Cationic Cobalt Porphyrin-Catalyzed Allylation of Aldehydes with Allyltrimethylsilanes

Article abstract of DOI:10.1021/acs.orglett.9b01303

Cationic cobalt porphyrin-catalyzed allylation of aldehydes with allyltrimethylsilanes is developed. The formation of the aldehyde-cobalt porphyrin complex, the key intermediate for the addition of allylsilanes, is confirmed by theoretical studies and syn

Full text of DOI:10.1021/acs.orglett.9b01303

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Cationic Cobalt Porphyrin-Catalyzed Allylation of Aldehydes with
Allyltrimethylsilanes
Rei Tomifuji, Shota Masuda, Takuya Kurahashi,* and Seijiro Matsubara*
Department of Material Chemistry, Graduate School of Engineering, Kyoto University, Kyoto 615-8510, Japan
S
* Supporting Information
ABSTRACT: Cationic cobalt porphyrin-catalyzed allylation of
aldehydes with allyltrimethylsilanes is developed. The formation of
the aldehyde−cobalt porphyrin complex, the key intermediate for the
addition of allylsilanes, is confirmed by theoretical studies and
synchrotron-based X-ray absorption fine-structure measurements.
Facile dissociation of the product by allylation from the cobalt
complex regenerates the active complex with the aldehyde. The readily
obtainable [Co(TPP)]SbF6 complex serves as an efficient catalyst for
this allylation.
etalloporphyrins are well-known to have important
biochemical functions in nature. Therefore, there has
Scheme 1. Lewis Acidity of [Co(TArP)]SbF6 CP1
M
been much interest in exploiting the unique character of
metalloporphyrins in organic synthesis.1 Recently, we reported
that metalloporphyrins could serve as a Lewis acid to catalyze
various reactions such as the formal hetero-Diels−Alder
reaction, cycloisomerization, and cycloaddition.2 Herein, we
report that cobalt porphyrin can catalyze the allylation of
aldehydes via the addition of allyltrimethylsilanes to give
homoallylic alcohols.
Allylation of aldehydes in the presence of a Lewis acid
catalyst is a fundamental transformation involving C−C bond
formation to afford homoallylic alcohols.3 The reaction
mechanism involves coordination and activation of the
aldehyde with a strong Lewis acid as the key step.
Consequently, there is often a trade-off between the activation
of the aldehyde for coordination with the Lewis acid and
product dissociation from the Lewis acid, thereby necessitating
the use of a stoichiometric or substoichiometric amount of the
Lewis acid catalyst. Therefore, the development of an efficient,
easy-to-prepare catalyst for the allylation, especially for
reactions with less reactive and thermodynamically stable
allylsilanes (e.g., allyltrimethylsilanes), remains a topic of
interest. We hypothesized that the cationic cobalt porphyrin
complex, by virtue of its high Lewis acidity, would catalytically
facilitate the activation of the aldehyde, thereby promoting the
addition of allyltrimethylsilane to afford the corresponding
homoallylic alcohol. Moreover, the steric hindrance of the
porphyrin ligand would promote the dissociation of the alcohol
so that a high turnover frequency and high turnover number
can be achieved.
higher Lewis acidity. The Lewis acidity of CP1 was comparable
to that of the rare-earth Lewis acid Sc(OTf)3.3a Encouraged by
these results, we next investigated the reaction of aldehyde 1a
and allylsilane 2a in the presence of a catalytic amount of CP1
(1 mol %). As shown in Scheme 2, the reaction proceeded at
25 °C and was completed in 1 h to afford homoallylic alcohol
3aa′ in 95% yield after protodesilylation of 3aa. To understand
the efficiency of the cobalt porphyrin catalyst, we carried out
density functional theory (DFT) calculations to determine the
Gibbs free energy change for the dissociation of 3aa in cobalt
Scheme 2. [Co(TArP)]SbF6-Catalyzed Allylation
To prove our hypothesis, we initially examined the Lewis
acidity of cationic cobalt porphyrin CP1 by IR spectroscopy.
