Synthesis of a hydrophilic phosphonic acid monomer for dental materials

Article abstract of DOI:10.1039/a909877a

A novel phosphonic acid monomer for use in dental composites was prepared by the base-catalysed rearrangement of the corresponding diethyl arylphosphonate.

Full text of DOI:10.1039/a909877a

Synthesis of a hydrophilic phosphonic acid monomer for dental materials
Liyuan Mou,a Gurdial Singh*a and John W. Nicholsonb
a Department of Chemistry, University of Sunderland, Sunderland, UK SR1 3SD.
E-mail: gurdial.singh@sunderland.ac.uk
b Dental Biomaterials Department, GKT Dental Institute, Kings College, University of London, London, UK SE1 9RT
Received (in Liverpool, UK) 16th December 1999, Accepted 25th January 2000
A novel phosphonic acid monomer for use in dental
composites was prepared by the base-catalysed rearrange-
ment of the corresponding diethyl arylphosphonate.
In general dentistry may be described as a profession that is
engaged with the repair and treatment of the teeth in order to
restore proper function and good aesthetics. One of the major
problems encountered in the treatment of teeth is dental caries,
which is a preventable condition but one that is widespread and
is possibly the most common disease in the world.1 The onset of
dental caries is due to metabolic activity by the bacteria of the
plaque leading to production of acids from sugar in the diet,
resulting in a localised attack on the enamel and dentine of the
tooth.2 The infection caused by caries can lead to other more
serious conditions and can be fatal in the case of endocarditis,3
and as a consequence dental repair can be seen as important.
Repair materials have been the subject of considerable research,
with recent emphasis on aesthetic materials that are capable of
bonding to the cleaned tooth surface.
In the present work, we have sought to prepare an acid-
functional monomer for use in dental restorative materials.
There are existing filling materials that contain acid-functional
monomers, but they are aliphatic, and the resulting materials
have relatively poor mechanical properties.4 We anticipate that
aromatic monomers, such as 1, will yield materials with
be unfruitful. The alternative strategy of lithio anion formation.7
followed by subsequent trapping led to complex mixtures that
contained a small amount of the desired product, as evidenced
by NMR.
Our successful approach to the synthesis of 1 is detailed in
reaction Scheme 1. Bisphenol 4 was treated with NaH and
chlorodiethylphosphite and afforded the di-O-phosphate 5 in
50% yield.† An improved yield (75%) of 5 was obtained if the
reaction was conducted using diethyl phosphite in the presence
of CCl4 and triethylamine.8 With the diphosphate 5 in hand we
examined its rearrangement to the C-phosphonate as similar
processes had been reported by Melvin at Pfizer9 for simpler
systems.10 Thus treatment of the diaryl phosphate 5 with a 2.5
fold excess of lithium diisopropylamide at 278 °C resulted in
the clean rearrangement to the diarylphosphonate 6 in 86%
isolated yield after recrystallisation. That the rearrangement had
occurred to give 6 was strongly supported by its spectral
1
properties, in particular the H, 13C and 31P NMR were most
improved mechanical properties, because of the similarity with
the aromatic dimethacrylate monomers currently used ex-
tensively in dentistry.
We also reasoned that the incorporation of a phosphonic
function into monomer structures would result in increased
biocompatability and adhesion to the tooth due to chelation with
calcium ions in the tooth surface.
Here we report on the synthesis of the novel monomer 1 in
which we have incorporated a phosphonic acid residue in both
aromatic rings as well as incorporating a methylmethacrylate
function that is capable of ionic or radical polymerisation.
At the outset we investigated the formation of the aryl
phosphonate 6 employing Michaelis–Arbuzov reaction with the
2,2A-dibromobisphenol 2 and triethyl phosphite and with O-
dimethoxy analogues under a wide range of conditions,5 and all
of these attempts proved to be unfruitful. As a result of these
findings we explored the chemistry of the di-O-isopropylidene
protected dibromoaryl ether 3 employing Michaelis–Arbuzov
conditions and also the Hirao modification.6 Once again these
attempts at the introduction of the phosphonate group proved to
Scheme 1 Reagents and conditions: i, NEt3 (2 equiv.), diethyl phosphite (2
equiv.), CCl4, room temp., 30 h; ii, LDA (5 equiv.), THF, 278 °C, 1 h, then
3 h at room temp.; iii, glycidol (5 equiv.), cat. NEt3, heat, 3 h; iv, Py,
methacrylic anhydride (3.5 equiv.), cat. DMAP, room temp., 18 h; v,
TMSBr (9 equiv.), room temp., 18 h, followed by aq. MeOH.
DOI: 10.1039/a909877a
Chem. Commun., 2000, 345–346
This journal is © The Royal Society of Chemistry 2000
345

informative. In the 1H NMR there was significant 1H–31P
coupling to the ortho and meta protons (JP–H 11.2 and 7.2 Hz)
undertake the removal of the phosphonate ester groups and
gratifyingly this afforded the desired acid monomer 1 in 60%.
Alternatively, it could be isolated as its dicyclohexylammonium
salt.
In summary, we have successfully developed the synthesis of
the phosphonic acids 1 employing a double phosphate re-
arrangement reaction in excellent overall yield.
Polymerisation studies of 1 will be reported elsewhere. To
date preliminary biocompatability studies are showing promise
and these are ongoing.
of the aromatic rings; this coupling was absent in 5. In the 13
C
NMR, coupling of 13C–31P was observed for the aromatic
resonances (JC–P 179.6, 13.5 Hz). The 31P NMR also provided
convincing support for the assignment, as the resonance due to
the phosphonate was observed at d 221.54 for 6 whilst in the
phosphate 5 the resonance was found to be at d 27.21.
This type of rearrangement reaction most likely proceeds via
ortho-directed metallation of the aromatic ring followed by
migration of the phosphorus to the cabanion centre, resulting in
the formation of the phenolic phosphate. During the course of
these studies we examined this rearrangement using 31P NMR
but we were unable to detect any intermediates during this
process. We believe that this is the first example of this type of
double phosphate rearrangement to provide the diphosphonate
6.
We thank the EPSRC for support of this work and for access
to central facilities for high resolution mass spectrometric data
at the University of Wales, Swansea.
Notes and references
Having the desired arylphosphonate 6 in hand alkylation of
both the phenolic functions was accomplished with glycidol in
the presence of triethylamine to afford 7 in 50% yield. Selective
methacrylation of the primary hydroxy functions at both termini
was accomplished with either methacroyl chloride in the
presence of pyridine or with methacrylic anhydride in the
presence of a catalytic amount of DMAP. The latter sequence
was preferred as it gave rise to the desired product 8 in 50%
yield as compared to 29% for the former reaction procedure. In
both of these esterification reactions we observed the formation
of doubly and singly esterifed products that were separated by
column chromatography. The mono ester thus obtained could
be converted to the diester on further treatment with either of the
two acylating agents.
At this juncture it remained for us to selectively remove the
ethyl ester protecting group of the phosphinates in 8. We
examined a range of procedures, including acid and base
hydrolysis, and all of these methods proved to be unfruitful,
with many resulting in C–O bond cleavage and removal of the
methacroyl group. As a result of this we undertook model
