Acceleration of Ammonium Nitrite Denitrification by ... - ACS Publications

A reaction of ammonium nitrite in ice was investigated. Upon freezing, some nitrite is oxidized by dissolved oxygen and some nitrite reacts with ammon...
0 downloads 0 Views 960KB Size
ARTICLE pubs.acs.org/JPCA

Acceleration of Ammonium Nitrite Denitrification by Freezing: Determination of Activation Energy from the Temperature of Maximum Reaction Rate Norimichi Takenaka,*,† Itaru Takahashi,† Hiroshi Suekane,† Koji Yamamoto,‡ Yasuhiro Sadanaga,† and Hiroshi Bandow† †

Laboratory of Environmental Chemistry, Department of Applied Chemistry, Graduate Scool of Engineering, Osaka Prefecture University, 1-1 Gakuen-cho, Naka-ku, Sakai-shi, Osaka 599-8531, Japan ‡ Osaka Prefecture University College of Technology, 26-12 Saiwai-cyo, Neyagawa-shi, Osaka 572-8572, Japan ABSTRACT: A reaction of ammonium nitrite in ice was investigated. Upon freezing, some nitrite is oxidized by dissolved oxygen and some nitrite reacts with ammonium to produce nitrogen and water in a denitrification reaction. The former reaction was accelerated only during freezing, and the latter one was accelerated even after the whole sample was frozen. The denitrification reaction proceeded at very low concentration in ice, which were conditions under which the reaction would not proceed in solution. The nitrogen production increased linearly with increasing initial concentration of ammonium nitrite. The concentration factor in the unfrozen solution in ice was estimated to be 50.6 when the initial concentration was 0.5 mmol dm3, as obtained from comparison of reaction rates in solution and in ice. A new method for determination of the activation energy is proposed that gives a value of 53 to 61 kJ mol1 for denitrification. The reaction order of the denitrification process is also determined using our method, and it is concluded to follow third-order kinetics.

’ INTRODUCTION It is not an uncommon phenomenon that reaction rates in ice are faster than those in solution.119 In most cases, the freezeconcentration into unfrozen solution in ice (when each small region of unfrozen solution existing in ice is indicated, we use the term “micro-pocket”) is the main reason for the increase in reaction rate. Pincock reported an acceleration theory for the freezeconcentration effect.20 Takenaka and Bandow also applied that theory to the freeze-concentration effect.21 It was shown that reactions that follow zero- and first-order kinetics, reactions of high concentration, and reactions having extremely high activation energies are not accelerated in ice. Reactions that follow second- and higher-order kinetics, reactions of low concentration and reactions having low activation energies can be accelerated greatly in ice. Concentrated ammonium nitrite is well-known to decompose to produce N2 in a chemical denitrification reaction, and this reaction is used for laboratory N2 generation (reaction 1).2227 NH4 NO2 ðaqÞ f N2 ðgÞ þ 2H2 O

salt effect and the effect of nitrous acid decomposition are both neglected, it is not possible to distinguish between the rate law proposed by Abel and that proposed by Dusenburry and Powell. If the ionic strength, or activity, of the solution is considered, then the rate calculated by the rate law of Dusenburry and Powell approaches the same value calculated by the Abel rate law. At low pH, nitrous acid decomposes through several reaction pathways. One of them results in the production of NO and HNO3 (reaction 3). The production of NO2 by reaction 4 is also possible but was not observed in the decomposition of ammonium nitrite.27 Ewing reported N2O production from the decomposition of ammonium nitrite (reaction 5), and the ratio of N2O: N2 produced was 2% and 20% at pH 4.29 and pH 5.03, respectively.24

ð1Þ

Abel proposed a third-order rate law for reaction 1 as follows:26 d½N2 =dt ¼ k½NH4 þ ½HNO2 ½NO2  

ð2Þ

However, Dusenburry and Powell proposed a second-order rate law.23 Ewing and Bauer24 demonstrated that when the secondary r 2011 American Chemical Society

