Reagent concentration effects in the REM resin solid phase synthesis of tertiary amines

Article abstract of DOI:10.1016/S0040-4020(03)00176-5

The use of reagent concentration has resulted in increased rates for all stages of the REM resin synthesis of tertiary amines. These increases in rate translate into faster reaction times, higher yields and lower reagent consumption. Of the methods examin

Full text of DOI:10.1016/S0040-4020(03)00176-5

Tetrahedron 59 (2003) 2137–2145
Reagent concentration effects in the REM resin solid phase
synthesis of tertiary amines
*
J. Richard Morphy, Zoran Rankovic and Mark York
Department of Medicinal Chemistry, Organon Laboratories Ltd., Newhouse, ML1 5SH, Scotland, UK
Received 6 November 2002; revised 10 January 2003; accepted 30 January 2003
Abstract—The use of reagent concentration has resulted in increased rates for all stages of the REM resin synthesis of tertiary amines. These
increases in rate translate into faster reaction times, higher yields and lower reagent consumption. Of the methods examined, the most
successful was the use of perfluorous solvents, either alone or with a small amount of organic co-solvent. q 2003 Elsevier Science Ltd. All
rights reserved.
1
. Introduction
which on exposure to a mild base releases the 38 amine
product 4 from the resin (Hofmann elimination) whilst
regenerating starting material 1 (Scheme 1).
The use of solid phase organic synthesis as a tool for the
synthesis of chemical libraries is an area of great importance
in both academia and in the pharmaceutical industry. One
1
of the drawbacks to such methodology, however, is the large
excesses of reagents and extended reaction times often
required to force solid phase reactions to completion. The
use of excess reagents is particularly undesirable from an
environmental point of view and from an economic
standpoint in the case of costly or commercially unavailable
reagents. One possible strategy to combat these drawbacks
is the use of reagent concentration effects, whereby the
effective concentration of reagent at the functionalised sites
of the resin is increased. The simplest way to concentrate
2
reagents is of course to lower solvent volumes, or remove
solvent from the process altogether. Another possible
approach is the use of solvents in which the reagents are
poorly soluble, resulting in an increased concentration of
3
reagents within the polymer. A particularly successful
group of solvents here are the perfluorous organic solvents
which are well known for their immiscibility with common
Scheme 1.
4
organic solvents, and have recently been shown to provide
5
reagent concentration in the REM resin cycle. Herein we
present our results using reagent concentration in the
synthesis of 38 amines on REM resin.
2. Results and discussion
2.1. Michael addition
REM resin methodology is an efficient approach for the
6
solid phase synthesis of 38 amines. In its simplest form the
REM resin, 1 undergoes Michael addition with a 28 amine to
give a polymer bound 38 amine, 2. Quaternisation of 2 with
an alkyl halide then gives a quaternary ammonium salt 3,
Our initial efforts into the application of reagent concen-
tration effects in the REM resin cycle were concentrated on
the Michael addition (1!2) where, under standard con-
ditions, 5–10 equiv. of amine and 18 h reaction times in
N,N-dimethylformamide (DMF) are required. In order to
investigate the possibility of reagent concentration within
this process REM resin 1 (polystyrene with 1% divinyl-
benzene crosslinking, 2 mmol/g) was suspended in various
Keywords: tertiary amines; perfluorous solvents; polymer support; reagent
concentration; solid-phase synthesis; REM resin.
*
Corresponding author; e-mail: r.morphy@organon.co.uk
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00176-5

2138
J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
Table 2.
a
Yield (%) 6
Entry
Perfluorohexane (mL) and DMF (0.03 mL)
1
2
3
4
5
6
7
8
9
0
0.05
0.1
0.15
0.2
0.25
0.5
0.75
1
13
15
17
16
50
.95
.95
.95
.95
.95
.95
Scheme 2.
1
1
0
1
2
5
solvents in the presence of amine 5 (2 equiv.) at 208C for 3 h
5
a
(
(
Scheme 2). This was followed by standard quaternisation
BnBr 10 equiv., DMF, 18 h) and cleavage (DIEA 2 equiv.,
Michael reaction performed with 0.05 mmol resin and 0.1 mmol amine at
2
08C for 3 h. The amount of perfluorohexane was varied as indicated.
Yields determined by H NMR using N-methylmaleimide as an internal
standard.
DCM, 6 h) to give the expected product 6. The yields are
shown in Table 1. Product purities were effectively
independent of solvent and were greater than 90% as
determined by HPLC.
a
1
effect within the resin. Presumably the perfluorous species
cannot enter the resin and so the only liquid within the resin
interior will be the amine reagent. By using a small amount
of co-solvent the liquid volume can be increased to a level
more conducive to diffusion throughout all the resin beads
and allow some resin swelling. The high yields seen with
perfluorous solvents were particularly surprising when it is
considered that the resins used showed little or no swelling
in these solvents (swelling of 1 in DMF¼4.4 mL/g, in
perfluorohexane¼0.4 mL/g, in perfluorohexane with 1 mL
As anticipated aliphatic perfluorous solvents gave signifi-
cantly greater yields than DMF, which failed to give any
product under standard dilution conditions (Table 1, entry 1
versus entries 14–17). Interestingly, aromatic perfluorous
solvents did not give any product. When perfluorohexane
was combined with a small amount of either DMSO or DMF
as a co-solvent the yield was generally greater than three
times that observed when that volume of co-solvent was
used alone (Table 1, entries 18, 19 versus entries 2, 4),
although the yield seen with only a small volume of co-
solvent was itself significantly greater than that seen when
the standard volume of solvent was used. The yield seen
when perfluorohexane was used in conjunction with a co-
solvent was similar to that seen when neat amine
7
DMF per mg resin¼1.7 mL/g). The use of other solvents in
which amine solubility and resin swelling were poor gave
mixed results with water and methylcyclohexane giving no
product and approximately 40% 6 seen when silane based
solvents were used in conjunction with a co-solvent (Table
1, entries 6, 7, 9, 11). In order to confirm that the elevated
yields seen with perfluorohexane were due to a reagent
concentration effect, the effect on the yield of 6 of varying
the amount of perfluorohexane was investigated (Table 2).
