ZrCl4 as a new and efficient catalyst for the opening of epoxide rings by amines

Article abstract of DOI:10.1016/j.tetlet.2003.09.046

Zirconium(IV) chloride catalyses the nucleophilic opening of epoxide rings by amines leading to the efficient synthesis of β-amino alcohols. The reaction works well with aromatic and aliphatic amines in short times at room temperature in the absence of solvent. Exclusive trans stereoselectivity is observed for cyclic epoxides. Aromatic amines exhibit excellent regioselectivity for preferential nucleophilic attack at the sterically less hindered position during the reaction with unsymmetrical epoxides. However, in case of styrene oxide, selective formation of the benzylic amine was observed during the reactions with aromatic amines.

Full text of DOI:10.1016/j.tetlet.2003.09.046

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 8315–8319
ZrCl4 as a new and efficient catalyst for the opening of epoxide
rings by amines
Asit K. Chakraborti* and Atul Kondaskar
Department of Medicinal Chemistry, National Institute of Pharmaceutical Education and Research (NIPER), Sector 67,
S. A. S. Nagar, Punjab 160 062, India
Received 15 May 2003; revised 23 August 2003; accepted 5 September 2003
Abstract—Zirconium(IV) chloride catalyses the nucleophilic opening of epoxide rings by amines leading to the efficient synthesis
of b-amino alcohols. The reaction works well with aromatic and aliphatic amines in short times at room temperature in the
absence of solvent. Exclusive trans stereoselectivity is observed for cyclic epoxides. Aromatic amines exhibit excellent regioselectiv-
ity for preferential nucleophilic attack at the sterically less hindered position during the reaction with unsymmetrical epoxides.
However, in case of styrene oxide, selective formation of the benzylic amine was observed during the reactions with aromatic
amines.
© 2003 Elsevier Ltd. All rights reserved.
b-Amino alcohols are versatile intermediates in the
synthesis of a vast range of biologically active natural
and synthetic products,1 unnatural amino acids,2 and
chiral auxiliaries for asymmetric synthesis.3 The nucle-
ophilic opening of epoxides with amines constitutes a
well recognised route for the synthesis of b-amino
alcohols.4 The classical approach,5 involving heating
epoxides with amines, works less well with poorly
nucleophilic amines. Moreover, the lack of appreciable
regioselectivity, the requirements for high temperature
(which poses problems in dealing with sensitive epox-
ides), and the need for an excess of amine in the
classical methods of b-amino alcohol synthesis have led
to the necessity for activation of the epoxides so as to
increase their susceptibility to nucleophilic attack by
amines. The various methodologies developed for this
purpose include the use of alumina,6 metal amides,7
metal alkoxides,8 metal triflates,9 transition metal
halides,10 alkali metal perchlorates,11 rare earth metal
halides,12 silica under high pressure,13 and montmoril-
lonite clay under microwave irradiation.14 However,
these methodologies suffer from one or more disadvan-
tages such as long reaction times, elevated tempera-
tures, moderate yields, use of air and/or moisture
sensitive catalysts, requirement of stoichiometric
amounts of catalyst, costly reagents/catalysts, potential
rearrangement to allylic alcohols,15 potential hazards in
handling pyrophoric/moisture sensitive reagents in the
preparation of the catalyst, and in most of the cases
being applicable to aromatic amines only. Therefore,
the development of a better catalyst for the activation
of epoxides rendering them more susceptible to nucle-
ophilic attack under milder conditions is in high
demand.
Considering the widespread applications of the resul-
tant b-amino alcohols, we felt that a catalyst of choice
should be one that is easily available and less costly,
less toxic, and operable under environmentally friendly
conditions so as to fulfill the ‘triple bottom line’16
philosophy of green chemistry. We reasoned that the
high abundance of zirconium in the earth’s crust17
should make zirconium(IV) compounds less costly.
Since zirconium, in its normal quadrivalent state dis-
plays no redox properties but can attain a covalency
maximum up to 8 and displays low toxicity,18 Zr(IV)
compounds should be ideal for catalytic applications.
Accordingly, Zr(IV) derivatives are finding increasing
commercial use in catalysis.19 Recently, zirconium sul-
fophenyl phosphate has been introduced as a heteroge-
neous catalyst in the preparation of b-amino alcohols
via opening of epoxide rings with amines.20 However,
this methodology was not applicable to aliphatic
amines and even with aromatic amines the reactions
were carried out for a prolonged period (2–22 h).
Therefore, we planned to employ various Zr(IV) deriva-
tives to evaluate their catalytic efficiency in carrying out
the ring-opening reaction of epoxides with amines and
Keywords: zirconium; epoxide; amine; regioselectivity.
