Direct and Indirect Radiolytic Effects in Highly Concentrated Aqueous


Direct and Indirect Radiolytic Effects in Highly Concentrated Aqueous...

1 downloads 297 Views 3MB Size

ARTICLE pubs.acs.org/JPCA

Direct and Indirect Radiolytic Effects in Highly Concentrated Aqueous Solutions of Bromide Anna Balcerzyk,† Jay LaVerne,‡ and Mehran Mostafavi*,† †

Laboratoire de Chimie Physique, UMR 8000 CNRS/Universite Paris-Sud 11, Faculte des Sciences d’Orsay, B^at. 349, 91405 Orsay Cedex, France ‡ Radiation Laboratory and Department of Physics, University of Notre Dame, Notre Dame, Indiana 46556, United States ABSTRACT: Highly concentrated aqueous solutions of bromide were used to examine the total radical yield in the direct decomposition of water by γ-rays. Bromide concentrations were varied up to 6 M at which almost all OH• radicals, H• atoms, and hydrated electrons produced in the picosecond range oxidize bromide to ultimately form Br3, a stable species that can easily be measured with a spectrometer. Considering only the decomposition of water in the presence of air and in acidic conditions, the apparent yield of oxidizing species is found to be around (10 ( 0.05)  107 mol J1. The absorption of irradiation dose by the solute at high concentration is discussed and quantitatively evaluated. At 6 M Br solutions, 38% of the dose is absorbed by solutes and Br is directly ionized. The optimal value for the initial yield of the radicals produced by direct radiolytic Br ionization is found to be (9.6 ( 0.5)  107 mol J1.

’ INTRODUCTION The decomposition of pure water under irradiation with γ-rays is well-known, and the radiolytic yields of different species are established at the end of the homogeneous stage, around 100 ns after energy deposition in the solution.1 Great progress has also been made during the past decade to determine the yield of the hydrated electron back to the picosecond range.2 Nevertheless, as the OH• radical absorption intensity is very low, the value of its radiolytic yield at the picosecond range is still under discussion. Scavenging methods are usually used to determine the radiolytic yield of OH• radical, but to obtain data at short times the concentration of the scavenger must be very high.3,4 Although the radiolysis of pure water or low concentration aqueous solutions is well-known, direct ionization of the solute molecule or ion at high concentrations (direct effect) is not quantitatively well established.5 The direct effect of ionizing radiation is also important for practical demands in nuclear energy technology and even for radiotherapy. Only a few studies have been performed to quantitatively estimate the direct effect of ionizing radiation on the solutes, but their conclusions were not always consistent. One of the first instances in the examination of the direct effect was with highly concentrated nitric acid, which is used in reprocessing of nuclear spent fuels. The direct effect was observed to lead to the formation of NO3• radical using pulse radiolysis in the nanosecond range.6 There are several conflicting results on the direct effect with halide solutions, but the conclusion is that the yield of the direct effect in aqueous solutions of Cl and I is around 7  107 and 7.3  107 mol J1, respectively.7 One of the main problems with these reported studies is that the radical yields from water in the picosecond r 2011 American Chemical Society

range were not as precisely determined as they are today; there was also some confusion between the scavenging power and the direct effect. Here, highly concentrated bromide solutions are proposed as appropriate candidates for measuring the direct radiolytic effect. The radiolytic oxidation of bromide ions, Br, in aqueous solution is well studied and involves several reactions. Rafi and Sutton8 first demonstrated that the γ-radiolysis of aerated aqueous solutions of potassium bromide, KBr, in the concentration range 1041 M generates bromine, Br2, in equilibrium with tribromide ion, Br3. Pulse radiolysis studies by Rabani and co-workers911 on deaerated KBr aqueous solutions established much of the mechanism and measured many of the rate constants. Later, D’Angelantonio et al.12 reexamined the pulse radiolysis of Br aqueous solutions and determined more of the rate constants for this system. They also observed the formation of Br3 at 265 nm and reported that the intensity of the signal depends on the pH and the concentration of Br in the solution. By careful chemical preparation, Wang et al.7 studied the equilibrium between Br3 and Br2 and evaluated the value of this equilibrium constant to be 16.1 ( 0.3 M1.13 In addition, they determined a reliable molar extinction coefficient for Br3 (ε266 nm = 40 900 ( 400 M1 cm1). Recent work in this laboratory proposed that aqueous solution containing 2 M Br can be used as a possible dosimeter system for steady state radiolysis. This system has several advantages as a Received: February 8, 2011 Revised: March 16, 2011 Published: April 01, 2011 4326

dx.doi.org/10.1021/jp2012528 | J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A

ARTICLE

Table 1. Chemical Composition of Solutions and Yields of Productsa [Br]

a

gas

gas concentration

solution density

yield of Br3

(mol L1)

pH

saturation

(103 M)