The results seemed to confirm the reagent concentration
hypothesis with quantitative yields of product at all volumes
(
150 equiv.) was used as solvent. The success when a
small amount of co-solvent was used in conjunction with a
perfluorous solvent was thought to be due to a solvation
Table 1.
a
Yield (%) 6
Entry
Solvent
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
DMF
,5
29
,5
24
.95
,5
,5
,5
38
8
43
,5
,5
58
65
46
57
88
Table 3.
DMF—0.03 mL
DMSO
DMSO—0.03 mL
Neat amine
Water
Methylcyclohexane
a
Yield (%)
b
Entry
Volume of
DMF added
DMF
b
only
Perfluorohexane
c
and DMF
(
mL)
Hexamethyldisiloxane
Hexamethyldisiloxane and 0.03 mL DMF
Silicone 200
c
1
2
3
4
5
6
7
8
0.01
0.03
0.04
0.05
0.075
0.1
0.125
0.15
0.175
0.2
0.3
0.4
18
12
76
75
74
76
65
64
64
60
60
59
30
18
,5
0
1
2
3
4
5
6
7
8
9
0
c
Silicone 200 and 0.03 mL DMF
,5
,5
,5
,5
,5
,5
,5
,5
,5
,5
,5
Hexafluorobenzene
Octafluorotoluene
Perfluorodecalin
Perfluorotripropylamine
Perfluoromethylcyclohexane
Perfluorohexane
9
c
Perfluorohexane and 0.03 mL DMSO
c
Perfluorohexane and 0.03 mL DMF
None
10
11
12
94
,5
1
3
0.5
Michael reaction performed with 0.05 mmol resin and 0.1 mmol amine at
208C for 3 h. Solvent volume was 1 mL unless otherwise stated.
Michael reaction (0.05 mmol resin, 0.1 mmol amine) performed for 2 h.
a
a
1
Yields determined by H NMR using N-methylmaleimide as an internal
1
Yields determined by H NMR using N-methylmaleimide as an internal
standard.
Approx. 150 equiv. amine used as solvent.
Represents the use of a standard amount (1 mL) of the named solvent with
standard.
Michael reaction performed in DMF.
Michael reaction performed in perfluorohexane (1 mL) with the stated
amount of DMF added.
b
c
b
c
0.03 mL DMSO/DMF.

J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
2139
Table 4.
a
Equiv. amine 5
Yield (%)
c
c
No
Solvent
DMF
DMF
0.03 mL
DMSO
0.03 mL
Perfluorohexane
Perfluorohexane and
d
DMF 0.03 mL
Perfluorohexane and
d
DMSO 0.03 mL
b
2
4
6
8
26
44
48
63
65
72
.95
,5
8
12
13
23
40
56
73
92
41
43
50
70
74
.95
.95
62
75
83
91
95
90
.95
.95
80
94
90
94
95
.95
95
.95
.95
.95
.95
.95
.95
14
30
28
39
73
10
15
20
Michael reaction performed for 2 h.
a
b
c
d
1
Yields determined by H NMR using N-methylmaleimide as an internal standard.
Amine was pipetted onto dry resin.
All reactions performed with 0.05 mmol resin in 1 mL solvent unless otherwise stated.
Represents the use of a standard amount (1 mL per 0.05 mmol resin) of perfluorohexane with 0.03 mL DMSO/DMF.
Table 5.
(Table 3). As can be seen, when DMF was used alone some
degree of reagent concentration could be seen at solvent
volumes up to 0.03 mL (Table 3, entries 1 and 2). With
perfluorohexane and a co-solvent, however, approximately
75% yield was seen for all solvent volumes up to 0.05 mL
(Table 3, entries 1–4). The yield then began to decrease as
the reagent solution was diluted, although increases in yield
compared to the same volume of DMF alone were seen with
up to 0.4 mL of DMF (Table 3, entries 5–12). This may
have implications for other reactions where appreciable
volumes of co-solvent are required to solubilise reagents.
a
Yield (%) 6
Entry
Solvent
1
2
3
4
5
6
7
8
9
DMF
DMF—0.2 mL
DMF—0.1 mL
DMF—0.05 mL
Perfluorohexane
,5
6
9
13
20
55
64
44
5
b
b
Perfluorohexane and 0.2 mL DMF
Perfluorohexane and 0.1 mL DMF
Perfluorohexane and 0.05 mL DMF
None
b
Michael reaction performed with 0.05 mmol PS-PEG resin and 0.1 mmol
amine at 208C for 3 h. Solvent volume was 1 mL unless otherwise stated.
The effect of amount of amine on the yield of 6 was also
investigated (Table 4). Here amine 5 was added in varying
amounts to REM resin and agitated at 208C for 2 h before
standard quaternisation and cleavage with benzyl bromide.
Under standard conditions (DMF, 1 mL) the reaction could
not be forced to greater than 73% completion even when
a
1
Yields determined by H NMR using N-methylmaleimide as an internal
standard.
Represents the use of a standard amount (1 mL) of the perfluorohexane
with the stated amount DMF.
b
20 equiv. of 5 were used. This situation could be improved
by using a small amount of DMF or DMSO as solvent, or by
performing the reaction under solvent-free conditions. In
agreement with previous observations the best results were
seen when perfluorohexane was used either alone or in
conjunction with a small amount of co-solvent.
