Effect of ultrasonic frequency on the mechanism of formic acid sonolysis

Article abstract of DOI:10.1021/jp109444h

The kinetics and mechanism of formic acid sonochemical degradation were studied at ultrasonic frequencies of 20, 200, and 607 kHz under argon atmosphere. Total yield of HCOOH sonochemical degradation increases approximately 6-8-fold when the frequency inc

Full text of DOI:10.1021/jp109444h

ARTICLE
pubs.acs.org/JPCB
Effect of Ultrasonic Frequency on the Mechanism of
Formic Acid Sonolysis
†
‡
†
†
‡,
Nathalie M. Navarro, Tony Chave, Patrick Pochon, Isabelle Bisel, and Sergey I. Nikitenko *
†
CEA/DEN/MAR/DRCP/SCPS/LPCP, CEA Centre de Marcoule, BP 17171, 30207 Bagnols sur C ꢀe ze Cedex, France
‡
Institut de Chimie S ꢁe parative de Marcoule, UMR 5257 ICSM Site de Marcoule, BP 17171, 30207 Bagnols sur C ꢀe ze Cedex, France
S Supporting Information
b
ABSTRACT: The kinetics and mechanism of formic acid sonochemical
degradation were studied at ultrasonic frequencies of 20, 200, and 607 kHz
under argon atmosphere. Total yield of HCOOH sonochemical degrada-
tion increases approximately 6-8-fold when the frequency increased from
2
0 to 200 or to 607 kHz. At low ultrasonic frequencies, HCOOH
•
degradation has been attributed to oxidation with OH radicals from
water sonolysis and to the HCOOH decarboxylation occurring at the
cavitation bubble-liquid interface. With high-frequency ultrasound,
the sonochemical reaction is also influenced by HCOOH dehydrat-
ion. Whatever the ultrasonic frequency, the sonolysis of HCOOH yielded
H and CO in the gas phase as well as trace amounts of oxalic acid
2
2
and formaldehyde in the liquid phase. However, CO and CH formations
4
were only detected under high-frequency ultrasound. The most striking
difference between low-frequency and high-frequency ultrasound is
that the sonolysis of HCOOH at high ultrasonic frequencies initiates
Fischer-Tropsch hydrogenation of carbon monoxide.
1. INTRODUCTION
compared to low-frequency ultrasound. For example, in pure water
saturated with argon, P ꢁe trier et al. showed that hydrogen peroxide
formation was faster at 514 kHz than at 20 kHz. Hua et al. found
Over the past decade, considerable interest has been expressed
6
in the application of advanced oxidation processes (AOP) to the
destruction of hazardous organic compounds in industrial waste
that the sonolytic production of hydrogen peroxide in water was
about 20-fold higher at 513 kHz than that at 20 kHz in the presence
1
streams. Among several different techniques (catalytic wet air
7
of argon. Furthermore, it was reported that the formation rate
oxidation, Fenton process, photocatalytic oxidation, ozonation),
sonochemical oxidation is considered to be promising in its
of hydrogen peroxide reached its maximum between 200 and
8-10
2
300 kHz.
The frequency’s effect on the organic compound
degradation of such pollutants. The ultrasonic irradiation of
degradation rate in aqueous solutions was found to be more
complicated and seemed to depend on the physical properties of
the molecules involved. For hydrophilic nonvolatile compounds
liquids generates cavitation bubbles, which produce transient
heat and highly reactive radical species during their implosive
collapse. In aqueous solutions, the homolytic split of H O
2
1
1-13
like phenol, the degradation rate is higher at high frequency
and reaches a maximum at 200 kHz,
molecules inside the cavitation bubbles leads to the formation
14,15
•
whereas for hydrophobic
of OH radicals, which can react with nonvolatile organic scaven-
3,4
volatile molecules like CCl , the reaction rates in general increase
gers in the liquid reaction zone surrounding the bubbles. Thus,
each cavitation bubble can be considered as a microreactor able
to destroy pollutants either by radical reactions or by pyrolysis in
bulk at ambient temperature and pressure.
4
14
with ultrasonic frequency. However, controversial results were
found by Hung et al. who reported an optimum frequency at
16
6
18 kHz for the sonolytic degradation of CCl4. Finally, the
degradation rate of hydrophilic volatile solutes like 1,4-dioxane
increases up to 358 kHz, and then decreases with higher ultrasonic
Sonochemical activity depends on several parameters includ-
ing the liquid’s properties (vapor pressure, viscosity, etc.), gas
atmosphere, temperature of sonicated solutions, and ultrasonic
17
frequencies.
