Photocatalytic degradation of dichloromethane by chlorocuprate(II) ions

Article abstract of DOI:10.1021/ic800684v

Near UV irradiation of aerated solutions of (Et₄N) ₂[CuCl₄] in dichloromethane causes the decomposition of CH₂Cl₂, as evidenced by the buildup of HCl, C ₂H₂Cl₄, and peroxides. A net reduction to [CuCl₂]^- occurs in the early stages, but is later reversed. In CH₂Cl₂, [CuCl₄]^2- is in equilibrium with [Cu₂Cl₆]^2-, and only the latter species is photoactive. The decomposition is initiated by the photodissociation of chlorine atoms, which propagate to peroxy radicals, CHCl₂OO. Experimental evidence, including a linear dependence of the decomposition rate on the incident light intensity and on the fraction of light absorbed by [Cu₂Cl₆]^2-, is consistent with a mechanism in which CHCl₂OO is reduced by electron transfer from [CuCl₂]^-, following which protonation yields CHCl ₂OOH. The hydroperoxide accumulates during irradiation and it too can reoxidize [CuCl₂]^-. The quantum yield for HCl production at the outset of irradiation at 313 nm is 1.3 mol/einstein, based on the fraction of light absorbed by [Cu₂Cl₆]^2-.

Full text of DOI:10.1021/ic800684v

Inorg. Chem. 2008, 47, 7029-7034
Photocatalytic Degradation of Dichloromethane by Chlorocuprate(II) Ions
Kevin J. Doyle, Huy Tran, Miguel Baldoni-Olivencia, Merve Karabulut, and Patrick E. Hoggard*
Department of Chemistry, Santa Clara UniVersity, Santa Clara, California 95053
Received April 16, 2008
Near UV irradiation of aerated solutions of (Et4N)2[CuCl4] in dichloromethane causes the decomposition of CH2Cl2,
as evidenced by the buildup of HCl, C2H2Cl4, and peroxides. A net reduction to [CuCl2]- occurs in the early stages,
but is later reversed. In CH2Cl2, [CuCl4]2- is in equilibrium with [Cu2Cl6]2-, and only the latter species is photoactive.
The decomposition is initiated by the photodissociation of chlorine atoms, which propagate to peroxy radicals,
CHCl2OO. Experimental evidence, including a linear dependence of the decomposition rate on the incident light
intensity and on the fraction of light absorbed by [Cu2Cl6]2-, is consistent with a mechanism in which CHCl2OO is
reduced by electron transfer from [CuCl2]-, following which protonation yields CHCl2OOH. The hydroperoxide
accumulates during irradiation and it too can reoxidize [CuCl2]-. The quantum yield for HCl production at the
outset of irradiation at 313 nm is 1.3 mol/einstein, based on the fraction of light absorbed by [Cu2Cl6]2-.
I. Introduction
The individual steps of the decomposition process in the
gas phase have been investigated by several groups,4,13–16
and may be represented as follows.
Halomethanes pose environmental concerns that include
chronic toxicity, mutagenicity, and carcinogenicity.1 Dichlo-
romethane, due to its low absorption cross-section for solar
radiation, also has the potential to deliver chlorine to the
stratosphere and has a half-life in the troposphere estimated
to be several months.1 The main source for dichloromethane
and other halomethanes in drinking water supplies is
municipal disinfection.2 A heterogeneous catalyst that could
promote decomposition of CH2Cl2 and other halomethanes
by solar radiation could prove very beneficial, and the present
study was initiated in an attempt to understand how such a
catalyst might function.
CH2Cl2
9
hν8 Cl + CH2Cl
(1)
(2)
(3)
(4)
k2
Cl + CH2Cl2
98 HCl + CHCl2
k3
CH2Cl + CH2Cl2
CHCl2 + O2
98 CH3Cl + CHCl2
k4
98 CHCl2OO
(5) Borisch, J.; Pilkenton, S.; Miller, M. L.; Raftery, D.; Francisco, J. S.