* Corresponding author. Tel.: +91-0172-214682-686; fax: +91-0172-
214692; e-mail: akchakraborti@niper.ac.in
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.09.046

8316
A. K. Chakraborti, A. Kondaskar / Tetrahedron Letters 44 (2003) 8315–8319
found that ZrCl4 acts as an excellent catalyst for the
desired transformation.
In order to delineate the standard operating conditions,
cyclohexene oxide 1 (2.5 mmol) was treated with aniline
(2.5 mmol) at room temperature under solvent-free
conditions in the presence of ZrCl4 (5 mol%). Complete
conversion took place in 15 min leading to a quantita-
tive yield of the resultant 2-phenylaminocyclohexanol,
identified as the trans isomer with characteristic 1H
NMR signals appearing at l 3.13 (ddd, 1H, J=10.0,
10.1, and 3.9 Hz) for the CHNH and at 3.33 (ddd, 1H,
J=10.5, 10.4, and 4.2 Hz) for the CHOH protons and
a
13C NMR signal at l 60 for the CHNH carbon.9e,11
The catalyst was recovered by filtration after diluting
the reaction mixture with Et2O and was reused repeti-
tively without any significant loss of catalytic activity.
To evaluate the generality, reactions of 1 were carried
out with various aromatic and aliphatic amines under
the catalytic influence of ZrCl4 and excellent results
were obtained (Table 1). The reaction proceeded well
with both aromatic and aliphatic amines.
Figure 1. Catalytic cycle during the ZrCl4-catalysed opening
of epoxide rings with amines.
The DSC endotherm of the recovered catalyst was
found to be identical with that of an authentic sample
of ZrCl4 indicating that no ligand exchange takes place
during the reaction. The exclusive formation of the
trans amino alcohol suggests that the initially formed
intermediate (TS-I) is weakly polar and no distinct
charge accumulation on the epoxide carbons takes
place.
The role of ZrCl4 in catalysing the opening of epoxide
rings with amines may be realised through the catalytic
cycle depicted in Figure 1. Coordination of Zr4+ with
the epoxide oxygen (TS-I) renders the epoxide suscepti-
ble to nucleophilic attack by the amine leading to
TS-II/TS-IIa followed by protonolysis (via intramolec-
ular proton transfer involving TS-IIa or intermolecular
proton transfer involving TS-II) leading to the forma-
tion of the amino alcohol and liberation of the catalyst.
We next planned to evaluate the regioselective outcome
of the ZrCl4-catalysed epoxide opening reaction with
various amines using styrene oxide 2 as a representative
unsymmetrical epoxide (Scheme 1) and the results are
summarised in Table 2.
In all cases the corresponding amino alcohols were
obtained in excellent yields. The two isomers could not
be separated by column chromatography. On each
occasion, the regioselectivity was determined by GC–
MS and 1H NMR. In the MS, the regioisomer 3
exhibited a daughter ion at m/z (M+−31) due to the loss
of CH2OH and the diagnostic feature in the mass
spectra of 4 was the ion peak at m/z (M+−106) arising
from the loss of PhCHO.
Table 1. Reactions of 1 with various amines in the pres-
a
ence of ZrCl4
For reactions with aromatic amines, the benzylic pro-
ton of 3 and 4 appeared ꢀl 4.5 and 5.0, respectively, in
the 1H NMR. During the reaction with aniline, the
GC–MS revealed the product to be a mixture of 3 and
4 in a ratio of 92:8 on the basis of the daughter ions at
m/z 213–31 and 213–106 corresponding to 3 and 4,
respectively. The signals at l 3.6–4.0 (m, 4H), 4.48–4.51
(m, 1H), 6.4–7.5 (m, 10H) in the 1H NMR of the
product could be assigned to 2-phenylamino-2-
phenylethanol 38b and the values at l 3.1–3.4 (m, 2H),
Scheme 1. Regioselectivity in the epoxide ring opening of
styrene oxide 2 with various amines catalysed by ZrCl4.

A. K. Chakraborti, A. Kondaskar / Tetrahedron Letters 44 (2003) 8315–8319
8317
Table 2. Reactions of 2 with various amines in the pres-
the presence of ZrCl4 (5 mol%) and the results are
summarised in Table 3.
a
ence of ZrCl4
Quantitative yields of the corresponding amino alcohols
were obtained in the reaction of cyclopentene oxide,
propylene oxide, glycidyl phenyl ether, and glycidyl
tert-butyl ether (entries 1–4). A trans stereospecificity
was observed in the cleavage of the epoxide ring during
the reaction of cyclopentene oxide. Completely selective
nucleophilic attack at the sterically less hindered termi-
nal carbon of the epoxide moiety in propylene oxide,
glycidyl phenyl ether, and glycidyl tert-butyl ether was
observed.
