SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 11 January 2008
Published in Soil Sci Soc Am J 72:83-89 (2008)
DOI: 10.2136/sssaj2007.0064
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Elprince, A. M.
Right arrow Articles by El-Wakil, E. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Elprince, A. M.
Right arrow Articles by El-Wakil, E. M.
Agricola
Right arrow Articles by Elprince, A. M.
Right arrow Articles by El-Wakil, E. M.
Related Collections
Right arrow Soil Chemistry

SOIL CHEMISTRY

Kinetics of Nonenzymatic Decomposition of Hydrogen Peroxide by Torrifluvents

Adel M. Elprincea,*, Wafaa H. Mohameda and Ekhlas M. El-Wakilb

a Dep. of Soils and Agricultural Chemistry, Alexandria Univ., Alexandria, Egypt
b Dep. of Soils and Waters, Tanta Univ.,Egypt

* Corresponding author (aelprince{at}gmail.com).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Peroxide application has been proposed as a method to control soil aeration. The objectives of this study were to formulate the rate law, identify the inorganic catalyst, identify preferred active centers, and hypothesize a mechanism for the H2O2 decomposition reaction in Torrifluvents. The rate of decomposition of H2O2(aq) by Torrifluvents (10 locations in the Nile Delta) and synthetic Mn oxides was determined by measuring with a gas burette the volume of O2(g) given off. Heat pretreatment showed that the catalytic activity of the air-dried soil is essentially nonenzymatic on the time scale of the reaction. Pretreatment of the soil samples by NH2OH–HCl (pH 2), designed for selectively dissolving MnO2(s), completely deactivated their catalytic capacity. The pseudo-first-order rate coefficient, k*, for Torrifluvents was expressed by the equation: k* = ko[MnO2(s)](10–pH/4), where ko is a constant. Contrary to cryptomelane and pyrolusite, values of k* for birnessite satisfied the above equation, indicating that pedogenic birnessite was the active catalyst in Torrifluvents. Diethylenetriamine pentaacetic acid (DTPA) was found to be a more effective inhibitor than ethylenediamine tetraacetic acid (EDTA). A high value of the activation energy (E) was connected with a high value of the pre-exponential factor (A) of the Arrhenius equation. It was suggested that the birnessite surface has two kinds of active centers: MnIII/MnII and MnIV/MnIII. While the former is less numerous (lnA = 37 ± 4) with low activation energy (E = 75 ± 4 kJ mol–1), the latter is numerous (lnA = 62 ± 3) and of high activation energy (E = 144 ± 17 kJ mol–1). The Habes and Weiss mechanism explained the experimental results obtained.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
The concentration of O2(aq) is related to the pH + pe index in the soil solution (Elprince, 1986, p. 229). The O2(aq) in soil solutions drops to near zero within 24 h of submergence (Patrick and Sturgis, 1955). This rapid depletion of O2(aq) inhibits plant growth (Rowell, 1981) and activates anaerobes to reduce oxidized C, N, Mn, Fe, S, As, Cr, Cu, and Pd (Sposito, 1981). Calcium peroxide, CaO2(s), has been used as a seed coating (Westcott and Mikkelsen, 1983) to enhance the emergence of rice (Oryza sativa L.) seedlings from flooded soils, applied to eliminate unwanted odor in sewage systems (Hoffmann, 1977), and used in the aeration of wastewater (Akira and Mitsuzawa, 1987). Elprince and Mohamed (1992) proposed MgO2(s) as a potential O2–generating agent in soil for a longer reaction time scale relative to ionic peroxides. The H2O2(aq) produced on hydrolysis of CaO2(s) or MgO2(s) decomposes, releasing the desired O2(aq).

Soil catalase is an enzyme that catalyzes H2O2(aq) decomposition (Perez-Mateos et al., 1988). Johnson and Temple (1964) concluded that the actual catalase activity is very low and most of the peroxide decomposition is nonenzymatic. Some researchers have suggested that most of the catalytic capacity is due to Mn oxides in soils (Skujins, 1967).

