Solid-phase synthesis on functionalised fluoropolymer resins. Part 1: Nafion resin sulfonamide-immobilised carboxylic acid derivatives and aryl vinyl sulfones

Article abstract of DOI:10.1016/S0040-4039(00)00646-8

The preparation and properties of Nafion resin sulfonamide systems derived from methyl glycinate and from 3-hydroxyethylaniline are described. Nafion derivatisation reactions were slow and maximum achievable functionalisations were 50% of the ion exchange

Full text of DOI:10.1016/S0040-4039(00)00646-8

Tetrahedron Letters 41 (2000) 4487±4491
Solid-phase synthesis on functionalised ¯uoropolymer resins.
Part 1: Na®on resin sulfonamide-immobilised carboxylic acid
derivatives and aryl vinyl sulfones
Mahmoud Akhtar, Friedrich E. K. Kroll and David Gani*
School of Chemistry, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
Received 1 March 2000; accepted 20 April 2000
Abstract
The preparation and properties of Na®on resin sulfonamide systems derived from methyl glycinate and
from 3-hydroxyethylaniline are described. Na®on derivatisation reactions were slow and maximum
achievable functionalisations were 50% of the ion exchange capacity. In aqueous solution the resin derivatives
were acid and base labile but in organic solvents the systems were stable to powerful nucleophiles and to
heat. An aryl vinyl sulfone derivative used in the synthesis of tertiary amines a€orded very pure products.
# 2000 Elsevier Science Ltd. All rights reserved.
Solid-phase organic synthesis (SPOS) is a useful and often complementary alternative to
conventional solution-phase synthesis. SPOS normally involves `linking' molecules containing a
reaction centre to a polymer support that is otherwise inert, and then elaborating the reaction
centre. The reaction centre may be a substrate or a reagent. Advances in the range of chemistries
that have been utilised on polymer supports are impressive1 3 as are the underpinning advances in
new resin materials,4 resin functional groups,4 linkers,5 reagents,6 resin cleavage protocols,7,8 and
analytical chemistry.8,9 To date almost all e€orts have been concerned with developing 1±2%
cross-linked polystyrene-based systems for SPOS. Multiparallel SPOS is particularly useful in
preparing libraries of related compounds and avoids many of the problems associated with
assessing reaction eciency, purity and `deconvolution' in `split and mix' protocols. Recently we
TM
described POSAM , an ecient and versatile SPOS protocol for preparing compound libraries
on the 0.01 to 1.0 mmol scale.10 Here, the apparatus is constructed only from glass and PTFE and
thus is suitable for reactions requiring particularly demanding conditions. Existing resin materials
based upon polystyrene limit the utility of SPOS because polystyrene is mechanically fragile,
unstable to oxidants, strong acids and bases and to temperatures above 120ꢀC. Mindful of such
limitations, we set-out to evaluate the potential of more stable polymers including those containing
* Corresponding author.
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00646-8

