A cleaner approach to solid-supported radical chemistry: Application of hypophosphite salts to intramolecular C-C bond formation on the solid-phase

Article abstract of DOI:10.1055/s-2005-862376

The application of hypophosphite salts to solid-supported radical chemistry has been explored. EPHP-mediated intramolecular cyclisations have been shown to give good yield of cyclised products on JandaJel resin. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2005-862376

LETTER
477
A Cleaner Approach to Solid-Supported Radical Chemistry:
Application of Hypophosphite Salts to Intramolecular C-C Bond Formation
on the Solid-Phase
A
Cleaner
A
ppr
o
a
ac
h
to Solid
r
-Suppor
o
ted
R
adical
C
hemis
i
try ne Chevet, Toby Jackson, Barbara Santry, Anne Routledge*
Department of Chemistry, University of York, Heslington, York, YO10 5DD, UK
Fax +44(1904)432516; E-mail: ar30@york.ac.uk
Received 14 July 2004
bound radical precursor 1a was treated with two commer-
Abstract: The application of hypophosphite salts to solid-support-
ed radical chemistry has been explored. EPHP-mediated intramo-
lecular cyclisations have been shown to give good yield of cyclised
products on JandaJel® resin.
cially available radical chain carriers, N-ethylpiperidine
hypophosphite (EPHP)9 and tributyltin hydride (Bu3SnH)
in anhydrous, degassed toluene under reflux. In all reac-
tions varying amount of cyclised product, benzofuran 4
was detected by HPLC analysis of the resin cleavage
mixtures after reaction10 (Table 1).
Key words: solid-phase, cyclizations, radical reactions
Organotin hydride-mediated radical chemistry is a partic-
ularly powerful method of forming carbon-carbon bonds
both in solution1 and on the solid-phase.2 However, the
well-established neurotoxicity of organotin reagents3 has
precluded their use in the solid-phase synthesis of drug
candidate libraries by the pharmaceutical industry. Solu-
tion-phase alternatives to organotin hydrides have includ-
ed silicon-based radical chain carriers such as
tris(trimethylsilyl)silane,4 but cost and ease of oxidation
has prohibited the widespread use of this reagent in solid-
phase library synthesis. More recently, the application of
phosphorous centred radicals (derived from H3PO2 salts)5
to intramolecular C-C bond formation has been demon-
strated in the solution-phase in a variety of cleaner sol-
vents such as ethanol6 and water.7 The application of these
cost effective, non-toxic phosphorous reagents to solid-
supported radical chemistry would enable library synthe-
ses to utilize radical-mediated bond formation without the
problem of potential contamination of library members
with toxic organotin residues.
HO
O
DIC, cat. DMAP
cis-buten-1,4-diol
O
I
O
2-iodophenol
O
O
HO
PPh3, DEAD
O
65% over two steps
1a
(i) cyclisation
(ii) NaOMe
OH
I
OH
OH
O
O
O
4
2
3
Scheme 1 Synthesis of iodoether 1a on carboxypolystyrene resin
Herein we report our preliminary work towards develop-
ing a cleaner approach to solid-supported radical chemis-
try by exploring the application of hypophosphite salts to
a representative solid-phase intramolecular radical reac-
tion on a variety of commercially available resins.
Table 1 Cyclisation of Resin-Bound Iodoether 1a in Toluene with
Bu3SnH (10 equiv) or EPHP (20 equiv)
Time
(h)
Chain
AIBN
Yield of 2 Yield of 3 Yield of 4
carriera
(equiv)a (%)b
(%)b
(%)b
As hypophosphite reagents had not previous been used to
effect resin supported radical reactions we chose to test
the reagents on a simple system that had been shown to
undergo efficient radical cyclisation on the solid-phase.2a,c
Iodoether radical precursor 1a was synthesised on 2%
divinylbenzene cross-linked carboxypolystyrene (1.34
mmol/g, Novabiochem) via a solid-phase Mitsunobu
reaction8 (Scheme 1). The loading of 1a was 0.63 mmol/g
as determined by iodine elemental analysis. The resin-
4
4
Bu3SnH
EPHP
EPHP
EPHP
1
4
77
38
58
1
95
1
4
19
48
18
3c
3d
5
57
Trace
42
a Equivalents are based on the initial loading of resin-bound iodoether
1a.
b Ratio determined by HPLC analysis of reaction mixture after cleav-
age from the resin.
c 3 Equiv of initiator based on the initial loading of resin-bound iodo-
ether 1a, 1 equiv added at t = 0 h, t = 18 h and t = 36 h.
d Drop-wise addition of AIBN over 18 h.