3HNO2 f 2NO þ HNO3 þ H2 O

ð3Þ

2HNO2 f NO þ NO2 þ H2 O

ð4Þ

4HNO2 f N2 O þ 2HNO3 þ H2 O

ð5Þ

2HNO2 þ O2 f 2Hþ þ 2NO3 

ð6Þ

Received: September 27, 2011 Revised: November 14, 2011 Published: November 14, 2011 14446

dx.doi.org/10.1021/jp2093466 | J. Phys. Chem. A 2011, 115, 14446–14451

The Journal of Physical Chemistry A At low pH, nitrous acid is oxidized by dissolved oxygen to produce nitrate (reaction 6). This reaction is second order in nitrous acid and first order in oxygen.28,29 Nitrite ion cannot serve as a reactant for this oxidation reaction, and therefore, the reaction rate is fast in acidic solution but very slow at neutral and high pH. An increase in the reaction rate of chemical denitrification in freezing soil moisture due to the freeze-concentration effect was reported by Christianson and Cho.30 They observed N2O and N2 as products using mass spectrometry and showed that the N2 production rate was approximately first order with respect to initial nitrite concentration. Furthermore, they reported activation energies for the denitrification reaction of 45.7 and 52.8 kJ mol1 for unfrozen soil and frozen soil, respectively. The activation energy of reaction 1 in solution at 450 °C was investigated by Nguyen et al., who reported a value of 65.8 ( 6.7 kJ mol1.31 This value is different from that in soil solution, reinforcing the idea that the reaction mechanisms in soil solution and aqueous solution may differ. Nikolic and Hultman reported an increase in the chemical denitrification reaction rate in freezing aqueous solution.32,33 They reported that the reaction proceeded with greatest efficiency at pH 4.0. In this work, we investigated the great acceleration of chemical denitrification in ice in detail. Furthermore, we have proposed a new method to determine activation energies of reactions, and from these results we determined the reaction order for the denitrification reaction.

’ EXPERIMENTAL SECTION All reagents, obtained from Wako Pure Chemicals, Inc., were reagent grade and used as received. Pure water was prepared by Milli-Q Labo using distilled water. The resistivity of the pure water was higher than 18.2 MΩ cm. A solution of ammonium sulfate and sodium nitrite was prepared so that ammonium and nitrite are equi-molar. An aliquot of this solution was put into a polypropylene syringe sealed with a silicon stopper and frozen in a coolant, normally at 258 ( 1 or 253 ( 1 K. The pH of the solution was adjusted using sulfuric acid. The pH was measured with an M-13 Horiba Co. Ltd. pH meter with a 606910C glass electrode, and the pH meter was calibrated with a phosphate buffer (pH 7) and phthalate buffer (pH 4). After the sample was frozen, it was kept in the coolant for a few seconds to 24 h. The sample was then thawed, and the concentrations of ammonium, nitrite, and nitrate were measured using a Dionex ion chromatography system, ICS-1500, or a Yokogawa Analytical Systems Co., Ltd., IC-7000 ion-chromatographic analyzer. For analysis of the inorganic anions a Dionex AS-12A column with 9 mM Na2 CO3 eluent was used, and for analysis of the cations a Yokogawa Analytical Systems, ICD-C25 column was used with a mixture of 24 mM boric acid, 5 mM L-(+)-tartaric acid, 1 mM 2, 6-pyridine dicarboxylic acid as the eluent. In order to confirm the composition of gas emitted during freezing, the gas composition in a syringe was measured with a Pfeiffer Vacuum Co. Omnistar GSD300 O1 mass spectrometer. For comparison, the same solution was kept at 4 or 20 °C for the same time as the freezing experiments. ’ THEORY Generally, a reaction rate in ice is measured after a sample is thawed and is called the observed reaction rate, Rm. Rm has been theoretically evaluated,20,21 and it is generalized by the

