Manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives

Article abstract of DOI:10.1016/j.tetlet.2019.02.028

Herein, we report a manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives through cyclic C–C bond cleavage. The reaction happens via a radical-mediated pathway to selectively generate 1,5-ketoesters. A variety of substrates with subst

Full text of DOI:10.1016/j.tetlet.2019.02.028

Tetrahedron Letters xxx (xxxx) xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Manganese-catalyzed ring-opening carbonylation of cyclobutanol
derivatives
⇑
Tim Meyer, Zhiping Yin, Xiao-Feng Wu
LIKAT (Leibniz-Institut für Katalyse e.V.), Albert-Einstein-Straße 29a, 18059 Rostock, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein, we report
a manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives
Received 23 January 2019
Revised 12 February 2019
Accepted 14 February 2019
Available online xxxx
through cyclic CAC bond cleavage. The reaction happens via a radical-mediated pathway to selectively
generate 1,5-ketoesters. A variety of substrates with substituents on the aromatic ring reacted with linear
alcohols of different chain lengths. Obtained aliphatic esters are very attractive since they are usually dif-
ficult to access.
Ó 2019 Elsevier Ltd. All rights reserved.
Keywords:
Carbonylation
Ring-opening
Manganese
1,5-Ketoesters
Introduction
(OAc)4 (LTA) as the one electron oxidation system and catalyst
[13]. Phenylcyclobutanol 1a could be successfully converted into
The development of catalysts with abundant and therefore
inexpensive 3d transition metal complexes is a good alternative
to the extensively researched use of noble metals [1]. Such systems
would be particularly attractive in combination with low-cost
commercially available ligands and give improved reaction effi-
ciency and reactivity [2]. Catalytic ring-opening reactions are
highly efficient because the starting material is fully incorporated
into the target molecule. Tertiary cycloalkanols caught our interest,
which have proven to be privileged starting materials for oxidative
ring opening reactions [3]. The using of palladium catalysts [4] and
rhodium catalysts [5] have been well established in this research
field. However, early publications showed that it is possible to
use non-noble 3d transition metals such as Mn(III) [6], Cr(IV) [7]
or V(V) [8] to induce single-electron oxidization of cyclobutanol
resulting in a ring-opened free-radical intermediate. Recently,
some successes have been reported with catalysts based on Mn
[9] or Ag [10] to generate distally functionalized ketones. With
regard to carbonylation reactions, ring opening or expansion of
epoxides were reported and allow the synthesis of products with
high added value (e.g. lactones [11] or succinic anhydrides [12]).
The carbonylative cleavage of cyclobutanols is more challenging
and much less explored. To the best of our knowledge, only one
synthetic pathway was reported by Ryu and Sonada in 1994. The
carbonylation could be achieved using 1.2 equivalents of Pb
the desired 4-benzoylbutyric acid 6 in 43% yield and benzoic acid
7 as the by-product in 4% yield. From this it was concluded that
b-scission a) was a minor competitive reaction (Scheme 1). This
system faces obvious drawbacks concerning the selectivity and
leads to relatively low yields, besides the use of LTA in stoichiomet-
ric quantities, which is very toxic and should be avoided concern-
ing green chemistry. In continuation of our research on first row
transition-metal-catalyzed carbonylations, we succeeded in
designing a new Mn-catalyzed method for 1,5-ketoesters produc-
tion. The above mentioned drawbacks could be eliminated.
⇑
Corresponding author.
E-mail address: Xiao-Feng.Wu@catalysis.de (X.-F. Wu).
Scheme 1. Ring-opening carbonylation of cyclobutanol.
https://doi.org/10.1016/j.tetlet.2019.02.028
0040-4039/Ó 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: T. Meyer, Z. Yin and X. F. Wu, Manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives, Tetrahedron Letters,
https://doi.org/10.1016/j.tetlet.2019.02.028

