Base-catalyzed direct transformation of benzylamines into benzyl alcohols

Article abstract of DOI:10.1055/s-0031-1290595

Benzylamines were directly transformed into benzyl alcohols in superheated aqueous methanol in the presence of a base catalyst. This process is simple and will provide an alternative industrial route to benzyl alcohols such as xylene glycols. Georg Thieme Verlag Stuttgart · New York.

Full text of DOI:10.1055/s-0031-1290595

7
06
LETTER
Base-Catalyzed Direct Transformation of Benzylamines into Benzyl Alcohols
T
Y
ransformation of
u
Benzylamine
t
s
into
a
Benzyl
A
lc
k
ohols a Kanbara,* Takafumi Abe, Norio Fushimi, Taketo Ikeno
Niigata Research Laboratory, Mitsubishi Gas Chemical Company, Inc., 182, Shinwari, Tayuhama, Niigata 950-3112, Japan
Fax +81(25)2584800; E-mail: yutaka-kanbara@mgc.co.jp
Received 13 November 2011
conventional
Abstract: Benzylamines were directly transformed into benzyl al-
1
+
+
HONO
R –N2
cohols in superheated aqueous methanol in the presence of a base
catalyst. This process is simple and will provide an alternative in-
dustrial route to benzyl alcohols such as xylene glycols.
1
1
1
R –NH2 + Tos-X
R –N(Tos)2
R –OH
2
1
+
2
–
+
R -X
R N R 3X
Key words: alcohols, amines, catalysis, substitution, solid-phase
synthesis
this work
Ar NH2
1
step
Ar
OH
basic conditions
Benzyl alcohols are widely used as raw materials for med-
icines, agrochemicals, plasticizing agents, solvents in
paints, and so on. Particularly, xylene glycols, which
Scheme 1 Transformation of the aliphatic amino group into a hy-
droxy group
1
have two hydroxy groups, have become increasingly im-
portant as raw materials for synthetic fibers and synthetic
resins, such as polyesters and polyurethanes, and an effi-
amino groups into hydroxy groups under base-mediated
conditions.
2
First, m-xylene diamine (MXDA) was adopted as a model
compound and its transformation into m-xylene glycol
cient manufacturing process is desired. Many methods
have been proposed for the preparation of xylene glycols,
however, there are few that are applicable on an industrial
scale. The hydrogen reduction of dimethyl terephthalate,
generally used in industrial processes, needs high temper-
ature and high pressure and the yield is relatively low due
(
MXG) was examined in methanol at 240 °C (Table 1).
Without catalyst, no reaction proceeded (entry 1). Next,
solid acid (SiO –Al O ) and solid base catalysts (MgO,
2
2
3
Ca(OH) ) were investigated, but no product was obtained
2
3
(entries 2–4). HCl also showed no catalytic activity; how-
ever, by using KOH as a catalyst, MXG was obtained in
to over-reduced byproducts.
On the other hand, large amounts of xylene diamines are
commercially produced for nylon by the hydrogen reduc-
tion of xylene dinitriles, which are prepared from xylenes
2
0% yield (entries 5 and 6). Hence, other homogeneous
base catalysts were screened and it was found that
NaOMe and NaOH were both effective, and MXG was
4
by ammoxidation. It was considered that if an amino
group could be easily transformed into a hydroxy group,
xylene diamines could become new starting materials for
xylene glycols. The amino groups on the benzene ring are
relatively easily transformed into hydroxy groups. For ex-
ample, active anilines can be transformed into phenols by
Table 1 Effect of Various Bases and Acid Catalystsa
catalyst
2
H N
NH2
HO
OH
MeOH
40 °C (7 MPa)
2 h
2
MXDA
MXG
5
heating under acidic conditions. However, the transfor-
mation of aliphatic amines into alcohols is relatively cum-6 Entry
Catalyst
–
Yield (%)b
bersome. Diazotization of an amino group by nitrite salt,
1
0
0
conversion into the quaternary ammonium salt by alkyl
7
halide, sulfonamidation by sulfonyl chloride, and the for-
2
3
4
5
6
7
SiO –Al O
2
2
3
8
mation of the pyridinium salt are common methods
MgO
0
(
Scheme 1). Because the aliphatic amino group is not a
good leaving group, transformation of the amino group
into a better leaving group is necessary. Therefore, these
methods require several steps and it is difficult to apply
this to commercialization. With this background, we in-
vestigated the direct transformation of amino groups into
hydroxy groups and, in this communication, we report a
cost-effective and more direct transformation of benzylic
Ca(OH)2
HClc
0
0
KOH
20
74
73
NaOMed
NaOH
8
a
Reaction conditions: MXDA (0.3 g, 2.2 mmol), catalyst (0.1 g),
MeOH (7.4 g), 240 °C (autoclave), 2 h.