4488
per¯uorocarbon backbones. Here we describe the preparation, properties and evaluation in SPOS
of Na®on-based systems.
The profound thermal, mechanical and chemical stabilities of PTFE are well established and
are ideally suited for applications in SPOS. However, suitably functionalised PTFE derivatives
are extremely dicult to prepare by copolymerisation with tetra¯uoroethene (TFE) because
substituents larger than an F-atom retard the rate of the radical polymerisation process sub-
stantially and lead to a low incorporation of the substituted comonomer.11 Moreover, TFE is a
highly toxic and explosive gas and requires pressurising and heating for controlled polymerisation.
Na®on resins (DuPont) are the products of copolymerisation reactions of TFE and tri¯uorovinyl
ethers possessing per¯uoroalkylsulfonyl ¯uoride side-chains. After hydrolysis the sulfonic acid
products, Fig. 1, are used as ion selective membranes in electrochemical cells by the chloralkali
industry, or, as resin-bound super acids.11 Recently, the cation exchange properties of Na®on
resin have been used to support Pd(II) catalysts.12
Figure 1. General structure of Na®on and similar resin materials
Na®on1 NR50 is a per¯uoroalkyl sulfonic acid resin (ion exchange capacity ca. 0.8 millimoles
g^1) available as 10±35 mesh cylindrical pellets. This material was used to prepare functionalised
¯uoropolymer model systems to evaluate the stability and utility of ¯uoropolymers in SPOS. To
increase the surface area of the Na®on resin and ensure that the sulfonic acid groups were
accessible to reagents, the pellets were ground at ca. ^180ꢀC, under liquid nitrogen, in a mortar to
give a coarse (ca. 50±100 mM) white powder. At higher temperatures the resin was too plastic and
became squashed rather than ground. In order to activate the ¯uoropolymer resin as its sulfonyl
chloride 2, the powdered polymer 1 was re¯uxed with excess thionyl chloride or excess phosphorus
pentachloride in toluene for up to several days. Phosphorus pentachloride treatment for 36 h was
found to give the best results as judged by analysing the DCMy washed, and then dried, products
from several repeat reactions of di€ering duration. The resin mass increased in accord with
expectations and resin 2 displayed a new IR absorbing band at 1180 cm^1. Reaction of a small
amount of resin 2 with disperse orange 11 dye gave a coloured sulfonamide derivative 3 con®rming
that the sulfonyl moiety was activated either as just the chloride or as a mixture of the chloride
and the anhydride.
y
Abbreviations: DCM, dichloromethane; TFA, tri¯uoroacetic acid; DMF, dimethylformamide; PyBOP, benzo-
triazole-1-yl-oxy-tris-pyrrolidino-phosphonium hexa¯uorophosphate; NMM, N-methyl morpholine; DIPEA, diiso-
propylethylamine.

4489
In order to quantify the site availability of the activated resin it was decided to react the
material with a radiolabelled amine of known speci®c activity. Accordingly, 14C-[U]-glycine was
converted to its methyl ester 4, using thionyl chloride and methanol, and the solvents and acidic
gases were removed under reduced pressure. The ®nal speci®c activity of the 14C-[U]-glycine
methylester hydrochloride salt 4 was 3.8Â106 dpm mmol^1. The reaction of the activated resin 2
with the labelled amine 4 was assessed using a range of solvents and bases under a wide variety of
conditions. Re¯uxing the resin 2 in dry DCM in the presence excess pyridine and amine 4 for 5
days, optimised conditions, resulted in the formation of sulfonamide 5 in a 50% radiochemical yield,
Scheme 1. This was calculated by comparing the speci®c radioactivity of the product (1.52Â106
dpm g^1) with the starting amine after determining the ion exchange capacity of the crushed resin
1 (0.8 mmol g^1). It was already becoming evident that reactions involving the displacement of
groups from the sulfonyl S-atom of the resin were slow in organic solvents, presumably due to
combined steric and ¯uorophilic e€ects of the polymer backbone and side-chains.
The stability of the sulfonamide linkage in the radiolabelled sulfonamide 5 was tested under
aqueous basic and acidic conditions by stirring samples of the resin and removing aliquots of the
supernatant solution for scintillation counting. Even under mild conditions (0.25 M HCl or 0.25 M
NaOH) the sulfonamide linkage was hydrolysed within several hours. However, the sulfonamide
group in 5 showed remarkable stability in organic solvents and was stable to 2.8 M n-butyl
lithium and lithium aluminium hydride for prolonged periods. The ester moiety of 5 was accessible
to n-butyl lithium and gave the tertiary alcohol as was determined by cleavage of the product
from the resin (in acid) and subsequent analysis by 1H NMR spectroscopy.
Scheme 1.
The activated resin 2 was also converted to its N-sulfonyl (2S)-alanine t-butyl ester derivative 6
{FTIR; 1735 (CˆO, ester), 1320 (sulfonamide) and 1210 and 1150 cm^1 (SO2)} in 80% recovery
and the t-butyl group was removed using TFA/DCM quantitatively,13 Scheme 2. The carboxylic
acid was activated using NMM/PyBOP and was then treated with 14C-[U]-labelled glycine methyl
ester 4. The N-sulfonyl (2S)-alanyl glycine methyl ester derivative 8 was obtained after only 24 h,
with a radiochemical yield of only 30% {FTIR; 1740 (CˆO, ester), 1680 (CˆO, amide), 1320
(sulfonamide) and 1210 and 1150 cm^1 (SO2)}. As longer reaction times did not improve the yield, it
seems probable that the shorter periods required for the coupling derive from the reduced steric and
¯uorophilic e€ects experienced by the incoming N-nucleophile at greater distances from the polymer
backbone. (The low overall yield may re¯ect anhydride formation during the activation step).
t
.
Scheme 2. Reagents and conditions: (i) Pyridine, DCM, 5 days; (ii) Ala-O Bu AcOH, pyridine, DCM, 96 h; (iii) TFA,
DCM, 12 h; (iv) 4, PyBOP, DMF, NMM

