Anti-Markovnikov rearrangement in sulfur mediated allylic C-H amination of olefins

Article abstract of DOI:10.1039/c6cc05128f

Cationic rearrangement reactions usually follow Markovnikov's rule to give more substituted carbocations as stable intermediates. During our study on sulfur mediated allylic C-H amination of olefins, very rare cases of anti-Markovnikov rearrangement from secondary carbocations toward primary carbocations or primary triflates were observed.

Full text of DOI:10.1039/c6cc05128f

ChemComm
View Article Online
View Journal
COMMUNICATION
Anti-Markovnikov rearrangement in sulfur
mediated allylic C–H amination of olefins†
Cite this:DOI: 10.1039/c6cc05128f
Zhong Zhang, Hongguang Du,* Jiaxi Xu* and Pingfan Li*
Received 20th June 2016,
Accepted 25th August 2016
DOI: 10.1039/c6cc05128f
www.rsc.org/chemcomm
Cationic rearrangement reactions usually follow Markovnikov’s rule
to give more substituted carbocations as stable intermediates.
During our study on sulfur mediated allylic C–H amination of
olefins, very rare cases of anti-Markovnikov rearrangement from
secondary carbocations toward primary carbocations or primary
triflates were observed.
The essence of Markovnikov’s rule is the stability of cationic
intermediates: tertiary carbocations are more stable than the
corresponding secondary ones, which are more stable than the
corresponding primary ones. Most organic chemistry textbooks
would state the general trend for cationic rearrangement reac-
tions: less stable carbocations would rearrange toward the
more stable ones, i.e. primary carbocations would rearrange
toward the secondary ones, and secondary carbocations would
rearrange toward the tertiary ones, through hydride shift, alkyl
shift, or aryl shift.1,2 On the other hand, anti-Markovnikov
rearrangement reactions which convert tertiary carbocations
to secondary ones are also well-known. Indeed, the seminal
work on Wagner–Meerwein rearrangement was intrigued by the
conversion of a-pinene to bornyl chloride, a secondary chloride
rather than the expected tertiary chloride (Scheme 1a).3,4 In this
case, the release of ring strain for the four-membered ring is
apparently the thermodynamic driving force for such abnormal
anti-Markovnikov rearrangement. Another famous example
is Corey’s hypothesis of anti-Markovnikov rearrangement in
the biosynthetic pathway of lanosterol, a precursor for many
steroids including cholesterol, for its C-ring expansion from a
tertiary carbocation to a secondary one (Scheme 1b),5,6 a much
debated puzzle for the past two decades which has led to a
number of experimental and theoretical studies.7–13 However,
Scheme 1 Anti-Markovnikov rearrangement reactions.
to the best of our knowledge, secondary to primary carbocation rearrangement has not been reported in the literature. During
our study on sulfur mediated allylic C–H functionalization
of olefins,14 serendipitous observation of some rearranged
products led us to believe the possibility of such hitherto
State Key Laboratory of Chemical Resource Engineering, Department of Organic
Chemistry, Faculty of Science, Beijing University of Chemical Technology,
Beijing 100029, China. E-mail: dhg@mail.buct.edu.cn, jxxu@mail.buct.edu.cn,
unknown chemistry involving anti-Markovnikov rearrangement
lipf@mail.buct.edu.cn
from secondary carbocations toward primary carbocations or
primary triflates (Scheme 1c).
† Electronic supplementary information (ESI) available: Experimental procedures
and characterization data of new compounds. See DOI: 10.1039/c6cc05128f
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
Table 1 Sulfur mediated allylic C–H amination of allylbenzene with
different amine nucleophilesa
R1 R2 3, yieldb (%) 4, yieldb (%) 5, yieldb (%)
2
Entry
1
2
3
4
5
6
7
i-Pr2NH (2a) i-Pr i-Pr 3aa, 12
Et2NH (2b) Et Et 3ab, 16
n-Pr2NH (2c) n-Pr n-Pr 3ac, 18
4aa, 70
4ab, 47
4ac, 56
4ad, 50
4ae, 26
4af, 50
4ag, 48
—
—
—
i-PrNH2 (2d) i-Pr
BnNH2 (2e) Bn
H
H
3ad, 8c
3ae, 9
5ad, 18c
5ae, 32
5af, 35
5ag, 20c
n-PrNH2 (2f) n-Pr H
t-BuNH2 (2g) t-Bu H
3af, 9
3ag, 14c
a
Reaction conditions: 1 eq. of 1a (0.4 mmol) and 1.2 eq. of Ph2SO in
2 mL of CH2Cl2 was treated with 1.2 eq. of Tf2O at ꢀ78 1C, and then
warmed up to 0 1C, added 5 eq. of 2 in 1 mL of CH2Cl2, and reacted at
Scheme 2 Proposed reaction mechanism involving anti-Markovnikov
rearrangement.
