A hydrogen chloride-free pinner reaction promoted by lewis acids

Article abstract of DOI:10.1002/hlca.201200435

A hydrogen chloride-free variation of the Pinner reaction was developed, in which alcohols react with carbonitriles to furnish carboxylates. Best results were achieved with aliphatic alcohols, and aliphatic or benzylic nitriles in the presence of 2 equiv. of trimethylsilyl triflate (Me₃SiOTf). With these substrates, yields exceeding 80% were achieved. A strictly neutral variation of this protocol is possible, when 1 equiv. of Et₃N is added to the reaction mixture. Copyright

Full text of DOI:10.1002/hlca.201200435

Helvetica Chimica Acta – Vol. 95 (2012)
1851
A Hydrogen Chloride-Free Pinner Reaction Promoted by Lewis Acids
by Dominik Pfaff, Gregor Nemecek, and Joachim Podlech*
Institut fꢀr Organische Chemie, Karlsruher Institut fꢀr Technologie (KIT), Fritz-Haber-Weg 6,
DE-76131 Karlsruhe (joachim.podlech@kit.edu)
Dedicated to Prof. Dr. Dieter Seebach on the occasion of his 75th birthday
A hydrogen chloride-free variation of the Pinner reaction was developed, in which alcohols react
with carbonitriles to furnish carboxylates. Best results were achieved with aliphatic alcohols, and aliphatic
or benzylic nitriles in the presence of 2 equiv. of trimethylsilyl triflate (Me3SiOTf). With these substrates,
yields exceeding 80% were achieved. A strictly neutral variation of this protocol is possible, when 1 equiv.
of Et3N is added to the reaction mixture.
Introduction. – The Pinner imidate formation was first published in 1877 by Pinner
and Klein [1][2]. Carbonitriles and alcohols react in the presence of gaseous HCl to
form imidate hydrochlorides. Aliphatic and aromatic nitriles can be used (if steric
hindrance is not too strong), and the second component is usually an aliphatic primary
or secondary alcohol. Phenols can be used similarly, while electron-rich phenols (e.g.,
dihydroxybenzenes) are substrates of a HoubenꢀHoesch reaction (electrophilic
aromatic acylation), which is possible under these reaction conditions [3][4]. The
imidate hydrochlorides can subsequently be transformed with H2O to carboxylates,
with alcohols into ortho esters, and with amines into amidines or amidine hydro-
chlorides. A less frequently used pyrolysis (above 08) leads to carboxamides.
Scheme 1. Pinner Reaction with Some Variations
The mechanism of this reaction involves protonation of the nitrile N-atom leading
to an activated nitrilium chloride, which is then attacked by the alcohol (Scheme 2).
Proton transfer yields the imido-ester hydrochloride. A possible intermediate
equilibrium formation of a carbonimidoyl chloride has no significant influence on
the reaction outcome [2c].
ꢁ 2012 Verlag Helvetica Chimica Acta AG, Zꢀrich