Excellent chemoselectivity was achieved with epichloro-
hydrin (entry 5) resulting in an 81% yield of the amino
alcohol corresponding to nucleophilic attack at the
terminal carbon of the epoxide moiety. No product
arising from nucleophilic displacement of the chlorine
could be detected through GC–MS analysis of the
reaction mixture. Nucleophilic attack on epichlorohy-
drin generally results in the formation of a new methyl-
oxirane 5. The reaction, in principle, may proceed via
two distinct pathways: (i) direct displacement of chlo-
rine (path a) or (ii) initial attack on the epoxide (path b)
followed by protonation of the amino alcohol or extru-
sion of the chlorine atom to give 5 (Scheme 2).22 It is
anticipated that the chlorine atoms in ZrCl4 make the
central metal sufficiently electrophilic so as to hold the
negative charge of the alkoxide generated after the
nucleophilic attack on the metal complexed epoxide.
Thus, the free alkoxide anion is not available for subse-
quent elimination of the chloride anion.
4.60 (bs, 2H), 4.89–4.97 (m, 1H), 6.5–7.8 (m, 10H) to
2-phenylamino-1-phenylethanol 4. A 92:8 ratio could
be determined for 3:4 taking into consideration the
integral values of the corresponding benzylic protons.
The benzylic proton of 3 and 4, arising from the
reactions using aliphatic amines, appeared at ꢀl 3.9
and 4.7, respectively. The ratio of the two regioisomeric
products was determined by GC–MS and 1H NMR
(based on the methine and methylene proton signals
corresponding to 3 and 4) and by comparison with the
reported values (in the case of the product obtained
with piperidine,21 and benzylamine11).
In conclusion, ZrCl4 is a new, highly efficient, and
reusable catalyst for the opening of epoxides with
amines leading to the synthesis of b-amino alcohols.
Table 3. Reaction of various epoxides with aniline in the
a
presence of ZrCl4
Selective formation of the regioisomeric product 3 aris-
ing from nucleophilic attack at the benzylic carbon of 2
was observed during the reactions with aromatic
amines (entries 1–5). Aliphatic amines exhibited a pref-
erence for nucleophilic attack at the terminal carbon
and the best selectivity (22:78) was observed in the case
of benzylamine (entry 6). A moderate selectivity in
favour of attack at the non-benzylic position was
observed for pyrrolidine, piperidine, and morpholine
(entries 7–9). The preference for the formation of 3
during the reaction with aromatic amines may be
accounted for by the fact that the phenyl group in 2
assists in the stabilization/accumulation of carboca-
tionic character at the benzylic carbon in TS-I. Thus,
aromatic amines, being less nucleophilic, react selec-
tively at the benzylic carbon of 2. The preference for
reaction at the terminal carbon of 2 by the aliphatic
amines may be explained due to the increased nucle-
ophilicity of aliphatic amines, compared to that of
aromatic amines, favouring a more SN2 process.
Finally, to claim ZrCl4 as a general catalyst for the
activation of epoxides for nucleophilic attack by
amines, different epoxides were treated with aniline in

8318
A. K. Chakraborti, A. Kondaskar / Tetrahedron Letters 44 (2003) 8315–8319
1991, 91, 437–475; (d) Mitsunobu, O. In Comprehensive
Organic Synthesis; Winterfeldt, E., Ed.; Pergamon Press:
New York, 1996; Vol. 6, Part 1.3.4.1; (e) Rao, A. S.;
Paknikar, S. K.; Kirtane, J. G. Tetrahedron 1983, 39,
2323–2367.
5. (a) Crooks, P. A.; Szyudler, R. Chem. Ind. (London) 1973,
1111–1112; (b) Deyrup, J. A.; Moyer, C. L. J. Org. Chem.
1969, 34, 175–179; (c) Freifelder, M.; Stone, G. R. J. Org.
Chem. 1961, 26, 1477–1480; (d) Mousseron, M.; Jullien,
J.; Jolchine, Y. Bull. Soc. Chim. Fr. 1952, 757–766; (e)
Lutz, R. E.; Freek, J. A.; Murphy, R. S. J. Am. Chem.
Soc. 1948, 70, 2015–2023.
Scheme 2. Reaction of epichlorohydrin with a nucleophile.
