The present study discloses the synthesis of aryl/vinyl carboxylic acids from Csp²-bound halides (Cl, Br, I) in a carbonylative path by using silyl formate (from CO₂ and hydrosilane) as an instant CO-surrogate. Hydrosilane provides hydride for reduction and its oxidation product silanol serves as a coupling partner. Mono-, di-, and tri-carboxylic acids were obtained from the corresponding aryl/vinyl halides.

Full text of DOI:10.1039/c8cc06820h

ChemComm
View Article Online
View Journal
COMMUNICATION
Tandem one-pot CO2 reduction by PMHS and
silyloxycarbonylation of aryl/vinyl halides to
access carboxylic acids†
Cite this:DOI: 10.1039/c8cc06820h
Received 22nd August 2018,
Accepted 10th September 2018
Kumaraswamy Paridala, ‡ Sheng-Mei Lu,‡ Meng-Meng Wang and Can Li
*
DOI: 10.1039/c8cc06820h
rsc.li/chemcomm
The present study discloses the synthesis of aryl/vinyl carboxylic acids
from Csp2-bound halides (Cl, Br, I) in a carbonylative path by using silyl
formate (from CO2 and hydrosilane) as an instant CO-surrogate.
Hydrosilane provides hydride for reduction and its oxidation product
silanol serves as a coupling partner. Mono-, di-, and tri-carboxylic acids
were obtained from the corresponding aryl/vinyl halides.
Fig. 1 Occurrence of carboxylic acids in polymers.
Aryl carboxylic acids are an important class of raw materials for formed from CO2 and hydrosilane has been explored as a CO
the manufacturing of pharmaceuticals, fertilizers, polymers surrogate for formylation of alkenes or aryl halides to form
and nutritional products.1 For example, p-phthalic acid and aldehydes (Scheme 1a and b).10 As a formate ester, we anticipated
2,6-naphthalenedicarboxylic acid are raw materials for PET its applicability in C–C bond formation through alkoxy carbonylation
and PEN respectively and BTC (benzene tricarboxylic acid) is type reactions like other formate esters,11 and we are interested
the organic constituent of useful MOFs (Fig. 1).2 Making aryl in developing a strategy to use silyl formate in carbonylation
carboxylic acids through integration of CO2 (carbon dioxide) is reactions for getting products other than aldehydes/alcohols.
an ideal process. On that account, extensive carboxylation Here, we present our preliminary results about getting carboxylic
methods were developed.3,4
Apart from direct use of CO2 to install a carboxyl group,4 silyloxycarbonylation reaction of aryl halides (Scheme 1c).
treating CO2 as a CO-surrogate for carbonylation reactions has In the reported process of formylating aryl halides or
acids through a Cu(II)–Pd(II) catalysed tandem CO2 reduction-
gained much attention in recent years5–7 and in certain cases, alkenes, the intermediate silyl formate was supposed to form
carboxylic acids or esters were obtained. For example, in 2013, CO and silanol by decomposition. The available hydride source
Leitner and co-workers reported hydrocarboxylation of alkenes (H2 or Si–H) in the reaction system led to the hydrogenolysis of
with CO2 and H2 to get carboxylic acids.6 In 2014, Beller et al. the acyl complex [R(CQO)ML], providing aldehyde as the final
used alcohols as the reductant to reduce CO2 forming CO in situ product and silanol as the byproduct.10b We assumed that the
and then applied it in the following alkoxycarbonylation reac- silanol could act as a nucleophilic coupling partner and attack
tion giving esters as the product.7 The key point in these the acyl complex giving a silyl ester as the product.