A reagent concentration effect was also observed with
poly(ethylene glycol), (PS-PEG) based graft co-polymers
(
(
Table 5). Here a PS-PEG REM resin was prepared
0.48 mmol/g) and suspended in various solvents in the
Scheme 3.
of perfluorohexane greater than that required to completely
cover the resin (Table 2, entries 1–5 versus 6–11).
presence of amine 5 (2 equiv.) at 208C for either 3 or 6 h
(Scheme 2). This was followed by standard quaternisation
and cleavage as outlined previously in the text. Whilst
yields were generally lower than those observed with
polystyrene based resins the same trend was observed with
Next, it was decided to investigate the effect of the volume
of co-solvent on reaction efficiency in the synthesis of 6
Table 6.
a
Yield (%) 7
Conditions
Total time (h)
Michael—10 equiv. 5, DMF, 18 h, quaternisation—5 equiv., DMF, 18 h, elimination—2 equiv. DIEA, DCM, 6 h
Michael—5 equiv. 5, perfluorohexane and DMSO (0.03 mL), 2 h, quaternisation—20 equiv., DMSO, 1 h,
elimination—4 equiv. DIEA, DCM, 1 h
42
4
68
.95
Michael—5 equiv. 5, DMF, 2 h, quaternisation—20 equiv., DMSO, 1 h, elimination—4 equiv. DIEA, DCM, 1 h
4
7
Reaction performed on 0.05 mmol. Solvent volume was 1 mL unless otherwise stated.
a
1
Yields determined by H NMR using N-methylmaleimide as an internal standard.

2140
J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
2.2. Quaternisation
The quaternisation reaction (Scheme 1, 2!3) has pre-
viously been identified as both the slowest and yield limiting
6
b
segment of the standard REM resin process, and so this
reaction seemed to offer an ideal opportunity for reagent
concentration. We have previously reported a reagent
concentration effect when using DMSO as the solvent for
quaternisation and subsequent increases in yield over the
6
c
standard DMF. In order to investigate the quaternisation
process REM resin bound N-methylphenethylamine (poly-
styrene with 1% divinylbenzene crosslinking, 1.3 mmol/g),
8
and tetrahydro isoquinoline (polystyrene with 1%
divinylbenzene crosslinking, 1.2 mmol/g), 9 were prepared
and quaternised in various solvents with either benzyl
bromide or ethyl bromoacetate to give, after standard
cleavage 6 or 10, respectively (Scheme 4). The results are
shown in Table 7.
Scheme 4.
the highest amount of product being seen when perfluoro-
hexane was used in conjunction with a small amount of co-
solvent. The fact that the optimal amount of co-solvent was
higher in this case is, in all likelihood, simply a consequence
of the larger amounts of the lower loading PS-PEG resin
required.
As can be seen increases in yield were also seen in the
quaternisation reaction on switching from the standard
solvents to perfluorous solvents and the best results were
observed when perfluorohexane was used in conjunction
with a small amount of DMSO as co-solvent. The thorough
washing of the resin to remove all traces of perfluorous
solvents and the excess quaternisation reagent trapped
within the resin was of paramount importance here as any
residual quaternisation reagent could go on to quaternisation
the product as it was cleaved from the resin in the next stage.
The use of a small amount of DMSO alone generally gave
similar yields to the normal amount of solvent (Table 7,
entry 2 versus 3). Interestingly, some product was also seen
when water was employed as solvent (Table 7, entry 5),
presumably also a result of the concentration of reagents
within the resin.
In order to demonstrate the utility of the reagent
concentration effect using perfluorous solvents the REM
resin synthesis of 7 from amine 5 and ethyl bromoacetate
was undertaken (Scheme 3, Table 6).
As can be seen, when the synthesis of 7 is attempted under
standard conditions it takes a total time of 42 h and 68% of
the theoretical yield is obtained. By using perfluorous
solvent in the Michael reaction and an excess of ethyl
bromoacetate in DMSO for the quaternisation the time taken
for the synthesis can be reduced to 4 h and the yield is
quantitative. Using the same regime with DMF as the
solvent for the Michael reaction solvent results in a poor 7%
yield.
One possible concern with the use of perfluorous solvents in
solid phase chemistry is an adverse effect on reaction
kinetics due to the lack of swelling with lightly cross-linked
resins. Macroporous resins, however, are designed to give
minimal swelling and so would seem ideally suited to
applications involving perfluorous solvents. With this in
mind, macroporous REM resins corresponding to 8
Table 7.
(
0
polystyrene with 20% divinylbenzene crosslinking,
.79 mmol/g) and 9 (polystyrene with 20% divinylbenzene
a
a
Entry
Solvent
Yield
%) 6
Yield
(%) 10
(
crosslinking, 0.74 mmol/g) were prepared by reaction with
the corresponding amine. These were then quaternised as in
Scheme 4 to give, after standard cleavage, tertiary amines 6
or 10 (Table 8).
1
2
3
4
5
6
7
8
9
DMF
DMSO
DMSO—0.03 mL
Neat quaternisation reagent
Water
Perfluoromethylcyclohexane
Perfluorotripropylamine
Perfluorohexane
75
43
31
88
29
46
53
12
NA
63
62
63
.95
47
b
c
49
c
NA
NA
92
.95
13
Table 8.
c
a
a
Entry
Solvent
Yield
(%) 6
Yield
(%) 10
d
Perfluorohexane and 0.03 mL DMSO
None
10
1
2
3
DMF
Perfluoromethylcyclohexane
Neat quaternisation reagent
44
65
31
17
56
14
Quaternisation reaction performed with 0.05 mmol resin and 0.1 mmol
quaternisation reagent (6) or 0.25 mmol quaternisation reagent (10) at 208C
for 3 h. Solvent volume was 1 mL unless otherwise stated.
b
a
1
Yields determined by H NMR using N-methylmaleimide as an internal
standard.
Approx. 150 equiv. quaternisation reagent used as solvent.
Reaction not attempted.
Represents the use of a standard amount (1 mL) of the named solvent with
Quaternisation reaction performed with 0.05 mmol macroporous resin and
0.1 mmol quaternisation reagent (6) or 0.25 mmol quaternisation reagent
(10) at 208C for 3 h. Solvent volume was 1 mL.
b
c
d
a
1
Yields determined by H NMR using N-methylmaleimide as an internal
standard.