ARTICLE

following equation. Rm ¼ Aexpð  Ea =RTÞ

Cfp CT

!a ½A

Cfp CT

!b ½B

CT Cfp

ð7Þ

Here, A, Ea, CT, and Cfp are a pre-exponential factor, activation energy, total concentration of solutes in an original solution and total concentration in micropocket, respectively. [A], a, [B], and b represent the concentration of reactant A, the reaction order of A, the concentration of reactant B and its reaction order, respectively. If Rm is differentiated with respect to temperature and the solution is set to zero, then the temperature at which the maximum reaction rate, Tmax is observed can be determined. In eq 7, exp(Ea/RT) and Cfp are temperature dependent, resulting in the following relations. !   dRm 1 Ea a b aþb1 aþb1 0 ¼ A½A ½B þ ðCfp Þ Cfp 2 CT dT RTmax  expð  Ea =RTmax Þ ¼ 0 Ea ð8Þ Cafpþ b  1 þ ðC afpþ b  1 Þ0 ¼ 0 2 RTmax If the temperature dependence of Cfp and the reaction orders are known, then the activation energy can be determined. Equation 8 can be expressed as eqs 9 and 10 for the secondorder and third-order reactions for a single reactant A, respectively, if Cfp is assumed to be equal to molality, and if a linear relationship between freezing point depression and molality is assumed to be established, that is, Cfp ≈ (Tfp  T)/Kfp. Here, Kfp indicates the freezing-point depression constant. Ea Ea Tmax  Tfp ¼ 0 ð9Þ R R   Ea Ea 3 2 2 Ea  2Tfp Tmax ¼0 2Tmax þ  2Tfp Tmax þ Tfp R R R ð10Þ

2 Tmax þ

As shown in eqs 9 and 10, Tmax depends only on the activation energy and reaction order. If the relationship between Cfp and temperature are known, then the activation energy for any reaction can be determined.

’ RESULTS AND DISCUSSION Denitrification in Ice. The chemical denitrification reaction proceeds at room temperature when the concentration is very high, on the order of several mol dm3. In this case, generation of N2 bubbles in the solution can be observed visually. However, the reaction rate at low concentrations, such as 10 mmol dm3, is very slow and visual observation of N2 bubbles cannot be confirmed. The reaction rate is first order in nitrous acid, nitrite ion and ammonium ion, giving a third-order rate overall. This means that a reaction rate decreases by a factor of 109 when the concentration is diluted by a factor of 1000. However, when a 10 mmol dm3 ammonium nitrite solution was frozen at 258 K for 24 h, several cm3 of gas were produced from a 50 cm3 sample. As a first step, we confirmed that the denitrification reaction actually proceeds in ice. Figure 1 shows the ammonium, nitrite, and nitrate concentrations after 72 h in solution at 277 K and in ice at 258 K. Dissolved oxygen was eliminated prior to each experiment by N2 bubbling 14447

dx.doi.org/10.1021/jp2093466 |J. Phys. Chem. A 2011, 115, 14446–14451

The Journal of Physical Chemistry A

Figure 1. Ion concentrations after 72 h in solution at 277 K, and in ice at 258 K. The initial concentration of ammonium nitrite was 50.0 mmol dm3 at pH 4.0. Dissolved oxygen in the sample solution was eliminated prior to each experiment by bubbling N2 for more than 5 min. Twenty cm3 of the sample solution was put into a syringe, the headspace was eliminated from the syringe and the syringe was sealed with a silicon rubber plug. Then, the samples were frozen at 258 K or left at 273 K for 72 h. The volume of gas generated in ice was about 12 cm3. Gray, white and black bars show NH4+, nitrite and NO3, respectively.