reactions of using CO2 as a CO surrogate is to reduce CO2 to
Among the available hydrosilanes, PMHS (polymethylhydro-
CO with a suitable reductant. Hydrosilanes, as non-toxic and siloxane) is the cheapest (1–5 $ per kilogram) hydride donor
moisture stable leftover products of the silicone industry, were and has low effective mass per hydride (60).8b Silyl formate was
proved as cheap hydride sources for hydride transfer reductions.8 prepared from CO2 (balloon) and R3SiH catalysed by Cu(OAc)2/
With hydrosilane, CO2 can be transformed into silyl formate under L1(1,2-bis(diphenylphosphanyl)-benzene (BDP)) following
a
mild conditions which can be hydrolysed to formic acid or further modified procedure (P1) from Baba et al.,9a which was promptly
reduced to methanol or methane.9 Recently, silyl formate in situ utilized as a CO surrogate to examine the conditions for
silyloxycarbonylation (P2). Using 2-bromo naphthalene 1a as
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics,
the substrate, we added Pd(OAc)2/xanthphos (L2)/Et3N into the
silyl formate solution and progressed the reaction at 100 1C in a
closed vessel. After normal workup, 2-naphthalene carboxylic
acid 2a was obtained in 60% yield along with some unidentified
mixtures (Table 1, entry 1), supporting that, in the absence of
Chinese Academy of Sciences, Zhongshan Road 457, Dalian 116023, China.
E-mail: canli@dicp.ac.cn
† Electronic supplementary information (ESI) available: Copies of 1H and 13C NMR
spectra. See DOI: 10.1039/c8cc06820h
‡ P. K. and S.-M. Lu contributed equally.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
Table 2 Silyloxycarbonylation of bromoarylsa
Scheme 1 Carbonylations with CO2 as a CO surrogate.
Entry
1, A =
Time (h)
Product 2 yieldb (%)
1
2
3
4
1b, H
5
1.5
4
2
7
0.5
7
3
10
7
4
2b, 72
2c, 94
2d, 72
2e, 97
2f, 96
2g, 77
2h, 90
2i, 96
2j, 77
2k, 45
2l, 84
hydride, silanol can serve as the nucleophile to yield silyl ester.
To improve the yield of 2a, we screened solvents for silyloxy-
carbonylation reaction (Table 1 and Table S1, ESI†), and found
that toluene is a better solvent. With a 1 : 2 ratio of dioxane–
toluene, the yield was improved a little with incomplete con-
version (Table 1, entry 2). With a 1 : 4 ratio of dioxane–decane,
both conversion and yield were increased (Table 1, entry 3).
When 1 : 4 dioxane–toluene was used, 98% isolated yield was
obtained (Table 1, entry 4). At this stage, a parallel study of
different silyl hydrides (Table 1, entries 4–6 and Table S2, ESI†)
revealed that PMHS and (EtO)3SiH are excellent reagents for the
silyloxycarbonylation reaction.
1c, 4-CO2CH3
1d, 4-COCH3
1e, 4-CHO
1f, 4-NO2
5
6c
7
1g, 4-Cl
1h, 4-OCH3
1i, 3,4-(methylenedioxy)
1j, 3-NO2
1k, 2-F
1l, 3-F-4-CH3
8
9
10
11
12
With PMHS and 1 : 4 dioxane–toluene, the relative perfor-
mance of various Pd(II)-catalysts was studied and endorsed that
Pd(OAc)2/L2 was the best for the reaction (Table S3, ESI†).
Decreasing the loading of L2 or Pd(OAc)2 resulted in a lower yield
of 2a (Table 1, entries 7 and 8). A comparative study affirmed that
Et3N was a preferable base as N,N-diisopropylethylamine (DIPEA)
showed lower yield (Table 1, entry 9). Finally, 3 mol% Pd(OAc)2
with 6 mol% L2 in a 1 : 4 ratio of dioxane–toluene was accepted as
the optimized conditions for the silyloxycarbonylation process
(Table 1, entry 4). Controlled experiments confirmed that all the
reagents, the catalyst, the solvent ratio and the temperature are
essential for the silyloxycarbonylation to produce carboxylic acid 2a
in higher yield (Table S4, ESI†).
a
b
Reaction conditions: P1, P2: see Table 1. Isolated yield based on 1.
