Iron/amino acid catalyzed direct N-alkylation of amines with alcohols

Article abstract of DOI:10.1002/anie.201006660

(Chemical Equation Presented) Ironing it out: The straightforward N-alkylation using alcohols and iron/amino acid catalysis is described (see scheme). The reaction does not proceed by the conventional "borrowing hydrogen" mechanism, but appears to involve a substitution pathway (S _N) at the sp₃ carbon atom bearing the hydroxy group of the alcohol. Developing a catalyst that is effective at a near neutral pH was key to the successful N-alkylation.

Full text of DOI:10.1002/anie.201006660

Communications
DOI: 10.1002/anie.201006660
N-alkylation
Iron/Amino Acid Catalyzed Direct N-Alkylation of Amines with
Alcohols**
Yingsheng Zhao, Siong Wan Foo, and Susumu Saito*
Higher amines are widely used in both the bulk and fine
chemical industries as fundamental materials, additives, dyes,
agrochemicals, etc.[1] Therefore, the development of a selec-
tive N-alkylation method that is cost effective, salt free, and
Scheme 1. General scheme of direct N-alkylation using an alcohol.
Cp*H=1,2,3,4,5-pentamethylcyclopenta-1,3-diene.
environmentally benign is of considerable interest. However,
the most common strategies to produce secondary or tertiary
amines involve treating a primary amine with an alkylating
reagent having a good leaving group (RX: X = halide, OTs,
OTf, etc.). The main drawback of this conventional approach
is the generation of a stoichiometric amount of wasteful
(in)organic salts, as well as low secondary/tertiary amine
product selectivity. In recent years, N-monoalkylation was
demonstrated, wherein an alcohol[2] is used in place of RX
and transition-metal catalysis involves Ru,[3] Ir,[4] Cu,[5] or
Ag[6] catalyst precursors. These transition-metal-catalyzed
N-alkylations are usually redox-type reactions involving
“borrowing hydrogen”[2a,b] or “hydrogen autotransfer”[2c]
mechanisms (alcohol oxidation/imine formation/imine hydro-
genation). Although these reactions have found many appli-
cations using benzylic-type and saturated alcohols as the
N-alkylating agents,[2] the supposedly generated metal hy-
dride species are incompatible with some functional groups,
including olefins. Herein we report a novel straightforward
method of N-alkylation using alcohols and iron catalysis,[7]
which involves a substitution (SN) at the sp3-carbon atom
bearing the hydroxy group of the alcohol (Scheme 1). The
development of a catalyst that is effective at near neutral pH
was key to the successful N-alkylation.
of 2a), dl-pyroglutamic acid (4a; 6 mol%) at 1608C for
24 hours gave, after column chromatography on silica gel, the
corresponding monoalkylated amine 3aa in an yield of 71%
upon isolation (Table 1, entry 1). Notably, the dialkylation
reaction led to less than 5% of the tertiary amine as
ascertained by GC/MS and 1H NMR analyses. By using a
smaller amount of 4a (3 mol%) under otherwise identical
reaction conditions, a slightly lower productivity was obtained
Table 1: N alkylation of 1a with 2a to give 3aa using different reaction
conditions.[a]
Entry
FeX3
Ligand
Cp*H [mol%]
Conversion [%][b]
1
2
FeBr3
FeBr3
FeBr3
FeCl3
FeBr3
FeBr3
none
FeBr3
FeBr3
FeBr3
FeBr3
FeBr3
FeBr3
FeBr3
FeBr3
4a
4a
4a
4a
none
none
4a
4b
4c
4d
4e
4 f
0
5
5
5
0
5
0
0
0
5
5
5
5
5
5
77 (71)
85 (82)
99 (94)
59 (55)
43 (41)
<30
<1
45
58
46 (40)
52 (49)
44 (39)
68 (66)
60 (57)
60 (53)
3[c]
4
5[c]
6
Treatment of a 1,2,4-trimethylbenzene (1,2,4-TMB) solu-
tion of aniline 1a (6 mmol) with benzyl alcohol 2a (3 mmol)
in the presence of FeBr3 (3 mol%, based on the molar amount
7
8
9
10
11
12
13
14
15
[*] Dr. Y. Zhao, S. W. Foo, Dr. S. Saito
Department of Chemistry, Graduate School of Science & Research
Center for Materials Science, Nagoya University
Chikusa, Nagoya 464-8602 (Japan)
4g
4h
4i
Fax: (+81)52-789-5945
E-mail: saito.susumu@f.mbox.nagoya-u.ac.jp
Homepage: http://www.chem.nagoya-u.ac.jp/~susumu/
Dr. S. Saito
Institute for Advanced Research, Nagoya University
Chikusa, Nagoya 464-8601 (Japan)
[a] Unless otherwise specified, the reaction conditions were: 1a
(6 mmol), 2a (3 mmol), FeBr3 (3 mol%), 4a (6 mol%), Cp*H
(5 mol%) in 1,2,4-TMB (0.3 mL) at 1608C for 18 h. [b] Conversion was
determined by GC analysis using biphenyl as internal standard. The
values in parentheses are the yields of the isolated product 3aa. [c] 24 h.