in order to suppress an oxidation reaction of nitrous acid by dissolved oxygen. In solution, the ammonium concentration and the nitrate plus nitrite concentration remained almost unchanged after 72 h, and a small amount of nitrate was produced as a result of reaction 3 or 5. In ice at 258 K, about 50% of ammonium and also about 50% of nitrite plus nitrate decreased. In this experiment, the sample was put in a sealed polypropylene syringe with no head space, and therefore, any gaseous components were concluded to derive from the reactions themselves. The possible gaseous components include N2, HONO, NO, NO2, N2O, and NH3. A strong signal was detected at m/e of 28, corresponding to molecular nitrogen. All of the nitrogen present in the gaseous phase were generated by the denitrification reaction because on thawing any nitrogen expelled from the original solution would redissolve. A small signal at m/e of 32 was also detected, but this could come from dissolved oxygen that was not removed by the N2 bubbling step. A very small signal at m/e of 30 was detected, corresponding to NO, NO2, and HONO.34 Compared with the intensity of the signal at 28, the signal at 30 was much smaller and was therefore neglected. Signals at 17 (NH3), 44 (N2O or CO2) and 46 (NO2) were not detected. In solution24,31 and from frozen soil moisture,31 it was reported that N2O could be detected. This discrepancy is probably due to the difference in pH and temperature between the reactions. From the calculation of gas volume and the concentration of ammonium ion in the original sample, more than 98% of the evolved gas can be explained by N2 production from denitrification. We concluded that the denitrification reaction and nitrite oxidation (when dissolved oxygen was present) proceeded in ice, and the other reactions were negligible over the course of 72 h. The effect of N2 bubbling on the reaction rate of the denitrification reaction in ice can be neglected since the equilibrium constant of reaction 1 calculated by using a standard free energy of the reaction (345.7 kJ mol1 35) is 4.0  1060, and also since the concentration of dissolved nitrogen in the micropocket is expected to be an equilibrium one21 and is much smaller than N2 produced by reaction 1. Time profiles of the change in nitrite, nitrate, and nitrogen concentrations when dissolved oxygen was not removed are shown

ARTICLE

Figure 2. Ion concentration as a function of time in a sample frozen at 253 K. The initial concentration of ammonium nitrite was 5.0 mmol dm3 at pH 4.0. Dissolved oxygen in the sample solution was equilibrated with ambient air. Several ten cm3 samples were frozen at 253 K, and one of the sample was thawed at the desired time, and ion concentrations were determined. Circles, squares, and triangles show nitrite, NO3 and N2, respectively. N2 was calculated from ammonium concentration, and the concentrations are converted to those in solution.

in Figure 2. The reaction of nitrite with dissolved oxygen on freezing is known to be very fast and is complete before the entire sample is frozen.9,10,12 Therefore, the changes in nitrite and nitrate at early times were as expected and are shown with dashed lines in Figure 2. In this case, concentration of dissolved oxygen in the unfrozen solution in ice is estimated to be an equilibrium one21 and 0.67 mmol dm3.35 Nitrite oxidation was almost complete when the entire sample was frozen. However, the chemical denitrification reaction was slower, and it was observed that the ammonium and nitrite concentrations gradually decreased in ice. In Figure 2, the production of nitrogen is calculated from the decrease in the concentration of ammonium and is shown to correlate with the initial concentration in solution. At 5.0 mmol dm3 of ammonium nitrite, more than 50% of the ammonium ion was converted to nitrogen. As shown above, the denitrification reaction can be observed in ice even at very low concentration, conditions under which the reaction is not observed in liquid solution. When aqueous solution is frozen, pure ice crystals form and the solutes are rejected from ice into unfrozen solution.36 The rejected solutes are concentrated into micropocket in ice. In ice chemistry, quasi-liquid layer (QLL) is sometimes discussed. However, thickness of QLL is affected by temperature, concentration of solutes and ice crystal plain (basal and prism plain). Therefore, it is difficult to distinguish that reactions proceed in unfrozen solution or in QLL. In this work, we use the terms including these layers as “unfrozen solution in ice” or “micro-pocket”. Concentration Dependence. Both the reaction of nitrous acid with dissolved oxygen and the denitrification reaction are slow in solution. In ice, these reactions are greatly accelerated. The concentration dependence of the reactions was investigated and the results are shown in Figure 3. In solution, nitrite oxidation by dissolved oxygen proceeded to a small extent after 24 h, and ratio of the nitrate formed to the initial nitrite concentration decreased with increasing concentration. Due to the fact that the concentration of dissolved oxygen was limited and the dissolved oxygen consumed by the reaction was not supplied from the atmosphere under the freezing condition of the present study, the ratio of the nitrate formed to the initial nitrite concentration decreased with increasing concentration. N2 forms only at high concentrations after 24 h in solution, while the ratio of N2 to the 14448

dx.doi.org/10.1021/jp2093466 |J. Phys. Chem. A 2011, 115, 14446–14451

The Journal of Physical Chemistry A

Figure 3. Dependence of reactions (24 h) on the initial concentration of ammonium nitrite. The solution pH was 4.0, and reaction temperature were 253 K in ice and 293 K in solution. Dissolved oxygen in the sample solution was equilibrated with ambient air. Part (a) shows NO3 concentrations, and part (b) shows N2 production. White and black circles show the results in solution and in ice, respectively.

