Kinetics, Mechanism, and Activation Energy of H 2O2Decomposition on the Surface of ZrO 2 [612461]

Kinetics, Mechanism, and Activation Energy of H 2O2Decomposition on the Surface of ZrO 2
Cla´udio M. Lousada* and Mats Jonsson
KTH Chemical Science and Engineering, Nuclear Chemistry, Royal Institute of Technology, SE-100 44
Stockholm, Sweden
Recei Ved: March 31, 2010; Re Vised Manuscript Recei Ved: May 10, 2010
The kinetics, mechanism, and activation energy of H 2O2decomposition in ZrO 2particle suspensions were
studied. The obtained first-order and second-order rate constants for the decomposition of H 2O2in the presence
of ZrO 2atT)298.15 K produced the values k1)(6.15 (0.04)×10-5s-1andk2)(2.39 (0.09)×10-10
m·s-1, respectively. The dependency of the reaction first-order rate constant with temperature was studied;
consequently, the activation energy for the reaction was obtained in the temperature interval 294.15 -353.15
K having yielded the value Ea)33(1.0 kJ ·mol-1. The dependency of the zeroth-order reaction rate constant
with pH was investigated and discussed. A mechanistic study encompassing the investigation of the dynamicsof formation of hydroxyl radicals during the course of the reaction was performed. A version of the modifiedHantzsch method was applied for this purpose, and it was verified that the dynamics of formation of hydroxylradicals during the reaction are in good agreement with the proposed reaction mechanism.
1. Introduction
Zirconium dioxide, mostly known as zirconia, is one of the
most versatile ceramic materials known. Its physical andchemical properties make it suitable for a wide range ofapplications including nanotechnology,
1catalysis and synthesis,2-5
medicine,6electronics and sensors,7and the manufacture of a
diversity of materials.8The fact that its radiation stability is
good,9it has a low neutron cross section, and it has low
solubility in water at high temperatures10makes it presently a
main favored candidate to be a component of the inert fuelmatrix in nuclear reactors.
11The presence of zirconium dioxide
in nuclear systems is not only restricted to its incorporation intothe fuel matrix. The fuel material in a nuclear reactor is protectedby cladding pipes made of alloyed zirconium. In contact withwater near and above its critical temperature, a corrosion layerof hydrated zirconium dioxide ZrO
2·nH2O is formed and has
implications on the chemistry of the reactor.12-14One of the
most important water radiolysis products to concern about inreactor chemistry is hydrogen peroxide.
15Hydrogen peroxide
is an important compound that has found innumerous uses suchas a bleaching agent,
16disinfectant,17oxidizer,18or as a nontoxic
monopropellant in rocket fuel.19Hydrogen peroxide is the main
oxidizing molecular product formed during the radiolysis ofwater. It is formed primarily by combination reactions of HOradicals produced in the radiolytic decomposition of water. Itsimportance due to an increase in its concentration is augmentedunder conditions subjected to radiation with high linear energytransfer, as in an operating nuclear power plant,
20where ∼2%
of the total fast neutrons and γ-ray energy released in the core
of an operating nuclear reactor is deposited in the cooling water21
or under conditions of storage of spent nuclear fuel.22The
importance of the system ZrO 2/H2O2lead us to develop a study
on the dynamics, energetics, and mechanism of the reactionbetween these two chemical species.
It was previously reported that the reaction of H
2O2on the
surface of ZrO 2in liquid water consists of the decompositionof H 2O2to produce water and oxygen. Given that the zirconium
in ZrO 2is in its highest oxidation state, no redox reactions are
involved in the process. This metal-oxide-catalyzed reaction isproposed to follow the scheme
23
H2O2+Mf2HO·+M (R1)
HO·+H2O2fHO2·+H2O (R2)
2HO2·fH2O2+O2 (R3)
where M represents an undefined site located at the oxide
surface. The decomposition of hydrogen peroxide on a solidsurface is a spontaneous process at temperatures that range fromroom temperature to 286 °C, and its reported activation energy
ranges from 20.93 to 96.30 kJ ·mol
-1, depending on the surface
type and on factors such as the oxidation state of the metal,among others.
24Reaction R3 corresponds to the chain termina-
tion and occurs via the disproportionation of two hydroperoxylradicals as represented. When reaction R3 occurs with purewater as a solvent, the activation energy is 25.0 kJ ·mol
-1in
the temperature range 274 -316 K.25
The evaluation of the effect of zirconium dioxide on the
kinetics and energetics of decomposition of hydrogen peroxidehas to be done at a temperature and pH range where thespontaneous uncatalyzed decomposition of hydrogen peroxideis negligible when compared with the rate of decomposition ofhydrogen peroxide on the surface of the oxide itself. In neutralwater, from the species involved in the reactions mentionedabove, only dissociation of HO
2·needs to be considered because
the p Kavalues for H 2O2,H O·, and HO 2·are 11.8, 11.9, and
4.88, respectively.26The HO 2·hydroperoxyl radical is a weak
acid and is also the protonated form of the superoxide radicalanion which is easily formed and sorbed at the surface of thezirconium dioxide according to the following reaction
27
HO2·98ZrO2
H++O2-·(R4)
The superoxide anion radical is stabilized by adsorption on
the surface of the zirconium dioxide and has been used* To whom correspondence should be addressed. Tel: (46) 8 790 87 89.