Approx. 150 equiv. quaternisation reagent used as solvent.
b
0.03 mL DMSO.

J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
2141
Scheme 5.
As anticipated, the use of perfluorous solvents resulted in
considerable increases in yield compared to the standard
solvents. However, as the yields seen here were consider-
ably lower than those seen for the corresponding micro-
porous resins no further studies with macroporous matrices
were undertaken.
1.23 mmol/g), was quaternised with halide 11 (20 equiv.,
DMSO, 208C, 1 h) and subjected to Hofmann elimination
conditions in the presence of varying amounts of methanol
5
(Scheme 5, X¼O, R¼CH ). The results are shown in
3
Table 9.
For reaction in DCM, although transesterification did occur,
the process did not proceed to completion with less than
2
.3. Hofmann elimination/transesterification
At first sight the Hofmann elimination reaction (Scheme 1,
3
Table 10.
!4) offers few opportunities for reagent concentration,
a
being a facile process which can be performed quanti-
tatively in 1 h (Table 6). By incorporating a transesterifica-
tion reaction into the Hofmann elimination step, however,
Entry
1
Alcohol
MeOH
Equiv.
5
Solvent
Conversion (%)
DCM
Perfluorohexane
None
49
.95
55
25
.95
65
5
b
an opportunity for reagent concentration was discovered.
The use of 2,2,2-trifluoroethanol esters as activating groups
in peptide coupling and transesterification reactions has
2
3
EtOH
10
5
DCM
Perfluorohexane
None
DCM
8
been previously reported. It was thought that by incorpor-
28
ating such a moiety into the alkyl halide quaternisation
reagent, a REM resin bound activated ester would be
obtained that could be cleaved from the resin and
transesterified in one-pot. In this way a large batch of
polymer bound quaternary ammonium salt 3 could be
prepared and then split into numerous smaller batches for
the synthesis of an array of esters and amides by
transesterification with alcohols and amines, respectively.
It was hoped that reagent concentration would favourably
affect the transesterification equilibrium. In order to test this
hypothesis, REM resin bound N-methylphenethyl amine 8
Perfluorohexane
None
DCM
.95
69
59
4
5
2.5
2.5
Perfluorohexane
None
DCM
.95
70
39
Perfluorohexane
None
DCM
.95
41
28
(
polystyrene with 1% divinylbenzene crosslinking,
6
10
Table 9.
a
Entry
Solvent
Equiv.
MeOH
Conversion
b
(%)
Perfluorohexane
None
DCM
.95
57
,5
7
8
9
20
10
20
1
2
3
4
5
6
7
8
9
DCM
DCM
DCM
DCM
DCM
2
5
10
20
30
5
5
2
5
5
18
49
76
86
.95
70
58
.95
.95
65
90
55
Perfluorohexane
None
DCM
64
,5
,5
DCM (0.1 mL)
DCM (0.05 mL)
Perfluorohexane
Perfluorohexane
Perfluorohexane
DCM
50
,5
1
1
1
0
1
2
Perfluorohexane and DCM (0.1 mL)
Perfluorohexane and DCM (0.05 mL)
None
5
5
Perfluorohexane
,5
Quaternisation (0.05 mmol resin, 1 mmol 11) performed for 1 h in DMSO.
Transesterification performed for 18 h (0.05 mmol resin, DIEA 2 equiv.,
Quaternisation (0.05 mmol resin, 1 mmol 11) performed for 1 h in DMSO.
Transesterification performed for 18 h (0.05 mmol resin, DIEA 2 equiv.,
K CO 4 equiv., ROH, solvent). Solvent volume was 1 mL.
2 3
K
2
CO
3
4 equiv., MeOH, solvent 1 mL).
Solvent was 1 mL unless otherwise stated.
a
b
1
Conversion determined by H NMR (CDCl
a
1
3
).
3
Conversion determined by H NMR (CDCl ).

2142
J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
Table 11.
a
Entry
Amine
Equiv.
5
Solvent
DCM
Conversion (%)
1
2
3
88
Perfluorohexane
None
DCM
.95
.95
78
5
5
Perfluorohexane
None
DCM
.95
.95
88
Perfluorohexane
None
DCM
.95
.95
.95
4
5
5
5
Perfluorohexane
None
DCM
.95
.95
20
Perfluorohexane
None
DCM
83
83
,5
6
7
20
20
Perfluorohexane
None
DCM
,5
,5
,5
Perfluorohexane
None
DCM
,5
,5
50
8
9
5
5
5
Perfluorohexane and DCM (0.1 mL)
DCM
88
,5
Perfluorohexane and DCM (0.1 mL)
DCM
,5
17
10
Perfluorohexane and DCM (0.1 mL)
81
2 3
Quaternisation (0.05 mmol resin, 1 mmol 11) performed for 1 h in DMSO. Transesterification performed for 18 h (0.05 mmol resin, DIEA 2 equiv., K CO
4
2
equiv., RNH , solvent). Solvent volume was 1 mL.
1
Conversion determined by H NMR (CDCl ).
3
a
3
0 equiv. of methanol (Table 9, entries 1–5). Clearly,
co-solvent tended to depress reaction in this case. It was
unclear at which point during the process the transesteri-
fication reaction occurred. Presumably resin bound
intermediate 12 would be more reactive towards transesteri-
fication due to the influence of the quaternary nitrogen.