initial ammonium concentration in ice increased linearly with increasing concentration. It has been reported that the dependence of the reaction rate in ice on concentration apparently does not obey the normal reaction rate kinetics,37 and this is the same result. When all ions are rejected from an ice phase into an unfrozen phase, the concentrations in the unfrozen solution are greatly elevated with respect to the initial solution. Depending on temperature, the concentrations in the unfrozen solution reach to the equilibrium concentration. The concentration factors are different if the initial concentrations are different, that is, the concentration factor is high at a low initial concentration and low at a high initial concentration. However, the reaction rates in the unfrozen solution remain the same. Furthermore, total amounts of reactants in ice are proportional to the initial concentration. As a result, the measured product concentrations increase with increasing the initial concentrations, as indicated in Figure 3(b). pH Dependence and Concentration Factor. The pH dependence of the reaction follows the reaction rate equation in solution, R = k[HNO2][NO2] [NH4+] = k0 [HNO2]2[NH3]. These two equations cannot be distinguished save for the difference in the reaction rate coefficients, where k0 = k 3 KHNO2/ KNH4+, and KHNO2 and KNH4+ are the acid dissociation constants of HNO2 and NH4+, respectively. Therefore, a maximum reaction rate is observed at a temperature determined by the acid dissociation constants38 of HNO2 and NH4+. The pH in the unfrozen solution is changed from the original solution by mainly two reasons, that is, (1) the concentration effect of ions and (2) pH change due to H+ or OH transfer from the ice to the unfrozen solution occurred by neutralizing the freezing potential generated during freezing. In this system, it is expected that the freezing potential of ice is positive because of the presence of ammonium ions.39 Consequently, the pH in the unfrozen solution decreased. However, it is expected that the amount of protons transferred from ice to solution is less than 104 mol dm3,17 and at lower than pH 5, the pH decreases mainly due to the concentration effect. Therefore, the concentration factor can be calculated from the pH at the maximum reaction rate, hereafter referred to as pHmax, at pH values less than 5. These results are shown in Figure 4. The maximum reaction rate in solution, pHmaxsolution is calculated to be 3.69 at 253 K.38 The maximum reaction rate in ice, pHmaxice was observed at pH 4.0 for the 1.0 mmol dm3 sample. The pHmaxice are summarized in Table 1.

ARTICLE

Figure 4. pH dependence of N2 production for denitrification in ice (24 h). The initial concentration of ammonium nitrite was 5.0 mmol dm3. The sample was frozen at 253 K and kept for 24 h. Dissolved oxygen in the sample solution was equilibrated with ambient air. Circles show N2 production and line shows calculated values in solution at 253 K.

Table 1. pHmax‑ice and Concentration Factors concentration factora

initial conc. 3

(mmol dm )

pHmax‑solution

pHmax‑ice

pHmax

reaction rate

0.50 1.0

3.69 3.69

4.5 4.0

6.5 2.0

50.6 41.2

5.0

3.69

3.8

1.3

13.4

a

Concentration factors were calculated from the pHmax change (pHmax‑solution and pHmax‑ice) and from the initial reaction rates.

The pHmaxice value decreased with increasing concentration. The calculated concentration factors are also shown in Table 1, and a concentration factor of 6.5 was the highest observed in our experiments. This value is too low to explain the acceleration of the reaction. Riordan et al. reported a more accurate acid dissociation constant for nitrous acid, reporting a pKa of 2.8 ( 0.1 at 298 K.40 If we use this pKa and assume that the activation energies for dissociation are the same, the concentration factors are 3.5, 5.6, and 17.8 at 5, 1, and 0.5 mmol dm3, respectively. The freezing potential could be positive, but these values are still too small to explain the rate acceleration. Therefore, it is concluded that estimation of concentration factors from pH values is difficult. We then tried to estimate the concentration factors from the reaction rate coefficient. The activation energy of reaction 1 was reported to be 65.8 kJ mol1.31 We also estimated the activation energy for reaction 1 in solution from the Arrhenius plot of the rate of decrease in ammonium ion concentration, which gave an Ea value of 63.2 kJ mol1. This value is in good agreement with the reported value. From the initial reaction rate in ice, the concentration factors can be calculated from eq 11, which takes pH into account. R1 ¼ k0 ½HNO2 2 ½NH3  ¼ k1 0

