Interaction of chlorinated ethylenes with chromium exchanged zeolite Y: An in situ FT-IR study

Article abstract of DOI:10.1006/jcat.1997.1443

The interaction of chlorinated ethylenes (vinyl chloride, 1,1 dichloroethylene, trichloroethylene, and perchloroethylene) with the surface of chromium exchanged zeolite Y (Cr-Y) catalyst has been studied by in situ FT-IR spectroscopy. The adsorptions were carried out on the in situ oxidized Cr-Y pellet at temperatures between 25 and 300°C in a dry nitrogen stream as a means of studying possible decomposition intermediates. The adsorption at 25°C was chiefly physical in nature although some dechlorination of the molecule was evident even at this temperature. At higher temperatures, between 100 and 300°C, an oxygen attack on the adsorbed molecule led to the formation of partially and/or fully oxygenated (but still adsorbed) species. These oxygenated species, including carboxylate and carbonate, were found to contain fewer chlorine atoms than the original feed molecule. The catalytic activity for the formation of these intermediates was found to diminish with increasing chlorine content of the feed molecule. Based on these results, a reaction pathway for the progressive catalytic oxidation of chlorinated ethylenes has been proposed.

Full text of DOI:10.1006/jcat.1997.1443

JOURNAL OF CATALYSIS 165, 12–21 (1997)
ARTICLE NO. CA971443
Interaction of Chlorinated Ethylenes with Chromium Exchanged
Zeolite Y: An in Situ FT-IR Study
Prashant S. Chintawar and Howard L. Greene1
Department of Chemical Engineering, The University of Akron, Akron, Ohio 44325-3906
Received February 27, 1996; revised August 15, 1996; accepted August 19, 1996
action of TCE with OH radicals leading to the formation
of CHCl2CHO intermediate which was further mineral-
ized to HCl and CO2. Dibble and Raupp (6) have detected
CHCl2CHO and DCE during photooxidation of TCE using
in situ FT-IR techniques. They proposed that CHCl2CHO
is then further attacked to yield complete oxidation prod-
ucts. Three possible reactive photocatalytic species, namely
HO2, HO , and O2 were suggested as responsible for the
attack to produce CHCl2CHO.
Kazanjian and Horell (7) have studied the radiation in-
duced oxidation of TCE using FT-IR spectroscopy. Irradi-
ations were carried out in a Gammacell-220, a commer-
cial source containing about 3200 Ci of cobalt-60. The
three major products identified were dichloroacetyl chlo-
ride (CHCl2COCl), phosgene (COCl2), and trichloroethy-
lene oxide. A free radical mechanism was proposed to ac-
count for the observed intermediates. Similarly, Sutherland
and Spinks (8) have studied radiolysis of PCE. Upon irradi-
ation, PCE oxidized to trichloroacetyl chloride which was
later found to hydrolyze to trichloroacetic acid (8).
The interaction of chlorinated ethylenes (vinyl chloride, 1,1
dichloroethylene, trichloroethylene, and perchloroethylene) with
–
the surface of chromium exchanged zeolite Y (Cr Y) catalyst has
been studied by in situ FT-IR spectroscopy. The adsorptions were
–
carried out on the in situ oxidized Cr Y pellet at temperatures be-
tween 25 and 300 C in a dry nitrogen stream as a means of study-
ing possible decomposition intermediates. The adsorption at 25 C
was chiefly physical in nature although some dechlorination of the
molecule was evident even at this temperature. At higher temper-
atures, between 100 and 300 C, an oxygen attack on the adsorbed
molecule led to the formation of partially and/or fully oxygenated
(but still adsorbed) species. These oxygenated species, including
carboxylate and carbonate, were found to contain fewer chlorine
atoms than the original feed molecule. The catalytic activity for
the formation of these intermediates was found to diminish with
increasing chlorine content of the feed molecule. Based on these
results, a reaction pathway for the progressive catalytic oxidation
c
of chlorinated ethylenes has been proposed.
1997 Academic Press, Inc.
INTRODUCTION
Arnold et al. (9) have studied the homogeneous reaction
of oxygen atoms with CEs using infrared chemiluminscence
techniques. The initial step was postulated to involve the
formation ofan aldehyde. Reaction productsobserved were
CO, CO2, Cl2, HCl, HCHO, and COCl2.
Chlorinated ethylenes (CEs) such as vinyl chloride (VC,
== ==
trichloroethylene (TCE, HClC CCl2), and perchloroeth-
H2C CHCl), 1,1 dichloroethylene (1,1 DCE, Cl2C CH2),
==
==
ylene (PCE, Cl2C CCl2) constitute a significant fraction of
More recently, Mortland and Boyd (10) studied the in-
teraction of 1,2 DCE, TCE, and PCE with Cu(II)-smectite
suspension in CCl4 solution at reflux temperature using a
combination of FT-IR, ESR, and NMR spectroscopy. The
interaction was found to have resulted in dechlorination
and the formation of a carbonyl group. Similarly, Song
et al. (11) have studied the reaction of unsaturated organic
halides with the gas phase vanadium clusters. The results
obtained were explained in terms of three different mecha-
nisms for the reaction of halopropene: halogen abstraction;
halogen substitution followed by molecular hydrogen evap-
oration; and molecular addition followed by evaporation of
the molecular hydrogen and hydrogen halide.
the hazardous air and water pollutants. Low temperature
catalytic destruction (1–4) appears to be the most popular
and effective method for their remediation. However, the
available literature still lacks heterogeneous thermal cata-
lytic reaction mechanisms for their oxidation.
The destruction mechanism of these CEs, especially
TCE, using other techniques has been widely studied. A
large number ofstudiesdealingwith photocatalyticdestruc-
tion of TCE has been reported in the literature. Pruden and
Ollis (5) have reported detection of dichloroacetaldehyde
(CHCl2CHO) intermediate during photodecomposition of
TCE over TiO2 in aqueous solution. They proposed inter-
The present study was undertaken to elucidate the path-
ways involved in the progressive oxidation of several CEs,
1
To whom correspondence should be addressed. Present address:
Department of Chemical Engineering, Case Western Reserve University,
Cleveland, OH 44106-7217.