However, trifluoroester 13 was found to undergo complete
transesterification in solution when exposed to identical
conditions to those used in Table 1, entry 5. No
transesterification was observed under any conditions
when the 2,2,2-trifluoroethanol ester was replaced with
simple alkyl esters. At first sight, the success of transesteri-
fication in perfluorohexane was thought to be a result of the
miscibility of the liberated trifluoroethanol with the
although this was not a problem in the case of methanol,
where the alcohol is readily available in large quantities and
any excess could be removed by evaporation, it could
become a problem when using commercially unavailable or
non-volatile reagents. The extent of transesterification could
be improved, however, with the use of low solvent volumes
(
Table 9, entry 6), solvent-free conditions (Table 9, entry
1
2) or perfluorous solvents (Table 9, entries 8!11). The
best results were seen when perfluorohexane was used as
solvent, with essentially complete transesterification when 2
or more equivalents of ethanol were applied. Interestingly,
the use of perfluorohexane with a small volume of DCM as

J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
2143
perfluorous solvent. This would effectively remove any
trifluoroethanol from the reaction and so help drive the
equilibrium towards complete transesterification. However,
this was shown not to be the case since the two species are in
fact immiscible.
analysis (N determination was performed on a Perkin–
Elmer 2400 elemental analyser). Solvents and commercially
available reagents were used as received with no further
purification. Nuclear magnetic resonance were recorded at
400 MHz using a Bruker DPX instrument, where
1
appropriate yield was determined by H NMR using
Having established the feasibility of the one-pot Hofmann
elimination/transesterification protocol the reaction was
performed on a number of different alcohols (Table 10).
None of the alcohols used gave complete conversion under
standard conditions, although in most cases an improvement
was seen when the reaction was performed under solvent-
free conditions. All of the 18 alcohols used gave complete
conversion to the desired ester when perfluorohexane was
used as the solvent (Table 10, entries 1!6). With standard
solvents no product could be detected when 28 alcohols
were used as the nucleophilic species although this was
increased to approximately 50% by the use of perfluoro-
hexane (Table 10, entries 7!8). No transesterification was
seen under any conditions with a 38 alcohol (Table 10, entry
N-methylmaleimide as an internal standard (CDCl3
solution, 0.05 mmol/mL). Mass spectra were recorded on
a API150EX LC-MS instrument, High resolution mass
spectra were recorded on a Mariner TOF instrument with
perfluorokerosene as an internal standard. Infrared spectra
of products were recorded on a Perkin–Elmer Spectrum
1000 FT-IR instrument as thin films.
4.2. General procedure for Michael addition N-benzyl-
N-methylphenethylamine 6
REM resin (0.05 mmol) was treated with N-methyl-
phenethylamine (0.1–1 mmol) and the appropriate solvent.
The mixture was then agitated at 208C for 2–3 h, filtered
and the resin washed (3£1 mL, DMF, DCM, MeOH). The
precipitate was then air-dried, resuspended in DMF (1 mL)
and treated with benzyl bromide (0.5 mmol). The suspen-
sion was agitated at 208C for 18 h, filtered, washed (3£1 mL
DMF, DCM, MeOH) and resuspended in DCM (1 mL). The
suspension was treated with DIEA (0.1 mmol) and K CO
9
).
The use of 18 amines to give, after Hofmann elimination,
a-amino acid amides (Scheme 5, X¼NH) was also
successful (Table 11). The conversions seen were generally
much higher than for the corresponding alcohols, with
complete conversion seen for many amines under either
solvent-free or perfluorous conditions. Hydroxylamines and
a-amino acids were also suitable nucleophiles for the
transesterification process (Table 11, entries 8 and 10).
Aniline gave no transesterification products (Table 11, entry
2
3
(0.2 mmol) and agitated at 208C for 6 h. The mixture was
then filtered, the resin washed (3£1 mL DCM) and the
organic washings concentrated in vacuo. The product
N-benzyl-N-methylphenethylamine 6 was spectroscopically
1
1
identical to the previously published data.
6) and the use of 28 amines to give 38 amides was equally
unsuccessful (Table 11, entries 7 and 9).
4.2.1. Procedure for the REM resin synthesis of
N-methylphenethylamino)-acetic acid ethyl ester (7) in
(
When volatile nucleophiles were used in the process, any
excess was simply removed by evaporation. Excesses of
nucleophile for which this treatment was not amenable were
removed by the use of a polymer-supported sulphonyl
chloride (for alcohols), or isocyanate (for amines)
scavenger reagent.
4 h. REM resin (0.05 mmol, 2 mmol/g, 25 mg) was treated
with N-methylphenethylamine (0.25 mmol), DMSO
(0.03 mL) and perfluorohexane (1 mL). The mixture was
then agitated at 208C for 2 h, filtered and the resin washed
(3£1 mL, DMF, DCM, MeOH). The precipitate was then
air-dried, suspended in DMSO (1 mL) and treated with
ethylbromoacetate (1 mmol). The suspension was agitated
at 208C for 1 h, filtered, washed (3£1 mL DMF, DCM,
MeOH) and resuspended in DCM (1 mL). The suspension
was treated with DIEA (0.2 mmol) and K CO (0.2 mmol)
9
10
3. Conclusions
2
3
In conclusion, we have shown that the use of reagent
concentration methods in the REM resin synthesis of 38
amines resulted in increased rates in comparison with the
standard solvent systems which can be translated into higher
yields, shorter reaction times and a reduction in the excesses
of reagents required. The methods that showed the best
results were the use of low solvent volume or solvent-free
conditions or more generally the use of perfluorous solvents,
either with or without a small amount of organic co-solvent.
and agitated at 208C for 1 h. The mixture was then filtered,
the resin washed (3£1 mL DCM) and the organic washing
concentrated in vacuo to give the product as a colourless oil.
1
2 1
This material has been previously reported.
(CDCl , 400 MHz) d
(7); 1.24 (t, 3H, J¼7.1 Hz, CH
2.47 (s, 3H, NCH ), 2.74–2.83 (m, 4H, CH N and CH Ar),
3.30 (s, 2H, CH ), 7.18–
CO), 4.21 (q, 2H, J¼7.0 Hz, OCH
H NMR
),
3
H
3
3
2
2
2
2
7.21 (m, 3H, ArH), 7.24–7.29 (m, 2H, ArH); IR (film) 1740,
1454, 1176 cm2 ; HRMS 222.1482 (MH ), C13
1
þ
H
20NO
2
requires 222.1488.