½Hþ   NðIIIÞ KHNO2 þ ½Hþ 

1 !2 0 KNHþ  AN 4 @ A KNHþ þ ½Hþ  4

ð11Þ

Here, N(III) and AN indicate total nitrite ([HNO2] + [NO2]) and total ammonium ([NH4 +] + [NH3 ]) concentrations, 14449

dx.doi.org/10.1021/jp2093466 |J. Phys. Chem. A 2011, 115, 14446–14451

The Journal of Physical Chemistry A

Figure 5. Dependence of calculated Tmax on activation energy for the second- and third-order reactions. Solid line: second-order reaction, dashed line: third-order reaction.

Figure 6. Dependence of N2 produced on the freezing temperature. The initial concentration of ammonium nitrite was 5 mmol dm3. The sample was frozen at each temperature and kept for 24 h. Dissolved oxygen in the sample solution was equilibrated with ambient air. Nitrogen concentrations were calculated from decreases in the ammonium ion concentration during 24 h. ): [NO2]0 = [NH4+]0 = 5.0 mmol dm3, b: [NO2]0 = 4.6 mmol dm3, [NH4+]0 = 5.3 mmol dm3. The error bars show one standard deviation for 5 measurements.

respectively. At 253 K, KHNO2 = 103.69,38 KNH4+ = 109.83,38 and k0 1 = 714 mol2 dm6 s1.31 As shown in Table 1, the resulting concentration factors were calculated to be 50.6, 41.2, and 13.4 for 0.50, 1.0, and 5.0 mmol dm3, respectively. These concentration factors can be used to explain the acceleration of the reaction in ice. Estimation of Activation Energy. Activation energy can be determined by Tmax. Figure 5 shows the dependence of Tmax on the activation energy for second- and third-order reactions calculated from eqs 9 and 10. Tmax decreases with increasing activation energy in both cases, and determination of activation energy can be conducted using Tmax. Using Tmax, at least two activation energies are obtained. For example, if Tmax is determined experimentally to be 264 K, the calculated activation energies are 64 and 125 kJ mol1 for second- and third-order reactions, respectively. The difference between these two values is extremely large. Therefore, it is easy to determine the activation energy from Tmax. For the chemical denitrification reaction, the reaction order is reported to be both second-23 and third-25order.

ARTICLE

Using the Tmax method, the activation energy and also reaction order are determined. The temperature dependence of the decrease in ammonium ion concentration for 24 h was investigated using a 5 mmol dm3 ammonium nitrite solution. The results are shown in Figure 6. The decreased concentrations of ammonium ion corresponding to N2 produced are indicated with black circles and white diamonds. Tmax was found to be about 253 K, which results in activation energies of 26 and 53 kJ mol1 for second-order and third-order kinetics, respectively. The reported Ea for this reaction is 65.8 kJ mol1,31 and our measured value is 63.2 kJ mol1. The theoretical fit to the temperature dependence of N2 production at an Ea of 63.2 kJ mol1, in the case of the third-order kinetics, is also shown in Figure 6 with a dashed line. The Tmax at 63.2 kJ mol1 is calculated to be 255.9 K, which differs slightly from the experimentally determined value. However, the theoretical line in the case of second-order kinetics differs greatly. Therefore, we concluded that reaction 1 obeys third-order kinetics. The calculated fit for activation energy of 56 kJ mol1 is also shown in Figure 6 with a solid line. From these discussions, we concluded that the activation energy of reaction 1 is estimated to be between 53 and 56 kJ mol1. This Tmax determination method for reactions in ice can be used as a new method to estimate activation energies and to determine reaction-orders. Competitive Reaction. The reaction of nitrous acid with dissolved oxygen obeys second-order kinetics for nitrous acid and first-order kinetics for dissolved oxygen in solution.28 However, it is considered that excess of oxygen in the unfrozen solution in ice is confined in the boundary layer as bubbles and is redissolved into the unfrozen solution because of the decrease in the concentration of dissolved oxygen during the reaction. As a result, the concentration of dissolved oxygen in the unfrozen solution remains constant, and theoretical data support this assumption.21 Therefore, the reaction of nitrous acid with dissolved oxygen in ice could obey pseudosecond-order reaction kinetics.10 The denitrification reaction obeys third-order kinetics in ice, as described above. If the volume of the unfrozen solution is reduced by a factor of 100 (the concentration increases by a factor of 100), then the reaction rates for nitrite oxidation and denitrification become 104 and 106 times faster, respectively. From Figure 2, we observe that the chemical denitrification reaction is slower than the nitrite oxidation reaction, but it could become faster at higher concentration. We calculated the concentration at which the chemical denitrification reaction becomes faster than the nitrite oxidation using eq 12. R6 ¼ k6 ½HNO2 2 ½O2 

ð12Þ

At 253 K, k6 is calculated to be 2.66 mol2 dm6 s1.37 If we assume that the concentration of dissolved oxygen is 0.436 mmol dm3, the reaction rates become equal at 3.3 mol dm3. Above 3.3 mol dm3, the denitrification reaction becomes faster than nitrite oxidation. However, as shown in Figure 2, nitrite oxidation is much faster than chemical denitrification. This is due to the fact that the nitrite oxidation is extremely fast during freezing and that it is complete when entire sample is frozen. In ice, the concentration factor was estimated to be 50.6, and in this case the concentration did not reach 3.3 mol dm3. Therefore, the chemical denitrification reaction proceeded slowly after the nitrite oxidation reaction was complete. From this discussion, the concentration factor of 50.6 is a reasonable value to explain our results. 14450

dx.doi.org/10.1021/jp2093466 |J. Phys. Chem. A 2011, 115, 14446–14451

The Journal of Physical Chemistry A The obtained results in this work can be useful for atmospheric heterogeneous chemistry in/on ice, cloud particles in polar stratosphere and upper troposphere, wastewater purification, medicine, cryobiology, etc.

’ AUTHOR INFORMATION Corresponding Author

*Phone: (+81)72-254-9322; Fax: (+81)72-254-9910; E-mail: [email protected].

