Kinetics and mechanism of the nitrosobenzene deoxygenation by trivalent phosphorous compounds

Article abstract of DOI:10.1007/s11172-013-0359-8

The reaction of aryl nitroso compounds with organic phosphines and phosphites in aerated media is a convenient non-photolytic procedure to generate aromatic nitroso oxides. The reaction rate constants and activation parameters of the key (for the proposed method of nitroso oxide generation) reaction of nitrosobenzene with tripenyl phosphite or para-substituted phosphines (4-RC ₆H₄)₃P (R = MeO, Me, H, F), as well as that of para-methoxynitrosobenzene with triphenylphosphine in acetonitrile were determined by kinetic spectrophotometry and chemiluminescence. A significant transfer of the electron density to the nitroso compound occurs in the transition state of the reaction as was revealed using the Hammett correlation analysis and DFT calculations in the M06L/6-311+G(d,p) approximation. The introduction of the electron-donor substituent MeO into the para-position of PhNO decreases the reactivity of the nitroso compound by two orders of magnitude. The reactivity of triphenyl phosphite in the oxygen atom transfer reaction is lower by two orders of magnitude compared to that of triphenylphosphine. In the case of the reactions of PhNO with phosphines, the apparent rate constant depends on the oxygen content in the reaction medium.

Full text of DOI:10.1007/s11172-013-0359-8

Russian Chemical Bulletin, International Edition, Vol. 62, No. 11, pp. 2477—2486, November, 2013
2477
Kinetics and mechanism of the nitrosobenzene deoxygenation
by trivalent phosphorous compounds*
V. S. Khursan, V. A. Shamukaev, E. M. Chainikova, S. L. Khursan, and R. L. Safiullin
Institute of Organic Chemistry, Ufa Scientific Center, Russian Academy of Sciences,
71 prosp. Oktyabrya, 450054 Ufa, Bashkortostan, Russian Federation.
Fax: +7 (347) 235 6066. Eꢀmail: kinetic@anrb.ru
The reaction of aryl nitroso compounds with organic phosphines and phosphites in aerated
media is a convenient nonꢀphotolytic procedure to generate aromatic nitroso oxides. The
reaction rate constants and activation parameters of the key (for the proposed method of nitroso
oxide generation) reaction of nitrosobenzene with tripenyl phosphite or paraꢀsubstituted phosꢀ
phines (4ꢀRC6H4)3P (R = MeO, Me, H, F), as well as that of paraꢀmethoxynitrosobenzene
with triphenylphosphine in acetonitrile were determined by kinetic spectrophotometry and
chemiluminescence. A significant transfer of the electron density to the nitroso compound
occurs in the transition state of the reaction as was revealed using the Hammett correlation
analysis and DFT calculations in the M06L/6ꢀ311+G(d,p) approximation. The introduction of
the electronꢀdonor substituent MeO into the paraꢀposition of PhNO decreases the reactivity of
the nitroso compound by two orders of magnitude. The reactivity of triphenyl phosphite in the
oxygen atom transfer reaction is lower by two orders of magnitude compared to that of triꢀ
phenylphosphine. In the case of the reactions of PhNO with phosphines, the apparent rate
constant depends on the oxygen content in the reaction medium.
Key words: nitrosobenzenes, triphenyl phosphite, triphenylphosphines, kinetic spectrophotoꢀ
metry, chemiluminescence, DFT calculations, mechanism, reaction kinetics.
Investigation of chemical properties of active intermeꢀ
diates in chemical reactions is one of the most intensely
studied directions of modern chemistry. Nitroso oxides
RNOO play an important role in the oxidation and photoꢀ
oxidation of a series of nitrogenꢀcontaining organic comꢀ
pounds. The chemical properties and reactivity of these
compounds are being actively studied in our research
group1—11 and all over the world.12—19 The most frequentꢀ
ly used method for nitroso oxide generation is the photoꢀ
oxidation of the corresponding azides that proceeds
through the nitrene intermediates (Scheme 1).
in the visible spectral range4,13,14 and the products of their
transformation can undergo photolytic transformations,
which impedes the interpretation of the experimental reꢀ
sults. The minor product of 6ꢀazidoquinoline photooxidaꢀ
tion is 6ꢀnitroquinoline formed due to the photoisomerꢀ
ization of the corresponding nitroso oxide.20,21 The prodꢀ
uct of thermal consumption of 6ꢀnitroquinoline is dioxꢀ
azole decomposing according to Scheme 2.
Therefore, it is necessary to develop a procedure for
the efficient generation of aromatic nitroso oxides under
the conditions excluding the photolysis of the reaction
system. It is known22 that the thermal deoxygenation of
nitrobenzens by trivalent phosphorus compounds (phosꢀ
phines and phosphites) is a convenient method for generꢀ
ation of the corresponding nitrenes (Scheme 3). If this
reaction occurs under an oxygen atmosphere, it can serve
as a source of nitroso oxides formed by the further oxidaꢀ
tion of nitrenes. The reduction of nitrosobenzene and paraꢀ
methoxynitrosobenzene in acetonitrile at room temperaꢀ
ture was studied23 using triphenylphosphine as an examꢀ
ple. Based on an analysis of the reaction products, the
authors suggested the chain character of the process and
partial regeneration of nitrosobenzene upon the reaction
of transꢀnitroso oxide with triphenylphosphine. However,
the kinetic regularities of reaction (2) were not studied
earlier. As far as we know, there is only one work24 preꢀ
Scheme 1
(1)
In particular, this method generates aromatic nitroso
oxides ArNOO both under the conditions of matrix isolaꢀ
tion13—16 and in solution.13,14,17—19 However, this methꢀ
od has an obvious drawback: under the stationary photoꢀ
lysis conditions, both aromatic nitroso oxides that absorb
* Dedicated to Academician of the Russian Academy of Sciences
M. P. Egorov on the occasion of his 60th birthday.
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 11, pp. 2477—2486, Novemer, 2013.
1066ꢀ5285/13/6211ꢀ2477 © 2013 Springer Science+Business Media, Inc.