4
. Experimental
4.2.2. General procedure for quaternisation (3,4-di-
hydro-1H)-isoquinolin-2-yl)-acetic acid ethyl ester (10).
4
.1. General
REM resin bound 1,2,3,4-tetrahydroisoquinoline, 9
(0.05 mmol, 1.21 mmol/g, 41 mg) was treated with ethyl-
All resins were purchased from Argonaut Technologies.
REM resin synthesis was performed according to published
procedures. Loading levels of REM resin were determined
by complete Michael reaction (by I.R.) and elemental
bromoacetate (0.055–1 mmol) and the appropriate solvent.
The mixture was then agitated at 208C for 2–3 h, filtered
and the resin washed (1£1 mL DCM, 3£1 mL 1:1
(v/v) toluene/heptane, DCM, DMF, DCM, MeOH). The
6
b

2144
J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
precipitate was then air-dried and resuspended in DCM
1 mL). The suspension was treated with DIEA (0.1 mmol)
and K CO (0.2 mmol) and agitated at 208C for 6 h. The
J¼6.5 Hz, OCH ), 7.13–7.21 (m, 3H, ArH), 7.23–7.30 (m,
2
(
3H, ArH), 7.33–7.40 (m, 4H, ArH); IR (film) 1739, 1497,
1454, 1167, 1058 cm ; m/z (ES ); 284 (MH ); HRMS
2
1
þ
284.1639 (MH ), C H NO requires 284.1645.
þ
2
3
þ
mixture was then filtered, the resin washed (3£1 mL DCM)
and the organic washings concentrated in vacuo to give the
product as a pale yellow oil. This material has been
1
8
22
2
4.2.7. (N-Methylphenethylamino)-acetic acid pyridin-3-
1
1
3 1
previously reported. H NMR (CDCl , 400 MHz) d (10);
3
yl methyl ester (14d). As a colourless oil. H NMR (CDCl ,
3
H
1
CH Ar), 3.30 (s, 2H, CH CO), 3.80 (s, 2H, CH Ar), 4.22 (q,
.27 (t, 3H, J¼7.3 Hz, CH ), 2.82–2.94 (m, 4H, CH N and
400 MHz) d (14d); 2.53 (s, 3H, NCH ), 2.79–2.91 (m, 4H,
H
CH N and CH Ar), 3.45 (s, 2H, CH CO), 5.19 (s, 2H,
2 2 2
3
2
3
2
2
2
2
H, J¼7.3 Hz, OCH ), 6.98–7.03 (m, 1H, ArH), 7.08–7.13
OCH ), 7.12–7.32 (m, 6H, ArH), 7.70 (dd, 1H, J¼2.0,
2
2
(
m, 3H, ArH).
8.0 Hz, ArH), 8.58 (d, 1H, J¼5.0 Hz, ArH), 8.63 (s, 1H,
2
1
þ
ArH); IR (film) 1742, 1430, 1204, 1058 cm ; m/z (ES );
þ þ
285 (MH ); HRMS 285.1611 (MH ), C H N O requires
17 21 2 2
4
1
.2.3. 2,2,2-Trifluoroethylbromoacetate (11). Compound
1 was prepared according to a modified version of the
285.1598.
1
4
procedure published by Hudson. 2,2,2-Trifluoroethanol
230 mmol) was cooled on an acetone/ice bath with
(
4.2.8. (N-Methylphenethylamino)-acetic acid phenethyl
1
magnetic stirring and treated dropwise with bromoacetyl
bromide (115 mmol). The mixture was then allowed to
warm to room temperature overnight and excess 2,2,2-
trifluoroethanol removed in vacuo. The residue was then
washed with saturated brine (3£10 mL), diluted with diethyl
ester (14e). As a colourless oil. H NMR (CDCl , 400 MHz)
3
d (14e); 2.39 (s, 3H, NCH ), 2.68–2.81 (m, 4H, CH N and
H
3
2
CH Ar), 2.92 (t, 2H, J¼6.8 Hz, CH Ar), 3.27 (s, 2H,
2
2
CH CO), 4.33 (t, 2H, J¼6.7 Hz, OCH ), 7.13–7.31 (m,
2
2
2
1
10H, ArH); IR (film) 1746, 1496, 1454, 1201 cm ; m/z
(ES ); 298 (MH ); HRMS 298.1796 (MH ), C H NO
19 24 2
þ
requires 298.1802.
þ
þ
ether (30 mL), dried (MgSO ) and filtered through a pad of
4
basic alumina (30 g). The alumina was flushed with diethyl
ether (100 mL) and the ethereal solution concentrated in
vacuo to afford 11 as a colourless liquid (20.6 g, 81%). H
NMR (CDCl , 400 MHz) d (11); 3.92 (s, 2H, CH Br), 4.55
1
4.2.9. (N-Methylphenethylamino)-acetic acid isopropyl
1
ester (14f). As a colourless oil. H NMR (CDCl , 400 MHz)
3
3
H
2
(
q, 2H, J¼8.0 Hz, CH CF ).
d (14f); 1.25 (d, 6H, J¼6.0 Hz, CH ), 2.46 (s, 3H, NCH ),
2
3
H
3
3
2.73–2.85 (m, 4H, CH N and CH Ar), 3.29 (s, 2H,
CH CO), 5.03–5.11 (m, 1H, OCH), 7.18–7.22 (m, 3H,
2 2
4
.2.4. General procedure for Hofmann elimination/
2
transesterification (N-methylphenethylamino)-acetic
acid methyl ester (14a). REM resin bound N-methyl-
phenethylamine, 8 (0.05 mmol) was treated with 11
ArH), 7.24–7.31 (m, 2H, ArH); IR (film) 1739, 1454,
1176 cm ; m/z (ES ); 236 (MH ); HRMS 236.1643
2
1
þ
þ
(MH ), C H NO requires 236.1645.