b
room temperature before aqueous quenching. Isolated yields after
repeated flash column chromatography and preparative thin layer
c
chromatography. NMR yield based on an isolated mixture of 3 and 5,
see the ESI for mixture spectra.
major product 4; when R2 = H, primary amine nucleophile 2 could
also undergo a double SN2 type reaction with triflate G to give the
The experiment that initiated this study was very straightforward: azetidine product 5. The variable product distribution shown in
electrophilic addition of allylbenzene (1a) with the sulfur reagent Table 1 should be the consequence of different reaction rates
generated from diphenylsulfoxide and triflic anhydride15–19 followed between these intermediates with different amine substrates.
by nucleophilic substitution with various amine nucleophiles (2)
We believe that the thermodynamic driving force for the
should give allylic C–H amination product 3 with concomitant conversion of secondary carbocation B to primary carbocation F
double bond migration (Table 1). A similar reaction sequence has or primary triflates G could be the result of intramolecular
previously been accomplished in Mukaiyama group’s study electrostatic repulsion in dicationic intermediates.20–25 For
using a-methylstyrene as an olefin substrate.18,19 However, when dicationic intermediate B, the cationic carbon center and the
terminal olefin 1a was examined under these conditions, the cationic sulfur center are separated by only one carbon atom,
originally expected products 3aa to 3ag turned out to be the and thus experience more severe repulsion; on the other hand,
minor products, while the major products were 4aa to 4ag. If for intermediate F, the cationic carbon center and the cationic
primary amine nucleophiles (2d to 2g) were used instead of sulfur center are separated by two carbon atoms, and have less
secondary amine nucleophiles (2a to 2c), azetidine products 5ad severe intramolecular electrostatic repulsion. Anti-Markovnikov
to 5ag were also generated in up to 35% isolated yield. The rearrangement reactions from secondary carbocations toward
formation of abnormal products 4 and 5 indicates that some primary carbocations or primary triflates are thus possible
rearrangement step must be taking place during these reactions. because the secondary carbocations are now the thermodyna-
These observations led us to propose the reaction mechanism mically less stable intermediates.
involving anti-Markovnikov rearrangement as shown in Scheme 2.
As shown in Table 2, we then conducted a survey of various
Reaction of highly electrophilic sulfur reagent A15–19 with allyl- substituted allylbenzenes (1b to 1j) to compare the product
benzene (1a) should follow Markovnikov’s rule, i.e. the terminal ratio of 3 and 4. We expect that the competition between the
position of olefin would be the nucleophilic reacting site, to give rearrangement pathway (from B to F, G or H) and the normal
secondary carbocation B. For the minor reaction pathway, B could reaction pathway (from B to C or D) would be influenced by the
lead to secondary triflate C or allyl sulfonium salt D; after the electronic nature of different aryl migrating groups. Indeed,
addition of amine nucleophile 2, triflate C would also be con- compared with substrates with more electron-rich aryl groups,
verted to D through an elimination reaction, which would then substrates with more electron-deficient aryl groups generally give
undergo an SN2 type reaction at the allylic position to generate the less rearranged products. While the 3 : 4 ratio for allylbenzene
originally expected product 3. On the other hand, for the major substrate 1a is around 1 :4 (Table 2, entry 1), 4-methyl substituted
reaction pathway, secondary carbocation B might rearrange allylbenzene 1b gave a 1: 9 ratio (entry 2). 4-Halogen substituted
through phenonium ion E to give the corresponding primary allylbenzenes (1c, 1d and 1e) gave a ratio of around 1 :3 (entries
carbocation F, which could lead to primary triflate G or allylic 3–5), still favouring the rearranged products 4ca, 4da and 4ea.