–
namely VC, 1,1 DCE, TCE, and PCE on a Cr Y zeolite
12
0021-9517/97 $25.00
c
Copyright
1997 by Academic Press, Inc.
All rights of reproduction in any form reserved.

–
INTERACTION OF CHLORINATED ETHYLENES WITH Cr Y
13
surface as a function of temperature and chlorine content of
In situ adsorption of CEs was carried out at three differ-
the feed, using infrared absorption spectroscopy to follow ent temperatures, viz., 25, 100, and 300 C, and at a totalpres-
–
the course of reaction. The Cr Y zeolite catalyst was cho- sure of 1 atm. The intermediate temperature of 100 C was
sen for this study because of its known high activity toward chosen because it is higher than the normal boiling point of
the complete oxidation of CEs (12). The adsorption of CEs all the feeds except PCE. For PCE, the intermediate in situ
was carried out at low (25 C) as well as high (300 C) tem- adsorption experiments had to be carried out at 130 C.
peratures to aid in detecting all the possible intermediates
RESULTS
(which might not otherwise be visible at high temperature)
leading to the formation of complete oxidation products.
The gas phase spectra of the four CEs in nitrogen and
their interpretation are documented in Table 1. With the ex-
EXPERIMENTAL
==
ception of PCE, all CEs show distinct C C stretching bands
( C C) in the 1575–1630-cm 1 region. In the gas phase, the
==
The Y zeolite (Si/Al = 2.5), containing 20 wt% alumina
binder, was received from UOP in the form of 1/1600 pellets
(LZY-64). Two ammonium exchanges, each lasting3h, were
carried out on these pellets at 60 C using 2.24 M NH4Cl so-
PCE molecule has symmetry described by the point group
D2h. This band is therefore IR forbidden although it is vis-
ible in the Raman spectrum at 1571 cm 1 (13).
–
lution in water. The NH4 Y so obtained was then subjected
Adsorption of VC
to chromium exchange with 42.9 mM CrN3O9 9H2O solu-
tion as a source of chromium cations, at 60 C for 72 h. The
initial pH of the solution was adjusted to 4.0 by the addi-
tion ofa fewdropsofaqueousNH4OH. After the chromium
exchange, the catalyst pellets were thoroughly washed with
deionized and double distilled water, dried at 100 C and
Figure 1B shows the spectra obtained during VC adsorp-
tion at 25 C. The adsorption is accompanied by the detec-
tion of gas phase CO2 as seen by a weak doublet at 2365 and
TABLE 1
Vibrational Frequencies (cm 1) of Gas Phase CEs
and Their Assignment
–
then calcined at 500 C for 12 h. This catalyst (Cr Y) was
found to contain 4.14% Cr (Phillips PV9550 X-ray fluo-
rescence spectrometer) and had a BET (Quantasorb Jr.
surface area analyzer) surface area of 430 m2/g.
In situ FT-IR experiments were performed in a Bio-Rad
model FTS-7 spectrometer at 4-cm 1 resolution. The spec-
trometer was equipped with purged bench option. The ex-
periments were conducted in a variable temperature trans-
mission cell (HTC-100) manufactured by Harrick Scientific.
VC
1,1 DCE
Assignment
Assignment
(from (32))
Frequency
Frequency
(from (13))
–
617
709
729
C Cl twist
455
610
793
CCl2 wag
–
C Cl stretch
CCl2 sym-stretch
CCl2 asym-stretch
CH2 wag
–
C Cl stretch
–
897
C Cl wag
868
–
The Cr Y catalyst samples for the reactions were made in
941
CH2 wag
CH2 rock
1088
1391
1462
1576
1614
CH2 rock
CH2 scissor
CH2 deformation
the form of self supported pellets 13 mm in diameter using
a pellet die also made by Harrick Scientific.
1023
1269
1290
1361
–
C H rock
–
==
==
C H rock
C
C
C stretch
C stretch
In a typical experiment, the catalyst pellet was calcined
in situ at 350 C in flowing air for 3 h. The temperature of
the pellet was then lowered to the desired adsorption tem-
perature followed by 1 h evacuation to 10 5 Torr (1 Torr =
133.3 N m 2). The FT-IR spectrum of the catalyst pellet
was collected at this point and later used as a reference for
subsequent adsorption spectra.
CH2
deformation
CH2
==
1387
1628
C
C stretch
deformation
==
==
1599
1605
3097
3130
C
C
C stretch
C stretch
–
C H stretch
–
C H stretch
In situ adsorption of CEs on these pellets was carried out
at 2000 PPM concentration in dry nitrogen (<8 PPM water)
for 30 min. Preliminary adsorption experiments with dry ni-
trogen only (without any CE) showed no change in the zeo-
lite spectrum indicating no contamination by the carrier gas.
Rapid spectra were collected during the adsorption process
TCE
PCE
Assignment
(from (22))
Assignment
(from (13))
Frequency
Frequency
–
–
–
635
781
847
C Cl stretch
C Cl stretch
C H bend
781
916
CCl2 asym-stretch
CCl2 sym-stretch
5
and the cell was again evacuated to 10 Torr to remove
–
the gas phase and physisorbed species. The spectrum ob-
tained after 1 h evacuation was assigned to the chemisorbed
species on the surface. Pure component gas phase spectra,
used for comparison, were obtained by separately filling the
empty reaction cell with 2000 PPM of CE in dry nitrogen.
941
C Cl stretch
–
1254
1564
1593
3098
CH Cl bend
==
==
C
C
C stretch
C stretch
–
C H stretch

14
CHINTAWAR AND GREENE
Figure 1C shows that evacuation of the gas phase species
1
leads to the disappearance of 729- and 710-cm bands
1
–
==
( C Cl) with the retention of the 1604-cm ( C C) and
1
1387-cm (CH2 deformation) bands indicating the disso-
–
ciative adsorption of VC on Cr Y surface. The retention
of the C O band at 1678 cm 1 after evacuation suggests
==
strong adsorption of aldehyde/ketone on the surface.
The spectra obtained during the adsorption of VC at
100 C are shown in Fig. 2B. These spectra are character-
1
ized by strong absorption bands at 1383 and 1607 cm
.
This doublet signifies the carboxylate species of the formate
type on the surface of the catalyst, the basis for which came
–
from the adsorption of formic acid (HCOOH) on Cr Y,
as seen in Fig. 2A. These absorption bands represent asym-
metric stretching ( COO , asym) and symmetric stretching
( COO , sym), respectively, of the carboxylate of the for-
mate type (17). It is evident from the spectra in Fig. 2B
that in addition to the carboxylate of the formate type, the
moderately strong band at 1672 cm 1, previously assigned
to carbonyl stretching fequency of aldehye or ketone, is also
FIG. 1. FT-IR spectra of the species arising from VC adsorption on
–
Cr Y at 25 C. A: Gas phase VC spectrum. B, Progress of adsorption:
1, 1 min; 2, 30 min. C: Spectrum obtained after evacuation to 10 5 Torr.
1
2330 cm 1. Additionally, weak bands are seen at 1678 cm
–
observed during VC adsorption on Cr Y at 100 C. Evacua-
and 1842 cm 1. The CO2 doublet disappears after a few
tion does not remove any of the bands suggesting that these
species are strongly held on the catalyst surface (Fig. 2C).
Unlike adsorption at 100 C, which chiefly results in the
formation of carboxylate of the formate type, the adsorp-
tion at 300 C leads to the formation of a mixture of nonco-
ordinated carbonate (CO23 ) and carboxylate of the CO2
type as shown in Fig. 3A. While the presence of a broad
band at 1576 cm 1 ( COO , asym) along with the band at
1375 cm 1 ( COO , sym) suggests the presence of carboxy-
late of the CO2 type on the surface (18), the absorption at
1479cm 1 suggestsCO23 firmlyattached to the surface (19).
Evacuation does not remove the carboxylate and carbonate
species, as seen in Fig. 3B.
minutes of adsorption, whereas the intensity of the 1678-
1
and 1842-cm
adsorption.
bands increases during the course of
1
The frequency 1678 cm is very close to the stretching
==
==
frequency of the C O bond ( C O) of various aldehy-
des and ketones. For example, Finocchio et al. (14) have
reported C O for methylethyl ketone at 1660 cm 1, and
==
for benzaldehyde adsorbed on MgCr2O4 catalyst at 1684
and 1653 cm 1. In a separate experiment, we carried out
–
the adsorption of CHCl2COCl on Cr Y and detected the
1
O band at 1660 cm 1. On this basis, the 1678-cm
==
C
band observed during VC adsorption in the present study
is assigned to the carbonyl stretching frequency of an ad-
sorbed aldehyde or ketone. The adsorption band observed
1
at 1842 cm can be ascribed to COHCl species, as the
==
COHF molecule has been reported to show C O absorp-
tion at 1837 cm 1 (15).
Comparison of the intensities of the gas phase bands
(Fig. 1A) and those observed during adsorption (Fig. 1B)
indicate that C Cl stretching bands (710 and 729 cm 1) are
–
weaker and the CH2 deformation band (1366 cm 1) signifi-
cantly stronger during the adsorption than in the gas phase.
However, the majority of the bands seen during the ad-
sorption correspond to those in the gas phase VC with only
minor shifts. These perturbations (changes in the relative
intensity) suggest the bending of adsorbed VC molecule
–
via the C Cl group. A similar observation was made by
Grabowski et al. (16) during the adsorption of C3H6 on the
–
–
CoO MgO MoO3 surface. The relative intensity of the CH
deformation mode was substantially higher in the adsorbed
phase than in the gas phase and this phenomenon was as-
cribed to the bending of adsorbed C3H6 molecule via CH3
group.
FIG. 2. FT-IR spectra of the species arising from VC adsorption on
–
Cr Y at 100 C. A: Bands due to carboxylate species of the formate type
–
(Adsorption of formic acid on Cr Y for comparison). B, Progress of ad-
sorption: 1, 1 min; 2, 30 min. C: Spectrum obtained after evacuation to
10 5 Torr.

–
INTERACTION OF CHLORINATED ETHYLENES WITH Cr Y
15
FIG. 3. FT-IR spectra of the species arising from VC adsorption on
FIG. 5. FT-IR spectra of the species arising from 1,1 DCE adsorption
–
–
Cr Y at 300 C. A, Progress of adsorption: 1, 1 min; 2, 30 min. B: Spectrum
obtained after evacuation to 10 5 Torr.
on Cr Y at 100 C. A: Bands due to carboxylate species of the acetate type
–
(adsorption of acetic acid on Cr Y for comparison). B, Progress of adsorp-
tion: 1, 1 min; 2, 7 min; 3, 30 min. C: Spectrum obtained after evacuation
to 10 5 Torr.
Adsorption of 1,1 DCE
The 1,1 DCE adsorption at 25 C is depicted in Fig. 4B.
The progress of adsorption is accompanied by an increase
in the intensity of all the bands. Evacuation leads to the
removal of all the bands (Fig. 4C) indicating the occurrence
of only physical adsorption.