3
Herein, we present a study of the effect of ultrasonic frequency
on formic acid sonolysis in aqueous solutions. Formic acid is a
intensity, and frequency. The latter parameter seems to be
important in the optimization of the sonochemical processes
5
both with volatile and nonvolatile reagents. However, the influ-
ence of ultrasonic frequency on sonochemical processes is as yet
poorly understood. Published results have indicated that ultrasonic
irradiation at high frequency enhanced the reaction kinetics
Received: October 1, 2010
Revised:
January 20, 2011
Published: February 14, 2011
r 2011 American Chemical Society
2024
dx.doi.org/10.1021/jp109444h
|J. Phys. Chem. B 2011, 115, 2024–2029

The Journal of Physical Chemistry B
ARTICLE
1
8
typical reaction intermediate in various AOP processes and it is
therefore important in understanding the mechanism of
HCOOH sonochemical degradation. This reaction has been
studied by several authors. Henglein et al. showed that at 300
kHz ultrasound, CO, CO , and H were the major sonolysis
Thermo Fisher). The multiple ion monitoring (MIM) mode was
employed to follow the evolution of H , CO , CO, and CH
4
2
2
during sonolysis. The yields of H , CO , and CO were deter-
2
2
mined after mass spectrometric data calibration with standard
H /Ar, CO /Ar, and CO/Ar gas mixtures (Messer). The
2
2
2
2
products and only small amounts of oxalic acid were identified in
statistical error for G(H ) and G(CO ) values was estimated
2 2
to be 10%. The accuracy of G(CO) measurements was some-
19
sonicated solutions of concentrated HCOOH. Haissinsky et al.
found that in diluted HCOOH solutions sonolytical formation of
what lower (∼20%) because of higher background signal from
20
CO was favored at pH <2. More recently, Gogate et al. reported
N admixtures. Water vapor in the outlet gas was trapped with
2
that the total rate of HCOOH sonolysis is higher at 590 kHz
molecular sieves (Aldrich, 3 Å) prior the mass spectrometric
analysis. The emission of CO, CO , and CH during sonolysis
was confirmed qualitatively using gaseous IR-spectroscopy
(Nicolet Magna-IRTM Spectrometer, model 750).
2.4. Liquid Phase Analysis. Water-soluble sonolysis pro-
ducts were analyzed with HPLC using a DIONEX Ultimate
3000 instrument. Carboxylic acids in sonicated solutions were
separated on a METACARB 67H column (Varian) pre-equili-
18,21
compared to that at 20-50 kHz.
However, the detailed
2
4
analysis of the sonolytical products at different frequencies was
not performed. In this work, the sonochemical degradation of
HCOOH was studied at 20, 200, and 607 kHz ultrasonic
frequencies with other operating parameters (temperature, gas,
etc.) strictly controlled. Furthermore, to elucidate the mechan-
ism of HCOOH sonolysis, gaseous, and water-soluble products
were analyzed under equivalent reactor conditions. We demon-
strated for the first time that an increase in ultrasonic frequency
causes not only a change in the total reaction rate but also leads to
a modification in the HCOOH sonolysis mechanism.
brated at 35 °C. The mobile phase was an 0.01 N H SO solution
2 4
-
1
(0.8 mL min ). Carboxylic acids were then measured using a
UV detector at 210 nm.
Formaldehyde was determined by DNPH (2,4-dinitro-
phenylhydrazine) derivatization followed by HPLC analysis with
a C18 column (Dionex) at 30 °C. The mobile phase was a
2. EXPERIMENTAL METHODS
-1
mixture of 50% water and 50% acetonitrile (1 mL min ).
Analysis was performed with UV detector at 360 nm. The
formaldehyde concentrations were estimated after a data calibra-
tion with different solutions prepared using a commercial formal-
dehyde solution (Aldrich, 36%) (Supporting Information).
Hydrogen peroxide was monitored by spectrophotometry with
2
.1. Materials. Formic acid (98%), hydrogen peroxide
(
30%), and other analytical grade chemicals were purchased
from Aldrich and used without further purification. Deionized
water (Milli-Q 18.2 MΩ cm) was used to prepare all aqueous
solutions. Argon at 99.999% purity was provided by Air Liquide.
-1
-1 22
2
.2. Reactor Setup and Procedure. Two kinds of ultrasonic
Ti(IV) in 1 M HNO (λ = 411 nm, ε = 707 cm M ). The
statistical error for H O formation rate was estimated to be 10%.