(g cm3)

107 mol J1)

[Br3]/[Br2]

yield of oxidized species calculated Br3 yield (107 mol J1)

(107 mol J1)

a

2

6

N2O

15

1.14

4.00

32.2

8.24

4.13

b

4

6

N2O

9

1.28

4.55

64.4

9.23

4.13

c

6

6

N2O

5

1.41

4.77

96.6

9.63

4.07

d

6

1

N2O

5

1.41

3.93

96.6

7.94



e

2

1

O2

0.13

1.14

2.24

32.2

5.02

3.19

f

4

1

O2

0.07

1.28

3.51

64.4

7.12

3.79

g h

6 6

1 6

O2 O2

0.04 0.04

1.41 1.41

5.00 1.61

96.6 96.6

10.04 3.26

4.02 

i

6

6

N2

0.074

1.41

1.48

96.6

3



j

6

1

N2

0.074

1.41

3.44

96.6

6.92



[Br3]/[Br2] is found according to the equilibrium value (equation 22). The total oxidized species are calculated according to [Br ] þ 2[Br3].

Figure 1. Absorption spectra of N2O-saturated solutions containing 6 M NaBr at pH = 6 irradiated by γ-rays as a function of absorbed dose (Gy). Dose rate = 218 Gy/h; optical path = 1 cm.

Figure 2. Absorbance at 266 nm as a function of the absorbed dose for the difference systems listed in Table 1 and absorbance at 304 nm for the Fricke solution. Dose rate = 218 Gy/h; optical path = 1 cm.

chemical dosimeter.14 First, the preparation of the solution is easy, and the solution is very stable. Second, the radiation generated product, Br3, is very stable and constitutes a single absorbing species exhibiting an intense absorption band at 266 nm with a precisely defined extinction coefficient. Third, the radiolytic yield of Br3 is high and valid for a wide range of dose, between 10 and 200 Gy. All of these advantages can be exploited to promote this system as a probe of ionizing radiation. In the present work, the radiolytic yield of Br3 has been determined in the γ-radiolysis of highly concentrated solutions of NaBr up to 6 M. Both aerated and N2O-saturated samples were examined. By combination of the experimental results with model calculations, the production of Br3 is used to probe the direct radiation effect and to measure quantitatively the radiolytic yield of Br oxidation.

was purified by passage through a Millipore purification system. The solutions were made to pH 6 and saturated with nitrous oxide, purity >99.98%, for 30 min or at pH 1 and O2-saturated. The pH of the solutions was adjusted with HClO4. Salt solutions were investigated at 2, 4, and 6 M NaBr. All experiments at Universite Paris-Sud were carried out in quartz cells. Solutions were prepared immediately before irradiation. The irradiations were performed in a panoramic γ-source under ambient conditions. Solutions were irradiated to different doses with a maximum of 300 Gy. The dose rate was measured with the Fricke dosimeter. The absorbed radiation dose of highly concentrated solutions was calculated by taking into account the density of solution and the electronic concentrations of the solutes and water. Every measurement was repeated at least four times, and the average of the optical density was taken to calculate the yield of formation of Br3. The absorption spectra of Br3 were recorded with a single beam diode array UVvisible spectrophotometer (Hewlett-Packard 8543 operated at a resolution of 2 nm). Absorbance was measured in a 1 cm optical path quartz cell before and after radiation exposure.

’ EXPERIMENTAL SECTION The chemical reagents were purchased from Sigma-Aldrich. The purity of NaBr was greater than 99.999%, which is needed for the γ-radiolysis of the high concentration solutions. Water

4327

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A

ARTICLE

Table 2. Reaction Scheme Used in Model Calculations reaction

rate or equilibrium constant 5.5  109 dm3 mol1 s1

 1 e aq þ eaq f H2 þ 2OH 3

2

e aq

þ

1.3  1010 dm3 mol1 s1

þ H2O f H 3 þ H2O 

2.5  1010 dm3 mol1 s1

3 eaq  þ H 3 f H2 þ OH 4

e aq



3.0  1010 dm3 mol1 s1

þ OH 3 f OH

7.8  109 dm3 mol1 s1

5 H 3 þ H 3 f H2

7.0  109 dm3 mol1 s1

6 H 3 þ OH 3 f H2O 

5.5  109 dm3 mol1 s1

7 OH 3 þ OH f H2O2 8 9

e aq e aq



þ N2O f N2 þ OH 3 þ OH þ O2 f O 2

2.1  1010 dm3 mol1 s1

10 H 3 þ O2 f HO2

2.1  106 dm3 mol1 s1

11 H 3 þ N2O f N2 þ OH 3 



12 H 3 þ Br f HBr 



13 HBr þ O2 f Br þ

Figure 3. Absorption spectra of air-saturated solutions irradiated by γ-rays to various doses.