90 1C.
c
With the optimized conditions in hand, the substrate scope
and generality of the reaction were explored. For the para-
substituted bromobenzenes, regardless of the electron-withdrawing
or -donating nature, the corresponding benzoic acids were obtained
in good to excellent yields (Table 2) via a silyloxycarbonylation
reaction starting from CO2 and PMHS. For example, simple benzoic
acid 2b was obtained in 72% yield from bromobenzene 1b (Table 2,
entry 1). Functional groups such as ester, ketone, aldehyde, nitro,
chloro and ether can all be tolerated in the reaction, giving yields in
the range of 72–97% (Table 2, entries 2–8). For substrate 1g with a Cl
group at the para-position, two potential reaction sites lead to a
mixture of two products, 4-chlorobenzoic acid 2g and terephthalic
acid. Therefore, after tuning the reaction time (30 min) we got 2g in
77% yield at 90 1C. Compared with 1f bearing a nitro group at the
4-position, 1j having a nitro group at the 3-position under the same
reaction conditions gave a lower yield of 77% (Table 2, entry 9). For
substrate 1k bearing fluoride at the ortho-position, the reaction
provided 2k in 45% yield because of the incomplete conversion
(Table 2, entry 10). The low conversion is due to the steric
hindrance of the fluoride. For 3-fluoro-4-methyl-bromobenzene
1l, the silyloxycarbonylation reaction still proceeded smoothly,
providing 2l in 84% yield (Table 2, entry 11). Heterocyclic bromides
1m–q were also good substrates for the reaction. Starting from CO2
and PMHS, corresponding carboxylic acids 2m–q were obtained in
good to excellent yields (Table 2, entry 12).
Table 1 Reaction condition optimizationa
Entry Dioxane/solvent (V/V) R3SiH
Convnb (%) Yieldb (2a %)
1
2
3
4
Dioxane (1 : 2)
Toluene (1 : 2)
nDecane (1 : 4)
Toluene (1 : 4)
Toluene (1 : 4)
Toluene (1 : 4)
Toluene (1 : 4)
Toluene (1 : 4)
Toluene (1 : 4)
PMHS
PMHS
PMHS
PMHS
Ph2SiH2
495
78
495
495
495
60
65
67
98c
82
5
6
(EtO)3SiH 495
93c
45
7d
8e
9 f
PMHS
PMHS
PMHS
495
495
495
90c
90c
From the reaction result of 1g, we can deduce that: (1) multi
carboxylic acids can be anticipated from aryl multi halides; (2)
chlorobenzenes can also be feasible substrates for the silyloxy-
a
Reaction conditions: P1: CO2 (balloon), Cu(OAc)2ꢀH2O (1 mol%), L1
(1.5 mol%), R3SiH (2.5 mmol); P2: 1a (1.0 mmol); Pd(OAc)2 (3 mol%), L2 carbonylation reaction. So, we expected that, by increasing the
(6 mol%), Et3N (2.5 mmol), 100 1C, 3 h. Calculated by 1H NMR, 495%
b
amount of PMHS and Pd(II)/L2, iterative silyloxycarbonylation
of dihalides could occur and yield dicarboxylated products.
After a few trials, we found that 5 mol% Pd catalyst and four
c
conversion means no substrate can be detected. Isolated yield.
d
e
f
Pd(OAc)2/L2 = 1/1, 6 h. Pd(OAc)2 (1 mol%), 5 h. DIEPA (2.5 mmol)
as the base, 4 h.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018

View Article Online
ChemComm
Communication
Table 3 Silyloxycarbonylation of aryl multi halidesa
Entry
1, A =
Time (h)
3, yieldb (%)
2, yieldc (%)
1
1g, 4-Cl
1r, 3-I
1s, 4-Br
1t, 4-Br-3-CH3
16
16
8
3a, 63
3b, 62
3a, 93
3c, 72
2g, 33
2r, 20
0
0
2d
3
4
20
5
a
Reaction conditions: P1: CO2 (balloon), Cu(OAc)2ꢀH2O (2 mol%), L1 (3 mol%); P2: 1 (1.0 mmol), Pd(OAc)2 (5 mol%), L2 (10 mol%), Et3N
(4.0 mmol), PMHS (4.0 mmol), 100 1C. Isolated yield based on 1. Calculated from 1H NMR. 2r = 3-bromo benzoic acid.