Fax: (46) 8 790 87 72. E-mail: cmlp@kth.se.J. Phys. Chem. C 2010, 114, 11202–11208 11202
10.1021/jp1028933 2010 American Chemical Society
Published on Web 06/04/2010

previously as a probe for surface cationic fields. The superoxide
radical adsorbs to the surface exclusively by coordination withexposed Zr
4+surface sites.27The superoxide radical anion is in
many cases a precursor of the highly reactive hydroxyl radicalfollowing a Fenton-type mechanism in systems where the metalcation can undergo further stages in oxidation.
28The formed
superoxide anion radical is an active reductive species and canreduce ions in the higher valence state, for example
25
O2-·+Fe3+fO2+Fe2+(R5)
The rate constant for the reaction of the hydroxyl radical with
organic molecules ranges from ∼4×106to 2×1011M-1·s-1.29
The surface of the oxide is capable of stabilizing the intermedi-
ary radical species formed during the decomposition of H 2O2
by means of interactions of the formed radicals with the oxidelattice.
27,30As shown above, the proposed mechanism of the
reaction of decomposition of hydrogen peroxide on the surfaceof an oxide that cannot undergo further oxidation is rathercomplex and has not been clearly elucidated. In this work, thereaction of H
2O2in the presence of ZrO 2was studied at different
temperatures and pH values. A mechanistic study was alsoperformed and consisted of developing a method to study thedynamics of formation of the intermediary radical species duringthe decomposition of H
2O2at the surface of ZrO 2. A study
involving water radiolysis in the presence of tris/HCl bufferallowed us to calibrate the method cited above and consequentlyto quantify the rate of formation of HO radicals during thereaction of H
2O2in the presence of ZrO 2.
2. Experimental Details
Instrumentation. Specific surface areas of the powders were
determined using the BET method of isothermal adsorption anddesorption of a gaseous mixture consisting of 30% N
2and 70%
He on a Micrometrics Flowsorb II 2300 instrument. γ-Irradiation
was performed using a MDS Nordion 1000 Elite Cs-137γ-source with a dose rate of 0.15 Gy ·s
-1; this value was
determined by Fricke dosimetry.31X-ray powder diffractograms
(XRD) were obtained at 293 K using Cu K Rradiation, on a
PANanalytical X’pert instrument. Powders were encapsulatedon Lindemann capillaries. The data was collected over the range3e2θe80°with a step size of 0.033 °(2θ). Data evaluation
was done using The High Score Plus software package, andthe PDF-2 database was used for matching the experimentallyobtained diffractograms. The samples were weighted to (10
-5
g, in a Mettler Toledo AT261 Delta Range microbalance. The
reactions were performed under an inert atmosphere with aconstant flux of N
2gas (AGA Gas AB) with a flow rate of 0.21
L·min-1that was also used for stirring the solutions. We kept
the temperature constant throughout the experiments by usinga Huber CC1 or a Lauda E100 thermostat calibrated against aTherma 1 thermometer coupled to a submersible K-type(NiCrNi) temperature probe with a precision of (0.1 K. UV/
vis spectra were collected using a WPA Lightwave S2000 or aWPA Biowave II UV/vis spectrophotometer. Trace elementalanalysis were performed using the technique of inductivelycoupled plasma spectroscopy on a Thermo Scientific iCAP 6000series ICP spectrometer. The analysis for Zr was performed atthe wavelength of 343.823 nm.
Reagents and Experiments. All solutions used in this study
were prepared using water from a Millipore Milli-Q system.
Zirconium dioxide (CAS[1314-23-4], Aldrich 99%, particle
size<5µm) was used without further purification. The powder
pattern was indexed as monoclinic. A Rietveld refinement usingICSD-26488 as a starting model was performed yielding thefollowing cell parameters: (a) 5.1458(2), (b) 5.2083(3), and (c)
5.3124(3) Å. These values are in good agreement with (a) 5.143,(b) 5.204, and (c) 5.311 Å attributed to the monoclinic phaseand published in a previous work.
32The specific surface area
of the ZrO 2powder, measured by the BET method produced
the value of 5.0 (0.2 m2·g-1. This value is the average of
three measurements, each one consisting of a sorption and adesorption isotherm whose values were also averaged. The ZrO
2
particle suspensions, where the reaction with H 2O2(CAS[7722-
84-1]) took place, consisted of 0.5 -4.5 g of ZrO 2in 50 mL of
H2O2solution with a H 2O2concentration that varied from 0.5
mM to 6 mM. The H 2O2solutions were prepared from a 30%
standard solution (Merck). After extraction of the sample fromthe reaction vessel, the sample was filtered through a GammaMedical 0.45 µm to 25 mm cellulose acetate syringe filter.
Subsequently a sample volume of 0.2 mL was used for themeasurement of hydrogen peroxide concentration. The concen-tration of hydrogen peroxide as a function of reaction time wasthen determined by the Ghormley triiodide method. In thismethod, I
-is oxidized to I 3-by the H 2O2.33,34The absorbance
of the product I 3-was measured spectrophotometrically at the
wavelength of 360 nm. A calibration curve where the absorbanceof I
3-was plotted as a function of the concentration of hydrogen
peroxide was obtained in the range of concentrations 0.02 to0.8 mM in H
2O2, resulting in a linear correlation between
absorbance and concentration. A mechanistic study of thedecomposition of H
2O2on the surface of ZrO 2was carried out
and involved verifying the presence and quantifying the rate of
production of hydroxyl radicals as intermediate product in thereaction of decomposition of H
2O2on the surface of ZrO 2.This
was done by means of the reaction between tris(hydroxymethy-l)aminomethane (tris buffer) (CAS[77-86-1], BDH Chemicals99%) and the hydroxyl radicals to produce formaldehyde.
35In
this study, the reaction between ZrO 2particle suspensions and
H2O2was performed at temperatures that ranged from 293.15
to 313.15 K at the midpoint of the buffering range of the trisbuffer. The reaction mixture consisted of 0.5 -4.5 g of ZrO
2in
50 mL solution of 0.5 -6.0 mM in H 2O2and 0.5 -20 mM in
tris buffer at a pH of 7.5; the pH was adjusted with HCl. Theformaldehyde produced was then quantified spectrophotometri-cally at 368 nm using a modified version of the Hantzschreaction.
36In this method, the formaldehyde reacted with
acetoacetanilide AAA (CAS[102-01-2], Alfa Aesar >98%) in
the presence of ammonium acetate (CAS[631-61-8], Lancaster98%) to form a dihydropyridine derivative, which has themaximum absorption wavelength at 368 nm. A calibration curveplotting the absorbance of the dihydropyridine derivative as afunction of formaldehyde concentration was obtained at 368nm, giving a linear correlation between absorbance and con-centration in the concentration range 0.15 µMt o1m Mi n
formaldehyde. The plotting of the calibration curve for form-aldehyde required the preparation of several solutions of CH
2O
with different rigorously known concentrations in the concentra-tion range mentioned above. It was then necessary to proceedto the accurate determination of the concentration of formal-dehyde in the solution used initially (CAS[50-00-0]), Aldrich3 7w t%i nH
2O) using the iodometric method.37The solutions
and respective standardizations necessary to follow the iodo-metric method procedure were prepared as stated in the citedpaper
37and as described elsewhere.38The error associated with
the determination of the concentration of formaldehyde in theinitial solution was <2%.H
2O2Decomposition on the Surface of ZrO 2 J. Phys. Chem. C, Vol. 114, No. 25, 2010 11203

3. Results and Discussion
Kinetics of H 2O2Decomposition on the Surface ZrO 2.It
has been previously reported that the catalytic decompositionof hydrogen peroxide follows first-order kinetics with respectto H
2O2.39-41In the presence of a solid, the reaction can be
approached to a pseudo-first-order when an excess of solidsubstance is present. Given this fact, and considering reactionR1, one can expect that the concentration of H
2O2evolves as a
function of time according to
-d[H2O2]
dt)k1[H2O2], which in its integrated form,
ln([H2O2]
[H2O2]o))- k1t(6)
where tis the reaction time, k1is the pseudo-first-order rate constant
at a given temperature, [H 2O2] is the concentration of H 2O2at a
time, t, and [H 2O2]ois the concentration of H 2O2att)0.
The set of data obtained for the variation in the concentration
of hydrogen peroxide at T)298.15 K as a function of reaction
time is represented in Figure 1a. Previous test reactions withconcentrations in hydrogen peroxide ranging from 0.5 -6.0 mM
in H
2O2and 0.5 -4.5 g of ZrO 2, in 50 mL of solution showed
that the concentration in H 2O2and the amount of ZrO 2that
provided better conditions for error minimization caused by tooshort reaction time or by a large number of dilutions for furtherconcentration measurement was the value of 0.5 mM in H
2O2
and 1.5 g of ZrO 2. The first-order rate constant obtained from
the slope in Figure 1b for the temperature T)298.15 K is k1
)(6.15 (0.04)×10-5s-1. We determined the second-order
rate constant by studying the pseudo-first-order rate constantas a function of solid-surface-area-to-solution volume ratio. Thesecond-order rate expression is given by
-d[H2O2]
dt)k2(SAZrO2
V)[H2O2] (7)
where SAZrO 2denotes the surface area of the zirconium dioxide
powder, Vis the volume of the solution where the reaction takes
place, and k2is the second-order rate constant. The first-order
rate constant is plotted against surface-area-to-solution volumeratio in Figure 2. The second-order rate constant obtained fromthe slope of Figure 2 for the reaction at 293.15 K is k
2)(2.39
(0.09)×10-10m·s-1. As expected, this value is very far from
a diffusion controlled reaction for which the rate constant is onthe order of 10
-5m·s-1for particles of this size.42
The variation of the rate constant with temperature generally
follows the Arrhenius equation
k1)Ae-Ea/RT(8)
where Eais the activation energy for the reaction, Ais the pre-
exponential or the frequency factor, Ris the gas constant, andTis the absolute temperature. To determine the activation energy
for the reaction, we obtained the rate constants as a function oftemperature in the temperature interval T)[294.15 -353.15]
K with a temperature step of 5 K. Some of the resulting first-order plots are represented in Figure 3. The plots of ln([H
2O2]/
[H2O2]0) as a function of reaction time show good agreement
with first-order kinetic behavior; the order of the reaction withrespect to H
2O2was confirmed by obtaining initial rates with
different initial concentrations of H 2O2while keeping the amount
of ZrO 2constant. A plot of ln(initial reaction rate) as a function
of ln(concentration of H 2O2) was obtained at T)313.15 K,
for a mass of ZrO 2equal to 1.5 g and concentrations in H 2O2
that ranged from 0.5 to 6 mM. A linear regression of theobtained data yielded a slope of 0.98 (0.04. Considering the
experimental error associated with this value, it is possible tosay that the reaction is first-order with respect to H
2O2. Because
the intercept of this plot is equal to ln k1, this value corresponds
to ak1value of (2.02 (0.09)×10-4s-1, which differs by less
than a factor of two from the first-order rate constant obtainedat this same temperature by the standard method, k
1)(1.22 (
0.13)×10-4s-1.
For each temperature value, three experiments were per-
formed, and the resulting rate constants were averaged for thecalculation of the activation energy. The logarithm of the rateconstants as a function of the inverse of the temperature,obtained in the temperature interval mentioned above, are
Figure 1. (a) Concentration of hydrogen peroxide as a function of reaction time at T)298.15 K. Initially, 0.5 mM in H 2O2and 1.5 g of ZrO 2in
50 mL of H 2O. (b) ln([H 2O2]/[H 2O2]0) as a function of reaction time for T)298.15 K.
Figure 2. First-order rate constant as a function of the ZrO 2surface-
area-to-solution volume ratio obtained at a temperature of 298.15 Kand with a initial concentration of 0.5 mM in H
2O2in a volume of 50
mL.
Figure 3. ln([H 2O2]/[H 2O2]0) as a function of reaction time for T)
298.15, 313.15, and 353.15 K.11204 J. Phys. Chem. C, Vol. 114, No. 25, 2010 Lousada and Jonsson

represented in Figure 4. As can be seen, the rate constant is
largely affected by the temperature, displaying Arrheniusbehavior. The rate constant varied from (5.670 (0.003) ×10
-5
to (4.50 (0.01)×10-4s-1, when the temperature varied from
298.15 to 353.15 K, respectively. The half-life varied from 198.0min for the reaction at 298.15 K to 26.3 min for the reaction at353.15 K. The determined activation energy for the decomposi-tion reaction of hydrogen peroxide at the surface of zirconiumdioxide in the temperature range T)[298.15 -353.15] K is 33
(1.0 kJ ·mol
-1. This value is in good agreement with previously
published values for similar systems.24When comparing this
value with the energy necessary for the cleavage of the O -O
bond in H 2O2,43208 kJ ·mol-1, it is obvious that the oxide -liquid
interface lowers the energy barrier for its cleavage substantially.
To verify the fact that zirconium dioxide acts only as a
catalytic support for the decomposition of the hydrogen peroxideand has no further involvement in the reaction, a study ofpossible changes in the surface area of the powder wasperformed prior to immersion in water, after immersion in waterduring a time interval equal to the reaction time and afterreaction with hydrogen peroxide. Prior to the BET specificsurface area determination, the solid was dried in vacuum at T
)353.15 K and P)0.1 Pa; the data obtained are summarized
in Table 1. As can be seen, no changes occurred in the specificsurface area of the solid during the reaction with hydrogenperoxide. We studied a possible change in the crystal structureof the powder by obtaining XRD diffractograms before and afterreaction. Prior to the collection of the XRD diffractogram, thereaction between ZrO
2and H 2O2was performed in 50 mL of
H2O with 0.5 mM H 2O2atT)298.15 K until complete
consumption of H 2O2, which implied a reaction time of around
13 h. The obtained cell parameters (a) 5.1497(7), (b) 5.2123(7),and (c) 5.3164(8) Å are in excellent agreement with the onesobtained before the reaction took place. This is an indicatorthat no change occurred in the crystal structure of the powderin the course of the reaction. We investigated the possiblereduction of Zr(IV) in the powder by means of the O
2-·
according to reaction R5 and the subsequent release of zirconium
in the solution where the reaction took place by measuringpossible traces of elemental zirconium in solution usinginductively coupled plasma spectroscopy (ICP). A blankmeasurement was previously performed on a sample from asuspension consisting of 1.5 g of zirconium dioxide in 50 mLof water and from which the solid particles were filtered aftera time of exposure in the solution that matched the reactiontime. The measurement to track the amount of zirconium
released during reaction was made after the reaction with H
2O2
reached the end point, corresponding to complete consumptionof the hydrogen peroxide initially present. The ICP spectroscopicmeasurement was performed after filtration of the solid particlesfrom the reactant solution, following a similar procedure andthe same conditions as the blank experiment. The measuredincrease in concentration of Zr in solution after the reactionwas (9.1 (0.2)×10
-8mol·dm-3. When comparing the amount
of zirconium in solution and the amount of hydrogen peroxidepresent that had reacted, one can conclude that the obtainedvalue for the concentration of zirconium in solution is negligible,and reaction R5, even if it would occur as a side reaction, wouldhave very little importance to the overall process in this system.However, the occurrence of reaction R5 causing a change in theoxidation state of Zr(IV) to other state besides Zr(0) could bethought to occur. The implication of the occurrence of this reactionin the crystal structure of the solid would be to create a defect inthe lattice by the replacement of a Zr(IV) atom by a Zr atom in adifferent oxidation state and consequent rearrangement of oxygenatoms to compensate for the non-neutrality in terms of the overallcharge of the new created lattice. This would translate in differentcell parameters before and after reaction. This is not observed, andso the reduction of zirconium by superoxide anion radical is notdetectable in this system.
In general, a continuously cycling surface catalyzed reaction can
be broken down into a short sequence of steps as follows:
44
The overall surface chemical reaction represents the sum of
the elementary steps mentioned above, each of them occurringwith different rate constants. When in the presence of a largeamount of active adsorbing sites when compared with the totalamount of adsorbate, the initial steep part of the curverepresenting the concentration of H
2O2as a function of reaction
time corresponds to a process dominated by the steps 1a and2a of Scheme 1, diffusion and adsorption of H
2O2into the
surface of ZrO 2, until an equilibrium of adsorbate in the surface
is reached. The period of time during which steps 1a and 2adominate the overall process can be reduced to a very shortperiod that is negligible when compared with the total time ofthe reaction. A behavior very close to zeroth-order kinetics canthen be obtained when the amount of H
2O2present is in large
excess compared with the number of adsorption sites availableinitially on the powder surface. In this way, it is possible to fitthe data to zeroth-order kinetics minimizing the error of such
Figure 4. ln(k) as a function of 1/ T(K) for T)[294.15 -353.15] K
with a temperature step of 5 K.
TABLE 1: Specific Surface Areas of the Fresh Powder Determined by the BET Method after Immersion in Water at 80 °C
and after Reaction with H 2O2at 80 °C
specific surface area (m2·g-1)
ZrO 2powder (grain size <5µm) fresh powder 5.0 (0.2
after immersion in water at 80 °C 5.0 (0.2
after reaction with H 2O2in aqueous media at 80 °C 5.0 (0.3SCHEME 1: Steps Involved in a Continuously Cycling
Surface Catalyzed Reaction
(1a) diffusion of adsorptive reactants to the active site on the
solid surface; (2a) adsorption of one or more reactants(adsorbates) onto the surface: if molecule A is chemicallyadsorbed onto one of the active sites, then a surface complex(S-A) is formed; (3a) surface reaction: A reacts forming the
products (B +C); (4a) desorption of products from the
surface: (B +C) escapes the site, thus regenerating site S;
and (5a) diffusion of products away from the surface.H
2O2Decomposition on the Surface of ZrO 2 J. Phys. Chem. C, Vol. 114, No. 25, 2010 11205

approximation. The lower limit of Zirconium mass, where the
reaction changes from pseudo-first-order to zeroth-order, whichimplies that the reaction rate constant will be independent ofthe initial concentration of H
2O2, was∼0.5 g, which corresponds
to a surface area of 2.5 m2. When the mass of zirconium was
<0 . 5gi n5 0m Lo f0 . 5m MH 2O2solution, zeroth-order kinetics
was approached. The zeroth-order rate constant obtained at T
)298.15 K with 0.224 g ZrO 2in 50 mL of 0.5 mM H 2O2
solution was k0)(2.0 (0.1)×10-5M·s-1.
The pH plays an important role in surface processes specif-
ically in the rate of uptake of an adsorbate by a surface.45In
the case of a system where hydrogen-bonded structures arepossible to form, the pH effect becomes even more important.Hydrogen peroxide is capable of forming stable cyclic hydrogen-bonded structures.
46At low pH values, metal oxide and
hydroxide surfaces tend to be positively charged, with an excessof protons bound to the surface, and thus these surfaces tend torepel positively charged ions and attract negatively charged ions.In the specific case of the system studied in this work, thisenvironment can trigger the formation of stable hydrogen-bonded clusters of hydrogen peroxide in solution and on thesurface, having the effect of stabilizing the hydrogen peroxide.At some intermediate pH value, the surface becomes chargeneutral. At the pH of the point of zero charge, which in thecase of ZrO
2is∼6.5,47electrostatic repulsion of a positively or
negatively charged ion would be minimized. At pH above thepoint of zero charge, the surface becomes negatively chargedbecause of the predominance of hydroxo (OH
-) or oxo (O2-)
groups on the surface. Under these conditions, a positivelycharged ion in solution would be attracted to the surface, but anegatively charged ion would be repelled. Anions show the
opposite behavior, with strong electrostatic attraction to metaloxide particle surfaces at low pH values and repulsion at highpH values. According to the proposed mechanism of thedecomposition of H
2O2on the surface of a solid, one of the
intermediate species formed is the superoxide radical (p Ka)
4.88). A change in the pH of the solution where the reaction takes
place can have the effect of altering the concentration of superoxideradical trapped on the surface by affecting the attractive/repulsiveforces between the superoxide radical and the surface and bypromoting or slowing down the rate of decomposition of the formeraccording to reaction R4. To evaluate the effect of pH changes onthe rate of decomposition of H
2O2by affecting the stability of
intermediate species, the zeroth-order rate was determined fordifferent pH values. The data obtained are represented in Figure5. As can be seen, the zeroth-order rate constant is linearlydependent on the pH of the solution. When considering the zeroth-order rate constant, where as described above the processes 3a and4a are rate-determining, one can have a picture of what is the effectin the reaction rate of the amount of superoxide radical present.Because at the range of studied pH values the deprotonation ofH
2O2and HO·radical are not to be considered because of their
higher p Kavalues, the pH effect translates mostly in the amount
of superoxide present on the surface. As expected when consideringthe amount of superoxide radical present on the surface as animportant factor affecting the reaction rate, the zeroth-order rate ishigher the less superoxide is present on the surface. Reaction R1is dependent on the number of available sites on the surface whereH
2O2can adsorb. The amount of superoxide accommodated on
the surface has an impact on the overall reaction rate. This factoccurs probably by alterations in the interactions between H
2O2
and the active sites in the surface due to the occupancy of the latterby the superoxide anion radical. Besides this, the fact that hydrogenperoxide forms stabilizing hydrogen-bonded clusters at lower pHvalues will increase the potential energy barrier for the reaction to
occur. This means that the rate at which the hydrogen peroxidereaches catalytic active surface sites will be diminished, causingthe rate constant to decrease. These two effects, hydrogen peroxidestabilization and superoxide presence in the surface, have to beconsidered when analyzing the dynamics of the reaction at differentpH values.
Mechanistic Study of the Decomposition of H
2O2on the
Surface of ZrO 2.A mechanistic study on the production of
HO radicals as intermediate species formed in the decompositionof H
2O2in the presence of ZrO 2according to reaction R1 was
performed. An assessment of the amount of hydroxyl radicalsproduced as intermediate species in the decomposition ofhydrogen peroxide was carried out. These studies involveddetermining the rate of formation of hydroxyl radicals andcompare its dynamics with the rate of consumption of hydrogenperoxide. The chemical yield for the formation of formaldehydeupon reaction of the hydroxyl radicals with tris buffer wasdetermined by quantifying the amount of formaldehyde pro-duced when a known amount of hydroxyl radicals are present
in the system. For that, a calibration curve was obtained byperforming an experiment where a known amount of hydroxylradicals was produced by radiolysis of water under γradiation
and where tris buffer was present triggering the formation offormaldehyde according to Scheme 2.
The method based on the Hantzsch reaction, introduced by
Nash
48for the spectrophotometric detection of formaldehyde,
has been used frequently in previous works. In this work,however, it was necessary to use a modified version of thismethod.
36The use of acetoacetanilide instead of acetylacetone
or 2,4-pentadione avoids interferences with H 2O2, which made
this technique possible to apply with good sensitivity to thesystem studied in this work.
Water radiolysis occurs when the water is exposed to ionizing
radiation. Upon radiolysis of water, HO
·,H 2O2,H 2O·,eaq-,H·,
and H 2are formed.49The main products formed in water radiolysis
are e aq-and HO·, which are formed in equal amounts. When the
solution is saturated with N 2O, the solvated electrons produced are
converted to HO·according to the following reactions50
eaq-+N2OfN2+O·-k)9.1×109L·mol-1·s-1
(R9)
O·-+H2OfHO·+OH-k)1.8×106L·mol-1·s-1
(R10)
The amount of hydroxyl radicals produced in water radiolysis
can be quantified from the dose rate of the radiation source and
Figure 5. Zeroth-order rate constant as a function of pH obtained at
298.15 K for 50 mL of solution, 0.5 mM in H 2O2, and 0.5 g ZrO 2. The
obtained correlation coefficient for the least-squares fit is equal to 1,and the associated error is equal to 5.16 ×10
-8.11206 J. Phys. Chem. C, Vol. 114, No. 25, 2010 Lousada and Jonsson

theGvalue (radiation chemical yield) for the hydroxyl radical
according to
C˙(OH·))D˙×G×F (11)
Where C˙(OH·) is the amount of OH radicals produced in
mol·dm-3·s-1,D˙is the dose rate of the radiation source and in
this case is equal to 0.15 Gy ·s-1,Gis the Gvalue for the
production of hydroxyl radicals and Fis the density of the
solvent which in this case is 1 as the solvent is water. Giventhat the solution was purged with N
2O gas, the Gvalue for the
production of hydroxyl radicals is equal to 5.5 ×10-7mol·J-1.51
From eq 11, one can obtain the amount of hydroxyl radicals
produced as a function of irradiation time C(HO·)t, where
C(HO·)thas the units of mol ·dm-3. The set of data obtained
for comparison of the amount of formaldehyde produced withthe amount of hydroxyl radicals present in solution uponirradiation of a solution 20 mM in tris/HCl buffer, pH )7.5 in
50 mL of H
2Oa t T)293.25 K, are represented in Figure 6.
The yield of the method when comparing the concentration offormaldehyde produced with a given concentration of hydroxylradicals present is 35%. This means that 35% of the hydroxylradicals present in the system react with tris buffer to produceformaldehyde. A result well below 100% would be expectedbecause according to Scheme 2, the reaction of the hydroxylradical with tris buffer can take place in other positions besidestheR-hydrogen atom of the alcohol group that results in the
formation of formaldehyde. Consequently, the reaction canfollow different pathways producing different compounds notdetectable by the modified Hantzsch method. However, the limitof detection obtained in this work for the detection of formal-dehyde was 0.5 µM, which corresponded to a concentration in
HO radicals equal to 1.43 µM. When comparing the limit of
detection of the method, obtained in this work for the detectionof formaldehyde, with a previously published value where nosolid was present, 0.1 µM,
36it can be asserted that this method
can be applied to systems similar to the one studied in this workwithout major changes in the detection limits caused byinterferences due to the presence of H
2O2and the solid oxide.
On the basis of the data from Figure 6, a study involvingdetermining the amount of formaldehyde and consequently the
amount of hydroxyl radicals produced during the course of thereaction between H
2O2and ZrO 2was performed. The reaction
was carried out in a suspension of 4.5 g ZrO 2at a temperature
of 293.25 K in 50 mL of H 2O with 20 mM tris buffer and 5
mM H 2O2. The pH was adjusted to 7.5 with HCl. Samples were
collected at different time intervals and filtered. Subsequently,1.5 mL of reactant solution was diluted in 2.5 mL of 4 Msolution of ammonium acetate and 1 mL of 0.2 M solution ofacetoacetanilide in ethanol. The modified Hantzsch reaction wasleft to react during 15 min at a temperature of 313.15 K. Theobtained set of data is represented in Figure 7.
According to the proposed reaction mechanism, the overall
stoichiometry of the reaction of H
2O2in the surface of ZrO 2
states that for each mol of H 2O2consumed, two moles of
hydroxyl radicals are produced. The determined concentrationof hydroxyl radical represented in Figure 7 lies around 1/10 ofthe concentration predicted by the reaction mechanism. This isdue to the competition between tris buffer and H
2O2to react
with the hydroxyl radical. The energetics of the bonds involvedin both reactions are approximately of the same magnitude, thecleavage of the O -H bond in tris buffer requires around
52431
kJ·mol-1of energy, the cleavage of the C -H bond requires
393 kJ mol-1, whereas the cleavage of the O -H bond in H 2O2
requires43429 kJ ·mol-1. Although the method used for the study
of the evolution of hydroxyl radicals with reaction time doesnot allow discriminating if the hydroxyl radicals are on thesurface of the powder or in solution when scavenged by thetris, previous studies show that for similar systems the hydroxylradicals can be trapped and stabilized on the surface of the metaloxide powder.
53The rate constant for H 2O2consumption
obtained with the presence of tris/HCl buffer at pH 7.5 wascompared with the rate constant obtained under normal condi-tions, no buffering system present, pH 7.0. The first-order rateconstant obtained for 313.15 K under normal conditionsproduced the value k
1)(1.22 (0.13)×10-4s-1, and the
first-order rate constant obtained with the presence of tri/HClbuffer at pH 7.5 produced the value k
1)(1.13 (0.70)×10-4
s-1. When taking the associated errors into account, the obtained
values are in good agreement, and it is possible to state that thebuffering system tris/HCl with a pH 7.5 is not affecting thereaction rate when compared with the reaction where no bufferis present.
4. Conclusions
The obtained activation energy for the reaction agrees with
previously published values for similar systems where differenttechniques were used for its determination. The mechanism of
Figure 6. Comparison between the rate of formation of hydroxyl
radicals in water radiolysis and the corresponding concentration offormaldehyde, obtained using the modified Hantzsch method.
Figure 7. Evolution in the concentrations of H 2O2and HO·as a
function of reaction time, in the reaction of H 2O2with ZrO 2.SCHEME 2: Formation of Formaldehyde by Means of
the Reaction of Tris(hydroxymethyl)aminomethane withHydroxyl Radical to Form Amongst Other Products,Formaldehyde and a Radical Species, Which CanFurther React to Form a More Stable Species
H2O2Decomposition on the Surface of ZrO 2 J. Phys. Chem. C, Vol. 114, No. 25, 2010 11207

the reaction of decomposition of H 2O2in the presence of ZrO 2,
although not completely elucidated, is confirmed to involve theformation of hydroxyl radicals as intermediate species, mostlikely sorbed to the surface of the solid. The dynamics of theformation of the HO radicals during the course of the reactionof decomposition of H
2O2was found to be in good agreement
with the predicted by the proposed reaction mechanism. Themodified Hantzsch method has proven an easy and sensitivemethod for the determination of the concentration of HO radicalsproduced as a byproduct of the decomposition of H
2O2. Because
the system is complex because of the possibility of innumeroussurface phenomena that could cause interferences with thereaction of production of formaldehyde and further with themodified Hantzsch reaction, one could expect that the sensitivityof this method would be reduced when applied to this type ofsystem. This was not verified, and the limit of detection obtainedin this work for the detection of formaldehyde is in the sameorder of magnitude of the limit of detection of the methodpreviously published and where no solid was present. It can beasserted that the modified Hantzsch method can be applied tosystems similar to the one studied in this work without majorchanges in the detection limits caused by interferences due tothe presence of H
2O2or the solid metal oxide.
Acknowledgment. The research described here was finan-
cially supported by the Swedish Centre for Nuclear Technology- SKC.
References and Notes
(1) Li, H. H.; Liu, S. Q.; Dai, Z. H.; Bao, J. C.; Yang, X. D. Sensors
2009 ,9, 8547.
(2) Ozawa, M. J. Alloys Compd. 1998 ,275, 886.
(3) Li, Q.; Yuan, L.; Ye, T.; Chen, Y. (University of China Science &
Technology). Patent CN101306370-A, 2009.
(4) Wulff-Doering, J.; Stichert, W.; Schueth, F. (BASF). Patent 849224,
1998.
(5) Fukunaga, T. (Idemitsu Kosan Co., Ltd. (IDEK)). Patent 1093852,
1999.
(6) Andreiotelli, M.; Wenz, H. J.; Kohal, R. J. Clin. Oral Implant. Res.
2009 ,20, 32.
(7) Niinisto, L.; Paivasaari, J.; Niinisto, J.; Putkonen, M.; Nieminen,
M.Phys. Status Solidi A-Appl. Res. 2004 ,201, 1443.
(8) Comte, M. J. M.; Comte, M. (Eurokera SNC). Patent 1957421,
2007.
(9) Sickafus, K. E.; Minervini, L.; Grimes, R. W.; Valdez, J. A.;
Ishimaru, M.; Li, F.; McClellan, K. J.; Hartmann, T. Science 2000 ,289,
748.
(10) Qiu, L. Y.; Guzonas, D. A.; Webb, D. G. J. Solution Chem. 2009 ,
38, 857.
(11) Radiation Effects in Solids ; Sickafus, Kurt E., Kotomin, E. A.,
Uberuaga, B. P., Eds.; Springer: Dordrecht, The Netherlands, 2007; Vol.235.
(12) Medek, J.; Weishauptova ´, Z. J. Nucl. Mater. 2009 ,393, 306.
(13) Cox, B. J. Nucl. Mater. 2005 ,336, 331.
(14) Motta, A. T.; Yilmazbayhan, A.; da Silva, M. J. G.; Comstock,
R. J.; Was, G. S.; Busby, J. T.; Gartner, E.; Peng, Q.; Jeong, Y. H.; Park,J. Y. J. Nucl. Mater. 2007 ,371, 61.(15) Wada, Y.; Uchida, S.; Nakamura, M.; Akamine, K. J. Nucl. Sci.
Technol. 1999 ,36, 169.
(16) Dishman, M. V.; Covey, D. A.; Baughan, L. W. Dent. Mater. 1994 ,
10, 33.
(17) Penna, T. C. V.; Ferraz, C. A. M. Infect. Control Hosp. Epidemiol.
2000 ,21, 499.
(18) Brodrecht, D. J.; Rusek, J. J. Appl. Energy 2003 ,74, 113.
(19) Hurlbert, E.; Applewhite, J.; Nguyen, T.; Reed, B.; Zhang, B. J.;
Wang, Y. J. Propul. Power 1998 ,14, 676.
(20) Pastina, B.; LaVerne, J. A. J. Phys. Chem. A 1999 ,103, 1592.
(21) Loveland, W. D.; Morrissey, D. J.; Seaborg, G. T. Modern Nuclear
Chemistry ; John Wiley & Sons: New York, 2006.
(22) Jonsson, M.; Nielsen, F.; Roth, O.; Ekeroth, E.; Nilsson, S.; Hossain,
M. M. EnViron. Sci. Technol. 2007 ,41, 7087.
(23) Hiroki, A.; LaVerne, J. A. J. Phys. Chem. B 2005 ,109, 3364.
(24) Lin, C. C.; Smith, F. R.; Ichikawa, N.; Baba, T.; Itow, M. Int.
J. Chem. Kinet. 1991 ,23, 971.
(25) Denisov, E. T.; Denisova, T. G.; Pokidova, T. S. Handbook of Free
Radical Initiators ; John Wiley & Sons: Hoboken, NJ, 2003.
(26) Takagi Junichi, I. K. Nucl. Sci. Eng. 1985 ,89, 177.
(27) Anpo, M.; Che, M.; Fubini, B.; Garrone, E.; Giamello, E.; Paganini,
M.Top. Catal. 1999 ,8, 189.
(28) Wardman, P.; Ross, A. B. Free Radical Biol. Med. 1991 ,10, 243.
(29) Anbar, M.; Neta, P. Int. J. Appl. Radiat. Isot. 1967 ,18, 493.
(30) Bahnemann, D. W.; Hoffmann, M. R. J. Electrochem. Soc. 1987 ,
134, C463.
(31) G. Choppin, J.-O. L.; Rydberg, J. Radiochemistry and Nuclear
Chemistry , 2nd ed.; Butterworth-Heinemann: Oxford, U.K., 1995; Vol. 46.
(32) Bailey, J. E. Proc. R. Soc. London, Ser. A 1964 ,279, 395.
(33) Hochanadel, C. J. J. Phys. Chem. 1952 ,56, 587.
(34) Ghormley, J. A.; Stewart, A. C. J. Am. Chem. Soc. 1956 ,78, 2934.
(35) Shiraishi, H.; Kataoka, M.; Morita, Y.; Umemoto, J. Free Radical
Res. 1993 ,19, 315.
(36) Li, Q.; Sritharathikhun, P.; Motomizu, S. Anal. Sci. 2007 ,23, 413.
(37) EN 717-2 ; European Committee for Standardization, Brussels, 1994.
(38) Food Chemicals Codex , 5th ed.; National Academy Press: Wash-
ington, D.C., 2003.
(39) Chirita, P. Chem. Biochem. Eng. Q. 2009 ,23, 259.
(40) Croiset, E.; Rice, F. S.; Hanush, G. R. Hydrogen Peroxide
Decomposition in Supercritical Water ; Wiley-Blackwell: Hoboken, NJ,
1997; Vol. 43.
(41) Murphy, M. D.; Griffiths, W. E.; Rowlands, C. C.; Hancock, E. F.;
Giamello, E. EPR Study of the H
2O2Interaction with TIO 2:EVidence for a
NoVel S)1 Surface Radical Pair ; Royal Society of Chemistry: Cambridge,
U.K., 1997.
(42) Jonsson, M. In Recent Trends in Radiation Chemistry ; Wishart,
J. F., Rao, B. S. M., Eds.; World Scientific: Singapore, 2010; p 301.
(43) McKay, D. J.; Wright, J. S. J. Am. Chem. Soc. 1998 ,120, 1003.
(44) Erbil, H. Y. Surface Chemistry Of Solid and Liquid Interfaces , 1st
ed.; Blackwell Publishing: Oxford, U.K., 2006.
(45) Nilsson, A.; Pettersson, L. G. M.; Nørskov, J. K. Chemical Bonding
at Surfaces and Interfaces ; Elsevier: Amsterdam, 2008.
(46) Kulkarni, A. D.; Pathak, R. K.; Bartolotti, L. J. J. Phys. Chem. A
2005 ,109, 4583.
(47) Kosmulski, M. J. Dispersion Sci. Technol. 2002 ,23, 529.
(48) Nash, T. Biochem. J. 1953 ,55, 416.
(49) Jacqueline Belloni, T. D. In Radiation Chemistry: From Basics to
Applications in Material and Life Sciences ; Mostafavi, M., Spotheim-
Maurizot, M., Eds.; EDP Sciences: Les Ulis Cedex, France, 2008.
(50) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J.
Phys. Chem. Ref. Data 1988 ,17, 513.
(51) Tabata, Y. Pulse Radiolysis of Irradiated Systems ; CRC Press:
Tokyo, 1991.
(52) Denisova, T.; Denisov, E. Kinet. Catal. 2006 ,47, 121.
(53) Giamello, E.; Calosso, L.; Fubini, B.; Geobaldo, F. J. Phys. Chem.
1993 ,97, 5735.
JP102893311208 J. Phys. Chem. C, Vol. 114, No. 25, 2010 Lousada and Jonsson