sulfonium salt H; alternatively, phenonium ion E could be opened While 4-nitro substituted allylbenzene 1f gave nearly equal
by a triflate anion to directly give primary triflate G. In the amounts of products 3fa and 4fa (entry 6), 4-trifluoromethyl
presence of amine nucleophile 2, elimination of G should substituted allylbenzene 1g gave the rearranged product as a
generate allylic sulfonium salt H, which would then undergo minor product (entry 7). This trend is in good accordance with
an SN2 type reaction at the allylic position to give the observed our proposed carbocation rearrangement reaction mechanism,
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016

View Article Online
ChemComm
Communication
Table 2 Sulfur mediated allylic C–H amination with different aryl migrating
groups
3, yielda (%) 4, yielda (%) 3 : 4 ratiob
1
Entry
1
2
3
4
5
1a
1b 3ba, 9
1c 3ca, 17
1d 3da, 20
3aa, 12
4aa, 70
4ba, 58
4ca, 38
4da, 67
4ea, 50
4fa, 24
18 : 82
10 : 90
26 : 74
27 : 73
27 : 73
46 : 54
1e
1f
3ea, 22
3fa, 20
3ga, 40
6
7
8
Scheme 3 Attempts for tertiary to primary carbocation rearrangement
and secondary to secondary carbocation rearrangement.
1g
4ga, 22
4ha, 43
60 : 40
39 : 61
1h 3ha, 26
9
1i
3ia, 46
4ia, 24
60 : 40
ratio would change to 25 : 75, indicating the reaction rates for
the two competing pathways were influenced differently by the
increase of reaction temperature.
10
1j
3ja, 44
4ja, 4
92 : 8
a
Isolated yields after flash column chromatography. b 1H NMR ratio
We next examined the possibility of tertiary to primary
carbocation rearrangement, and secondary to secondary carbo-
cation rearrangement reactions, which could be driven by
intramolecular electrostatic repulsion in dicationic intermediates.
for the crude reaction mixture, see the ESI for mixture spectra.
since the tendency for aryl shift should decrease with more As shown in Scheme 3, with substrate 1k and amine nucleophile
electron-deficient aryl groups. In addition, the trend (entries 5 2g, only the normal allylic C–H amination products 3kg and 3kg0
and 8–10) for 4-chloro (1e), 3-chloro (1h), 3,5-dichloro (1i), and were formed, indicating the absence of phenyl shift, which would
2-chloro (1j) substituted allylbenzenes seems to indicate that result in a tertiary to primary carbocation rearrangement and give
the steric effect was also important for the rearrangement the corresponding azetidine type substitution product 5kg. In this
to occur: for ortho-substituted substrate 1j, presumably the pheno- case, intramolecular electrostatic repulsion is not enough to
nium type intermediate (similar to structure E in Scheme 2) overcome the energetic cost for anti-Markovnikov rearrangement
would be very much hindered from forming, and thus very little of a relatively more stable tertiary carbocation. In contrast, with
rearranged product 4ja was produced (entry 10); on the other substrate 1l and amine nucleophiles 2a and 2d, only rearranged
hand, para-substituted substrate 1e would have the least steri- allylic C–H amination products 4la and 4ld/4ld0 were formed,
cally hindered migrating group, and thus gave more rearranged indicating the high tendency for a secondary to secondary carbo-
product 4ea (entry 5). Substrate 1h with only one meta-chloro cation rearrangement in these cases, which should decrease
substituent should be more electron-rich and less sterically the intramolecular electrostatic repulsion in the corresponding
hindered than 3,5-dichloro substituted substrate 1i, and thus dicationic intermediates.
the aryl migration product 4ha was formed in a higher ratio
compared with 4ia (entries 8 and 9).
In conclusion, we have observed a new type of anti-
Markovnikov rearrangement reactions converting secondary
For the reaction between 1a and 2a, we also examined the carbocations toward primary carbocations or primary triflates
concentration effect and the temperature effect. If we doubled in sulfur mediated allylic C–H amination of terminal olefins,
or quadrupled the amount of CH2Cl2 solvent used for the first which could be driven by intramolecular electrostatic repulsion
step, which decreased the concentration of substrate 1a from in dicationic intermediates. These reactions also provide a
0.2 M to 0.1 M or 0.05 M, the 3aa : 4aa ratio turned out to be direct method for converting terminal olefins to the corres-
18 : 82 or 19 : 81, which remained virtually unchanged. This ponding allylic amines or azetidines.
suggests that the reaction rates for both competing pathways
This research was supported in part by the National Natural
shown in Scheme 2 changed proportionally by the change Science Foundation of China (21402005, 21372025), the
in the substrate concentration. On the other hand, if triflic National Basic Research Program of China (2013CB328905),
anhydride was added at ꢀ45 1C instead of ꢀ78 1C, the 3aa : 4aa Fundamental Research Funds for the Central Universities
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
12 M. Nishizawa, Y. Asai and H. Imagawa, Org. Lett., 2006, 8,
5793–5796.
13 N. Chen, S. Wang, L. Smentek, B. A. Hess and R. Wu, Angew. Chem.,
Int. Ed., 2015, 54, 8693–8696.
(buctrc201402), and the Public Hatching Platform for Recruited
Talents of Beijing University of Chemical Technology.
14 G. Hu, J. Xu and P. Li, Org. Lett., 2014, 16, 6036–6039.
15 J. B. Hendrickson and S. M. Schwartzman, Tetrahedron Lett., 1975,
16, 273–276.
16 V. G. Nenaidenko and E. S. Balenkova, Russ. J. Org. Chem., 2003, 39,
291–330.
17 B. A. Garcia, J. L. Poole and D. Y. Gin, J. Am. Chem. Soc., 1997, 119,
7597–7598.
18 H. Yamanaka, J. Matsuo, A. Kawana and T. Mukaiyama, Chem. Lett.,
2003, 32, 626–627.
Notes and references
1 F. A. Carey and R. J. Sundberg, Advanced Organic Chemistry, Part A:
Structure and Mechanisms, 4th edn, 2000.
2 M. B. Smith and J. March, March’s Advanced Organic Chemistry:
Reactions, Mechanisms, and Structure, 6th edn, 2007.
3 H. Meerwein and K. Emster, Ber. Dtsch. Chem. Ges. A/B, 1922, 55,
2500–2528.
4 L. Birladeanu, J. Chem. Educ., 2000, 77, 858–863.
5 E. J. Corey, S. C. Virgil, H. Cheng, C. H. Baker, S. P. T. Matsuda, 19 H. Yamanaka, J. Matsuo, A. Kawana and T. Mukaiyama, ARKIVOC,
V. Singh and S. Sarshar, J. Am. Chem. Soc., 1995, 117, 11819–11820.
6 E. J. Corey and H. Cheng, Tetrahedron Lett., 1996, 37, 2709–2712.
7 C. Jenson and W. L. Jorgensen, J. Am. Chem. Soc., 1997, 119,
10846–10854.
8 M. Nishizawa, Y. Iwamoto, H. Takao, H. Imagawa and T. Sugihara,
Org. Lett., 2000, 2, 1685–1687.
2004, iii, 42–65.
20 G. A. Olah, G. K. S. Prakash and T. N. Rawdah, J. Am. Chem. Soc.,
1980, 102, 6127–6130.
21 G. K. S. Prakash, T. N. Rawdah and G. A. Olah, Angew. Chem.,
Int. Ed., 1983, 22, 390–401.
22 D. A. Klumpp, P. S. Beauchamp, G. V. Sanchez, S. Aguirre and
S. Leon, Tetrahedron Lett., 2001, 42, 5821–5823.
9 B. A. Hess, J. Am. Chem. Soc., 2002, 124, 10286–10287.
10 M. Nishizawa, A. Yadav, H. Imagawa and T. Sugihara, Tetrahedron 23 D. A. Klumpp, R. Rendy and A. McElrea, Tetrahedron Lett., 2004, 45,
Lett., 2003, 44, 3867–3870. 7959–7961.
11 M. Nishizawa, A. Yadav, Y. Iwamoto and H. Imagawa, Tetrahedron, 24 A. Li, P. J. Kindelin and D. A. Klumpp, Org. Lett., 2006, 8, 1233–1236.
2004, 60, 9223–9234.
25 D. A. Klumpp, Chem. – Eur. J., 2008, 14, 2004–2015.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016