deprotection experiments with both the phosphonates 5 and 7
using trimethylsilyl bromide.11 This treatment resulted in the
formation of the respective phosphonic acids 9 and 10 as
† All new compounds gave satisfactory spectral, microanalytical and/or
high resolution mass spectrometry. Selected data for 1: dH(270 MHz,
CDCl3) 1.65 (s, 6H), 1.95 (s, 6H), 3.76–4.29 (m, 12H), 5.66 (m, 2H), 6.18
(s, 2H), 6.85 (br d, 2H, J 5.1), 7.28 (br d, 2H, J 8.5), 7.84 (br d, 2H, J 14.1);
dP(109.25 MHz, CDCl3) +18.48; dC(67.8 MHz, CDCl3) 30.55, 41.83,
41.85, 64.62, 67.87, 70.92, 112.92 (JC–P 10.9), 117.04 (JC–P 189), 125.93,
131.62 (JC–P 13.5), 132.35 (JC–P 4.5), 135.84, 143.11 (JC–P 21), 158.39
(JC–P 3.6 Hz), 167.04; nmax(thin film)/cm21 3324–2716, 1720, 1600, 1490,
1294, 1168, 941 (found: M+ + H, 673.1840, C29H39O14P2 requires
673.1815; dicyclohexylammonium salt (FAB) found: M+ + H, 1035,
C53H85N2O14P2 requires 1035). For 5: dH(270 MHz, CDCl3) 1.28 (t, 12H,
J 7.2), 1.62 (s, 6H), 4.12 (q, 8H, J 7.2), 7.03–7.33 (m, 8H); dP(109.25 MHz,
CDCl3) 27.21; dC 15.98 (d, JC–P 6.7), 30.81, 42.17, 64.43 (JC–P 6.0), 119.31
(JC–P 4.9), 127.96, 146.98 (JC–P 1.3), 148.63 (JC–P 7.1); nmax(thin film)/
cm21 3451, 3281, 1604, 1505, 1271, 1033, 967. For 6: mp 93–94 °C;
dH(270 MHz, CDCl3) 1.28 (t, 12H, J 7.2), 1.62 (s, 6H), 4.10 (q, 8H, J 7.2),
6.86 (dd, 2H, J 9.5, 7.2), 7.20 (dd, 2H, J 11.2, 2.6), 7.23 (dd, 2H, J 9.5, 2.6);
dP(109.25 MHz, CDCl3) 221.54; dC(67.8 MHz, CDCl3) 16.09 (d, JC–P 6.2),
30.67, 41.65, 64.64 (JC–P 4.7), 108.04 (JC–P 179.6), 117.38 (JC–P 13), 128.80
(JC–P 7), 134.06 (JC–P 2.1), 141.42 (JC–p 12.7), 160.17 (JC–P 7.2);
nmax(KBr)/cm21 3451, 3137, 1602, 1405, 1209, 1027, 979.
1 T. R. Pitt Ford, The restoration of teeth, Blackwell Scientific, Oxford,
1985.
2 J. W. Nicholson, Educ. Chem., 1994, 31, 12.
3 A comprehensive guide to clinical dentistry, ed. A. H. Rowel, Class,
London, 1989, vol. 3.
4 K. Miyazaki, T. Horibe, J. M. Antoncucci, S. Takagi and L. C. Chow,
Dent. Mater., 1993, 9, 41; J. W. Nicholson and M. Alsarhedd, J. Oral
Rehabil., 1998, 25, 616.
5 P. Tavs and F. Korte, Tetrahedron, 1967, 23, 4677; P. Tavs, Chem. Ber.,
1970, 103, 2428.
6 T. Hirao and T. Masunaga, Synthesis, 1981, 56.
7 V. E. Baulin, V. Kh. Syundyukova and E. N. Tsvetkov, Zh. Obsch.
Khim., 1989, 59, 62.
8 B. Dhawan and D. Redmore, J. Org. Chem., 1984, 49, 4018.
9 L. S. Melvin, Tetrahedron Lett., 1981, 22, 3375.
10 For the S-variant, see: S. Masson, J.-F. Saint-Clair and M. Saquet,
Synthesis, 1993, 485.
dicyclohexylammonium salts, after hydrolysis, in yields of 55
and 96%. As a result of these findings we used trimethylsilyl
bromide followed by hydrolysis with aqueous methanol to
11 G. A. Olah and S. C. Narang, Tetrahedron, 1982, 38, 2225.
Communication a909877a
346
Chem. Commun., 2000, 345–346