Br?nsted Acid-Catalyzed Intramolecular Nucleophilic Substitution of the Hydroxyl Group in Stereogenic Alcohols with Chirality Transfer

Article abstract of DOI:10.1021/jacs.5b02013

The hydroxyl group of enantioenriched benzyl, propargyl, allyl, and alkyl alcohols has been intramolecularly displaced by uncharged O-, N-, and S-centered nucleophiles to yield enantioenriched tetrahydrofuran, pyrrolidine, and tetrahydrothiophene derivatives with phosphinic acid catalysis. The five-membered heterocyclic products are generated in good to excellent yields, with high degree of chirality transfer, and water as the only side-product. Racemization experiments show that phosphinic acid does not promote S_N1 reactivity. Density functional theory calculations corroborate a reaction pathway where the phosphinic acid operates as a bifunctional catalyst in the intramolecular substitution reaction. In this mechanism, the acidic proton of the phosphinic acid protonates the hydroxyl group, enhancing the leaving group ability. Simultaneously, the oxo group of phosphinic acid operates as a base abstracting the nucleophilic proton and thus enhancing the nucleophilicity. This reaction will open up new atom efficient techniques that enable alcohols to be used as nucleofuges in substitution reactions in the future.

Full text of DOI:10.1021/jacs.5b02013

Communication
pubs.acs.org/JACS
Brønsted Acid-Catalyzed Intramolecular Nucleophilic Substitution of
the Hydroxyl Group in Stereogenic Alcohols with Chirality Transfer
Anon Bunrit,† Christian Dahlstrand,† Sandra K. Olsson,† Pemikar Srifa,† Genping Huang,‡
Andreas Orthaber,§ Per J. R. Sjoberg,† Srijit Biswas,*,† Fahmi Himo,*,‡ and Joseph S. M. Samec*,†
̈
†Department of Chemistry, BMC, Uppsala University, Box 576, 75123 Uppsala, Sweden
‡Department of Organic Chemistry, Stockholm University, 106 91 Stockholm, Sweden
§Department of Chemistry, Ångstrom Laboratories, Uppsala University, Box 523, 75120 Uppsala, Sweden
̈
S
* Supporting Information
Scheme 1. Current Substitution Methodologies Based on
Catalysis Proceed via Achiral Intermediates to Yield Racemic
Products
ABSTRACT: The hydroxyl group of enantioenriched
benzyl, propargyl, allyl, and alkyl alcohols has been
intramolecularly displaced by uncharged O-, N-, and S-
centered nucleophiles to yield enantioenriched tetrahy-
drofuran, pyrrolidine, and tetrahydrothiophene derivatives
with phosphinic acid catalysis. The five-membered
heterocyclic products are generated in good to excellent
yields, with high degree of chirality transfer, and water as
the only side-product. Racemization experiments show
that phosphinic acid does not promote SN1 reactivity.
Density functional theory calculations corroborate a
reaction pathway where the phosphinic acid operates as
a bifunctional catalyst in the intramolecular substitution
reaction. In this mechanism, the acidic proton of the
phosphinic acid protonates the hydroxyl group, enhancing
the leaving group ability. Simultaneously, the oxo group of
phosphinic acid operates as a base abstracting the
nucleophilic proton and thus enhancing the nucleophil-
icity. This reaction will open up new atom efficient
techniques that enable alcohols to be used as nucleofuges
in substitution reactions in the future.