We measured the carbonyl stretching vibration of the 2,6-
dimethyl-γ-pyrone complexes of Lewis acid in 30 mM CH2Cl2
solution in a sealed liquid cell at 25 °C (Scheme 1); a larger
shift (ΔνCO) of the carbonyl absorption peak indicated
Received: April 14, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b01303
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Gibbs Free Energy Change for Dissociation of
Product 3aa from [Co(TPP)]SbF6 and Coordination of
Aldehyde 1a to [Co(TPP)]SbF6 CP1
Figure 2. (a) Major contributor to the aldehyde−cobalt porphyrin
interactions in second-order perturbation theory analysis of the Fock
matrix in NBO basis of CP2. LP and LAO denote lone pair and Lewis
acceptor orbital, respectively. (b) Counterpoise-corrected interaction
energy (Eint) analysis based on the fragment molecular orbital scheme.
involved in the complex formation also suggested weak
interactions (Figure 1b, −10.3 kcal/mol). On the other
hand, the interaction between 1a and the cobalt porphyrin in
CP2 was found to be strong. The Lewis acid−Lewis base
interaction was calculated to be 92.1 kcal/mol by NBO
analysis (Figure 2a). Furthermore, Eint analysis suggested
strong interaction between 1a and cobalt porphyrin (Figure 2b,
−33.6 kcal/mol).
porphyrin CP3 to regenerate active catalyst CP1 for the next
catalytic cycle and the coordination of 1a to CP1 to generate
active species CP2 for the addition of 2a (Scheme 3). The
Gibbs free energy change for the dissociation of 3aa in CP3
was calculated to be −7.1 kcal/mol; that is, the reaction is
exergonic, has a favorable equilibrium constant, and can occur
spontaneously. Meanwhile, the Gibbs free energy change for
the coordination of 1a to CP1 was calculated to be −24.9 kcal/
mol relative to the Gibbs energy of CP3; this reaction also has
a favorable equilibrium constant and would be spontaneous.
Thus, the dissociation of 3aa in CP3 to afford CP1 and the
subsequent coordination of 1a to the CP1 are exergonic,
consistent with the short reaction time under mild conditions.
To understand why the dissociation of product 3aa and the
coordination of aldehyde 1a are facilitated, the interaction
energy of 3aa and the cobalt porphyrin in CP3, as well as that
of 1a and the cobalt porphyrin in CP2, were investigated. The
weak interactions in CP3 could be confirmed by the analysis of
noncovalent interactions. As shown in Figure 1a, only weak
interactions were observed between 3aa and the cobalt
porphyrin. The counterpoise-corrected interaction energy
(Eint) analysis based on the fragment molecular orbital scheme
To gain further insight into the in situ formation of CP2 via
the coordination of 1a to CP1, we performed solution-phase in
situ Co K-edge X-ray absorption spectroscopy (XAS) at a
synchrotron radiation facility. XAS measurements were carried
out using a toluene solution of CP1 and 1a at 25 °C. The
results of extended X-ray absorption fine structure (EXAFS)
analysis are shown in Figure 3. The spectral features arising
from the coordination sphere of the cobalt atoms, i.e., the
radial distance of 1.1−2.2 Å, matched well with the spectra
simulated using the DFT-calculated structure of complex CP2
−
(Rf = 1.7%), in which 1a and counteranion SbF6 were
coordinated to the cationic cobalt porphyrin. EXAFS analysis
of the coordination sphere of cobalt indicated that the cobalt
atoms had octahedral coordination geometry, with the oxygen
Figure 1. (a) NCI (noncovalent interactions) plot of CP3. The
gradient isosurfaces (s = 0.5 au) are colored over the range −0.02 to
+0.02 au in terms of sign (λ2)ρ. (b) Counterpoise-corrected
interaction energy (Eint) analysis based on the fragment molecular
orbital scheme.
Figure 3. Solution-phase Co K-edge EXAFS analysis (Fourier
transform of k3-weighted spectrum, Δk = 3.0−11.8 Å−1, ΔR = 1.1−
2.2 Å, spatial resolution δR = 0.133 Å, Rf = 1.7%): simulated spectrum
based on DFT structure of CP2 (red line) and experimental spectrum
(black line; mixture of CP1 and 1a).
B
DOI: 10.1021/acs.orglett.9b01303
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Table 1. Allylation of Aldehydes 1 with Allylsilane 2
Figure 4. Solution-phase Co K-edge XANES spectra: CP1 (black
line), CP2 (red line; mixture of CP1 and 1a), and CP3 (blue line;
mixture of CP1 and 3aa).
atom of 1a and the nearest fluoride atom of SbF6− located at a
distance of 1.89 and 2.01 Å from cobalt, respectively. The
structure of complex CP2 calculated by DFT was consistent
with the observed coordination geometry of the cobalt atoms
and bond lengths in the coordination sphere (Co−O, 1.91 Å,
and Co−F, 2.02 Å). The Co K-edge XANES spectra of three
different solutions are shown in Figure 4 for the following
cases: (1) solution of CP1 without the addition of 1a or 3aa;
(2) mixture of CP1 with 1a as a representative solution of
CP2; and (3) mixture of CP1 with 3aa as a representative
solution of CP3. Clearly, the unique spectral feature was
observed only for the mixture of CP1 with 1a, suggesting the
formation of CP2. Meanwhile the addition of 3aa to CP1 did
not lead to any change in the spectra as compared with that of
CP1. Thus, the Gibbs energy changes calculated by DFT for
the dissociation of 3aa and coordination of 1a were also
consistent with the observed changes in these three XANES
spectra (vide supra, Scheme 3).
Based on the theoretical study and X-ray spectroscopic
analysis, we envisioned that the cobalt porphyrin complex
[Co(TPP)]SbF6, which can be readily prepared using the
commercially available [Co(TPP)]Cl and AgSbF6, would also
promote the allylation. When using [Co(TPP)]SbF6 in place
of CP1, the reaction temperature had to be elevated to address
the insolubility of the complex in organic solvents attributed to
the strong π−π stacking intermolecular interactions. Nonethe-
less, the reaction of aldehyde 1a with allylsilane 2a in the
presence of cationic cobalt porphyrin catalyst [Co(TPP)]SbF6
(1 mol %) afforded homoallylic alcohol 3aa′ in 97% yield at
100 °C. Thus, the cationic cobalt porphyrin [Co(TPP)]SbF6-
catalyzed allylation of aldehydes was briefly examined; the
results are summarized in Table 1. Under the optimized
conditions, aromatic aldehydes possessing variety of functional
groups on the phenyl moiety reacted with allylsilane 2a to
afford the corresponding substituted product in good to
moderate yields. Aliphatic aldehydes could also participate in
the reaction to afford homoallylic alcohols. Of note, the
reaction of dicarbonyl compounds such as 1m and 1s afforded
3ma′ and 3sa′ in 99% and 57% yields, respectively, while the
use of Sc(OTf)3 as the catalyst resulted in lower yields (52%
a
Reaction conditions: [Co(TPP)]SbF6 (1 mol %), aldehyde 1 (0.2
mmol), and allylsilane 2 (0.5 mmol) in 1 mL of DCE for 4 h.
b
c
Isolated yields are given. Reaction with Sc(OTf)3 catalyst instead of
d
cobalt porphyrin catalyst [Co(TPP)]SbF6. Ratio of isomers was
determined by H NMR analysis of the crude mixture.
1
and 14%, respectively). These results highlight the unique
reactivity of the [Co(TPP)]SbF6 catalyst. For example, in case
of the reaction using 1s, which has two carbonyl moieties, 1s
acts as a bidentate chelating ligand toward Sc(OTf)3; thus,
dissociation of the product is prevented. Meanwhile, the Co
porphyrin has only one d-orbital for available coordination
sites, so 1s behaves as a monodentate ligand; consequently, a
higher yield of the product is obtained as compared with that
in the case of the Sc(OTf)3 catalyst. The use of
methylallyltrimethylsilane 2b instead of 2a also provided the
corresponding product 3pb′. However, the reaction with
crotyltrimethylsilane did not afford the desired product. Note
that the drawback of the [Co(TPP)]SbF6‑ catalyzed reaction is
the formation of a side product when employing an aromatic
aldehyde with an electron-donating substituent, as is
commonly observed with other Lewis acid and Brønsted acid
systems.4 For example, the reaction of anisaldehyde with 2a
afforded a diallylated product 4ua in 32% yield as the major
product, while the desired homoallylic alcohol was not formed
(SI).