J. Phys. Chem. B 2004, 108, 5640–5646.
Under aerobic conditions, thermal or photochemical ho-
molysis of a C-Cl bond in dichloromethane causes decom-
position to HCl and CO.3,4 Wavelengths below approxi-
mately 225 nm are required to have significant absorption
of light by CH2Cl2. In the presence of TiO2 and other
heterogeneous catalysts, decomposition of dichloromethane
to HCl and CO2 can be accomplished with UV irradiation
above 300 nm.5–12
(6) Calza, P.; Minero, C.; Pelizzetti, E. J. Chem. Soc., Faraday Trans.
1997, 93, 3765–3771.
(7) Gutierrez-Ortiz, J. I.; Lopez-Fonseca, R.; Aurrekoetxea, U.; Gonzalez-
Velasco, J. R. J. Catal. 2003, 218, 148–154.
(8) Halmann, M.; Hunt, A. J.; Spath, D. Sol. Energy Mater. Sol. Cells
1992, 26, 1–16.
(9) Hsiao, C. Y.; Lee, C. L.; Ollis, D. F. J. Catal. 1983, 82, 418–423.
(10) Pinard, L.; Magnoux, P.; Ayrault, P.; Guisnet, M. J. Catal. 2004, 221,
662–665.
(11) Sabin, F.; Tuerk, T.; Vogler, A. J. Photochem. Photobiol., A 1992,
63, 99–106.
* To whom correspondence should be addressed. E-mail: phoggard@
scu.edu.
(12) Tanguay, J. F.; Suib, S. L.; Coughlin, R. W. J. Catal. 1989, 117, 335–
347.
(1) U.S. Environmental Protection Agency. Ambient water quality criteria
for halomethanes; EPA 440 5-80-051; 1980.
(13) Catoire, V.; Lesclaux, R.; Schneider, W. F.; Wallington, T. J. J. Phys.
Chem. 1996, 100, 14356–14371.
(2) U.S. Environmental Protection Agency. Preliminary assessment of
suspected carcinogens in drinking water [Report to Congress]; EPA
560475003, 1975.
(14) Hisatsune, I. C.; Heicklen, J. Can. J. Spectrosc. 1973, 18, 77–81.
(15) Rayez, J. C.; Rayez, M. T.; Halvick, P.; Duguay, B.; Lesclaux, R.;
Dannenberg, J. J. Chem. Phys. 1987, 116, 203–213.
(16) Sheps, L.; Crowther, A. C.; Elles, C. G.; Crim, F. F. J. Phys. Chem.
A 2005, 109, 4296–4302.
(3) Hoare, M. R.; Norrish, R. G. W.; Whittingham, G. Proc. R. Soc.
(London) 1959, A250, 197–211.
(4) Sanhueza, E.; Heicklen, J. J. Phys. Chem. 1975, 79, 7–11.
10.1021/ic800684v CCC: $40.75  2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 15, 2008 7029
Published on Web 07/01/2008

Doyle et al.
chlorocuprate(II) complexes.23,32–37 If the chlorocuprate(I)
product from the photolysis of CuCl42- in a halocarbon could
be reoxidized to copper(II), as occurs with iron(II/III), the
net result of an oxidation-reduction cycle would be the
degradation of the halocarbon. We undertook an investigation
of this process in dichloromethane, with the assumption that
absorption into the ligand-to-metal charge transfer (LMCT)
bands of CuCl42-, which extend from the UV into the blue
part of the spectrum,38,39 would cause chlorine atom dis-
sociation, initiating the decomposition process (eq 9).
k5
2CHCl2OO
CHCl2O
CHClO
98 2CHCl2O + O2
(5)
(6)
(7)
k6
98 CHClO + Cl
k7
9
8 HCl + CO
Light-induced dissociation is followed by hydrogen ab-
straction,16 leaving dichloromethyl radicals, which add O2
to form dichloromethylperoxy radicals.13 These decompose
bimolecularly to form dichloromethoxy radicals,13 which
expel chlorine atoms,15 leaving HClO. Formyl chloride
decomposes to CO and HCl with a half-life of 10 min at
room temperature.14
hν
[CuCl4]2- 98 [CuCl3]2- + Cl
(9)
To complete this cycle, copper(I) might be reoxidized by
the peroxy radical or, if it is formed at all in dichloromethane,
the hydroperoxide.
Another potential reaction that must be considered is
hydrogen abstraction by the peroxy radical.
k8
II. Experimental Section
CHCl2OO + CH2Cl2
98 CHCl2OOH + CHCl2
(8)
Tetraethylammonium tetrachlorocuprate(II), 1,1,2,2-tetrachloro-
ethane, hexachloroethane, and HPLC-grade dichloromethane, sta-
bilized with ∼0.1% amylenes, were obtained from Aldrich. For
most experiments, the stabilizer was removed by shaking the
dichloromethane with 1 M aqueous sulfuric acid, then a sodium
carbonate solution, and finally several times with water. Measure-
ments of HCl production could not be done accurately with
stabilized CH2Cl2 because it added to the alkene on a time scale of
minutes.40,41
Photolyses were carried out by pipetting 3.0 mL of a solution
into a fused silica cuvette and irradiating it with either a 350-W
mercury lamp (Oriel) with a Schott WG-320 or WG-395 cutoff
filter, or a 500-W Hg/Xe lamp (Oriel) passed through a 25 cm
monochromator. Light intensities were measured with an Oriel
70260 radiant power meter. UV-visible spectra were monitored
with a Cary 50 spectrophotometer or an HP 8453 diode array
spectrometer. GC-mass spectrometry was carried out with a
Shimadzu QP-5000 with a Restek Corp. XTI-5 column. The oven
start temperature was 40 °C and a linear temperature gradient of
20° min-1 was applied to 260 °C. The instrument was operated in
splitless injection mode. Chlorine-containing species were identified
from their mass spectra.
Because the O-H bond energy in hydroperoxides is
generally between 370 and 380 kJ/mol,17,18 hydrogen
abstraction from carbon is competitive with termination and
other reactions only for relatively weak C-H bonds.19 The
C-H bond energy in CH2Cl2 is reported to be 402.5 kJ/
mol;20 thus, the abstraction in eq 8 is unfavorable and, in
fact, is not observed in the gas phase.4,13 Nevertheless, it
should be considered as a potential reaction in the liquid
phase. Hydroperoxides normally oxidize substrates by break-
ing the O-O bond,21–23 and the resulting dichloromethoxy
radicals would decay as in eqs 6 and 7.
We wish to intervene in this process using longer
wavelength irradiation. We have observed the photodegra-
dation of chloroform to be catalyzed by iron(III) chloride
with visible light.24 Irradiation of FeCl3 solutions causes
chlorine atom dissociation and reduction to Fe(II), which is
then reoxidized thermally by CCl3OOH, an intermediate in
the chloroform degradation process that begins with hydrogen
abstraction from chloroform by the chlorine radicals.24 The
C-H bond energy is smaller in chloroform than it is in
dichloromethane,25 which might explain why the hydroper-
oxide accumulates in irradiated chloroform.26
Two methods were used to measure HCl production during
photolysis: (1) 200 µL aliquots were removed periodically from
the photolysate and shaken with 3.0 mL of DI water that had been
Several photoreactions of copper(II) complexes in organic
solvents are known in which reduction to copper(I) oc-
curs,27–31 including several studies of copper(II) chloride and
(28) Matsushima, R.; Ichikawa, Y.; Kuwabara, K. Bull. Chem. Soc. Jpn.
1980, 53, 1902–1907.
(29) Ng, W.; Hoggard, P. E. Inorg. Chim. Acta 2001, 321, 1–4.
(30) Sundararajan, S.; Wehry, E. L. J. Phys. Chem. 1972, 76, 1528–1536.
(31) Tanaka, K. J. Inorg. Nucl. Chem. 1981, 43, 2999–3001.
(32) Cervone, E.; Camassei, F. D.; Giannini, I.; Sykora, J. J. Photochem.
1979, 11, 321–332.
(17) Benson, S. W.; Shaw, R. AdV. Chem. Ser. 1968, 75, 288–294.
(18) Kondo, O.; Benson, S. W. J. Phys. Chem. 1984, 88, 6675–6680.
(19) Neta, P.; Huie, R. E.; Ross, A. B. J. Phys. Chem. Ref. Data 1990, 19,
413–513.
(33) Horvath, E.; Sykora, J.; Gazo, J. Z. Anorg. Allg. Chem. 1978, 442,
235–244.
