C-8-selective allylation of quinoline: A case study of β-hydride vs β-hydroxy elimination

Article abstract of DOI:10.1021/acs.orglett.6b01845

An unprecedented C(8)-H bond allylation of quinoline with allyl carbonate and allyl alcohol catalyzed by Cp?Co(III) using a traceless directing group via β-oxygen and β-hydroxy elimination is described. This site-selective allylation reaction proceeds smo

Full text of DOI:10.1021/acs.orglett.6b01845

Letter
pubs.acs.org/OrgLett
C‑8-Selective Allylation of Quinoline: A Case Study of β‑Hydride vs
β‑Hydroxy Elimination
Deepti Kalsi,† Roshayed A. Laskar,† Nagaraju Barsu,† J. Richard Premkumar,‡ and Basker Sundararaju*,†
†Fine Chemical Laboratory, Department of Chemistry, Indian Institute of Technology Kanpur, Kanpur, Uttar Pradesh 208016, India
‡Center for Molecular Modelling, Indian Institute of Chemical Technology, Hyderabad 500007, India
S
* Supporting Information
ABSTRACT: An unprecedented C(8)−H bond allylation of quinoline with allyl carbonate and allyl alcohol catalyzed by
Cp*Co(III) using a traceless directing group via β-oxygen and β-hydroxy elimination is described. This site-selective allylation
reaction proceeds smoothly with various functional group tolerance including quinoxaline and phenanthridine. Under the
nonoxidative reaction conditions, the difference in selectivity between Rh(III) and Co(III), which proceeds through β-hydride
and β-hydroxy elimination using allyl alcohol, is shown for the first time.
ver the past decade, activation of C−H bonds emerged as
conditions, which proceeds via β-hydride elimination path-
Oan important tool for atom- and step-economic syn- way.14 This distinctive behavior of Co(III) and Rh(III) raised
thesis.1 In particular, Cp*Rh(III) has attained its pinnacle over
our curiosity to explore further reactions which can answer a
few more queries. (1) Is β-hydroxy elimination substrate
dependent or catalyst dependent? (2) If it is catalyst (metal)
dependent, why did Rh(III) gave poor yield compared to
Co(III) (for example, see ref 13c)? (3) Do properties of group
9 elements (Co, Rh, Ir) such as atom size, electronic
configuration, nucleophilicity, and oxophilicity have any effect
in elimination pathway? (4) Can we obtain β-hydroxy and β-
hydride elimination products selectively with allyl alcohol using
two different metals from group 9 under the same reaction
conditions (preferably nonoxidative conditions)?
To gain perspicuity and understanding to answer the above
questions, we decided to scrutinize our results on allylation
with simultaneous use of Co(III) and Rh(III) employing allyl
alcohol as a coupling partner. Owing to the higher activity
exhibited against hepatitis by 8-allylquinoline and its lack of
presence in the literature for direct synthesis of this novel
compound,15a we studied the sequential allylation/alkene
metathesis catalysis and selective linkage of functionalized
alkenes for further applications.15b These multifold advantages
encouraged us to choose quinoline N-oxide as a traceless
directing group for remote C−H bond allylation. In order to
establish the reactivity of Cp*Co(III) for C−H bond allylation
with quinoline N-oxide, we have chosen allyl carbonate as the
coupling partner, Cp*Co(III) as a catalyst, and carboxylate as
an additive in trifluoroethanol as solvent. To determine the best
conditions for the reaction, various catalysts, additives, and
temperatures were screened thoroughly, and the optimization
the past few years for a vast number of synthetic trans-
formations through C−H bond functionalization.2,3 Due to its
meager resources and its extortionate prices, however, we need
an alternative for bringing our research to a more pragmatic or
industrial level. In 2013, Kanai and Matsunaga showed
Cp*Co(III), which is not only an alternate catalyst for its
noble counterpart [Rh(III)] but also unique in its reactivity and
selectivity.4 However, often the requirement of strong chelating
groups such as pyridine and its analogues disfavors its
expediency in practical applications.5 Glorius,6 Chang,7
Matsunaga,8 Ackermann,9 Sundararaju,10 and others11 circum-
vented this problem recently in a few cases, but the potential
for exploiting the unique reactivity of Cp*Co(III) is still great
when weakly chelating/traceless directing groups are used.