2 2
3
equipment were used in this study. The experiments at 20 kHz
were performed in a thermostatically controlled closed batch2
reactor containing 50 mL of solution, and equipped with a 1 cm
irradiating surface area titanium probe and piezoelectric trans-
ducer powered by a 20 kHz generator (750 W Sonics). The
probe was immersed reproducibly below the surface of the
sonicated solution.
The high-frequency reactor consisted of a thermostatted batch
reactor equipped with a piezoelectric transducer providing
00 kHz or 607 kHz ultrasound. The transducer was fitted at
the bottom of the reactor and connected to a generator with a
maximum electrical power of 125W (ELAC Nautik). The
volume of sonicated solution was 250 mL. Images of both
reactors used in this work are given in Supporting Information.
For all experiments, the solutions were sparged with argon
about 1 h before sonication and during the ultrasonic treatment
at a controlled rate of 90 mL min . The temperature in the
reactor during sonolysis was maintained at 20 °C with a Huber
Unistat Tango thermo-cryostat and measured by a thermocouple
immersed approximately 3 cm below the surface of solution. The
3. RESULTS AND DISCUSSION
3
.1. Water Sonolysis. The kinetics of H and H O forma-
2
2
2
tion during pure water sonolysis at selected ultrasonic frequen-
cies was studied during the first stage of work. Formation of
both products follows zero order kinetics at all studied ultra-
sonic frequencies. It is generally recognized that in argon
atmosphere H and H O are formed due to the homolytical
2
2
2
2
split of water molecules inside the cavitation bubbles, followed
•
•
by the mutual recombination of H and OH species inside
•
•
•
•
the bubble (H þ H ) and at the bubble-liquid (OH þ OH )
3
interface:
-
1
•
•
H O - ÞÞÞ f H þ OH
ð1Þ
ð2Þ
2
•
•
H þ H f H
2
-
1
•
•
acoustic power density, P (W mL ), transmitted to the solu3-
ac
OH þ OH f H O
ð3Þ
2
2
tion, was measured using a conventional thermal probe method.
The calibration procedure details are presented in the Support-
Table 1 shows the yields of H
and H
O
during water
2
2
2
ing Information. The P values at 20, 200, and 607 kHz were set
at 0.40, 0.18, and 0.26 W mL respectively. To compare the
sonolysis obtained at different ultrasonic frequencies. These
ac
-
1
6
results confirm published data that G(H
high-frequency ultrasound than at low frequencies. For example,
Table 1 demonstrates that at 607 kHz the G(H ) value is
almost seven times higher than that obtained at 20 kHz.
Surprisingly, the observed G(H ) at 20 kHz is lower than
that of H obtained at the same conditions. To the best of our
O ) is greater under
2 2
efficiency of the sonochemical processes at different frequencies,
-
1
the sonochemical yields of the products (G, μmol kJ ) were
O
2 2
applied instead of widely used sonochemical reaction rates (R, M
-
1
min ) because G values are independent from the volume of
treated solutions and Pac.
O
2 2
2
2
.3. Gas Phase Analysis. Gaseous products of sonolysis were
knowledge, this is the first observation of such a phenomenon.
In contrast, at high-frequency ultrasounds the yields of both
identified using a quadrupole mass spectrometer (PROLAB 300,
2
025
dx.doi.org/10.1021/jp109444h |J. Phys. Chem. B 2011, 115, 2024–2029

The Journal of Physical Chemistry B
ARTICLE
-
1
Table 1. Effect of Ultrasonic Frequency on G Values for
H O and H during Water Sonolysis under Argon at 20°C,
Table 2. Yield of H O (μmol kJ ) during Formic Acid
2 2
Sonolysis in Argon at 20°C
2
2
2
σ = (10%
HCOOH (mol L-
1
)
G (μmol kJ-
1
)
frequency (kHz)
water
0.1
1.0
3.0
frequency (kHz)
H
2
O
2
H
2
2
0
0.06
0.28
0.46
0.06
0.17
0.32
0.02
0.07
0.15
0.01
0.04
0.05
2
0
0.06
0.05
0.28
0.23
0.46
0.12
-
200
607
a
2
0
2
00
0.28
-
a
5
14
07
Ref 6; 24 °C, Ar.
6
0.49
a
Figure 2. Evolution of HCHO concentration with sonication time during
-1,
HCOOH sonolysis [HCOOH] = 3M, 20 °C, Ar; P =0.40WmL 0.18 W
ac
-1
-1
mL and 0.26 W mL for 20, 200, and 607 kHz, respectively.