(20) Seetula, J. A. J. Chem. Soc., Faraday Trans. 1996, 92, 3069–3078.
(21) Coffman, D. D.; Jenner, E. L. J. Am. Chem. Soc. 1958, 80, 2872–
2877.
(34) Horvath, O. J. Photochem. Photobiol., A 1989, 48, 243–248.
(35) Sykora, J.; Horvath, E.; Gazo, J. Z. Chem. 1978, 18, 346–347.
(36) Sykora, J.; Giannini, I.; Diomedi Camassei, F. J. Chem. Soc., Chem.
Commun. 1978, 207–208.
(22) Kharasch, M. S.; Fono, A. J. Org. Chem. 1959, 24, 72–78.
(23) Kochi, J. K. J. Am. Chem. Soc. 1962, 84, 2121–2127.
(24) Hoggard, P. E.; Gruber, M.; Vogler, A. Inorg. Chim. Acta 2003, 346,
137–142.
(37) Sykora, J.; Jakubcova, M.; Cvengrosova, Z. Collect. Czech. Chem.
Commun. 1982, 47, 2061–2068.
(38) Sharnoff, M.; Reimann, C. W. J. Chem. Phys. 1967, 46, 2634–2640.
(39) Desjardins, S. R.; Penfield, K. W.; Cohen, S. L.; Musselman, R. L.;
Solomon, E. I. J. Am. Chem. Soc. 1983, 105, 4590–4603.
(40) Poutsma, M. L. J. Am. Chem. Soc. 1965, 87, 42854293.
(41) Pocker, Y.; Stevens, K. D.; Champoux, J. J. J. Am. Chem. Soc. 1969,
91, 4199–4205.
(25) Tschuikow-Roux, E.; Paddison, S. Int. J. Chem. Kinet. 1987, 19, 15–
24.
(26) Ga¨b, S.; Turner, W. V. Angew. Chem. 1985, 97, 48.
(27) Gafney, H. D.; Lintvedt, R. L. J. Am. Chem. Soc. 1971, 93, 1623–
1628.
7030 Inorganic Chemistry, Vol. 47, No. 15, 2008

Photodegradation of CH2Cl2 by (Et4N)2CuCl4
has also been observed by Golubeva et al. in chlorobenzene
solutions.47,48
Anhydrous HCl was added to solutions of (Et4N)2[CuCl4]
in CH2Cl2, with the expectation that H2CuCl4 (in which the
protons would be associated with one or more chloride
ligands) would be formed, analogous to the protonation
observed with PdCl4 in chloroform,49 with a spectrum
2-
shifted somewhat from that of {2Et4N+;CuCl42-}. Instead,
the peaks associated with CuCl42- disappeared, while peaks
at 259, 324, and 475 nm remained. This spectrum, shown in
Figure 1, obeyed Beer’s Law at different copper concentra-
2-
tions and matched that of the Cu2Cl6 ion in the solid
state50,51 and in solution.47 This implies a strong association
beween HCl and chloride ion and, in fact, the equilibrium
constant constant for the process
HCl + Cl- h HCl-2
(10)
2-
2-
Figure 1. Spectra of CuCl4 (solid line) and 1/2 × Cu2Cl6 (dashed
2-
2-
line) in CH2Cl2; CuCl4 spectrum in presence of excess Cl-, Cu2Cl6 in
presence of excess HCl.
in CH3NO2 has been suggested to be so large that the
association is virtually stoichiometric.41 It may be presumed
that the anions in eq 10 are predominantly associated with
Et4N+ cations but, for simplicity, the ion pair notation will
not be used.
boiled to expel CO2, after which the pH of the aqueous phase was
recorded; (2) 50 µL aliquots were removed periodically and added
to 3.0 mL of a solution of tetraphenylporphyrin (H2TPP) in CHCl3,
the spectrum of which was then measured to determine the amount
of H4TPP2+ formed, by use of the extinction coefficients of the
porphyrin species.42,43 The equilibria strongly favor the diprotonated
over the monoprotonated porphyrin.44 There was good agreement
between the two methods, but the TPP method had better reproduc-
ibility, provided the stabilizer was removed from the dichlo-
romethane, because dissolved CH2Cl2 remaining in the aqueous
phase caused the pH meter to drift.