1
==
(18) suggests the value of 1693 cm for C O of gaseous
1
formaldehyde (HCHO). Therefore, the band at 1696 cm
can be assigned to the gas phase HCHO. A pair of intense
bands detected at 1445 and 1578 cm 1 (Fig. 5B) represent
the carboxylate of the acetate type, which is not removed
by evacuation (Fig. 5C). The basis for the assignment of the
frequencies to the vibrations of carboxylate of the acetate
type came from the adsorption of acetic acid (CH3COOH)
1
In addition to the gas phase bands at 1611 and 1626 cm
1
in the 1400–2000 cm wavenumber region, as shown in
Fig. 5B, 1,1 DCE adsorption at 100 C is accompanied by the
formation of new bands at 1445, 1578, and 1696 cm 1. The
–
on Cr Y as seen in Fig. 5A, where strong bands are regis-
1
intensity of the weak band at 1696 cm increases during
1
tered at 1418, 1478, and 1587 cm
.
the initial 7 min of adsorption and subsequently disappears.
–
Interaction of 1,1 DCE with Cr Y at 300 C is depicted in
Fig. 6. Very intense bands due to carboxylate of the acetate-
type species are seen at 1576 and 1443 cm at the com-
==
Mortland and Boyd (20) have suggested the
C
O value
1
==
of 1700 cm for the hydrogen saturated C O group and
1740 cm for the chlorine saturated C O group. Davydov
1
1
==
mencement of adsorption. Another moderate adsorption
FIG. 4. FT-IR spectra of the species arising from 1,1 DCE adsorp-
–
tion on Cr Y at 25 C. A: Gas phase 1,1 DCE spectrum. B, Progress of
FIG. 6. FT-IR spectra of the species arising from 1,1 DCE adsorp-
–
adsorption: 1, 1 min; 2, 30 min. C: Spectrum obtained after evacuation to
10 5 Torr.
tion on Cr Y at 300 C. A, Progress of adsorption: 1, 1 min; 2, 30 min.
B: Spectrum obtained after evacuation to 10 5 Torr.

16
CHINTAWAR AND GREENE
band at 1483 cm 1 suggests the presence of CO23 species,
as in the case of VC. As seen from Fig. 6B, the evacuation
does not remove carboxylate and carbonate species from
the surface.
Adsorption of TCE
TCE adsorption at 25 C isshown in Fig. 7A. In addition to
all the gas phase frequencies, a strong negative absorption
is detected at 3744 cm 1, which is not observed during VC
and 1,1 DCE adsorption at 25 C (Figs. 1B and 4B). This fre-
–
quency corresponds to silanol (Si OH) groups on the outer
surface of the crystallites; i.e., OH groups which terminate
the faces of zeolite crystallites at positions where bonding
in the interior would occur with adjacent tetrahedral Si or
Al ions (21). A negative band at this position indicates the
TCE interaction with these silanol groups.
FIG. 8. FT-IR spectra of the species arising from TCE adsorption on
–
Cr Y at 100 C. A, Progress of adsorption: 1, 1 min; 2, 30 min. B: Spectrum
obtained after evacuation to 10 5 Torr.
Most of the frequencies show a negligible shift from the
–
gas phase values with the exception of the C H band at
1
3082 cm and the
C
C band at 1584 cm 1. The C H
==
–
1
==
and C O vibrations, respectively, of CHCl2CHO as sug-
band shows a net drop of 16 cm over the gas phase fre-
gested by Phillips and Raupp (22). A weak negative band
is again registered at 3752 cm indicating the interaction
–
quency. This perturbation suggests that the C H bond is
1
==
involved in the adsorption process. The C C bond shows
–
of TCE molecules with the Si OH groups. However, the
two stretching frequencies in the gas phase at 1564 and
1593 cm 1 (Fig. 7A), therefore the value of 1584 cm 1 ob-
tained during the adsorption cannot be interpreted prop-
erly. However, the intense nature of this band suggests the
majority of the bands is due to gas phase TCE with neg-
ligible shifts. After 15 min of adsorption, a strong band is
also seen at 1418 cm 1. This band along with the band at
1584 cm 1 indicate the formation of carboxylate of the ac-
etate type since the adsorption of CH3COOH gives rise to
==
strong polarization of the C C bond during adsorption.
The evacuation leads to the removal of all the bands indi-
cating that the adsorption is physical only, as depicted in
Fig. 7C.
TCE adsorption at 100 C is shown in Fig. 8A. In addition
to all the gas phase frequencies, two prominent bands are
seen at 1387 and 1751 cm 1. The intensity of these bands
is found to increase during the course of adsorption. These
1
bands at 1418, 1478, and 1587 cm as shown in Fig. 5A.
After evacuation, only the bands due to adsorbed car-
boxylate of the acetate type and CHCl2CHO are seen
(Fig. 8B).
As seen from Fig. 9A, unlike adsorption at lower temper-
ature, the adsorption at 300 C does not show any interme-
diate species like CHCl2CHO. However, very intense bands
==
==
bands can be assigned to the HC O bending ( HC O)
FIG. 7. FT-IR spectra of the species arising from TCE adsorption on
FIG. 9. FT-IR spectra of the species arising from TCE adsorption on
–
–
Cr Y at 25 C. A: Gas phase TCE spectrum. B, Progress of adsorption: 1,
Cr Y at 300 C. A, Progress of adsorption: 1, 1 min; 2, 30 min. B: Spectrum
1 min; 2, 30 min. C: Spectrum obtained after evacuation to 10 5 Torr.
obtained after evacuation to 10 5 Torr.

–
INTERACTION OF CHLORINATED ETHYLENES WITH Cr Y
17
1
due to carboxylate of the acetate type (1568 and 1445 cm
)
are seen at the commencement of adsorption. Evacuation
leads to the removal of gas phase TCE and only the car-
boxylate species of the acetate type remains on the surface
(Fig. 9B).
Adsorption of PCE
==
As already mentioned, the C C stretching band is not
visible in gas phase PCE due to the symmetric nature of this
molecule. However, as seen from Fig. 10B, an intense vibra-
1
==
tion is registered at 1603 cm due to C C during adsorp-
tion of PCE at 25 C. This band, at higher frequency than the
==
C
C bond stretching vibration of the free molecule in the
vapor phase (1571 cm 1), indicates the interaction of PCE
molecules with surface adsorption sites through the dissoci-
FIG. 11. FT-IR spectra of the species arising from PCE adsorption on
–
ation of one or more C Cl bonds. As shown in Table 1, the
==
–
Cr Y at 130 C. A, Progress of adsorption: 1, 1 min; 2, 30 min. B: Spectrum
obtained after evacuation to 10 5 Torr.
