New synthesis of biaryls via Rh-catalyzed decarbonylative Suzuki-coupling of carboxylic anhydrides with arylboroxines

Article abstract of DOI:10.1002/adsc.200404190

The catalytic cross-coupling of aromatic carboxylic anhydrides or acid chlorides with triarylboroxines has been achieved for the first time under decarbonylation, giving rise to the unsymmetrical biaryls rather than the expected diaryl ketones. This new decarbonylative Suzuki coupling, catalyzed by a [Rh(ethylene)₂Cl]₂/KF system, can be applied to aromatic, heteroaromatic and vinylic carboxylic anhydrides, potentially opening up new perspectives for the use of carboxylic acid derivatives as substrates in biaryl syntheses.

Full text of DOI:10.1002/adsc.200404190

COMMUNICATIONS
New Synthesis of Biaryls via Rh-Catalyzed Decarbonylative
Suzuki-Coupling of Carboxylic Anhydrides with Arylboroxines
L. J. Gooßen,a, b,* J. Paetzolda
a
Max-Planck Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr, Germany
Present address: Rheinisch-Westfälische Technische Hochschule Aachen, Institut für Organische Chemie,
b
Professor Pirlet Str. 1, 52074 Aachen, Germany
Fax: (þ49)-241-80-92385, e-mail: goossen@oc.rwth-aachen.de
Received: June 21, 2004; Accepted: October 22, 2004
SupportingInformation for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The catalytic cross-couplingof aromatic
carboxylic anhydrides or acid chlorides with triaryl-
boroxines has been achieved for the first time under
decarbonylation, giving rise to the unsymmetrical
biaryls rather than the expected diaryl ketones.
This new decarbonylative Suzuki coupling, catalyzed
by a [Rh(ethylene)2Cl]2/KF system, can be applied to
aromatic, heteroaromatic and vinylic carboxylic an-
hydrides, potentially openingup new perspectives
for the use of carboxylic acid derivatives as sub-
strates in biaryl syntheses.
extension of this powerful synthetic methodology to
other broadly available substrate classes such as carbox-
ylic acids would be of high interest.
It has recently been demonstrated that carboxylic acid
derivatives, e. g., carboxylic anhydrides or esters, can ox-
idatively add to transition metal catalysts, giving rise to
acyl-metal complexes.[7] At elevated temperatures,
these intermediates decarbonylate to yield the corre-
spondingalkyl- or aryl-metal species. This behavior
has been utilized in decarbonylative Heck olefina-
tions[8,9] and decarbonylative elimination reactions,[10]
in which carboxylic acid derivatives serve as synthetic
equivalents of aryl or alkyl halides. However, in catalytic
cross-couplingreactions, the decarbonylation step was
found to be less favorable, so that exclusively aryl ke-
tones are formed in the Suzuki couplingof carboxylic
anhydrides with boronic acids.[7,11] To the best of our
knowledge, only one special case of a cross-coupling re-
Keywords: biaryls; carboxylic acids; cross-coupling;
decarbonylation; rhodium
Substituted biphenyls are a key structural motif in nu- action has been reported that proceeds under decarbon-
merous biologically active compounds.[1] Due to the ylation, but it calls for stoichiometric quantities of nickel
mild reaction conditions, the high yields and the excel- salts to trap the carbon monoxide in the form of toxic
lent functional group tolerance, the Suzuki coupling of nickel carbonyl compounds.[12]
arylboronic acids with aryl halides is usually the method
Our target was to identify conditions to access biaryls
of choice for their synthesis.[2] Besides aryl halides,[3] aryl via a decarbonylative couplingof aromatic carboxylic
and alkyl sulfonates,[4] diazonium salts[5] and anilinium anhydrides with arylboronic acids, in the presence of
salts[6] can be used as the electrophilic couplingpartner.
only a catalytic amount of a transition metal complex.
Since these do not always satisfy the synthetic needs, an A plausible mechanism for this transformation would
Scheme 1. Cross-couplingreaction between 3-methylbenzoic anhydride and boronic acid derivatives.
Adv. Synth. Catal. 2004, 346, 1665–1668
DOI:10.1002/adsc.200404190
ꢀ 2004 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
1665