[**] This work was supported by a Grant-in-Aid for Basic Research (B)
and Scientific Research on Priority Areas “Advanced Molecular
Transformations of Carbon Resource” from the MEXT, as well as by
funds from Asahi Glass, Asahi Kasei Chemicals, Daiso, Kuraray,
Mitsubishi Chemical, Mitsui Chemicals, Nissan Chemical Ind., and
Sumitomo Chemical. S.S. wishes to thank Prof. R. Noyori (Nagoya
University & RIKEN) for fruitful discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201006660.
3006
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3006 –3009

Table 2: Coupling between various amines 1 and alcohols 2.[a]
(conversion of 2a: 68%). Without 4a, the yield of 3aa
decreased by almost half, even with a reaction time of
24 hours (entry 5). The use of other amino acids such as 4b–g
gave moderate yields (entries 8–13). The ester and amide
derivatives 4h and 4i were also unable to give satisfactory
results (entries 14 and 15, respectively). The combination of
FeBr3/4a with Cp*H (5 mol%) showed the highest catalytic
activity, giving 3aa in 82% yield (entry 2). The use of other
diene additives, such as cycloocta-1,5-diene (cod) or 1,2,3,4,5-
pentaphenylcyclopentadiene, gave results (3aa: ca. 70%)
comparable to those obtained without using any diene
additives (entry 1). By prolonging the reaction time to
24 hours the conversion of 2a was maximized, thus giving a
94% yield of 3aa upon isolation (entry 3). In the absence of
4a, a combination of catalytic FeBr3 (3 mol%) and Cp*H
(5 mol%) gave a yield of 3aa that was lower than that
obtained with FeBr3 alone (entry 5 vs. 6). FeCl3 was less
effective (3aa: 55%) than FeBr3 (entry 4), and the reaction
did not take place with 4a alone (entry 7). When we used a
1:1:1 ratio of 1a/2a/PhNMe2 and 1a/2a/iPr2NEt under other-
wise identical reaction conditions, 3aa was obtained in 30%
and < 5%, respectively.