2478
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013
Khursan et al.
Scheme 2
to the forbidden transition n  *. Each experiment was carried
out at least three times. Experiments with 4ꢀmethoxynitrosoꢀ
benzene were monitored at the wavelength corresponding to the
absorption band maximum max = 730 nm. This wavelength was
chosen because the used aromatic phosphorus(III) compounds,
intermediates, and reaction products are transparent in this wave
range. The digitized kinetic curves at the reaction depth at least
90% were processed in the framework of the firstꢀorder kinetic
equation, and the effective rate constant of nitrosobenzene conꢀ
sumption (keff) was determined. This value depended linearly on
the initial concentration of the phosphorus(III) compound, and
the angular coefficient of the dependence keff/[R3P]0 was nuꢀ
merically equal to the rate constant of reaction (2). The activaꢀ
tion parameters of reaction (2) were determined in the temperaꢀ
ture range from 281 to 321 K using the above described proceꢀ
dure of monitoring the nitrobenzene concentration under the
conditions of a significant (as a rule, 20ꢀfold) excess of R3P.
Chemiluminescence method. The reactions of nitrosobenzene
with triarylphosphines were carried out in a glass reactor eqꢀ
uipped with a stirrer, a bubblier (experiment was made in air and
under argon), and an injector using acetonitrile as a solvent with
temperature varying from 293 to 323 K. A solution (1.5 mL) of
nitrosobenzene was injected into a solution (13.5 mL) of triꢀ
phenylphosphine with a known concentration and the change
in the CL intensity in time was detected. Under the experꢀ
imental conditions, the time of mixing reactants did not exꢀ
ceed 1 s. A FEUꢀ119 photoelectron multiplier, whose specꢀ
tral sensitivity range varies from 330 to 650 nm, was used as
a detector.
senting the kinetic data on the reaction of 2ꢀnitrosobiꢀ
phenyl with triethyl phosphite in triethyl phosphate taken as
a solvent, where it was found that logA = 6.9 (L mol–1 s–1),
Ea = 48 kJ mol–1. Note that the rate constant of this reacꢀ
tion characterizes the efficiency of nitrene generation (and,
therefore, nitroso oxides) and is the key characteristic of
the proposed nonꢀphotocatalytic method of ArNOO
generation.
Scheme 3
(2)
This work is devoted to studying the kinetics of the
reactions of nitrosobenzene with a series of paraꢀsubstiꢀ
tuted phosphines. The kinetic data for the reactions of
paraꢀmethoxynitrosobenzene with triphenylphosphine and
of nitrosobenzene with triphenyl phosphite are given for
a comparison. The kinetics of the process was studied by
spectrophotometry from the absorbance decay of the
nitroso compound and by chemiluminescence (CL),
which is observed in the reactions of ArNO with aromatic
phosphines.25
Experimental
DFT calculations. Theoretical simulation of the reaction of
nitrosobenzene with R3P was performed using the Gaussianꢀ09,
Revision C1 program package.29 The ChemCraft program30 was
used for the visualization of the obtained results. The DFT methꢀ
ods B3LYP,31,32 CAMꢀB3LYP,33 TPSSTPSS and TPSSh,34
LCꢀwPBE,35 M06,36 and M06L37 in combination with the
6ꢀ311+G(d,p) basis set38,39 were used to optimize the geometric
parameters and to calculate the vibrational spectra. The correꢀ
spondence of the determined structures to minima on the potenꢀ
tial energy surface was established by the absence of negative
elements in the diagonalized Hesse matrix. The influence of the
solvent was taken into account using the polarizable continuum
model IEFPCM.40 Wave functions of transition states were anaꢀ
lyzed by the NBO41 and AIM42 methods. All calculations were
performed at the Supercomputer Cluster of the Center of Colꢀ
lective Use of the Institute of Organic Chemistry (Ufa Scientific
Center, Russian Academy of Sciences).
Acetonitrile (special purity grade (0)) for HPLC (Kriokhrom)
was used without preliminary purification. Nitrobenzene and
4ꢀmethoxynitrosobenzene were synthesized according to known
procedures.26,27 Phosphines (4ꢀRC6H4)3P (R = MeO, Me,
H, F) were recrystallized from ethanol. Triphenyl phosꢀ
phite ((PhO)3P) was synthesized by the reaction of PCl3 with
phenol.28
Method of kinetic spectrophotometry (KSP). The studies were
carried out on a Shimadzu UVꢀ365 spectrometer. A temperaꢀ
tureꢀmaintained cell with an optical path length of 1 cm served
as a reactor. A solution of phosphine (phosphite) in acetoꢀ
nitrile, whose concentration was varied in the range
(0.5—4)•10–3 mol L–1, was preliminarily flushed with argon or
oxygen and nitrobenzene (2•10–4 mol L–1) was rapidly added.
The absorbance decay was detected in time at the longꢀwaveꢀ
length absorption maximum (max = 755 nm), which corresponds