L. J. Gooßen, J. Paetzold
COMMUNICATIONS
Table 1. Decarbonylative Suzuki-couplingof carboxylic anhydrides.
Entry
Ph-M
Precatalyst
Ligand/Additive
Solvent
Yield of 6a [%]
Yield of 7a [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14[b]
15
16
17
18
19
20
21
22[c]
23
24
25
26[e]
2a
2a
2a
2a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
4a
5a
5a
Pd(PPh3)4
PdCl2
PdCl2
[Rh(ethylene)Cl]2
PdCl2
PdCl2
[Rh(ethylene)2Cl]2
RhCl3
(Rh(nbd)Cl)2
(Rh(COD)Cl)2
Rh(acac)(CO)2
[Rh(ethylene)2Cl]2
“
“
“
“
–
LiCl
DPE-Phos
–
LiCl
DPE-Phos
–
–
NMP
NMP
DMPU
mesitylene
NMP
DMPU
mesitylene
“
“
“
“
“
“
“
“
“
<2[a]
2[a]
5[a]
10[a]
2
5
40
0
20
20
40
25
65
65
40
55
40
40
40
60
10
30
<2
50[d]
10
65
20
2
15
5
5
10
10
0
5
5
–
–
–
25
25
20
20
5
15
10
10
25
10
50
25
<2
15
15
20
PPh3
KF
KF
LiF
NaF
Na2CO3
K3PO4
KF
KF
KF
KF
KF
–
“
“
“
“
“
“
“
“
diglyme
n-Bu2O
NMP
mesitylene
“
“
“
“
PdCl2
[Rh(ethylene)2Cl]2
“
“
–
–
Conditions: 1.00 mmol 3-methylbenzoic anhydride, 1.50 mmol 2a–5a, 0.03 mmol precatalyst, 0.10 mmol additive, 1608C, 16 h;
GC yields using n-hexadecane as internal standard.
[a]
Main product: benzoic acid.
1.00 mmol additive.
1408C.
Side product: 3,3’-dimethylbiphenyl (15%).
With 1.00 mmol 3-methylbenzoyl chloride (8a) instead of 1a.
[b]
[c]
[d]
[e]
involve the oxidative addition of the carboxylic anhy- was not unexpected, since boronic acids are known to
dride to a transition metal catalyst, followed by decar- liberate water at elevated temperatures to form cyclic
bonylation of the resultingacyl-metal species and trans-
boroxines. We attempted to work around this side reac-
metallation with the boronic acid derivative. Thus, the tion of the carboxylic anhydrides by directly usingpre-
key to high product selectivity would be to assist the de- formed triphenylboroxine (3a). Indeed, use of 3a led
carbonylation step and avoid premature transmetalla- to less hydrolysis and improved yields of the cross-cou-
tion. The biaryl product would finally be liberated by re- plingproducts (entries 5–23). In the presence of various
ductive elimination, regenerating the initial transition Pd catalysts, predominantly the non-decarbonylated
metal complex.
product 7a was formed even under conditions that had
We chose the reaction of 3-methylbenzoic anhydride been successfully used in decarbonylative Heck olefina-
1a with several benzeneboronic acid derivatives 2a–5a tions or decarbonylative eliminations of carboxylic an-
as our model system (Scheme 1) and screened various hydrides (entries 5 and 6).[8e,10] Only in the presence of
transition metal complexes under different conditions [Rh(ethylene)2Cl]2 and under conditions similar to
to elucidate the factors facilitatingthe formation of de-
carbonylated product 6a over the ketone 7a. Selected re- acid chlorides,[9] were significant amounts of the desired
sults are displayed in Table 1. biaryl 6a formed (entry 7). Encouraged by this finding,
those used for decarbonylative Heck olefinations of
Usingbenzeneboronic acid ( 2a) as the carbon nucleo- we set out to further optimize the catalytic system and
phile, hydrolysis of the carboxylic anhydride predomi- tested various rhodium complexes in combination with
nated under various conditions, and only small quanti- several additives (entries 7–18). Best results were ob-
ties of cross-couplingproducts were observed despite
usinganhydrous solvents (Table 1, entries 1–4). This
tained with [Rh(ethylene)2Cl]2, while similar complexes
with more strongly bound ligands showed significantly
1666
ꢀ 2004 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1665–1668