Manganese oxides are one group of soil colloids that have been shown to be capable of sorbing a large amount of trace metals (McKenzie, 1989; Scheckel and Sparks, 2001; Kay et al., 2001); oxidizing Co(II) to Co(III) (Traina and Doner, 1985; Crespo and Lunar, 1997; Fendorf et al., 1999), Cr(III) to Cr(VI) (Bartlett and James, 1979; Negra et al., 2005), As(III) to As(V) (Scott and Morgan, 1995; Tournassat et al., 2002; Tani et al., 2004), and Pu(IV)or Pu(V) to Pu(V/VI) (Powell et al., 2006); and catalyzing organic matter formation (Bartlett and James, 1993).

Colloid-chemical properties of a synthetic MnO2(s) have been reported by Morgan and Stumm (1964), indicating that the zero point of charge (zpc) is somewhat below pH 3. Oscarson et al. (1983) presented transmission electron micrographs, x-ray diffraction data, and reported zpc values of 2.3, 2.8, and 6.4 for birnessite, cryptomelain, and pyrolusite, respectively, with a specific surface of 0.277, 0.346, and 0.008 km2 kg–1, respectively. The zpc of birnessite enables a net negative surface charge even under low-pH conditions, unlike many other soil sorbents such as Fe oxides or organic matter (Negra et al., 2005).

McKenzie (1971) stated that, among Mn oxides, birnessite appears to be the most common pedogenic form; it forms a very complex series, and various replacements or defects may occur in natural samples. The basic birnessite-type structure consists of Mn octahedral sheets separated by ~0.7- or ~1.0-nm interlayer regions filled with cations and water (Johnson and Post, 2006). Three oxidation states, namely Mn(IV), Mn(III), and Mn(II), have been recognized in birnessite. Manganese x-ray absorption near-edge structure spectra for air-dried soil samples have indicated that Mn sites are ~2/3 Mn(IV) and ~1/3 Mn(II), with minor amounts of Mn(III) (Schulze et al., 1995). Synchrotron powder diffraction data for Na birnessite indicated that Mn sites are ~2/3 Mn(IV) and ~1/3 Mn(III) with minor amounts of Mn(II) (Post et al., 2002). Birnessite types rich in Mn(III) seem to be the most common (Drits et al., 1997).

Torrifluvents are soils formed in recent alluvial deposits along waterways and deltas. When irrigated, the soils can be highly productive, as has been demonstrated in the Nile, Senegal, and Niger valleys in Africa, and the Indo-Gangetic Plain of Pakistan and India in Asia. In addition, Torrifluvents can be found in the Imperial and Mexicali valleys, with their extension into the Coachella valley, the Yuma area, and the floodplain at the mouth of the Colorado River in arid North America. Most of the Torrifluvents are moderately to slowly permeable and thus water logging is a major problem affecting their productivity and sustainability (Dregne, 1976; Soil Survey Staff, 1999).

The aim of this study was to extend the birnessite study by Elprince and Mohamed (1992) by investigating the catalytic decomposition kinetics of H2O2(aq) by Torrifluvents. The specific objectives were to formulate the rate law, identify the soil inorganic catalyst, identify preferred active centers, and hypothesize a mechanism for the reaction by examining the dependence of the pseudo-first-order rate coefficient (k*) on pH, temperature, and the catalyst, EDTA or DTPA, concentration.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
The Soil Samples
The soils used in this study were clayey, mixed, thermic Typic Torrifluvents. They were sampled from the 0- to 30-cm depth at 10 locations within the Nile Delta, Egypt. They were selected to obtain samples of a broad range in MnO2(s) contents. The Alexandria site was sampled (24 random soil samples from a 20-ha field) to determine the spatial variability of the rate coefficient in the field. The soil samples were air dried, ground to pass through a 2-mm sieve, and placed in closed containers. Soils from the Nile Delta are composed primarily of alluvial flood deposits transported from the volcanic Ethiopian Highlands. Montmorillonite, kaolinite, illite, quartz, and feldspar are the dominant minerals in these soils (Sheta et al., 1981). Some of the characteristics of the soils used are given in Table 1 . Calcium carbonate was determined manometrically (Page et al., 1982, p. 183), organic C by the Walkley–Black method (Page et al., 1982, p. 570), and texture by the pipette method (Klute, 1986, p. 404). Soil MnO2(s) was determined by atomic absorption spectrometer in H2O2–NaOAc (pH 5) extract (Jackson, 1956, p. 31).


View this table:
[in this window]
[in a new window]

  		Table 1. Some of the characteristics of the soils used.