1852
Helvetica Chimica Acta – Vol. 95 (2012)
Scheme 2. Mechanism of the Pinner Reaction. P.T., Proton transfer.
The rather drastic reaction conditions of the original Pinner reaction do not allow
its broad application, especially not for the transformation of acid-sensitive substrates.
Occasionally, milder conditions for the Pinner reaction have been reported. Jiang et al.
reported on a variation, in which HCl was replaced with an ionic liquid based on a
sulfonate. Nevertheless, these are similarly Brønsted-acidic conditions, and this
methods has only been applied to aliphatic nitriles [5]. A significantly milder variation
uses dihydridotetrakis(triphenylphosphano)ruthenium ([RuH2(PPh3)4]) as a catalyst,
where its catalytic activity is most probably due to its Lewis acidity [6]. Only poor yields
have been reported with various Rh, Pd, Cu, and Mo catalysts. Primary aliphatic nitriles
and alcohols were reacted by this method, and intramolecular variations (with
formation of five-membered rings) were possible. This intramolecular variations
furthermore allowed the utilization of secondary alcohols with good yields [6].
Reactions between electron-deficient nitriles (F- or Cl-substituted) and hydroxides
with increased nucleophilicity (NH2OH, oximes, or hydroxamic acids) are possible at
elevated temperatures without further activation [7]. Formation of imidates from
nitriles and alcohols is furthermore possible with base catalysis. Setting of an
equilibrium usually results in slow reactions with poor yields, rendering this variation
significantly less useful [8]. Nevertheless, a similar reaction is used in the most
prominent trichloroacetimidate activation of sugars as glycosyl donors, developed by
Schmidt et al. [9]. Furthermore, Luo and Jeevanandam reported on a variation in which
HCl was generated in situ by reaction of Me3SiCl with EtOH (both in excess) [10].
During elaboration of a total synthesis, we incidentally found that OH groups react
with the solvent MeCN in the presence of 2 equiv. of hafnium triflate (¼ hafnium
trifluoromethanesulfonate; Hf(OTf)4). We considered this Lewis acid too expensive to
be useful in a broadly applied method and started to work on a comprehensive
optimization, which is reported below.
Results and Discussion. – We considered the reaction of (9H-fluoren-9-yl)methanol
with MeCN to be a representative example, which was used for the optimization of the
reaction conditions. With this substrate, a simple UV detection in thin layer
chromatography was warranted, and its molecular weight prevents losses during
evaporation procedures (Scheme 3). All given yields refer to isolated and purified
substrates. Contrary to the originally proposed Pinner reaction, after aqueous workup
we always observed formation of the ester instead of the imidate.

Helvetica Chimica Acta – Vol. 95 (2012)
Scheme 3. Reaction Used for Optimizations
1853
This reaction proceeded with good yield (72%), when 2 equiv. of Hf(OTf)4 were
used as Lewis acid in neat MeCN as the solvent (Table 1); presence of an inert-gas
atmosphere was not necessary. Utilization of catalytic amounts of this Lewis acid led to
significantly reduced yields.
Table 1. Optimization of Reaction Conditions with Hf(OTf)4 (cf. Scheme 3)
Entry
Lewis acid (equiv.)
Conditions
Yield (s.m.)a) [%]
1
2
3
4
5
6
Hf(OTf)4 (2)
MeCN, r.t., 48 h
MeCN, r.t., 65 h
MeCN/H2O 3 : 1, r.t., 65 h
MeCN/H2O 1 : 1, r.t., 65 h
MeCN/H2O 5 : 1, r.t., 65 h
MeCN/H2O 10 : 1, r.t., 65 h
72 (15)
25 (69)
9 (79)
4 (87)
17 (71)
3 (81)
Hf(OTf)4 (0.2)
Hf(OTf)4 (0.2)
Hf(OTf)4 (0.2)
Hf(OTf)4 (0.2)
Hf(OTf)4 (0.2)
a) Yields of recovered starting material (s.m.) given in parentheses. All yields refer to isolated and
purified substrates.
Various Lewis acids were tested (Table 2). Addition of AlBr3 gave quite similar
yields (50%) as have been achieved with Hf(OTf)4. Since the former Lewis acid is
significantly cheaper than the latter, we tested whether over-stoichiometric amounts of
AlBr3 together with Hf(OTf)4 as catalyst leads to higher yields. Nevertheless, the
reaction outcome could not be improved with this variation. Best results were achieved
with 2 equiv. of Me3SiOTf at room temperature (83%). Neither higher nor lower
reaction temperatures led to an improvement. Replacement of this Lewis acid with the
cheaper Me3SiCl resulted in non-satisfactory yields (33%).
The optimized conditions have been applied to the reaction of four primary
aliphatic alcohols and a selection of commonly used nitriles (Table 3). We combined
(9H-fluoren-9-yl)methanol, decan-1-ol, 6-chlorohexan-1-ol, and diethylene glycol
monoethyl ether with MeCN, acrylonitrile, benzonitrile, and benzyl cyanide (BnCN).
Best results were obtained in the reactions of MeCN and BnCN, i.e., in the reaction of
aliphatic nitriles. Yields from 75 to 90% were obtained with these substrates. The yields
were not satisfactory with acrylonitrile and benzonitrile. Nevertheless, a previously
published alternative synthesis of (9H-fluoren-9-yl)methyl acrylate starting with (9H-
fluoren-9-yl)methanol and the moisture-sensitive acryl chloride (in the presence of
Et3N) was conducted at 08 under a N2 atmosphere and gave a yield of only 33% after
31 h [11]. The here presented protocol provides higher yields (52%, Entry 2) and
allows the utilization of a less-sensitive substrate. The identity and purity of all products
were determined by NMR spectroscopy.

