The gas-phase on-line production of phosphoryl halides, POX3 and their identification by infrared spectroscopy

Article abstract of DOI:10.1016/S1386-1425(98)00019-5

A new route has been devised, leading to the production of POX₃ molecules where X = F, Br and I by an on-line process using phosphoryl chloride, POCl₃ as a starting material passed over heated sodium fluoride, NaF, potassium bromide, KBr and potassium iodide, KI at 535, 690 and 480°C, respectively. The products have been characterised by the infrared (IR) spectra of their vapours. The low resolution gas-phase Fourier transform infrared spectra reported for the first time show strong bands centered at 1416.6, 1312.9, 1297.9 and 1285 cm^-1, assigned to ν₁(a₁), the O=P stretching fundamental of POF₃, POCl₃, POBr₃ and POI₃, respectively.

Full text of DOI:10.1016/S1386-1425(98)00019-5

Spectrochimica Acta Part A 54 (1998) 921–926
The gas-phase on-line production of phosphoryl halides, POX3
and their identification by infrared spectroscopy
Abdul W. Allaf *
Atomic Energy Commission of Syria, Department of Chemistry, P.O. Box 6091, Damascus, Syria
Received 28 October 1997; received in revised form 19 December 1997; accepted 23 December 1997
Abstract
A new route has been devised, leading to the production of POX molecules where X=F, Br and I by an on-line
3
process using phosphoryl chloride, POCl as a starting material passed over heated sodium fluoride, NaF, potassium
3
bromide, KBr and potassium iodide, KI at 535, 690 and 480°C, respectively. The products have been characterised
by the infrared (IR) spectra of their vapours. The low resolution gas-phase Fourier transform infrared spectra
−
1
reported for the first time show strong bands centered at 1416.6, 1312.9, 1297.9 and 1285 cm , assigned to w (a ),
1
1
the OꢀP stretching fundamental of POF , POCl , POBr and POI , respectively. © 1998 Elsevier Science B.V. All
3
3
3
3
rights reserved.
Keywords: Gas-phase; Phosphoryl halides; Infrared spectroscopy
1. Introduction
made in 1973 by the iodination of POCl with LiI,
3
or by reacting ROPI with iodine [2]:
2
Phosphoryl chloride, POCl can be made by the
3
ROPI +I 2“RI +POI
3
2
2
oxidation of PCl with pure oxygen or slightly
3
above room temperature similar to that used on
an industrial scale [1]. Phosphoryl bromide,
POBr3 can be made also by a similar method
Most of the phosphoryl halide compounds are
colourless gases or volatile liquids but POI3 is
dark violet, m.p. 53°C. All of them are
monomeric tetrahedral (C ) symmetry point
using PBr3 instead of PCl . Another route for
3
36
preparing POX3 where X=Cl and Br is as
follows:
group molecules. PO Cl produced by a photo-
2
chemical reaction between ozone and OPCl (ob-
tained from POCl ) in solid argon has been
3
P O + 6 PX “10 OPX
3
4
10
5
investigated by IR spectroscopy and ab initio
SCF calculations [3]. The same molecule is
formed by the high temperature reaction between
POCl , O and Ag. The paper includes all the
Phosphoryl fluoride, POF can be made by the
3
fluorination of POCl using a metal fluoride (ea.
3
M=Na, Mg, Zn, Pb, Ag, etc.). POI was first
3
3
2
frequencies observed for the above molecule. The
−
1
*
Tel.: +963 6111926/7; fax: +963 6112289.
frequencies were: 1122, 586 and 1429 cm , as-
1
386-1425/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved.
PII S1386-1425(98)00019-5