To investigate the diastereoselectivity of the cobalt
porphyrin-catalyzed allylation, the reaction of 2a with 2-
C
DOI: 10.1021/acs.orglett.9b01303
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
phenylpropionaldehyde 1t was examined, and homoallylic
alcohol 3ta′ was obtained in 92% yield as a 1:1 diastereomeric
mixture. This result indicated that the allylation proceeded via
an acyclic transition state for the addition of the allylsilane to
the carbonyl carbon of the aldehyde and not via a chairlike six-
membered transition state (the Zimmerman−Traxler model).
In order to elucidate the reaction mechanism, we examined the
allylation with allylsilane 2d possessing a sterically hindered
TBDMS group. The reaction did not afford the allylation
product 3aa′. This result, along with the reaction scope
provided in Table 1, clearly suggested that the plausible rate-
determining step in the catalytic cycle is silyl group migration
and not addition of the allylsilane to the aldehyde carbonyl
carbon or dissociation to regenerate the initial active Lewis
acid cobalt porphyrin complex.
In conclusion, we have demonstrated for the first time that a
cobalt porphyrin Lewis acid can catalyze the allylation of
aldehydes with allyltrimethylsilanes. The use of a cationic
cobalt porphyrin complex as the Lewis acid catalyst facilitates
activation of the aldehyde via coordination and dissociation of
the product, so that the reaction can proceed to completion
within a short time under mild conditions. The allylation
proceeds with 1 mol % of the cobalt porphyrin catalyst CP1 at
25 °C within 1 h. The coordination of the aldehyde to the
cobalt porphyrin catalyst is demonstrated by Co K-edge XAS
analysis and DFT calculations. Moreover, allylation with a
diverse range of aldehydes can be promoted by the readily
available [Co(TPP)]SbF6. Since the use of a metalloporphyrin
Lewis acid as the catalyst enables allylation within a short time,
further research into metalloporphyrin Lewis acid catalysts may
reveal the intrinsic catalytic activity of first-row transition
metals. Studies in this regard are ongoing in our laboratory,
and the results will be reported in due course.
Prof. Tsunehiro Tanaka (Kyoto University), Prof. Masaharu
Nakamura (Kyoto University), Prof. Hikaru Takaya (Kyoto
University), Dr. Tetsuo Honma (Japan Synchrotron Radiation
Research Institute; JASRI), Dr. Hironori Ofuchi (JASRI), Dr.
Masafumi Takagaki (JASRI), and Mr. Kyohei Fujiwara
(Ajinomoto Co., Inc.) for valuable support in the X-ray
absorption fine structure analysis. Part of this study was
performed at the beamline of the SPring-8 synchrotron
radiation facility (Proposal Nos. 2016A1549 and 2019A1712).
REFERENCES
■
(1) For examples of the use of metalloporphyrins in nonoxidative
bond formation, see: (a) Suda, K.; Baba, K.; Nakajima, S.-I.;
Takanami, T. Chem. Commun. 2002, 2570. (b) Suda, K.; Kikkawa,
T.; Nakajima, S.-I.; Takanami, T. J. Am. Chem. Soc. 2004, 126, 9554.
(c) Suda, K.; Baba, K.; Nakajima, S.-I.; Takanami, T. Tetrahedron Lett.
1999, 40, 7243. (d) Schmidt, J. A. R.; Lobkovsky, E. B.; Coates, G. W.
J. Am. Chem. Soc. 2005, 127, 11426. (e) Zhou, C.-H.; Chan, P. W. H.;
Che, C.-M. Org. Lett. 2006, 8, 325. (f) Tachinami, T.; Nishimura, T.;
Ushimaru, R.; Noyori, R.; Naka, H. J. Am. Chem. Soc. 2013, 135, 50.
