Polymerization of glycidyl phenyl ether with a thiophosphonic acid ester as a novel thermally latent initiator

Article abstract of DOI:10.1246/cl.2000.100

S,S-Di-t-butyl phenylthiophosphonate (1) served as a novel thermally latent initiator in the polymerization of glycidyl phenyl ether (GPE) in the presence Of ZnCl₂.

Full text of DOI:10.1246/cl.2000.100

100
Chemistry Letters 2000
Polymerization of Glycidyl Phenyl Ether with a Thiophosphonic Acid Ester
as a Novel Thermally Latent Initiator
Moonsuk Kim, Fumio Sanda, and Takeshi Endo*
Research Laboratory of Resources Utilization, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 226-8503
(Received November 9, 1999; CL-990959)
S,S-Di-t-butyl phenylthiophosphonate (1) served as a
novel thermally latent initiator in the polymerization of gly-
cidyl phenyl ether (GPE) in the presence of ZnCl2.
Latent initiators can generate active species for polymer-
ization by external stimulation such as heating or photoirradia-
tion, which have been widely used in industrial fields such as
coatings, adhesives, and inks. Crivello et al.1 and we2 have
developed several onium salts such as sulfonium, pyridinium,
phosphonium, diaryl iodonium, and triaryl sulfonium salts,
which are useful as efficient latent thermal and photo initiators.
Meanwhile, N-substituted phthalimides,3 aminimides,4 car-
boxylic acid esters,5 and sulfonic acid esters6 have succeeded
as non-salt-type latent initiators, overcoming the problems of
onium salts such as low solubility in monomers and solvents,
remaining of inorganic compounds in polymers, and high cost.
Recently, we have reported that phosphonic acid esters can
serve as non-salt-type latent initiators in the polymerization of
glycidyl phenyl ether (GPE).7 We report herein a novel latent
initiator, thiophosphonic acid ester (1) in comparison with the
corresponding phosphonic acid ester (2).
S,S-Di-t-butyl phenylthiophosphonate (1) was synthesized
by the reaction of phenylthiophosphonic dichloride with t-butyl
mercaptan in the presence of sodium hydride, and the structure
1
was confirmed by H, 13C, 31P NMR, and IR spectroscopy
besides element analysis.8 Polymerization of GPE with 1 (1
mol%) was carried out at 170-190 °C for 12 h. The phenylth-
iophosphonate (1) was completely soluble in GPE at ambient
temperature and the polymerization proceeded homogeneously,
but the conversion of GPE was below 4% under these condi-
tions to obtain only a trace amount of GPE oligomer (Mn 400).
The nucleophilic thiophosphonate anion formed by the thermal
decomposition of 1 might terminate the cationic polymeriza-
tion of GPE. We have recently found that ZnCl2 is effective to
decrease the nucleophilicity of a phosphonate anion generated
from a phosphonic acid ester, resulting in a high initiator activ-
ity.7 Therefore, the polymerization of GPE with 1 (1 mol%)
was carried out in the presence of ZnCl2 (0.2 mol%) at 90-170
°C for 12 h (Scheme 1). The polymerization of GPE did not
proceed below 90 °C, but proceeded rapidly above 90 °C to
afford the polymers with Mn of 2500-3500, as shown in Figure
1 and Table 1. GPE was converted quantitatively at 170 °C.
The activity of thiophosphonate 1 was higher than that of the
corresponding phosphonate 2.9,10
ference of the initiator activity of the thiophosphonate 1 and
the phosphonate 2.11 As shown in Table 2, the t-butyl proton
of 1 showed a larger atomic charge and longer C-H bond
length than those of 2. It was confirmed that the t-butyl proton
of 1 would be more easily eliminated to form phenyl thiophos-
phonic acid compared with 2, resulting in the higher activity of
1 because the initiating species should be the eliminated proton
Ab initio calculations were carried out to examine the dif-

Chemistry Letters 2000
101
Allen, Applied Science Publishers: Essex, England (1981),
Chap. 1.
2
3
T. Endo and F. Sanda, Macromol. Symp., 107, 237 (1996).
T. Takata, Y. Z. Menceloglu, and T. Endo, J. Polym. Sci.,
Part A: Polym. Chem., 30, 501 (1992).
4
5
6
S. D. Lee, F. Sanda, and T. Endo, J. Polym. Sci., Part A:
Polym. Chem., 35, 689 (1997).
T. Moriguchi, Y. Nakane, T. Takata, and T. Endo,
Macromolecules., 28, 4334 (1995).
a) S. D. Lee, T. Takata, and T. Endo, Macromolecules.,
29, 3317 (1996). b) S. D. Lee, T. Takata, and T. Endo, J.
Polym. Sci., Part A: Polym. Chem., 37, 293 (1999).
M. S. Kim, F. Sanda, and T. Endo, Macromolecules in
press.
7
8
1: Yield: 76.2%. mp: 60.5-61.5 °C. IR (KBr, cm-1): 3057,
2961, 2922, 2858, 1456, 1437, 1365, 1157, 1093, 1026,
802, 713, 688, 671; 1H NMR (CDCl3): δ 8.22-8.15, 7.52-
7.26 (m, 5H, -C6H5), 1.45 (S, 18H, -6(CH3)); 13C NMR
(CDCl3): δ 139.1, 138.4, 131.4, 128.2, 54.7, 32.3; 31P
NMR (CDCl3): δ 64.43. Anal. Found: C, 52.71; H, 6.89;
S, 30.09%. Calcd for C14H23PS3: C, 52.80; H, 7.28; S,
30.20%.
in this polymerization.7 The polymer obtained in the cationic
polymerization of GPE with 2 in the presence of ZnCl2 at 110
°C exhibited a bimodal GPC peak, probably due to two kinds
of active species; free ion and ion pair.7 Meanwhile, the poly-
mer obtained with 1 under the similar conditions showed a uni-
modal GPC curve, although it contained a small shoulder at a
high molecular weight region.
In summary, although the thiophosphonate 1 alone showed
little activity as the initiator of the polymerization of GPE, it
served as a good thermally latent initiator in the presence of
ZnCl2. Ab initio calculation could explain the higher activity
of 1 than the corresponding phosphonate 2. As far as we
know, this is the first attempt to enhance the activity of a latent
initiator by replacing the oxygen atom into a sulfur atom.
9
The regio regularities (head-to-tail contents) of polyGPEs
obtained by the polymerization at 150 °C with 1 and 2 as
the initiators were estimated to be 79 and 74%, respective-
ly, by comparing the 13C NMR integration ratio of the
regio regular and irregular parts (NNE mode, 125.65
MHz, in CDCl3). The assignment was based on Ref 10.
10 J. C. Ronda, A. Serra, A. Mantecon, and V. Cadiz,
Polymer, 36, 471 (1995).
11 All calculations were carried out with the Gaussian 94
programs on a Silicon Graphics Indigo 2 IMPACT 10000.
Geometries were fully optimized by the HF/STO-3G basis
set, followed by a single point calculation by the HF/6-
311G** basis set.
References and Notes
1
a) J. V. Crivello and H. W. Lam, J. Polym. Sci., Polym.
Chem. Ed., 17, 977 (1979). b) J. V. Crivello, in
“Developements in Polymer Photochemistry,” ed by N. S.