9
22
A.W. Allaf / Spectrochimica Acta Part A 54 (1998) 921–926
not been reported previously, using POCl as a
signed to a (PO s-stretch), PCl stretch and b
2
3
1
2
precursor passed over heated NaF, KBr and KI
for POF3, POBr3 and POI3 productions,
respectively.
(
PO a-stretch) of PO Cl, respectively. Andrews
2 2
and Withnall reported the formation of P O, P O,
4
2
PO and PO species and their matrix isolation
2
infrared spectra by the photolysis and discharge
reaction of oxygen atoms with P [4]. Binnewies
4
and Schn o¨ ckel mentioned five types of reactions
that may lead to OEX or SEX molecules produc-
tion where, E=P, AS and Sb. One of these
reactions is the reduction of phosphoryl halides,
2. Experimental
The first step was to optimize the production
conditions of phosphoryl chloride, POCl , before
3
POX with elementary silver at high temperature
attempting to observe POF , POBr and POI .
3
3
3
3
[
5]. Boghosian et. al. recorded the gas phase Ra-
POCl (Merck, 99.9%) was degased in the vacuum
3
man and UV/Vis spectra of the liquid and solid
complexes of POCl ꢁFeCl molten mixtures under
system before use with liquid nitrogen. The on-
line process involved warming the POCl3 with
heating tape to 55°C to obtain enough vapor
pressure and then pass through an empty tube
kept at 475°C. The remaining conditions were the
fast flow reaction under 0.3 Torr pressure.
3
3
static equilibrium conditions [6]. Moores and An-
drews reported some interesting work concerning
the oxidation of phosphorus(III) halides by red
photolysis of ozone complexes in solid argon [7].
The photolysis of PCl /O and PBr /O complexes
Fig. 1 shows the experimental set up for the
synthesis and spectrometric observation of phos-
3
3
3
3
by red visible radiation in solid argon gives phos-
phoryl halides. This work has been extended by
examining the photolysis of ozone in matrices
containing the phosphorus(III) halides PF , PCl
phoryl halide, POX where X=F, Br and I. The
3
synthesis routes for POF , POBr and POI can be
3
3
3
written as follows:
3
3
and PBr to generate simple compounds of phos-
3
fast flow
phorus halogen and oxygen and to explore how
the products depended on applied wavelength [7].
This paper investigates the gas phase IR spectra
and the on-line productions of phosphoryl
POCl3(g)+3NaF(s)
“3NaCl +POF
(g)
3(g)
5
35°C
fast flow
POCl3(g)+3KBr(s)
“3KCl +POBr
(g)
3(g)
6
90°C
halides, POX where X=F, Br and I which have
3
Fig. 1. The experimental set-up for the synthesis and spectrometric observation of phosphoryl halide, POX3.

A.W. Allaf / Spectrochimica Acta Part A 54 (1998) 921–926
923
−
about +1%. Comparison between the 1312.7 cm
1
, w (a ) stretching mode in POCl observed by the
1
1
3
matrix isolation technique after photolysis of PCl3/
O using 520 nm filter [7,8], with that observed by
3
Fig. 2. The gas phase IR spectrum of POCl3.
fast flow
POCl3(g)+3KI(s)
“3KCl +POI
(g) 3(g)
4
80°C
Note that, the first empty furnace (F ) is kept at
1
475°C in all the above reactions. The salts contain-
ing halogens used in these experiments, placed in
the second furnace (F ), were obtained from
2
Aldrich with a purity (99.9%).
The products from these reactions were passed
directly into an IR cell (Fig. 1) fitted with two KBr
windows. The distance between the second furnace
(
F2) and the IR cell is 3 cm. The cell was pumped
out via a liquid nitrogen trap with a rotary pump
3
−1
(
E2M8, Edwards) displacing 9.5 m
h
. The
IR-spectra were obtained on a JASCO 300E FTIR
−
1
spectrometer at a resolution of 2 cm
.
3. Results and discussion
Fig. 2 shows the result of passing POCl gas over
3
an empty tube heated at 475°C. Two strong bands
were observed within the spectrometer range at
−
1
1
1
312.9 and 590 cm . The first band centered at
−
1
312.9 cm is assigned to w (a ), the OꢀP stretch-
1
1
ing mode of POCl and shows typical PQR-type
3
structure and with a strong Q-head. The second
−
1
band at 590 cm
is assigned to w (e), mode of
4
POCl . It is known empirically that there could be
3
a shift in the band position between the matrix
isolation method and the gas-phase technique of
Fig. 3. Full IR spectra of POX where X=F, C1, Br and I in
3
−
1
the range 400–1500 cm
.