SYNLETT 2012, 23, 706–710
Advanced online publication: 28.02.2012
DOI: 10.1055/s-0031-1290595; Art ID: U66211ST
x
x.
x
x.
2
0
1
2
b
GC yield. Tridecane was used as an internal standard.
Aqueous solution was used.
c
d
Methanol solution was used.
©
Georg Thieme Verlag Stuttgart · New York

LETTER
Transformation of Benzylamines into Benzyl Alcohols
707
obtained in 74 and 73% yields, respectively (entries 7 and an equivalent amount of HCl were heated in supercritical
8
).
methanol, benzyl methyl ether was produced as a main
product and benzyl alcohol was afforded in 25% yield as
a byproduct.10 Katritzky and coworkers reported that
Various reaction conditions were examined to increase
the yield (Table 2). When the amount of NaOH was re-
duced to 0.8 molar equivalent against MXDA (0.4 molar
equivalent with respect to the amino group), the reaction
was retarded and the yield was reduced. On the other
hand, the yield was not improved when the amount of
NaOH was increased to 2.7 molar equivalents (entries 1–
when benzylamine was heated in aqueous formic acid at
c
3
50 °C, benzyl alcohol was obtained in 41% yield.9 It
seems that a key for improving the yield in the process re-
ported here was that basic catalysts were used instead of
11
acidic catalysts. In order to examine whether subcritical
conditions were important or not, a reaction was per-
formed in 1-undecanol at reflux (235 °C) under atmo-
spheric pressure. Since, under these conditions, MXG was
produced in 36% yield, it was considered that pressure
was not so important and that the reaction temperature
was more critical.
3
). Excess methanol was needed; when the amount of
methanol was reduced to 14 molar equivalents against
MXDA, the yield was decreased (entries 2, 4, and 5). Con-
cerning the temperature, 240 °C was found to be optimal.
The reaction was retarded at 220 °C and some undesired
byproducts were observed at 260 °C, thus the yields were
decreased in both cases (entries 2, 6, and 7).
Table 3 Effects of Protic Solventsa
Table 2 Reaction Conditions for MXDAa
Yield (%)b
Entry
Solvents
MeOH
EtOH
Entry
NaOH/MXDA MeOH/MXDA Temp
Yield
(%)c
1
2
3
73
77
74
ratiob
0.80
1.4
ratio
110
110
110
56
(°C)
240
240
240
240
240
220
260
1
2
3
4
5
6
15
73
71
71
60
28
63
OH
4
0.7
0
2.7
OH
1.4
5
6
H O
2
1.4
14
MeOH–H O (93:7 wt)
83
2
1.4
110
110
93 (89)c
7
MeOH–H O (70:30 wt)
2
a
7
1.4
Reaction conditions: MXDA (0.3 g, 2.2 mmol), NaOH (0.1 g,
.5 mmol), alcohol (7.4 g), 240 °C (autoclave), 2 h.
GC yield. Tridecane was used as internal standard.
Isolated yield is shown in parenthesis.
2
a
Reaction conditions: MXDA (0.3 g, 2.2 mmol), NaOH in MeOH,
h, autoclave.
When NaOH/MXDA molar ratio was 0.8, the NaOH/amino group
b
2
c
b
molar ratio was 0.4.
c
GC yield. Tridecane was used as internal standard.
Through the use of these optimized conditions, a variety
of benzyl amine derivatives were investigated (Table 4).
To examine the substituent effect on the nitrogen atom,
benzylamine, N-methylbenzylamine, and N,N-dimethyl-
benzylamine were tested. Benzyl alcohol was obtained in
high yield from benzylamine, however, the yields were
very low for the other amines. Hence, it was found that
only the primary amino group could be transformed into
the hydroxy group (entries 1–3). Alkyl groups on the ben-
zene ring were tolerated, and o-, m-, p-methyl and p-tert-
butyl benzylamine gave the corresponding products in
high yields (entries 4–7). m-Chlorobenzylamine reacted
in high yield, although the yields of o- and p-chlorobenzyl-
amines were lower, due to partial methoxylation (entries
Next, protic solvents were examined (Table 3). Primary
alcohols, such as methanol, ethanol, and 1-propanol, gave
similar results (entries 1–3). In contrast, almost no reac-
tion proceeded in a secondary alcohol such as 2-propanol
(
entry 4). The reaction also did not proceed in H O (entry
2
5
). However, it was found that the addition of water to
methanol reduced the amount of byproducts and im-
proved the yield. When the reactions were performed in
mixed solvents (MeOH–H O, 97:3 and 70:30 wt%), the
2
GC yields were improved to 83 and 93%, respectively
(
entries 6 and 7). The crude MXG (93% GC yield) was
easily purified by Kugelrohr distillation to afford the pure
product in 89% yield. These strong solvent effects of al-
cohols and water clearly indicated that alcohols took part
in the reaction and played a significant role.