b
c
d
Table 4 Silyloxycarbonylation of chloro and iodo arylsa
equivalents of PMHS were effective enough for successful
dicarboxylation. So, under these conditions, dihalo benzene
1g produced a mixture of terephthalic acid 3a (63%) along with
2g (33%) (Table 3, entry 1). Similarly, 1r resulted in a mixture
of 3b (62%) and 3-Br-benzoic acid 2r (20%) (Table 3, entry 2)
(vide infra). Encouragingly, exclusive dicarboxylic acids 3a, 3c and
3d–f were obtained from the corresponding aryl di-bromides 1s–w
(Table 3, entries 3–5). Utilization of CO2 to make these dicarboxylic
acids was not reported in the literature, and for the first time we
made these dicarboxylic acids through CO2 incorporation using the
present method. To check the extent of iterative silyloxycarbonyla-
tion, tribromobenzene 1x was subjected to tricarboxylation with
an excess of PMHS and catalyst, and trimesic acid 4 (BTC) was
obtained in almost quantitative yield (97%, Scheme 2).
Enlightened from the dicarboxylation of 1g and 1r that
chloro and iodo aryls can also participate in the carboxylation
process, we next focused on the conversion of iodo and chloro
aryls. Under the same conditions of bromo aryls, iodoaryls 5a–d
yielded the carboxylic acids 2b, 2d, 2f, and 2h in comparable
yields with their bromo-congeners (Table 4, entries 1–4). Similarly,
meta-substituted iodoaryl 5e gave 2s in 76% yield (Table 4, entry 5).
In the case of 1r, having I and Br at the 1,3-position, the mono
Entry
5 or 6, X/A
Time (h)
2, yielda (%)
1
2
3
4
5
6
7
8
5a, I/H
5b, I/4-COCH3
5c, I/4-NO2
5d, I/4-OCH3
5e, I/3-COCH3
1r, I/3-Br
6a, Cl/4-CO2CH3
6b, Cl, 4-COCH3
6c, Cl/4-CHO
5
4
10
5
2
2
16
14
5
14
20
10
5
2b, 84
2d, 85
2f, 85
2h, 60
2s, 76b
2r, 52
2c, 83
2d, 45
2e, 93
2f, 60
2t, 20
2u, 45
2v, 45
9
10
11
12
13
6d, Cl/4-NO2
6e, Cl/3-CO2CH3
6f, 2,6-dichloropyridine
6g, 2-chloroquinoline
a
b
The same as that in Table 2. Treated with MeOH/H+ and isolated as
methyl ester.
silyloxycarbonylation produced 3-Br-benzoic acid 2r in 52% yield corresponding acids 2u and 2v respectively in 45% yield (Table 4,
indicating that the reaction of iodide is favored over bromide entries 12 and 13). Overall, in the present carboxylation process, the
(Table 4, entry 6). Only activated chloroaryls 6a–d with electron iodo congeners showed similar reactivity while the chloro congeners
withdrawing groups at the para position underwent the silylox- showed suppressed yields and took longer reaction times.
ycarbonylation reaction and produced the corresponding acids
We found that the present strategy also functioned on vinyl
2c–f (Table 4, entries 7–10) in good yields but with relatively long bromides. Alkenoic acids 8a–c were obtained in excellent yields
reaction times owing to their lower reactive profiles. But chloro aryl from the corresponding bromides 7a–c under the standard
6e produced 2t only in 20% yield (Table 4, entry 11). Also, conditions of mono carboxylation. The products maintained
2,6-dichoropryridne 6f and 2-chloroquinoline 6g yielded the retention of geometry from the corresponding bromoalkenes
(Scheme 3).
During the mechanism study, to rule out the role of formic
acid in the reaction,12 prior hydrolysis of silyl formate and
then carbonylation was performed and no product was formed
(see Table S5, ESI†). Here, under our conditions, CO gas was
detected by GC upon heating the silyl formate in the presence
of Et3N,13 which implied the possibility of CO involved
Scheme 2 Silyloxycarbonylation of 1,3,5-tribromobenzene.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
Conflicts of interest
There are no conflicts to declare.