SYNLETT 2005, No. 3, pp 0477–0480
1
6.
0
2.
2
0
0
5
Advanced online publication: 04.02.2005
DOI: 10.1055/s-2005-862376; Art ID: D19304ST
© Georg Thieme Verlag Stuttgart · New York

478
C. Chevet et al.
LETTER
R = H
Cyclisation with EPHP was sluggish compared to
Bu3SnH with only 57% conversion after 48 hours even
with the addition of three equivalents of AIBN.
acetic anhydride
NaH
OR
45%
5 R = Ac
HO
It was noted that the EPHP salt was not soluble in toluene
so a variety of solvent systems were screened as alterna-
tives, in these systems the resin was in a swollen state and
the EPHP salt in solution. The reactions were performed
with either AIBN or V-5011 as the radical initiator. In all
cases the conversion to benzofuran 4 was not significantly
improved (Table 2).
5, 2-iodophenol
PPh3, DEAD
63%
I
OR
O
6 R = Ac
2, DIC
NaOMe
I
O
Table 2 Cyclisation of Resin-Bound Iodoether 1a with EPHP (20
2 R = H
equiv) and AIBN or V-50 (1 equiv)
64%
O
80%
O
1b
Solvent
Time
(h)
Yield of 2 Yield of 3 Yield of 4
(%)a
(%)a
(%)a
19
1
(i) cyclisation
(ii) NaOMe
Toluene
4
4
77
4
THF–H2O (4:1)
THF–H2O (4:1)b
THF–EtOH (4:1)
THF–EtOH (4:1)c
99
Trace
Trace
2
OH
4
98
2
I
OH
OH
4
78
20
63
O
4
O
O
48
31
6
2
3
a Ratio determined by HPLC analysis of reaction mixture after cleav-
age from the resin.
Scheme 2 Synthesis of iodoether 1b on NovaSyn® TG carboxy
resin
b Initiator V-50 (1 equiv).
c 3 Equiv of initiator based on the initial loading of resin-bound iodo-
ether 1a, 1 equiv added at t = 0 h, t = 18 h and t = 36 h.
Table 3 Cyclisation of NovaSyn® TG-Bound Iodoether 1b with
either Bu3SnH (10 equiv)/AIBN (1 equiv) or EPHP (20 equiv)/AIBN
or V-50 (1 equiv)
As EPHP is a salt the sluggish reaction on divinylbenzene
cross-linked polystyrene could be due to low concen-
tration of the reagent within the hydrophobic polymer Solvent
matrix. This prompted us to explore the same reaction on
Chain
carrier
Time
(h)
Yield of 2 Yield of 3 Yield of 4
(%)a
(%)a
(%)a
97
24
8
a more hydrophilic resin, NovaSyn® TG carboxy resin
Toluene
Bu3SnH
EPHP
EPHP
EPHPb
EPHP
4
18
18
18
18
18
2
1
(0.25 mmol/g, Novabiochem).
Toluene
74
2
The use of this resin could also potentially allow the cycli-
sation to be carried out in less toxic solvents such as water
or ethanol, which do not swell divinylbenzene cross-
linked polystyrene resins. The radical precursor 2 was
synthesized in solution12 and loaded onto NovaSyn® TG
carboxy resin via an esterification reaction to give the res-
in-bound radical precursor 1b (Scheme 2). The loading of
1b was 0.15 mmol/g as determined by iodine elemental
analysis. Radical precursor 1b was treated with EPHP and
either AIBN or V-50 in a variety of solvents (Table 3).
EtOH
H2O
92
Trace
0
100
98
0
H2O
Trace
3
2
THF:EtOH EPHP
(4:1)13
33
64
THF:EtOH EPHP
(4:1)c
48
5
6
89
a Equivalents are based on the initial loading of resin-bound iodoether
The EPHP mediated reactions (Table 3, entries 2–6)
ranged from 0% to 89% conversion with reduced product
3 detected in all cases. In comparison, reaction with tribu-
tyltin hydride and AIBN gave 97% conversion to benzo-
furan 4 (Table 3, entry 1) with only a trace amount of the
reduced product 3 detected.
1a.
b Initiator V-50 (1 equiv).
c An additional equiv of initiator added at 24 h.
Macroporous resin (ArgoPore®) gave a conversion to the
cyclised product 4 similar to cross-linked polystyrene but
had a higher percentage of reduced product 3. The PEG
grafted resins, HypoGel® and ArgoGel® all resulted in
conversion to the cyclised product with similar product
distribution to NovaSyn® TG resin. The polytetrahydro-
In order to explore the influence of the solid-support on
the product ratio of the EPHP-mediated cyclization, iodo-
ether 2 was loaded onto a variety cross-linked and PEG
grafted solid-supports14 then subjected to radical cycliza-
tion.15 After reaction cleaved material was analysed by
HPLC (Table 4).
Synlett 2005, No. 3, 477–480 © Thieme Stuttgart · New York

LETTER
A Cleaner Approach to Solid-Supported Radical Chemistry
479
Table 4 Cyclisation of Resin-Bound Iodoether 2 with EPHP (20
In conclusion, it appears from the limited reactions stud-
ied that N-ethylpiperidine hypophosphite (EPHP) could
be used as a cleaner alternative to tributyltin hydride for
solid-supported intramolecular radical cyclisation. The
choice of polymer support may be critical to the success
of the reaction but the cyclisations studied proceed effi-
ciently in THF–EtOH (4:1) on JandaJel® resin and this
may reflect the concentration of EPHP/AIBN within the
polymer matrix of this resin. Further studies to explore the
partitioning of EPHP within a resin bead and subsequent
effects on product ratio of radical cyclisations are in
progress.
equiv) and Initiator (2 equiv)a 48 h in THF–EtOH (4:1)
Solid-support
Yield of 2 Yield of 3 Yieldof4
(%)
(%)
(%)
O
O
31
14
55
N
H
OH
ArgoPore® HL 0.84 mmol/gb
COOH
6
5
7
7
87
88
HypoGel® 0.54 mmol/gb
O
O
N
H
OH
OH
Acknowledgment
ArgoGel®, 0.35 mmol/gb
We thank the ERASMUS scheme (CC) and the University of York
for financial support. In addition, we also thank Dr A F Parsons and
Mr. C Jessop (University of York) for helpful discussions.
O
O
0
2
98
N
H
JandaJel® 0.32 mmol/gb
References
a 1 Equiv added at t = 0 h, 1 equiv added at t = 24 h.
(1) Jasperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. Rev. 1991,
91, 1237.
b Loading of iodoether, determined by elemental analysis for iodine.
(2) (a) Routledge, A.; Abell, C.; Balasubramanian, S. Synlett
1997, 61. (b) Miyabe, H.; Tanaka, H.; Naito, T. Tetrahedron
Lett. 1999, 40, 8387. (c) Berteina, S.; DeMesmaeker, A.
Tetrahedron Lett. 1998, 39, 5759.
(3) Yallapragada, P. R.; Vig, P. J. S.; Kodavanti, P. R. S.;
Desaiah, D. J. Toxicol. Environ. Health 1991, 34, 229.
(4) Chatgilialoglu, C.; Griller, D.; Lesage, M. J. Org. Chem.
1988, 53, 3641.
(5) (a) Barton, D. H. R.; Jang, D. O.; Jaszberenyi, J. C.
Tetrahedron Lett. 1992, 33, 5709. (b) Graham, S. R.;
Murphy, J. A.; Coates, D. Tetrahedron Lett. 1999, 40, 2415.