þ
1
4
22
2
(
1 mmol) and DMSO (1 mL). The mixture was then agitated
at 208C for 1 h, filtered and the resin washed (1£1 mL
DCM, DMF, DCM). The precipitate was dried in vacuo and
treated with methanol (0.25 mmol), K CO (0.2 mmol),
4.2.10. (N-Methylphenethylamino)-acetic acid cyclo-
1
hexyl ester (14g). As a colourless oil. H NMR (CDCl ,
3
400 MHz) d (14g); 1.20–1.93 (m, 10H, CH ), 2.44 (s, 3H,
2
2
3
H
DIEA (0.1 mmol) and the appropriate solvent. The resulting
suspension was agitated at 208C for 18 h. The mixture was
then filtered, the resin washed (1£1 mL DCM, 3£1 mL 1:1
NCH ), 2.73–2.85 (m, 4H, CH N and CH Ar), 3.28 (s, 2H,
3 2 2
CH CO), 4.88–4.93 (m, 1H, OCH), 7.18–7.22 (m, 3H,
2
ArH), 7.24–7.31 (m, 2H, ArH); IR (film) 1740, 1452,
1172 cm ; m/z (ES ); 276 (MH ); HRMS 276.4029
2
1
þ
þ
(MH ), C H NO requires 276.4023.
(
v/v) toluene/heptane, DCM) and the organic washings
þ
concentrated in vacuo to give the product as a colourless oil.
1
1
7
26
2
H NMR (CDCl , 400 MHz) d (14a); 2.45 (s, 3H, NCH ),
3
H
3
2
CH CO), 3.79 (s, 3H, OCH ), 7.18–7.22 (m, 3H, ArH),
.72–2.83 (m, 4H, CH N and CH Ar), 3.32 (s, 2H,
4.2.11. (N-Methylphenethylamino)-N-propyl acetamide
1
2
2
(14h). As a colourless oil. H NMR (CDCl , 400 MHz) d
3
2
3
H
21
7
m/z (ES ); 208 (MH ); HRMS 208.1331 (MH ),
.24–7.31 (m, 2H, ArH); IR (film) 1737, 1452, 1170 cm
þ
;
(14h); 0.83 (t, 3H, J¼7.5 Hz, CH ), 1.38 (S , 2H, J¼7.5 Hz,
3
e
þ
þ
CH ), 2.35 (s, 3H, NCH ), 2.69 (t, 2H, J¼6.5 Hz, CH Ar),
2 2
2 3 2
C H NO requires 208.1332.
1
2.76 (t, 2H, J¼6.5 Hz, CH N), 3.00–3.08 (m, 4H, CH CO
2
2
18
2
and CH N), 6.80 (s, br, 1H, NH), 7.17–7.31 (m, 5H, ArH);
2
1
þ
4
.2.5. (N-Methylphenethylamino)-acetic acid allyl ester
1
IR (film) 1662, 1527, 1454, 1148 cm ; m/z (ES ); 235
þ þ
(MH ); HRMS 235.1813 (MH ), C H N O requires
14 23 2
(
14b). As a colourless oil. H NMR (CDCl , 400 MHz) d
3
H
(
14b); 2.46 (s, 3H, NCH ), 2.77–2.84 (m, 4H, CH N and
3
235.1805.
2
CH Ar), 3.35 (s, 2H, CH CO), 4.62 (d, 2H, J¼5.5 Hz,
2
2
OCH ), 5.24 (dd, 1H, J¼1.5, 10.5 Hz, vCH –cis), 5.32
4.2.12. (N-Methylphenethylamino)-N-allyl acetamide
1
2
2
(
vCH), 7.18–7.22 (m, 3H, ArH), 7.24–7.31 (m, 2H, ArH);
dd, 1H, J¼1.5, 17.1 Hz, vCH –trans), 5.88–5.97 (m, 1H,
(14i). As a colourless oil. H NMR (CDCl , 400 MHz) d
3
2
H
(14i); 2.37 (s, 3H, NCH ), 2.66–2.79 (m, 4H, CH Ar and
3 2
2
1
þ
MH ); HRMS 234.1497 (MH ), C H NO requires
IR (film) 1740, 1454, 1171, 1058 cm ; m/z (ES ); 234
þ
CH N), 3.04 (s, 2H, CH CO), 3.70 (t, 2H, J¼5.5 Hz, CH N),
2
2
2
þ
34.1489.
(
5.06–5.09(m, 2H, vCH ),5.64–5.72(m, 1H, vCH),6.89(s,
2
1
4
20
2
2
br, 1H, NH), 7.16–7.31 (m, 5H, ArH); IR (film) 1664, 1521,
þ
2
1
þ
(MH ), C H N O requires 233.1649.
1
454, 1275 cm ; m/z (ES ); 233 (MH ); HRMS 233.1654
þ
4.2.6. (N-Methylphenethylamino)-acetic acid benzyl
1
ester (14c). As a colourless oil. H NMR (CDCl3,
1
4 21 2
4
4
00 MHz) dH (14c); 2.53 (s, 3H, NCH ), 2.77–2.86 (m,
3
4.2.13. (N-Methylphenethylamino)-N-benzyl acetamide
1
H, CH N and CH Ar), 3.44 (s, 2H, CH CO), 4.13 (t, 2H,
2
(14j). As a colourless oil. H NMR (CDCl , 400 MHz) d
3
2
2
H

J. R. Morphy et al. / Tetrahedron 59 (2003) 2137–2145
2145
þ þ þ
(ES ); 265 (MH ); HRMS 265.1539 (MH ), C H N O
14 21 2 3
(
14j); 2.35 (s, 3H, NCH ), 2.66–2.73 (m, 4H, CH Ar and
3
2
CH N), 3.08 (s, 2H, CH CO), 4.27 (d, 2H, J¼6.0 Hz,
requires 265.1547.
2
2
ArCH N), 7.08–7.33 (m, 10H, ArH); IR (film) 1668, 1520,
2
2
1
þ
HRMS 283.1814 (MH ), C H N O requires 283.1805.
