Cyclic phosphonitrilic compounds bearing -N=PCL3 group as homogenous catalyst towards the silanol condensation

Article abstract of DOI:10.1080/10426500490466571

Condensation reactions of 1,1,3,3,3 pentamethyldisiloxane-ol (MDH), in toluene catalysed by two types of phosphonitrilic catalyst bearing a phosphazene -N=PCl₃, were studied using sampling and gas chromatographic analysis method. The condensation kinetics reactions were compared for both phosphazene compounds. The process is selective, leading to linear decamethyltetrasiloxane MD₂M, (where D denotes the dimethylsiloxane unit and M denotes the trimethyloxysilane unit) as almost the exclusive primary product. The mechanisms of the MDH condensation are discussed using the ³¹P NMR sampling technique.

Full text of DOI:10.1080/10426500490466571

This article was downloaded by: [North Dakota State University]
On: 02 November 2014, At: 11:53
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Phosphorus, Sulfur, and Silicon and the Related
Elements
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gpss20
CYCLIC PHOSPHONITRILIC COMPOUNDS BEARING
─
N═PCl GROUP AS HOMOGENOUS CATALYST TOWARDS
3
THE SILANOL CONDENSATION
a
a
a
b
Yann Vaugeois , Ahmed Mazzah , Roger Jaeger & Jean Habimana
a
Laboratoire de Spectrochimie Infrarouge et Raman, CNRS, Université des Sciences et
Technologies de Lille , France
b
Dow Corning S. A. Parc Industriel , Seneffe, Belgium
Published online: 16 Aug 2010.
To cite this article: Yann Vaugeois , Ahmed Mazzah , Roger Jaeger & Jean Habimana (2004) CYCLIC PHOSPHONITRILIC
COMPOUNDS BEARING ─N═PCl GROUP AS HOMOGENOUS CATALYST TOWARDS THE SILANOL CONDENSATION, Phosphorus,
3
Sulfur, and Silicon and the Related Elements, 179:9, 1819-1840, DOI: 10.1080/10426500490466571
To link to this article: http://dx.doi.org/10.1080/10426500490466571
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Phosphorus, Sulfur, and Silicon, 179:1819–1840, 2004
Copyright ꢀC Taylor & Francis Inc.
ISSN: 1042-6507 print / 1563-5325 online
DOI: 10.1080/10426500490466571
CYCLIC PHOSPHONITRILIC COMPOUNDS BEARING
N PCL GROUP AS HOMOGENOUS CATALYST
3
TOWARDS THE SILANOL CONDENSATION
a
a
a
Yann Vaugeois, Ahmed Mazzah, Roger de Jaeger,
b
and Jean Habimana
Laboratoire de Spectrochimie Infrarouge et Raman, CNRS,
a
Universit e´ des Sciences et Technologies de Lille, France; and
b
Dow Corning S. A. Parc Industriel, Seneffe, Belgium
(
Received July 15, 2003; accepted January 22, 2004)
Condensation reactions of 1,1,3,3,3 pentamethyldisiloxane-ol (MDH),
in toluene catalysed by two types of phosphonitrilic catalyst bear-
ing a phosphazene N PCl , were studied using sampling and gas
3
chromatographic analysis method. The condensation kinetics reactions
were compared for both phosphazene compounds. The process is selec-
2
tive, leading to linear decamethyltetrasiloxane MD M, (where D denotes
the dimethylsiloxane unit and M denotes the trimethyloxysilane unit)
as almost the exclusive primary product. The mechanisms of the MDH
3
1
condensation are discussed using the P NMR sampling technique.
Keywords: Cyclic phosphazenes; polysiloxane; reactivity mechanism;
silanol condensation
INTRODUCTION
There has been considerable interest in the chemistry of phosphazene
compounds as catalysts for the polycondensation and polymerization
of siloxane polymers.1
–16
Thus, silanol-ended linear dimethylsiloxane
oligomers may be transformed at ambient temperature within min-
5
utes into a polymer of high molecular weight, above 10 , using per-
chlorooligophosphazenium salts. The advantage of using these cata-
lysts is that the product of the polycondensation does not contain cyclic
The authors are grateful to Dow Corning Company for financial support and for the
permission to publish this data.
Address correspondence to Ahmed Mazzah, Laboratoire de Spectrochimie Infrarouge
et Raman, CNRS, UMR 8516, B aˆ t. C5, Universit e´ des Sciences et Technologies de Lille,
France. E-mail: Ahmed.Mazzah@univ-lille1.fr
1
819