’ ACKNOWLEDGMENT This work was partially supported by the 2004 Showa Shell Sekiyu Foundation for Promotion of Environmental Research Grant No. B033. ’ REFERENCES (1) Grant, N. H.; Clark, D. E.; Alburn, H. E. J. Am. Chem. Soc. 1961, 83, 4476–4477. (2) Brown, W. D.; Dolev, A. J. Food Sci. 1963, 28, 211–213. (3) Weatherburn, M. W.; Logan, J. E. Clin. Chim. Acta 1964, 9, 581–584. (4) Pincock, R. E. Acc. Chem. Res. 1969, 2, 97–103. (5) Lund, D. B.; Fennema, O.; Powrie, W. D. J. Food Sci. 1969, 34, 378–382. (6) Fan, T.-Y.; Tannenbaum, S. R. J. Agric. Food Chem. 1973, 21, 967–969. (7) Hatley, R. H. M.; Franks, F.; Day, H.; Byth, B. Biophys. Chem. 1986, 24, 41–46. (8) Hatley, R. H. M.; Franks, F.; Day, H. Biophys. Chem. 1986, 24, 187–192. (9) Takenaka, N.; Ueda, A.; Maeda, Y. Nature 1992, 358, 736–738. (10) Takenaka, N.; Ueda, A.; Daimon, T.; Bandow, H. J. Phys. Chem. 1996, 100, 13874–13884. (11) Emoto, M.; Honda, K.; Yamazaki, N.; Mori, Y.; Fujieda, S. Abstract(I) of 79th Annual meeting of Japan Chemical Society, 2001, p 622, 3PA104. (in Japanese) (12) Betterton, E.; Anderson, D. J. J. Atmos. Chem. 2001, 40, 171–189. (13) Takenaka, N.; Furuya, S.; Sato, K.; Bandow, H.; Maeda, Y.; Furukawa, Y. Int. J. Chem. Kinet. 2003, 35, 198–205. (14) Sato, K.; Furuya, S.; Takenaka, N.; Bandow, H.; Maeda, Y.; Furukawa, Y. Bull. Chem. Soc. Jpn. 2003, 76, 1139–1144. (15) Arakaki, T.; Shibata, M.; Miyake, T.; Hirakawa, T.; Sakugawa, H. Geochem. J. 2004, 38, 383–388. (16) Champion, D.; Simatos, D.; Kalogianni, E. P.; Cayot, P.; Meste, M. L. J. Agro. Chem 2004, 52, 3399–3404. (17) Takenaka, N.; Tanaka, M.; Okitsu, K.; Bandow, H. J. Phys. Chem. A 2006, 110, 10628–10632. (18) O’Driscoll, P.; Minogue, N.; Takenaka, N.; Sodeau, J. J. Phys. Chem. A 2008, 112, 1677–1682. (19) O’Sullivan, D.; Sodeau, J. R. J. Phys. Chem. A 2010, 114, 12208–12215. (20) Pincock, R. E. Acc. Chem. Res. 1969, 2, 97–103. (21) Takenaka, N.; Bandow, H. J. Phys. Chem. A 2007, 111, 8780–8786. (22) Mellor, J. W. A Comprehensive Treatise on Inorganic and Theoretical Chemistry; Wen Wai Press: London, 1931; Vol. VIII (N, P), pp 470529. (23) Dusenburry, J. H.; Powell, R. E. J. Am. Chem. Soc. 1951, 73, 3266–3268. (24) Ewing, G. J.; Bauer, N. J. Phys. Chem. 1958, 62, 1449–1452. (25) Meyet, J.; Trutzer, E. Z. Electrochem. 1911, 14, 69–76. (26) Abel, E. H. Z. Phys. Chem. 1931, 510–522. (27) Rubin, M. B.; Noyes, R. M.; Smith, K. W. J. Phys. Chem. 1987, 91, 1618–1622.

ARTICLE

(28) Damshen, D. E.; Martin, L. B. Atmos. Environ. 1983, 17, 2005–2011. (29) Mudgal, P. K.; Bansal, S. P.; Gupta, K. S. Atmos. Environ. 2007, 41, 4097–4105. (30) Christianson, C. B.; Cho, C. M. Soil Sci. Soc. Am. J. 1983, 47, 38–42. (31) Nguyen, D. A.; Iwaniw, M. A.; Fogler, H. S. Chem. Eng. Sci. 2003, 58, 4351–4362. (32) Nikolic, A.; Hultman, B. Water Sci. Technol. 2005, 52, 509–516. (33) Nikokic, A.; Hultman, B. Vatten 2003, 59, 39–45. (34) Caloz, Franc-ois, Doctor thesis, Ecole Polytechnique Federale: Lausanne, 1997, Thesis No. 1628 (http://library.epfl.ch/en/theses/?nr)1628). (35) Lide, D. R., Frederikse, H. P. R.CRC Handbook of Chemistry and Physics, 76th ed.; CRC Press, Inc.: Boca Raton, FL, 1995. (36) Bogdan, A. J. Phys. Chem. A 2010, 114, 10135–10139. (37) Takenaka, N.; Ueda, A.; Maeda, Y. Proc. NIPR Symp. Polar Meteorol. Glaciol. 1993, 7, 24–32. (38) Nair, S. K.; Peter, L. K. Atmos. Environ. 1989, 23, 1399–1423. (39) Cobb, A. W.; Gross, G. W. J. Electrochem. Soc. 1969, 116, 796–804. (40) Riordan, E.; Minogue, N.; Healy, D.; O’Driscall, P.; Sodeau, J. R. J. Phys. Chem. A 2005, 109, 779–786.

14451

dx.doi.org/10.1021/jp2093466 |J. Phys. Chem. A 2011, 115, 14446–14451