Entry Amine 1
Alcohol 2
Product 3
Yield
[%][b]
1
2
3
1a (X=H)
2b (Y=4-Cl)
2c (Y=3-Br)
2d (Y=2-I)
2e (Y=4-CF3) 3ae (Y=4-CF3)
2 f (Y=3-Me) 3af (Y=3-Me)
2g (Y=3-OMe) 3ag (Y=3-OMe)
3ab (Y=4-Cl)
3ac (Y=3-Br)
3ad (Y=2-I)
91
1a
1a
1a
1a
1a
94
88[e]
92
4[c]
5
90
6
93
7
8
9
10
11
1b (X=4-OMe) 2a (Y=H)
1c (X=2-OMe) 2a
3ba (X=4-OMe)
3ca (X=2-OMe)
3da (X=2-F)
3ea (X=3-CF3)
3 fa (X=3-CN)
77[f]
78
1d (X=2-F)
2a
2a
2a
90
1e (X=3-CF3)
1 f (X=3-CN)
90[g]
64[g]
12
1g
2a
3ga
78
(X=3-SO2NH2)
Encouraged by this discovery, we next examined a range
of different substrates under the optimal reaction conditions
(FeBr3 (3 mol%), 4a (6 mol%), Cp*H (6 mol%) in 1,2,4-
TMB (Fe: 0.075m), 1608C, 24 h). The results are summarized
in Table 2. Coupling reactions between benzylic-type primary
alcohols and anilines having different electronic properties
mostly gave the expected products in high yields (entries 1–
15). Even when the (halophenyl)methanols 2b–d or 6-amino-
2-bromopyridine (1h) were used, the results afforded the
N-alkylation product in good yield and carbon–halogen bond
cleavage did not occur (entries 1–3 and 13). There are a
paucity of reports on direct coupling between anilines and
allylic alcohols;[8] however, allylic alcohols and benzylic-type
secondary alcohols can be classified as a family of compounds
that readily generate carbocation equivalents and are far
more reactive than the benzylic-type primary alcohols 2a–g.
In fact, under the reaction conditions for the Lewis acid
catalyzed SN-type N-alkylations,[9] in which carbocation
species may favorably form from allylic alcohols, the
N-alkylation using 2a rarely proceeded. Additionally, under
13
14
1h
2a
2a
3ha
83
87
PhNHMe
1i
PhN(Bn)Me
3ia
15
1j
2a
2a
3ja
89
90
16[c] 1k
3ka
17[c] 1l
2a
2a
3la
86
83
acidic conditions such as those employing FeCl3,[10b] NbCl5,[10c]
18[d] 1m
3ma
[11]
and AuCl3
as catalyst precursors, electron-rich aromatic
[a] Reaction conditions: 1/2/FeBr3/4a/Cp*H=200:100:3:6:6, in 1,2,4-
TMB at 1608C for 24 h. [b] Yield of isolated, purified product based on 2.
Yield of the N-dialkylation product was consistently <5%. [c] 1/2/FeBr3/
4a/Cp*H=200:100:5:10:10 at 1808C for 36 h. [d] 1/2/FeBr3/4a/
Cp*H=200:100:10:20:20 at 2008C for 36 h. [e] 36 h. [f] 48 h. [g] 1408C.
substrates were prone to undergoing Friedel–Crafts-type
alkylation[10] with alcohols. Fortunately, this undesirable
alkylation, which should have occurred on the aromatic ring
of the solvent (1,2,4-TMB) or aniline derivatives,[10b] was
almost completely suppressed. The finding suggests that a
more neutral pH environment was important.
The NH2 of sulfonamides were readily alkylated with
alcohols using several of the previously described meth-
ods,[2c,3h,i,5b,c,9] but this sulfonamide NH2 remained untouched
in the presence of the NH2 group of aniline when using the
present reaction (entry 12). The benzyl amines 1k–m that
have a nitrogen atom which is more basic than that of anilines,
were also suitable substrates, albeit at a higher loading of
FeBr3 and an elevated reaction temperature (1808C;
entries 16–18).
The least reactive alcohols, such as fully saturated
alcohols, were also tested (Table 3). By merely increasing
the reaction temperature, the secondary amines derived from
monoalkylation were obtained (entries 1–5). Even at a high
temperature, the secondary alcohol 2k and the interior olefin
of 2l were well tolerated (entries 4 and 5). The reaction
pathway involving a carbocation (SN1) is unlikely to occur
with primary alcohols. The a/g (> 99%) and E/Z (or Z/E)
(> 99%) selectivities of the reaction with allylic alcohols
Angew. Chem. Int. Ed. 2011, 50, 3006 –3009
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3007

Communications
Table 3: Coupling between 1a and various alcohols 2.[a]
using different alcohols to check the likely mechanistic
scenario (Scheme 2). (3-Methylphenyl)methanol (2 f) was
converted into 3af in 94% yield, and was not labeled with
Entry Alcohol 2 Product 3
T
t
Yield
[%][b]
[8C] [h]
1
2
3
2h
2i
200 36 3ah
200 36 3ai
200 36 3aj
88
85
86
2j
Scheme 2. Experiment using deuterium-labeled alcohol [D2]-2a.