Nitrosobenzene deoxygenation by PIII compounds
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013 2479
Results and Discussion
Equation (II) with a high correlation coefficient is fulꢀ
filled in the whole range of times and temperatures, which
makes it possible to calculate the rate constant of reacꢀ
tion (2a), k2a, we are interested in (Table 1). The data of
Table 1 were processed in the coordinates of the Arrhenius
equation, and the activation parameters of reaction (2a)
were determined (Table 2). The obtained results with alꢀ
lowance for differences in the objects of the study and the
solvent are well consistent with the published data24: the
preꢀexponential factors coincide within the experimental
error and the activation energy of the reaction of 2ꢀnitroꢀ
sobiphenyl with triethyl phosphite in triethyl phosphate is
lower by 12 kJ mol–1 than that found by us. The low actiꢀ
vation energy provides the rate constant determined for
trialkyl phosphite higher by approximately an order of
magnitude, and the triphenyl phosphite used is less active
in the deoxygenated of nitrosobenzenes.
Reaction of nitrosobenzenes with phosphines. To conꢀ
tinue the investigation of the kinetics of ArNO deoxygenꢀ
ation by organic phosphorus(III) compounds, we studied
the kinetic regularities of the reactions of nitrosobenzꢀ
ene (1a) and paraꢀmethoxynitrosobenzene (1b) with a series
of substituted aromatic phosphines 2a—d in acetonitrile.
The primary products of these reactions are paraꢀsubstiꢀ
tuted phenylnitrenes and paraꢀsubstituted aromatic phosꢀ
phine oxides (Scheme 5).
Reaction of nitrosobenzene with triphenyl phosphite.
Kinetic spectrophotometry is a convenient method to study
the kinetics of the reactions of phosphites with substituted
nitrosobenzenes, since the absorption of one of the comꢀ
ponents (ArNO) can be separated from the absorption of
other compounds of the reacting system. For example, it
is known that triphenyl phosphite and triphenyl phosꢀ
phate absorb in the UV spectral range with the band
maxima at 265 nm ( = 1750 L mol–1 cm–1) and 262 nm
( = 810 L mol–1 cm–1), respectively.43 Nitrosobenꢀ
zene exhibits an intense absorption at 250 nm ( =
= 19 000 L mol–1 cm–1 43
and a weak band in the far
)
visible spectral range corresponding to the forbidden tranꢀ
sition n  *. According to our data, the longꢀwavelength
absorption maxima of nitrobenzene and paraꢀmethoxyꢀ
nitrosobenzene in acetonitrile lie at 755 nm ( =
= 43 L mol–1 cm–1) and 730 nm ( = 52 L mol–1 cm–1),
respectively.
It turned out that the reaction of nitrosobenzenes with
phosphites occurs with fairly low rates, which makes this
system to be less promising from the viewpoint of efficient
generation of considerable amounts of nitroso oxide.
Therefore, in this work, we restricted the study by the
interaction of triphenyl phosphite (PhO)3P with nitroꢀ
sobenzene PhNO. In this case, the equation of reaction (2)
takes the form shown in Scheme 4.
Scheme 5
Scheme 4
(2a)
(2b)
1: R1 = H (a), MeO (b); 2: R2 = H (a), MeO(b), Me (c), F (d)
The kinetics of reaction (2a) was studied at the absorption
maximum of the nitroso compound under the conditions
of the pseudoꢀfirst order from its concentration, i.e., under
the conditions that [(PhO)3P]0 = 5•10–2 mol L–1 
 [PhNO]0 = 2•10–3 mol L–1. Acetonitrile was used as a
solvent and the reaction was carried out in aerated soluꢀ
tions in the temperature range from 293 to 333 K. The
further temperature rise is impossible because of a subꢀ
stantial evaporation of the solvent. According to the law of
active masses, the change in the PhNO concentration is
the following:
On the whole, the procedure of studying the kinetics of
reaction (2b) is similar to that described above. Since phosꢀ
phines are more reactive than triphenyl phosphite, the iniꢀ
tial concentration of the nitroso compound was decreased
by approximately an order of magnitude ([PhNO]0 
 2•10–4 mol L–1) to attain reasonable reaction times. As
a result, the absorbance of ArNO did not exceed 0.01 in
the course of the reaction. Nevertheless, computer simulaꢀ
tion of the digitized signal made it possible to obtain rather
reliable and reproducible data. Figure 1 shows the typical
kinetic curves of absorbance decay. It is seen that A deꢀ
creases to zero, i.e., the reaction products do not absorb at
the monitoring wavelength. In fact, according to pubꢀ
lished data,43 phosphine oxide Ph3PO has an absorption
band in the UV range with a maximum at 265 nm ( =
= 1550 L mol–1 cm–1). It was found that keff calculated by
kinetic data processing increased linearly with an increase
in the initial phosphine concentration (Fig. 2). The rate constant
of the reaction of 1a with phosphines 2a—d was determined
from the slope ratio of this dependence (see Table 1).
–d[PhNO]/dt = k2a[PhNO](PhO)3P].
(I)
Taking into account that a 25ꢀfold excess of triphenyl
phosphite is used in the reaction, it can be considered that
its concentration remains almost unchanged in the reacꢀ
tion course and this kinetic equation can be written in the
integral form
[PhNO] = [PhNO]0•exp(–kefft),
where keff = k2a[(PhO)3P]0.
(II)

2480
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013
Khursan et al.
Table 1. Rate constants of the reactions of nitrosobenzene (1a) with organic phosphorus(III) compounds in
acetonitrile
a
R3P
Atmosphere
Т/К
[R3P]0/mol L–1
k2 /L mol–1 s–1
Method
(PhO)3P
Air
Air
Air
Air
Air
293
303
313
323
333
5•10–2
5•10–2
5•10–2
5•10–2
5•10–2
6.2•10–4
2.3•10–3
3.1•10–3
5.5•10–3
1.6•10–2
KSP
KSP
KSP
KSP
KSP
Ph3P
Air
Air
Air
Air
Air
Argon
Argonb
Argon
Argon
Argon
Argon
293
293
303
313
323
293
293
281
291
301
321
(0.5—4)•10–3
1•10–2
3.2
3.1
4.4
6.5
8.8
5.6
7.2
4.6
6.1
11
KSP
CL
CL
CL
CL
KSP
KSP
KSP
KSP
KSP
KSP
1•10–2
1•10–2
1•10–2
(0.5—3)•10–3
(0.5—3)•10–3
1•10–3
1•10–3
1•10–3
1•10–3
20
Ph3Pc
Air
293
(2—8)•10–3
4.4•10–2
KSP
(4ꢀMeOC6H4)3P
Air
Air
Air
Air
Air
293
293
303
313
323
293
(0.5—2)•10–3
5•10–3
19
15
18
21
23
KSP
CL
CL
CL
CL
5•10–3
5•10–3
5•10–3
Argon
(0.5—2)•10–3
30
KSP
(4ꢀMeOC6H4)3Pc
(4ꢀMeC6H4)3P
Air
293
(2—10)•10–3
0.23
KSP
Air
Air
Air
Air
Air
Argon
Argon
Argon
Argon
Argon
293
293
303
313
323
293
281
291
301
311
(1—4)•10–3
5•10–3
4.7
13
17
19
21
22
18
21
25
30
KSP
CL
CL
CL
CL
KSP
KSP
KSP
KSP
KSP
5•10–3
5•10–3
5•10–3
(0.5—1.5)•10–3
1•10–3
1•10–3
1•10–3
1•10–3
(4ꢀMeC6H4)3Pc
(4ꢀFC6H4)3P
Air
293
(2—6)•10–3
0.12
KSP
Air
Air
Air
Air
Air
Argon
Argonb
Argon
Argon
Argon
Argon
Argon
293
293
303
313
323
293
293
281
291
301
311
321
(0.5—4)•10–3
5•10–3
2.8
2.8
5.0
6.6
9.5
4.3
4.7
1.8
4.3
4.9
8.9
11
KSP
CL
CL
CL
CL
KSP
KSP
KSP
KSP
KSP
KSP
KSP
5•10–3
5•10–3
5•10–3
(0.5—2)•10–3
(0.5—2)•10–3
1.5•10–3
1.5•10–3
1.5•10–3
1.5•10–3
1.5•10–3
(4ꢀFC6H4)3Pc
Air
293
(4—8)•10–3
1.4•10–2
KSP
a The inaccuracy of the kinetic curve approximations, as a rule, was 1—2% and did not exceed 10%. In all cases,
the initial concentration of the phosphorus(III) compound exceeded the initial nitrobenzene concentration by at
least an order of magnitude.
b Argon was specially purified from oxygen.
c The reaction with paraꢀmethoxynitrosobenzene (1b).

Nitrosobenzene deoxygenation by PIII compounds
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013 2481
Table 2. Arrhenius parameters of the rate constant according to the experimental data and the DFT calculations of the transition state
for the reactions of nitrobenzene with the organic phosphorus(III) compounds
Reaction
KSP
CL
M06L/6ꢀ311+G(d,p)
logA (logk)
Ea
logA
Ea
H
(gas phase)
Gsolv
H
(MeCN)
–S
/kJ mol–1
/kJ mol–1
/kJ mol–1 K–1
kJ mol–1
PhNO + 2b
PhNO + 2c
PhNO + 2a
PhNO + 2d
4ꢀMeOC6H4NO + 2a (–1.36, 293K)
PhNO + (PhO)3P 7.5 1.2
(1.28, 293K)
3.6 0.1
—
3.1 0.1
3.2 0.2
5.4 0.1
6.0 0.5
5.6 0.2
—
11 1
12 1
28 1
31 3
40 1
—
31.7
33.6
39.4
40.4
55.5
69.5
–19.7
–17.0
–14.7
–14.4
–8.6
12.1
16.6
24.7
26.1
46.8
76.4
188
200
188
196
181
219
13 1
29 3
33 1
—
5.9 0.4
6.5 0.7
60 7
6.9
Notes. A (L mol–1 s–1) is the preꢀexponential factor, Ea is the activation energy, H is the activation enthalpy, Gsolv is the solvation
energy, and S is the activation entropy.
A (arb. units)
0.010
It was found that the rate constant k2b is sensitive to
the oxygen concentration in the reaction system. Similar
kinetic pattern is observed when the reaction is carried out
with continuous argon bubbling: complete nitrosobenzene
consumption in the course of deoxygenation, the pseudoꢀ
firstꢀorder reaction kinetics under the conditions of a phosꢀ
phine excess over 1a, and retention of the overall second
order (the first order with respect to both reactants). The
apparent rate constant k2b determined under an argon atꢀ
mosphere increases by approximately two times (by four
times for 2c) (see Table 1). Moreover, it was assumed that
oxygen present in technical argon can affect the value of
k2b. We specially purified argon from oxygen and found for
phosphines 2a and 2d that in "pure" argon the rate conꢀ
stant k2b increases additionally by 10—20% (see Table 1).
The explanation of the influence of oxygen on the kiꢀ
netics of the reaction of ArNO with phosphines is worth of
more detailed investigation. Now we will advance hypothꢀ
eses that explain the observed regularities. As already menꢀ
tioned above, the partial regeneration of nitrosobenzene
according to Scheme 6 (reaction (5)) was concluded23 on
the basis of the reaction product analysis.
1
2
3
4
0.008
0.006
0.004
0.002
0
200
400
600
800 t/s
Fig. 1. Typical kinetic curves of the absorbance decay in the
reactions of nitrosobenzenes with aromatic phosphines for the
reaction of PPh3 (2a) with nitrosobenzene (1a) at the initial
concentrations of 2a equal to 5•10–4 (1), 1•10–3 (2), 2•10–3 (3),
and 4•10–3 mol L–1 (4). Reaction conditions: initial nitrosoꢀ
benzene concentration [1a]0 = 2•10–4 mol L–1, MeCN, 293 К,
oxygen bubbling.
keff/s–1
1
0.04
0.03
0.02
0.01
Scheme 6
ArNO + R3P
1ArN:
1ArN: + R3PO
(2)
(3)
3ArN:
2
3
3ArN: + O2
 transꢀArNOO +
+ (1 – )cisꢀArNOO
(4)
4
transꢀArNOO + R3P
1.0
2.0
3.0
C•103/mol L–1
ArNO + R3PO
(5)
(6)
Fig. 2. Effective rate constants of the reactions of 1a with phosꢀ
phines 2b (1), 2c (2), 2a (3), and 2d (4) vs concentration of
introduced phosphine (C). Reaction conditions: MeCN, 293 К,
oxygen bubbling.
ArN: + ArNO
ArN=N(O)Ar
 is the content of the transꢀisomer of formed nitroso oxide.

2482
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013
Khursan et al.
The cisꢀisomer of ArNOO undergoes intramolecꢀ
ular isomerization (see Scheme 2, thermal route),
and the mechanism of this reaction was studied earliꢀ
er.20,21,44—46
for kinetic measurements of the reactions between tripheꢀ
nylphosphine derivatives and nitrobenzene in which the
photoreaction is a kinetic indicator for the rateꢀdeterminꢀ
ing step of the process. The reaction rate was monitored
under the conditions of a considerable phosphine excess
over PhNO. The CL intensity decay rigidly corresponded
to an exponential law. All measurements in the temperaꢀ
ture range 293—323 K were carried out at least three times.
The reaction rate constants k2b are given in Table 1. The
results of calculation for the activation parameters of the
reaction are presented in Table 2. The high reproducibility
of the results obtained by two methods (KSP and CL)
confirms correctness of the kinetic analysis for the CL
scheme. In addition, the CL method made it possible
to study the temperature dependence of the reaction
rate constant for highly reactive paraꢀmethoxysubstituted
phosphine.
We do not reject the possibility of this mechanism for
ArNO regeneration and note an alternative possibility for
the observed effect manifestation. Under an oxygen atmoꢀ
sphere nitrenes are efficiently replaced by the correspondꢀ
ing nitroso oxides (see Scheme 6, reaction (4)). However,
the reaction of nitrene with the nitroso compound to form
an azoxy product is known under the conditions of oxygen
deficiency (see Scheme 6, reaction (6)). For the complete
interception of nitrene, the stoichiometry of consumption
of ArNO and R3P should tend to the ratio 2 : 1, which
approximately corresponds to the scale of increasing
the rate constant k2b in argon compared to that in oxygen.
It is most likely that the presence of O2 in the system
decreases the probability of reaction (6) and, as a conseꢀ
quence, the value of rate constant k2b. At the moment it is
difficult to conclude which explanation is more probable.
However, the combined action of both factors cannot be
excluded. From the practical viewpoint, the rate constant
k2b measured in an aerated medium seems interesting,
because nitroso oxides are generated just in these conꢀ
ditions.
The activation parameters of the rate constants of reꢀ
action (2b) were determined for all studied phosphines,
except 2b (see Table 2). As in the reaction involving triphꢀ
enyl phosphite, the observed low values of preꢀexponenꢀ
tial factors indicate substantial losses of the entropy in the
transition state. A strong decrease (by 16 kJ mol–1) in the
activation energy was observed upon the introduction of
the methyl substituent with usually weak electronic effects
into the aromatic ring of phosphine. The introduction of
the fluorine atom into the aromatic ring, on the contrary,
decreases the reactivity and somewhat increases the actiꢀ
vation energy.
Electronic effects in the reaction of nitrobenzenes with
phosphines. The presented results show that a strong influꢀ
ence of substituents on the reactivity is manifested in the
studied reaction. To take an additional information, we
studied the kinetics of reaction (2b) involving paraꢀmethꢀ
oxysubstituted nitrosobenzene 1b. The kinetics of the reꢀ
action of 1b with phosphines 2a—d was studied using the
above described procedure at the absorption maximum of
1b (730 nm). The results of determination of the rate conꢀ
stant k2b are presented in Table 1. The kinetic studies were
carried out under oxygen at room temperature, and the
initial concentration of 1b was, as a rule, 3•10–4 mol L–1
.
A comparison of the experimental results for PhNO and
4ꢀMeOC6H4NO shows that the introduction of the elecꢀ
tronꢀdonor substituent MeO into the aromatic ring of the
nitroso compound sharply decreases the reactivity of
the latter. The scale of the effect of H atom replaceꢀ
ment by the methoxy group in nitrobenzene is two orders
of magnitude. This surprisingly strong influence of the
structure of ArNO on their reactivity in the reaction with
phosphines provides possibilities to control the nitroso
oxide generation rate by choosing ArNO, which is another
advantage of the proposed method for the chemistry of
nitro oxides.
Finally, our results make it possible to evaluate the
influence of the phosphine structure on the reactivity in
the framework of the widely used Hammett correlations.
To estimate the electronic substituent effects, the Hamꢀ
mett constants  (see Ref. 47), which describe the total
effect of substituent including the inductive and resonance
components, were used for the substituents. Obviously,
the steric effect of the studied reaction is insubstantial. In
three reaction series (the reactions of PhNO with phosꢀ
phines under either O2 or Ar and of 4ꢀMeOC6H4NO with
phosphines in the presence of oxygen), the following corꢀ
relation was studied:
Chemiluminescence study of the kinetics of the reaction
of nitrosobenzenes with phosphines. Chemiluminescence
was recently found25 in the deoxygenation of nitroso comꢀ
pounds by triphenylphosphine, and its intensity maximum
is observed in the visible spectral range at 570 nm. The
scheme of the process performed earlier25 is similar to
Scheme 6. The emitter of the CL found in the reaction
with triphenylphosphine is triphenylphosphine imine
formed by the interaction of triphenylphosphine and triꢀ
plet nitrene (Scheme 7).
Scheme 7
3PhN: + Ph3P
3PhN=PPh3
PhN=PPh3 + h570
The kinetic analysis of the process indicates that the
CL decay rate is limited by the rate of interaction for the
reactants.25 This makes it possible to use the CL method
logk2b = logk02b + ,
(III)

Nitrosobenzene deoxygenation by PIII compounds
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013 2483
where  is the constant of the reaction series reflecting the
scale of the substituent effect on the rate constant, and  is
the substituent constant in the Hammett scale.
entropy are explained by the formation of the transition
state with charge separation and a loss of degrees of freedom.
It is assumed that an intermediate of the ArN–—O—P+R3
type is formed in the transition state and then decomposes
to the reaction products. It is likely that this assumption
can be extended to the reaction of nitrosobenzenes with
phosphines.
In all cases, the angular coefficient is negative; i.e., in
all reaction series, reaction (2b) is accelerated by the elecꢀ
tronꢀdonor substituents in a phosphine molecule. This reꢀ
sult and the above mentioned decrease in the reactivity of
4ꢀMeOC6H4NO compared to that of PhNO indicate that
the nitroso compounds manifest the electrophilic properꢀ
ties in the studied reaction. The angular coefficient of
the Hammett dependence has the following values:  =
= –2.7 0.3 (correlation coefficient r = 0.99) for the reacꢀ
tion of nitrosobenzene with argon bubbling into the reaction
system,  = –2.3 0.8 (r = 0.90) in the reaction of nitroꢀ
sobenzene in the presence of oxygen, and  = –3.4 0.6
(r = 0.97) for the reaction of paraꢀmethoxynitrosobenzꢀ
ene under an O2 atmosphere.
In connection with the above, the DFT theoretical
simulation of the studied reactions was performed. A seꢀ
ries of different functionals (M06L, TPSSTPSS, M06,
TPSSh, B3LYP, CAMꢀB3LYP, LCꢀwPBE) was tested in
combination with the Pople splitꢀvalence triple polarized
basis set 6ꢀ311+G(d,p). The equilibrium structures of all
reactants were calculated and the transition states (TS) of all
studied transformations were localized using the indicated
approximations. The TS of the reaction of 1a with 2d
and triphenyl phosphite determined by the M06L/6ꢀ
311+G(d,p) method are exemplified in Fig. 3. A strong
effect of the density functional nature on the calculated
value of the activation barrier was revealed in the course of
simulation. For example, for reaction 1a + 2a (ignoring
solvation effects), the enthalpy of activation (H) inꢀ
creases from 39.4 (M06L) to 112.7 (LCꢀwPBE) kJ mol–1
in the order of functionals presented above. It can be asꢀ
sumed that the multiconfigurational character of the wave
function for the TS is a reason for the observed nonꢀreproꢀ
ducibility of the values of H, since singlet nitrene is one
of the reaction products. The static component of electron
correlation plays a substantial role in these systems and is
ignored in the framework of oneꢀdeterminant description
of the wave function. The observed scatter of H values
shows the ability of different density functionals to deꢀ
scribe similar states more or less adequately.48
Calculation of the reaction mechanism by the DFT method.
The activation parameters of the reaction of nitrosoꢀ
benzene with triarylphosphines are linearly interrelated
(see Table 2); i.e., the studied reaction exhibits the comꢀ
pensation effect
logA = (1.56 0.07) + (0.142 0.003)Ea,
r = 0.9995.
The compensation dependence of the activation enerꢀ
gy on the preꢀexponential factor indicates the general reꢀ
action mechanism for all phosphines. In addition, the low
value of logA (the "normal" value for the bimolecular reacꢀ
tion is 10—11) indicates a substantial entropy loss in the
transition state of the studied reaction. The authors24 sugꢀ
gest for the related reaction of 2ꢀnitrosobiphenyl with triꢀ
ethyl phosphite that the low value of logA and, correꢀ
spondingly, the high negative value of the activation of
In spite of considerable differences in estimation of the
activation barrier for the reaction of nitrosobenzenes with
a
b
1.368
1.748
1.332
1.946
F
P
N
O
C
F
F
Fig. 3. Transition states of the reactions of nitrosobenzene (1a) with (4ꢀFC6H4)3P (2d) (a) and triphenyl phosphite (b) localized in the
M06L/6ꢀ311+G(d,p) approximation. Interatomic distances are given in Å.

2484
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013
Khursan et al.
Table 3. Chemical potentials of the reagents and properties of the transition state of the reactions of nitrosobenzene with organic
phosphorus(III) compounds
Reaction
/au
Properties of transition state of reaction
Properties of BCP P···O
PhNO
R3P

r(NO)
r(P···O)
QPCM
/au

L
| / |
1 3
imag
/cm–1
Å
au
PhNO + 2b
PhNO + 2c
PhNO + 2a
PhNO + 2d
4ꢀMeOC6H4NO + 2a
PhNO + (PhO)3P
0.162
0.162
0.162
0.162
0.149
0.162
0.112
0.116
0.125
0.136
0.125
0.131
272i
277i
294i
287i
292i
292i
1.329
1.328
1.331
1.332
1.341
1.368
1.979
1.970
1.952
1.946
1.906
1.748
0.383
0.362
0.330
0.323
0.216
0.327
0.104
0.105
0.108
0.110
0.116
0.149
0.014
0.009
0.009
–0.009
–0.046
–0.030
0.503
0.514
0.514
0.549
0.636
0.615
Notes. Accepted designations:  is the chemical potential of reagents;  = –(HOMO + LUMO)/2, where  are the energies of the
HOMO and LUMO, respectively; BCP P···O is the bond critical point according to the results of AIM analysis of the electron
density; 
is the only one imaginary frequency in the calculated vibrational spectrum verifying the transition state; r are
imag
intermolecular spacings; QPCM is the excessive electron density on nitrobenzene in the transition state (Mulliken analysis with
allowance for the polarization of the TS by the solvent (MeCN) in the framework of the IEFPCM model);  is the electron density;
2
L is the electron density Laplacian, L =  ;  and  are eigenvalues of the electron density Hessian.
1
3
R3P, all functionals used correctly reproduce the relative
reactivity in a series of phosphines 2a—d and a substantial
increase in the activation energy of the reaction 1b + 2a
and in the reaction of PhNO with (PhO)3P which correꢀ
sponds to the experimental data (see Table 2). Solvation
that was taken into account in the framework of the
IEFPCM polarizable continuum is an important stabiliꢀ
zation factor. The results of DFT calculations obtained by
the most adequate for the system functional (M06L) are
given in Tables 2 and 3.
An analysis of the geometric structure and the electron
distribution in the TS of the studied reactions showed,
first, their high correspondence to the experimental data
and, second, complete consistency of the hypothesis proꢀ
posed24 for the reaction mechanism.
results of NBO and AIM analyses of the wave function,
according to which 0.7—0.8 au of the negative charge is
localized on the nitroso compound.
(3) As mentioned above, the incipient P···O bond is
the most remarkable geometric characteristic of the studꢀ
ied reaction. The AIM analysis of the TS wave function
made it possible to characterize the properties of the bond
critical point (BCP) for this pair of atoms (see Table 3).
Note a comparatively low electron density  in the BCP
and the minimum value of  in a series of studied TS is
observed for the reaction of PhNO with 2b, i.e., for the
reaction that proceeds with the lowest activation barrier
and, vice versa, the maximum value of  is observed for the
TS of the reaction PhNO + (PhO)3P.
(4) According to the published work,49 the type of
chemical bond can be characterized by such BCP paramꢀ
eters as the ratio of minimum and maximum eigenvalues
of the Hessian  taken by modulus |1/3| and the electron
density Laplacian L[]. The data in Table 3 show that
|1/3| < 1 and L is low in all cases. This interactions of
atoms is named "closedꢀshell interaction" and is characꢀ
teristic of ionic and strongly polar covalent bonds, since
the withdrawal of electron density from the BCP towards
nuclei predominates, and its measure is 3.49 For the same
reason, the electron density in the BCP is fairly low
(see above).
We revealed a series of the regularities.
(1) The change in the P···O interatomic distance in the
TS and the value of activation barrier are antibate: the
maximum length of the formed bond (1.979 Å) is observed
in the TS of the reaction of PhNO with the most reactive
phosphine 2b, whereas in the TS of the reaction of nitroꢀ
sobenzene with triphenyl phosphite r(P···O) is much shortꢀ
er (1.748 Å).
(2) An excessive electron density is accumulated on
the nitroso compound in the transition state, and the deꢀ
gree of electron density transfer correlates with the elecꢀ
tronic properties for the paraꢀsubstituent in a series of
phosphines 2a—d. The Mulliken analysis (with allowance
of the polarization of the TS by the solvent) gives a change
in the excessive electron density from 0.323 au for 2d to
0.383 au for 2c (see Table 3). In this series, the tendency is
more important that the absolute value because reliability
of the Mulliken analysis is often low especially when difꢀ
fuse functions in the basis set are used.48 A significant
charge transfer from R3P to ArNO is confirmed by the
Thus, the results of DFT simulation of the reaction of
ArNO with R3P are well consistent with the kinetic data,
explain the differences observed in the reactivity of the
reactants, and theoretically confirm Cadogan´s hypoꢀ
thesis22,24 that the reaction proceeds through the TS with
a high degree of charge separation on the reaction center.
This results in substantial losses of the activation entropy
(see Table 2). In addition, the high polarity of the TS
assumes its efficient solvation. Perhaps, peculiarities of

Nitrosobenzene deoxygenation by PIII compounds
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013 2485
13. E. A. Pritchina, N. P. Gritsan, J. Photochem. Photobiol. A,
1988, 43, 165.
14. N. P. Gritsan, E. A. Pritchina, J. Inform. Rec. Mater., 1989,
17, 391.
15. S. Ishikawa, S. Tsuji, Y. Sawaki, J. Am. Chem. Soc., 1991,
113, 4282.
16. S. Ishikawa, T. Nojima, Y. Sawaki, J. Chem. Soc., Perkin
Trans. 2, 1996, 127.
17. T. Harder, P. Wessig, J. Bendig, R. Stösser, J. Am. Chem.
Soc., 1999, 121, 6580.
18. H. Inui, M. Irisawa, S. Oishi, Chem. Lett., 2005, 34, 478.
19. E. A. Pritchina, N. P. Gritsan, T. Bally, Phys. Chem. Chem.
Phys., 2006, 8, 719.
20. E. M. Chainikova, R. L. Safiullin, A. N. Teregulova, L. V.
Spirikhin, E. G. Galkin, Dokl. Chem. (Engl. Transl.), 2012,
442, 30 [Dokl. Akad. Nauk, 2012, 442, 493].
21. E. Chainikova, R. Safiullin, L. Spirikhin, A. Erastov, Tetraꢀ
hedron Lett., 2013, 54, 2140.
specific solvation of the TS, which were ignored in the
framework of our theoretical model, result in the appearꢀ
ance of a compensation effect in a series of paraꢀsubstitutꢀ
ed phosphates.
Thus, the kinetics of the reactions of nitrosobenzenes
with trivalent phosphorus compounds was studied by two
independent methods: kinetic spectrophotometry and
chemiluminescence. The obtained data create a theoretiꢀ
cal basis for a substantiated choice of conditions for the
dark (i.e., nonꢀphotolytic) generation of aromatic nitroso
oxides. The revealed strong dependence of the rate conꢀ
stant on the structures of the reactants makes it possible to
generate ArNOO in a wide range of rates, which can be
useful for choosing optimum experimental conditions to
study the chemical properties of nitroso oxides.
This work was financially supported by the Division of
Chemistry and Materials Science of the Russian Academy
of Sciences (Program "Intermediates of Chemical Reacꢀ
tions, Their Direct Detection and Stabilization, and Deꢀ
termination of Structural Parameters") and the Russian
Foundation for Basic Research (Project No. 13ꢀ03ꢀ00201).
22. J. I. G. Cadogan, Q. Rev., Chem. Soc., 1968, 22, 222.
23. E. M. Chainikova, R. L. Safiullin, Russ. Chem. Bull. (Int. Ed.),
2009, 58, 926 [Izv. Akad. Nauk, Ser. Khim., 2009, 906].
24. J. I. G. Cadogan, A. Cooper, J. Chem. Soc. B, 1969, 883.
25. V. A. Shamukaev, A. N. Teregulova, S. S. Ostakhov, R. L.
Safiullin, High Energy Chem. (Engl. Transl.), 2013, 47 [Khim.
Vys. Energ., 2013, 47, 66].
26. G. H. Coleman, Ch. M. McCloskey, F. A. Stuart, Org. Synth.,
1955, Coll. Vol. 3, 668.
References
27. I. Guben, Metody organicheskoi khimii [Methods of Organic
Chemistry], Goskhimizdat, Moscow—Leningrad, 1949
(in Russian).
1. E. M. Chainikova, S. L. Khursan, R. L. Safiullin, Dokl. Phys.
Chem. (Engl. Transl.), 2003, 390, 163 [Dokl. Akad. Nauk,
2003, 390, 796].
28. V. M. Plets, Organicheskie soedineniya fosfora [Organic Phosꢀ
phorus Compounds], Oborongiz, Moscow, 1940 (in Russian).
29. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fuꢀ
kuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J. Montgomery, Jr., J. E.
Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E.
Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Moroꢀ
kuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Danꢀ
nenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresꢀ
man, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussianꢀ09,
Rev. C1, Gaussian, Inc., Wallingford (CT), 2009.
30. G. A. Zhurko, ChemCraft, Version 1.6 (build 332); http://
www.chemcraftprog.com.
31. A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
32. C. Lee, W. Yang, R. G. Parr, Phys. Rev. B, 1988, 37, 785.
33. T. Yanai, D. P. Tew, N. C. Handy, Chem. Phys. Lett., 2004,
393, 51.
34. J. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys.
Rev. Lett., 2003, 91, 146401.
35. O. A. Vydrov, G. E. Scuseria, J. Chem. Phys., 2006,
125, 234109.
36. Y. Zhao, D. Truhlar, Theor. Chem. Acc., 2008, 120, 215.
2. R. L. Safiullin, S. L. Khursan, E. M. Chainikova, V. T. Daꢀ
nilov, Kinet. Catal. (Engl. Transl.), 2004, 45, 640 [Kinet.
Katal., 2004, 45, 680].
3. E. M. Chainikova, S. L. Khursan, R. L. Safiullin, Kinet.
Catal. (Engl. Transl.), 2004, 45, 794 [Kinet. Katal., 2004,
45, 842].
4. E. M. Chainikova, S. L. Hursan, R. L. Safiullin, [Kinet.
Catal. (Engl. Transl.), 2006, 47, 549 [Kinet. Katal., 2006,
47, 566].
5. M. R. Talipov, A. B. Ryzhkov, S. L. Khursan, R. L. Safiulꢀ
lin, J. Struct. Chem. (Engl. Transl.), 2006, 47, 1051 [Zh. Strukt.
Khim., 2006, 47, 1062].
6. E. M. Chainikova, R. L. Safiullin, Kinet. Catal.
(Engl. Transl.), 2009, 50, 527 [Kinet. Katal., 2009, 50, 551].
7. V. S. Khursan, E. M. Chainikova, R. L. Safiullin, High Enerꢀ
gy Chem. (Engl. Transl.), 2009, 43, 467 [Khim. Vys. Energ.,
2009, 43, 523].
8. E. M. Chainikova, R. L. Safiullin, Russ. Chem. Bull.
(Int. Ed.), 2009, 58, 2437 [Izv. Akad. Nauk, Ser. Khim.,
2009, 2359].
9. M. R. Talipov, S. L. Khursan, R. L. Safiullin, J. Phys. Chem.
A, 2009, 113, 6468.
10. V. S. Khursan, O. A. Kovaleva, E. M. Chainikova, M. R.
Talipov, R. L. Safiullin, High Energy Chem. (Engl. Transl.),
2010, 44, 284 [Khim. Vys. Energ., 2010, 44, 313].
11. M. R. Talipov, S. L. Khursan, R. L. Safiullin, [Russ. J. Phys.
Chem. A (Engl. Transl.), 2011, 85, 364 [Zh. Fiz. Khim., 2011,
85, 427].
12. Y. Sawaki, S. Ishikawa, H. Iwamura, J. Am. Chem. Soc.,
1987, 109, 584.

2486
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 11, November, 2013
Khursan et al.
37. Y. Zhao, D. G. Truhlar, J. Chem. Phys., 2006, 125, 194101.
38. A. D. McLean, G. S. Chandler, J. Chem. Phys., 1980,
72, 5639.
45. E. M. Chainikova, R. L. Safiullin, L. V. Spirikhin, Dokl.
Chem. (Engl. Transl.), 2012, 442, 12 [Dokl. Akad. Nauk, 2012,
442, 200].
39. K. Raghavachari, J. S. Binkley, R. Seeger, J. A. Pople,
J. Chem. Phys., 1980, 72, 650.
40. J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev., 2005,
105, 2999.
41. A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev., 1988,
88, 899.
42. R. F. W. Bader, Atoms in Molecules: A Quantum Theory,
Oxford Univ. Press, Oxford, 1994, 458 pp.
43. V. Talrose, A. N. Yermakov, A. A. Usov, A. A. Goncharova,
A. N. Leskin, N. A. Messineva, N. V. Trusova, M. V. Efimkiꢀ
na, in NIST Chemistry WebBook, NIST Standard Reference
Database Number 69, Eds P. J. Linstrom, W. G. Mallard,
National Institute of Standards and Technology, Gaithersꢀ
burg (MD), 2012, p. 20899.
46. E. M. Chainikova, R. L. Safiullin, L. V. Spirikhin, M. F.
Abdullin, J. Phys. Chem. A, 2012, 116, 8142.
47. A. J. Gordon, R. A. Ford, The Chemist´s Companion: A Handꢀ
book of Practical Data, Techniques, and References, Wiley,
New York, 1972, 537 pp.
48. F. Jensen, Introduction to Computational Chemistry, John
Wiley and Sons, Chichester, 2007.
49. I. S. Bushmarinov, K. A. Lyssenko, M. Yu. Antipin, Russ.
Chem. Rev., 2009, 78, 283.
44. M. R. Talipov, S. L. Khursan, R. L. Safiullin, Russ. J. Phys.
Chem. A (Engl. Transl.), 2012, 86, 235 [Zh. Fiz. Khim., 2012,
86, 292].
Received May 27, 2013