(g) Ushimaru, R.; Nishimura, T.; Iwatsuki, T.; Naka, H. Chem. Pharm.
Bull. 2017, 65, 1000. (h) Liu, Y.; Wei, J.; Che, C.-M. Chem. Commun.
2010, 46, 6926.
(2) (a) Fujiwara, K.; Kurahashi, T.; Matsubara, S. J. Am. Chem. Soc.
2012, 134, 5512. (b) Wakabayashi, R.; Kurahashi, T.; Matsubara, S.
Org. Lett. 2012, 14, 4794. (c) Ozawa, T.; Kurahashi, T.; Matsubara, S.
Org. Lett. 2012, 14, 3008. (d) Ozawa, T.; Kurahashi, T.; Matsubara, S.
Synlett 2013, 24, 2763. (e) Teranishi, S.; Kurahashi, T.; Matsubara, S.
Synlett 2013, 24, 2148. (f) Shiba, T.; Kuroda, D.; Kurahashi, T.;
Matsubara, S. Synlett 2014, 25, 2005. (g) Maeda, K.; Terada, T.;
Kurahashi, T.; Matsubara, S. Org. Lett. 2015, 17, 5284. (h) Kuwano,
T.; Kurahashi, T.; Matsubara, S. Chem. Lett. 2013, 42, 1241.
(i) Teranishi, S.; Maeda, K.; Kurahashi, T.; Matsubara, S. Org. Lett.
2019, 21, 2593.
(3) For selected examples forallylation of aldehyde with
allyltrimethylsilane by Lewis acid catalyst, see: (a) Aggarwal, V. K.;
Vennall, G. P. Synthesis 1998, 1998, 1822. (b) Akiyama, T.; Iwai, J.;
Sugano, M. Tetrahedron 1999, 55, 7499. (c) Marx, A.; Yamamoto, H.
Angew. Chem., Int. Ed. 2000, 39, 178. (d) Yadav, J. S.; Chand, P. K.;
Anjaneyulu, S. Tetrahedron Lett. 2002, 43, 3783. (e) Watahiki, T.;
Oriyama, T. Tetrahedron Lett. 2002, 43, 8959. (f) Lingaiah, B. V.;
Ezikiel, G.; Yakaiah, T.; Reddy, G. V.; Rao, P. S. Tetrahedron Lett.
2006, 47, 4315. (g) Yadav, J. S.; Reddy, B. V. S.; Vishnumurthy, P.;
Chary, C. J. Tetrahedron Lett. 2007, 48, 5915. (h) Wang, M. W.;
Chen, Y. J.; Wang, D. Synlett 2000, 385. (i) Mahlau, M.; García-
García, P.; List, B. Chem. - Eur. J. 2012, 18, 16283. (j) Hosomi, A.;
Endo, M.; Sakurai, H. Chem. Lett. 1976, 5, 941. (k) Ishihara, K.;
Mouri, M.; Gao, Q. Z.; Maruyama, T.; Furuta, K.; Yamamoto, H. J.
Am. Chem. Soc. 1993, 115, 11490.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b01303.
Experimental procedures including spectroscopic and
analytical data of new compounds (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
(4) Ishihara, K.; Hiraiwa, Y.; Yamamoto, H. Synlett 2001, 30, 1851.
*E-mail: kurahashi.takuya.2c@kyoto-u.ac.jp.
*E-mail: matsubara.seijiro.2e@kyoto-u.ac.jp.
ORCID
Takuya Kurahashi: 0000-0001-8417-7660
Seijiro Matsubara: 0000-0001-8484-4574
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Advanced Catalytic Trans-
formation Program for Carbon Utilization (ACT-C No.
JPMJCR12YE) from Japan Science and Technology Agency
(JST) and a Grant-in-Aid for Scientific Research (18H04253,
17KT0006, and 15H05845) from the Ministry of Education,
Culture, Sports, Science and Technology (Japan). T.K.
acknowledges the Asahi Glass Foundation. We are grateful to
D
DOI: 10.1021/acs.orglett.9b01303
Org. Lett. XXXX, XXX, XXX−XXX