The reactions 4 and 5 can be summarized as:
Figure 1. Effect of ultrasonic frequency (20 and 200 kHz) on the
sonolysis of H O mol L under argon, 20 °C.
2 2
HCOOH þ H2O2 f CO2 þ 2H2O
ð6Þ
-
1
From reaction 6 it follows that, for the radical mechanism,
G(CO ) should not be higher than G(H O ) in pure water.
2
2 2
-
5
-4
-1
products are fairly similar, which fits well with the reaction
In the liquid phase, small amounts (∼10 to 10 mol L
scheme (1-3). It has been suggested that H O can be degraded
after 3 h of sonolysis) of oxalic acid were identified at all
2
2
at low ultrasonic frequency. To verify this hypothesis, we
ultrasonic frequencies and for all concentrations of formic acid,
-
3
19
performed experiments on the sonolysis of 10 M H O2
which is in agreement with Henglein’s work. It is likely that
2
•
aqueous solutions at 20 and 200 kHz in the presence of argon.
H C O is formed due to the HCOO radical recombination at
2
2 4
19
Figure 1 clearly shows a difference in H O behavior between
the bubble liquid interface:
2
2
2
0 and 200 kHz ultrasound. During sonication at 20 kHz, the
•
2
HCOO f H2C2O4
ð7Þ
concentration of H O decreases, whereas at 200 kHz an increase
2
2
in H O concentration can be observed. The degradation of
2
2
Surprisingly, formaldehyde was also detected in the sonicated
HCOOH solutions. A clear signal was detected by means of
HPLC with UV detector at 360 nm after sonicated sample pre-
treatment with DNPH. Figure 2 shows the linear increase of
HCHO concentration with sonication time. After 3 h of sonolysis
at high frequency, the HCHO concentration reached approxi-
H O at low-frequency ultrasound is most probably related to its
2
2
catalytic decomposition on titanium particles coming from the
cavitation erosion of the titanium probe. A control experiment
under mechanical stirring in the presence of titanium powder
(
Supporting Information) showed relatively rapid H O decom-
2 2
-
4
-1
position, confirming this hypothesis. The cavitation erosion is
mately (2.5-3.7) ꢀ 10 mol L , whereas at 20 kHz its con-
centration was at least 10 times lower. To the best of our
knowledge, this is the first observation of HCHO formation
during sonolysis of carboxylic acid solutions. A probable me-
chanism for its formation will be discussed later.
known to be much stronger at low ultrasonic frequency com-
23,24
pared to high-frequency ultrasound.
Therefore, the catalytic
decomposition of H O would be expected to be more effective
2
2
at 20 kHz than under high-frequency ultrasound.
.2. HCOOH Sonolysis. Table 2 shows the G(H O ) values
3
2
2
The mass spectrometric analysis of the outlet gas revealed the
as a function of HCOOH concentration at the ultrasonic
frequencies studied. It can be seen that whatever the ultrasonic
frequency, the yield of hydrogen peroxide decreases with in-
creasing HCOOH concentration, indicating effective scavenging
formation of CO, CO , and H as the principal gaseous products
2
2
of HCOOH sonolysis (Figure 3), in agreement with previously
19,20
published data.
Figure 3 clearly shows a more rapid HCOOH
sonolysis under high-frequency ultrasound compared to that at
•
of OH radicals:
2
0 kHz.
The G(CO ) values increased with ultrasonic frequency and
2
•
•
HCOOH þ OH f HCOO þ H2O
ð4Þ
ð5Þ
HCOOH concentration. In diluted HCOOH (0.1 M), the yield
of H O was comparable with the value obtained during pure
2
2
•
•
HCOO þ OH f CO þ H O
2
2
water sonolysis, indicating that the formation of CO via radical
2
2
026
dx.doi.org/10.1021/jp109444h |J. Phys. Chem. B 2011, 115, 2024–2029

The Journal of Physical Chemistry B
ARTICLE
Table 3. Total Yield of HCOOH Sonolysis G(HCOOH) =
G(CO) þ G(CO ) and the Yield Ratio G(CO)/G(CO ) as a
2
2
Function of Ultrasonic Frequency and HCOOH Concentra-
tion; Ar, 20°C, σ=(30%
frequency
kHz)
[HCOOH]
G(HCOOH)
G (CO) /
G(CO
1
(μmol kJ-1
(
(mol L-
)
)
2
)
2
0
0.1
1
0.1
0.1
0.2
0.1
0.6
1.4
0.7
1.3
1.8
below the quantification
limit
3
2
6
00
07
0.1
1
1.1
2.0
1.5
1.5
1.7
3
0.1
1
3
HCOOH - ÞÞÞ f CO þ H O
ð9Þ
2
The yield of CO is extremely sensitive to ultrasonic frequency.
For 20 kHz, figure 3 reveals an amount of CO in the outlet gases
below the detection limits in all the HCOOH concentrations
studied. At 200 kHz, the formation of CO was always observed
except in 0.1 M HCOOH solution. For more concentrated
solutions, the yield of CO is comparable to that of CO and
2
increased with the HCOOH concentration. At 607 kHz, the
formation of CO was observed for all HCOOH concentrations.
The G(CO) values increased with the HCOOH concentration,
similarly to observations at 200 kHz.
Table 3 demonstrates that under high-frequency ultrasound,
the ratio of the sonochemical yields of CO and CO increased
2
with HCOOH concentration from ∼1.1 (1 M HCOOH) to
∼
2.0 (3.0 M HCOOH). This ratio is much less than that
reported for gas-phase HCOOH thermolysis ([CO]/[CO ]
2
Figure 3. Yields of H
2
, CO
2
, and CO during HCOOH sonolysis at 20,
27
∼
10). Therefore, the G(CO)/G(CO ) ratio observed in our
2
200, and 607 kHz under argon atmosphere, 20 °C.
experiments confirms the assumption that HCOOH sonolysis
occurs in the liquid reaction zone rather than in the gas phase of
the cavitation bubble. Several studies have shown that H O
molecules catalyze HCOOH decarboxylation under hydrother-
reaction with hydroxyl radical (reaction 5) is insignificant. The
2
formation of CO can be explained by HCOOH decarboxyla-
2
mal conditions, providing a higher yield of CO compared to that
tion:
2
28-30
of CO.
Thus, presuming HCOOH sonochemical degrada-
HCOOH - ÞÞÞ f CO2 þ H2
ð8Þ
tion in the liquid phase, the G(CO)/G(CO ) ratio should be
2
smaller than 1. However, Table 3 indicates a clearly higher yield
for CO observed in sonochemical experiments. Furthermore, the
relatively high influence of the decarboxylation reaction pathway
In 1-3 M HCOOH, the yield of H O decreases, and the
2
2
contribution of radical reactions becomes more important.
However, in contrast to eq 6, the G(CO ) values are higher
2
in HCOOH sonochemical degradation as well as H formation
2
than G(H O ) obtained during pure water sonolysis, indicating
2
2
during water sonolysis supposes observed G(H ) values higher
2
that degradation by radical reactions and by HCOOH decarbox-
ylation occurred simultaneously.
than G(H ) values obtained during pure water sonolysis. This is
2
true for low-frequency ultrasound (Figure 3). However, Figure 3
also shows that under high-frequency ultrasound the G(H2)
value is lower than that in pure water for all concentrations of
HCOOH. These contradictions between observed and expected
reaction yields can be understood considering the secondary gas
phase sonochemical processes initiated under high-frequency
Partial vapor pressure of HCOOH in relatively diluted aqu-
eous solutions (0.1-3.0 M) is known to be lower than that of
25,26
water at room temperature.
Consequently, the gas phase of
the cavitation bubble would be depleted on HCOOH molecules
under our conditions, and reaction 8 should occur preferably in
the liquid reaction zone surrounding the cavitation bubble. The
temperature at the interface is indeed high enough to consider a
ultrasound. It has been reported that CO can be reduced
2
sonochemically to CO in water due to the reaction with
4
hydrothermal decomposition of formic acid in this region. It is
31,32
H coming from sonochemical water split:
2
well known that there are two reaction channels in the thermal
decomposition of formic acid: HCOOH decarboxylation
CO2 þ H2 - ÞÞÞ f CO þ H2O
ð10Þ
(reaction 8) and the dehydratation reaction that produces carbon
monoxide:
In our system, H from water sonolysis and decarboxylation
2
2
027
dx.doi.org/10.1021/jp109444h |J. Phys. Chem. B 2011, 115, 2024–2029

The Journal of Physical Chemistry B
ARTICLE
indicates a local temperature much higher than 300 °C during
sonochemical hydrogenation of CO, according to thermody-
34
namic data.
The total yield of HCOOH sonochemical degradation can be
expressed as:
GðHCOOHÞ ¼ GðCOÞ þ GðCO2Þ þ GðCH4Þ
1
þ
GðH2C2O4Þ þ GðH2COÞ
ð13Þ
2
1
Presuming G(CO) þ G(CO ) . G(CH ) þ / G(H C O )
2
4
2
2 2 4
þ G(H CO), eq 13 can be simplified as: G(HCOOH) = G(CO)
2
þ G(CO ). Table 3 summarizes G(HCOOH) values at different
2
ultrasonic frequencies. Generally speaking, Table 3 reveals that
the sonolysis of HCOOH is approximately 6-8 times more
effective under high-frequency ultrasound than under 20 kHz
ultrasound. Several reasons can be given for such a significant
difference. First, the distribution of the cavitation field is different
Figure 4. IR-spectrum of the outlet gas obtained after 1 h of 3 M
HCOOH sonolysis, 20 °C, 200 kHz, argon.
35
between low- and high-frequency ultrasound. A 20 kHz horn
system gives a limited conical cavitation zone near the horn tip
while high frequencies (f g 200 kHz) tend to give a more diffuse,
widely distributed zone of cavitation. It can be assumed that a
much larger number of cavitation bubbles can be generated
under high-frequency ultrasound in a treated volume of liquid at
any time for the same specific absorbed acoustic power. Conse-
quently, at high frequency a larger volume of solution is directly
submitted to ultrasonic irradiation compared to 20 kHz ultra-
sound. Second, the resonance size of a single cavitation bubble is
reaction 8 reacted with CO formed after the radical oxidation
2
(
4-5) and decarboxylation (8) of HCOOH. Note here that CO2
is more soluble in water than Ar and thus can accumulate in the
3
2
reaction medium. At 20 kHz ultrasound, the amount of CO2
produced is low and, consequently, the amount of CO formed in
the secondary processes is also insignificant. At low-frequency
ultrasound, the H formed does not participate in any secondary
2
reactions and its yield in HCOOH solutions is higher than in
pure water. However, under high-frequency ultrasound the
17
inversely correlated to the ultrasonic frequency:
concentration of CO is sufficient to provide significant con-
2
sumption of H2.
3
Fωr2
KP0
Rr2
¼
ð14Þ
The mass spectra of the outlet gas at high-frequency sonolysis
revealed the appearance of a signal with m/z = 16 simultaneously
with the CO signal (m/z = 28) (Supporting Information). This
signal could be assigned to methane formation. Indeed, the FTIR
2
where Rr is the resonant bubble radius, K = C /C is the
polytropic index, P is the hydrostatic pressure, F is the density
of the liquid, and ω is the radial resonant frequency. As the
p
v
0
spectra of the outlet gas collected under these conditions
Figure 4) exhibited a peak at 2900 cm- typical for C-H
1
r
(
frequency increases, the resonant radius of the cavitation bubbles
decreases. Smaller bubbles require fewer acoustic cycles before
reaching their resonant size. This leads to cavitation events
occurring at a faster rate at higher ultrasonic frequency. A
decrease in the bubble size causes a significant increase in its
surface area to volume ratio, SA/V. For example, the SA/V ratio
stretching vibrations in CH molecules. This result clearly points
4
to methane formation during the sonolysis of HCOOH under
high-frequency ultrasound, which is in agreement with the recent
finding of Harada who reported the sonochemical reduction of
CO to CH in water in the presence of a CO -H gas
2
4
2
2
3
3
mixture. The FTIR spectra also indicate the presence of CO
-1
increases from 0.02 to 0.52 μm when the ultrasonic frequency
and CO in the outlet gas. Under our conditions, the formation of
2
increases from 20 to 607 kHz. In other words, the relative
effective volume of the liquid overheated shell for each cavitation
bubble rises considerably as frequency increases. Both factors,
total volume of the cavitation zone and single bubble SA/V ratio,
are very important for the sonochemical reactions occurring at
bubble/liquid interface, as in the HCOOH sonolysis studied in
this work.
The ultrasonic frequency also could influence the intrabubble
conditions. Until recently, it was thought that low-frequency
ultrasound primarily generated transient cavitation and high-
frequency ultrasound primarily generates stable cavitation. How-
ever, recent studies of multibubble sonoluminescence and so-
nochemistry have revealed the existence of transient cavitation at
CH can be explained by the secondary CO hydrogenation inside
4
the cavitation bubble, similar to the well-known Fischer-
Tropsch synthesis:
3
4
CO þ 3H2 - ÞÞÞ f CH4 þ H2O
ð11Þ
This reaction is not observed under low-frequency ultrasound
because of the negligibly low yield of CO at all HCOOH
concentrations studied. In view of the proposed hydrogenation
mechanism, the formation of formaldehyde could be considered
as an intermediate stage of CH formation:
4
CO þ H - ÞÞÞ f H CO
ð12Þ
2
2
36
Formation of oxygenated impurities during the catalytic
hydrogenation of CO has been reported in the presence of large
amounts of water vapor. It should be noted here that Fischer-
Tropsch synthesis under silent conditions needs catalysts. In our
experiments, CO hydrogenation occurred due to the extreme
local conditions created by acoustic cavitation inside the cavita-
tion bubbles. The absence of hydrocarbons heavier than CH4
high ultrasonic frequency. Moreover, it has been shown very
recently that the increase of ultrasonic frequency causes a
population of electronically OH(C Σ ) and vibrationally OH-
34
2
þ
2
þ
(A Σ , ν
) excited radicals during water sonolysis in the pre-
37
1
sence of noble gases. Potentially these species would have
3
7
advanced oxidizing properties. Further studies are required
to elucidate the impact of ultrasonic frequency on extreme
2
028
dx.doi.org/10.1021/jp109444h |J. Phys. Chem. B 2011, 115, 2024–2029

The Journal of Physical Chemistry B
ARTICLE
conditions created by acoustic bubble collapse and related
sonochemical activity.
(6) Petrier, C.; Jeunet, A.; Luche, J. L.; Reverdy, G. J. Am. Chem. Soc.
1992, 114, 3148–3150.
(
7) Hua, I.; Hoffmann, M. R. Environ. Sci. Technol. 1997, 31, 2237–
243.
8) Mendez-Arriaga, F.; Torres-Palma, R. A.; Petrier, C.; Esplugas,
2
4. CONCLUSIONS
(
Ultrasonic frequency is a critical parameter in the sonolysis of
S.; Gimenez, J.; Pulgarin, C. Water Res. 2008, 42, 4243–4248.
(9) Jiang, Y.; Petrier, C.; Waite, T. D. Ultrasonics Sonochemistry 2006,
formic acid in aqueous solutions. The total yield of 0.1-3.0 M
HCOOH sonochemical degradation increases approximately 6-
to 8-fold when the frequency increases from 20 kHz to 200 or
1
3, 415–422.
10) Koda, S.; Kimura, T.; Kondo, T.; Mitome, H. Ultrasonics
Sonochemistry 2003, 10, 149–156.
11) Petrier, C.; Lamy, M. F.; Francony, A.; Benahcene, A.; David,
B.; Renaudin, V.; Gondrexon, N. J. Phys. Chem. 1994, 98 (41), 10514–
0520.
12) Berlan, J.; Trabelsi, F.; Delmas, H.; Wilhelm, A. M.; Petrignani,
(
6
07 kHz under argon atmosphere. In agreement with the
(
literature, CO , CO, H , and oxalic acid are the main products
2
2
of HCOOH sonolysis. Moreover for the first time, formaldehyde
and methane were detected. In a diluted solution (0.1 M
HCOOH), sonochemical degradation occurs mostly by HCOOH
decarboxylation in the liquid reaction zone at the bubble/solution
interface for all ultrasonic frequencies studied. When the HCOOH
concentration becomes higher (1.0-3.0 M), degradation also
1
(
J. F. Ultrasonics Sonochemistry 1994, 1, S97–S102.
(13) Chen, J. W.; Chang, J. A.; Smith, G. V. Chem. Eng. Prog. S. Ser.
1971, 67, 18–26.
(14) Petrier, C.; Francony, A. Water Sci. Technol. 1997, 35, 175–180.
•
(
15) Okouchi, S.; Nojima, O.; Arai, T. Water Sci. Technol. 1992, 26,
053–2056.
16) Hung, H. M.; R. Hoffmann, M. R. J. Phys. Chem A 1999, 103,
734–2739.
occurs via simultaneous OH radical scavenging. Under high-
2
2
frequency ultrasound the reaction is also influenced by HCOOH
dehydration. The most striking feature of this work is that the high-
frequency ultrasonic treatment initiated secondary hydrogenation
(
(
(
17) Becket, M. A.; Hua, I. J. Phys. Chem. A 2001, 105, 3796–3802.
18) Gogate, P. R.; Mujumdar, S.; Pandit, A. B. Adv. Environ. Res.
reactions of CO and CO. Under high ultrasonic frequency CO2
2
reacted with H to produce CO. Thereby, at 200 and 607 kHz CO
2
2
003, 7, 283–299.
is the major product and the yield of H is even lower than that
2
(19) Hart, E. J.; Henglein, A. Radiat. Phys. Chem. 1988, 32, 11–13.
obtained during pure water sonolysis. Sonochemically driven
Fischer-Tropsch hydrogenation of CO led to the formation of
methane and formaldehyde as byproduct of HCOOH degradation.
Finally, we found catalytic decomposition of sonochemically
formed H O on the surface of titanium particles formed during
cavitation erosion of the ultrasonic probe under low-frequency
ultrasound. This finding is important for the proper interpretation
of sonochemical experiments.
(20) Haïssinsky, M.; Klein, R. J. Chim. Phys. Phys. Chim. Biol. 1968,
65, 336–344.
(21) Gogate, P. R.; Pandit, A. B.; Wilhelm, A. M.; Ratsimba, B.;
Delmas, H. Water Res. 2006, 40, 1697–1705.
(
22) Nikitenko, S. I.; Le Naour, C.; Moisy, P. Ultrasonics Sonochem-
istry 2007, 14, 330–336.
23) Karimi, A.; Martin, J. L. Cavitation Erosion of Materials. Int.
Met. Rev. 1986, 31, 1–26.
24) Fatjo, G. A.; P ꢁe rez, A. T.; Hadfield, M. Ultrasonics Sonochemistry
011, 18, 73–79.
2
2
(
(
2
’
ASSOCIATED CONTENT
(25) Ito, T.; Yoshida, F. J. Chem. Eng. Data 1963, 8, 315–320.
(
26) Conti, J. J.; Othmer, D. F.; Gilmont, R. J. Chem. Eng. Data 1960,
S
Supporting Information. Additional figures and tables,
b
5, 301–307.
with a more detailed description of the materials and methods
used for the study. This material is available free of charge via the
Internet at http://pubs.acs.org.
(27) Saito, K.; Kakumoto, T.; Kuroda, H.; Torii, S.; Imamura, A.
J. Chem. Phys. 1984, 80, 4989–4994.
(28) Yu, J.; Savage, P. E. Ind. Eng. Chem. Res. 1998, 37, 2–10.
(
29) Chen, H. T.; Chang, J. G.; Chen, H. L. J. Phys. Chem. A 2008,
12, 8093–8099.
30) Bjerre, A. B.; Sorensen, E. Ind. Eng. Chem. Res. 1992, 31, 1574–
1577.
1
’
AUTHOR INFORMATION
(
Corresponding Author
(
(
(
31) Henglein, A. Z. Naturforsch. 1985, 40b, 100–107.
32) Harada, H. Ultrasonics Sonochemistry 1998, 5, 73–77.
33) Harada, H.; Hosoki, C.; Ishikane, M. J. J. Photochem. Photobiol.
*
E-mail: serguei.nikitenko@cea.fr. Phone: þ33(0) 466 339 251.
Fax: þ33(0) 466 796 027.
A 2003, 160, 11–17.
34) Anderson, R. B. The Fischer-Tropsch Synthesis; Academic
Press, Inc.: New York, 1984.
35) Price, G. J.; Harris, N. K.; Stewart, A. J. Ultrasonics Sonochemistry
010, 17, 30–33.
36) Ashokkumar, M.; Lee, J.; Iida, Y.; Yasui, K.; Kozuka, T.; Tuziuti,
T.; Towata, A. Phys. Chem. Chem. Phys. 2009, 11, 10118–10121.
37) Pflieger, R.; Brau, H.-P.; Nikitenko, S. I. Chem.—Eur. J. 2010,
6, 11801–11803.
’
ACKNOWLEDGMENT
(
N. Navarro thanks CEA/DEN/MAR/DRCP/SCPS for finan-
(
cial support of her PhD thesis. The authors are grateful to
J.P Donnarel and D. Sans for their help in HPLC and FTIR
analysis.
2
(
(
’
REFERENCES
1
(1) Andreozzi, R.; Caprio, V; Insola, A.; Marotta, R Catal. Today
1
999, 53, 51–59.
(
(
2) Adewuyi, Y. G. Ind. Eng. Chem. Res. 2001, 40, 4681–4715.
3) Mason, T. J.; Lorimer; J; P. Applied Sonochemistry. The Uses of
Power Ultrasound in Chemistry and Processing; Wiley-VCH: Weinheim,
2
1
2
002.
(
4) Suslick, K. S.; Hammerton, D. A.; Cline, R. E. J. Am. Chem. Soc.
986, 108, 5641–5642.
(
5) Petrier, C.; Francony, A. Ultrasonics Sonochemistry 1997, 4,
95–300.
2
029
dx.doi.org/10.1021/jp109444h |J. Phys. Chem. B 2011, 115, 2024–2029