Decarbonylative Suzuki-Couplingof Carboxylic Anhydrides with Arylboroxines
COMMUNICATIONS
Table 2. Scope of the reaction accordingto Scheme 2.
ꢀ
Anhydrideþ Product 6 Ar Ar’
Yield of Yield of
Boroxine
6 [%][a] 7 [%]a
1aþ3a
6a 65 (64)
20
<5
<5
Scheme 2. Cross-couplingreaction of carboxylic anhydrides
and aryl boroxines.
1aþ3b
1aþ3c
6b 70 (66)
6c 65 (48)
lower activities (entries 7–12). The addition of catalytic
quantities of fluoride salts and in particular KFappeared
to improve the activity of the rhodium catalyst, leading
to significantly higher yields of the product 6a, whereas
the addition of other inorganic halides or bases show no
enhancement of the catalytic activity (entries 13–18).
At this point, we cannot provide an unambiguous ex-
planation for this interestingcounter ion effect.
Mesitylene proved to be the optimal solvent for this
reaction in terms of yield. The highest selectivity (6:1)
for the decarbonylated product was reached in n-Bu2O,
albeit with a slightly lower overall yield (entry 20).
More strongly coordinating solvents such as NMP appa-
rently retarded the decarbonylation step so that more
ketone 7a was formed (entry 21). Similar to other decar-
bonylative processes, elevated temperatures are essen-
tial (entry 22).
1aþ3d
1aþ3e
6d 65 (55)
6e 70 (61)
20
<5
1aþ3f
1aþ3g
1aþ3h
6f
65 (57)
<5
15
6g 70 (55)
6h 50 (37)
40
Under optimized conditions, satisfactory yields of
biaryl 6a were also obtained when employingthe cyclic
phenyl boronate 4a or sodium tetraphenylborate (5a)
(entries 24 and 25) as carbon nucleophiles. Furthermore,
3-methylbenzoyl chloride (8a) can also serve as the aryl
source instead of the carboxylic anhydride 1a, especially
in combination with sodium tetraphenylborate (5a) (en-
try 26).
Havingestablished an efficient reaction protocol for
the transformation depicted in Scheme 2, we next set
out to investigate its scope. Various functionalized aryl-
boroxines 3a–i were generated from commercially
available boronic acids 2a–i by heatingthem under re-
duced pressure, or alternatively by azeotropic distilla-
tion of a toluene solution of 2a–i. The boroxines were
successfully coupled with aromatic, heteroaromatic
and vinylic carboxylic acid anhydrides 1a–f, giving rise
to the correspondingbiaryls 6a–m in moderate to
good yields.[13] Selected results are shown in Table 2.
The selectivity of decarbonylated product 6 vs. ketone
7 was found to be higher for electron-poor than for elec-
tron-rich boroxines (e.g., entries 6b and 6 h). This can be
rationalized by the higher nucleophilicity of electron-
rich boroxines, favoringtransmetallation over the com-
petingdecarbonylation.
1bþ3a
1cþ3a
6i
6j
60 (55)
40 (35)
25
15
1dþ3a
1aþ3i
6k 30 (16)
<5
6l
60 (57)
25
1eþ3a
1fþ3a
6m 25 (21)
6n 25 (21)
15
5
Conditions: 1.00 mmol carboxylic anhydride, 0.50 mmol bor-
oxine, 0.015 mmol [Rh(ethylene)2Cl]2, 0.10 mmol KF, 4 mL
mesitylene, 1608C, 8 h.
[a]
GC yields, isolated yields in brackets. The residue derived
from the anhydride is displayed on the left hand side of the
structures.
terestingfrom a mechanistic standpoint, it moreover
In summary, for the first time, a reaction protocol has bears the potential of becomingthe basis for the use of
been identified for the cross-couplingof aromatic car- aromatic carboxylic acid derivatives as substrates for
boxylic acid derivatives with organometallic species biaryl synthesis. Improvement of the reaction protocol
that proceeds mainly under extrusion of carbon monox- towards in situ activation of carboxylic acids is currently
ide to give a biaryl product. The reaction is not only in- under investigation in our laboratory.
Adv. Synth. Catal. 2004, 346, 1665–1668
asc.wiley-vch.de
ꢀ 2004 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
1667