8
–10). In the case of methoxybenzylamines, the yield was
moderate because hydroxybenzyl alcohols were obtained
as byproducts (entries 11–13). The yield of 1-phenyletha-
nol was also low. It is considered that the steric bulk
around the amino group encumbered the reaction (entry
Reactions in supercritical and subcritical aqueous (protic)
media have recently garnered a lot of attention, and the re-
actions of benzylamines in such media were frequently re-
1
4). No product was obtained from aniline or 1,3-bis(ami-
nomethyl)cyclohexane (entries 15 and 16).
9
ported. However, there have been no reports in which
Not only benzylamine but other aromatic and hetero aro-
matic methylamines could also be used. Naphthylmethyl-
benzyl alcohols were obtained in high yield. For example,
Deka and coworkers reported that when benzylamine and
©
Thieme Stuttgart · New York
Synlett 2012, 23, 706–710

7
08
Y. Kanbara et al.
LETTER
amine and pyridinylmethylamine gave the corresponding MeOH by using NaOMe as a catalyst, no benzyl methyl
1
2
alcohols in high yields (entries 17–20).
In order to obtain information about the reaction mecha-
ether derivatives were produced (Table 1, entry 7). When
m-bis(methoxymethyl)benzene was reacted, no reaction
occurred and almost all the starting material was recov-
ered, which showed that the benzyl methyl ethers were
stable under the reaction conditions. Based on these re-
sults, it was concluded that the reaction did not proceed
1
3
nism, the time course of the reaction of MXDA was an-
alyzed (Scheme 2). Soon after the reaction started, the
amount of MXDA reduced and the amount of mono alco-
hol increased. After 50 min, the amount of mono alcohol
decreased and MXG increased. From these observations,
it was apparent that the amino groups were being trans-
formed into hydroxy group in a stepwise manner. As the
reaction proceeded, an equimolar amount of methylamine
compared to reacted amino group was generated.14–16 The
through the normal S 2 mechanism. A more precise anal-
N
ysis of the reaction is ongoing.
H2N
NH2
NaOH HO
MeOH–H O
OH
+
2MeNH2
2
2
40 °C
reaction was catalyzed by NaOH in H O–MeOH in
2
Figure 1, however, when the reaction was performed in
Scheme 2
Table 4 Substrates Scope and Limitationsa
Entry
Amine
Product
Substituent
Yield (%)b
1
2
1
2
3
R1
R = H, R = H
91 (86)
6
2
OH
1
2
N
R2
R = Me, R = H
1
2
R = Me, R = Me
c
4
5
6
o-Me
m-Me
p-Me
92 (89)
NH2
OH
93 (79)
96 (93)
Me
Me
NH2
OH
7
93 (91)
74d
t-Bu
t-Bu
8
9
0
o-Cl
m-Cl
p-Cl
NH2
OH
e
91 (86)
Cl
Cl
1
35f
1
1
1
1
2
3
o-OMe
m-OMe
p-OMe
48e
65
32
NH2
OH
e,g
MeO
MeO
1
1
4
5
23
NH2
OH
NH2
OH
0
0
NH2
NH2
NH2
OH
OH
1
1
6
7
OH
93 (85)
1
1
2
8
9
0
2-
3-
4-
82 (75)
99 (94)
70
N
N
NH2
OH
a
b
c
d
e
f
Reaction conditions: Amine/MeOH/H O/NaOH = 1:110:70:1.4 (molar ratio), 240 °C, 2 h.
GC yield. Tridecane was used as internal standard. Isolated yields after distillation or column chromatography are shown in parentheses.
Tetradecane was used as an internal standard.
2
o-Methoxybenzylalcohol was observed in 11% yield.
Amine/MeOH/H O/NaOH = 1:55:35:0.7 (molar ratio),
2
p-Methoxybenzylalcohol was observed in 32% yield.
m-Hydroxybenzylalcohol was observed in 14% yield.
g
Synlett 2012, 23, 706–710
© Thieme Stuttgart · New York

LETTER
Transformation of Benzylamines into Benzyl Alcohols
709
Tetrahedron Lett. 1981, 22, 199. (b) Müller, P.; Thi, M. P.
N. Helv. Chim. Acta 1980, 63, 2168. (c) Kato, Y.; Yen, D.
H.; Fukudome, Y.; Hata, T.; Urabe, H. Org. Lett. 2010, 12,
4
137. (d) Katritzky, A. R. Tetrahedron 1980, 36, 679.
9) (a) Siskin, M.; Katritzky, A. R. Chem. Rev. 2001, 101, 825.
b) Katritzky, A. R.; Nichols, D. A.; Siskin, M.; Murugan,
R.; Balasubramanian, M. Chem. Rev. 2001, 101, 837.
c) Katritzky, A. R.; Lapucha, A. R.; Siskin, M. Energy
(
(
(
Fuels 1990, 4, 555. (d) Katritzky, A. R.; Nichols, D. A.;
Shipkova, P. A.; Siskin, M. ACH – Models Chem. 1998, 135,
5
53.
10) (a) Deka, D. C. Indian J. Chem. Technol. 1995, 2, 197.
b) Nandi, D. K.; Palit, S. K.; Deka, D. C. J. Chem. Technol.
(
(
Figure 1 Time course of the reaction. Reaction conditions: MXDA/
Biotechnol. 1987, 38, 243. (c) Deka, D. C.; Nandi, D. K.;
MeOH/H O/NaOH = 1:110:70:1.4 (molar ratio), 240 °C. Methylam-
Palit, S. K. J. Chem. Technol. Biotechnol. 1988, 41, 95.
2
1
1b
ine yield = (methylamine mol/MXDA mol) × 0.5.
(11) (a) After our patent application, another patent
1
1c
application was published. It was reported that when
benzylamine and an equivalent amount of glycolic acid were
heated in supercritical methanol, the amide was probably
generated in situ, and then it was converted into benzyl
alcohol with good yield. (b) Kanbara, Y.; Abe, T.; Fushimi,
N. Jpn. Patent Appl. 288352, 2010. (c) Kamimura, A.;
Kaiso, K.; Sugimoto, T. PCT Int. Appl. WO 016409, 2011.
12) We examined the reaction of MXG and methylamine as a
representative of the reverse reaction, that is, transformation
of benzyl alcohols into benzylamines. So far, MXDA was
not obtained.
In summary, we have discovered that benzylamines can
be easily transformed into benzyl alcohols in high yields
by simple heating in aqueous methanol under basic condi-
1
7,18
tions.
The method is straightforward and cost-effec-
tive. Hence, we hope it will become a useful alternative
industrial route to benzyl alcohols such as xylene glycols.
(
(
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett.
13) In order to check metal leaching from the stainless-steel
autoclave, ICP analysis of the reaction solution was
performed [reaction conditions: MXDA/MeOH/H O/
2
NaOH = 1:110:70:1.4 (molar ratio), 240 °C, 2 h]. The
concentrations of Fe, Cr, Ni, and Mo were 0.018, 0.058, 0.0,
and 3.0 ppm, respectively. The reaction also conducted in a
glass vessel when it was carried out in 1-undecanol under
reflux at atmospheric pressure. Therefore, it was considered
that metal leaching from the reaction vessel was minute and
its effect was marginal.
References and Notes
(
1) (a) Ringk, W.; Theimer, E. T. In Kirk-Othmer Encyclopedia
of Chemical Technology, 3rd ed., Vol. 3; Grayson, M., Ed.;
John Wiley & Sons: New York, 1978, 796. (b) Brühne, F.;
Wright, E. In Ullmann’s Encyclopedia of Industrial
Chemistry, 5th ed., Vol. A4; Gerhartz, W., Ed.; VCH:
Weinheim, 1985, 6.
(14) MXDA, mono-alcohol, MXG, MeOH, and methylamine
were only observed, and no other byproducts or inter-
mediates were detected in the time-course GC analysis of the
reaction.
(
2) (a) Bungs, J. A. U. S. Patent 2967854, 1961. (b)Nishimura,
A. A. Polymer 1967, 8, 446. (c) Laguerre, A.; Pilard, J.-F.;
Couvercelle, J.-P.; Bunel, C.; Kebir, N.; Campistron, I. e-
Polymers 2006, no. 048, 1. (d) Ho, T.; Wynne, K. J.; Nissan,
R. A. Macromolecules 1993, 26, 7029.
(15) Reactions under N and H (initial pressure: 1 MPa) were
2
2
investigated and compared with those at 1 atm N . Even
2
(
3) (a) Adkins, H. Org. React. (N. Y.) 1954, 8, 1. (b) Onda, Y.;
under H pressure, the reaction smoothly proceeded and no
2
Kosuge, F.; Tsunoda, M. U. S. Patent 4990690, 1989.
differences were observed. Combined with the fact that no
byproducts or intermediates were detected, it was thus not
considered probable that the reaction proceeded via imine or
aldehyde intermediates.
(
c) Toba, M.; Tanaka, S.; Niwa, S.; Mizukami, F.; Koppany,
Z. Appl. Catal., A 1999, 189, 243.
(
4) (a) Weissermel, K.; Arpe, H.-J. Industrial Organic
Chemistry, 3rd ed.; VCH: Weinheim, 1997, 400.
(16) From the fact that the yields of N-methyl and N,N-dimethyl
benzylamine were very low, it was apparent that N-
methylation by methanol did not occur during the reaction
(17) Preparation of m-xylene glycol (MXG); Typical
Procedure (Table 1)
(
b) Wittcoff, H. A.; Reuben, B. G.; Plotkin, J. S. Industrial
Organic Chemicals, 2nd ed.; Wiley-Interscience: New
Jersey, 2004, 327. (c) Saunders, J. H. In Kirk-Othmer
Encyclopedia of Chemical Technology, 3rd ed., Vol. 18;
Grayson, M., Ed.; John Wiley & Sons: New York, 1982,
To a 30-mL autoclave was added MXDA (0.30 g, 2.2
400. (d) Kohan, M. I. In Ullmann’s Encyclopedia of
mmol), catalyst (0.1 g), and MeOH (7.4 g), and then the
Industrial Chemistry, 5th ed., Vol. A21; Elvers, B.;
Hawkins, S.; Schulz, G., Eds.; VCH: Weinheim, 1992, 180.
5) (a) Clarke, H. T.; Hartman, W. W. Org. Synth., Coll. Vol. I
inner gas was replaced by N . The mixture was heated at
2
240 °C for 2 h, and then cooled in an ice–water bath. The
yield was determined by GC analysis using tridecane as
internal standard. The crude products were purified by
Kugelrohr distillation.
(
(
1
932, 455. (b) Boswell, D. E.; Brennan, J. A.; Landis, P. S.
Tetrahedron Lett. 1970, 60, 5265.
1
6) (a) Baumgarten, R. J. J. Chem. Educ. 1966, 43, 398.
Spectral data of MXG (Table 3, entry7): H NMR (500
MHz, CD OD): d = 4.60 (s, 4 H), 7.24–7.34 (m, 4 H). 13C
(b) Kirmse, W. Angew. Chem. Int. Ed. Engl. 1976, 15, 251.
(c) Naik, R.; Pasha, M. A. Synth. Commun. 2005, 35, 2823.
(d) Moss, R. A. Chem. Eng. News 1971, 49, 28.
3
NMR (126 MHz, CD OD): d = 65.2, 126.6, 126.9, 129.4,
3
142.8.
(
7) Brasen, W. R.; Hauser, C. R. Org. Synth., Coll. Vol. IV 1963,
82.
8) (a) Curtis, V. A.; Kuntson, F. J.; Baugarten, R. J.
(18) Preparation of benzyl alcohol derivatives (Table 4): To a
30-mL autoclave was added benzylamine derivative (4.4
mmol for entry 9, 11, and 12, and 2.2 mmol for the others),
5
(
©
Thieme Stuttgart · New York
Synlett 2012, 23, 706–710

7
10
Y. Kanbara et al.
LETTER
powdered NaOH (0.1 g, 2.5 mmol), MeOH (7.4 g), and H O
other products were purified by Kugelrohr distillation.
Spectral data of benzyl alcohol (Table 4, entry1): H NMR
2
1
(
2.9 g), then the inner gas was replaced by N . The mixture
2
was heated at 240 °C for 2 h, and then cooled in an ice–water
bath. The reaction mixture was neutralized with 1 M HCl
(500 MHz, CD OD): d = 4.59 (s, 2 H), 7.23–7.33 (m, 5 H).
3
1
3
C NMR (126 MHz, CD OD): d = 65.2, 126.6, 126.9, 129.4,
3
(2.5 mL, 2.5 mmol). GC yield was determined by GC
142.8.
analysis using tetradecane (entry 4) and tridecane (for the
others) as internal standard. For entry 1, the solutions were
dried over Na SO , concentrated, and purified by silica-gel
All other products were characterized by comparison of GC
retention time with chemical reagents purchased from
commercial suppliers.
2
4
column chromatography (hexane–EtOAc, 75:25 v/v). The
Synlett 2012, 23, 706–710
© Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.