The specific advantages are that the reactions are car-
ried out under mild conditions in short times, with
excellent stereo-, regio-, and chemoselectivity, and work
for aromatic and aliphatic amines. The low cost and
6. Alumina: Posner, G. H.; Rogers, D. Z. J. Am. Chem.
Soc. 1977, 99, 8208–8214.
7. (a) Et2AlNHR: Overman, L. E.; Flippin, L. A. Tetra-
hedron Lett. 1981, 22, 196–198; (b) Silicon amides: Pap-
ini, A.; Ricci, I.; Taddei, M.; Seconi, G.; Dembach, P. J.
23
lack of appreciable toxicity of ZrCl4 are consistent
with increasing environmental concerns.24 The solvent-
free reaction conditions employed in the present
method will make it ‘environmentally friendly’ and
potentially useful for industrial applications.
Chem. Soc., Perkin Trans. 1 1984, 2261–2265; (c)
R2NMgBr: Carre, M. C.; Houmounou, J. P.; Caubere, P.
Tetrahedron Lett. 1985, 26, 3107–3110; (d) R%PbNR2:
3
Yamada, J.-I.; Yumoto, M.; Yamamoto, Y. Tetrahedron
Lett. 1989, 30, 4255–4258; (e) Cu amide: Yamamoto, Y.;
Asao, N.; Meguro, M.; Tsukade, N.; Nemoto, H.; Ada-
yari, N.; Wilson, J. G.; Nakamura, H. J. Chem. Soc.,
Chem. Commun. 1993, 1201–1203.
Typical procedure: 2-(phenylamino)cyclohexanol: ZrCl4
(0.03 g, 5 mol%) was added to a magnetically stirred
mixture of 1 (0.25 mL, 2.5 mmol) and aniline (0.225
mL, 2.5 mmol) at room temperature under nitrogen.
After completion of the reaction (15 min, TLC), the
reaction mixture was diluted with Et2O (15 mL) and the
precipitated catalyst was separated by decantation of
the supernatant ethereal solution. The catalyst was
washed with Et2O (10 mL) and the combined ethereal
solutions were dried (Na2SO4) and concentrated in
vacuo to afford 2-(phenylamino)cyclohexanol (0.475 g,
8. (a) Ti(OPri)4: Sagava, S.; Abe, H.; Hase, Y.; Inaba, T. J.
Org. Chem. 1999, 64, 4962; (b) DIPAT (diisopropoxyalu-
minium trifluoroacetate): Rampalli, S.; Chaudhari, S. S.;
Akamanchi, K. G. Synthesis 2000, 78–80.
9. (a) Ph4SbOTf: Fujiwara, M.; Imada, M.; Baba, A.; Mat-
suda, H. Tetrahedron Lett. 1989, 30, 739–742; (b)
La(OTf)3: Chini, M.; Crotti, P.; Favero, L.; Machhia, F.;
Pineschi, M. Tetrahedron Lett. 1994, 35, 433–436; (c)
Yb(OTf)3: Meguro, M.; Asao, N.; Yamamoto, Y. J.
Chem. Soc., Perkin Trans. 1 1994, 2597–2601; (d) LiOTf:
Auge, J.; Leroy, F. Tetrahedron Lett. 1996, 37, 7715–
7716; (e) Cu(OTf)2 and Sn(OTf)2: Sekar, G.; Singh, V. K.
J. Org. Chem. 1999, 64, 287–289.
1
100%), mp 57–59°C (lit.9e 58–59°C); H NMR (CDCl3)
l 1.03–1.42 (m, 4H), 1.72–1.78 (m, 2H), 2.10–2.16 (m,
2H), 2.8–3.0 (m, 2H, D2O exchangeable), 3.13 (ddd,
1H, J=3.9, 10.0, 10.1 Hz), 3.33 (ddd, 1H, J=4.2, 10.4,
10.5 Hz), 6.7–7.2 (m, 5H); 13C NMR (CDCl3) l 24.27,
25.02, 31.62, 33.15, 60.17, 74.55, 114.40, 118.38, 129.34,
147.81; EIMS (m/z) 191 (M+).
10. (a) CoCl2: Iqbal, J.; Pandey, A. Tetrahedron Lett. 1990,
31, 575–576; (b) TaCl5: Chandrasekhar, S.; Ramchander,
T.; Prakash, S. J. Synthesis 2000, 1817–1818.
11. Chini, M.; Crotti, P.; Macchia, F. Tetrahedron Lett. 1990,
31, 4661–4664.
References
12. (a) SmI2: Weghe, V.; Collin, J. Tetrahedron Lett. 1995,
36, 1649–1652; (b) SmCl3: Fu, X.-L.; Wu, S.-H. Synth.