At the outset of our continuous efforts to examine new
reactivity of Cp*Co(III) which includes oxidant-free annula-
tion,10d sequential C−H activation/oxygen atom transfer,10b
C(sp3)−H bond alkenation,10a and C−H bond allylation using
allylic substrate including allyl alcohol are reported in this
study. Direct allylation of arenes and heteroarenes are of
significant value due to its efficacious modification of olefins
upon allylation. Though Cp*Co(III)-catalyzed allylation of C−
H bonds was demonstrated by Glorius under mild conditions,
this required a strong chelating group such as pyridine or its
analogues and activated allylic substrate such as carbonate or
acetate.12 Later, Kanai/Matsunaga and Glorius showed that
activation of allyl alcohol with other directing groups including
amide proceeds via a β-hydroxy elimination pathway using
Co(III) as a catalyst.13 It was in contrast with allylation shown
by Glorius and others with Rh(III) and Ru(II) under oxidative
Received: June 25, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b01845
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
results are shown in Table S1 (see the Supporting
Information).
51% yield (Scheme 2). This initial result probed us to
investigate the allylation with allylic alcohol using Co(III).
The desired product 3aa was obtained in 66% yield using
Cp*Co(CO)I2 (A) (5 mol %) along with NaOPiv (30 mol %)
as an additive after a brief screening of metal carboxylates at 80
°C (Table S1, entry 1−6). Neither changing the solvent nor
increasing the temperature provides augmentation in the
product formation (entry 7). Next, we investigated the effect
of ionization by the addition of various silver salts. Silver triflate
gave allylated product 3aa in 67% yield, whereas silver
hexafluoroantimonate gave only 35% yield (entries 8 and 9).
Further exploration of different cobalt catalysts including
preformed cationic cobalt complex (C) revealed that
[Cp*CoI2]2 (B)13d provides the best yield at 100 °C for 24
h (entry 10−12). In addition, our controlled experiments
showed that all three components are necessary to obtain 3aa
in high yield (entryies 13 and 14). Additional experiments were
carried out with Rh(III) and Ir(III); while the former gave 73%
yield of 3aa, the latter did not give any product (entries 15 and
16).
Scheme 2. C(8)−H Bond Allylation of 1a with Allyl Alcohol
The scope of allylation with cobalt(III) was carried out using
α-vinylbenzyl alcohol as the allylic substrate and quinoline N-
oxide as a traceless directing group (Table 1). Methyl
Table 1. Allylation vs β-Aryl Ketone Catalyzed by Co(III)
a
and Rh(III)
With optimized conditions in hand, we further explored the
scope of quinoline N-oxide (Scheme 1). Modest yields were
Scheme 1. Scope of Quinoline N-Oxide
obtained when substrates changed from electron-rich to
electron-poor quinoline N-oxide (3ba−ha). Various substitu-
ents at the C-3 position of 1a such as Ph, electron-poor, and
electron-rich olefins resulted in moderate yield (3ia−la). A
wide range of substrates having functional groups that are
known to be sensitive such as nitro and ketone groups were
tolerable under our reaction conditions (3ma,ha). Analogues of
quinoline N-oxides such as quinoxaline gave the allylation
product in good yield, whereas acridine gave exclusively the bis-
allylation product albeit in low yield.
a
Next, we investigated allyl alcohol as a substrate with 1a
under standard conditions. To our delight, we obtained 3aa in
Numbers in parentheses are the ratio of E and Z isomers; Co(III) =
[Cp*CoI2]2; Rh(III) = [Cp*RhCl2]2.