Recently, Aponick, Widenhoefer, and Uenishi demonstrated
chirality transfer in gold- and palladium-catalyzed intramolecular
amination and etherification reactions of stereogenic allylic
alcohols that proceeded through an SN2′-type transition state
(Scheme 2).7 A mechanistic study by density functional theory
Scheme 2. Gold- and Palladium-Catalyzed SN2′-Type Allylic
Amination and Etherification
unctional group interconversion (FGI) of nonderivatized
Fstereogenic alcohols by direct substitution of the hydroxyl
group with chirality transfer is a major challenge in C-heteroatom
bond-forming reactions.1 Traditional methodologies include:
use of stoichiometric and harmful reagents, multistep reaction
sequences, and purification issues, all giving low atom economy
and high environmental factors.2,3 Therefore, the FGI of
nonderivatized alcohols was recently voted as the second most
desired reaction that pharmaceutical companies wanted a
greener version of.4 Efficient methodologies to substitute the
hydroxyl group without racemizing the stereogenic center during
a substitution reaction are important for the development of
techniques to utilize biomass as starting material in organic
synthesis.5 There has been a significant advancement in the direct
substitution of the hydroxyl group of alcohols by Lewis and
Brønsted acid catalysis (Scheme 1A).1 However, these reactions
proceed via an SN1-type pathway in which the enantioenrich-
ment of the substrate is lost.2 Another methodology is the
elimination−addition that also suffers from the formation of an
achiral carbonyl intermediate (Scheme 1B).6
(DFT) calculations suggested a unique transition-state stabiliza-
tion through intramolecular hydrogen bonding to achieve high
degree of chirality transfer (Scheme 2A).8 No reactivity was
observed when the position of the double bond was changed to
the other vicinal side of the hydroxyl group (Scheme 1B). Thus,
the reported methodologies are limited to allyl alcohols that can
form the bicyclic transition state (Scheme 2A).9 To our
knowledge, the direct substitution of a hydroxyl group in
nonallylic stereogenic alcohols with chirality transfer has not yet
been reported and motivated further study.10
Received: February 24, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b02013
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
We herein report a phosphinic acid-catalyzed intramolecular
substitution of the hydroxyl group of enantioenriched aryl, allyl,
propargyl, and alkyl alcohols by O-, N-, and S-centered
nucleophiles with good to excellent chirality transfer to generate
enantioenriched five-membered tetradydrofuran, pyrrolidine,
and tetrahydrothiophene derivatives with inversion of config-
uration (Scheme 3).
based on iron(III), gold(I), or gold(III) previously reported to
activate the hydroxyl group in nucleophilic substitution of
alcohols generated 2a in high yields but with poor chirality
transfers (Table 1, entries 10−12).13 The phosphinic acid used in
this study was a 50% aqueous solution. The other catalysts used
were dry. Therefore, the effect of water was investigated for both
phosphinic acid and p-TSA. It was found that neither the initial
rate nor the chirality transfer was affected by the water
concentration (SI).
Scheme 3. Phosphinic Acid-Catalyzed SN2-Type
Intramolecular Substitution of the Hydroxyl Group in
Alcohols Proceeds with Chirality Transfer
O-, N-, and S-centered nucleophiles were utilized in the
intramolecular substitution of different enantioenriched benzylic
alcohols (Table 2). Substrates with N- and S-centered
a b
,
Table 2. Intramolecular Substitution of Benzylic Alcohols
The weakly nucleophilic O-centered (S)-1-phenylbutane-1,4-
diol (1a) was chosen as the model substrate for screening the
chirality transfer in the intramolecular substitution reaction to
yield (R)-2-phenyltetrahydrofuran (2a) (Table 1). Performing
a
Table 1. Optimization of Reaction Conditions
T
t
yield
b
chirality transfer
c
entry
catalyst
H3PO2
pKa14
(°C) (h)
(%)
(%)
a
1
2
1.07
1.86
2.0
2.12
2.3
-2.8
-12
4.76
-3
80
80
80
80
80
40
40
80
40
40
80
40
12
48
12
48
48
12
12
48
12
12
12
12
99
0
91
0
Reaction Condition: 1 (1 mmol), DCE (2 mL) and catalyst (10 mol
b
c
%) was heated in an oil bath. Isolated yields. Reaction temp. 65 °C;
PMP = para-methoxyphenyl. Reaction temp. 50 °C, time 24 h.
PhH2PO2
H3PO3
H3PO4
Ph2HPO2
p-TSA
d
3
0
0
4
0
0
d
nucleophiles generated 1,2-diphenylpyrrolidine (2b) and 2-
phenyltetrahydrothiophene (2c) in excellent yields with 96% and
91% chirality transfer, respectively (Table 2, entries 2 and 3).