Notes and references
¨
1 (a) F. Rohrscheid, Ullmann’s Encyclopedia of Industrial Chemistry,
2000; (b) Carboxylic Acids and Esters, ed. S. Patai, John Wiley & Sons,
Ltd., Chichester, UK, 1969.
Scheme 3 Silyloxycarbonylation of vinyl bromides.
2 PET, PEN, MOFs (a) E. Gubbels, T. Heitz, M. Yamamoto, V. Chilekar,
¨
S. Zarbakhsh, M. Gepraegs, H. Kopnick, M. Schmidt, W. Bru¨gging,
J. Ru¨ter and W. Kaminsky, Ullmann’s Encyclopedia of Industrial
Chemistry, Wiley, 2018; (b) S. L. Sakellarides, Encyclopedia Of Polymer
Science and Technology, Wiley, 2004; (c) T. M. Tovar, J. Zhao, W. T.
Nunn, H. F. Barton, G. W. Peterson, G. N. Parsons and M. D. LeVan,
J. Am. Chem. Soc., 2016, 138, 11449.
3 For recent reviews: (a) S.-S. Yan, Q. Fu, L.-L. Liao, G.-Q. Sun, J.-H. Ye,
L. Gong, Y.-Z. Bo-Xue and D.-G. Yu, Coordin. Chem. Rev., 2018,
374, 439; (b) Chemical Transformations of Carbon Dioxide, ed.
X.-F. Wu and M. Beller, Springer International Publishing,
2018; (c) M. Aresta, A. Dibenedetto and E. Quaranta, J. Catal.,
¨
2016, 343, 2; (d) M. Borjesson, T. Moragas, D. Gallego and
Scheme 4 Proposed mechanism of silyloxycarbonylation.
R. Martin, ACS Catal., 2016, 6, 6739.
4 For recent reports, see: (a) J. Song, Q. Liu, H. Liu and X. Jiang, Eur.
J. Org. Chem., 2018, 696; (b) C. S. Yeung and V. M. Dong, Top. Catal.,
2014, 57, 1342; (c) J. Hou, J.-S. Li and J. Wu, Asian J. Org. Chem., 2018,
7, 1439; (d) F. Tan and G. Yin, Chin. J. Chem., 2018, 36, 545;
(e) Y. Cao, X. He, N. Wang, H.-R. Li and L.-N. He, Chin. J. Chem.,
carbonylation. After releasing CO, silanol R3SiOH was left and it
was supposed to participate in the reaction to form the silyl
ester. To check the silyl ester, the reaction mixture before
hydrolyzing was analyzed by NMR (Fig. S1–S5, ESI†). The results
of 1H NMR showed the presence of (EtO)3SiOSi(EtO)2OH and a
little up field shift for Ar-protons of 2a (Fig. S1 and S2, ESI†).
In 13C NMR spectra (Fig. S3 and S4, ESI†), chemical shifts at
d 165.0 and d 171.3 ppm were ascribed to 2a0 (ArCO2Si(OEt)3)
and 2a respectively. Also, 29Si NMR showed two peaks at d
ꢁ85.3 and ꢁ85.5 ppm attributed to (EtO)3SiOSi(EtO)2OH. The
third peak at d ꢁ92.5 ppm implied the signal for 2a0 (Fig. S5,
ESI†). Based on these results, we proposed the mechanism
shown in Scheme 4. The formation of free acid could be
possible by dimerization of silanol while hydrolysing the silyl
ester. Usual water work-up hydrolyses the silyl ester and liberates
the carboxylic acid.