2
T. Meyer et al. / Tetrahedron Letters xxx (xxxx) xxx
system as well, but only 15% of the desired product could be
Results and discussion
obtained. With the suitable ligand and oxidant at hand, we
started a thorough screening of solvents. Trifluorotoluene (TFT),
which was used from the beginning, emerged as the best solvent.
However, the same yield could be achieved with m-xylene (51%,
entry 16). Interesting regarding green chemistry would be if a
mixture of isomers would give comparable results, since xylenes
are petrochemically produced on large scale by catalytic
reforming. Other tested solvents all gave decreased yields
(entries 16–20). Interestingly, the use of less polar ‘‘non-
fluorinated” toluene resulted in a significantly lower yield (31%,
entry 17). In a comparative reaction with methanol as reactant
For our preliminary studies we have chosen 1-phenyl-cyclobu-
tanol 1a as the model substrate to establish this procedure [9,14].
We conducted comprehensive optimization of the reaction condi-
tions that are outlined in Table 1. Of all the manganese-based cat-
alysts tested, manganese acetate has proven to be the most
effective one (entries 1–5). To provide a more efficient catalytic
system, we examined various commercially available ligands
(entries 6–12, 3a–3f). The N,N-bidentate ligands (3c, 3d and 3f)
tested all gave worse results. The uses of simple pyridine as the
ligand could not futher improve the yield (entries 10–12). In order
to demonstrate the importance of the ligand to be bidentate, 2-
phenylpyridine 3e was used, which differs from 3a only by a miss-
ing nitrogen, but the yield was 11% lower. With a monodentate
ligand having a strong electron-donating group in unobstructed
para-position such as DMAP 3b, it was possible to get the same
results as with BiPy (entry 6). The effect of different oxidants
was explored subsequently. In line with the recently reported
results for oxidative ring-opening reactions [9a], the hypervalent
iodine oxidants (entry 13, 4b) stand out due to their high
reactivity. It is noteworthy that Ce(SO4)2 provided similar results
compared to iodosylbenzene 4c (entries 13–15). PIFA ((bis
(trifluoroacetoxy)iodo)benzene) was tested as oxidant in our
and solvent, only
a minor amount of desired product was
generated (8%, entry 20). Finally, the CO pressure was increased
from 40 to 60 bar and the yield improved dramatically (80%,
entry 21). In order to establish direct comparability with the
aforementioned system of Ryu and Sonada, LTA was tested under
the final conditions and lead to a low yield here (33%, entry 22).
With the optimal conditions in hand, our method was applied
on several substrates in combination with different alcohols
(Table 2) [15]. The reaction was studied with linear alcohols of dif-
ferent chain lengths (2a–2e) and several substrates with sub-
stituents on the aromatic ring of different electronic properties
(1a–1g) including di-substituted ones (1e, 1g). In general, good
yields of the corresponding products with different aliphatic alco-
hols (5aa–5ae) can be achieved. Among them, ester product 5ac
could be obtained in an excellent yield of 91% (entry 3). To our
delight, it was also possible to use long chained alcohols like n-
octanol 2e to generate the elusive aliphatic ester 5ae in a synthet-
ically useful yield (73%, entry 5). However, only trace amount of
the desired product could be detected when i-proponal or tert-
butanol was tested. Our carbonylation system exhibited functional
group tolerance with fluoro and methoxy groups on the aromatic
ring. Encouraged by the good results with n-butanol, n-butanol
and methanol were tested with the substrates, respectively.
Remarkably, the trend of increased reactivity in combination with
n-butanol and the associated increased yields also continued for
the various substituted cyclobutanols. In case of the alkyl-substi-
tuted substrate 1c together with methanol 2a, the corresponding
product 5ca was obtained in moderate yield (55%, entry 7) and
in presence of n-butanol corresponding product 5cc could be
obtained in a good yield (74%, entry 8). This phenomenon could
also be observed for single and double substituted products, with
a dramatic improvement in yields for para-methoxy substituted
esters 5da (40%, entry 9) and 5dc (89%, entry 10). The same was
true for the meta-fluoro substituted esters 5fa (30%, entry 13)
and 5fc (86%, entry 14). In both cases the respective yield has more
than doubled when n-butanol was used instead of methanol. Di-
substituted products with a methoxy group in para- and meta-
position 5ea (31%, entry 11) and 5ec (46%, entry 12) also showed
an improvement, but less significant. In the case of difluorosubsti-
tuted substrate, the desired product could only be isolated in
traces, even with n-butanol 5gc (<12%, entry 15). Additionally,
alkyl substituted and non-substituted cyclobutanol were tested
under standard conditions as well, but no desired products could
be detected.
Table 1
Optimization of Reaction Conditions.
Entrya
Deviation from standard conditions
Yield (%)b
1
2
3
4
5
6
7
8
None
51
41
38
26
19
51
50
41
41
36
34
27
34
28
27
51
31
26
25
8
Mn(CO)5Br as catalyst
MnCl2 as catalyst
Mn(acac)2 as catalyst
MnBr2 as catalyst
DMAP 3b as ligand
Batophen 3c as ligand
1,10-Phenantroline 3d as ligand
2-Phenylpyridine 3e
Pyridine as ligand
No ligand
TMEDA 3f as ligand
BI-OH 4b as oxidant
Ce(SO4)2 as oxidant
4c as oxidant
m-Xylene as solvent
Toluene
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Dimethylcarbonate
Dioxane
MeOH
CO (60 bar)
CO (60 bar), Pb(OAc)4 4e as oxidant
A plausible mechanism is proposed based on experimental
observations and literatures (Scheme 2). First the hypervalent
iodine reagent oxidizes the catalyst and together with cyclobutanol
a to give the Mn(V) species b. Subsequent single-electron transfer
(SET) releases the cyclobutyloxy radical c, which undergoes a ‘rad-
ical clock’-type ring opening tautomerization, leading to alkyl rad-
ical d [14a]. Under the given pressure, it is likely that together with
CO the acyl radical e will be formed. It can coordinate to the
manganese center to generate f. Then X ligand exchange leads to
a Mn(V) complex g which is able to give the final ester product h
80 (79c)
33
a
Unless otherwise noted, all the reactions were conducted on a 0.2 mmol scale in
the presence of the alcohol 2 (5.0 mmol) using an autoclave
b
GC yields were determined by GC-FID analysis using n-hexadecane as internal
standard
c
The isolated yield of isolated product 5aa was obtained from a reaction on a
0.3 mmol scale.
Please cite this article as: T. Meyer, Z. Yin and X. F. Wu, Manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives, Tetrahedron Letters,
https://doi.org/10.1016/j.tetlet.2019.02.028