Derivatives of the products 2b and 2c were crystallized, and the
absolute configurations were determined by X-ray crystallog-
raphy showing R-configurations of 2b and 2c (Figure 1). This
5
0
0
6
98
91
0
84
23
0
7
TfOH
8
AcOH
9
H2SO4
FeCl3
92
96
88
96
0
10
11
12
−
−
−
0
AuCl
0
NaAuCl4
19
a
Reaction Condition: 1a (1 mmol), solvent (2 mL), and catalyst (10
mol %) was heated in an oil bath. NMR Yield. By chiral HPLC
analysis. When the reaction was performed at 120 °C, poor reactivity
b
c
d
was observed (SI).
the reaction using 10 mol % phosphinic acid in 1,2-dichloro-
ethane generated 2a in a quantitative yield with 91% chirality
transfer (Table 1, entry 1). The reaction is sensitive to the
reaction temperature where the chirality transfer drops to 9%
when the reaction is performed at 110 °C (SI). Other solvents
were inferior with respect to reactivity and chirality transfer (SI).
Similar phosphorus acids were also evaluated in which no
formation of 2a was observed even after prolonged reaction
times (Table 1, entries 2−5).11 Other Brønsted acids such as p-
toluenesulfonic acid (p-TSA), trifluoromethanesulfonic acid
(TfOH), acetic acid (AcOH), and sulfuric acid (H2SO4) having
a OX−OH structural motif similar to phosphinic acid were
also studied and found to be inferior in terms of products’ yields
and chirality transfers (Table 1, entries 6−9).12 There is no clear
correlation between yield, ee, and pKa. Lewis acidic catalysts
Figure 1. X-ray structures of the HCl salt of 2b and the corresponding
sulfone of 2c showing the absolute configurations (R) ORTEP plots
with thermal ellipsoids drawn at the 50% probability level.
demonstrates that an inversion of configuration at the chiral
center has occurred under the substitution reactions. Substrates
with electron-withdrawing p-fluoro and p-trifluoromethyl
substituents in the benzyl group generated products 2d−2g in
good to excellent yields and with high degrees of chirality
transfers (Table 2, entries 4−7). A decrease in chirality transfer
was observed for 1h with an electron-donating p-methoxy group
in the phenyl-ring (Table 2, entry 8). Excellent chirality transfer
could be obtained also for a substrate with an electron-rich benzyl
group by exchanging the nucleophile from aniline to the stronger
B
DOI: 10.1021/jacs.5b02013
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
nucleophilic p-anisidine (Table 2, entry 9). The 1,4 relation
between nucleofuge and nucleophile is pivotal for reactivity.
Phosphinic acid did not promote the substitution reaction with
substrates that generate three-, four-, or six-membered
heterocyclic compounds or in intermolecular substitutions
(vide infra).
To expand the substrate scope of the phosphinic acid-
catalyzed substitution with chirality transfer, more challenging
nonbenzylic alcohols were screened (Table 3). Alcohol 1j
reactivity to yield 2m under these reaction conditions. Using the
stronger acidic p-TSA gave full conversion to the symmetrical
and fully racemized 2m (Table 4, entry 2), which is produced via
an SN1 reaction pathway. Using TfOH as catalyst gave full
conversion of 1m to yield racemized 2m and side-products at
40 °C. Also, iron(III) gave full conversion to racemic 2m. These
experiments demonstrate that the activation of phosphinic acid is
unique where no SN1-reactivity is observed under the reaction
conditions employed. They furthermore show that a favored five-
membered ring formation is pivotal for the substitution reaction
to proceed, where no intermolecular substitution reaction of 1m
to generate the symmetrical ether 2m occurred in the presence of
phosphinic acid. The racemization observed with the stronger
Brønsted acids, such as p-TSA, could be explained by that the
stronger acids promote a more SN1-like transition state (TS).
This is also in accordance with the results in which p-TSA and
TfOH gave lower chirality transfer than phosphinic acid in the
transformation of 1a to 2a (Table 1, entries 6 and 7). To get
further insight into the reaction mechanism, a rate-order
determination in catalyst was performed. The reaction follows
first-order dependence in catalyst concentration (SI).
Table 3. Intramolecular Substitution of Allyl, Propargyl, And
a b
,
Alkyl Alcohols
a
Reaction Condition: 1 (1 mmol), DCE (2 mL) and catalyst (10 mol
The experiments disclosed above suggest that the phosphinic
acid promotes intramolecular substitution reactions that proceed
through a unique reaction mechanism. Depending on the
electronic properties of both the nucleophile and the nucleofuge,
the intramolecular reactions proceed with 80−100% chirality
transfer (Tables 2 and 3). Furthermore, negligible reactivity was
observed in the intermolecular reaction (Table 4). Hence, from
this it can be concluded that phosphinic acid does not promote
an SN1 reaction. Plausibly, the catalyst can act in a bifunctional
Brønsted acid/Brønsted base fashion to activate both the
nucleophile and the leaving group in an SN2-type reaction.
This would be analogous to the bifunctional role of the
phosphoric acid catalyst demonstrated in a number of organo-
catalytic reactions.11,16 To investigate the feasibility of this kind
of mechanism and to characterize the TS involved, we used DFT
methodology to study the transformations from 1a−c to 2a−c.17
The optimized structures of the TSs are given in Figure 2, along
with the calculated barriers and reaction free energies. In these
TSs, the catalyst has a bifunctional role in which the acidic proton
of the phosphinic acid protonates the oxygen of the leaving
hydroxyl group to promote the C−O bond cleavage.
Concomitantly, the oxo-group of the catalyst deprotonates the
nucleophile, thus increasing the nucleophilicity. This results in an
SN2-type reaction mechanism with an inversion of configuration
at the carbon center, in accordance with the experimental
findings. Indeed, the TSs were found to have plausible energies.
The barriers for the O-, N- and S-centered nucleophiles (i.e.,
transformations 1a → 2a, 1b → 2b and 1c → 2c, respectively)
were calculated to be 27.3, 23.1, and 26.8 kcal/mol, respectively.
These values are consistent with the fact that the N-centered
nucleophiles were reactive at 60 °C, but not the S- or O-centered
nucleophiles, for which the reactions proceed only at 80 °C. For
comparison, we have also calculated the TSs for the O-centered
nucleophile using p-TSA, acetic acid, and diphenyl phosphinic
acid as catalysts. The results are given in the SI.
b
%) was heated in an oil bath. Isolated yields.
activated with an allyl group generated 2j in a 85% yield and 93%
chirality transfer (Table 3, entry 1). It is noteworthy that this class
of allyl alcohols has not previously been transformed in
intramolecular substitution reactions with chirality transfer
(Scheme 2B).5,10 Less reactive propargyl alcohol 1k,1 generated
enantioenriched 2-(phenylethynyl)tetrahydrothiophene 2k in a
92% yield and 100% chirality transfer (Table 3, entry 2). This is,
to our knowledge, the only example of a substitution reaction of a
propargyl alcohol with chirality transfer.1f Secondary aliphatic
alcohols were anticipated to be most challenging due to poor
reactivity and the risk for competing elimination reactions. To
date, only a handful of reports are available for direct nucleophilic
substitution of secondary aliphatic alcohol, and these give
racemic mixtures.6,15 Gratifyingly, transformation of aliphatic
alcohol 1l generated the product 2-methyl-1-phenylpyrrolidine
(2l) in a 81% yield and 100% chirality transfer (Table 3, entry 3).
To get a better insight into the unique activation of the
hydroxyl group by phosphinic acid and to compare to other acids,
racemization experiments were performed. Enantioenriched 1-
phenylethanol (1m) was exposed to substitution reaction
conditions with different catalysts (Table 4). Using phosphinic
acid as catalyst, negligible racemization of 1m and trace amount
formation of symmetrical ether 2m were observed (Table 4,
entry 1). This shows that phosphinic acid does not promote SN1-
a
Table 4. Racemization Experiment of 1m
entry
catalyst
2m
T (°C)
recovered 1m
1
2
3
4
H3PO2
p-TSA
TfOH
FeCl3
<5%
<95%
>20%
>95%
80
40
40
40
>95% (S/R = 85:15)
A Brønsted acid-catalyzed intramolecular reaction that
promotes a direct substitution of the hydroxyl group in
enantioenriched alcohols in good to excellent yields with high
degree of chirality transfer has been developed. The hydroxyl
group of aryl, allyl, propargyl, and alkyl alcohols was substituted
by O-, S-, and N-centered nucleophiles to generate enantiomeri-
cally enriched five-membered heterocyclic compounds. DFT
0%
0%
0%
a
Reaction Condition: 1 (1 mmol), DCE (2 mL), and catalyst (10 mol
1
%) were heated in an oil bath. Yields were determined by H NMR,
and enantioenrichment by chiral HPLC.
C
DOI: 10.1021/jacs.5b02013
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Figure 2. Optimized TS structures and calculated energies (kcal/mol) for the intramolecular direct substitution utilizing O, N, and S nucleophiles.
(8) (a) Ghebreghiorgis, T.; Biannic, B.; Kirk, B. H.; Ess, D. H.;
Aponick, A. J. Am. Chem. Soc. 2012, 134, 16307. (b) Ghebreghiorgis, T.;
Biannic, B.; Kirk, B. H.; Aponick, A.; Ess, D. H. J. Org. Chem. 2013, 78,
7664.
calculations corroborate a bifunctional role of the phosphinic
acid, i.e., promoting an SN2-type reaction mechanism by
activation of both nucleofuge and nucleophile in a bridging TS.
(9) (a) Ozawa, F.; Okamoto, H.; Kawagishi, S.; Yamamoto, S.; Minami,
T.; Yoshifuji, M. J. Am. Chem. Soc. 2002, 124, 10968. (b) Sawadjoon, S.;
ASSOCIATED CONTENT
* Supporting Information
Experimental details and characterization data. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
Sjoberg, P. J. R.; Orthaber, A.; Matsson, O.; Samec, J. S. M. Chem.Eur.
̈
J. 2014, 20, 1520.
(10) Chirality transfer was recently reported for intramolecular
Friedel−Crafts reaction, see: Jefferies, L. R.; Cook, S. P. Org. Lett.
2014, 16, 2026.
(11) Chiral phosphorous based Brønsted acids have been reported in
different reactions, see: (a) Chen, D.-F.; Han, Z.-Y.; Zhou, X.-L.; Gong,
L.-Z. Acc. Chem. Res. 2014, 47, 2365. (b) Reuping, M.; Kuenkel, A.;
Atodiresei, I. Chem. Soc. Rev. 2011, 40, 4539.
(12) For a recent report on boronic acid catalyzed nucleophilic
substitution of allylic alcohols, see: Zheng, H.; Ghanbari, S.; Nakamura,
S.; Hall, D. G. Angew. Chem., Int. Ed. 2012, 51, 6187.
AUTHOR INFORMATION
■
Corresponding Authors
*joseph.samec@kemi.uu.se
*himo@organ.su.se
*srijit_biswas@yahoo.co.in
Notes
The authors declare no competing financial interest.
(13) (a) Kischel, J.; Mertins, K.; Michalik, D.; Zapf, A.; Beller, M. Adv.
Synth. Catal. 2007, 349, 865. (b) Biswas, S.; Maiti, S.; Jana, U. Eur. J. Org.
Chem. 2009, 2354. (c) Nishimoto, Y.; Onishi, Y.; Yasuda, M.; Baba, A.
Angew. Chem., Int. Ed. 2009, 48, 9131. (d) Iovel, I.; Mertins, K.; Kischel,
J.; Zapf, A.; Beller, M. Angew. Chem., Int. Ed. 2005, 44, 3913.
(14) (a) Bjerum, J.; Schwarzenbach, G.; Sillen, L. G. In Stability
Constants; Chemical Society: London, 1958. (b) Edmundson, R. S. In
Dictionary of Organophosphorous Compounds; Chapman and Hall:
London, 1988. (c) Quin, L. D. A. In A Guide to Organophosphorous
Chemistry; Wiley: New York, 2000.
ACKNOWLEDGMENTS
J.S.M.S. thanks the Swedish Research Council, Olle Engkvist
■
Byggmastare, and Wenner-Gren for financial support. F.H.
̈
acknowledges financial support from the Goran Gustafsson
̈
Foundation and the Knut and Alice Wallenberg Foundation.
Computer time was generously granted by the Swedish National
Infrastructure for Computing. P.S. thanks Erasmus Mundus.
(15) (a) For two isolated examples, see: Yi, W.-B.; Cai, C. J. Fluorine
Chem. 2005, 126, 831. (b) For a recent Ru-catalyzed arene activation,
see: Lee, D.-H.; Kwon, K.-H.; Yi, C. S. J. Am. Chem. Soc. 2012, 134, 7325.
(c) For redox chain reaction, see: Han, X.; Wu, J. Angew. Chem., Int. Ed.
2013, 52, 4637.
(16) See the following computational studies on the bifunctional
activation mode of phosphoric acids of carbonyls/imines and
nucleophiles in transfer hydrogenation, Mannich, and Strecker
REFERENCES
■
(1) (a) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66.
(b) Emer, E.; Riccardo, S.; Capdevila, M. G.; Petruzziello, D.; Vincentiis,
F. D.; Cozzi, P. G. Eur. J. Org. Chem. 2011, 647. (c) Tamaru, Y. Eur. J.
Org. Chem. 2005, 2647. (d) Muzart, J. Eur. J. Org. Chem. 2007, 3077.
(e) Muzart, J. Tetrahedron 2005, 61, 4179. (f) Biswas, S.; Samec, J. S. M.
Chem. Asian J. 2013, 8, 974.
(2) Katritzky, A. R.; Brycki, B. E. Chem. Soc. Rev. 1990, 19, 83.
(3) (a) Sheldon, R. A. Pure Appl. Chem. 2000, 72, 1233. (b) Trost, B.
M. Science 1991, 254, 1471.
́
reactions: (a) Simon, L.; Goodman, J. M. J. Am. Chem. Soc. 2008, 130,
8741. (b) Marcelli, T.; Hammar, P.; Himo, F. Chem.Eur. J. 2008, 14,
8562. (c) Yamanaka, M.; Itoh, J.; Fuchibe, K.; Akiyama, T. J. Am. Chem.
(4) Constable, D. J. C.; Dunn, P. J.; Hayler, J. D.; Humphrey, G. R., Jr.;
Leazer, J. L.; Linderman, R. J.; Lorenz, K.; Manley, J.; Pearlman, B. A.;
Wells, A.; Zaks, A.; Zhang, T. Y. Green Chem. 2007, 9, 411.
(5) Corma, A.; Iborra, S.; Velty, A. Chem. Rev. 2007, 107, 2411.
(6) (a) Krische, M. Angew. Chem., Int. Ed. 2009, 48, 34. (b) Kumagai,
N.; Shibasaki, M. Angew. Chem., Int. Ed. 2011, 50, 4760. (c) Dobereiner,
G. E.; Crabtree, R. H. Chem. Rev. 2010, 110, 681. (d) Guillena, G.;
Ramon, D. J.; Yus, M. Chem. Rev. 2010, 110, 1611.
Soc. 2007, 129, 6756. (d) Simon
2009, 131, 4070.
(17) The calculations were performed with the B3LYP functional, and
the reported Gibbs free energies are corrected for solvation and
dispersion. See SI for full details.
́
, L.; Goodman, J. M. J. Am. Chem. Soc.
(7) (a) Aponick, A.; Biannic, B. Org. Lett. 2011, 13, 1330.
(b) Mukherjee, P.; Widenhoefer, R. A. Chem.Eur. J. 2013, 19, 3437.
(c) Mukherjee, P.; Widenhoefer, R. A. Org. Lett. 2011, 13, 1334.
(d) Mukherjee, P.; Widenhoefer, R. A. Org. Lett. 2010, 12, 1184.
(e) Kawai, N.; Jean-Marie, L.; Ohmi, M.; Uenishi, J. J. Org. Chem. 2006,
71, 4530.
D
DOI: 10.1021/jacs.5b02013
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX