Photocyclization of a mixed phosphonium-iodonium ylide with phenylacetylene: A formal kinetic approach to the mechanistic study

Article abstract of DOI:10.1134/S0023158415040138

The recently discovered photoheterocyclization of a mixture of mixed phosphonium-iodonium ylides and acetylene compounds to give λ⁵-phosphinoline, a new phosphorus-containing heterocycle, occurs with self-acceleration and an induction period at relatively high concentrations of the reactants. The reaction was found to be catalyzed by the acid formed during photolysis. We developed a spectrophotometric method to study the kinetics of photocyclization between phenylacetylene and a phosphonium-iodonium ylide in the presence of trifluoroacetic acid and studied how the yield of target λ⁵-phosphinoline and kinetic parameters of the reaction, such as the induction period and maximum rate of product formation, depend on the initial concentrations of the starting compounds and catalyst. It was shown that the addition of the acid has no effect on the critical character of the dependence of the reaction on the concentration of the starting compound: the reaction proceeds only at an ylide concentration of >0.02 mol/L in a microheterogeneous suspension and the self-accelerating character of λ⁵-phosphinoline formation is preserved. The optimum reagent ratios maximizing the yield of the target product were determined. The acid was found to play a dual role: it catalyzes the reaction only at concentrations not higher than the initial ylide concentration. At higher acid concentrations, no formation of λ⁵-phosphinoline was observed.

Full text of DOI:10.1134/S0023158415040138

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2015, Vol. 56, No. 4, pp. 403–411. © Pleiades Publishing, Ltd., 2015.
Original Russian Text © T.D. Nekipelova, M.A. Taranova, E.D. Matveeva, V.A. Kuz’min, N.S. Zefirov, 2015, published in Kinetika i Kataliz, 2015, Vol. 56, No. 4, pp. 411–420.
Photocyclization of a Mixed Phosphonium–Iodonium Ylide
with Phenylacetylene: A Formal Kinetic Approach
to the Mechanistic Study
a, b,
b
b, c
a, b
b, c
T. D. Nekipelova *, M. A. Taranova , E. D. Matveeva , V. A. Kuz’min , and N. S. Zefirov
a
Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, Moscow, 119334 Russia
b
Faculty of Chemistry, Moscow State University, Moscow, 119991 Russia
c
Institute of Physiologically Active Substances, Russian Academy of Sciences, Chernogolovka, Moscow oblast, 142432 Russia
*
eꢀmail: nekip@sky.chph.ras.ru
Received September 3, 2014
Abstract—The recently discovered photoheterocyclization of a mixture of mixed phosphonium–iodonium
5
ylides and acetylene compounds to give
λ
ꢀphosphinoline, a new phosphorusꢀcontaining heterocycle, occurs
with selfꢀacceleration and an induction period at relatively high concentrations of the reactants. The reaction
was found to be catalyzed by the acid formed during photolysis. We developed a spectrophotometric method
to study the kinetics of photocyclization between phenylacetylene and a phosphonium–iodonium ylide in the
5
presence of trifluoroacetic acid and studied how the yield of target
λ
ꢀphosphinoline and kinetic parameters
of the reaction, such as the induction period and maximum rate of product formation, depend on the initial
concentrations of the starting compounds and catalyst. It was shown that the addition of the acid has no effect
on the critical character of the dependence of the reaction on the concentration of the starting compound:
the reaction proceeds only at an ylide concentration of >0.02 mol/L in a microheterogeneous suspension and
5
the selfꢀaccelerating character of
λ
ꢀphosphinoline formation is preserved. The optimum reagent ratios maxꢀ
imizing the yield of the target product were determined. The acid was found to play a dual role: it catalyzes
the reaction only at concentrations not higher than the initial ylide concentration. At higher acid concentraꢀ
5
tions, no formation of
λ
ꢀphosphinoline was observed.
Keywords: phosphonium–iodonium ylides,
λ
5ꢀphosphinoline, photoheterocyclization, phenylacetylene,
acid catalysis
DOI: 10.1134/S0023158415040138
The synthesis and structural and reactivity studies covered, which yield pseudocycloaddition or heterocyꢀ
of mixed phosphonium–iodonium ylides allowed clization products, viz., phosphorusꢀcontaining hetꢀ
design of new reactants for organic synthesis based on erocyclic compounds [7–11]. The composition of the
these ylides. This class of compounds with several products depends on the nature of the triple bond. The
reaction sites opens up new routes to difficultꢀtoꢀsynꢀ photolysis of phosphonium–iodonium ylides in the
thesize and new heterocyclic compounds [1–3]. In presence of nitriles RCN affords phosphonium–subꢀ
thermal reactions, mixed phosphonium–iodonium stituted oxazoles (key precursors in the synthesis of
ylides exhibit the properties of alkenyliodonium salts bioactive compounds) in high yields, PhI being a leavꢀ
and phosphorus ylides. Mixed ylides (tetrafluoroboꢀ ing group [7, 11]. The photolysis of a mixture of phosꢀ
rates and tosylates) do not enter into the Wittig reacꢀ phonium–iodonium ylides (
1
) and acetylenes (
ꢀphosphinolines ( ), phosꢀ
very good nucleofuge, viz., the phenyliodonium phoniumꢀsubstituted furans ( ), PhI, and phosphoꢀ
group, allowed design of processes in which these nium salts ( ) (Scheme 1, the BF4 counter ion is omitꢀ
ylides not only are substrates in nucleophilic substituꢀ ted) [9, 10, 12]. The relative yields of products
2)
5
tion with different aldehydes [4]. The presence of a yields several products:
λ
3
4
–
5
3–5
tion but also act as Oꢀnucleophiles [1–3].
depend significantly on the acetylene structure [10].