1
820
Y. Vaugeois et al.
oligosiloxanes. These phosphonitrile chloride oligomers also promote
the redistribution of linear polysiloxanes without formation of a signif-
icant amount of cyclic species.1
7,18
The key to understanding the catal-
ysis in this process is to gain knowledge of the kinetics of the condensa-
tion as well as information on the reactivity of phosphazene compounds
with silicon–oxygen compounds such as silanol, siloxanes, and alkoxysi-
lanes. According to the recent literature, the mechanisms of these im-
portant reactions have extensively been investigated by Chojnowski
1
9,20
et al.
We should also note the theoretical study concerning the con-
densation reactions in silanol-water clusterrs using B3LYP and MP2
2
1
methods, which have been recently investigated by Ignatev et al. The
aim of this study was to explore the catalytic effect of water in the
2
H3SiOH + nH2O (n = 1, 2) systems.
As part of our study, we have recently studied the kinetics and
mechanism of reaction between oligosiloxanes and P-Trichloro-N-
dichlorophosphoryl monophosphazene Cl3P N P(O)Cl2.22 The pur-
pose of this work was to explore the action of the N PCl3 unit
toward the oligosiloxane compounds and the role of the reaction prod-
ucts in the siloxane polycondensation or redistribution process. Indeed,
it was demonstrated in this work that only the monosubstitution of
a chlorine atom by the (OSiMe2)n-OSiMe3 species n = 0, 2, 3 with
elimination of trimethylchlorosilane occurred, leading to the deriva-
tives Cl2OP N PCl2O-(SiMe2-O)n SiMe3. A two-step mechanism is
proposed, consisting in a nucleophilic substitution, involving a trico-
ordinate phosphazenium intermediate, followed by the formation of an
active ionic centre probably an oxonium ion.
In this article we present a kinetic study of the model silicon func-
tional compound pentamethyldisiloxane-ol (MDH) catalyzed by model
cyclic phosphazene compounds bearing a ( N PCl3) group.
The preparation of different cyclic phosphazenes compounds (2)–(8)
are described in (Scheme 1):
1
,1,3,3,3-Pentamethyldisiloxane-1-ol (MDH) was used as the model
substrate in this polycondensation study because it is the simplest sta-
ble oligosiloxanol and its condensation leads to only one product, MD2M,
where D and M denote the dimethylsiloxane and M the trimethylsilox-
ane units, respectively, according to Equation (1):
Well-defined cyclic phosphazene species containing two or one
N PCl3 units (3) and (8), respectively, were selected to promote the
MDH condensation. The cyclic phosphazene compounds (3) and (8) are

Cyclic Phosphazenes
1821
SCHEME 1 Synthesis of cyclic phosphazene compounds (2)–(8).
well-known and well-characterized chemical compounds, their purity
was controlled by 31P NMR (nuclear magnetic resonance).
RESULTS AND DISCUSSION
Kinetics of Condensation of 1,1,3,3,3-
Pentamethyl-disiloxane-1-ol
Reactions were performed in toluene. The initial concentration of
[
(
1
MDH]0 was 1 Mol/l and 0.7 mol/l in the cases of the catalysts (3) and
8), respectively. In both cases the concentration of the catalyst was
−3
0
mol/l. Kinetics were investigated by sampling the reaction mix-
ture and determining the substrate MDH concentrations by gas chro-
matography. Reactions were followed to a conversion of 60–98%.

1
822
Y. Vaugeois et al.
Since equilibrium constants of the silanol condensation reactions are
relatively high,2
3–25
the reaction in Equation (1) is assumed here to be
irreversible. In this case, Equation (1) is represented by the following
kinetic equation:
1
d[MDH]
dt
α
β
Rate = −2
= K.[MDH] .[cata]
m = α + β
where m presents the global order of the reaction, α and β represent the
partial orders, and K is the rate constant of the condensation reaction.
If we consider that the concentration in catalyst remains constant,
the rate is reduced to the form
−
d[MDH]
dt
α
β
=
2Kapp[MDH] , with Kapp = K.[cat]
Kapp being the apparent rate constant.
Study of the Catalytic Activity of (3)
The influences of the initial concentration of compound (3), of the tem-
perature and of the HCl (stemming from the hydrolysis of the exocyclic
P Cl bonds) on the kinetics of the reaction were examined.
Determination of the Partial Orders with Regard to the
MDH and to the Catalyst
◦
At first, we studied the kinetics of the reaction at 40 C with initial
−
3
concentrations in MDH and (3) around 1 and 10 mol/l, respectively.
The reaction proceeds according to the second-order kinetic law for
the catalyst (3) until 80% MDH conversion (Figures 1 and 2). Afterwards
the condensation reaction slowed down, presumably as a result of the
irreversible loss on decomposition of an active form of the catalyst. The
−
3
constant rate deducted from the curve of Figure 2 is Kapp = 1.45 10 l.
−1
−1
mol .s
.
A considerable amount of water was released during the reaction,
and a dispersed water phase appeared in the early stage of the process.
Thus, a stationary water concentration in the organic phase is estab-
lished very soon after the beginning of the reaction. For this purpose,
this kinetic reaction was compared with that of a MDH condensation
◦
carried out at 40 C in the toluene, catalyzed by HCl in the concentra-
−2
tion of 10 mol/l. Indeed, the hydrolysis of (3) led to the formation of
such acid, which catalyzes the condensation of silandiols. The HCl
2
6

