Solid-phase synthesis of 4-aminopiperidine analogues using the Alloc protecting group: An investigation of Alloc removal from secondary amines

Article abstract of DOI:10.1016/S0040-4039(01)00755-9

A useful method for Alloc removal from secondary amines on solid-phase has been optimised. The use of Me₂NH·BH₃ (40 equiv., 40 min) as scavenger of the allyl cations in a palladium-catalysed process with Pd[PPh₃]₄ leads to quantitative removal of the Alloc group without any allyl back alkylation. Other scavengers such as morpholine or PhSiH₃ are clearly inferior. Furthermore, this study has highlighted differences in the reaction kinetics of the deprotection step between secondary and primary amines such as those from α-amino acids.

Full text of DOI:10.1016/S0040-4039(01)00755-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 4471–4474
Solid-phase synthesis of 4-aminopiperidine analogues using the
Alloc protecting group: an investigation of Alloc removal from
secondary amines†
Dolors Ferna´ndez-Forner,a,* Gaspar Casals,a Elo´ısa Navarro,a Hamish Rydera and
Fernando Albericiob
aResearch Center, Almirall Prodesfarma, Cardener 68-74, E-08024 Barcelona, Spain
bDepartment of Organic Chemistry, University of Barcelona, E-08028 Barcelona, Spain
Received 27 April 2001; revised 27 April 2001; accepted 6 May 2001
Abstract—A useful method for Alloc removal from secondary amines on solid-phase has been optimised. The use of Me2NH·BH3
(40 equiv., 40 min) as scavenger of the allyl cations in a palladium-catalysed process with Pd[PPh3]4 leads to quantitative removal
of the Alloc group without any allyl back alkylation. Other scavengers such as morpholine or PhSiH3 are clearly inferior.
Furthermore, this study has highlighted differences in the reaction kinetics of the deprotection step between secondary and
primary amines such as those from a-amino acids. © 2001 Elsevier Science Ltd. All rights reserved.
In the area of combinatorial chemistry and solid-phase
synthesis, there is a constant need for new methods that
easily introduce and cleanly remove protecting groups.1
Since the 4-aminopiperidine substructure is central to a
variety of drug-like molecules,2 our group was inter-
ested in developing an efficient and reliable solid-phase
synthesis for the preparation of libraries of 4-
aminopiperidine derivatives. One of the main concerns
with the synthetic approach selected was the choice of a
suitable protecting group for the secondary amine func-
tion of the 4-aminopiperidine building block. Initially,
the use of 4-amino-1-Fmoc-piperidine as a building
block was planned, as Fmoc is compatible with major
synthetic strategies used in solid-phase and can be
removed easily under mild basic conditions.3 However,
its synthesis in reasonable yield failed, probably due to
incompatibility between the free amino function and
the Fmoc group. As an alternative to the Fmoc group,
various examples of the protection of amines using the
allyloxycarbonyl (Alloc) group have been described.4
However, use of the Alloc protecting group in solid-
phase synthesis for secondary amines like piperidine has
not been reported before.5 Secondary amines are more
nucleophilic than primary amines and are therefore
more prone to suffer allyl back alkylation, the main
drawback of the use of allyl based protecting groups.4
We present here the results obtained using 4-amino-1-
Alloc-piperidine (1) for the solid-phase synthesis of the
4-aminopiperidine amide analogues (3) and the depro-
tection conditions explored for removal of the allylic
carbamate whilst avoiding the formation of the allyl-
amine derivative and other by-products. Our synthetic
strategy was (i) attachment of 4-amino-1-Alloc-pipe-
ridine to a BAL linker6 derivatised resin through a
reductive amination; (ii) acylation of the secondary
amine thus obtained with a carboxylic acid; (iii)
removal of the Alloc group; (iv) derivatisation via the
amino function of piperidine, and (v) cleavage from the
resin (Scheme 1).
4-Amino-1-Alloc-piperidine (1) was obtained in two
steps in good yield starting from the piperidone, as
outlined in Scheme 2.7
Keywords: allyl protecting groups; combinatorial chemistry; diversity;
reductive amination.