An analysis of Scheme 1 [12] demonstrated the folꢀ
lowing features of the reaction: firstly, reaction prodꢀ
The phenyliodonium group as a part of mixed
ylides determines their ability to photodecomposition
on exposure to light to form reactive ionic and radical
ion intermediates, as it was observed for diaryliodoꢀ
nium salts [5, 6]. In recent years, two new photochemꢀ
ical reactions of mixed phosphonium–iodonium ylides
ucts
3–5 contain no iodine, which is released as pheꢀ
5
nyl iodide; secondly,
λ
ꢀphosphinoline is a neutral
compound whose formation from the ylide (cation)
and acetylene is accompanied by proton elimination,
−
with compounds containing a triple bond have been disꢀ i.e., the formation of acid with counterion (BF );
4
403

404
NEKIPELOVA et al.
COR
Ph
+
+
Ph3P
−
(
Ph)3P
Ο
P
+
R1 C CH
h
ν
, CH
–
2
Cl
2
Ph
+ (
β
)
+ (
γ
)(Ph) P CH CR
3
2
C C
+
+
+
PhI
αH
PhI
(
α
)
R1
R1
–
R
R
O
O
1
2
3
4
5
α
+
β
+
γ
= 1
Scheme 1.
1
thirdly, the formation of phosphonium salt
5
requires
2
(R = Ph) yielding
3
occurs only at relatively high
two additional hydrogen atoms (СН2 group of the concentrations of
salt). While the formation of furan derivative can be concentrations, ylide
assumed to be analogous to the formation of oxazoles CH Cl2 and produces a weakly opalescent suspension.
1
(
≥
0.02 mol/L in CH Cl2). At these
1 is not soluble completely in
2
4
2
upon the photolysis of 1 in solutions of nitriles [7, 11],
The solution is light yellow, whereas the starting ylide
ꢀphosphinoline formation, a in the crystalline state is a white powder. The reaction
quite unusual heterocycle, has not been clarified comꢀ is initiated by light with a wavelength of 350–450 nm,
pletely. To date, a great number of substituted a range in which ylide
does not noticeably absorb
ꢀphosphinolines have been synthesized from ylides light at concentrations of 10 mol/L, which are comꢀ
with different substituents at the double bond and hetꢀ monly used in spectrophotometric measurements.
eroaryl substituents at the phosphorus atom using However, when the concentration of
is higher or a
acetylenes with different structures [13–15]. In some cell with an optical path length of 5 cm is used, the
1
5
λ
the mechanism of
1
5
–4
λ
1
5
cases, the corresponding
photolysis product (for example, for phenylacetylene region up to 460 nm. The development of the reaction
the yield of is 60% and the yield of
is less than 7% is unusual. An induction period is observed during
9]). Note that the formation of
occurs in a oneꢀpot metalꢀfree system at room temꢀ spectrum at the wavelengths at which
perature within less than 30 min, which is not quite line absorbs (Fig. 1). Thereafter, the reaction develꢀ
ops spontaneously, starting locally either on heterogeꢀ
neous particles of ylide or on the cell wall or at the
λ
ꢀphosphinoline is the main “tail” of UV absorption is observed in the visible
3
4
λ
5
[
ꢀphosphinoline which there are no changes in the visible absorption
5
λ
ꢀphosphinoꢀ
3
common for organic synthesis reactions.
1
As was noted earlier [12], the formation of
ꢀphosphinoline has some specific features that are
solution surface, and this is manifested as the appearꢀ
ance of a brightꢀbrown spot. Within 1–2 min, the
reaction propagates throughout the bulk of the soluꢀ
tion and stops. The final solution is brown and transꢀ
5
λ
not typical of homogeneous organic reactions. It was
established that the reaction between
1
(R = Ph) and
parent. The visible absorption spectrum of the colored
5
product corresponds to that of
λ
ꢀphosphinoline 3.
А
The development of the reaction suggests that it
occurs via a autocatalytic or chain mechanism, and the
principal question arising here concerns the nature of
the products ensuring autocatalysis or chain propagaꢀ
tion and branching. The aboveꢀnoted acid formation
that follows from the reaction stoichiometry (Scheme 1)
is confirmed by a comparison between the pH values
of aqueous extracts of the starting and final solutions,
which indicates a noticeable increase in the acidity of
the reaction mixture during the reaction (pH 6.0 and
pH 2.0, respectively). It was assumed [12] that the acid
released initiates the photodecomposition of the ylide
and, upon photoinitiation in the close proximity of a
heterogeneous particle, where the local concentrations
of the reagents are higher, due to the photodecomposiꢀ
tion of the ylide the acid concentration reaches the critꢀ
ical value necessary for acid catalysis before it reaches
the same value in the solution bulk. This initiates a
spontaneous development of the acidꢀcatalyzed process
and its propagation throughout the bulk of the mixture.
Thus, light acts as an initiator triggering the selfꢀcataꢀ
lyzed process. Preliminary experiments have shown
1
1
0
.5
.0
.5
6
1
1
6
6
1
0
2
25
275
325
375
425 475
Wavelength, nm
5
Fig. 1. Absorption spectrum of
λ
ꢀphosphinoline as a funcꢀ
4
tion of the TFA concentration, 10 mol/L: (
1
) 0, (
) 12.0. The solvent is
= 0.4 cm. The dashed line indicates the registraꢀ
tion wavelength in the kinetic experiments.