Cyclic Phosphazenes
1823
FIGURE 1 Substrate conversion time dependences for the condensation
◦
−1
of MDH catalyzed by (3) in toluene at 40 C, [MDH]
0
= 1 mol. l , [(3)] =
−3
−1
10
mol.l .
FIGURE 2 The second-order plot for the substrate conversion versus time.

1
824
Y. Vaugeois et al.
FIGURE 3 Substrate conversion time dependences for the condensation of
◦
−1
MDH catalyzed by HCl in toluene at 40 C, [MDH]
0
= 1 mol.l , [HCl] =
−2
−1
10
mol.l .
concentration chosen in this study corresponds to the total concentra-
tion of chlorine atoms in compounds (3).
1
1
The curves [MDH] = f(time) and
−
= f(time) are repre-
[
MDH]
[MDH]0
sented in Figures 3 and 4. In this case, we should note that the reaction
also obeys a second-order kinetic law, with an apparent rate constant
−3
−1 −1
Kapp = 0.9 10 mol .s
.
The necessary time to have the conversion of 80% of the MDH is
about 3600 s. This comparison shows that MDH’s condensation in the
presence of (3) cannot be explained by a simple catalysis of HCl pro-
duced from the hydrolysis of the P Cl bonds.
FIGURE 4 The second-order plot for the substrate conversion versus time.

Cyclic Phosphazenes
1825
TABLE I Comparison of Second-Order
Rate Constant for the Condensation of MDH
−
1
in the Presence of (3), [MDH]
0
= 1 mol l
−1
−1 −1
[
3] mol.l
Kapp (l.mol .s )
−
−
−
−
3
3
3
3
−3
−3
−3
−3
5
.10
1.375.10
0.3.10
0.25.10
0.143.10
1
0
0
.85.10
.5.10
0
A series of runs was also performed at different catalyst concentra-
−
3
−3
−3
−3
◦
tions (5.10 , 10 , 0.85.10 , and 0.5. 10 mol/l) at 25 C. The initial
concentration in MDH in all cases was equal to 1 mol/l. The apparent
rate constant for each experiment is summarized in Table I. The curve
in Figure 5 represents the evolution of Ln Kapp versus Ln (3). The slope
in a double logarithmic plot of the rate-concentration dependance gives
a value equal to 1, which would tend to show that a mole of (3) catalyses
the reaction of condensation of two MDH moles. It is interesting to note
2
0
that this value is different from that proposed by Chojnowsky et al.,
who quoted a value of 1.5 for both catalysts Cl3P N P(O)Cl2 (P2O) and
Cl3P N PCl3] [SbCl6] (P ).
+
−
+
[
2
The results of the experiments performed at different temperatures
allowed the estimation of the activation energie of the catalyst. Indeed,
FIGURE 5 Dependence of Ln of the second-order rate constant on the Ln of
the concentration of catalyst for the condensation of MDH in the presence of
= 1 mol l− , 25 C.
1
◦
(3), [MDH]
0

1
826
Y. Vaugeois et al.
FIGURE 6 Substrate conversion time dependences for the condensation of
MDH catalyzed by (3) at different temperatures in toluene, [MDH]
0
= 1 mol
−1
−3
l
, [(3)] = 10 mol l-1, ◦ 298 K, • 313 K, ꢀ 333 K.
three studies of the reaction of condensation were made at three differ-
ent temperatures: 25, 40, and 60 C, with initial concentrations of MDH
◦
−
3
and catalyst of [MDH]0 = 1 mol/l and (3) = 10 mol/l. In Figures 6 and
1
1
7
are represented the curves [MDH] = f(time) and
−
=
[
MDH]
[MDH]0
f(time). We can note that the rate of condensation increases with the
FIGURE 7 Second-order plot for the condensation of MDH in toluene in the
−
1
−3
presence of (3) at different temperature, [MDH]
0
= 1 mol l , [(3)] = 10 mol
−1
l
, ◦ 298 K, • 313 K, ꢀ 333 K.