* Corresponding author. Fax: 34 93 291 34 20; e-mail: dfernand@
almirallprodesfarma.com
The choice of an appropriate allyl scavenger is clearly
the key factor in establishing a general and convenient
strategy for the removal of Alloc groups from primary,
and, more importantly, secondary amines. In the litera-
ture several protocols using a large variety of nucle-
†
Abbreviations used: BAL, backbone amide linker, specifically as
implemented in the tris(alkoxy)benzylamide system; HATU, N-
[(dimethylamino)-1H-1,2,3-triazolo-[4,5-b]pyridin-1-yl-methylene]-
N-methylmethanaminium hexafluorophosphate N-oxide; NDMBA,
N,N%-dimethyl barbituric acid.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00755-9

4472
D. Ferna´ndez-Forner et al. / Tetrahedron Letters 42 (2001) 4471–4474
Scheme 1. Reagents and conditions: (i) NaBH3CN, DMF–DCM (8:2), 18 h, rt; (ii) HATU, DIEA, DMF–DCM (1:9), 2 h, rt; (iii)
see Table 1; (v) TFA–CH2Cl2 (1:1).
Scheme 2. Reagents: (a) H2O, K2CO3; (b) NaBH3CN, NH4OAc, MeOH.8
ophilic species with soluble palladium catalyst have
been described.4 In order to test some different proce-
dures, biphenyl-2-carboxylic acid piperidin-4-ylamide
was synthesised by acylation of the 4-amino-1-Alloc-
piperidine anchored to the BAL-resin with the o-
phenylbenzoic acid, followed by removal of the Alloc
group, and cleavage from the resin before further reac-
tion with R2 containing reagents (Scheme 1). The prod-
ucts obtained were analysed by HPLC–MS.
morpholine, which is used broadly for cleavage of allyl
esters,11 and finally the carbon nucleophile N,N%-
dimethyl barbituric acid (NDMBA).12
The best preliminary results were obtained using 20
equiv. of the amino–borane complex (entry 1) which
lead, after 2 h of treatment, to 85% of the deprotected
NH-piperidine, while only 7% of the N-allylpiperidine
analogue was obtained. In the case of morpholine,
significant amounts of the desired NH derivative (67%)
were only obtained when 5 equiv. of HOAc13 were
added to the 20 equiv. of scavenger (entry 3). In the
same experiment a significant amount of N-allylated
amine was obtained (19%). A large excess of HOAc
(entry 4) did not improve the results. On the other
hand, with NDMBA (entry 5) only the unchanged
1-Alloc-piperidine derivative was recovered after 2 h of
reaction time. In the best two experiments (entries 1
and 3), besides the N-allyl product, we detected 10–14%
of the same impurity that was initially observed in the
PhSiH3 experiment. This was characterised by HPLC–
MS, and later confirmed by synthesis as the N-
formylpiperidine derivative.
In the first deprotection experiment catalysed with
Pd[PPh3]4, the hydride donor PhSiH3 was chosen as a
neutral and easy to handle allyl group scavenger.9 After
treatment of the resin 2 with 20 equiv. of PhSiH3 and
0.1 equiv. of Pd[PPh3]4 (2×10 min, see Ref. 10 for
experimental details), only 52% of the expected NH
piperidine was obtained, while 42% of the allyl pipe-
ridine derivative was detected together with 6% of
another impurity, which initially we ignored. Although
some back alkylation (maximum of 5%) has been
reported in the deprotection of N-Alloc-N-methylben-
zylamine in solution where 2 equiv. of PhSiH3 were
used,4 the amount of the allylpiperidine derivative
obtained in this solid-phase approach, where a 20-fold
excess of scavenger was used, was quite unexpected. It
was apparent that a careful systematic approach was
required in order to optimise the reaction conditions.
It is important to point out that, with regard to reactiv-
ity, N-Alloc-piperidines seemed to undergo deprotec-
tion on solid-phase much more slowly than other allyl
carbamates (Me2NH·BH3 has been reported to lead to
complete deprotection of N-Alloc-amines in less than
10 minutes and NDMBA in less than 90 minutes).5
For this systematic study, an initial set of experiments
(Table 1, entries 1–5) with a small selection of scav-
engers and two different reaction times (20 and 120
min) was set up. The hydride donor Me2NH·BH3 was
selected,5 as well as the nucleophilic secondary amine
From the above experiments, the amino–borane com-
plex Me2NH·BH3 appeared to be the most suitable

D. Ferna´ndez-Forner et al. / Tetrahedron Letters 42 (2001) 4471–4474
4473
Table 1. Alloc deprotection of 4-aminopiperidine derivatives catalysed by Pd(PPh3)4
Scavenger
Additive13
% NH
120 min
% N-Allyl
20 min 120 min
% N-Alloc
20 min 120 min
% N-Formyl
20 min
20 min
120 min
1
2
3
4
5
Me2NH·BH3 (20 equiv.)
Morpholine (20 equiv.)
Morpholine (20 equiv.)
Morpholine (20 equiv.)
NDMBA (20 equiv.)