9
24
A.W. Allaf / Spectrochimica Acta Part A 54 (1998) 921–926
Table 1
Observed gas phase IR frequencies of POX in comparison with matrix isolation and Raman results including the P–R separation
3
−
1
(cm
)
Molecule
Mode
Assignment
Observed
P–R separation
Matrix [7]
isolation
Raman [9]
gas-phase
This work (IR)
gas-phase
Obs.
43.1
Cal.
POF3
w1
w4
w2
w3
a1(OꢀP str.)
e(PꢁF str.)
a1 PF3 umbrella
a1 FPO def.
1410.5
990
875
1417
991
873
472
1416.6
991.2
872
41.6
470
472.3
POCl3
POBr3
POI3
w1
w4
a1(OꢀP str.)
e(PꢁCl str.)
1312.7
589
1312.7
589.5
1312.9
590
15.4
8.7
16.2
8.8
w1
w4
a1(OꢀP str.)
e(PꢁBr str.)
1295.1
488
—
—
1297.9
490
w1
w4
a1(OꢀP str.)
e(PꢁI) str.
—
—
—
—
1285
426
3.9
3.4
gas phase Raman spectroscopy of POCl at
served gas phase IR frequencies are identical to
those obtained by gas phase Raman technique
and lies within an acceptble range with matrix
isolation technique.
3
−
1
1312.7 cm
[9] and this gas phase infrared spec-
−
1
troscopy work at 1312.9 cm , is reasonable and
fits very nicely especially with the gas phase Ra-
man results. These gas phase IR results fit also
within the expected shift between matrix isolation
and gas phase results.
There is very little trace of POCl as can be seen
3
in spectrum A and it is estimated that the purity is
about 99%. The POF sample was collected using
3
Fig. 3 shows three new spectra A, C, and D
400–1500 cm ) where spectrum B is the precur-
cold slush bath of acetonitrile mixed with liquid
nitrogen at −41°C and the IR spectrum was
taken for the collected material. The obtained
infrared spectrum was exactly identical to that
observed by the on-line process.
−
1
(
sor (POCl ). Spectrum A is the result of passing
3
POCl gas over an empty tube kept at 475°C and
3
also over heated NaF at 535°C. Four new bands
were observed at 1416.6, 991.2, 872 and 472.3
Spectrum C in Fig. 3 shows the result of pass-
−
1
−1
cm . The first band at 1416.6 cm
is assigned
ing POCl gas over heated KBr at 690°C. Two
3
−
to w (a ), the OꢀP stretching mode of POF3 and
new bands were observed at 1297.9 and 490 cm
1. The first band at 1297.9 cm
1
1
−
1
show a PQR type structure with a very strong
Q-head. This result is consistent with expectation
and the band is shifted to higher frequency when
is assigned to
w (a ), the OꢀP stretching mode of POBr and also
1
1
3
shows a PQR type structure. The shift to lower
frequency again fits with expectation after
fluorine replace the chlorine in POCl to form
3
POF which is similar to the NSX work [10]. The
bromine replaces the chlorine in order to form
3
−
1
−1
second band at 991.2 cm
is assigned to w (e)
POBr . Comparison of the 1295.1 cm
(most
4
3
−
1
POF , the third band at 872 cm
is assigned to
w (a )POF , this band has a PQR type structure
intense peak [7]) the w (a ), the OꢀP stretching
3
1
1
mode of POBr observed by matrix isolation tech-
2
1
3
3
−
1
and the final band at 472.3 cm
is assigned to
nique after full arc photolysis of PBr and ozone
3
−
1
w (a )POF and also has a PQR type structure.
[7] and these gas phase IR results at 1297.9 cm
is justified. The second band at 490 cm
,
3
1
3
−
1
The obtained gas phase IR results for POF are
is
3
compared with matrix isolation after full arc pho-
assigned to w (e) of POBr . To confirm this reac-
4
3
tolysis of PF /O [7] and with gas phase Raman
results too [9] (see Table 1). Note that, the ob-
tion, POBr3 pure sample was obtained from
Merck with a purity of 99.9% and replaced in the
3
3

A.W. Allaf / Spectrochimica Acta Part A 54 (1998) 921–926
925
gas flow system. The sample has very high vapor
pressure so the IR spectrum is recorded without
any heating and the result was similar spectrum
observed by the new route applied in this work.
The final step is shown in Fig. 3 (spectrum D)
which is the result of passing POCl3 gas over
heated KI at 480°C. The gas phase IR spectrum
of POI is recorded here for the first time and no
3
Therefore, POBr can be collected by our route
below 55°C which is the melting point of pure
POBr3.
other spectroscopic data is available. It should be
pointed out that the producing of POI3 is in
general far from being trivial due to easy sublima-
tion of iodine at high temperature of heated KI.
Nevertheless, the two chraracterized bands of
3
POI are observed. Because, there is no previous
3
experimental spectroscopic data for POI , the as-
3
signment of the obtained spectrum could be
analysed considering the previous analogous
molecules. The two bands correspond to POI are
3
−
1
centred at 1285 and 426 cm . The first band
positioned to the right side POCl mode centred
3
−
1
at 1285 cm , is assigned to w (a ), the OꢀP
1
1
stretching mode of POI . The observed band has
3
a PR type structure and this experiment has been
−
1
performed especially under 1 cm
resolution in
order to resolve this structure. The PR profile can
be seen in Fig. 3 (bottom spectrum), the Q-head
was not observed but that could be due to de-
creases in electronegativity from fluorine up to
iodine which may be responsible for broder band
in POI than of the other halides in preceding
3
−
1
spectra. The second band at 426 cm
is very
small and tentatively assigned to w (e), of POI .
4
3
As can be seen from the bottom spectrum, the
concentration of POCl is too high and it was
3
very difficult to obtain pure POI from POCl3
3
owing to easy sublimation of iodine at high tem-
perature of heated KI. Therefore, the concentra-
tion of iodine in the vacuum system is very low
compared with POCl which could lead to the
3
formation of POCl I or POClI in some stages
2
2
before forming the interested POI molecule (Fig.
3
4
).
All the observed frequencies (cm ) of POX3
−
1
determined by this gas phase IR spectroscopy
work in comparison with the data obtained by
matrix isolation and gas phase Raman results are
listed in Table 1. The Table also includes the
observed and calculated values using the method
of Paul and Dijkstra [11] for the w (a ) POX
1
1
3
where X=F, Cl, Br and I. The overall shift
between the matrix isolation and these gas phase
IR spectra of w (a ), the OꢀP main stretching
Fig. 4. The w (a ), OꢀP stretching band of POX in the range
1
1
−
3
1
(1200–1500) cm
.
1
1

9
26
A.W. Allaf / Spectrochimica Acta Part A 54 (1998) 921–926
mode of POF , POCl and POBr , is 6.1, 0.2 and
gratefully acknowledges TWAS for research grant
No. 94-025 RG/CHE/AF/AC. I am very grateful
to Drs R. Suffolk, D.R.M. Walton, H. Klawi, G.
Zazafoon and Z. Ajji for their valuable discus-
sion, as well as Miss Wafa’a Alibrahim and Miss
B. Alrez for the help.
3
3
3
−
1
2.8 cm , respectively. This shift is in agreement
with the less 1% shift between matrix isolation
and gas phase IR results.
It is concluded that the present work reports
for the first time, the on-line production and
detection of POF , POCl , POBr and POI , in the
3
3
3
3
gas-phase. These experimental data agree with
expectations and are comparable with the data
obtained by matrix isolation and gas phase Ra-
man spectroscopy. The frequency shift between
matrix isolation and this gas phase IR results is
less than 1% and in good agreement with the two
techniques. Concerning our gas phase infrared
and Raman results, both results are in excellent
.
References
[
1] N.N. Greenwood, A. Earnshaw, Chemistry of the Ele-
ments, 1st ed., Pergamon Press, Oxford, 1989.
[2] A.V. Kirsanov, Zh.K. Gorbatenko, N.G. Feshchenko,
Chemistry of phosphorus iodides, Pure Appl. Chem. 44
(
1975) 125.
[
[
3] R. Ahlrichs, C. Ehrhardt, M. Lakenbrink, S. Schunck, H.
Schn o¨ ckel, J. Am. Chem. Soc. 108 (1986) 3596.
4] L. Andrews, R. Withnall, J. Am. Chem. Soc. 110 (1988)
5605.
agreement. POF can be collected by our route at
3
−
41°C using cold slush bath of acetonitrile mixed
with liquid nitrogen. POBr can be collected also
3
[
[
5] M. Binnewies, H. Schn o¨ ckel, Chem. Rev. 90 (1990) 321.
6] S. Boghosian, G.A. Voyiatzis, G.N. Papatheodoron, J.
Chem. Soc. Dalton Trans 16 (1996) 3505.
by the new applied route at around 55°C.
[
[
7] B.W. Moors, L. Andrews, J. Phys. Chem. 93 (1989) 1902.
8] A.J. Downs, G.P. Gaskill, S.P. Saville, Inorg. Chem. 21
Acknowledgements
(
1982) 3385.
[
9] R.J.H. Clark, D.M. Rippon, Mol. Phys. 28 (1974) 305.
I would like to thank Professor Dr I. Othman
DG) and Professor Dr Y. Koudsi (Head of
Chemistry Department) for their encouragement
as well as Mr N. Odeh and Miss D. Naima for
their effort in setting up the experiments and
[
[
10] A.W. Allaf, G.Y. Matti, R.J. Suffolk, J.D. Watts, Elec-
tron Spectrosc. Relat. Phenom. 48 (1989) 411; and Chem.
Phys. Lett. 155 (1989) 32.
11] W.A. Seth Paul, G. Dijkstra, Spectrochim. Acta Part A
23A (1967) 2861.
(
.