 
Synthetic MnO2 Preparation
Birnessite, {delta}-MnO2(s), was prepared by adding concentrated HCl to a boiling solution of KMnO4; cryptomelane, K1/8MnO2(s), by adding a solution of KMnO4 to MnSO4 in acetic acid and heating to 80°C; and pyrolusite, β-MnO2(s), by evaporating a solution of Mn(NO3)2 nearly to dryness and then heating at 180°C (McKenzie, 1971; Oscarson et al., 1983; Elprince and Mohamed, 1992). The oxides were washed with deionized-distilled water and diluted to make 10 g L–1 suspensions. Cobalt-birnessite samples were prepared, with a fractional coverage ({propto}Co) ranging from 0.09 to 0.99 as outlined by Elprince and Mohamed (1992).

Soil Pretreatments
There were two pretreatments: (i) heating the soil samples at 120°C for 24 h, and (ii) selective dissolution of soil Mn oxides by NH2OH–HCl solution adjusted to pH 2, as outlined by Shuman (1982).

Aqueous H2O2 Decomposition Experiments
Analytical-grade H2O2 (8.8 mol L–1) was used to prepare a standard solution of 0.1 mol L–1. The solution was standardized before use by iodometric titration (Kolthoff and Sandell, 1948). An initial H2O2(aq) concentration of 60 mmol L–1 was selected for the kinetic experiments at the desired pH (tris buffer of 10 mmol L–1) with 0.25, 0.5, 1, 1.5, or 2 g soil or 50 or 100 mg Mn oxide in 25-cm3 aqueous solution volume. An initial concentration of 2, 5, 10, or 20 mol m–3 of ethylenediamine tetraacetic acid (EDTA) or diethylenetriamine pentaacetic acid (DTPA) was used in the reaction cells for some of the kinetic experiments. The volume of O2(g) given off, under constant temperature and controlled shaking, was determined using a gas burette. The suspension shaking was to eliminate film diffusion (Ogwada and Sparks, 1986). The temperature was controlled using a Julabo circulator (Julabo Labortechnik, Schwartzland, Germany) mounted to a closed water bath.

Preliminary Experiment on the Effect of Other Possible Reactions
Possible reactions other than the self-decomposition of H2O2(aq) were the oxidation of soil organic matter by H2O2 and the decomposition of soil CaCO3(s) and subsequent production of CO2(g). It was found in preliminary experiments that the volume of the evolved O2(g) in the presence of a CO2 trap [supersaturated solution of Ba(OH)2] was not different from that evolved in the absence of a trap. The final volume (Vf) of the evolved O2(g) (measured at 0.101 MPa pressure) agreed with the calculated volume (Vfc) from the initial concentration of H2O2 (using the ideal gas law) with an absolute error (|Vf – Vfc|100)/Vfc < 2%. Thus consumption of H2O2 in reactions other than its self-decomposition appeared to be insignificant on the time scale of the reaction (t1/2 = 0.693/k*).

Fitting the Kinetic Law to Experimental Data
The kinetics of chemical reactions in soils has been discussed by Sposito (1994, p. 58), who suggested that rate laws often expressed in a pseudo-first-order form:

Formula 1[1]
where X is a species whose reaction is monitored with time and k*(Env) is a "pseudo-first-order" rate coefficient that depends on one or more chemical "environmental factors," Env, which are held fixed as the reaction is monitored. Environmental factors could be the temperature, the solution pH, the ionic strength, and the concentration of an inhibitor.

A pseudo-first-order rate law was found applicable for the catalytic decomposition kinetics of H2O2(aq) by synthetic birnessite (Elprince and Mohamed, 1992). The integrated form of a pseudo-first-order kinetic law is

Formula 2[2]
where [H2O2]o is the initial concentration and k* is a pseudo-first-order rate coefficient. If Vf is the total volume of O2(g) evolved during the whole reaction and if V is the volume evolved up to time t, then [H2O2]o is proportional to Vf and [H2O2] is proportional to (Vf – V). Thus Eq. [2] can be rewritten in the form