´
´
2018, 36, 644; ( f ) F. Julia-Hernandez, M. Gaydou, E. Serrano, M. van
Gemmeren and R. Martin, Top. Curr. Chem., 2016, 374, 45;
(g) I. Tommasi, Catalysts, 2017, 7, 380; (h) S. Wang, G. Du and
C. Xi, Org. Biomol. Chem., 2016, 14, 3666.
5 (a) M. Jensen, M. Rønne, A. Ravn, R. Juhl, D. Nielsen, X.-M. Hu,
S. Pedersen, K. Daasbjerg and T. Skrydstrup, Nat. Commun., 2017,
8, 489; (b) M. Ali, A. Gual, G. Ebeling and J. Dupont, ChemCatChem,
2014, 6, 2224; (c) X. Ren, Z. Zheng, L. Zhang, Z. Wang, C. Xia and
K. Ding, Angew. Chem., Int. Ed., 2017, 56, 310; (d) D.-G. Yu, Z. Zhang,
T. Ju and J.-H. Ye, Synlett, 2017, 741.
6 T. G. Ostapowicz, M. Schmitz, M. Krystof, J. Klankermayer and
W. Leitner, Angew. Chem., Int. Ed., 2013, 52, 12119.
7 L. Wu, Q. Liu, I. Fleischer, R. Jackstell and M. Beller, Nat. Commun.,
2014, 5, 3091.
8 (a) J. Pesti and G. L. Larson, Org. Process Res. Dev., 2016, 20, 1164;
(b) J. M. Lavis, R. E. Maleczka, V. Chandrasekaran and S. J. Collier,
Encycl. Reagents Org. Synth., 2016; (c) D. P. Xu, W. J. Xiao, J. J. Peng,
J. Y. Li and Y. Bai, Chin. J. Org. Chem., 2014, 34, 2195.
In conclusion, we have developed an efficient silyloxycarbonyla-
tion strategy to make carboxylic acids from CO2 utilizing PMHS as a
source of both non-metallic reductant and nucleophilic coupling
partner. The strategy works in one-pot tandem Cu(II)-catalysed
reduction–Pd(II)-mediated silyloxycarbonylation reaction involving
silyl formate as a CO-surrogate. The method is applicable to
produce mono-, di-, and tricarboxylic acids from iodo/bromo/
chloro aryls and vinyl bromides, in good to excellent yields. This
work provides a novel strategy of utilizing CO2 as a CO-surrogate
to get carboxylic acids. Further application of this strategy to other
carbonylative reactions is underway.
9 CO2 reduction by Si–H: (a) K. Motokura, D. Kashiwame, N. Takahashi,
A. Miyaji and T. Baba, Chem. – Eur. J., 2013, 19, 10030; (b) K. Motokura,
M. Naijo, S. Yamaguchi, A. Miyaji and T. Baba, Chem. Lett., 2015,
44, 1464; (c) D. S. Morris, C. Weetman, J. T. C. Wennmacher, M. Cokoja,
M. Drees, F. E. Ku¨hn and J. B. Love, Catal. Sci. Technol., 2017, 7, 2838;
(d) M. Rauch and G. Parkin, J. Am. Chem. Soc., 2017, 139, 18162;
(e) J. Chen, L. Falivene, L. Caporaso, L. Cavallo and E. Y. X. Chen,
J. Am. Chem. Soc., 2016, 138, 5321; ( f ) D. Mukherjee, D. F. Sauer,
A. Zanardi and J. Okuda, Chem. – Eur. J., 2016, 22, 7730.
10 (a) X. Ren, Z. Zheng, L. Zhang, Z. Wang, C. Xia and K. Ding, Angew.
Chem., Int. Ed., 2017, 56, 310; (b) B. Yu, Z. Yang, Y. Zhao, L. Hao,
H. Zhang, X. Gao, B. Han and Z. Liu, Chem. – Eur. J., 2016, 22, 1097.
11 Selected reviews: (a) H. Konishi and K. Manabe, Synlett, 2014, 1971;
(b) H. Konishi, Chem. Pharm. Bull., 2018, 66, 1.
12 In a separate experiment formic acid was used as a CO surrogate
and the carboxylation reaction was checked (see Table S6, ESI†).
13 Enhanced CO production was found compared to simple thermal
decomposition.
We thank the National Key R&D Program of China
(2017YFB0702800) and the National Natural Science Founda-
tion of China (21621063).
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018