The spectrum of Cu2Cl62- shown in Figure 1 was arrived
at by determining the extinction coefficient at 1 nm intervals
from spectra at six concentrations by linear regression. In
each case, after adding HCl and recording the spectrum, more
HCl was bubbled into the solution to ensure that the spectrum
2-
was constant. The spectrum of CuCl4 was also arrived at
from the spectra at several concentrations. However, it was
found that addition of too much chloride ion caused reduction
to occur thermally by a process not understood, but alluded
to in the literature for Cu2+ with excess halide ions.52
Consequently, Cl- was added in small portions until the 474
nm band, characteristic of Cu2Cl62-, just disappeared, with
Peroxide concentrations were measured by mixing 1.00 mL of
the photolysate with 2.00 mL of approximately 0.01 M Bu4NI in
-
CHCl3 and determining the resulting I3 concentration from the
extinction coefficient, 2.50 × 104, at 365 nm.45
The concentration of C2H2Cl4 in photolysates was determined
from the peak area in the gas chromatogram relative to an internal
C2Cl6 standard.
-
isosbestic points still intact, before net reduction to CuCl2 ,
which lowered the absorbance at all wavelengths. Prepared
thusly, we found that Beer’s Law with respect to the copper
concentration was adhered to very well, and R2 for the linear
regression exceeded 0.99 at all wavelengths except between
550 and 600 nm, where the absorbance is very low.
III. Results
A. Chlorocopper Complex Speciation. UV-visible
absorption spectra of solutions of (Et4N)2[CuCl4] in CH2Cl2
did not follow Beer’s Law. Peaks were apparent in the
spectrum at approximately 251, 300, 409, and 474 nm. When
an excess of Et4NCl was added, the spectrum was altered to
one with peaks at 233, 294, and 410 nm (Figure 1). This
spectrum did follow Beer’s Law throughout and can readily
2-
2-
The concentrations of CuCl4 and Cu2Cl6 in solutions
of (Et4N)2[CuCl4] in CH2Cl2 were determined from absorp-
tion spectra by multicomponent analysis. The apparent
equilibrium constant for the dissociation process,
2CuCl24-hCu2Cl62- + 2Cl-
(11a)
2-
be identified as CuCl4 by comparison with both solid
state39,46 and solution47 spectra. This suggests that a rapid
decreased with increasing copper concentration. Extrapolated
to zero concentration, K was 5 × 10-5, about ten times
greater than the value found for chlorobenzene solutions.47
The equilibrium shifted with temperature, favoring dissocia-
tion at higher T. A plot of ln K versus 1/T (Supporting
2-
dissociation equilibrium is operative. Dissociation of CuCl4
(42) Du, H.; Fuh, R.-C. A.; Li, J.; Corkan, L. A.; Lindsey, J. S. Photochem.
Photobiol. 1998, 68, 141–142.
(43) Lindsey, J. S., PhotochemCAD, spectra recorded by Junzhong Li and
Richard W. Wagner. http://omlc.ogi.edu/spectra/PhotochemCAD/html/
index.html (accessed Mar 15, 2007).
(48) Golubeva, E. N.; Kokorin, A. I.; Kochubei, D. I.; Pergushov, V. I.;
Kriventsov, V. V. Kinet. Catal. 2002, 43, 408–411.
(44) Stone, A.; Fleischer, E. B. J. Am. Chem. Soc. 1968, 90, 2735–2748.
(45) Solis Montiel, E.; Solano, H. J. A. Ingenieria y Ciencia Quimica 1986,
10, 45–48.
(49) Le, A. L.; Hoggard, P. E. Photochem. Photobiol. 2008, 84, 86–89.
(50) Desjardins, S. R.; Wilcox, D. E.; Musselman, R. L.; Solomon, E. I.
Inorg. Chem. 1987, 26, 288–300.
(46) Smith, D. W. Coord. Chem. ReV. 1976, 21, 93–158.
(47) Kharitonov, D. N.; Golubeva, E. N. Kinet. Catal. 2003, 44, 513–517.
(51) Willett, R. D.; Liles; O, L., Jr. Inorg. Chem. 1967, 6, 1666–1669.
(52) Furlani, C.; Sgamellatti, A.; Ciullo, G. Ric. Sci., Rend. 1964, 49–60.