1854
Helvetica Chimica Acta – Vol. 95 (2012)
Table 2. Optimization of the Lewis Acid (cf. Scheme 3)
Entry
Lewis acid (equiv.)
Conditionsa)
Yield (s.m.)b) [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Hf(OTf)4 (2)
LaCl3 (2)
ZnCl2 (2)
ZnBr2 (2)
FeCl3 (2)
CuBr2 (2)
CeCl3 · 6 H2O (2)
AlCl3 (2)
AlBr3 (2)
AlBr3 (2)
AlCl3 (2), Hf(OTf)4 (0.1)
AlBr3 (2), Hf(OTf)4 (0.1)
TMSOTf (2)
TMSOTf (2)
TMSOTf (2)
TMSOTf (1)
TMSOTf (0.5)
TMSCl (2)
r.t., 48 h
r.t., 65 h
r.t., 65 h
r.t., 65 h
r.t., 65 h
r.t., 90 h
r.t., 90 h
r.t., 65 h
r.t., 65 h
508, 90 h
r.t., 65 h
r.t., 65 h
r.t., 65 h
508, 65 h
08, 65 h
72 (15)
2 (83)
4 (87)
1 (87)
3 (82)
5 (68)
1 (87)
7 (83)
50 (39)
65 (20)
9 (82)
64 (24)
83 (9)
78 (5)
60 (27)
77 (5)
41 (49)
33 (59)
26 (64)
80 (10)
508, 65 h
508, 65 h
r.t., 65 h
508, 65 h
Et3N, r.t., 65 hc)
TMSCl (2)
TMSOTf (2)
a) Reactions performed in MeCN without inert-gas atmosphere. b) Yields of recovered starting material
c
(s.m.) given in parentheses. All yields refer to isolated and purified substrates. ) TMSOTf was added to a
solution of (9H-fluoren-9-yl)methanol and Et3N (1 equiv.) in MeCN.
Table 3. Variation of Some Alcohols and Carbonitrilesa)
Entry
R1
R2
Yield [%]b)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
(9H-Fluoren-9-yl)methyl
(9H-Fluoren-9-yl)methyl
(9H-Fluoren-9-yl)methyl
(9H-Fluoren-9-yl)methyl
Me(CH2)9
Me(CH2)9
Me(CH2)9
Me(CH2)9
Cl(CH2)6
Cl(CH2)6
Cl(CH2)6
Cl(CH2)6
EtO(CH2)2O(CH2)2
EtO(CH2)2O(CH2)2
EtO(CH2)2O(CH2)2
EtO(CH2)2O(CH2)2
Me
Vinyl
Ph
83
52
44
86
80
29
23
85
84
38
27
90
75
23
26
85
Bn
Me
Vinyl
Ph
Bn
Me
Vinyl
Ph
Bn
Me
Vinyl
Ph
Bn
a) Reactions performed in the respective nitrile as solvent at room temperature for 65 h. b) All yields
refer to isolated and purified substrates.