B
DOI: 10.1021/acs.orglett.6b01845
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
substituents at various positions of quinoline (C-2, C-4, and C-
6) gave optimum yields with linear selectivity (3pa−sa). The
reaction proceeded smoothly even in the presence of sensitive
but useful groups such as Br- and CO2Me-, which furnished
moderate reaction results (3ta−ua). Other heterocycles such as
quinoxaline also accomplished the reaction (3va) with 37%
yield. We exposed other aryl-substituted α-vinyl allyl alcohols
under the same reaction conditions with Co(III), which
progressed placidly (3wa). While the manuscript was in
preparation, Matsunaga reported C−H bond allylation of
amide starting from allylic alcohols using Co(III).13a
Scheme 3. Proposed Mechanism
Subsequently, we carried out the allylation with Rh(III) to
compare the reactivity and selectivity we obtained with Co(III)
(Table 1). As per our expectations, allylation with α-vinylbenzyl
alcohol using Rh(III) under our optimized conditions gave β-
aryl ketone rather than 3aa, which is confirmed through NMR
and MS analysis. To prove the generality of the reaction, it was
performed with various substrates, where we obtained the
product with excellent selectivity under the same reaction
conditions (Table 1, 4a−h). It is important to note that similar
compounds were obtained by Glorius and others using allyl
alcohol under oxidative conditions.14a Kim showed recently
that reaction could be performed with a substoichiometric
amount of Cu(II), but the reaction proceeded only with α-
alkyl-substituted allyl alcohol,14c which compliments to our
presented results. To prove the concept, we chose N-
pyrimidinylindole as the directing group and performed the
reaction under our optimized conditions with Rh(III) and
Co(III), and to our delight, we observed similar results. With
Rh(III), we obtained 48% isolated yield of allylated product and
β-aryl ketone product 35% yield, respectively.16
Scheme 4. Diversification of 3aa
Based on the previous results of Glorius and Kanai/
Matsunaga,13 it was believed that β-hydroxy or β-oxygen
elimination were facile with Co(III) compared with Rh(III),
and the density functional theory results obtained by Kanai/
Matsunaga et al. suggest that the energy required for β-hydroxy
elimination is comparatively lower with Co(III) than it requires
for Rh(III). We looked at the transition states leading to β-
hydroxide elimination and β-hydride elimination products for
[Cp*RhIII] catalyst as shown in Figure S1.16 The transition
state 5TSβ‑H is 5.0 kcal/mol lower in energy compared to
2TSβ‑OH. The DFT results clarify the experimental observations
as [Cp*RhIII] favors the β-hydride elimination pathway.16,17
Based on previous reports on allylation with Rh(III) and
Co(III),13,14 we propose the possible mechanism in Scheme 3.
Initial ionization in the presence of metal carboxylate led to
intermediate D. Cationic complex D will then be coordinated
to the substrate 1a through oxygen, and subsequent
deprotonation led to the cyclometalated complex F. This will
further coordinate with allylic substrate via olefin coordination
followed by 1,2-insertion, leading to intermediate H. Inter-
mediate H will undergo β-hydride or β-hydroxy elimination
depends on the metal we employ. For example, Co(III)
undergoes β-hydroxy/oxygen elimination to lead to allylated
product, whereas Rh(III) undergoes β-hydride elimination over
β-hydroxy elimination to lead to β-aryl ketone, which was
further supported by density functional theory calculations
(Figure S1).16
incorporated at the C-2 position of quinoline using cobalt-
catalyzed sequential addition/oxygen transfer with BuNC. A
t
similar transformation was observed with Ag(I) salt as catalyst,
but it does not give satisfactory results with 3aa. However, we
were pleased to find better product formation of 6 with
cobalt(III) as mentioned in Scheme 4.18 C-2-selective
amidation and quinolone were also obtained in good yield.
In conclusion, we have demonstrated unprecedented
Cp*Co(III)-catalyzed remote C(8)−H bond allylation of
quinoline using N-oxide as a traceless directing group and
allyl carbonate as a coupling partner. An oxidant-free, good
functional group tolerance allylation of C−H bonds was
achieved under mild conditions. Diverse selectivity was
obtained with Rh(III) and Co(III) via β-hydride and β-hydroxy
elimination. This difference in selectivity was shown for the first
time under nonoxidative conditions and further supported by
DFT calculations.
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.6b01845.
Quinoline N-oxide is a versatile substrate, and it acts as a
traceless directing group that allows further diversification at
late stages by keeping the allyl fragment intact as depicted in
Scheme 4. Selective removal of oxygen mediated by PCl3 was
achieved in 60% yield. N-tert-Butylformyl groups were
Experimental methods, optimization of reaction con-
ditions, and other supplementary data (PDF)
C
DOI: 10.1021/acs.orglett.6b01845
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(10) (a) Sen, M.; Emayavaramban, B.; Barsu, N.; Premkumar, J. R.;
Sundararaju, B. ACS Catal. 2016, 6, 2792−2796. (b) Barsu, N.; Sen,
M.; Sundararaju, B. Chem. Commun. 2016, 52, 1338−1341. (c) Barsu,
N.; Kalsi, D.; Sundararaju, B. Chem. - Eur. J. 2015, 21, 9364−9368.
(d) Sen, M.; Kalsi, D.; Sundararaju, B. Chem. - Eur. J. 2015, 21,
15529−15533. (e) Kalsi, D.; Sundararaju, B. Org. Lett. 2015, 17,
6118−6121.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: basker@iitk.ac.in.
Notes
The authors declare no competing financial interest.
(11) (a) Muralirajan, K.; Kuppusamy, R.; Prakash, S.; Cheng, C.-H.