Formula 3[3]
The fitting of Eq. [3] to data was done by the method of least squares (Daniel and Wood, 1971) to compute a value for k*, a standard error {sigma}k*, and a correlation coefficient square, R2. The R2 values in all kinetic experiments varied from 0.962 to 0.999.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Pretreatments
The heat pretreatment lowered the average value of k* ± {sigma}k* from (1.8 ± 0.3) x 10–3 to (1.7 ± 0.3) x 10–3 s–1 for the 24 soil samples from the Alexandria 20-ha field. The heat pretreatment lowered the catalytic activity by only 6%. On the other hand, pretreatment of the soil samples by NH2OH–HCl (pH 2), designed for selectively dissolving MnO2(s), completely deactivated their catalytic capacity. The NH2OH–HCl (pH 2) treatment causes selective dissolution of birnessite together with other Mn oxide forms as revealed by x-ray diffraction analysis (Tokashiki et al., 1985). We therefore concluded that the catalytic activity of the air-dried soils is, essentially, nonenzymatic on the time scale of the H2O2(aq) decomposition reaction, and soil MnO2(s) is the active catalyst.

Effect of Manganese Dioxide Content in Soil on the Decomposition Rate
As the concentration of MnO2(s) in the reaction cell increased from 54 to 637 mmol m–3, the k* value increased from (7.9 ± 0.1) x 10–5 to (1.11 ± 0.05) x 10–3 s–1 for H2O2(aq) decomposition by soil samples (Table 2 ). Assuming that k* varies with [MnO2] according to the equation k* {propto} [MnO2]m, the order of the reaction with respect to [MnO2] was found to be 0.97 ± 0.15 (Fig. 1 ), making m ~ 1. Plotting values of k* for synthetic birnessite, cryptomelane, and pyrolusite (Table 2) on the log k* vs. log[MnO2] plot reveals that, among the three Mn oxides, birnessite is the closest to the extrapolated line (Fig. 1). Thus soil birnessite seems to be the active catalyst in the inorganic catalytic decomposition of H2O2(aq) in Torrifluvents. This is in agreement with McKenzie (1971), who stated that, among Mn oxides, birnessite appears to be the most common pedogenic form.


View this table:
[in this window]
[in a new window]

  		Table 2. The pseudo-first-order rate coefficient (k*) values as a function of MnO2(s) concentration for the decomposition of H2O2(aq) at an initial H2O2 concentration of 60 mol m–3, pH 7.6, and temperature 30 ± 0.2°C.

 

Figure 1
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 1. Log of the pseudo-first-order rate coefficient (k*) vs. the concentration of MnO2(s) in the reaction cell for the decomposition of H2O2(aq) by Torrifluvent soil samples (T) and three synthetic Mn oxides: birnessite (B), cryptomelane (C), and pyrolusite (P); initial H2O2(aq) concentration and pH equal to 60 mmol m–3 and 7.6, respectively, at 30 ± 0.2°C.

 
The measured k* and MnO2(s) content values from the Alexandria site (24 random soil samples from a 20-ha field) bring out two important features. First, the observed spatial variation in k* is normally distributed, with a mean and standard deviation equal to 1.8 x 10–3and 0.3 x 10–3 s–1, respectively. On the other hand, the observed variations in the k* of birnessite (k*B) [also MnO2(s) content] are more accurately described as lognormal than normal (Analytical Software, 1996). The mean, standard deviation, mode, and median are 1160, 470, 1120, and 1140 s–1 kg–1 birnessite for k*; and 777, 281, 720, 740 mg kg–1 for [MnO2(s)]. The second is the marked contrasts in the variability pattern of k* compared with k*B and, hence, the number of soil samples required to estimate their means at different levels of precision.

Effect of pH on the Decomposition Rate by Soil
Figure 2a shows the values of ln[(Vf – V)/Vf] against t for the decomposition of H2O2(aq) by the Etay Elbaroud soil (40 kg m–3) at 30.0 ± 0.2°C and pH 6.0, 6.5, 7.0, 7.5, and 8.0. Figure 2b shows the calculated values of log k* against the pH. The value of k* increases from 2.4 x 10–4 to 1.0 x 10–3 s–1 as the pH value increases from 6.0 to 8.0. Assuming that k* varies with [H+] according to the equation k* {propto} [H+]h, the order of the reaction with respect to [H+] is 0.26 ± 0.01 (Fig. 2b), making h ~ 1/4. A similar value for h has been reported for synthetic birnessite (h = 0.23 ± 0.04 ~ 1/4) (Elprince and Mohamed, 1992). Combining this result with the above-mentioned dependence of k* on [MnO2], one obtains

Formula 4[4]
where [MnO2] is in the range 50 to 640 mmol m–3 and pH is in the range 6.0 to 8.0 at 30 ± 0.2°C.


