to radical attack, stabilize the radical formed, or provide an
initiating/leaving group within the polymerization.
Trithiocarbonates 7-9 are classified as RAFT CTAs
whereas the MADIX process relies on the use of xanthates.
The generic strategy using TCDI, outlined in Scheme 2, may
be used to form xanthates through a similarly facile
procedure, sharing all of the benefits of the laboratory scale
syntheses of RAFT agents. Xanthates are of particular interest
as they have been reported to successfully mediate the
controlled polymerization of vinyl acetate (VAc).22 Con-
trolled radical polymerization of VAc is difficult to obtain
because of the highly reactive nature of the monomer radical,
which has a tendency to add to activated bonds such as Cd
S. The resultant radical is relatively stable and therefore
unlikely to fragment in the required fashion. The Centre for
Advanced Macromolecular Design (CAMD) group investi-
gated the effect of different Z groups within the xanthate in
order to avoid these complications and concluded that by
incorporating an electron-withdrawing Z group, fragmenta-
tion could be promoted leading to enhanced control of the
polymerization.22 A Z group derived from 2-propanol, was
shown to be particularly effective. The choice of R group is
The arrangement of substituents around the thiocarbonyl-
thio group makes 2-ethylsulfanylthiocarbonylsulfanyl-pro-
pionic acid ethyl ester, 7, an excellent candidate CTA for
controlled RAFT polymerization. The R group, originating
from the ethyl 2-mercaptopropionate, is expected to perform
well in the polymerization of methyl acrylate (MA), as the
radical formed is structurally similar to the MA propagating
radical.15 The S-ethyl group, acting as the Z group, is
expected to sufficiently activate the CdS group toward
radical addition (relative to the corresponding O-ethyl or
N-ethyl groups but will deactivate the CdS group with
respect to alkyl or aryl groups) and stabilize the resultant
intermediate radical. The successful synthesis of the trithio-
carbonate 7 was confirmed by electrospray MS (m/z)
1
239.023, MH+), H NMR and 13C NMR (δ ) 221.9 ppm,
CdS).
The formation of dibenzyl trithiocarbonate, 8, was achieved
in a one-pot synthesis, by reacting TCDI with 2 equiv of
primary thiol. Compound 8 has been previously reported16-20
using conventional CTA synthetic routes and acts as a
disubstituted CTA, having two identical groups capable of
leaving and reinitiating polymerization. Using conventional
syntheses, reported yields of 8 vary considerably (40-
95+%). The use of TCDI however gave 8 in a respectable
purified yield (81%) using a one-pot facile reaction over 6
also crucial and groups derived from primary thiols have
22,23
been successfully used.
To produce such a xanthate,
monosubstitution of TCDI was achieved by reacting 1 equiv
of 2-propanol with TCDI, followed by addition of benzyl
mercaptan without purification of the intermediate. The
resulting xanthate, 10, was purified by flash chromatography
1
(68% purified yield) and was characterized by H NMR (δ
) 1.4 ppm, 6H, CH3; δ ) 4.3 ppm, 2H, Ar-CH2; δ ) 5.8
ppm, 1H, R2CH; δ ) 7.2 ppm, 5H, Ar-H), 13C NMR (δ )
213.2 ppm, CdS) and electrospray MS (m/z ) 227.056,
MH+).
1
h. The success of the reaction was determined by H NMR
(δ ) 4.6 ppm, 4H, Ar-CH2; δ ) 7.3 ppm, 10 H, Ar-H),
13C NMR (δ ) 222.8 ppm, CdS), and electrospray MS (m/z
) 291.034, MH+).
The attempted synthesis of a xanthate by initial mono-
substitution of TCDI with ethyl 2-mercaptopropionate,
followed by addition of benzyl alcohol, led to the coupling
of two benzyl alcohol residues via the thiocarbonyl group.
A novel CTA, substituted with two potential R groups,
has also been synthesized using the TCDI approach. 2-Ben-
zylsulfanylthiocarbonylsulfanyl-propionic acid ethyl ester, 9,
is an asymmetrical trithiocarbonate synthesized using the
monosubstitution of TCDI with 1 equiv of secondary thiol
followed by addition of one equivalent of a primary thiol
(64% purified yield). Compound 9 was characterized using
1H NMR (δ ) 1.3 ppm, 3H, CH3; δ ) 1.6 ppm, 3H, CH3;
4.6 ppm, 2H, Ar-CH2; δ ) 7.3 ppm, 5H, Ar-H), 13C NMR
(δ ) 220.2 ppm, CdS) and electrospray MS (m/z ) 269.067,
(M+ - S)H+).21
1
This was confirmed by H NMR and 13C NMR. It is clear
that the formation of the symmetric thiocarbonate was
achieved through the displacement of the secondary thiol
by the excess benzyl alcohol, possibly after initial xanthate
formation. We have previously reported the displacement
of tertiary alcohols from carbonates by neighboring hydroxyl
groups to form cyclic carbonate structure,8 but the displace-
ment of secondary thiol groups by excess alcohol was not
expected.
We report the use of TCDI in a one-pot controlled
formation of an asymmetric dithiocarbamate through the
coupling of benzylamine and ethyl 2-mercaptopropionate.
Although this structure is not expected to function well as a
RAFT/MADIX CTA, the formation of dithiocarbamates via
the same strategy was designed to demonstrate the scope of
the TCDI synthetic approach. Katritzky et al. have recently
reported a similar approach for the synthesis of dithiocar-
bamates, via the reaction of thiols with thiocarbamoylben-
zotriazoles (yields 60-99%).14 To our surprise, the expected
ethyl ester containing product was not formed, but the
(15) Chong, Y. K.; Krstina, J.; Le, T. P. T.; Moad, G.; Postma, A.;
Rizzardo, E.; Thang, S. H. Macromolecules 2003, 36, 2256-2272.
(16) Lima, V.; Jiang, X.; Brokken-Zijp, J.; Scohenmakers, P. J.;
Klumperman, B.; Van Der Linde, R. J. Polym. Sci., Part A: Polym. Chem.
2005, 43, 959-973.
(17) You, Y. Z.; Hong, C. Y.; Bai, R. K.; Pan, C. Y.; Wang, J. Macromol.
Chem. Phys. 2002, 203, 477-483.
(18) Tamami, B.; Kiasat, A. R. J. Chem. Res., Synop. 1998, 8, 454-
455.
(19) Jian-Jun Yuan, Rui Ma, Qing Gao, Yi-Feng Wang, Shi-Yuan Cheng,
Lin-Xian Feng, Zhi-Qiang Fan, Lei Jiang, J. Appl. Polym. Sci. 2003, 89,
1017-1025.
(20) Islam, N. B.; Kwart, H.; Khan, S. J. Chem. Eng. Data 1985, 30,
509-512.
(21) As carbonyls and thiocarbonyls are particularly susceptible to
fragmentation, the major peak in the MS analyses is assigned to the
molecular ion, minus the carbonyl oxygen (sulfur). The nominal MS data
show peaks with a lower intensity that do represent the expected molecular
ion, but these peaks could not be analyzed with high-resolution MS because
of too much interference from other fragments. The data are still correct
and consistent with what could be expected from high-resolution MS.
(22) Stenzel, M. H.; Cummins, L.; Roberts, E.; Davis, T. P.; Vana, P.;
Barner-Kowollic, C. Macromol. Chem. Phys. 2003, 204, 1160-1168.
(23) Boschmann, D.; Vana, P. Polym. Bull. 2005, 53, 231-242.
Org. Lett., Vol. 8, No. 4, 2006
555