Commun. 1997, 27, 1677–1684; (c) CeCl3: Reddy, L. R.;
Reddy, M. A.; Bhanumathi, N.; Rao, K. R. Synthesis
2001, 831–832.
13. (a) Kotsuki, H.; Shimanouchi, T.; Teraguchi, M.;
Kataoka, M.; Tatsukawa, A.; Nishizawa, H. Chem. Lett.
1994, 2159–2162; (b) Kotsuki, H.; Hayashida, K.; Shi-
manouchi, T.; Nishizawa, H. J. Org. Chem. 1996, 61,
984–990.
14. Mojtahedi, M. M.; Saidi, M. R.; Bolourtchian, M. J.
Chem. Res. (S) 1999, 128–129.
15. Metal alkylamides induce the rearrangement of epoxides
to allylic alcohols. See: (a) Crandall, J. K.; Apparu, M.
Org. React. 1983, 29, 345–443; (b) Kissel, C. L.; Rick-
born, B. J. Org. Chem. 1972, 37, 2060–2063.
1. (a) Gehlert, D. R.; Goldstein, D. J.; Hipskind, P. A. Ann.
Rep. Med. Chem. 1999, 201–210; (b) Corey, E. J.; Zhang,
F. Angew. Chem., Int. Ed. 1999, 38, 1931–1934; (c) Johan-
nes, C. W.; Visser, M. S.; Weatherhead, G. S.; Hoveyda,
A. H. J. Am. Chem. Soc. 1998, 120, 8340–8347; (d) Chng,
B. L.; Ganesan, A. Bioorg. Med. Chem. Lett. 1997, 7,
1511–1514; (e) Rogers, G. A.; Parsons, S. M.; Anderson,
D. C.; Nilsson, L. M.; Bahr, B. A.; Kornreich, W. D.;
Kaufman, R.; Jacobs, R. S.; Kirtman, B. J. Med. Chem.
1989, 32, 1217–1230; (f) De Cree, J.; Geukens, H.; Leem-
poels, J.; Verhaegen, H. Drug Dev. Res. 1986, 8, 109–117.
2. (a) O’Brien, P. Angew Chem., Int. Ed. 1999, 38, 326–329;
(b) Li, G.; Chang, H.-T.; Sharpless, K. B. Angew Chem.,
Int. Ed. 1996, 35, 451–454.
3. Ager, D. J.; Prakash, I.; Schaad, D. R. Chem. Rev. 1996,
96, 835–876.
16. Elkington,
J.
http://www.sustainability.co.uk/
4. (a) Bergmeier, S. C. Tetrahedron 2000, 56, 2561–2576; (b)
Hodgson, D. M.; Gibbs, A. R.; Lee, G. P. Tetrahedron
1996, 52, 14361–14384; (c) Hanson, R. M. Chem. Rev.
sustainability.htm
17. Huheey, J. E. Inorganic Chemistry, 3rd ed.; Roger &
Harper Row: Singapore, 1990; Chapter 18.

A. K. Chakraborti, A. Kondaskar / Tetrahedron Letters 44 (2003) 8315–8319
8319
18. (a) Moles, P. J. Adhesion Sci. Technol. 1992, 6, 61–71; (b)
Farnworth, F. Jones, S. L.; McAlpine, I. Speciality Inor-
ganic Chemicals, Special Publication No. 40, Royal Soci-
ety of Chemistry, London 1980; (c) LD50 (oral, rat) of
ZrCl4 is 1688 mg kg−1 [Emsley, J. The Elements; 3rd ed.
Clarendon: Oxford, 1988].
Mendenhall, D.; Vanderplas, B. C. Tetrahedron: Asym-
metry 1999, 10, 2655–2663.
22. McClure, D. E.; Arison, B. H.; Baldwin, J. J. J. Am.
Chem. Soc. 1979, 101, 3666–3668.
23. ZrCl4 has low toxicity [LD50 (rat)=1688 mg kg−1] and is
not considered poisonous. See: Emsley, J. The Elements,
3rd ed.; Clarendon: Oxford, 1998.
19. Moles, P. J. http://www.zrchem.com
24. Garrett, R. L. In Designing Safer Chemicals; Garrett, R.
L.; De Vilto, S. C., Eds.; American Chemical Society
Symposium Series 640: Washington DC, 1996; Chapter 1.
20. Curini, M.; Epifano, F.; Marcotullio, M. C.; Rosati, O.
Eur. J. Org. Chem. 2001, 4149–4152.
21. Anderson, S. A.; Ayers, J. T.; DeVries, K. M.; Ito, F.;