C
C of CEs is related to the number of chlorine atoms
==
==
attached to the C C bond. An increase in the C C there-
fore indicates a loss of one or more chlorine atoms of the
PCE molecule. A similar phenomenon has been observed
–
teraction with the Cr Y surface. The adsorbed species can
be easily removed by evacuation (Fig. 10D).
==
PCE adsorption at 130 C is shown in Fig. 11. Unlike with
the other feeds at this temperature, there is no evidence
of carbonyl formation. The strong band at 1609 cm lies
by Phillips and Raupp, where the C C of TCE was found
1
to increase from 1597 cm 1 to 1609 cm upon Cl abstrac-
–
1
tion by coordinatively unsaturated Ti cations of the TiO2
surface (22). In this reference, the spectrum of adsorbed
TCE is matched to that of DCE.
==
in the C C region and is most probably due to 1,1 DCE,
1
==
since, 1,1 DCE shows the C C stretching band at 1614 cm
(Table 1). An intense band at 1416 cm due to the CH2
1
At 25 C, the bands observed during PCE adsorption
scissor of adsorbed 1,1 DCE is also seen. The adsorption in
this case, however, is irreversible since this species cannot
be removed by evacuation (Fig. 11B).
PCE adsorption at 300 C is shown in Fig. 12A. As in the
case of other feeds at this temperature, these spectra are
characterized by strong bands due to carboxylate of the ac-
etate type at 1591 and 1408 cm 1. The adsorption is also
accompanied by a weak band at 1782 cm 1, suggesting the
formation of small amounts of monochloroacetaldehyde
(Fig. 10B) are very similar to those found for VC. Hence,
1
1603 cm corresponds to C C and 1354 cm 1 to the CH2
==
deformation band of adsorbed VC (Fig. 10C) while the
1
bands at 916 and 779 cm can be assigned to gas phase
PCE (Fig. 10A) as seen from Fig. 10. Strong silanol inter-
action is again observed at 3732 cm 1. Strong perturbation
==
of the C C bond is evident from the fact that the
==
C
C
band (invisible in the gas phase) becomes visible upon in-
FIG. 10. FT-IR spectra of the species arising from PCE adsorption on
–
Cr Y at 25 C. A: Gas phase PCE spectrum; B: Spectrum obtained after
FIG. 12. FT-IR spectra of the species arising from PCE adsorption on
–
–
30 min adsorption of PCE; C: Spectrum of VC adsorbed on Cr Y at 25 C
Cr Y at 300 C. A, Progress of adsorption: 1, 1 min; 2, 30 min. B: Spectrum
(for comparison); D: Spectrum obtained after evacuation to 10 5 Torr.
obtained after evacuation to 10 5 Torr.

18
CHINTAWAR AND GREENE
TABLE 2
Species Observed during Various Stages of Adsorption
Temperature of
adsorption ( C)
Feed
VC
Species and the bands detected
==
25
During adsorption: gas phase CO2,
C
O of aldehyde/ketone (1678 cm 1),
COHCl (1842 cm 1), gas phase VC
==
1
–
After evacuation:
C
O of aldehyde/ketone (1678 cm ), VC without C Cl bond
VC
VC
100
During adsorption: gas phase CO2, carboxylate of the formate type (1383 and 1607 cm 1),
==
C
O of aldehyde/ketone (1672 cm 1), gas phase VC
After evacuation: carboxylate of the formate type (1383 and 1607 cm 1),
1
==
C
O of aldehyde/ketone (1672 cm
)
300
During adsorption: gas phase CO2, carboxylate of the CO2 type
(1375 and 1576 cm 1), CO23 (1479 cm 1), gas phase VC
After evacuation: carboxylate of the CO2 type (1375 and 1576 cm 1), CO32 (1479 cm
)
1
1,1 DCE
1,1 DCE
25
During adsorption: gas phase CO2, gas phase 1,1 DCE
After evacuation: flat spectrum
During adsorption: gas phase CO2, HCHO (1696 cm 1), carboxylate of the acetate type
(1445 and 1578 cm 1), gas phase 1,1 DCE
100
1
After evacuation: carboxylate to the acetate type (1445 and 1578 cm
)
1,1 DCE
300
During adsorption: gas phase CO2, carboxylate of the acetate type (1443 and 1576 cm 1),
CO23 (1483 cm 1), gas phase 1,1 DCE
After evacuation: carboxylate of the acetate type (1443 and 1576 cm 1), CO32 (1483 cm
)
1
TCE
TCE
25
During adsorption: gas phase CO2, strong silanol interaction (3744 cm 1), gas phase TCE
After evacuation: flat spectrum
100
During adsorption: gas phase CO2, moderate silanol interaction (3752 cm 1), CHCl2CHO
(1387 and 1751 cm 1), gas phase TCE, carboxylate of the acetate type
(1418 and 1584 cm 1) (after 15 min adsorption)
After evacuation: carboxylate to the acetate type (1418 and 1584 cm 1),
1
CHCl2CHO (1387 and 1751 cm
)
TCE
300
During adsorption: gas phase CO2, carboxylate of the acetate type
(1445 and 1568 cm 1), gas phase TCE
1
After evacuation: carboxylate of the acetate type (1445 and 1568 cm
)
PCE
PCE
PCE
25
130
300
During adsorption: gas phase CO2, adsorbed VC, gas phase PCE
After evacuation: flat spectrum
During adsorption: gas phase CO2, partially dechlorinated PCE, gas phase PCE
After evacuation: partially dechlorinated PCE
During adsorption: gas phase CO2, carboxylate of the acetate type (1408 and 1591 cm 1),
CH2ClCHO (1782 cm 1), strong silanol interaction (3731 cm 1), gas phase PCE
1
After evacuation: carboxylate of the acetate type (1408 and 1591 cm
)
(CH2ClCHO) (23). Strong silanol interaction is also reg- the temperature. This gas phase CO2 is shown as a dou-
istered as a negative absorption at 3731 cm 1. Evacuation blet at 2360 and 2340 cm 1 in Figs. 1B, 4B, and 7B. With
leads to the removal of gas phase PCE; carboxylate of the the progress of time of adsorption, however, the gas phase
acetate type, along with the negative band at silanol are the CO2 disappears and the CEs subsequently react with the
–
only remaining absorptions (Fig. 12B).
Cr Y surface giving rise to various partial oxygenated prod-
Table 2 summarizes the species observed during various ucts such as COHCl, CHCl2CHO, CH2ClCHO, HCHO, car-
stages of adsorption for all the feeds as a function of tem- boxylate of the type CO2 , formate, and acetate. Recall that
perature.
adsorptions are carried out in the absence of the oxygen
in the feed stream and only the initial chemisorbed oxygen
is present on the catalyst surface. It is speculated that the
chemisorbed oxygen (and not the gas phase oxygen) is the
reactive species. This implies that the reaction pathway de-
DISCUSSION
Common Adsorption Features
As shown in Table 2, a common feature of all the ad- veloped here is initially applicable to the reactions carried
sorption spectra is the formation of gas phase CO2 at the out in the presence of gas phase oxygen also, except that
commencement of adsorption, irrespective of the feed and intermediates will be much more clearly visible.

–
INTERACTION OF CHLORINATED ETHYLENES WITH Cr Y
19
Rationale for Carboxylate Assignment
etate type carboxylate stretching frequencies is a function
of number of chlorine atoms in the feed molecule. So it
is possible that this carboxylate formed is more complex
than simple acetate. According to Bellamy (15), in esters,
acids, and ketones the substitution of a chlorine atom on
the -carbon atom to the carbonyl group results in a small
shift in the COO (asym) to a higher value. Bellamy has
shown that in the vapor phase, ClCH2COOH absorbs at
1794 cm 1, while CH3COOH absorbs at 1785 cm 1. There-
fore, it is possible that the carboxylate formed from other
CEs is not completely devoid of chlorine.
As already seen, with the exception of VC, the adsorption
of all CEs at 300 C leads to the formation of carboxylate of
the acetate type. In addition, the adsorption of 1,1 DCE and
TCE at 100 C also yields the same carboxylate. Conversely,
the adsorption of VC at 100 C results in the appearance of
carboxylate of the formate type and at 300 C it yields CO32
and carboxylate of the CO2 type.
The assignment of a pair of bands at 1568–1591 and
1
1408–1444 cm to the carboxylate of the acetate type
deserves some explanation. This species is referred to as
acetate because the bands are similar to those observed
after the chemisorption of CH3COOH on Cr Y. Various
Reaction Pathway
–
researchers have previously found these bands at similar
As mentioned earlier, the adsorption of CEs was car-
positions. Kuznetsov et al. (24) found the CH3COOH ad- ried out at three different temperatures, viz., 25, 100 (130
1
sorption bands on Cr2O3 at 1435 and 1555 cm and as- for PCE), and 300 C to aid in detecting all the possible in-
signed them to COO (sym) and COO (asym), respec- termediates leading to the formation of complete oxidation
tively, of the acetate. Busca and Lorenzelli (25) detected products. The reaction scheme developed here assumesthat
1
the same bands at 1440 and 1540 cm on Fe2O3 after the complete oxidation reaction of CEs proceeds through
CH3COOH adsorption. Datka and Eischens (26) produced the formation of intermediates which are more easily de-
acetate by exposure of Pt/Al2O3 to C2H2 at 250 C and found tected at low temperatures and with less than stoichiomet-
1
the bands at 1460 and 1580 cm
.
ric oxygen present. At temperatures of 300 C and above, it
The conventional coke, which is polymeric carbon, shows is hypothesized that these intermediates are not observed
1
==
C
C (aromatic) at 1591 cm along with CH2 and CH3 because of their rapid decomposition into deeper oxida-
1
bending absorptions at 1458, 1394, and 1378 cm (26). tion products. Thus, the earliest intermediates would be
Since these four bands are not observed upon CE adsorp- observed at the lowest temperature and, as the tempera-
tion at 100 and 300 C, and the position of the bands ob- ture of adsorption is increased, these intermediates would
served is very close to that of acetate as reported by vari- undergo further reaction, forming deeper oxidation prod-
ous researchers and as seen by CH3COOH adsorption on ucts.
–
–
Cr Y, we assign them to the carboxylate of the acetate type.
At 25 C, only the interaction of VC with the Cr Y
Table 3 shows the acetate type carboxylate stretching fre- surface was observed to result in the formation of a
quencies of these CEs only at 100 and 300 C, since no such strongly adsorbed carbonyl compound. The dechlorina-
==
species is formed at 25 C.
tion of the molecule was also observed since the C C
VC is excluded from Table 3 since it forms carboxylate (1605 cm 1) and CH2 deformation (1366 cm 1) bands were
–
of the CO2 type and not acetate. Table 3 shows that there is seen, whereas C Cl was not seen in the adsorbed species.
a distinct trend in the value of COO (sym) and COO With adsorption of 1,1 DCE and TCE, no new species
(asym) with the chlorine content of the feed molecule. The were detected at 25 C, indicating the occurrence of physi-
1
COO (sym) decreases from 1445 cm for 1,1 DCE to cal adsorption only. The PCE adsorption at 25 C resulted in
1408 for PCE. Conversely, there is a small increase in the the partial dechlorination (but no oxygen addition) of the
1
==
value of COO (asym) from 1576 cm for 1,1 DCE to molecule as suggested by an increase in the C C stretch-
1
1591 cm for PCE. This suggests that the position of ac- ing frequency of adsorbed PCE as compared to the vapor
phase molecule.
At 100 C, the VC molecule interacted further with the
–
TABLE 3
Cr Y surface leadingto the appearance ofboth an adsorbed
carbonyl compound and carboxylate of the formate type.
At 100 C, 1,1 DCE reacted with the surface leading to the
detection of gas phase HCHO and small amounts of the
carboxylate of the acetate type. Similarly, the TCE interac-
tion resulted in the formation of CCl2HCHO and the same
carboxylate. For PCE, except for partial dechlorination, the
interaction at 130 C did not lead to the formation of any
oxygenated species.
FT-IR Assigned Carboxylate of the Acetate Type Stretching
Frequencies of CEs at Different Temperatures
100 C
300 C
COO (sym) COO (asym) COO (sym) COO (asym)
1
1
1
1
Feed
(cm
)
(cm
)
(cm
)
(cm )
1,1 DCE
TCE
PCE
1445
1418
none
1578
1584
none
1443
1445
1408
1576
1568
1591
At 300 C, VC adsorption led to the detection of car-
boxylate of the CO2 type and CO23 on the Cr Y surface.
–

20
CHINTAWAR AND GREENE
–
Similarly, 1,1 DCE interaction at this temperature resulted position of CEs on the Cr Y surface in the absence of gas
chiefly in the formation of carboxylate of the acetate type phase oxygen:
and small amounts of CO23 . Conversely, the TCE adsorp-
tion at 300 C led to the appearance of only carboxylate of
CE adsorption through chlorine abstraction
⇒ oxygen insertion ⇒ carbonyl formation
⇒ oxygen insertion ⇒ carboxylate formation
⇒ carbonate, carbon dioxide formation.
–
the acetate type. Similarly, PCE interacted with the Cr Y
forming small amounts of CH2ClCHO and carboxylate of
the acetate type.
It is postulated that the first step in CE interaction with
–
the Cr Y surface occurs through the abstraction of one or
A separate investigation carried out in this laboratory
has previously demonstrated that the above scheme is truly
catalytic in nature. In this investigation, the stability of
chromium exchanged zeolites was studied during complete
more chlorine atoms. The chlorine abstraction is evident
from the spectra obtained during VC and PCE adsorption
at 25 C.
The chlorine abstraction as the first step in chlorinated
hydrocarbon destruction has been proposed by other re-
searchers also. For example, Hatje et al. (27) have studied
1
oxidation of several CEs at 2400 h space velocity and
1% CE in excess air at 500 C for 51 h. The fresh catalyst
exhibited 92.5% TCE destruction efficiency which even-
tually dropped to 67.5% after aging. This experiment cor-
responded to 1270 turnovers/site, indicating a sustainable
catalytic reaction.
The chlorine abstraction as the first step, followed by the
scission of the remaining molecule, parallels the conclusion
drawn by Finocchio et al. (14). In this in situ FT-IR study
of oxidation of light hydrocarbons, it was proved that these
molecules are first activated at their weakest point (i.e., the
–
–
the interaction of C6H5Cl with Pt Y and Pd Y surfaces.
–
The formation of the Pt Cl complex during the reaction
was proved using XPS. Kovacs and Solymosi (28) studied
the thermal and photo-induced dissociation of C2H5I on
Pd(100) surface. The primary products of dissociation were
found to be C2H5 and I . It has been shown by Phillips
and Raupp (22) that TCE adsorbs dissociatively on the
surface of TiO2 at room temperature in the absence of
UV radiation. The chlorine of the TCE is abstracted by
a coordinatively unsaturated Ti cation, and the rest of the
dichloroethylene fragment is attached to the neighboring Ti
site.
At the intermediate temperature of 100 C, mainly an
oxygenated species (i.e., carbonyl) is observed. This indi-
catesthat the partiallydechlorinated CE molecule observed
after the first step is then attacked by oxygen to yield car-
bonyl species, which may or may not contain chlorine. In the
case of 1,1 DCE and TCE, the adsorption at 100 C led to
the formation of HCHO and CHCl2CHO species, respec-
tively. The carbonyl compounds formed are seen to contain
fewer chlorine atoms than the feed molecule, thereby rein-
forcing the hypothesis that chlorine abstraction is the first
step.
–
–
C H bond). In the present case of CEs, and C Cl bond is
–
weaker than the C H bond and hence gets activated first.
The final formation of CO2 is not seen here, probably
because of the lack of available oxygen. However, earlier
work in this laboratory has shown that CO2 is the only prod-
–
uct when air or oxygen is passed over Cr Y catalyst pellets
initially containing adsorbed carboxylate.
This scheme is also consistent with those proposed in the
literature by others (14, 29, 30). For example, Kuznetsov
et al. (29) have studied the interaction of hydrocarbons such
as C2H4, C3H6, and C4H8 with a Cr2O3 surface, known to
be a good complete oxidation catalyst. The oxidation of ad-
sorbed molecules by the oxygen of the catalyst was found to
take place. Surface reactions were reported to occur with
the formation of various oxygen containing species such
as carboxylate and carbonate which were bound to the
surface. These complexes underwent further oxidation at
higher temperatures to yield complete oxidation products
such as CO and CO2.
The formation of a carbonyl compound is invariably fol-
lowed at 300 C by the detection of carboxylate and car-
bonate species. Therefore, it is postulated that in the next
step(s), the carbonyl species is further attacked by the re-
active oxygen forming the highly oxygenated products.
Although the exact fate of chlorine abstracted in the first
Effect of the Chlorine Content of the Feed Molecule
step is not known, three distinct possibilities exist. The ab-
1
–
sence of H Cl band at 2885 cm in all the IR spectra
The data shown in Table 2 strongly suggest a sequential
reaction pathway which is not only a function of temper-
ature but also a function of chlorine content of the CEs.
Thus, VC forms significant deep oxidation products at 25
and 100 C while more highly chlorinated feeds require pro-
gressively higher temperatures to obtain the same degree
of oxidation.
suggests that adsorbed or gas phase HCl is probably not
formed. The formation of molecular chlorine (Cl2), an IR
invisible species, cannot be confirmed. It is hypothesized
that one or more chlorine atoms abstracted in the first step,
–
–
form a Cr Cl surface complex similar to the Pt Cl complex
as suggested by Hatje et al. (27).
From the observed results, the following reaction scheme
is proposed for the progressive catalytic oxidative decom-
The diminishingcatalyticactivitywith increasingchlorine
content of the feed molecule is probably due to the chloride

–
INTERACTION OF CHLORINATED ETHYLENES WITH Cr Y
21
poisoning of the catalyst surface. As proposed in the reac- 10. Mortland, M. M., and Boyd, S. A., Environ. Sci. Technol. 23, 223(1989).
11. Song, L., Freitas, J. E., and El-Sayed, M. A., J. Phys. Chem. 94, 1604
tion pathway, one or more chlorine atoms are abstracted
from the CE molecule during the first step of its interaction
(1990).
12. Chatterjee, S., Greene, H. L., and Park, Y. J., J. Catal. 138, 179 (1992).
13. Shimanouchi, T., “Tables of Molecular Vibrational Frequencies,”
National Bureau of Standards, Washington, 1972.
–
with the Cr Y surface. Thus, with increasing chlorine con-
tent of the feed molecule, the chloride concentration of the
surface is hypothesized to increase, which probably leads to 14. Finocchio, E., Busca, G., Lorenzelli, V., and Willey, R. J., J. Catal. 151,
204 (1995).
the observed drop in the catalytic activity. Similar conclu-
sions were drawn by Windawi and Wyatt (31) who studied
the complete oxidation of several halogenated VOCs on a
Pt-supported Al2O3 catalyst. Based on theoretical consid-
15. Bellamy, L. J., in “The Infrared Spectra of Complex Molecules,
Vol. 2,” 2nd ed. Chapman and Hall, New York, 1980.
16. Grabowski, R., Efremov, A. A., Davydov, A. A., and Haber, E., Kinet.
Katal. 22(4), 1014 (1981).
erations, the authors argued that the rate limiting step in 17. Kuznetsov, V. A., Geri, S. V., Gorokhovat-skii, Y. B., and Rozhkova,
E. V., Kinet. Katal. 18(2), 418 (1977).
18. Davydov, A. A., in “Infrared Spectroscopy of Adsorbed Species on
the oxidation of halogenated VOCs on this catalyst is the
removal of halide from the surface.
the Surface of Transition Metal Oxides,” p. 39, 1st ed. Wiley, New York,
1990.
19. Kiselev, V. F., and Krylov, O. V., “Adsorption and Catalysis on Tran-
ACKNOWLEDGMENTS
sition Metals and Their Oxides,” p. 241. Springer-Verlag, New York,
1989.
Partial funding for this research was obtained from the U.S. Environ-
mental Protection Agency, the U.S. Air Force, and SERDP is acknowl-
edged with appreciation. Zeolite samples were obtained from UOP. The
contents of this paper should not be construed to represent EPA policy.
20. Mortland, M. M., and Boyd, S. A., Environ. Sci. Technol. 23, 223
(1989).
21. Hughes, T. R., and White, H. M., J. Phys. Chem. 71(7), 2192 (1967).
22. Phillips, L. A., and Raupp, G. B., J. Mol. Catal. 77, 297 (1992).
23. Bellamy, L. J., and Williams, R. L., J. Chem. Soc., 3465 (1968).
24. Kuznetsov, V. A., Geri, S. V., Gorokhovat-skii, Y. B., and Rozhkova,
E. V., Kinet. Katal. 18(2), 418 (1977).
REFERENCES
1. Manning, M. P., Hazard.Waste 1(1), 41 (1984).
25. Busca, G., and Lorenzelli, V., J. Catal. 66, 28 (1980).
2. Welden, J., and Senken, S. M., Combust. Sci. Technol. 47, 229 (1986). 26. Datka, J., Sabak, Z., and Eischens, R. P., J. Catal. 145, 544 (1994).
3. Spivey, J. J., Ind. Eng. Chem. Res. 26(11), 2165 (1987).
4. Lester, G. R., “Proceedings, Air and Waste Management Association,
82nd Annual Meeting and Exposition, June 25–30, Anaheim, 1989.”
5. Pruden, A. L., and Ollis, D. F., J. Catal. 82, 404 (1983).
6. Dibble, L. A., and Raupp, G. B., Environ. Sci. Technol. 26, 492
(1992).
7. Kazanjian, A. R., and Horell, D. R., J. Phys. Chem. 75(5), 613
(1971).
8. Sutherland, J. W., and Spinks, J. W. T., Can. J. Chem. 37, 79 (1959).
9. Arnold, S. J., Kimbell, G. H., and Snelling, D. R., Can. J. Chem. 52,
2608 (1974).
27. Hatje, U., Hagelstein, M., and Fo¨rster, H., in “Zeolites and Re-
lated Microporous Materials: State of the Art 1994” (J. Weitkamp,
H. G. Karge, H. Pfeifer, and W. Ho¨lderich, Eds.), Vol. 84. Elsevier,
Amsterdam, 1995.
28. Kovacs, I., and Solymosi, F., J. Phys. Chem. 97, 11056 (1993).
29. Kuznetsov, V. A., Geri, S. V., Gorokhovat-skii, Y. B., Kinet. Katal.
18(3), 710 (1977).
30. Jiang, Z., Li, C., and Xin, Q., Cuihua Xuebao 14(3), 229 (1993).
31. Windawi, H., and Wyatt, M., Platinum Metals Rev. 37(4), 186
(1993).
32. Gullikson, C. W., and Nielsen, J. R., J. Mol. Spectrosc. 1, 158 (1957).