Figure 2
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 2. Effect of pH on H2O2(aq) decomposition by the Etay Elbaroud soil sample (40 kg m–3) for an initial H2O2 concentration of 60 mmol L–1 at 30.0 ± 0.2°C: (a) ln(Vf – V)/Vf vs. time, where Vf is the total volume of O2(g) evolved during the whole reaction and V is the volume evolved up to time t, and (b) log pseudo-first-order rate coefficient (k*) vs. pH.

 
Effects of EDTA and DTPA on the Decomposition of Hydrogen Peroxide
If the reaction H2O2 + Mn2+ -> Mn3+ + OH– + OH is a rate-determining step, decomposition will be retarded by the addition of a complexing agent (L). If the complex MnL is assumed to be in thermodynamic equilibrium with the reactants, we can assume that k* {propto} KMnL[L]m, where KMnL is the dissociation constant of the complex made of Mn2+ and the ligand L. The concentration of L (2, 5, 10, or 20 mol m–3 of EDTA or DTPA) was in excess as the H2O2(aq) decomposition reaction was monitored. Figures 3a and 3b show that log k* varies linearly with –log[L] for EDTA and DTPA, respectively, for Etay Elbaroud soil (40 kg m–3) at pH 7.6 and 30 ± 0.2°C.


Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 3. Log of the pseudo-first-order-rate coefficient (k*) vs. the negative log of the concentration (pL) of (a) EDTA and (b) DTPA for the decomposition of H2O2(aq) by Etay Elbaroud soil.

 
The slopes of the lines in Fig. 3 yield m ~ –1/4 for EDTA and m ~ –1/2 for DTPA. The intercepts yield KMnEDTA/KMnDTPA ~ 8, which is close to the dissociation constant ratio reported by Lindsay (1979, p. 234) for the complexes MnEDTA and Mn2DTPA, namely 10–16.91/10–17.60 ~ 5.

Effect of Temperature
The Compensation Effect
The pseudo-first-order rate coefficient, k*, is expressed by the Arrhenius equation: k* = A exp(–E/RT), where E and A are the activation energy and the pre-exponential factor, respectively, R is the gas constant, and T is the absolute temperature. The E values calculated from our data range from 69 to 147 kJ mol–1 (Table 3 ). Values of E >42 kJ mol–1 indicate surface-controlled reactions for reactions free of physical (diffusion and thermal) resistances (Ogwada and Sparks, 1986). We can therefore conclude that the E values calculated from our data suggest that the H2O2 decomposition process is not a diffusion process but a chemical reaction controlled process proceeding on the surface of the birnessite catalyst.


View this table:
[in this window]
[in a new window]

  		Table 3. Activation energy (E) and pre-exponential factor (A) for H2O2 decomposition. Temperature range was 20 to 40°C.

 
As seen in Table 3, a high value of E is connected with a high value of A. The existence of such a relation is the more surprising since it would seem that there should be no relation between E and A. The existence of such a law, however, leads to very great difficulties in the comparative assessment of the activity of catalysts. Indeed, the transition to a catalyst having a higher activation energy ought to lead to a lower overall activity. But a simultaneous and, as a rule, proportional growth of the pre-exponential factor levels out the action of the activation term. The most rational explanation of this compensation effect was given in Pachenkov and Lebedev (1976, p. 470). According to these researchers, the probability of the entire surface of a catalyst being active is quite small, and what we have to deal with are active centers. For statistical thermodynamic reasons, the active centers should be more numerous (high value of A) with decreasing bond energy (high value of E). It thus becomes clear why a high activation energy is connected with a high value of the pre-exponential factor.