Adv. Synth. Catal. 2016, 358, 774−783. (b) Kong, L.; Yu, S.; Zhou, X.;
Li, X. Org. Lett. 2016, 18, 588−591. (c) Sivakumar, G.; Vijeta, A.;
Jeganmohan, M. Chem. - Eur. J. 2016, 22, 5899−5903. (d) Lade, M.
D.; Pawar, A. B. Org. Chem. Front. 2016, 3, 836. (e) Boerth, J. A.;
Hummel, J. A.; Ellman, J. A. Angew. Chem. 2016, DOI: 10.1002/
ange.201603831.
ACKNOWLEDGMENTS
■
Financial support provided by SERB (EMR/2016/000136) to
carryout this research is gratefully acknowledged. D.K. and N.B.
thank IITK and CSIR for their fellowships.
́ ́
(12) (a) Yu, D.-G.; Gensch, T.; de Azambuja, F.; Vasquez-Cespedes,
REFERENCES
■
S.; Glorius, F. J. Am. Chem. Soc. 2014, 136, 17722. Also see:
(b) Moselage, M.; Sauermann, N.; Koeller, J.; Liu, W.; Gelman, D.
Synlett 2015, 26, 1596−1600. Also see with Ru(II): Kumar, G. S.;
Kapur, M. Org. Lett. 2016, 18, 1112.
(1) Li, J. J. C−H Bond Activation in Organic Synthesis; CRC Press,
2015; pp 1−327.
(2) Dixneuf, P. H.; Doucet, H. C−H Bond Activation and Catalytic
Functionalization I; Springer, 2016; pp 1−76.
(13) (a) Bunno, Y.; Murakami, N.; Suzuki, Y.; Kanai, M.; Matsunaga,
(3) Selected recent reviews on C−H bond functionalization:
(a) Song, G.; Li, X. Acc. Chem. Res. 2015, 48, 1007. (b) Ye, B.;
Cramer, N. Acc. Chem. Res. 2015, 48, 1308. (c) Patureau, F. W.;
Wencel-Delord, J.; Glorius, F. Aldrichimica Acta 2012, 45, 31−41.
(d) Davies, H. M. L.; Du Bois, J.; Yu, J.-Q. Chem. Soc. Rev. 2011, 40,
1855−1856. (e) Satoh, T.; Miura, M. Chem. - Eur. J. 2010, 16, 11212−
11222. (f) Colby, D. A.; Bergman, R. G.; Ellman, J. A. Chem. Rev.
2010, 110, 624−655.
́ ́
S. Org. Lett. 2016, 18, 2216−2219. (b) Gensch, T.; Vasquez-Cespedes,
S.; Yu, D. − G.; Glorius, F. Org. Lett. 2015, 17, 3714−3717. (c) Suzuki,
Y.; Sun, B.; Sakata, K.; Yoshino, T.; Matsunaga, S.; Kanai, M. Angew.
Chem., Int. Ed. 2015, 54, 9944−9947. (d) Sun, B.; Yoshino, T.;
Matsunaga, S.; Kanai, M. Chem. Commun. 2015, 51, 4659−4661.
(14) Under oxidative conditions using Cp*Rh(III) favors β-hydride
elimination, see: (a) Shi, Z.; Boultadakis-Arapinis, M.; Glorius, F.
Chem. Commun. 2013, 49, 6489−6491. (b) Huang, L.; Wang, Q.; Qi,
J.; Wu, X.; Huang, K.; Jiang, H. Chem. Sci. 2013, 4, 2665−2669.
(c) Han, S. H.; Choi, M.; Jeong, T.; Sharma, S.; Mishra, N. K.; Park, J.;
Oh, J. S.; Kim, W. J.; Lee, J. S.; Kim, I. S. J. Org. Chem. 2015, 80,
11092−11099.
(4) Yoshino, T.; Ikemoto, H.; Matsunaga, S.; Kanai, M. Angew.
Chem., Int. Ed. 2013, 52, 2207−2211.
(5) Selected examples on Cp*Co(III)-catalyzed C−H bond
functionalization using pyridine or its analogues as directing group:
(a) Kim, J. H.; Greßies, S.; Glorius, F. Angew. Chem., Int. Ed. 2016, 55,
5577−5581. (b) Barsu, N.; Atiur Rahman, Md.; Sen, M.; Sundararaju,
B. Chem. - Eur. J. 2016, 22, 9135. (c) Prakash, S.; Muralirajan, K.;
Cheng, C. − H. Angew. Chem., Int. Ed. 2016, 55, 1844. (d) Li, J.;
Ackermann, L. Angew. Chem., Int. Ed. 2015, 54, 3635−3638. (e) Li, J.;
Ackermann, L. Angew. Chem., Int. Ed. 2015, 54, 8551−8554.