Helvetica Chimica Acta – Vol. 95 (2012)
1855
A proposal of a plausible reaction mechanism includes formation of an N-
silylnitrilium ion [12] (Scheme 4), which is thus activated for a nucleophilic attack. The
nucleophilicity of an intermediate silyl ether should be higher compared to the
nucleophilicity of the corresponding alcohol (albeit its steric hindrance is increased).
Formation of the Brønsted acid TfOH is likely under these reaction conditions, but it
seems to have no influence on the reaction outcome, since a similar reaction with
addition of 1 equiv. of Et3N leads to comparable yields (Table 2, Entries 13 and 20).
Further experiments to gain a more detailed insight in the mechanism of this reaction
are currently under investigation in our laboratories.
Scheme 4. Proposal for a Reaction Mechanism
Conclusions. – The here presented variation of the Pinner reaction with mild
reaction conditions should be of significant use for the transformation of nitriles to
carboxylates, for the acylation of alcohols, and thus for the protection of alcohols.
Currently, we are trying to broaden the scope of this reaction.
This work was supported by the Landesgraduiertenfçrderung Baden-Wꢀrttemberg (grant to G. N.).
Experimental Part
General Procedure for Reactions with MeCN and Acrylonitrile. Me3SiOTf (336 mg, 1.51 mmol,
2.00 equiv.) was added to a soln. of the alcohol (0.76 mmol, 1.00 equiv.) in the nitrile (3 ml), and the
mixture was stirred at r.t. for 65 h. H2O (25 ml) and brine (25 ml) were added, and the mixture was
extracted with AcOEt (3 ꢁ 30 ml). The combined org. layers were dried (Na2SO4) and concentrated. The
crude product was purified by chromatography (silica gel).
General Procedure for Reactions with BnCN and PhCN. Me3SiOTf (336 mg, 1.51 mmol, 2.00 equiv.)
was added to a soln. of the alcohol (0.76 mmol, 1.00 equiv.) in the nitrile (3 ml), and the mixture was
stirred at r.t. for 65 h. H2O (25 ml) and brine (25 ml) were added, and the mixture was extracted with
AcOEt (3 ꢁ 30 ml). The combined org. layers were dried (Na2SO4) and concentrated, and excess nitrile
was distilled off at reduced pressure (908, 0.1 mbar). The crude product was purified by chromatography
(silica gel).
REFERENCES
[1] A. Pinner, F. Klein, Ber. Dtsch. Chem. Ges. 1877, 10, 1889; A. Pinner, F. Klein, Ber. Dtsch. Chem. Ges.
1878, 11, 1475; A. Pinner, Ber. Dtsch. Chem. Ges. 1883, 16, 1643.
[2] a) T. K. Brotherton, J. W. Lynn, Chem. Rev. 1959, 59, 841; b) R. Roger, D. G. Neilson, Chem. Rev.
1961, 61, 179; c) E. N. Zilꢂberman, Russ. Chem. Rev. 1962, 31, 615; d) D. G. Neilson, in ꢃThe
chemistry of functional groups: the chemistry of amidines and imidatesꢂ, Ed. S. Patai, John Wiley &
Sons, London, 1975, pp. 385 – 489; e) D. G. Neilson, in ꢃThe chemistry of amidines and imidatesꢂ,
Eds. S. Patai, Z. Rappoport, John Wiley & Sons, Chichester, 1991, pp. 425 – 483.
[3] K. Hoesch, Ber. Dtsch. Chem. Ges. 1915, 48, 1122; J. Houben, Ber. Dtsch. Chem. Ges. 1926, 59, 2878.
[4] P. E. Spoerri, A. S. DuBois, Org. React. (N.Y.) 1949, 5, 387.
[5] D. Jiang, Y. Y. Wang, M. Tu, L. Y. Dai, React. Kinet. Catal. Lett. 2008, 95, 265.

1856
Helvetica Chimica Acta – Vol. 95 (2012)
[6] T. Naota, Y. Shichijo, S.-i. Murahashi, J. Chem. Soc., Chem. Commun. 1994, 1359.
[7] N. A. Bokach, V. Y. Kukushkin, Russ. Chem. Rev. 2005, 74, 153.
[8] F. C. Schaefer, G. A. Peters, J. Org. Chem. 1961, 26, 412.
[9] R. R. Schmidt, J. Michel, Angew. Chem. 1980, 92, 763; Angew. Chem., Int. Ed. 1980, 19, 731.
[10] F.-T. Luo, A. Jeevanandam, Tetrahedron Lett. 1998, 39, 9455.
[11] H. G. Schild, D. A. Tirrell, Macromolecules 1992, 25, 4553.
[12] R. Caputo, C. Ferreri, G. Palumbo, Tetrahedron Lett. 1984, 25, 577.
Received August 9, 2012