þ
1
496, 1454, 1265, 1123 cm ; m/z (ES ); 283 (MH );
þ
1
8 23 2
References
4
mide (14k). As a colourless oil. H NMR (CDCl3,
.2.14. (N-Methylphenethylamino)-N-phenethyl aceta-
1
1. Thompson, L. A.; Ellman, J. A. Chem. Rev. 1996, 96,
555–600.
4
00 MHz) dH (14k); 2.28 (s, 3H, NCH ), 2.62–2.72 (m,
3
6
H, CH Ar, CH Ar and CH N), 2.98 (s, 2H, CH CO), 3.31
2
2. (a) Eichler, J.; Houghten, R. A.; Lebl, M. J. Pept. Sci. 1996, 2,
240–244. (b) Morphy, J. R.; Rankovic, Z. R.; York, M.
Tetrahedron Lett. 2002, 43, 5973–5975.
2
2
2
(
q, 2H, J¼6.5 Hz, CH N), 6.87 (s, br, 1H, NH), 7.11–7.31
2
2
1
(
m, 10H, ArH); IR (film) 1668, 1524, 1454, 1124 cm ; m/z
þ
þ
requires 297.1961.
þ
(
ES ); 297 (MH ); HRMS 297.1971 (MH ), C H N O
19 25 2
3. Harrison, C. R.; Hodge, P.; Hunt, B. J.; Khoshdel, E.;
Richardson, G. J. Org. Chem. 1983, 48, 3721–3728.
4
. Horvath, I. T. Acc. Chem. Res. 1998, 31, 641–650, and
references therein.
4
mide (14l). As a colourless oil. H NMR (CDCl , 400 MHz)
.2.15. (N-Methylphenethylamino)-N-cyclohexyl aceta-
1
5. (a) Morphy, J. R.; Rankovic, Z. R.; York, M. Tetrahedron Lett.
2001, 42, 7509–7511. (b) Morphy, J. R.; Rankovic, Z. R.;
York, M. Tetrahedron Lett. 2002, 43, 6413–6415.
6. (a) Morphy, J. R.; Rankovic, Z.; Rees, D. C. Tetrahedron Lett.
1996, 37, 3209–3212. (b) Brown, A. R.; Rees, D. C.;
Rankovic, Z. R.; Morphy, J. R. J. Am. Chem. Soc. 1997,
119, 3288–3295. (c) Cameron, K. S.; Morphy, J. R.;
Rankovic, Z. R.; York, M. J. Comb. Chem. 2002, 4, 199–203.
7. Resin swelling determinations were made on 500 mg dry resin
using the method reported in: Santini, R.; Griffith, M. C.; Qi,
M. Tetrahedron Lett. 1998, 39, 8951–8954.
3
dH (14l); 0.98–1.83 (m, 10H, CH ), 2.33 (s, 3H, NCH ),
2
3
2
CH CO), 3.63–3.72 (m, 1H, NCH), 6.87 (s, br, 1H, NH),
.66–2.82 (m, 4H, CH N and CH Ar), 3.04 (s, 2H,
2 2
2
7
1
.18–7.33 (m, 5H, ArH); IR (film) 1668, 1520, 1452, 1150,
þ
2
1
þ
053 cm ; m/z (ES ); 275 (MH ); HRMS 275.2113
þ
(
MH ), C H N O requires 275.2118.
7 27 2
1
4
.2.16. N-Methoxy-2-(N-methylphenethylamino)-aceta-
1
mide (14m). As a colourless oil. H NMR (CDCl3,
00 MHz) dH (14m); 0.36 (s, 3H, NCH ), 2.68 (t, 2H,
4
3
J¼6.0 Hz, CH Ar), 2.75 (t, 2H, J¼6.0 Hz, CH N), 3.09 (s,
8. Kaldor, S. W.; Siegel, M. G.; Fritz, J. E.; Dressman, B. A.;
Hahn, P. J. Tetrahedron Lett. 1996, 37, 7193–7196.
9. Rueter, J. K.; Nortey, S. O.; Baxter, E. W.; Leo, G. C.; Reitz,
A. B. Tetrahedron Lett. 1998, 39, 975–978.
2
2
2
8
1
H, CH CO), 3.59 (s, 3H, OCH ), 7.18–7.34 (m, 5H, ArH),
2 3
.86 (s, br, 1H, NH); IR (film) 1673, 1495, 1455 1124, 1080,
þ
2
1
þ
050 cm ; m/z (ES ); 223 (MH ); HRMS 223.1435
þ
(
MH ), C H N O requires 223.1441.
12 19 2 2
10. Kaldor, S. W.; Siegel, M. G.; Fritz, J. E.; Dressman, B. A.;
Hahn, P. J. Tetrahedron Lett. 1996, 37, 7193–7196.
11. Lepley, A. R.; Giumanini, A. G. J. Org. Chem. 1967, 32,
1706–1713.
4.2.17. [2-(N-Methylphenethylamino)-acetylamino]-
1
acetic acid methyl ester (14n). As a colourless oil. H
NMR (CDCl , 400 MHz) dH (14n); 2.40 (s, 3H, NCH3),
3
12. Baltzly, R.; Phillips, A. P. J. Am. Chem. Soc. 1949, 71(119),
3288–3295.
13. Wedekind, E. Chem. Ber. 1903, 36, 1161.
2
CH N), 3.07 (s, 2H, CH CO), 3.73 (s, 3H, OCH ), 3.84 (d,
.71 (t, 2H, J¼6.5 Hz, CH Ar), 2.78 (t, 2H, J¼6.5 Hz,
2
2
2
3
2
(
H, J¼6.0 Hz, CH COCH ), 7.18–7.32 (m, 5H, ArH); IR
14. Goerger, M. M.; Hudson, B. S. J. Org. Chem. 1988, 53,
3148–3153.
2
3
film) 1752, 1673, 1524, 1454, 1437, 1206, 1128 cm21; m/z