deuterium. In contrast, full incorporation (> 99%) of deute-
rium atoms into the benzylic position of product [D2]-3aa was
rigorously ascertained by 1H NMR and HRMS methods. This
result is in sharp contrast to the observation of D–H exchange
in Ru-[3h,i] and Cu-mediated[5b] N-alkylation with alcohols.
Selective mono- or dimethylation of the aniline nitrogen
atom is also possible by varying the ratio of aniline and
CH3OH used (Scheme 3). Three deuterium atoms of CD3OH
were preserved (deuterium incorporation into N,N-dimethy-
laniline: > 99%) at the same carbon atom.
4
2k
200 36 3ak
70
5
6
7
2l
200 36 3al
140 24 3am
100 36 3an
85
2m
2n
83[c]
73[c]
Scheme 3. N alkylation of 1a using either CH3OH or CD3OH. Reaction
conditions: [a] 1a/CH3OH/FeBr3/4a/Cp*H=250:100:5:10:10, in 1,2,4-
TMB at 2008C for 36 h; [b] 1a/CD3OH/FeBr3/4a/
8
2o
2p
100 36 3ao
70[d]
Cp*H=100:266:5:10:10, in 1,2,4-TMB at 2008C for 36 h.
9[e]
100 24 3ap
60 24 3aq
80
74
In summary, we have developed a highly selective and
practical procedure for producing higher amines using acid-
base cooperative catalysis[12] based on an iron(III)/amino acid
combination that obviates the need for activated leaving
groups. The broad substrate scope and functional group
tolerance were highlighted with respect to each amine and
alcohol component. These characteristic features provide a
mechanistically distinct alternative to borrowing hydrogen
strategies. The catalytic system enabled selective N-mono-
alkylation (over-alkylation < 5%) in most cases. Because of
the inherent nature of more neutral pH conditions, elevated
temperatures were required to obtain a good product yield.
Nonetheless, prolonged heating at 160–2008C is acceptable
for an industrial process because the extra heat energy
recovered can be effectively reused for other purposes.
10[e] 2q
[a] Reaction conditions: 1/2/FeBr3/4a/Cp*H=200:100:3:6:6, in 1,2,4-
TMB. [b] Yield of isolated, purified products based on 2. Yield of N-
dialkylation product was consistently <5%. [c] a/g=>99:1; E/Z=>
99:1. [d] a/g=>99:1; E/Z=1:>99. [e] Run in toluene instead of 1,2,4-
TMB.
2m–o were excellent (entries 6–8), thus supporting the
proposed SN2 pathway. If the reaction mechanism involved
either a borrowing hydrogen or an SN1 pathway, the
Z geometry of 2o should have been lost.
So far, the mechanism of other transition-metal-catalyzed
-alkylations of amines using simple saturated alcohols has
been uniformly considered a borrowing hydrogen path-
way.[1–6] In contrast, the present catalysis appears to involve
a different mechanism. We performed a crossover experiment
Experimental Section
Typical procedure: Iron(III) bromide (27 mg, 0.09 mmol), dl-pyro-
glutamic acid (4a; 23.5 mg, 0.18 mmol), 1,2,3,4,5-pentamethylcyclo-
penta-1,3-diene (Cp*H: 23.5 mL 0.15 mmol), benzyl alcohol (2a;
3008 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3006 –3009

324.3 mg, 3.0 mmol), aniline (1a; 558.6 mg, 6.0 mmol), and 1,2,4-
trimethylbenzene (0.3 mL) were added to a dry flask that was fitted
with a Young’s stopcock in an argon (Ar) atmosphere. The reaction
mixture was stirred at 1608C under Ar for 24 h and then cooled to
room temperature and diluted with ethyl acetate (50 mL). The
resulting mixture was washed with saturated aqueous NH4Cl (2 ꢀ
20 mL) and then the aqueous phase was extracted with ethyl acetate
(2 ꢀ 20 mL). The combined organic phases were dried over sodium
sulfate and filtered. The filtrate was then concentrated under reduced
pressure using a rotary evaporator. The crude reaction mixture was
purified by column chromatography on silica gel (n-hexane/ethyl
acetate 98:2) to give the N-phenylbenzylamine (3aa) as a white solid
in 94% yield (516 mg, 2.82 mmol).