4490
The sulfonyl chloride 2 was reacted with 3-aminobenzyl alcohol 9 to give a stable red coloured
sulfonanilamide 10 in quantitative recovery.13 Likewise, reaction of the activated resin 2 with
2-(3-aminophenylsulfonyl)ethanol gave the resin-bound sulfonanilamide 11 {FTIR; 1510 and
1480 (CH2), 1320 (sulfonamide) and 1210 and 1150 (SO2) and 750 and 680 cm-1 (m-disubstituted
aromatic)}, Scheme 3.
Scheme 3. Reagents and conditions: (i) Ms-Cl, Et3N, 20 h; (ii) piperazino-4-acetophenone, DMF, 72 h; (iii) PhCH2Br,
DMF, 24 h; (iv) DIPEA, DMF, 24 h
Treatment of the terminal alcohol 11 with mesyl chloride and elimination of methanesulfonic
acid from the product (using triethylamine) gave the aryl vinyl sulfone 12. The conjugate addition
of piperazino-4-acetophenone to the vinyl sulfone slowly gave the expected resin bound tertiary
amine 13 which displayed new absorbances in its FTIR spectrum at 2950, 2850 and 1650 (CˆO)
cm^1. The resin bound diamine 13 was quaternised at its non-deactivated N-atom, using benzyl
bromide, as has been described for the analogous polystyrene-based vinyl sulfones.14 Treatment
of the quaternised salt 14 with DIPEA regenerated the sulfonanilimide vinyl sulfone 12 and
a€orded the tertiary amine, N-benzylpiperazino-4-acetophenone 15 in greater than 98% purity,
1
as judged by H and 13C NMR spectroscopy,15 albeit in low overall yield (ca. 5%), Scheme 3.
N-Benzyl tetrahydroisoquinoline was also prepared in a similarly high purity and in low yield.
Again it was evident that all reactions were very slow and interestingly, the carbonyl group of the
resin-bound tertiary amine 13 reacted very slowly (if at all) with Grignard and aryl lithium
reagents, as judged by FT-IR spectroscopy and preparative work-up. Recovered material showed
no mass loss and contained no polymer decomposition products such as is common for poly-
styrene based systems.
From the results reported here it is evident that ¯uoropolymers o€er considerable potential as
inert support materials for SPOS, although Na®on derivatives are not ideal. Two properties of
Na®on NR50 probably limit its suitability for SPOS most. The ®rst is `clustering' of functional
sites which appears to allow site interference and the formation of symmetric sulfonic and carboxylic
acid anhydrides during activation steps. This property, of course, restricts initial loadings at least.
The second is the combined highly polar nature of the functional groups and non-lipophilic
per¯uorinated nature of the polymer side-chains. These are sterically demanding and do not solvate
in organic solvents to allow the resin to swell, ultimately causing poor reaction kinetics, unlike in