T. Meyer et al. / Tetrahedron Letters xxx (xxxx) xxx
3
Table 2
Manganese-Catalyzed Ring-Opening Carbonylation of Cyclobutanol Derivatives: Substrate Scope.
Entry
1
Cyclobutanol
Alcohol
Product
Yield [%][b]
79
2
3
4
5
6
7
61
91
70
73
47
55
8
74
40
89
31
46
9
10
11
12
13
14
30
86
(continued on next page)
Please cite this article as: T. Meyer, Z. Yin and X. F. Wu, Manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives, Tetrahedron Letters,
https://doi.org/10.1016/j.tetlet.2019.02.028

4
T. Meyer et al. / Tetrahedron Letters xxx (xxxx) xxx
Table 2 (continued)
Entry
15
Cyclobutanol
Alcohol
Product
Yield [%][b]
<12
[a] Unless otherwise noted, all the reactions were conducted on a 0.3 mmol scale in the presence of the alcohol 2 (5.0 mmol) using an autoclave.
[b] Isolated yields.
b) J.-B. Peng, F.-P. Wu, X.-F. Wu, Chem. Rev. (2018), in press;
c) X.-X. Guo, D.-W. Gu, Z. Wu, W. Zhang, Chem. Rev. 115 (2014) 1622–1651;
d) C.-L. Sun, B.-J. Li, Z.-J. Shi, Chem. Rev. 111 (2010) 1293–1314;
e) A. Fürstner, Angew. Chem. 121 (2009) 1390–1393, Angew. Chem. Int. Ed.
2009, 48, 1364-1367;
f) A. Correa, O.G. Mancheño, C. Bolm, Chem. Soc. Rev. 37 (2008) 1108–1117.
[2] a) B.M. Trost, Angew. Chem. 107 (1995) 285–307, Angew. Chem. Int. Ed. 1995,
34, 259-281;
b) B.M. Trost, Science 254 (1991) 1471–1477.
[3] a) I. Marek, A. Masarwa, P.O. Delaye, M. Leibeling, Angew. Chem. 127 (2015)
424–439, Angew. Chem. Int. Ed. 2015, 54, 414-429;
b) T. Seiser, T. Saget, D.N. Tran, N. Cramer, Angew. Chem. 123 (2011) 7884–
7896, Angew. Chem., Int. Ed. 2011, 50, 7740-7752;
c) H. Fujioka, H. Komatsu, A. Miyoshi, K. Murai, Y. Kita, Tetrahedron Lett. 52
(2011) 973–975;
d) R. Ren, H. Zhao, L. Huan, C. Zhu, Angew. Chem. Int. Ed. 54 (2015) 12692–
12696;
e) D. Wang, R. Ren, C. Zhu, J. Org. Chem. 81 (2016) 8043–8049;
f) L. Huan, C. Zhu, Org. Chem. Front. 3 (2016) 1467–1471;
g) M. Wang, Z. Wu, C. Zhu, Org. Chem. Front. 4 (2017) 427–430;
h) X. Wu, C. Zhu, Chem. Rec. 18 (2018) 587–598.
[4] a) T. Nishimura, S. Matsumura, Y. Maeda, S. Uemura, Chem. Commun. (2002)
50–51;
b) T. Nishimura, K. Ohe, S. Uemura, J. Am. Chem. Soc. 121 (1999) 2645–2646;
c) T. Nishimura, S. Uemura, J. Am. Chem. Soc. 121 (1999) 11010–11011.
[5] a) N. Ishida, Y. Nakanishi, M. Murakami, Angew Chem. 125 (2013) 12091–
12094, Angew. Chem. Int. Ed. 2013, 52, 11875-11878;
b) T. Seiser, N. Cramer, J. Am. Chem. Soc. 132 (2010) 5340–5341.
[6] J. Rocek, A.E. Radkowsky, J. Org. Chem. 38 (1973) 89–95.
[7] J. Rocek, A.E. Radkowsky, J. Am. Chem. Soc. 90 (1968) 2986–2988.
[8] J. Rocek, D.E. Aylward, J. Am. Chem. Soc. 97 (1975) 5452–5456.
[9] a) R. Ren, Z. Wu, Y. Xu, C. Zhu, Angew. Chem. 128 (2016) 2916–2919, Angew.
Chem., Int. Ed. 2016, 55, 2866-2869;
Scheme 2. Proposed mechanism for the ring-opening carbonylation of
cyclobutanol.
b) R. Ren, Z. Wu, C. Zhu, Chem. Commun. 52 (2016) 8160–8163.
[10] a) X. Fan, H. Zhao, J. Yu, X. Bao, C. Zhu, Org. Chem. Front. 3 (2016) 227–232;
b) H. Zhao, X. Fan, J. Yu, C. Zhu, J. Am. Chem. Soc. 137 (2015) 3490–3493.
[11] a) J.A. Schmidt, E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc. 127 (2005)
11426–11435;
after reductive elimination. Evidence for the formation of benzoic