(15) (a) Okamoto, T.; Kobayashi, T.; Yoshida, S. Med. Chem. 2007,
3, 35−44. (b) Bilel, H.; Hamdi, N.; Zagrouba, F.; Fischmeister, C.;
Bruneau, C. Catal. Sci. Technol. 2014, 4, 2064−2071.
(16) See the Supporting Information for more details.
(17) Dai, H.; Yu, C.; Wang, Z.; Yan, H.; Lu, C. Org. Lett. 2016, 18,
3410.
(18) See also a similar transformation with Ag(I): Chen, Z.; Yu, X.;
Su, M.; Yang, X.; Wu, J. Adv. Synth. Catal. 2009, 351, 2702.
́
(f) Sauermann, N.; Gonzalez, M.; Ackermann, L. Org. Lett. 2015,
17, 5316−5319. (g) Zhao, D.; Kim, J. H.; Stegemann, L.; Strassert, C.
A.; Glorius, F. Angew. Chem., Int. Ed. 2015, 54, 4508−4511. (h) Patel,
P.; Chang, S. ACS Catal. 2015, 5, 853−858. (i) Pawar, A. B.; Chang, S.
Org. Lett. 2015, 17, 660−663. (j) Liang, Y.; Liang, Y. − F.; Tang, C.;
Yuan, Y.; Jiao, N. Chem. - Eur. J. 2015, 21, 16395−16399. (k) Hummel,
J. R.; Ellman, J. A. J. Am. Chem. Soc. 2015, 137, 490−498. (l) Hummel,
J. R.; Ellman, J. A. Org. Lett. 2015, 17, 2400−2403. (m) Ozkal, E.;
Cacherat, B.; Morandi, B. ACS Catal. 2015, 5, 6458−6462. (n) Liu, X.
− G.; Zhang, S. − S.; Wu, J. − Q.; Li, Q.; Wang, H. Tetrahedron Lett.
2015, 56, 4093−4095. (o) Zhang, Z. − Z.; Liu, B.; Wang, C. − Y.; Shi,
B. − F. Org. Lett. 2015, 17, 4094−4097. (p) Sun, B.; Yoshino, T.;
Matsunaga, S.; Kanai, M. Adv. Synth. Catal. 2014, 356, 1491−1495.
(q) Grigorjeva, L.; Daugulis, O. Angew. Chem., Int. Ed. 2014, 53,
10209−10212.
(6) (a) Lerchen, A.; Vas
Int. Ed. 2016, 55, 3208−3211. (b) Kim, J. H.; Greßies, S.; Glorius, F.
Angew. Chem., Int. Ed. 2016, 55, 5577−5581. (c) Lu, Q.; Vasquez-
Cespedes, S.; Gensch, T.; Glorius, F. ACS Catal. 2016, 6, 2352−2356.
́ ́
quez-Cespedes, S.; Glorius, F. Angew. Chem.,
́
́
(7) (a) Park, J.; Chang, S. Angew. Chem., Int. Ed. 2015, 54, 14103−
14107. (b) Figg, T. M.; Park, S.; Park, J.; Chang, S.; Musav, D. G.
Organometallics 2014, 33, 4076−4085.
(8) (a) Sun, B.; Yoshino, T.; Kanai, M.; Matsunaga, S. Angew. Chem.,
Int. Ed. 2015, 54, 12968−12972. (b) Suzuki, Y.; Sun, B.; Yoshino, T.;
Matsunaga, S.; Kanai, M. Tetrahedron 2015, 71, 4552−4556.
(c) Ikemoto, H.; Yoshino, T.; Sakata, K.; Matsunaga, S.; Kanai, M. J.
Am. Chem. Soc. 2014, 136, 5424−5431.
́
(9) (a) Wang, H.; Moselage, M.; Gonzalez, M. J.; Ackermann, L. ACS
Catal. 2016, 6, 2705−2709. (b) Li, J.; Tang, M.; Zang, L.; Zhang, L.;
Zhang, Z.; Ackermann, L. Org. Lett. 2016, 18, 2742−2745. (c) Wang,
H.; Koeller, J.; Liu, W.; Ackermann, L. Chem. - Eur. J. 2015, 21,
15525−15528.
D
DOI: 10.1021/acs.orglett.6b01845
Org. Lett. XXXX, XXX, XXX−XXX