2) 0.55,
(3
) 1.13, (
2
4) 2.4, (5) 6.0, and (6
CH Cl ,
l
2
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

PHOTOCYCLIZATION OF A MIXED PHOSPHONIUM–IODONIUM YLIDE
405
that the addition of strong organic acids (trifluoroacetic length of 0.4 cm and was placed into the cell compartꢀ
acid, ꢀtoluenesulfonic acid, as well as HBF4) shortens ment of the spectrophotometer, an appropriate
p
the induction time and accelerates the reaction [12]. amount of a TFA solution in CH Cl2 (0.2 mL of TFA
2
The addition of weak acetic acid has no effect on the in 2 mL of CH Cl2, [TFA] = 1.35 mol/L) was added
2
0
course of the reaction.
using a micropipette, and monitoring the formation
kinetics of the colored product (475 nm) was begun.
The concentration of the acid was varied in the 0.009–
Thus, the process under study is an example of
reaction that can be regarded as a photoinitiated reacꢀ
tion catalyzed by one of the final products, viz., an
acid. The kinetics of this reaction under steadyꢀstate
photolysis conditions is difficult to study, since the
reaction proceeds at high concentrations of the startꢀ
ing compound, at which a direct spectrophotometric
study is impossible and, therefore, the samples must be
diluted 50ꢀ to 100ꢀfold for making a spectrophotometꢀ
ric analysis. One should take into account that slight
stirring of the mixture disturbs the course of the proꢀ
cess by changing the local concentrations of the reacꢀ
tants. Finally, after the end of the induction period, the
reaction develops very rapidly (within 2–3 min) and,
under these conditions, it is difficult to take a suffiꢀ
0
.675 mol/L range. A comparison between the formaꢀ
tion kinetics of compound in the presence of atmoꢀ
3
spheric oxygen and after purging the mixture with an
inert gas indicated no effect of oxygen on the reaction,
and, for this reason, the reaction was performed in air.
We studied how the induction period of compound
3
formation and the maximum rate of the process after
the end of the induction period depend on the initial
concentrations of ylide, phenylacetylene, and added
acid. The change in absorbance per unit time (dA/dt)
at the measurement wavelength (475 nm) was taken to
be the reaction rate.
The solution of ylide in CH Cl2 was observed to
2
cient number of samples for kinetic measurements. undergo changes when it was left to stand even for a
The possibility of the photochemical reaction turning short time (30 min). In particular, it was shown that
into an acidꢀcatalyzed one allowed us to study the forꢀ the ylide particles undergo aggregation, which affects
5
mal kinetics of
λ
ꢀphosphinoline formation in the the induction time and the process rate, but not the
reaction between ylide 1 and phenylacetylene 2 and to yield of the final product. For this reason, a new ylide
determine some kinetic parameters of the reaction and sample was used in each experiment, and it was disꢀ
the critical concentrations of the reactants. This can solved in CH Cl2 immediately before the experiment.
2
provide a key to determining the optimum conditions Due to the slight quantitative discrepancy between the
5
for the preparation of
λ
ꢀphosphinoline in the course results obtained using different batches of ylide, which
of the complex photocatalytic process.
could contain trace amounts of an acid, the quantitaꢀ
tive parameters given in the table for different reaction
conditions refer to certain experimental series, each
performed for 1 day with the same batches of ylide and
EXPERIMENTAL
Ylide
procedure described in [1, 8, 9].
The reaction between ylide
lacetylene in the presence of trifluoroacetic acid
1
(R = Ph) was synthesized according to the solvent. It should be noted that, regardless of the ylide
batch used, all reaction regularities are qualitatively
reproduced.
1
(R = Ph) and phenyꢀ
2
When the reaction was complete, the final solution
(
TFA), which was found to be most convenient from was diluted 50ꢀ to 100ꢀfold with CH Cl2 and the
2
the experimental standpoint, was studied in the cell absorption spectrum was recorded on a UVꢀ3101 PC
compartment of a UVꢀ3101 PC spectrophotometer spectrophotometer, from which the final concentraꢀ
(
Shimadzu, Japan) by monitoring the formation of the tion of compound
3
and its yield were calculated. As
5
colored reaction product,
λ
ꢀphosphinoline (
3
), at follows from Fig. 1, compound
3
has two protonated
4
75 nm in the kinetic mode. As was noted above, in forms (the absorption maxima at 439, 418, and 391 nm
steadyꢀstate photolysis, the reaction proceeds only at correspond to the neutral and protonated forms,
sufficiently high concentrations of the starting comꢀ respectively) with similar absorption coefficients
4
–1
–1
pounds (>0.02 mol/L); for this reason, the formation
(
~1.25
2
×
10 L mol cm ). Upon dilution with
of the final colored product was monitored on the CH Cl2, the spectrum of the product corresponded to
longꢀwavelength absorption wing of compound
3
the protonated form with the absorption maximum at
(
2
2
RESULTS
than, the concentration of
was poured into a quartz cell with an optical path an acid occurs at a ylide
1
. The resulting solution pound
3
upon steadyꢀstate photolysis in the absence of
concentration above
1
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

4
06
NEKIPELOVA et al.
5ꢀphosphinoline (
3) yield in relation to the initial concentrations of ylide 1, phenyꢀ
Induction time, maximum rate, and
lacetylene , and trifluoroacetic acid
λ
2
[1
] ,
[
2
] ,
[TFA]0,
0
0
Series
tind, s
wmax*, s–1
Yield of 3, %
mol/L
0
.025
0.075
0.0375
0.075
0.15
0
>10000
>2400
388
–
0
0.04
0.15
0.24
0
0
0.0125
0.023
0.023
0.023
0.023
1
0.025
32 ± 3
38 ± 4
37 ± 4
0
0
.050
.10
200
113
0.30
0
0
0
0
0
0
.075
.075
.075
.10
.11
.12
0.225
0.225
0.225
0.30
0.33
0.36
0.023
0.046
0.07
0.023
0.023
0.023
180
127
125
150
133
120
0.07
0.11
0.13
0.11
0.13
0.14
–
2**
0
0
0
0
0
0
.024
.048
.072
.096
.144
.192
730
730
411
285
207
196
0.013
0.014
0.04
0.09
0.09
0.09
17 ± 2
19 ± 2
32 ± 3
32 ± 3
32 ± 3
32 ± 3
3
0.025
0.025
0.023
0
0
0
0
0
0
0
0
.009
.023
.045
.09
.18
.45
970
490
750
0.01
32 ± 3
32 ± 3
2 ± 0.5
0.026
0.0014
0
4
>10
>10
0
0
0
0
0
4
0.12
4
0
20
20
0
0.0008
0.0014
0.006
.54
.675
*
The reaction rate is expressed in terms of a change in absorbance per second.
*
* In series 2, the yield of was not determined.
3
0
.02 mol/L [12]. Since in the present study the reacꢀ
Thus, the lower concentration limit of ylide
1
at
5
λ
tion was carried out under different conditions (withꢀ which the formation of
ꢀphosphinoline occurs is the
same for both the steadyꢀstate photolysis conditions
and the reaction occurring without external irradiaꢀ
tion in the spectrophotometer cell compartment in the
presence of the acid. The plot of the maximum rate of
the process as a function of the initial concentration of
out using an external source of radiation in the specꢀ
trophotometer cell compartment in the presence of
acid), we performed series of experiments at different
initial concentrations of ylide
1
in the presence of a
2 at an acid concenꢀ
3ꢀfold excess of phenylacetylene
ylide
1 is nearly linear and intersects the x axis at [1]0 =
tration of 0.023 mol/L (table, series 1). The experiꢀ
ment carried out in the absence of an acid showed that
no formation of colored products occurs upon
0
.016–0.018 mol/L, the point corresponding to the
lower limit of reactive ylide concentration. Note that,
although a colored product is not produced at ylide
3ꢀhꢀlong incubation of a solution of the (1 + 2) mixꢀ
concentrations of
ture in the spectrophotometer cell compartment. In that ylide undergoes no conversion under these condiꢀ
the presence of the acid, the reaction pattern changes tions, since when ylide
is exposed to light and acid,
dramatically (Fig. 2a). At ]0
0.025 mol/L, the forꢀ it turns slowly into salt and other compounds having
mation of
is observed after an induction period and, no visible absorption. It is important that in this case
there are no traces of the photoheterocyclization
1 < 0.02 mol/L, this does not mean
1
5
[1
≥
3
with an increasing
[1]0, the induction time decreases
product (table, series 1, run 2).
3
and the reaction rate after the end of the induction
period increases (table, series 1 and 2). The yield of the
target compound at an ylide concentration of 0.05
A comparison between the induction periods for
acidꢀcontaining solutions showed that these periods
mol/L almost reaches the maximum value of ~40% of are significantly shorter when the mixture is placed in
the ylide added to the reaction (table).
the spectrophotometer, as well as when it is exposed to
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

PHOTOCYCLIZATION OF A MIXED PHOSPHONIUM–IODONIUM YLIDE
a) (b)
407
(
A
A
A
0
0
0
0
.5
.4
.3
.2
0
0
0
0
0
.5
0
0
0
.6
.4
.3
.2
.1
0
6
2
1
0.1
.4
.2
0
0.2 0.4 0.6
[TFA], mol/L
1
5
4
4
3
2
3
0
200
400
600
800
1000
300
400
500
600 700
Wavelength, nm
t
, s
Fig. 2. (a) Kinetics of the formation of the compounds produced in the reaction between
presence of TFA at concentrations of ( ) 0.009, ( ) 0.023, ( ) 0.045, ( ) 0.45, ( ) 0.54, and (
CH Cl . (b). Absorption spectra of the final reaction products at different TFA concentrations, mol/L: (
1
(0.025 mol/L) and
2
(0.12 mol/L) in the
1
2
3
4
5
6
) 0.675 mool/L.
λ
= 475 nm,
req
1) 0.009, (
2) 0.045, (3) 0.09,
2
2
and (
added acid.
4) 0.675. 1 : 50 dilution with CH Cl . Inset: absorbance of the final product at 400 nm as a function of the concentration of
2
2
scattered light, than when the solution is stored in the starting ylide
dark. Thus, the lowꢀintensity light of the probing lamp the concentration of
1
, the reaction yield does not depend on
2. This conclusion was also conꢀ
with a wavelength of 475 nm initiates reaction in the firmed at other initial concentrations of the ylide. The
presence of the acid but does not initiate the reaction induction period of the reaction decreases with an
in the absence of an acid. The appearance of small increasing concentration of 2, and it tends to a limit at
fluctuations in absorbance shortly before the end and sixfold excess of the ylide. The maximum reaction rate
simultaneously after of the induction period (Fig. 2a) after the induction period does not depend on the conꢀ
suggests that the reaction in the spectrophotometer centration of
2 at its fourfold excess. Note that at all
does not start simultaneously in the entire bulk (simiꢀ ylide and phenylacetylene concentrations examined,
larly to the photolysis under external irradiation). This an induction periods and a subsequent selfꢀaccelerꢀ
is confirmed by the behavior of the solution that was ated progress of the reaction were observed. Thus, the
left in the light: the reaction commences at the surface optimum initial concentration of phenylacetylene is
of the solution
its 3ꢀ or 4ꢀfold excess over the ylide concentration.