L. J. Gooßen, J. Paetzold
COMMUNICATIONS
4176–4211; d) S. Kotha, K. Lahiri, D. Kashinath, Tetra-
hedron 2002, 58, 9633–9695; e) A. Suzuki, in: Metal-Cat-
alyzed Cross-Coupling Reactions, (Eds.: F. Diederich,
P. J. Stang), VCH, Weinheim, 1998, pp. 49–97.
[4] a) Z. Y. Tang, Q. S. Hu, J. Am. Chem. Soc. 2004, 126,
3058–3059; b) V. Percec, G. M. Golding, J. Smidrkal, O.
Weichold, J. Org. Chem. 2004, 69, 3447–3452.
[5] a) S. Darses, T. Jeffery, J. P. Genet, J. L. Brayer, J. P. De-
moute, Tetrahedron Lett. 1996, 37, 3857–3860; b) S.
Darses, J. P. Genet, J. L. Brayer, J. P. Demoute, Tetrahe-
dron Lett. 1997, 38, 4393–4396; c) K. Kikukawa, K. Ike-
naga, F. Wada, T. Matsuda, Chem. Lett. 1983, 1337–1340;
d) K. Kikukawa, K. Kono, F. Wada, T. Matsuda, J. Org.
Chem. 1983, 48, 1333–1336.
Experimental Section
Synthesis of 3-Methylbiphenyl (6a)
A flame-dried, 20-mL reaction vessel equipped with a pressure
equalizer and magnetic stirrer was charged with 2,4,6-triphe-
nylboroxine (3a; 0.50 mmol, 155 mg), and dry potassium fluo-
ride (0.10 mmol, 5.80 mg). The reaction vessel was closed and
dried for 15 minutes under vacuum at 1008C. After cooling
to room temperature, di-m-chlorotetrakis(h2-ethylene)dirho-
dium(I) (5.80 mg, 0.015 mmol) and 3-methylbenzoic anhy-
dride (1a; 254 mg, 1.00 mmol) were added via syringe as a stock
solution in 4 mL anhydrous mesitylene. The mixture was then
stirred at 1608C and the progress of the reaction was monitored
by GC. After completion of the reaction (usually after 8 h), the
volatiles were removed under vacuum and the residue was pu-
rified by column chromatography (SiO2, hexane), affording 6a
as a white solid; yield: 108 mg(64%). The product was identi-
[6] S. B. Blakey, D. W. C. MacMillan, J. Am. Chem. Soc.
2003, 125, 6046–6047.
[7] a) L. J. Gooßen, K. Ghosh, Angew. Chem. 2001, 113,
3566–3568; Angew. Chem. Int. Ed. 2001, 40, 3458–
3460; b) R. Kakino, H. Narahashi, I. Shimizu, A. Yama-
moto, Bull. Chem. Soc. Jpn. 2002, 75, 1333–1345; c) R.
Kakino, S. Yasumi, I. Shimizu, A. Yamamoto, Bull.
Chem. Soc. Jpn. 2002, 75, 137–148; d) L. J. Gooßen, K.
Ghosh, Chem. Commun. 2001, 2084–2085.
1
fied by means of H and 13C NMR as well as by GC-MS and
HRMS to be 3-methylbiphenyl, CAS registry number [643-
93-6].
The side product 7a (40 mg, 20%), CAS registry number
[643-65-2], was isolated after further elution with hexanes/eth-
yl acetate.
[8] a) H.-U. Blaser, A. Spencer, J. Organomet. Chem. 1982,
233, 267–274; b) M. S. Stephan, A. J. J. M. Teunissen,
G. K. M. Verzijl, J. G. de Vries, Angew. Chem. 1998,
110, 688–690; Angew. Chem. Int. Ed. 1998, 37, 662–
664; c) L. J. Gooßen, J. Paetzold, Angew. Chem. 2002,
114, 1285–1289; Angew. Chem. Int. Ed. 2002, 41,
1237–1241; d) L. J. Gooßen, J. Paetzold, Angew. Chem.
2004, 116, 1115–1118; Angew. Chem. Int. Ed. 2004, 43,
1095–1098; e) A. F. Shmidt, V. V. Smirnov, Kinet. Catal.
2000, 41, 743–744.
Compounds 6b–n were synthesized accordingly (see Sup-
portingInformation).
Acknowledgements
We thank Prof. Dr. M. T. Reetz for generous support and con-
stant encouragement and gratefully acknowledge the financial
support of the DFG, the FCI, and the BMBF.
[9] Sugihara, T. Satoh, M. Miura, M. Nomura, Angew.
Chem. 2003, 115, 4820–4822; Angew. Chem. Int. Ed.
2003, 42, 4672–4674.
References and Notes
[10] L. J. Gooßen, N. Rodriguez, Chem. Commun. 2004, 724–
725.
[1] Z. Tan, E. Negishi, Synthesis of Natural Products via Pal-
ladium-Catalyzed Cross-Coupling, in: Handbook of Or-
ganopalladium Chemistry for Organic Synthesis, (Ed.:
E. Negishi), John Wiley & Sons, New York, 2002, Vol.
1, pp. 863–942.
[2] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483.
[3] For recent reviews see: a) N. Miyaura, in: Topics in Cur-
rent Chemistry – Cross-Coupling Reactions, Vol. 219,
(Ed.: N. Miyaura), Springer, Heidelberg, 2002, pp. 12–
59; b) K. Fagnou, M. Lautens, Chem. Rev. 2003, 103,
169–196; c) A. F. Littke, G. C. Fu, Angew. Chem. 2002,
114, 4350–4386; Angew. Chem. Int. Ed. 2002, 41,
[11] C. G. Frost, K. J. Wadsworth, Chem. Commun. 2001,
2316–2317; b) K. Oguma, M. Miura, T. Satoh, M. No-
mura, J. Organomet. Chem. 2002, 648, 297–301.
[12] E. M. OꢁBrien, E. A. Bercot, T. Rovis, J. Am. Chem. Soc.
2003, 125, 10498–10499.
[13] The difference to 100% yield is made up mostly by car-
boxylic acid, probably due to hydrolysis caused by in-
complete dehydration of the boronic acid in the preced-
ingstep.
1668
ꢀ 2004 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1665–1668