(6) Yorimitsu, H.; Shinokubo, H.; Oshima, K. Bull. Chem. Soc.
Jpn. 2001, 74, 225.
furan cross-linked resin, JandaJel® gave excellent conver-
sion, with only a small amount of reduced product 3
detected. In order to determine if the optimisation was
specific to iodoether 2, an alternative radical precursor
was attached to JandaJel® resin, a primary radical precur-
sor 7 (Figure 1).
I
OH
O
(7) Kita, Y.; Nambu, H.; Ramesh, N. G.; Anilkumar, G.;
Matsugi, M. Org. Lett. 2001, 3, 1157.
(8) Rano, T. A.; Chapman, K. T. Tetrahedron Lett. 1995, 36,
3789.
CH3O
7
Figure 1
(9) Carboxypolystyrene resin loaded with 1a (100 mg, 0.063
mmol) was suspended in anhyd degassed toluene (1 mL),
EPHP (225 mg, 1.26 mmol) and AIBN (10 mg, 0.06 mmol)
were added, and then the reaction mixture was heated under
reflux for 4 h. The resin was washed with DMF (3 × 5 mL),
THF (3 × 5 mL), THF–H2O (1:1, 3 × 5 mL), THF (3 × 5
mL), CH2Cl2 (3 × 5 mL) then dried under vacuum for 48 h.
(10) Resin was suspended in THF–MeOH (4:1), excess MeONa
was added and the reaction mixture was agitated for 24 h.
The resin was filtered off and the solution was eluted
through a plug of ion exchange resin (Dowex 50W-X8 H+)
then analysed by HPLC on a Alltech econosil Si column
eluting with hexane–i-PrOH (0–5 min 100% hexane, 5–35
min gradient 0–30% i-PrOH): iodoether 2, 18.4 min; benzyl
ether 3, 17.3 min; benzofuran 4, 19.9 min; 220 nm detection.
(11) 2,2¢-azobis(2-amidinopropane) dihydrochloride, a water
soluble azo radical initiator.
Both 2 and 7 were cyclised using the optimised reaction
conditions. The cyclisations were scaled up to allow de-
termination of a crude yield rather than just a conversion
to either cyclised or reduced product as had previously
been determined (Table 5).
Table 5 Cyclisation of JandaJel®-Bound Iodoethers
Radical precursor
Crude yield (%)a
Purity of cyclised
product
2
7
97
92
77b
69b (2:1, trans:cis)c
a Based on the loading of resin-bound iodoether.
b Determined by HPLC analysis at 220 nm detection.
c Determined by NMR.
(12) The stepwise solid-phase synthesis (Scheme 1) could not be
applied to NovaSyn® TG carboxy resin as reaction with cis-
butan-1,4-diol resulted in extensive cross-linking of the
resin.
Synlett 2005, No. 3, 477–480 © Thieme Stuttgart · New York

480
C. Chevet et al.
LETTER
(14) Carboxy-functionalised resins were either commercially
(13) NovaSyn ® TG resin loaded with 1b (100 mg, 0.015 mmol)
was suspended in degassed solvent (0.25 mL), EPHP (54
mg, 0.30 mmol) and AIBN (2.5 mg, 0.015 mmol) or V-50 (4
mg, 0.015 mmol) were added then the reaction mixture was
heated under reflux for 18 h. The resin was washed with
THF (3 × 5 mL), THF–H2O (1:1, 3 × 5 mL), THF (3 × 5
mL), CH2Cl2 (3 × 5 mL) then dried under vacuum for 48 h.
available (HypoGel®) or synthesized from commercially
available amino functionalised resin by reaction with
glutaric anhydride (JandaJel®, ArgoGel® and ArgoPore®).
(15) The reactions were performed on 100 mg of resin and the
amount of solvent used in the reaction was dependent on the
loading of the resin (100 mg of resin with 1 mmol g–1
loading = 1.5 mL solvent for reaction), 1 equiv of initiator
added at t = 0 h and t = 24 h.
Synlett 2005, No. 3, 477–480 © Thieme Stuttgart · New York