The yield of phosphinoline
3 in the solution incuꢀ
bated in the dark was by 15% less than that in the soluꢀ
tion incubated in the spectrophotometer. The specꢀ
trum of the product displayed slight absorption at λ ≥
Kinetic Parameters of the Reaction as a Function
of the Concentration of Trifluoroacetic Acid
5
00 nm, where phosphinoline
yield of in the case of reaction in the presence of TFA
~40%) was lower than that under the conditions of
3 does not absorb. The
Figure 2a shows the kinetic curves as changes in
absorbance at 475 nm in the presence of TFA at differꢀ
ent concentrations (table, series 4). Correlating the
kinetic data with the analysis of the reaction products
3
(
steadyꢀstate photolysis of a neutral solution of the
ylide (at most 60%). The causes of this effect will be
discussed below.
(
Fig. 2b) shows that a highꢀyield formation of
only at [TFA]0 ]0. As the TFA concentration is furꢀ
as a funcꢀ ther increased, the yield of the target product
tion of phenylacetylene concentration were studied at decreases dramatically and the induction time
constant concentrations of ylide (0.024 mol/L) and increases (Fig. 2a, curve 3). At 2[ ]0 < [TFA]0 < 15[ ]0
TFA (0.023 mol/L) and different concentrations of colored products does not form at all. At acid concenꢀ
varying from 0.024 to 0.192 mol/L (table, series 3). trations of TFA]0 > 2[ ]0, the solution becomes
The results obtained show that, at the concentration of homogeneous and colorless; i.e., the ylide dissolves
that is 3 times higher than the concentration of the completely in CH Cl2. A further increase in the TFA
3 occurs
≤
[1
Kinetic parameters of the formation of
3
1
1
1 ,
2
[
1
2
2
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

408
NEKIPELOVA et al.
concentration ([TFA]0
≥
15[
1
]0) results in the formaꢀ This result suggests that the formation of 3 involves
tion of colored products virtually without an induction only the starting form of ylide
period (Fig. 2a, curves ). These products absorb in transformation in the acidic medium.
the visible region up to 700 nm with absorbance maxꢀ
In the experiments at an ylide concentration of
ima at 600, 480, and 400 nm (Fig. 2b, curve ). Probꢀ 0.075 mol/L, we studied in detail how the induction
ably, they are oligomeric products of reactions of pheꢀ time and the reaction rate depend on the TFA concenꢀ
nylacetylene with other products of ylide decomposiꢀ tration at [TFA]0
]0 (table, series 2). In this case,
tion in the acidic medium. The kinetic curves for the the induction time decreases with an increasing acid
formation of these products have a pronounced break, concentration in the [TFA]0 < [
]0 range; however, at
indicating two steps of the process at high acid conꢀ [TFA] of 0.046–0.07 mol/L the induction time
1, but not products of its
4–6
4
≤
[1
1
0
centrations.
remains virtually unchanged and the reaction rate
The observed dependence of the reaction pathway increases after the end of the induction time.
on the acid concentration was unexpected, and, for
this reason, the behavior of ylide 1 in the presence of
DISCUSSION
different concentrations of the acid was studied in
greater detail. The experiments were performed at an
Photochemical reactions in the ylides under study
–4
ylide concentration of
1 10 mol/L. Up to a twofold are caused by the presence of the phenyliodonium
×
excess of the acid, the absorption spectrum of the ylide fragment. Diaryliodonium salts are widely used as
changed only slightly. At higher concentrations of the polymerization photoinitiators, and the mechanism of
acid, the spectrum of the ylide displays significant their photodecomposition was comprehensively studꢀ
changes: a new peak with
λ
max
= 269 nm appears. At a ied [5, 6]. On being exposed to light, these compounds
2
0ꢀfold excess of the acid, this peak appears almost undergo decomposition both via a heterolytic mechaꢀ
immediately after mixing the reagents. At lower conꢀ nism as a result of photoinduced electron transfer to
centrations of TFA, the formation of the product corꢀ form PhI and ionic intermediates and via a homolytic
responding to this peak is complete in 15 min. After mechanism to form radicals and radical ions. Extenꢀ
the extraction of the acid with water, the spectrum of sion of the decomposition mechanism of diaryliodoꢀ
this product remained unchanged; i.e., the product nium salts to the photodecomposition of the mixed
does not result from the acid–base equilibrium and phosphonium–iodonium ylide can be represented as
the reaction of the ylide with excess acid is irreversible. follows:
+
+
−
Ο
−
−
Ο
(
Ph)3P
(
Ph)3P +
Ο
(Ph)3P
+C C
+
PhI
(I)
+
C C
C C
••
−
Ο
(
Ph)3P
Ph
Ph
Ph
C C
h
ν
+
PhI
Ph
+
−
Ο
−
(
Ph)3P
•C C
(
Ph)3P
Ο
•
+
+ PhI
(II)
•C C
Ph
Ph
Scheme 2.
The heterolytic (I) and homolytic (II) photodeꢀ proceeds via two parallel pathways limits the yields of
composition of ylide
results in the formation of a target products. In contrast to pseudocycloaddition,
1
complex mixture of reactive ionic, radical ion, radical, which occurs in a homogeneous solution even at low
–
4
and, possibly, carbene intermediates that can react concentrations of ylides (10 mol/L), heterocyꢀ
with the solvent (CH Cl2) and with the added reagent, clization is observed only at ylide concentrations
2
for example, phenylacetylene.
above 0.02 mol/L in a microsuspension in CH Cl2,
with the acid released in the reaction playing an
2
The study of the product composition and mechꢀ
anism of photolysis of mixed ylides with different
acetylenes and nitriles showed that the pseudocyꢀ
cloaddition reactions occur with a high probability
due to the heterolytic decomposition of ylide in the
reaction of the added compound containing a triple
bond with an ionic intermediate, and the phosphoꢀ
nium salt and other transformation products of ylide
important role. Note that the reaction of the solvent
•+
with the radical ion PhI
homolytic decomposition of ylide
formed upon the
(reaction (II))
1
also results in the formation of an acid according to
reaction (III) [5, 6]:
PhI•+ + CH Cl2
→
PhI + H + C HCl . (III)
2
+
•
2
Therefore, one can assume that it is via reaction
result from transformations of radical intermediates (III) that the acid forms initially upon the steadyꢀstate
11, 12]. The fact that photodecomposition of ylides photolysis of the (
) mixture as a result of the
[
1 + 2
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

PHOTOCYCLIZATION OF A MIXED PHOSPHONIUM–IODONIUM YLIDE
409
homolytic decomposition of the ylide and this acid
triggers heterocyclization. To verify this assumption,
we performed two experiments. In the first experiꢀ
ment, in the steadyꢀstate photolysis of the mixture of
(a)
A
2
.5
1
(0.05 mol/L) and
period was 4 min (Fig. 3a). In the second experiment,
ylide (0.05 mol/L) was photolyzed for 4 min and
compound (0.15 mol/L) was then added. The forꢀ
mation of started only 4 min after the addition of
Fig. 3b); i.e., the acid formed in the steadyꢀstate phoꢀ
tolysis of upon its homolytic decomposition is not
2 (0.15 mol/L) the induction
2.0
1
2
3
2
(
1.5
1
6
1
accumulated and seems to be consumed in the formaꢀ
tion of phosphonium salt, as was suggested in [11, 12].
Thus, it is the mixture of the reagents that creates a
medium necessary for the initiation of heterocyclizaꢀ
1
.0
.5
6
tion. This suggests that complexation between
1 and 2
0
in the heterogeneous system ( ) favors the photoꢀ
1
+ 2
1
decomposition of the ylide via the heterolytic mechaꢀ
nism. Such a complex was actually observed in the
reactant mixture by spectrophotometry. The activaꢀ
tion of the triple bonds of arylalkynes by iodonium
salts forming in situ was assumed, for example, in the
organic synthesis of spiro compounds [16]. Apparꢀ
ently, such activation can occur also upon the comꢀ
plexation of a mixed ylide with acetylenes. In this
regard, our result on the photoinitiation of the reacꢀ
tion in the presence of TFA by light with a wavelength
of 475 nm, at which the ylide absorbance is negligible,
is likely explained by the specific photophysical and
photochemical properties of the nanoparticles formed
by the ylide in its complex with phenylacetylene.
5
0
2
50
300
350
400
450 500
Wavelength, nm
(b)
A
2
2
1
1
.5
.0
.5
.0
The question now arises of what kinds of chemical
processes occur during the induction period.
6
1
Spectra
ing products form during the induction period of the
) mixture. Upon photolysis of the individual
ylide, the spectral changes at the same point in time
are significantly less (Fig. 3b, spectra ). Unfortuꢀ
1–3 in Fig. 3a clearly show that UVꢀabsorbꢀ
(
1
+
2
6
1, 2
1
nately, the mass spectrometry of the samples taken
during the induction period detected no noticeable
formation of any unknown products with an ionic
mass above 100 Da. The samples taken before the end
of the induction period contain, besides ions of the
starting compounds, small amounts of ions with
0.5
5
0
250
300
350
400
450 500
Wavelength, nm
masses corresponding to the final products, viz.,
ꢀphosphinoline and the phosphonium salt. These
5
Fig. 3. (a) Evolution of the absorption spectrum in the
steadyꢀstate photolysis of a mixture of (0.05 mol/L) and
(0.15 mol/L); photolysis time, min: ( ) 0, ( ) 1, ( ) 2,
) 3, ( ) 4, and ( ) 7. (b) Evolution of the absorption specꢀ
trum in the steadyꢀstate photolysis of ylide (0.05 mol/L);
photolysis time, min: ( ) 0 and ( ) 4; after the addition of
(0.15 mol/L) for the total photolysis time, min: ( ) 4,
) 7, ( ) 8, and ( ) 10 ( ). = 400 nm, 1 : 200 dilution
with CH Cl .
λ
1
data suggest that, although the formation of
UVꢀabsorbing intermediates upon the photolysis of
the mixture was observed during the induction
period, these intermediates, on the one hand,
decompose during the mass spectrometric analysis
and do not manifest themselves and, on the other
hand, result in the formation of the target product
and acid.
2
1
2
3
(
4
5
6
1
1
2
2
3
(
4
5
6
6
λ
exc
2
2
The complicated dependences of the product yield tion of an acid, the concentration of liberated Н+ canꢀ
and kinetic parameters of the reaction on the acid not be higher than the initial ylide concentration;
concentration are due to the reactions of ylides with therefore, in this case the acid concentration remains
excess acid. It should be noted here that, under the at the optimum level throughout the process. This fact
steadyꢀstate photolysis conditions without the addiꢀ is responsible for the phosphinoline
3 yield in the
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

410
NEKIPELOVA et al.
decomposition of the starting phosphonium–iodoꢀ
nium ylide and upon the formation of [5, 6, 12]. The
reaction under study proceeds in the phenylacetyꢀ
lene–ylide complex in excess phenylacetylene, and,
therefore, in the first approximation it can be regarded
as an autocatalytic firstꢀorder reaction catalyzed by
the final product rather than the target product.
(
a)
3
A
4
3
3
To fit experimental data, we used the equation
describing an autocatalytic firstꢀorder reaction [17].
The value measured here is the absorbance due to
2
1
0
2
product 3. All remaining components of the mixture,
including the initiating admixture of TFA, show no
noticeable absorbance in the visible region, so the
equation for the autocatalytic reaction was used in the
following form:
1
+
+
[
H ] [1] (exp(k([1] + [H ] )t) − 1)
0 0 0 0
+ +
A =
γ,
(1)
[
1] + [H ] exp(k([1] + [H ] )t)
0 0 0 0
50
100
150
200
t
, s
where
at the registration wavelength (475 nm),
are the initial concentrations of ylide and catalyzing
admixture, respectively, is the rate constant of the
secondꢀorder reaction (L mol s ),
is the factor converting the concentration of
the measured absorbance, which is determined by the
molar absorption coefficient of at the registration
A
is the current absorbance of phosphinoline 3
+
[1]0 and [H ]0
(
b)
k
A
–1 –1
t
is time (s), and
into
2
2
1
1
0
.5
γ
3
3
.0
.5
.0
.5
0
wavelength (475 nm), optical path length (0.4 cm),
and an additional factor accounting for the fact that
the target product yield is significantly lower than
100% (in L/mol). To estimate the rate constant, the
experimental curves obtained in series 2 were fitted to
Eq. (1).
2
3
1
Assuming that [H+] = [TFA] = 0.023 mol/L, it is
impossible to describe the formation kinetics of
3 at
varied concentrations of ylide (series 2), since a
1
high concentration of the catalyzing admixture
results in the absence of an induction period. Howꢀ
ever, the calculated curves adequately reflect the forꢀ
mation regularities of
3 under variations of both the
80
120
160
200
t
, s
ylide concentration at [TFA] = 0.023 mol/L (Fig. 4a)
and the TFA concentration from 0.023 to 0.07 mol/L
and at a constant ylide concentration (Fig. 4b) at a
Fig. 4. Experimental kinetic curves for the change in
absorbance at 475 nm (points) for ( + TFA) mixꢀ
tures and their fits to Eq. (1) (solid curves): (a) at different
values of [ ) 0.075, ( ) 0.10, ( ) 0.11, and
) 0.12, [TFA] = 0.023 mol/L; (b) at different values of
+
1
+
2
concentration of catalyzing additive [
H
] of about
–
12
–13
10 –10 mol/L. At a constant acid concentration
1
]0, mol/L: (
1
2
3
+
–13
of [H ] = 5.4
×
10 mol/L, regardless of the initial
(
[
4
0
ylide concentration, we obtained close values of the
TFA] , mol/L: (
1
) 0.023, (
2
) 0.046, and ( ) 0.07, [
]0 = 3[1] .
0
3
1
]0 =
0
–1 –1
rate constant
35 3) L/mol. By fitting the kinetic curves at differꢀ
ent TFA concentrations and constant concentration
steadyꢀstate photolysis being higher than the yield of [ ] = 0.075 mol/L (Fig. 4b), the following values
observed upon the addition of the catalyzing acid, were obtained: [H ] = 5.4
k
= (1.57
±
0.15) L mol
s
and
γ
=
0
.075 mol/L. In all experiments, [
2
(
±
1
+
–13
–13
×
10 , 10.8
×
10
,
–
13
since in the latter case the acid concentration during and 16.2
×
10
mol/L and k = 1.68, 2.35, and
2
0
.45 L mol s–1 at [TFA] = 0.023, 0.046, and
–1
the reaction much earlier begins to exceed the current
ylide concentration and the acid converts the ylide
into its inactive form.
.07 mol/L, respectively. The tenꢀorders difference
between the concentration of the added acid and the
+
calculated concentration of catalyzing additive
is a complex hetꢀ caused by several factors and reflects, first of all, the
erogeneousꢀhomogeneous reaction induced by light low degree of dissociation of the acid in the nonpolar
H is
The formation of phosphinoline
3
+
and catalyzed by
Н that forms both upon the photoꢀ organic solvent. Note that the estimate of the paramꢀ
KINETICS AND CATALYSIS Vol. 56
No. 4 2015

PHOTOCYCLIZATION OF A MIXED PHOSPHONIUM–IODONIUM YLIDE
411
eter does not contradict the experimental values of by the Department of Chemistry and Materials Sciꢀ
γ
molar absorption coeffiicient and reaction yield.
The runs of the calculated curves coincide incomꢀ
pletely with the experimental dependences, which is
undoubtedly due to the simplified character of the
model applied. In the experiment at all concentrations
of the ylide and acid, a steeper beginning and end of the
reaction were observed. The former is due to the heterꢀ
ogeneous character of the reaction, and the latter is due
to the accumulation of the acid, which transforms the
ylide into its inactive form. It is interesting that at high
concentrations of the ylide (>0.1 mol/L, Fig. 4a,
ence of the Russian Academy of Sciences.
REFERENCES
1
2
3
4
. Matveeva, E.D., Podrugina, T.A., Grishin, Yu.K.,
Tkachev, V.V., Zhdankin, V.V., Aldoshin, S.M., and
Zefirov, N.S., Russ. J. Org. Chem., 2003, vol. 39, p. 536.
. Matveeva, E.D., Podrugina, T.A., Grishin, Yu.K., Pavꢀ
lova, A.S., and Zefirov, N.S., Russ. J. Org. Chem., 2007,
vol. 43, p. 201.
. Matveeva, E.D., Podrugina, T.A., Pavlova, A.S.,
curves 3, 4), the initial reaction rates after the end of the
Mironov, A V., and Zefirov, N.S., Russ. Chem. Bull.
008, vol. 57, p. 400.
,
induction period virtually do not depend on the ylide
concentration. This is likely caused by the large size of
the ylide particles at these concentrations, which result
in the absence of a dependence of the reaction rate on
the ylide concentration at the initial stage. In the develꢀ
oped reaction at a higher concentration of the ylide, the
maximum reaction rate is higher in both the experiment
2
. Zhdankin, V.V., Maydanovych, O., Herschbach, J.,
Bruno, J., Matveeva, E.D., and Zefirov, N.S., J. Org.
Chem., 2003, vol. 68, p. 1018.
5
6
7
. Dektar, J.L. and Hacker, N.P., J. Org. Chem., 1990,
vol. 55, p. 639.
. Narewska, J., Strzelczyk, R., and Podsiadly, R.,
J. Photochem. Photobiol., A, 2010, vol. 212, p. 68.
(table) and calculations.
The results obtained emphasize once again the
complex mechanism of the reaction, which cannot be
described completely by the simplified model. The
. Matveeva, E.D., Podrugina, T.A., Pavlova, A.S.,
Mironov, A.V., and Zefirov, N.S., Russ. Chem. Bull.
008, vol. 57, p. 2237.
,
2
dependence of the rate constant
tration possibly means that, during the formation of
k on the acid concenꢀ
8
. Matveeva, E.D., Podrugina, T.A., Pavlova, A.S.,
Mironov, A.V., Gleiter, R., and Zefirov, N.S., Eur. J.
Org. Chem., 2009, no. 14, p. 2323.
3,
the acid is not only involved in the initiation of the
reaction, but it also carries out the acid catalysis in
transformations of intermediates in the formation of
the target product. One cannot exclude also that the
increase in the acid concentration results in the formaꢀ
tion of one more species catalyzing the reaction.
In conclusion, it should be emphasized that this
experimental spectrophotometric and formal kinetic
9. Matveeva, E.D., Podrugina, T.A., Pavlova, A.S.,
Mironov, A.V., Borisenko, A.A., Gleiter, R., and
Zefirov, N.S., J. Org. Chem., 2009, vol. 74, p. 9428.
10. Matveeva, E.D., Podrugina, T.A., Taranova, M.A.,
Borisenko, A.A., Mironov, A.V., Gleiter, R., and
Zefirov, N.S., J. Org. Chem., 2011, vol. 76, p. 566.
studies of the complex photochemical synthesis of 11. Nekipelova, T.D., Kuzmin, V.A., Matveeva, E.D.,
5
λ
ꢀphosphinoline under conditions of pure photoꢀ
Gleiter, R., and Zefirov, N.S., J. Phys. Org. Chem.
013, vol. 26, p. 137.
,
2
chemical initiation changing to initiation in the presꢀ
ence of acid allowed determination of the optimum 12. Nekipelova, T.D., Taranova, M.A., Matveeva, E.D.,
ratios of the reagents maximizing the yield of the tarꢀ
get product. The kinetic parameters of the process
were found to depend on the concentration of the
added acid in a complex manner. This study raised a
set of questions that cannot be answered within a forꢀ
malized kinetic approach and require studying the
mechanism of photoinitiation in the presence and in
the absence of an acid, the role of heterogeneous
state of the starting ylide, and the effect of particle
size on the course of the reaction. A detailed spectral
kinetic analysis of reactive intermediate species (ions
and radicals) using pulse methods and EPR spectrosꢀ
copy would allow determination of their roles in the
formation of final products.
Podrugina, T.A., Kuzmin, V.A., and Zefirov, N.S.,
Dokl. Chem., 2012, vol. 447, part 1, p. 262.
1
1
1
3. Matveeva, E.D., Podrugina, T.A., Taranova, M.A.,
Ivanova, A.M., Gleiter, R., and Zefirov, N.S., J. Org.
Chem., 2012, vol. 77, p. 5770.
4. Matveeva, E.D., Podrugina, T.A., Taranova, M.A.,
Vinogradov, D.S., Gleiter, R., and Zefirov, N.S., J. Org.
Chem., 2013, vol. 78, p. 11691.
5. Matveeva, E.D., Podrugina, T.A., Taranova, M.A.,
Melikhova, E.Yu., Gleiter, R., and Zefirov, N.S.,
Tetrahedron, 2013, vol. 69, p. 7395.
1
6. Dohi, T., Nakae, T., Ishikado, Y., Kato, D., and Kita, Y.,
Org. Biomol. Chem., 2011, vol. 9, p. 6899.
1
7. Emanuel’, N.M. and Knorre, D.G., Kurs khimicheskoi
kinetiki (Chemical Kinetics), Moscow: Vysshaya
Shkola, 1969.
ACKNOWLEDGMENTS
This work was supported by the Russian Foundaꢀ
tion for Basic Research (grant no. 14ꢀ03ꢀ00795) and
Translated by K. Utegenov
KINETICS AND CATALYSIS Vol. 56
No. 4 2015