Cyclic Phosphazenes
1827
TABLE II Comparison of Second-Order Rate
Constant for the Condensation of MDH Catalyzed by
−
1
(
[
3) at Different Temperatures, [MDH]
0
= 1 mol l
,
(3)] = 10− mol l
3
−1
Temperature (K)
Kapp (l.mol 1.s−1
−
)
−
−
−
3
3
3
2
3
3
98
13
33
0.3.10
1.45.10
2.95.10
temperature. The values of the apparent constant rate according to the
temperature are summarized in Table II.
Even if it is not possible to determine the real value of the energy
of activation, Arrh e´ nius’s law nevertheless allows us to calculate its
apparent value connected with Kapp according to the following relation:
Kapp = K0 exp−
Eapp/RT
in which Eapp represents the energy of activation and K0 the factor of
frequency. The curve of Figure 8 represents the variation of the Ln Kapp
1
versus . We found a value of 12.70 kcal/mol, which is very close to that
T
2
0
determined by Chojnowsky et al. for a similar study carried out with
catalysts P2NOCl5 and [PCl3NPCl3] [SbCl6] (9.1 and 12.9 kcal/mol,
respectively).
+
−
FIGURE 8 Dependence of Ln of the second-order rate constant versus 1/T
= 1 mol l− , [(3)] = 10 mol l
1
−3
−1
.
(
Determination of Eapp), [MDH]
0

1
828
Y. Vaugeois et al.
FIGURE 9 Substrate-conversion-time dependences for the condensation of
MDH catalysed by (8) in toluene at two different temperatures, [MDH]
0
=
−1
−3
−1
0
, 7 mol l , [(8)] = 10 mol l , ꢁ 313 K, ꢀ 333 K.
Study of the Catalytic Activity of (8)
An interesting observation is the considerable difference in the catalytic
activity of (3) compared to (8) in which the five chlorine atoms are re-
placed by phenoxy group. Kinetics of condensation of MDH catalyzed
◦
by (8) were carried out at two temperatures, 40 and 60 C, with initial
−
1
−3
−1
concentrations in silanol and catalyst at 0.7 mol l and 10 mol.l
respectively. The catalyst (3) reacts faster than (8) by a factor of about
0, as shown in Figures 9 and 10 and Table III. This decrease in the
,
1
FIGURE 10 Second-order plot for the condensation of MDH in toluene in the
−
1
−3
presence of (8) at a different temperature, [MDH]
mol l , ꢁ 313 K, ꢀ 333 K.
0
= 0.7 mol l , [(8)] = 10
−1

Cyclic Phosphazenes
1829
TABLE III Comparison of Second-Order Rate
Constant for the Condensation of MDH
Catalyzed by (8) at Two Different Temperatures,
= 0, 7 mol l− , [(8)] = 10 mol l
1
−3
−1
[MDH]
0
Temperature (K)
Kapp (l.mol 1.s−1
−
)
−
−
4
4
3
3
13
33
0.62.10
1.28.10
rate of the MDH condensation reaction may be explained by assuming
that the substituted cyclophosphazene compounds (8) exhibit only one
N PCl3 group by the bulk effect of the phenoxy groups and also by its
lower acidity.
MECHANISM OF THE CONDENSATION PROCESS
All of these results prompted us to investigate in detail the reactivity
of the model compound (3) towards the linear siloxane oligomers MDH
and MD2M in an attempt to gain more knowledge concerning the silox-
ane condensation and redistribution and to establish the condensation
reaction mechanism of MDH.
To our knowledge, no study concerning the hydrolysis by the liquid
water and the action of alcohols (in absence of tertiary amine), on com-
pound (3) was published. Lehr27 showed, however, that the action of two
ꢀꢀ
equivalents of formic acid on (3) leads to the hydrolysis product ( 3 ).
The 31P NMR spectrum of this compound is characterized by a dou-
2
blet at δ = 21.4 ppm (PCl2 intracyclic) ( JPP = 46 Hz), a singulet at
δ = 0.7 ppm ( NH P(O)Cl2), and a multiplet in δ = −13.1 ppm (PN4).
Some authors also studied the substitution of the chlorine atoms in (3),
ꢀ
ꢀ
(
3 ), and in a compound similar to (8) (N3P3(OC6H4X)5 N PCl3, X = p-
28,29
OCH3,
by reaction of the latter with alcohols (in presence or not

1
830
Y. Vaugeois et al.
of tertiary amine), or with alcohol salts. In the three cases, they indi-
cate the formation of products of types N3P3(OR)4(N P(OR)2OH)2 and
N3P3(OR)5(N P(OR)2OH and also postulate, on the basis of infrared
(
IR) spectra, the existence of the equilibrium in Equation (2) between
two tautomeric forms.
(2)
The analogy can also be made with the reactivity of P2NOCl5 with
3
0
alcohols (in the presence of an excess of latter), which first leads
to HN(POCl2)2 and then to the esters of imidodiphosphorique acid
HN(PO(OR)2)2.
Another fact that is worth mentioning concerns the highest reactivity
of the N PCl3 group in (3) compared to NPCl2 intracyclic group. In-
deed it is possible by reaction of (3) with the sodium phenol salt at ambi-
3
1
ent temperature to synthesize the compound N3P3Cl4(NP(OC6H5)3)2.
This should be attributed to the easier ionization of NPCl3 compared to
NPCl2.
We approached these results by investigation of the reaction of com-
pound (3) with siloxanes in the presence of H2O, with H2O alone, and
finally with MDH. The reaction in all cases was investigated by a sam-
pling technique. 31P NMR spectroscopic analysis permitted the reaction
to be followed directly. For that purpose, three tubes were prepared that
contained the following mixtures and that were maintained at ambient
◦
temperature (20 C):
tube 1, (3)/MD2M/H2O/toluene, molar ratio: [MD2M]/(3) = 2 and
[
H2O]/(3) = 10;
tube 2, (3)/H2O/toluene, molar ratio: (3)/[H2O] = 1; and
tube 3, (3)/MDH/toluene, molar ratio: [MDH]/(3) = 2.
The evolution of 31P NMR spectra versus time, in the case of tube 1,
is represented in Figure 11. The reaction takes place in heterogeneous
media (H2O is miscible neither with toluene nor with MD2M), so we
can expect a slow hydrolysis of (3).
Concerning the absence of the reactivity of (3) towards MD2M at am-
3
1
bient temperature, as long as H2O is not present no evolution of the
P
NMR spectra has been detected. Once H2O is introduced, the spectrum
is characterized first of all by the appearance of four supplementary
signals compared to the virgin spectrum of (3) (Figure 11a), a doublet

Cyclic Phosphazenes
1831
3
1
FIGURE 11
MD
b) 10 min, (c) 100 min.
P NMR proton decoupled spectra of the solution (3)/H
2
O/
◦
2
M/ toluene (tube 1), [MD
2
M]/[(3)] = 2, [H
2
O]/[(3)] = 10 at 20 C, (a) 0 min,
(

1
832
Y. Vaugeois et al.
centred at δ = 20.8 ppm (J = 59 Hz), a singlet at δ = 2.8 ppm, a dou-
blet at δ = −5.6 ppm (J = 40.9 Hz), and a multiplet at δ = −17 ppm.
On the basis of the chemical shift and the coupling constants, we
can, by comparison to the spectrum of (3), attribute the doublet at
δ = 20.8 ppm to an phosphor atom (P1) of intracyclic NPCl2 moiety and
that at δ = −5.6 ppm to an phosphor atom (P2) in exocyclic NPCl3
group. These results are in accordance with theformation of the mono-
ꢁ
hydrolyzed compound, namely (3 ), shown below:
In that case, the singlet at δ = 2.8 ppm could be attributed to
the hydrolyzed N PCl3 group (P3), to which two tautomeric forms,
N PCl2(OH) or NH P(O)Cl2, correspond. The multiplet at δ =
−
17 ppm could be then attributed to the intracyclic (P4) phosphor of
PN4 moiety.
The evolution of 31P NMR spectra versus time is characterized, first
of all (Figure 11b) by the appearance of two other doublets of weak
intensities, the one (P ), overlapping with the doublet corresponding
ꢁ
1
ꢁ
to P1 centered at δ = 20.4 ppm (J = 38.3 Hz), with the other one (P )
3
centered at δ = 0.9 ppm (J = 22 Hz). We should note that the sum of
the intensities (INPCl3 in (3) + IP2 + IP3 + IPꢁ3) is equal to that of the
intensities (INPCl2 in (3) + IP1 + IPꢁ1), and equals twice of (IPN4 in (3) +
IP4). The structure of compound (3) is well preserved.
Secondly (Figure 11c), with the appearance of a wide signal badly
ꢁ
ꢁ
ꢁꢁ
resolved (P ), centered at δ = 23.7 ppm, a multiplet P increases the
1
4
ꢁ
intensity with that of the doublet (P ), of a signal badly resolved at δ =
1
−
0.8 ppm (P5) and finally by a net increase of the doublet at δ = 0.9 ppm
ꢁ
(
P ). The attribution of these signals already mentioned will be done
3
later on in this study.
We compared this evolution to that obtained for the mixture (3)/(H2O)
toluene (molar ratio (3)/H2O = 1, tube 2). In Figure 12 are repre-
3
1
sented the P NMR spectra of the mixture (3)/H2O/MD2M/toluene
(
Figure 12a) and (3)/H2O/toluene (Figure 12b) after 30 minutes of

Cyclic Phosphazenes
1833
FIGURE 12 31P NMR proton decoupled spectra of the solution, (a)
3)/H O/MD M/toluene (tube 1) and (b) (3)/H O/toluene (tube 2) after 30 min of
reaction.
(
2
2
2
reaction. By comparing the two spectra, we can note that they are
ꢁ
ꢁ
strictly identical except for the multiplet of (P ), which is hardly percep-
1
tible. It is necessary to note that in both cases an important precipitate
arises in the beginning of the reaction, so the spectra correspond only
to the soluble products in the media.
In Figure 13, we have reported the 31P NMR spectrum of compound
3) introduced into the mixture of MDH/toluene, (MDH/(3) = 2) after
0 min of reaction, (tube 3). In this case, the formation of H2O result-
ing from MDH condensation is progressive. From the beginning of the
(
4

1
834
Y. Vaugeois et al.
FIGURE 13 31P NMR proton decoupled spectrum of the solution
3)/MDH/toluene (tube 3), [MDH]/[(3)] = 2 after 40 min of reaction.
(
reaction (after 6 min), in addition to the signals already attributed to
ꢀ
the monohydrolyzed compound (3 ), two signals of equal intensity ap-
pear, a little perceptible doublet at δ = 24.2 ppm (J = 49 Hz) as well as
a singlet at δ = 0.8 ppm. These values, compared to those obtained by
2
6
Lehr in the study of the hydrolysis of (3) by the formic acid, lead us
ꢁ
ꢁ
to attribute the doublet to the intracyclic phosphor NPCl2 (P ) and the
1
ꢁ
ꢁ
singlet to the exocyclic phosphor NHP(O)Cl2 (P ) in the dihydrolyzed
3
ꢀꢀ
compound (3 ), below:
We should also note that the signals in the range of the chemical shift
δ = 1.3 ppm to δ = −4 ppm change. Indeed, the badly resolved signal
which was attributed to P5 in the case of the mixture (3)/MD2M/H2O
was transformed to a singulet at δ = −0.4 ppm as well as a wide sig-
nal centered at δ = −2.1 ppm, the latter becoming more intense. In
this domain of chemical shift corresponds to the phosphor atoms of
P(O)(OH)nCl2−n (n
ꢂ= 0) or NHP(O)(OH)2 which is in agreement with
ꢀꢀ
a supplementary hydrolysis of (3 ).

Cyclic Phosphazenes
1835
The fact that the sum of the intensities of the signals in the range
of δ = 3.1 and δ = −4, 5 ppm becomes superior to that corresponding
ꢁ
ꢁꢁ
ꢀꢀ
to P1 + P + P shows that the structure of type (3 ) is not preserved
1
1
anymore. This could be due to the cleavage of the exocyclic NP bonds.
In this case, the precipitate is less important compared to the previous
experiments. In this study we note that even after 75 min of reaction,
55% of the phosphor atoms associated to the initial compound (3) still
remain in solution.
The presence of the doublets of equal intensities at δ = 20.4 ppm
ꢁ
ꢁ
and δ = 0.9 ppm (P and P , Figures 11b and 12), in the case of the
1
3
hydrolysis of compound (3) in MD2M and toluene or only in toluene, can
be explained by an equilibrium between two tautomeric phosphazene–
phosphazane forms according to the reaction in Equation (3).
Then, in the phosphazane form (for the exocyclic phosphor atoms) the
doublet would correspond at δ = −5.6 ppm (NPCl3) and the singlet at
δ = 2.8 ppm ( NHP(O)Cl2), and in the phosphazene form the doublet
would correspond at δ = 0.9 ppm. The uniqueness of this last signal
can be explained by a reversible and fast migration of the hydroxyl
group between the exocyclic phosphor atoms according to the reaction
in Equation (4).
(4)
This migration was already reported by Kabachnik et al.,32 in the
◦
case of the silylated monophosphazene compounds above −30 C ac-
cording to the reaction in Equation (5).

1
836
Y. Vaugeois et al.
(5)
However, this phenomenon is quasi nonexistent in the case of the
reaction of (3) with MDH. Indeed, only the hydrolysis products of (3),
ꢀ
ꢀꢀ
namely (3 ) and (3 ), were formed. (the analogy can be made with the
reaction of Cl3PNP(O)Cl2 with MDH2 ). The formation of such com-
pounds can be explained by the succession of the reactions in Equations
0,21
(
8)–(10). The first step consists in the formation of the silylated inter-
mediate of type (I), according to the reaction in Equation (6):
(6)
The immediate passage to the hydroxylated form (II) can be made
by the action of HCl on (I) followed by the cleavage of P O Si bond
according to the reaction in Equation (7) and by a desilylation reaction,
similar to that of the silylated esters of the imidodiphosphoric acid with
alcohols in excess,32 which in our case led to the intermediate (II) ac-
cording to Equation (8), the reaction in the right direction 1 being faster:
(
(
7)
8)

Cyclic Phosphazenes
1837
We can also think out the reaction of H2O with (I) according to
Equation (9):
(9)
This reaction must, however, be less important in the beginning of
condensation, a period in which the concentration of H2O is weak with
regard to that of the MDH.
The reactions in Equations (6)–(8) can be considered as initiation
reactions for the MDH condensation. The active entities could be the
ꢀ
hydroxylated compounds (3) and (3 ), (whose concentration should be
proportional to compound (3)), associated to the released HCl, which
can also play a role but, as has been shown, alone could not explain the
kinetics of the condensation reaction.
According to the mechanism proposed to explain the acid catalysis
of the polycondensation of silanols, it is then logical to suppose the
ꢀ
formation of an oxonium ion from (3) and (3 ) according to the reaction
in Equation (10).
(10)
Followed by the reaction of the silanol on this ion according to the
reaction in Equation (11).
(11)
So, the order 1 with regard to the catalyst can be explained by the fact
that the protonated form (CatH) is present in a quickly reached equilib-
rium, which does not determine the rate of the condensation reaction.

1
838
Y. Vaugeois et al.
So, the concentration of the oxonium ions is proportional to CatH—in
our case to (3). Therefore, the rate of the condensation reaction which
is imposed by the reaction in Equation (11) is second order in MDH
according to
k3k1
k2
[catH][MDH]2
V =
×
−
[cat ]
Everything that precedes brings us to suggest the reactional mech-
anism of Scheme 2.
SCHEME 2
EXPERIMENTAL SECTION
Chemicals
Me3SiOSiMe2OH was synthesized by hydrolysis of 1,1,1,3,3-
pentamethyldisiloxane (ABCR reagent grade) catalyzed by Aldrich 5%
palladium on actived carbon and purified by distillation. The purity
cheeked by gas chromatography was 99%.
Compounds used for the determination of response factors in
gas chromatography, i.e., Me3Si[OSiMe2]n-OSiMe3, n = 0, 1, 2, and

Cyclic Phosphazenes
1839
Me3SiOH, were obtained and purified using well-known procedures.
Their purities determined by gas chromatography were at least 99%.
All experiments were carried out under an atmosphere of dry oxygen-
free nitrogen. Solvents were dried over molecular sieves and distilled
prior to use. The preparation of different cyclic phosphazenes com-
pounds (2)–(8), have been performed according to the different liter-
2
7–33
34
35
36
27
28
ature: (2);
(3); (4) and (5); (6); (7); and (8).
Kinetics Studies
The condensation process was carried out in a glass 10 ml reactor fit-
ted with a three-way precision stopcock. The reactor was purged with
dry nitrogen and charged with the required amounts of solvent, stan-
dard for gas chromatography (dodecane), and substrates by means of
precision Hamilton syringes, keeping a stream of dry nitrogen through
the stopcock. The reactor was thermostated by Maton TH30 thermo-
◦
stat, keeping the temperature constant within ± 0.1 C. The zero sample
was withdrawn by means of a Hamilton syringe, and then the required
volume of the stock solution of catalyst was introduced by precision
Hamilton syringe. Samples (50 µl) were withdrawn with a Hamilton
syringe at required time intervals. They were introduced to Ependorfer
tubes containing a quenching agent (i.e., 20 µl of ET3N). Samples were
then subjected to gas chromatography analysis.
Gas Chromatography
Gas choromatography analyses were performed with a Shimadzu GC-
1
4A gas chromatograph working with Shimadzu CR-4A integrator.
The chromatograph was fitted with a flame ionization detector, semi-
capillary column CPSIL 5CB (l = 30 m, d = 0.53 mm), temperature 50–
◦
◦
−1
2
80 C, programmed 10 C min . Detector and injector temperatures
◦
were 300 C. Dodecane was used as internal standard.
NMR Spectroscopy
31P NMR analysis were made with a Bruker 300 WP instrument, work-
ing at 121.49 MHz frequency.
REFERENCES
[
[
1] Wacker Chemie, British Patent, 1,406,340, (1973).
2] J.-M. Gilson and J. Habimana, (Dow Corning S. A. Seneffe Bel), British Patent,
2,257,975A, (1992).

1
840
Y. Vaugeois et al.
[
[
[
[
[
[
[
3] S. Nitzche and M. Wick, (Wacker Chemie GMBH), German Patent, 1,279,019, (1968).
4] S. Nitzsche and M. Wick, Wacker Chemie, U.S. Patent, 3,839,388, (1974).
5] Wacker Chemie, U.S. Patent US-PS, 3,652,711, (1969).
6] Dow Corning, British Patent, 2,252,975A, (1992).
7] Dow Corning, U.S. Patent, 5,210,131, (1992).
8] Wacker Chemie, U.S. Patent, 2,830,967, (1958).
9] J. S. Razzano, D. P. Thompson, P. P. Anderson, and S. Rubinsztajn, (General Electric
Company, Schenectady, NY, USA), British Patent, 2,279,959, (1995).
[
[
[
[
[
10] Wacker Chemie, German Patent, 3,725,377, (1989).
11] Wacker Chemie, European Patent, 3,812,04, (1990).
12] Wacker Chemie GMBH, German Patent, 2,229,51B, (1978).
13] Wacker Chemie GMBH, German Patent, 3,725,377A, (1989).
14] R. Lehnert, H. Kr o¨ ning, H. R o¨ ster, H. Schickmann, H. Rautschek, and H. D. Wedt,
(
(
Cheminwerk N u¨ nchritz GMBH, N u¨ nchritz DE), German Patent, 4,3231,88A1,
1995).
[
15] H. Schickmann, H.-D. Wendt, and R. Lehnert, (Chemiwerk N u¨ nchritz GMBH,
N u¨ nchritz), German Patent, 4323185A1, (1995).
[
[
16] R. Hager and J. Weis, Z. Naturforsch, 49, 1774 (1994).
17] S. Rubinsztjn, (General Electric Compagny, Schenectady, NY, USA), British Patent,
2
279945A (1987).
18] Wacker Chemie GMBH, European Patent applications, 0626414A1 and 0626415A1,
1994).
19] J. Chojnowski, M. Cypryk, W. Fortuniak, K. Kazmierski, and R. Taylor, J.
Organomet. Chem., 526, 351 (1996).
20] J. Chojnowski, W. Fortuniak, J. Habimana, and R. G. Taylor, J. Organomet. Chem.,
[
[
[
[
[
[
(
5
34, 105 (1997).
21] I. S. Ignatyev, F. Partal, and J. J. Lopez Gonzalez, Chem. Phys. Lett., 368(5,6), 616
2003).
(
22] A. Meliani, Y. Vaugeois, H. Bali, A. Mazzah, and R. De Jaeger, Phosphorus, Sulfur,
and Silicon, 166, 283 (2000).
23] J. Chojnowski, In Siloxane Polymers, edited by J. Clarson and J. A. Semlyen (Ellis
Horwood PTR Prentice Hall, Englewood Cliffs, 1993), p. 1.
[
[
[
[
[
24] J. C. Saam and D. J. Huebner, J. Poly. Sci. Polym. Chem., 20, 3351 (1982).
25] L. Wilczek and J. Chojnowski, Makromol. Chem., 184, 77 (1983).
26] Z. Lasocki and S. Chrzczonowicz, J. Polymer Sci., 59, 259 (1962).
27] W. Lehr, Z. Anorg. Allg. Chem., 350, 18 (1967).
28] A. A. Volodin, V. V. Kireev, V. V. Korshak, E. A. Filippov, and V. M. Chukova, Zh.
Obshch. Khim., 42, 1503 (1972).
[
29] A. A. Volodin, V. V. Kireev, V. V. Korshak, and S. N. Zelenetskii, Zh. Obshch. Khim.,
42, 1506 (1972).
[30] L. Riesel, H. H. Patzman, and H. P. Baritch, Z. Anorg. Allg. Chem., 404, 219 (1974).
[31] H. R. Allcock and D. Ngo, Macromolecules, 25, 2802 (1992).
[32] A. M. Kabachnik, N. N. Zaslavaskaya, V. A. Gilyarov, P. V. Petrovski, and V. A.
Svoren, Doklady Akademii Nauk SSSR, 228(4), 849 (1976).
[
[
[
[
33] A. M. De Fiqelmont, Ann. Chim., 12, 169 (1939).
34] G. R. Feistel and T. Moeller, J. Inorg. Nucl. Chem., 29, 2731 (1967).
35] M. Kajiwara and Y. Kurachi, Polyhedron, 2, 1211 (1983).
36] F. Spizzo, Ph.D. thesis, University of Padova, Italy, (1992).