2009, 131, 1775 – 1779; j) B. Gnanaprakasam, J. Zhang, D.
Milstein, Angew. Chem. 2010, 122, 1510 – 1513; Angew. Chem.
Int. Ed. 2010, 49, 1468 – 1471; k) S. Imm, S. Bꢂhn, L. Neubert, H.
Neumann, M. Beller, Angew. Chem. 2010, 122, 8303 – 8306;
Angew. Chem. Int. Ed. 2010, 49, 8126 – 8129.
[4] a) K.-i. Fujita, K. Yamamoto, R. Yamaguchi, Org. Lett. 2002, 4,
2691 – 2694; b) K.-i. Fujita, Y. Enoki, R. Yamaguchi, Tetrahedron
2008, 64, 1943 – 1954; c) B. Blank, M. Madalska, R. Kempe, Adv.
Synth. Catal. 2008, 350, 749 – 758; d) D. Balcells, A. Nova, E.
Clot, D. Gnanamgrai, R. H. Crabtree, O. Eisenstein, Organo-
metallics 2008, 27, 2529 – 2535; e) B. Blank, S. Michlik, R.
Kempe, Adv. Synth. Catal. 2009, 351, 2903 – 2911; f) H. Aramoto,
Y. Obora, Y. Ishii, J. Org. Chem. 2009, 74, 628 – 633; g) B. Blank,
R. Kempe, Chem. Eur. J. 2009, 15, 3790 – 3799.
[5] Copper/aluminum hydrotalcite: a) P. R. Likhar, ; R. Arundhathi,
M. L. Kantam, P. S. Prathima, Eur. J. Org. Chem. 2009, 5383 –
5389; R. Arundhathi, M. L. Kantam, P. S. Prathima, Eur. J. Org.
Chem. 2009, 5383 – 5389; b) F. Shi, M. K. Tse, X. Cui, D. Gꢄrdes,
D. Michalik, T. Y. Deng, M. Beller, Angew. Chem. 2009, 121,
6026 – 6029; Angew. Chem. Int. Ed. 2009, 48, 5912 – 5915; c) X.
Cui, F. Shi, M. K. Tse, D. Gꢄrdes, K. Thurow, M. Beller, Y. Deng,
Adv. Synth. Catal. 2009, 351, 2949 – 2958.
[6] B. Sreedhar, P. S. Reddy, M. A. Reddy, B. Neelima, R. Arund-
hathi, Tetrahedron Lett. 2007, 48, 8174 – 8177.
[7] A. Correa, O. G. Manecheno, C. Bolm, Chem. Soc. Rev. 2008, 37,
1108 – 1117.
[8] a) F. Ozawa, H. Okamoto, S. Kawagishi, S. Yamamoto, T.
Minami, M. Yoshifuji, J. Am. Chem. Soc. 2002, 124, 10968 –
10969; b) Y. Yamashita, A. Gopalarathnam, J. F. Hartwig, J.
Am. Chem. Soc. 2007, 129, 7508 – 7509; c) M. Utsunomiya, Y.
Miyamoto, J. Ipposhi, T. Ohshima, K. Mashima, Org. Lett. 2007,
9, 3371 – 3374.
[9] a) V. Terrasson, S. Marque, M. Georgy, J.-M. Campagne, D.
Prim, Adv. Synth. Catal. 2006, 348, 2063 – 2067; b) H. Qin, N.
Yamagiwa, S. Matsunaga, M. Shibasaki, Angew. Chem. 2007,
119, 413 – 417; Angew. Chem. Int. Ed. 2007, 46, 409 – 413; c) C. R.