4491
aqueous solutions. A polymer possessing a per¯uorinated backbone and lipophilic side-chains
that are capable of solvating in organic solvents is described in the following article.
Acknowledgements
We thank the Universities of Birmingham and St. Andrews for ®nancial support.
References
1. Balkenhohl, F.; Bussche-Hunnefeld, C.; Lansky, A.; Zechel, C. Angew. Chem., Int. Ed. Engl. 1996, 35, 2288±2337.
2. Gallop, M. A.; Barrett, R. W.; Dower, W. J.; Fodor, S. P. A.; Gordon, M. A. J. Med. Chem. 1994, 37, 1233±1251;
Gordon, E. M.; Barrett, R. W.; Dower, W. J.; Fodor, S. P. A. J. Med. Chem. 1994, 37, 1385±1401.
3. Hermkens, P. H. H.; Ottenheim, H. C. J.; Rees, D. Tetrahedron 1996, 52, 4527; Maehr, H. Bioorg. Med. Chem.
1997, 5, 473±491.
4. See Novabiochem Catalogue 1999 for new resins and new immobilised functional groups.
5. Backes, B. J.; Ellman, J. A. J. Am. Chem. Soc. 1994, 116, 11171±11172; Holmes, C. P.; Jones, D. G. J. Org. Chem.
1995, 60, 2318±2319; Plunkett, M. J.; Ellman, J. A. J. Org. Chem. 1995, 60, 6006±6007; Chenera, B.; Finkelstein,
J. A.; Veber, D. F. J. Am. Chem. Soc. 1995, 117, 11999±12000.
6. Hormozdiari, P.; Gani, D. Tetrahedron Lett. 1996, 37, 8227±8130; Ley, S. V.; Thomas, A. W.; Finch, H. J. Chem.
Soc., Perkin Trans. 1 1999, 669±671.
7. Conti, P.; Demont, D.; Cals, J.; Ottenheijm, H. C. J.; Leysen, D. Tetrahedron Lett. 1997, 38, 2915±2918.
8. Stones, D.; Miller, D. J.; Beaton, M. W.; Rutherford, T. J.; Gani, D. Tetrahedron Lett. 1997, 38, 4875±4878.
9. Nestler, H. P.; Bartlett, P. A.; Still, W. C. J. Org. Chem. 1994, 59, 4723±4724; Nicolaou, K. C.; Xiao, X.-Y.;
Parandoosh, Z.; Senyei, A.; Nova, M. P. Angew. Chem., Int. Ed. Engl. 1995, 34, 2289±2291; Egner, B. J.; Cardno,
M.; Bradley, M. J. Chem. Soc., Chem. Commun. 1995, 2163±2164.
10. Gani, D.; Akhtar, M.; Kroll, F. E. K.; Smith, C. F. M.; Stones, D. Tetrahedron Lett. 1997, 38, 8577±8580.
11. Fielding, H. C.; ICI Chemicals and Polymers, Fluoropolymer Membranes, Fluoroploymer Conference, 1992.
12. Seen, A. J.; Townsend, A. T.; Bellis, J. C.; Cavell, K. J. J. Mol. Catal. 1999, 149, 233±242.
13. All yields and recoveries were assessed by mass changes after thorough washing and drying.
14. Kroll, F. E. K.; Morphy, R.; Rees, D.; Gani, D. Tetrahedron Lett. 1997, 38, 8573±8576.
15. Data for 15: {(HRMS: found: [M+H]+, 295.1805. Calcd for C19H22N2O: 295.1810); ꢀmax(thin ®lm)/cm^1 2950,
2842, 1670 (CˆO) and 1599 (aromatic); ꢁH(300 MHz; C2HCl3) 2.51 (3H, s, CH3), 2.57±2.61 (4H, m, piperazine),
3.34±3.38 (4H, m, piperazine), 3.57 (2H, s, benzyl), 6.84±6.87 (2H, m, Ar±H), 7.26±7.40 (5H, m, Ar-H) and 7.85±
7.88 (2H, m, Ar-H); m/z (CI) 295 [M+H]+}.