Inorganic Chemistry, Vol. 47, No. 15, 2008 7031

Doyle et al.
Figure 3. Initial rate of HCl formation in solutions of (Et4N)2CuCl4 in
unstabilized CH2Cl2, irradiated at 313 nm, as a function of the fraction of
light absorbed by the Cu2Cl62- component; I0 ) 5.4 × 10-10 einstein/s, R2
) 0.99.
Figure 2. UV-visible spectral changes during the irradiation of a 3.8 ×
10-4 M solution of (Et4N)2CuCl4 in aerated CH2Cl2 (350-W Hg lamp, 320-
nm cutoff filter, 1-min irradiation intervals).
Information, Figure S1) yielded 69 kJ/mol for ∆H and 166
J/mol · K for ∆S.
When an aerated sample of (Et4N)2CuCl4 with an excess
of Et4NCl was irradiated under the same conditions, the
photoreaction was completely suppressed. The CuCl4
spectrum remained unaltered, and no HCl was formed. On
the other hand, when the dissociation equilibrium was shifted
almost completely toward Cu2Cl62- by addition of anhydrous
HCl, irradiation caused reduction, followed by reoxidation.
The initial conversion to Cu2Cl62- can readily be ascribed
to the photoproduction of HCl which, as discussed above,
shifts the dissociation equilibrium.
B. Chlorocopper Photochemical Changes. Irradiation (λ
> 320 or 395 nm) of deaerated solutions of (Et4N)2[CuCl4]
in CH2Cl2 led to the loss of spectral features above 250 nm,
with the growth of a peak at 234 nm (see Supporting
Information, Figure S2). An isosbestic point was maintained
at 239 nm. The final spectrum was consistent with that of
2-
CuCl2 .53 Very little HCl was produced during the photore-
-
duction. After complete reduction to CuCl2- in eight separate
experiments, a maximum of 0.1 equiv of HCl, scaled against
the total copper present, was found in the solution. We were
unable to detect any C2H2Cl4 by GC-MS, which implies that
no significant formation of CHCl2 radicals occurred, as would
be expected if Cl atoms had formed and abstracted hydrogen.
When aerated solutions of (Et4N)2CuCl4 in CH2Cl2 were
exposed either to λ > 320 nm or to λ > 395 nm broadband
2CuCl24- + 2HCl h Cu2Cl62- + 2HCl-2
(11b)
C. Dichloromethane Decomposition Products. The
initial rate of HCl formation under 313 nm irradiation,
determined by adding small aliquots from the photolysate
to a solution of H2TPP, was measured over a range of starting
concentrations of (Et4N)2[CuCl4]. The rates correlated well
with the fraction of light absorbed by Cu2Cl62-, as shown in
Figure 3. The concentrations of Cu2Cl62- and CuCl42- in each
solution were determined by fitting the absorption spectrum
to the spectra of the pure components.
The initial rate of HCl formation was also studied as a
function of the incident light intensity. The rate varied
linearly with intensity, as shown in Figure 4. The slope
corresponded to a quantum yield, ꢀ, of 0.6, based on the
total light absorbed, but based only on the fraction of light
absorbed by Cu2Cl62-, ꢀ was 1.3.
Upon extended broadband (λ > 320 nm) irradiation in
unstabilized CH2Cl2, HCl continued to form, the development
being approximately linear well beyond stoichiometric
quantities, regardless of whether Cu(I) or Cu(II) predomi-
nated in the solution, as shown for one example in Figure 5.
In the presence of the amylene stabilizer, there was a
considerable induction period, following which HCl formed
at approximately the same rate as in unstabilized dichlo-
romethane.
irradiation, a series of spectral changes ensued, depicted in
2-
Figure 2. Initially, the 474 nm peak, due to Cu2Cl6
increased in intensity while the 404 nm peak, due to CuCl4
,
,
2-
decreased in intensity, and isosbestic points at 235, 247, 271,
312, 383, and 450 nm corresponded to the wavelengths at
2-
which the extinction coefficients of (1/2)Cu2Cl6 and
2-
CuCl4 (Figure 1) are equal.
After a period of conversion of CuCl4 to Cu2Cl6
2-
2-
,
continued irradiation of aerated solutions led to a loss of the
-
initial isosbestic points and net reduction to CuCl2 , similar
to the reaction occurring in deaerated solution. However, the
reduction eventually halted and reversed, regenerating Cu(II)
2-
as Cu2Cl6 (see Supporting Information, Figure S3). The
presence of an isosbestic point at 240 nm during the
reoxidation suggests that the only copper-containing species
-
present were CuCl2 and Cu2Cl62-. Further irradiation led
to no spectral changes as long as access to air was allowed.
When cuvettes were stoppered at this point, continued
-
irradiation eventually led to rereduction to CuCl2 .
Several halocarbon products were identified from GC-MS
analysis of photolysates at various reaction times, most
(53) Stevenson, K. L.; Braun, J. L.; Davis, D. D.; Kurtz, K. S.; Sparks,
R. I. Inorg. Chem. 1988, 27, 3472–3476.
7032 Inorganic Chemistry, Vol. 47, No. 15, 2008

Photodegradation of CH2Cl2 by (Et4N)2CuCl4
the peak area was less than 1% of that of chlorocyclohexane,
formed by HCl addition. We conclude that Cl2 was a minor
contributor to iodide oxidation. Figure 5 shows the total
peroxide concentration as a function of irradiation time. The
hydroperoxide did not reach a steady state during any of the
photolyses for which we tracked its concentration.
The buildup of (presumably) CHCl2OOH can be expected
eventually to be limited by slower thermal54 and acid-
catalyzed55 hydroperoxide decomposition routes, but no
decrease in rate was evident during the up to 2 h irradiation
times employed in this study.
No reaction was observed to take place (by GC-MS, UV
spectral changes, or HCl production) in neat, aerated CH2Cl2
under any of our experimental conditions,.
IV. Discussion
Figure 4. Initial rate of HCl formation in a 5 × 10-4 M solution of
(Et4N)2[CuCl4] in unstabilized CH2Cl2 as a function of the incident light
intensity. The slope is 2.2 × 102 M/einstein; R2 ) 0.99.
The most direct way to effect photoreduction and initiate
decomposition of CH2Cl2 would be by the photodissociation
of CuCl3- to produce CuCl2- and a chlorine radical. Though
2-
it is possible that Cu2Cl6 is in rapid equilibrium with
-
CuCl3 , it seems more likely that fragmentation occurs after
2-
excitation of Cu2Cl6
,
hν
Cu2Cl26- 98 CuCl3- + CuCl-2 + Cl
(12)
-
followed by association of CuCl3 ions to reconstitute
Cu2Cl62-. Chlorine atoms may then initiate decomposition
as in eqs 1–7. The rate for this step may be expressed as
I0fRꢀ/V, where I0 is the incident light intensity, fR the fraction
of light absorbed by the reactant, Cu2Cl62-, ꢀ the quantum
yield for bond homolysis, and V the solution volume.
The linear dependence of the initial rate of HCl formation
on the incident light intensity and on the fraction of light
2-
absorbed by the Cu2Cl6 species implies that there is a
Figure 5. Evolution of HCl, peroxide, and C2H2Cl4 during irradiation (λ
> 320 nm) of a solution of (Et4N)2CuCl4 in aerated, unstabilized CH2Cl2.
Concentrations are relative to [Cu] ) 6.0 × 10-5 M.
radical termination pathway other than dimerization because
a biradical termination alone would lead to a square root
dependence of the reaction rate on the absorbed light
intensity. Such a pathway exists if the peroxy radicals (eq
4) oxidize Cu(I) back to Cu(II) by electron transfer.
prominently 1,1,2,2-tetrachloroethane, 1,1,2,3,3-pentachlo-
ropropane, and chloroform. The concentration of C2H2Cl4
as a function of irradiation time is also shown in Figure 5.
Although its concentration increased approximately linearly
with irradiation time, the relatively small yield compared to
other products implies that dimerization of · CHCl2 is not
the major radical termination step.
A key question with respect to photodecomposition in
liquid dichloromethane is whether it proceeds through a
hydroperoxide, as is known to occur with chloroform,26
despite the apparently unfavorable energetics for the forma-
tion of the hydroperoxide by hydrogen abstraction from
CH2Cl2. Although we were unable to analyze specifically
for this compound (it eluted with the solvent under all GC-
MS conditions that were tried), an analysis of total peroxide
was performed by the oxidation of iodide to triiodide.45
Hydroperoxides other than CHCl2OOH could in any case
function in like manner to reoxidize Cu(I) to Cu(II). To test
whether iodide might have been oxidized by Cl2 rather than
hydroperoxides, cyclohexene was added to a photolysate. A
small amount of 1,2-dichlorocyclohexane was formed, but
k13
CHCl2OO + CuCl-2 98 CHCl2OO- + CuCl2 (13)
k14
9
CHCl2OO- + HCl
8 CHCl2OOH + Cl-
(14)
(15)
k15
CuCl2 + Cl- 98 CuCl3-
Hydroperoxides generally react with oxidizable substrates
by breaking the O-O bond.21–23 Dichloromethylhydroper-
-
oxide can thus serve as a second oxidant for CuCl2 , yielding
thereby dichloromethoxy radicals, which decompose uni-
molecularly15 (eq 6).
(54) Kirillov, A. I. Zh. Org. Khim. 1965, 1, 1226–1230.
(55) Kharasch, M. S.; Fono, A.; Nudenberg, W. J. Org. Chem. 1950, 15,
748–752.
Inorganic Chemistry, Vol. 47, No. 15, 2008 7033

Doyle et al.
k16
This is consistent with the observed behavior in which
reduction of copper(II) to CuCl2- upon irradiation is reversed,
eventually reverting almost completely to the +2 state as
the concentration of the hydroperoxide increases.
Although eq 19 is consistent with our observation for
aerated solutions, it does not account for the lack of
stoichiometric HCl production in deoxygenated CH2Cl2, and
this remains unexplained.
CHCl2OOH + CuCl-2 98 CuCl2(OH)- + CHCl2O (16)
k17
9
CuCl2(OH)- + HCl
8 CuCl-3 + H2O
(17)
Given some dimerization of CHCl2 radicals, as was found
experimentally,
k18
V. Conclusion
2CHCl2
98 C2H2Cl4
(18)
Under near-UV, and even visible, irradiation, dissolved
(Et4N)2[CuCl4] catalyzes the decomposition of CH2Cl2, the
photoactive species being Cu2Cl62-. The photodissociation
of a chlorine atom initiates the decomposition. During the
process, a great deal of the hydroperoxide CHCl2OOH builds
up, even though abstraction of hydrogen by peroxy radicals
is unfavorable. It is proposed that the hydroperoxide is
formed instead through electron transfer from copper(II) to
the peroxy radical, followed by protonation of the resulting
peroxide anion.
a steady state in all intermediates except CHCl2OOH leads
to the relationship
d[CHCl2OOH] I0fRφ
d[HCl]
+
)
(19)
dt
dt
V
Thus, the sum of HCl and CHCl2OOH is predicted to grow
in direct proportion to the incident light intensity and the
fraction of light absorbed by the photoactive copper species.
Assuming this mechanism to be operative, the fractions of
HCl and CHCl2OOH produced are controlled by the relative
rates of peroxy radical formation (eq 4) and biradical
termination (eq 18). A faster termination step is predicted
to lead to a greater fraction of HCl while a more rapid
CHCl2OO formation step is predicted to lead to a higher
fraction of CHCl2OOH. The latter condition was observed
experimentally, which is consistent with the low rate
observed for the formation of the C2H2Cl4 termination
product.
Acknowledgment. This work was supported by the US
National Science Foundation through grants CHE-0079152
and CHE-0453460 (REU).
Supporting Information Available: Figures showing the tem-
perature dependence of the dimerization equilibrium constant,
sequential spectra from the irradiation of a deoxygenated solution
of (Et4N)2[CuCl4] in CH2Cl2, and sequential spectra from the
irradiation of an aerated solution, showing the conversion of Cu(II)
to Cu(I) and back to Cu(II) (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
Once the copper species reach a photostationary state, the
mechanism predicts that the concentration of CuCl2- should
be inversely proportional to the hydroperoxide concentration.
IC800684V
7034 Inorganic Chemistry, Vol. 47, No. 15, 2008