acid via b-scission as
a competitive reaction could not be
provided [13]. We assume that our catalytic system selectively
generates cyclobutyloxy radical c during the step of single-electron
transfer. However, we also can not excluded a Mn(III)/Mn(II) cat-
alytic cycle.
In conclusion, we described a method for the ring-opening car-
bonylation of cyclobutanols. Several elusive aliphatic 1,5-ketoe-
sters could be generated with our system, using simple
bipyridine as ligand and manganese acetate as catalyst both from
chemical feedstock.
b) M. Allmendinger, R. Eberhardt, G.A. Luinstra, F. Molnar, B. Rieger, Z. Anorg.
Allg. Chem. 629 (2003) 1347–1352;
c) V. Mahadevan, Y.D. Getzler, G.W. Coates, Angew Chem. 114 (2002) 2905–
2908, Angew. Chem. Int. Ed. 2002, 41, 2781-2784;
d) Y.D. Getzler, V. Mahadevan, E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc.
124 (2002) 1174–1175;
e) J.T. Lee, P. Thomas, H. Alper, J. Org. Chem. 66 (2001) 5424–5426.
[12] J.M. Rowley, E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc. 129 (2007) 4948–
4960.
[13] S. Tsunoi, I. Ryu, Y. Tamura, S. Yamasaki, N. Sonoda, Synlett (1994) 1009–1011.
[14] a) R. Ren, C. Zhu, Synlett 27 (2016) 1139–1144;
b) K.E. Liu, C.C. Johnson, M. Newcomb, S.J. Lippard, J. Am. Chem. Soc. 115
(1993) 939–947;
Acknowledgments
c) D. Griller, K.U. Ingold, Acc. Chem. Res. 13 (1980) 317–323.
[15] General procedure: A 4 mL screw-cap vial equipped with a septum, a small
We gratefully acknowledge the analytic support of Dr. W. Bau-
mann, Dr. C. Fischer, Dr. A. Spannenberg, S. Buchholz, and S. Schar-
eina. We thank Professor Matthias Beller and Professor Armin
Börner for providing a perfect working environment.
cannula, and a stirring bar was charged with starting material 1 (300
BiPy (30 mol%, 14.1 mg), Mn(OAc)3Á2H2O (20 mol%, 16.1 mg), TFT (3 mL), PIDA
4a (750 mol, 242 mg), and alcohol 2 (5 mmol). The vial was sealed, connected
lmol),
l
to atmosphere with a cannula, purged with argon three times, placed on an
alloy plate, and transferred into a 300 mL stovetop autoclave (4560 series from
Parr instrument companyÒ). The autoclave was flushed two times with argon
and two times with CO. It was then placed into an aluminum block on a
magnetic stirrer. The reaction mixture was stirred (600 rpm) for 20 h at 100 °C
and under pressure of 60 bar CO. Then it was cooled to room temperature and
the pressure was released carefully. As an internal standard, n-hexadecane (30
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2019.02.028.
lL, 100 lmol) was added to the reaction mixture. An aliquot (1 mL) of the
mixture was purified with a pipette flash column using EtOAc (3 mL) as the
eluent. Samples prepared in this way were subjected to GC analysis. Isolated
products were obtained by column chromatography on silica gel.
References
[1] a) Y. Li, Y. Hu, X.-F. Wu For selected reviews on catalyses with 3d transition
metal complexes see: Chem. Soc. Rev. 47 (2018) 172–194;
Please cite this article as: T. Meyer, Z. Yin and X. F. Wu, Manganese-catalyzed ring-opening carbonylation of cyclobutanol derivatives, Tetrahedron Letters,
https://doi.org/10.1016/j.tetlet.2019.02.028