50
15
67
0
85
41
67
9
7
2
19
0
7
10
19
1
33
78
0
100
100
1
45
0
90
98
3
0
14
0
10
2
14
1
HOAc (5 equiv.)
HOAc (40 equiv.)
0
2
0
0
0
0
40 min
120 min
40 min
120 min
40 min
120 min
40 min
120 min
6a Me2NH·BH3 (40 equiv.)
88
81
88
81
0
5
0
5
0
0
0
0
12
14
12
14
7
Me2NH·BH3 (20
equiv.)+piperidine (20
equiv.)
8
9
Me2NH·BH3 (40 equiv.)
NaBH3CN (40 equiv.)
2% H2O
1% HOAc
61
54
–
54
30
0
–
0
0
36
–
36
9
2
–
2
a This experiment has been repeated with freshly distilled DMF after the preparation of this manuscript: 99% of NH product was obtained and
less than 1% of N-formyl was detected.
scavenger, although back alkylation still occurred to a
significant extent. To improve the yield of the deprotec-
tion step, a second set of experiments was carried out
with this hydride donor, trying different reaction condi-
tions and also testing the pseudometallic hydride
NaBH3CN (entries 6–9).
alkylation. Other scavengers such as morpholine or
PhSiH3 give clearly inferior results. The formation of
N-formyl derivatives seems to be related to the removal
step and/or the work-up rather than to the stability of
the Alloc derivative, as the quantity of this side product
correlates with the yield of deprotected amine. This
undesired side reaction has been practically avoided
(<1%) subsequently by using freshly distilled DMF in the
work-up procedure. Furthermore, this study has high-
lighted differences in reaction kinetics of the deprotec-
tion step between the secondary amines described
herein and primary amines such as those from a-amino
acids.
Results illustrated in Table 1 (entry 6) show that using
a 40-fold excess of Me2NH·BH3 eliminates the compet-
ing side-reaction leading to allylation. The excess of
nucleophile also improved the reaction rate leading to
complete deprotection of the allyl carbamate in 40
minutes. The presence of water (entry 8) was detrimen-
tal, favouring the back alkylation. Furthermore, the
combination of the hydride and piperidine (entry 7),
which could compete with the piperidine bound to the
resin, did not totally eliminate the side reaction.
Finally, although NaBH3CN (entry 9) was very effec-
tive as a scavenger in preventing the back alkylation,
the kinetics of deprotection were slower than for
Me2NH·BH3 and the crude product displayed more
impurities by HPLC.
Acknowledgements
The authors wish to thank Josep Ma6 Huerta for supply-
ing HPLC–MS data. The work at the UB was partially
supported by CICYT (BQU2000-0235).
In order to compare results with the reportedly more
efficient but toxic tributyltin hydride,14 the deprotection
experiment was repeated using the same conditions as
in entry 1 but with nBu3SnH (20 equiv.) instead of
Me2NH·BH3. After 20 minutes of reaction time, results
equivalent to using 40 equiv. of Me2NH·BH3 (entry 6)
were obtained. Nevertheless, in both cases 12% of the
formyl derivative was observed. Despite these promis-
ing results the use of tributyltin hydride was avoided
mainly because of its toxicity.
References
1. For reviews regarding protecting groups, see: (i) Kocien-
ski, P. J. In Protecting Groups; Georg Thieme Verlag:
Stuttgart, Germany, 1994; (ii) Greene, T. W.; Wuts, P. G.
M. In Protective Groups in Organic Synthesis, 3rd ed.;
John Wiley & Sons: New York, 1999; (iii) Albericio, F.
Biopolymers (Peptide Science) 2000, 55, 123–139.
2. Some gastrointestinal related drugs are 4-aminopiperidine
amide derivatives, for example: ‘Clebopride’, US
4438492, 1975; ‘Cinitapride’, US 5026858, 1991.
3. Fields, G. B.; Noble, R. L. Int. J. Peptide Protein Res.
1990, 35, 161–214.
In conclusion, a useful method for Alloc removal from
secondary amines in solid-phase has been optimised.
The use of Me2NH·BH3 (40 equiv., 40 min) as scav-
enger of the allyl cations in a palladium-catalysed
deprotection process with Pd[PPh3]4 leads to quantita-
tive removal of the Alloc group without any allyl back
4. For reviews regarding allyl based protecting groups, see:
Guibe´, F. Tetrahedron 1997, 53, 13509–13556, and 1998,
54, 2967–3042.
5. The Alloc derivative of the less nucleophilic secondary

4474
D. Ferna´ndez-Forner et al. / Tetrahedron Letters 42 (2001) 4471–4474
amine N-methylbenzylamine has been taken as a model
F.; Loffet, A. Tetrahedron Lett. 1995, 36, 5741–5744; (b)
Thieriet, N.; Go´mez-Mart´ınez, P.; Guibe´, F. Tetrahedron
Lett. 1999, 40, 2505–2508.
for the study of the Alloc removal in solution. Go´mez-
Mart´ınez, P.; Dessolin, M.; Guibe´, F.; Albericio, F. J.
Chem. Soc., Perkin Trans. 1 1999, 2871–2874.
10. General procedure for Alloc removal: In a typical experi-
ment, resin 2 (50 mg, 0.05 mmol) was swelled with
CH2Cl2 (4×30 s) and a 1 or 2 M solution of scavenger (20
or 40 equiv.) in CH2Cl2 (for NaBH3CN experiments,
THF was used as a solvent) was first added under an Ar
atmosphere and stirred for 5 min. Then, a 0.01 M solu-
tion of Pd[PPh3]4 (0.1 equiv.) in CH2Cl2 was added and
the resin was allowed to stir. After 10, 20 or 60 minutes,
depending on the experiment, the solution was drained
off, the resin washed with CH2Cl2 (4×30 s) and the
deprotection procedure was repeated once. Finally, the
resin was washed successively with CH2Cl2 (4×30 s),
DMF (4×30 s), dioxane–water (9:1) (3×1 min), MeOH
(4×1 min), DMF (4×30 s) and CH2Cl2 (4×30 s).
11. (a) Lloyd-Williams, P.; Jou, G.; Albericio, F.; Giralt, E.
Tetrahedron Lett. 1991, 32, 4207–4210; (b) Kates, S. A.;
Sole´, N. A.; Johnson, C. R.; Hudson, D.; Barany, G.;
Albericio, F. Tetrahedron Lett. 1993, 34, 1549–1552; (c)
Salvino, J. M.; Mervic, M.; Mason, H. J.; Kiesow, T.;
Teager, D.; Airey, J.; Labaudiniere, R. J. Org. Chem.
1999, 64, 1823–1830.
6. (a) Jensen, K. J.; Alsina, J.; Songster, M. F.; Va´gner, J.;
Albericio, F.; Barany, G. J. Am. Chem. Soc. 1998, 120,
5441–5452; (b) Alsina, J.; Jensen, K. J.; Albericio, F.;
Barany, G. Chem. Eur. J. 1999, 5, 2787–2795.
7. N-Alloc-4-piperidone. Allyl chloroformate (21 mL, 1.5
equiv.) was added to a solution of 4-piperidone monohy-
drate hydrochloride (20 g, 0.13 mol) and K2CO3 (27 g,
1.5 equiv.) in water (300 mL) at 5°C, and the reaction
mixture was allowed to stir for 1 h at 5°C and for another
additional hour at rt. The aqueous solution was saturated
with NaCl and extracted with diethyl ether (4×100 mL).
The combined organic phases were initially washed with
a cold 5% aqueous NH3 solution and brine, dried over
MgSO4, and evaporated to dryness to give the title
product (23.7 g, quantitative yield), which was used with-
out further purification.
8. 4-Amino-N-Alloc-piperidine. A solution of the above
ketone (10 g, 0.05 mol), NaBH3CN (24 g, 7.6 equiv.) and
ammonium acetate (42 g, 10 equiv.) in dry methanol (500
mL) was heated under reflux for 1 h. Methanol was
removed by evaporation and the residue suspended in 300
mL of H2O. The solution was acidified with 2N HCl to
pH 2 (caution, HCN evolution!!), left to stand 15 min with
agitation and washed with EtOAc (2×100 mL). The
aqueous layer was saturated with K2CO3 and extracted
with EtOAc (3×100 mL). The combined organic phases
were washed with brine, dried over MgSO4, and evapo-
rated to dryness to give the title product (7.2 g, 78%
12. Kunz, H.; Ma¨rz, J. Angew. Chem., Int. Ed. Engl. 1988,
27, 1375–1377.
13. The presence of an acid and even of water has been
postulated to be beneficial for increasing the yields of the
cleavage of allyl esters; see Ref. 11a and Kates, S. A.;
Daniels, S. B.; Albericio, F. Anal. Biochem. 1993, 212,
303–310.
1
14. Dangles, O.; Guibe´, F.; Balavoine, G.; Lavielle, S.; Mar-
quet, A. J. Org. Chem. 1987, 52, 4984–4993.
yield), which was shown to be pure by H NMR.
9. (a) Dessolin, M.; Guillerez, M. G.; Thieriet, N.; Guibe´,
.