We have examined reaction rate theories for a clue about the functional relationship between E and A. It is the Rice–Ramsperger–Kassel–Marcus model (Masel, 2001, p. 381) that gives the dependence of A on E:

Formula 5[5]
where k is the Boltzmann constant (1.380658 x 10–23 J K–1), h is Plank's constant (6.6260755 x 10–34 J s), T is absolute temperature in Kelvin, v is the number of vibration modes that can store energy, q!=ABC is the partition functions per unit volume for the hot reacting complex A–B–C, qA and qBC are the partition functions per unit volume for the reactants A and BC. The value of v is related to the number of atoms (N) that form the nonlinear molecular structure for the transition state by the equation N = (v + 7)/3.

Fitting Eq. [5] to the experimental E–A data (Table 3) gives the equation

Formula 6[6]
where A and E are in s–1 kg–1 birnessite and kJ mol–1, respectively, v = 38 ± 2, and N = 15 ± 2. It is hard to specify the chemical formula of the transition-state molecule based on the observed N value only. In addition, the high value of the standard error in the intercept prevents a reasonable estimation for the partition functions ratio.

The Cobalt Poisoning Effect
As stated above, three oxidation states, namely Mn(IV), Mn(III), and Mn(II), have been recognized in birnessite (Schulze et al., 1995; Drits et al., 1997; Post et al., 2002). Thus, three possible redox sites may exist, namely Mn(IV)/Mn(III), Mn(III)/Mn(II), and Mn(IV)/Mn(II). Thermodynamics indicates that all three redox couples can be catalysts for the H2O2(aq) decomposition reaction, but not the Co(III)/Co(II) redox (Elprince and Mohamed, 1992). The data in Table 3 shows that for Co-fractional coverage, {propto}Co ≤ 0.19, Co(II) is poisoning the birnessite (k* becomes smaller). A possible pathway is the blocking of the active centers Mn(III)/Mn(II) (not Mn(IV)/Mn(II) and Mn(IV)/Mn(III), as will be shown below) by substitution of octahedral Mn(III) by Co(III) and reduction of Mn(III) to exchangeable Mn(II):

Formula 7[7]
Several experimental studies provide support for this pathway. Murray and Dillard (1979), using x-ray photoelectron spectroscopy, presented the first direct chemical evidence for the formation of Co(III) when Co(II) is sorbed by birnessite. Traina and Doner (1985) presented mass balance evidence suggesting that Mn release during Co(II) sorption results not only from the oxidation of Co(II) to Co(III), but also from a direct exchange of Co(II) for Mn(II) produced during the redox reaction.

On the other hand, for {propto}Co ≥ 0.48 (Table 3), the values of E and lnA seem constants independent of {propto}Co, i.e., Co(II) becomes useless for poisoning. At this stage, H2O2 is catalyzed by active centers that are unreachable by Co(II) and at the same time are numerous in number, as indicated by their high mean lnA value (Table 3). Below we will show that Mn(IV)/Mn(III) sites satisfy both conditions for a birnessite rich in Mn(III). As stated above, birnessite types rich in Mn(III) seem the most common (Post et al., 2002; Drits et al., 1997).

The Mn(IV)/Mn(II) sites do not react with Co(II) because its Mn(II) is in the edge-shared [MnO6] octahedron while Co(II) requires a tetrahedral combination. In the meantime, the Mn(IV)/Mn(III) sites do not react with Co(II) because they cannot be reduced by Co(II) and at the same time release Mn(II)adsorbed as required by the Traina and Doner (1985) experimental observations. Since birnessite is often rich in Mn(III) (Post et al., 2002), Mn(IV)/Mn(III) sites are the numerous while Mn(IV)/Mn(II) sites are less numerous. Thus we can infer that Mn(IV)/Mn(III) sites are active centers for the catalytic decomposition of H2O2 by birnessite.

Based on the above thermodynamics and structure arguments, the birnessite surface seems to have two kinds of active centers, namely Mn(III)/Mn(II) and Mn(VI)/Mn(III), capable of catalyzing the H2O2 decomposition reaction. Based on the A and E data in Table 3, we can assign A and E values for both active centers. When 0.48 ≥ {propto}Co ≥ 0.99, the catalytic activity of birnessite seems entirely due to the Mn(VI)/Mn(III) sites, which are numerous (mean lnA = 62 ± 3) and of high E value (mean E = 144 ± 17 kJ mol–1). On the other hand, the catalytic activity in unpoisoned birnessite ({propto}Co = 0) and pedogenic birnessite in bulk soil seems essentially due to the Mn(III)/Mn(II) active centers, which are less numerous (mean lnA = 37 ± 4) and of low E value (mean E = 75 ± 4 kJ mol–1).

The Reaction Mechanism
Habes and Weiss in 1934 (cited by Ahuja et al., 1988) proposed the following mechanism for the decomposition of H2O2(aq) by metal ions of variable valency (Mez):

Formula 8A[8a]

Formula 8B[8b]
The radicals then enter into a chain reaction with H2O2(aq) to form O2(g) and H2O(l) (Pachenkov and Lebedev, 1976, p. 470). Elprince and Mohamed (1992) reported the applicability of this mechanism for the decomposition of H2O2(aq) by Mnz+1/Mnz active centers of synthetic birnessite. The present experimental results indicate that the Habes and Weiss mechanism seems applicable for Torrifluvents as well. This conclusion is based on the following: (i) the E value is relatively high, compared with diffusion processes, indicating a surface-controlled reaction (Table 3); (ii) since Mn(II), Mn(III), H+, and OH– are in the elementary reactions Eq. [8a] and [8b], the rate coefficient k* is found to be [MnO2] and pH dependent (Eq. [4]); and (iii) since EDTA and DTPA form complexes with Mn(II) of the elementary reactions, the k* value decreases as the chelating agent concentration increases, and DTPA is found to be a more effective inhibitor than EDTA, as expected, due to the difference in their dissociation constants with Mn(II) (Fig. 3).

In summary, this study resulted in five important findings regarding the catalytic decomposition of H2O2(aq) in Torrifluvents. First, a pseudo-first-order law was found to be applicable for H2O2(aq) decomposition in Torrifluvents. Second, the birnessite mineral was found to be the active inorganic catalyst in Torrifluvents. Third, the birnessite surface seems to have two types of active centers: Mn(III)/Mn(II) and Mn(IV)/Mn(III), characterized by lnA values of 37 ± 4 and 75 ± 4 s–1 kg–1, respectively, and E values of 62 ± 3 and 144 ± 17 kJ mol–1, respectively. The fourth important finding is the compensation effect (the dependence of A on E), interpreted using statistical thermodynamics and the Rice–Ramsperger–Kassel–Marcus reaction rate theory. Fifth, the Habes and Weiss mechanism seems to explain the dependence of the H2O2 decomposition rate on environmental factors in Torrifluvents.

These findings can be of practical significance. As proposed by Elprince and Mohamed (1992), peroxide application could be a method to control soil aeration in the root zone. A representative k* value for H2O2(aq) decomposition in Torrifluvents is 2.3 x 10–4 s–1. Subsequently, the time for complete (99%) conversion of H2O2(aq) is 4.6/k* = 5.5 h. Soils that consume O2(g) at rates of 0.12 to 0.35 cm3 m–3 s–1 (Wesseling, 1974; Rowell, 1981) need 22 to 45 L of H2O2 (8.8 mol L–1) to generate the required O2(g) for the top 0.1-m layer of 1 ha of cropped soil and would last for 5.5 h. The decomposition of H2O2(aq) in soil is undoubtedly not as simple as implied by this computation. The results at least suggest, however, that H2O2(aq) could be a potential O2–generating agent in Torrifluvents in case a rapid O2 supply is required. This would be the case, for instance, (i) in transient periods of water logging because of heavy rainfall exceeding the ability of the upper soil layers to drain, (ii) where water tables rise into the root zone, (iii) in the presence of crops susceptible to anoxic conditions if they coincide with critical stages of crop growth, and (iv) where the accumulation of toxic substances such as NO2–, Mn2+, Fe2+, and S– together with microbial metabolites exist in concentrations injurious to root metabolism. In the meantime, looking for practical methods to increase the time scale of the H2O2(aq) decomposition reaction is highly desirable. Also, while soil birnessite may get chemically poisoned, Mn oxide depositing fungus (Tani et al., 2004) probably could be a potential catalyst-generating agent in bulk soil.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication February 15, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Elprince, A. M.
Right arrow Articles by El-Wakil, E. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Elprince, A. M.
Right arrow Articles by El-Wakil, E. M.
Agricola
Right arrow Articles by Elprince, A. M.
Right arrow Articles by El-Wakil, E. M.
Related Collections
Right arrow Soil Chemistry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome