Kinetics and mechanism of the enhanced reductive degradation of nitrobenzene by elemental iron in the presence of ultrasound

Article abstract of DOI:10.1021/es990385p

The degradation of nitrobenzene (NB) and aniline (AN) were facilitated by using sonolysis, reduction by Fe⁰ and a combination of the two processes. The rates of NB reduction by Fe⁰ increased in the presence of ultrasound. The first-order rate constant for NB degradation by ultrasound was 1.8 x 10^-3/min. The rate was considerably faster in the presence of Fe⁰. The similar degradation rates for AN in each system indicated that the sonication process was unaffected by the presence of Fe⁰. The rate enhancements for the NB degradation can be explained chiefly by the continuous cleaning and chemical activation of the Fe⁰ surfaces by acoustic cavitation and to accelerated mass transport rates of reactants, intermediates and products between the solution phase and the Fe⁰ surface. The relative concentrations of nitrosobenzene and AN, the main reaction intermediates produced by Fe⁰ reduction, were significantly changed in the presence of ultrasound.

Full text of DOI:10.1021/es990385p

Environ. Sci. Technol. 2000, 34, 1758-1763
nitrobenzene and the square-root of mixing rate (rpm)
Kinetics and Mechanism of the
Enhanced Reductive Degradation of
Nitrobenzene by Elemental Iron in
the Presence of Ultrasound
indicated that the observed reaction rates were controlled
0
by mass transfer of the substrates to the Fe surfaces (14).
Since acoustic cavitation is known to increase the surface
area of reactive solids by causing particles to rupture, Reinhart
et al. (15) investigated the dechlorination of trichloroethylene
0
by Fe
16), the kinetics and mechanism of carbon tetracholoride
in the presence of ultrasound. In our previous paper
(
0
degradation by Fe in the presence of ultrasound was
reported, and it was noted that the rate of CCl degradation
H U I - M I N G H U N G , F R A N K H . L I N G , A N D
M I C H A E L R . H O F F M A N N *
4
was enhanced by a factor of 40 compared to the same reaction
system in the absence of ultrasound.
W. M. Keck Laboratories, California Institute of Technology,
Pasadena, California 91125
In this study, we investigate the combination of ultrasound
0
and Fe for the reduction of nitrobenzene and aniline, the
0
byproduct from the nitrobenzene reduction by Fe . The
influence of ultrasonic power density is also discussed.
0
Sonolysis, reduction by elemental iron (Fe ), and a
combination of the two processes were used to facilitate
the degradation of nitrobenzene (NB) and aniline (AN)
in water. The rates of reduction of nitrobenzene by Fe are
enhanced in the presence of ultrasound. The first-order
rate constant, kUS, for nitrobenzene degradation by ultrasound
Technical Background
0
Mechanism of Nitrobenzene Sonolysis. The sonolytic deg-
radation of nitrobenzene has been studied by Weavers et al.
(17). Pyrolysis and hydroxylation are the main pathways for
the sonolytic degradation of nitrobenzene as follows where
-
3
-1
0
is 1.8 × 10 min , while in the presence of Fe , the
rate was found to be substantially faster. The observation
of similar degradation rates for aniline in each system
suggests that the sonication process was not affected by
′
)))′ refers to the application of ultrasound and [O] refers to
2 2
the ‚OH radical or other oxidants (e.g., H O ) produced by
ultrasound:
0
the presence of Fe . The observed rate enhancements
for the degradation of nitrobenzene can be attributed primarily
to the continuous cleaning and chemical activation of
0
the Fe surfaces by acoustic cavitation and to accelerated
mass transport rates of reactants, intermediates, and
0
products between the solution phase and the Fe surface.
The relative concentrations of nitrosobenzene and
aniline, the principal reaction intermediates generated by
0
Fe reduction, are altered substantially in the presence
of ultrasound.
Introduction
Ultrasonic irradiation has been applied as an advanced
oxidation technology for water treatment (1, 2). The chemical
effects of ultrasound are due to the phenomenon of acoustic
cavitation (3-5). Temperatures inside of the cavitation bubble
are on the order of 5000 K (4), and pressures of the order of
1
000 bar have been calculated (4). In homogeneous reactions,
the destruction of organic compounds occurs inside the
caviation bubbles by a pyrolysis reaction or in the nearby
interfacial region by hydroxylation (3, 5). In heterogeneous
reactions, the primary mode of action by sonication is erosion
involving the removal of oxide layers and impurities together
with pitting of metal surfaces. It is thought that sonication
serves to sweep reactive intermediates or products from metal
surfaces thereby reactivating and cleaning the surfaces for
subsequent reactions (3, 5).
0
The formation of m-nitrophenol (MNP), p-nitrophenol
PNP), and 4-nitrocatechol (4-NC) was observed during
Elemental iron (Fe ) has been proposed as a suitable
(
source of electrons for the in situ remediation of contami-
sonolysis of nitrobenzene in water at 20 kHz (17). However,
at 500 kHz, o-nitrophenol (ONP) was also observed as an
intermediate. With continuous irradiation these intermedi-
nated groundwater (6-14). Tratnyek and co-workers (7-10)
0
have used Fe to reduce chlorinated hydrocarbons to non-
chlorinated products and to reduce nitroaromatic com-
pounds (NACs) to anilines (13, 14). A linear correlation
between the observed rate constants for the reduction of
ates are further degraded to inorganic products, e.g., HNO
CO , and H O.
Mechanism of Nitrobenzene Reduction by Fe . According
3
,
2
2
0
to Agrawal and Tratnyek (14), the pathway for nitrobenzene
*
Corresponding author phone: (626)395-4391; fax: (626)395-3170;
e-mail: mrh@its.caltech.edu.
reduction by elemental iron can be described by the following
1
7 5 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 9, 2000
10.1021/es990385p CCC: $19.00

2000 Am erican Chem ical Society
Published on Web 03/23/2000

stoichiometric reactions
TABLE 1. HPLC Retention Times and Specific Topic Data for
Nitrobenzene (NB) and Its Sonolysis Reaction Intermediatesa
av retention
time, min
E, M- cm-1
1
λ, nm
6
NB
NSB
AN
ONP
MNP
PNP
267
307
224
224
224
318
224
∼4.5
∼5.5
∼2.0
∼1.5
∼2.4
∼2.2
∼3.7
6.36 × 10
6
6.60 × 10
6
4.85 × 10
6
4.56 × 10
6
7.37 × 10
6
8.53 × 10
6
4
-NC
5.71 × 10
a
NB ) nitrobenzene; NSB ) nitrosobenzene; AN ) aniline; ONP )
o-nitrophenol; MNP ) m -nitrophenol; PNP ) p-nitrophenol; and 4-NC
4-nitrocatechol.
)
solution. The reactor was made gastight with two O-ring
seals in a threaded Teflon collar that connects the glass cell
to the stainless steel probe. In addition, sampling ports were
sealed with Teflon valves and covered with rubber septa.
The ultrasonic power output was measured by calorimetry.
Chem ical Analyses. Sample aliquots of 0.5 mL were
collected at designated times and filtered through Teflon
syringe filters (Gelman Acrodisc) into amber colored liquid
chromatography vials. Quantification of nitrobenzene and
its intermediates was performed on a Hewlett-Packard Series
II 1090 Liquid Chromatograph with a 3 µm 100 × 4 mm
Hypersil BDS-C18 column (Hewlett-Packard). The method
where the overall stoichiometry is given by
Based on the overall reaction depicted in eq 9, the primary
product observed after several hours of reaction is aniline
with nitrosobenzene found as the only intermediate (eq 6).
-1
called for a 1 mL min flow rate, 40 °C temperature, and a
2
5 µL injection volume. The runs were 6 min long with an
isocratic mixture (70:30) of water to acetonitrile (EM Science,
HPLC grade). The postrun time was 1 min.
Experimental Procedures
High-purity reagents, nitrobenzene (Aldrich, 99%), ni-
trosobenzene (Fluka, 99%), and aniline (Fluka) were used
without further purification. Standard solutions of nitroben-
zene and aniline were made with water purified by a Millipore
Milli-Q UV Plus system (18.2 MΩ-cm resistivity). For ni-
trosobenzene, the stock solution was prepared with 50%
acetonitrile (EM Science, GR). The initial reaction volume
was 0.285 L, and the initial nitrobenzene concentration was
Results and Discussion
Experiments were performed at an initial nitrobenzene
concentration of 25 µM and an initial pH of 6 without an
added buffer. The loss of nitrobenzene was found to be less
than 5% in control experiments in the absence of ultrasound
and iron. After completion of each kinetic run with ultrasound
alone, the final observed pH was close to 5.5. With Fe alone
the final observed pH was close to 7, and with the combina-
tion of ultrasound and Fe , the observed pH was near 8. In
the reactions involving Fe , a rise in pH is consistent with the
0
2
5 µM. Solutions were not buffered.
Zerovalent iron particles (Fluka) were prepared by sieving
0
raw particles through a 20-mesh screen (1.0 mm opening
size). The elemental iron was not pretreated with acid prior
0
following stoichiometries:
0
to use. The degradation reactions with Fe were carried out
in tightly closed systems with reaction volumes of 300 mL.
Sample aliquots were obtained by syringe through a Teflon
septum. The reaction vessels were wrapped in aluminum to
limit exposure to light. Reaction solutions were preequili-
brated at 15 °C in a circulating water bath.
0
2+
-
Fe + 2H O h Fe + H + OH
(10)
(11)
2
2
0
2+
-
2
Fe + O + 2H O h 2Fe + 4OH
2 2
0
Kinetic runs were initiated by adding a measured amount
The observed higher pH with the combined US/ Fe systems
indicates that more Fe0 is involved in the degradation of
nitrobenzene or oxidized by water and oxygen in the presence
of ultrasound.
Nitrosobenzene (NSB) and aniline (AN) were the major
products observed in the reduction of nitrobenzene by Fe .
0
of Fe to the reactor system which was placed on a variable-
speed rotary shaker. The entire shaker was kept in an air-
cooled refrigeration unit at a constant temperature of 15 °C.
Sonolyses were performed with a VCX-400 Vibracell
0
(
Sonics and Materials, Inc., Danbury, CT) operating at 20
kHz. The temperature was maintained at 15 ( 2 °C with a
Haake A80 Refrigerated Bath and Circulator. Replaceable
titanium tips with diameter of 1.27 cm on the transducer
were polished, and the transducer was tuned before each
use to give a minimum power output when vibrating in air.
The tuning process is a standard procedure to bring the
transducer into resonance as part of the complete probe
assembly. Sonolytic reactions at 20 kHz were performed in
a 300 mL airtight reactor cell with air equilibrium. The physical
dimensions and fundamental characteristics of the trans-
ducer and the reactor configuration have been presented
However, during sonication the principal observed inter-
mediates were o-nitrophenol (ONP), m-nitrophenol (MNP),
p-nitrophenol (PNP), and 4-nitrocatechol (4-NC). The key
UV-vis spectroscopic and chromatographic data for ni-
trobenzene and its intermediates are presented in Table 1.
0
First-order plots of ln{[NB]/ [NB] }vs time during sonolysis
0
at 20 kHz and during reduction by Fe are shown in Figure
1. The slopes obtained from linear regression of the data
yield the observed first-order rate constants for nitrobenzene
degradation. The corresponding reaction rate for nitroben-
zene degradation by ultrasound in the presence of iron was
0
0
(
16). The bottom of the glass reactor was designed with a 1
also found to be first-order at lower Fe loadings, [Fe ]. At
0
cm indentation in the center for reflection of the sound waves
and for an even distribution of the cavitation bubbles in the
higher [Fe ], the second-order polynomial function (dashed
line) gives a better experimental fit than a simple first-order
VOL. 34, NO. 9, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 7 5 9

FIGURE 1. The degradation of nitrobenzene by ultrasound (b), 10
rpm/70 g L Fe (2), and US/70 g L Fe (9); the solid lines are the
0
-
1
0
-1
0
FIGURE 2. The fitted results from eq 18 in the combined US/Fe
systems as a function of [Fe ].
0
fitted results from the first-order analysis, and the dashed line is
-
1
0
the second-order fitting from eq 18 for US/70 g L Fe .
intermediates and nitrobenzene, and [Xi,j] is effective con-
centration of the newly generated Fe surface sites (i) and
the reactive intermediates (j). Since the reactive intermediates
will be consumed very fast with any reactants in the solution,
the contribution in [Xi,j] from the reactive intermediates can
be ignored. If we assume that [Xi,j] is mainly contributed
0
TABLE 2: Comparison of k Values for Nitrobenzene
0
Degradation in the Combined US/Fe System (Ultrasound
Power Amplitude: 35%) and on the Rotary Shaker System
0
0
0
0
[
Fe ],
kUS + kFe ,
kFe ,
kseckp,
k10rpm/Fe ,
-
1
-1
3
min-1 ‚ 103
min ‚ 105
-2
min ‚ 103
-1
from the newly generated Fe surface by ultrasound, the total
activated Fe surface (x) increasing rate is proportional to x
0
g L
min ‚ 10
0
0
1.8 (( 0.1)
2.9 (( 0.1)
5.9 (( 0.2)
6.9 (( 0.4)
7.8 (( 0.4)
12.4 (( 1.5)
0
0
0
0
with a rate constant, k
a
, as follows:
1
3
5
7
8
8
5
3
0
8
1.1
4.1
5.1
6.0
10.6
0.8
1.5
3.5
4.8
9.9
∼0
dx
2.8 ( 0.5
9.8 ( 0.5
21.0 ( 2.4
) kax
dt
(14)
x as a function of time can be solved as follows
relationship (solid line) as shown in Figure 1. The average
deviation for nitrobenzene reductions determined in replicate
experiments was < 10%.
k t
a
x ) x + [X ] ) x e
(15)
0
i,j
0
0
0
where x is the initial activated Fe surface before ultrasound
Reduction by Elem ental Iron Surfaces. Agrawal and
Tratnyek (14) reported that the nitrobenzene reduction rate
constants exhibited a linear relationship with respect to the
0
is applied. Since the influence of ultrasound at low [Fe ] can
be ignored, the exponential term in eq 15 could be expanded
into polynomial terms. Without consideration of the high
order terms, [Xi,j] can be expressed as follows
0
.5
square root of the stirring rate (i.e., rate ∝
ω ), which
indicated that the apparent reaction rate was limited by mass
transport. In this study, we mixed the suspension with a 360°
rotary shaker at ω ) 10 rpm. The rate of reduction of
[
X ] ) x k t ) k t
(16)
i,j
0
a
p
0
nitrobenzene by Fe is observed to be first-order. The
where k
p
) x k is a zero-order rate constant that increases
with added [Fe ]. Substitution of eqs 13 and 16 into eq 12,
yields the following time-implicit equations:
0 a
0
observed reaction rate constants, k10rpm/Fe0, as a function of
0
[
Fe ] are listed in Table 2. As expected, k10rpm/Fe0 increases
0
with [Fe ].
In the combined US/ Fe0 systems, the overall rate of
d[NB]
nitrobenzene disappearance can be described by a linear
combination of first-order terms (17)
-
) (kUS + kFe
0
+ kseckpt)[NB]
(17)
dt
The integration of eq 17 gives
d[NB]
dt
-
) kUS[NB] + kFe
0
[NB] + k′[NB] )
kUS + kFe0 + k′) [NB] (12)
[
NB]
1
2
2
(
- ln
) (kUS + kFe
0
)t + k k t
(18)
sec
p
[
[NB]0
]
where kUS, kFe0, and k′ are, respectively, the first-order
degradation rate constants for sonolysis, reduction by Fe ,
0
Values of kUS were determined from the separate sonication-
and the synergistic kinetic effect achieved by combining the
two systems. k′ may be related to the generation of increased
surface area or reactive intermediates, which are capable of
degrading nitrobenzene. The k′ should be dependent on the
newly created reactive surface sites and on the concentration
of the reactive intermediates. k′ can be expressed by the
following generalized equation
only experiments. Figure 2 shows the fitted results from eq
18 for various [Fe ]. The individual rate constants are listed
0
in Table 2.
The values of kFe0
in the combined US/ Fe
0
system are
slightly higher than k10rpm/Fe0
values in part because of the
higher relative mixing rates achieved during sonication. The
new surface area and species created by ultrasound are
0
reflected in ′ksec
k
p
′, which increases with Fe loading at higher
0
k′ ) ksec[Xi,j]
(13)
[Fe ]. The higher values of ′ksec
k
p
′ obtained with increasing
Fe ] are due primarily to the effects related to the direct
0
[
where ksec is the average second-order rate constant for the
reaction between the Fe surface sites or the reactive
reactions with elemental iron. Since ksec is a rate constant
quantifying the impact of the secondary effects on nitroben-
0
1
7 6 0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 9, 2000

0
an input power density of 139 W L-1. In this case, there is
zene, it is independent of [Fe ]. On the other hand, k
p
should
0
0
increase with [Fe ] as mentioned above.
The higher degradation rates observed in the combined
US/ Fe systems can be attributed to the indirect chemical
effects associated with the continuous ultrasonic cleaning
and activation of the Fe surface, the enhanced rate of mass
transport resulting from the turbulent effects of cavitation,
and the H released to solution from the HNO
during the sonolysis of nitrobenzene and water (eq 3). The
shock wave and microjets formed during cavitational bubble
collapse are primarily responsible for the surficial cleaning
actions of ultrasound. Transient cavitation results in turbulent
flow conditions within the reactor that enhance overall mass
transport. The iron corrosion accelerated by ultrasound may
also contribute to the reduction of the nitro group to an
amine group via eq 9.
no enhancement due to the presence of Fe surfaces because
0
aniline cannot be reduced by Fe . The observed rate
0
enhancements in the reduction of nitrobenzene in the
0
combined US/ Fe system could also due to the degradation
0
of aniline by sonolytic cavitation. The degradation of aniline
0
-1
with US or the combined US/ Fe (35 g L ) system has the
same observed rate constants. This result indicates that the
+
3
produced
0
sonolyses are not affected by Fe and the enhancement of
0
nitrobenzene degradation by US/ Fe is mainly due to a faster
rate of nitrobenzene reduction by Fe0 in the presence of
ultrasound.
The rate constants for the sonolytic degradations of
nitrobenzene and aniline are smaller than those observed
for CCl (16). This may be due to the lower tendency for the
4
substituted benzenes to partition into the vapor phase of the
cavitation bubbles due to their lower Henry’s Law constants
(17). The principal observed intermediate during aniline
sonolysis is benzoquinone. The possible mechanism for the
degradation of aniline by sonication is as follows.
It has long been recognized that air-saturated water, which
is exposed to the action of ultrasound, leads to the formation
-
-
of NO
2
and to a lesser extent NO as well as H O (18-21).
3
2
2
The observed pH decrease during sonication should result
0
in a faster rate of reduction of Fe (eq 6). Since the sonolytic
3
degradation of nitrobenzene produces HNO according to
eqs 1-3, we expect to see a faster reduction rate in the
combined reaction system, but this contribution can be
0
ignored because of higher pH observed at US/ Fe systems.
+
The H produced during sonolysis should be consumed by
0
Fe oxidation or nitrobenzene reduction according to the
following stoichiometry:
Aqueous-phase Fe2 (i.e., Fe(H
+
O)
6
2+) and the Fe2+ sorbed
, which is
2
0
on the surface of Fe can also react with H
2
O
2
produced from the sonication of H
2
O (20) as follows:
2
+
3+
-
Fe(H O)6 + H O f Fe(H O)
6
+ OH + ‚OH
(20)
2
2
2
2
2
+
3+
-
{
Fe‚Fe } + H O f {Fe‚Fe } + OH + ‚OH (21)
2 2
The ‚OH radical will be available for further reactions such
as the degradation of nitrobenzene and aniline. Since high
2
+
0
concentrations of Fe are produced from oxidation of Fe ,
the subsequent Fenton’s Reagent reaction should be inde-
0
pendent of [Fe ]. The contribution of the Fenton’s reaction
in the combined US/ Fe systems should be relatively constant
0
and thus reflected in the second term of eq 18. Since a similar
degradation rate for aniline is observed in both the US and
0
combined US/ Fe systems, the secondary Fenton’s reaction
Identification of Byproducts and Reaction Interm edi-
ates. The intermediates observed at relatively high concen-
trations during nitrobenzene sonication at 20 kHz are
m-nitrophenol, p-nitrophenol, and 4-nitrocatechol (17).
These intermediates are eventually decomposed sonochem-
appears to be negligible compared to other secondary effects.
Ultrasound also enhances the mass transfer of the product,
aniline, away from the Fe surface.
0
The observed nitrobenzene reduction rates in our study
are somewhat lower than those reported by Agrawal and
Tratnyek (14). These differences may be due to the specific
nature of the buffers and the elemental iron used in each
case. In our study, the elemental iron was used without
pretreatment, and therefore the initial reactive surface areas
may have been somewhat different than those used by
Agrawal and Tratnyek (14).
2 3
ically to CO and HNO . In this study, nitrosobenzene and
aniline were observed as the main intermediates in the
0
presence of Fe . Other reaction intermediates such as
4-nitrocatechol and benzoquinone have concentrations less
0
than 1 µM in the combined US/ Fe systems and are ignored
in Figure 3 (a). The nitrophenols (eq 4) were not observed.
This may be due to a fast transformation rate to 4-nitro-
catechol such that the nitrophenols concentrations were kept
below the detection limits under current experimental
conditions. The relative effects of sonication on the observed
intermediates and the total mass balance around nitroben-
zene are shown in Figure 3. After 3 h of reaction, the total
mass of aromatic compounds does not change in the
Aniline Sonolysis. The final observed product of ni-
trobenzene reduction by Fe is aniline, which is a well-known
0
corrosion inhibitor. The mechanism of inhibition is believed
to involve an interference with mass transport of the primary
oxidant to the metal surface, which in turn is strongly
influenced by the orientation of the sorbed aniline. The
sonolytic degradation of aniline is a first-order reaction with
0
mechanically agitated Fe system but is decreased by 12% in
-
3
-1
0
an apparent rate constant of 2.4 × 10 min at 20 kHz with
US system and by 40% in the combined US/ Fe system. The
VOL. 34, NO. 9, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 7 6 1

TABLE 3: Ultrasonic Power Amplitudes and the Corresponding
Absorbed Power as Determined by Calorimetry
applied power (W)
140
80
40
absorbed power (W)
40
139
21
73
11
39
-
1
power density, W L
-
1
-3
-3
-3
-3
-3
-3
kUS, m in
1.8 × 10
1.4 × 10
0.7 × 10
kUS+35gL-1Fe0, m in-
1
6.1 × 10
4.8 × 10
2.4 × 10
increased another 50%. As reported previously (16), the
efficiency of energy transfer to solution decreases with higher
power densities due to a greater loss of input energy as heat.
The observed rate constants are directly proportional to the
power density at low power inputs; however, at high input
powers the relationship becomes nonlinear. Furthermore,
0
at higher sonication intensities, the Fe -corrosion pathways
(
eqs 10 and 11) are enhanced, and thus the net degradation
rate of nitrobenzene is no longer directly proportional to the
sonication power.
In this study, we demonstrated that the rate of nitroben-
0
zene reduction by Fe could be enhanced in the presence of
ultrasound. The higher apparent rates of reaction in the
0
0
combined US/ Fe systems compared to Fe reaction alone
can be attributed primarily to the continuous cleaning and
0
activation of the Fe surface by the chemical and physical
effects of ultrasound. Furthermore, the rates of mass transport
rates of nitrobenzene, nitrosobenzene, and aniline to and
0
from the Fe surface are enhanced by hydrodynamic cavita-
tion. Oxidation of the reaction intermediates during soni-
cation also contributes to the observed rate enhancements.
The relative concentrations of the principal reaction inter-
mediates in US/ Fe0 systems, nitrosobenzene, and aniline
appear to be influenced significantly by total available surface
FIGURE 3. The concentrations of the reactant, intermediates, and
-
1
0
products as a function of time: (a) in the US/70 g L Fe system
0
-
1
0
area of Fe . In conclusion, the combination of ultrasound
and (b) in the 10 rpm/70 g L Fe system.
0
and Fe appears to have a positive synergistic effect on the
reduction of nitro aromatic compounds.
principal intermediate, aniline, which is not degraded by
Fe , is eventually destroyed by ultrasound. The array of
0
Acknowledgments
aromatic compounds observed during nitrobenzene deg-
radation should be more efficiently decomposed in the
Financial support from the Office of Naval Research (NAV5-
N0001492J1901; NAV1-N47408-97-M-0771) and the Depart-
ment of Energy (DOE 1 963472402) is gratefully acknowl-
edged.
0
combined US/ Fe systems. The higher concentration of
0
aniline in US/ Fe systems suggests that the nitrobenzene
degradation rate is apparently enhanced from the reduction
0
of nitrobenzene by Fe .
Effects of Ultrasonic Power Density. The reduction of
nitrobenzene by Fe occurs at the Fe / H O interface. The
2
Literature Cited
0
0
(
1) Cheung, H. M.; Bhatnagar, A.; Jansen, G. Environ. Sci. Technol.
1991, 25, 1510-1512.
heterogeneous reactions involves five steps (3, 22): (i) mass
0
transfer of the reactant to the Fe surface from the bulk
(2) Hua, I.; Hoffmann, M. R. Environ. Sci. Technol. 1996, 30, 864-
8
71.
solution; (ii) adsorption of the reactant on the surface; (iii)
chemical reaction at the surface; (iv) desorption of the
products from the surface; and (v) mass transfer of the
products into the bulk solution. Any one of these steps may
be rate limiting and, therefore, be reflected in the values of
(
(
3) Mason, T. J.; Lorimer, J. P. Sonochemistry: Theory, Application
and Uses of Ultrasound in Chemistry; John Wiley & Sons: New
York, 1988.
4) Suslick, K. S.; Hammerton, D. A.; Cline, R. E. J. Am. Chem. Soc.
1986, 108, 5641-5642.
k
obs reported in this study. To properly interpret trends in the
(5) Mason, T. J. Practical Sonochemistry: User’s Guide to Application
in Chemistry and Chemical Engineering; Ellis Horwood:, 1991.
observed reaction rates, it is important to distinguish between
mass transport and reaction-limited kinetics.
(
6) Fennelly, J. P.; Roberts, A. L. Environ. Sci. Technol. 1998, 32,
980-1988.
7) Matheson, L. J.; Tratnyek, P. G. Environ. Sci. Technol. 1994, 28,
2045-2053.
1
A common criterion for detecting mass transfer-limited
kinetics is a variation in the reaction rate with the intensity
of mixing. An increase in ultrasonic power will increase the
mixing intensity due to the turbulence generated by cavi-
tational bubble collapse. The ultrasonic power levels adjusted
by changing the ultrasound power amplitude and the
corresponding calorimetry results are given in Table 3. The
reactions taking place at an Fe loading of 35 g L are pseudo-
first-order. Under these conditions, the relative enhancement
in the nitrobenzene reduction rate increases with increasing
sonication power. For example, when the applied power
(
(8) Tratnyek, P. G.; Johnson, T. L.; Scherer, M. M.; Eykholt, G. R.
Groundwater Monitoring Remediation 1997, XVII, 108-114.
9) Scherer, M. M.; Westall, J. C.; Ziomek-Moroz, M.; Tratnyek, P.
(
G. Environ. Sci. Technol. 1997, 31, 2385-2391.
(
10) Johnson, T. L.; Fish, W.; Gorby, Y. A.; Tratnyek, P. G. J. Contam.
Hydrol. 1998, 29, 377-396.
0
-1
(11) Gillham, R. W.; Ohannesin, S. F. Ground Water 1994, 32, 958-
967.
(
(
12) Lipczynskakochany, E.; Harms, S.; Milburn, R.; Sprah, G.;
Nadarajah, N. Chemosphere 1994, 29, 1477-1489.
13) Johnson, T. L.; Scherer, M. M.; Tratnyek, P. G. Environ. Sci.
Technol. 1996, 30, 2634-2640.
-
1
density is increased from 39 to 73 W L , the reaction rate
is approximately doubled. However, when the power density
is increased further to 139 W L , the reaction rate is only
(14) Agrawal, A.; Tratnyek, P. G. Environ. Sci. Technol. 1996, 30, 153-
-
1
160.
1
7 6 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 9, 2000

(
15) Reinhart, D. R.; Clausen, C.; Geiger, C.; Ruiz, N.; Afiouni, G. F.
Enhancement of In-Situ Zero-Valent Metal Treatment of
Contaminated Groundwater. In Nonaqueous phase liquids in
surface environment: assessment and remediation; American
Society of Civil Engineers: Washington, DC, 1996; pp 323-332.
16) Hung, H.-M.; Hoffmann, M. R. Environ. Sci. Technol. 1998, 32,
(20) Hua, I.; Hoffmann, M. R. Environ. Sci. Technol. 1997, 31, 2237-
2243.
(
21) Kotronarou, A.; Mills, G.; Hoffmann, M. R. J. Phys. Chem. 1991,
5, 3630-3638.
9
(
(
(
(
(22) Spiro, M. Heterogeneous Catalysis of Solution Reactions. In
Chemical Kinetics; Elsevier: Amsterdam, 1989; pp 69-166.
3
011-3016.
17) Weavers, L. K.; Ling, F. H.; Hoffmann, M. R. Environ. Sci. Technol.
998, 32, 2727-2733.
1
Received for review April 5, 1999. Revised manuscript re-
ceived September 17, 1999. Accepted December 29, 1999.
18) Wakeford, C. A.; Blackburn, R.; Lickiss, P. D. Ultrasonics
Sonochemistry 1999, 6, 141-148.
19) Carvalho, L. R. F.; Souza, S. R.; Martinis, B. S.; Korn, M. Anal.
Chim. Acta 1995, 317, 171-179.
ES990385P
VOL. 34, NO. 9, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 7 6 3