Reddy, P. P. Madhavi, A. S. Reddy, Tetrahedron Lett. 2007, 48,
7169 – 7172; d) U. Jana, S. Maiti, S. Biswas, Tetrahedron Lett.
2008, 49, 858 – 862.
[10] a) M. Noji, T. Ohno, K. Fuji, N. Futaba, H. Tajima, K. Ishii, J.
Org. Chem. 2003, 68, 9340 – 9347; b) I. Iovel, K. Mertins, J.
Kischel, A. Zapf, M. Beller, Angew. Chem. 2005, 117, 3981 –
3985; Angew. Chem. Int. Ed. 2005, 44, 3913 – 3917; c) J. S. Yadav,
D. C. Bhunia, K. V. Krishna, P. Srihari, Tetrahedron Lett. 2007,
48, 8306 – 8310.
Received: October 23, 2010
Revised: January 13, 2011
Published online: February 25, 2011
Keywords: alcohols · amino acids · homogeneous catalysis ·
.
iron · N-alkylation
[1] a) A. Seayad, M. Ahmed, H. Klein, R. Jackstell, T. Gross, M.
Beller, Science 2002, 297, 1676 – 1678; b) S. A. Lawerencem,
Amines: Synthesis Properties and Applications, Cambridge
University, Cambridge, 2004; c) B. R. Brown, The Organic
Chemistry of Aliphatic Nitrogen Compounds, Cambridge Uni-
versity, Cambridge, 2004.
[2] Reviews: a) M. H. S. A. Hamid, P. A. Slatford, J. M. J. Williams,
Adv. Synth. Catal. 2007, 349, 1555 – 1575; b) T. D. Nixon, M. K.
Whittlesey, J. M. J. Williams, Dalton Trans. 2009, 753 – 762; c) G.
Guillena, D. J. Ramꢁn, M. Yus, Chem. Rev. 2010, 110, 1611 –
1641; d) G. E. Dobereiner, R. H. Crabtree, Chem. Rev. 2010, 110,
681 – 703.
[3] Selected examples: a) S. Murahashi, K. Kondo, T. Hakata,
Tetrahedron Lett. 1982, 23, 229 – 232; b) Y. Watanabe, Y. Tsuji,
H. Ige, Y. Ohsugi, T. Ohta, J. Org. Chem. 1984, 49, 3359 – 3363;
c) A. Tillack, D. Hollmann, K. Mevius, D. Michalik, S. Bꢂhn, M.
Beller, Eur. J. Org. Chem. 2008, 4745 – 4750; d) M. H. S. A.
Hamid, J. M. J. Williams, Chem. Commun. 2007, 725 – 727; e) S.
Naskar, M. Bhattacharjee, Tetrahedron Lett. 2007, 48, 3367 –
3370; f) S. Bꢂhn, D. Hollmann, A. Tillack, M. Beller, Adv.
Synth. Catal. 2008, 350, 2099 – 2103; g) C. Gunanathan, D.
Milstein, Angew. Chem. 2008, 120, 8789 – 8792; Angew. Chem.
Int. Ed. 2008, 47, 8661 – 8664; h) M. H. S. A. Hamid, C. L. Allen,
G. W. Lamb, A. C. Maxwell, H. C. Maytum, A. J. A. Watsom,
J. M. J. Williams, J. Am. Chem. Soc. 2009, 131, 1766 – 1774; Ru/
Fe3O4: i) F. Shi, M. K. Tse, S. Zhou, M.-M. Pohl, J. Radnik, S.
Hꢃbner, K. Jꢂhnisch, A. Brꢃckner, M. Beller, J. Am. Chem. Soc.
[11] P. Rubenbauer, T. Bach, Adv. Synth. Catal. 2008, 350, 1125 –
1130.
[12] Since our system comprises FeBr3/4a·amine as a weak acid and
amine as a weak base, both should be working cooperatively.
Angew. Chem. Int. Ed. 2011, 50, 3006 –3009
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3009