Irradiations at the University of Notre Dame were carried out using a Shepherd 109-68 60Co γ-source. The dose rate was 67.4 Gy/min as estimated by the Fricke dosimeter, and the dose absorbed by the samples was determined as described above. Samples were made to the correct pH and NaBr concentration and then purged with N2O or O2 and flame-sealed in Pyrex cells 1 cm in diameter by 10 cm long. Following irradiation, the samples were crushed in the inlet line of a SRI 8610 apparatus equipped with a thermal conductivity detector and ultrapure argon as the carrier gas. The chromatographic column was a 6.4 mm diameter 13 molecular sieve 3 m long, maintained at 40 °C. Calibration was performed by injection of pure H2 at normal conditions.

’ RESULTS Spectroscopic Measurements. Ten different sample configurations of aqueous Br solutions as listed in Table 1 were irradiated by γ-rays to determine the radiolytic formation yield of Br3. The samples a, b, c, and d were N2O-saturated, e, f, g, and h were air-saturated, and i and j were N2-saturated. Gas concentrations are known to decrease with increasing salt concentration. The values of gas solubility are calculated according to the model proposed by Schumpe and are in good agreement with results reported in the literature.15 After each irradiation dose, the absorption spectrum of the sample is recorded to show the formation of an absorption band with a maximum located at 266 nm, which corresponds to the Br3 anion. Typical spectra for N2O-saturated solutions of 6 M NaBr are shown in Figure 1 for a variety of doses. The spectra are seen to be of a similar shape for all doses. At pH 6, for a given dose, the intensity of the absorption spectra at 266 nm increases with increasing Br concentration. This effect is not strong; for example, the absorbance at the maximum of the absorption band for a dose of 25 Gy is 0.41, 0.46, and 0.48 for 2, 4, and 6 M Br, respectively. Figure 2 shows the variation in the absorbance at 266 nm as a function of the absorbed dose for the different aqueous solutions listed in Table 1. Also shown in this figure are the results for the Fricke solution (measured at 304 nm used as a dosimeter). The radiation chemical yields of Br3 are proportional to the slopes of the lines in Figure 2 and are

9.1  109 dm3 mol1 s1 1.9  1010 dm3 mol1 s1

1.7  106 dm3 mol1 s1 1.0  1010 dm3 mol1 s1

HO 2 

14 OH 3 þ Br f BrOH 3 15 BrOH 3  f Br þ OH 3

1.1  1010 dm3 mol1 s1 3.3  107 dm3 mol1 s1

16 BrOH 3  f Br 3 þ OH 17 Br 3 þ OH f BrOH 3  18 BrOH 3  þ H3Oþ f Br þ 2H2O

4.2  106 s1 1.3  1010 dm3 mol1 s1

19 Br þ Br f Br 2

1.2  1010 dm3 mol1 s1

 20 Br 2 f Br þ Br

1.9  104 s1

   21 Br 2 þ Br2 f Br3 þ Br

2.4  109 dm3 mol1 s1

22 Br2þ BrTBr 3

K = 16.1 dm3 mol1

  23 e aq þ Br2 f 2Br

1.3  1010 dm3 mol1 s1

24 H 3 þ f H þ 2Br  þ  25 HO þ Br 2 2 f 2Br þ O2 þ H

1.4  1010 dm3 mol1 s1 1.0  108 dm3 mol1 s1

   26 e aq þ Br3 f Br þ Br2

2.7  1010 dm3 mol1 s1

Br 2

þ

4.4  1010 dm3 mol1 s1

27 H 3 þ f Br þ þH 1.2  1010 dm3 mol1 s1     þ 28 HO2 þ Br3 f Br2 þ Br þ O2 þ H 1.2  107 dm3 mol1 s1 Br 3





Br 2

þ

found to be 4.00  107, 4.55 107, and 4.77  107 mol J1 for the samples a, b, and c, respectively. At pH 1, under N2O-saturated 6 M Br solutions, the yield is slightly lower (3.93  107 mol J1). The results are very reproducible, and the error is less than 5% for each sample. Results for all the sample systems are given in Table 1. An important consideration is that Br3 is in equilibrium with Br2 with an equilibrium constant of K = 16.1 ( 0.3 M1. The absorption of Br2 is negligible at 266 nm (ε266(Br2) = 47 M1 cm18). Therefore, this equilibrium must be taken into account in order to determine the total oxidation yield of Br. For solutions containing 6 M, almost 99% of oxidized atoms are in the form of Br3 and 1% as Br2. Combining the equilibrium constant with the data for the yields of Br3 leads to the total oxidation yield of Br. The results are listed in Table 1. O2-saturated solutions (eg) at pH 1 give spectra similar to that for N2O-saturated solutions. The shape of the absorption spectra recorded after the irradiation does not depend on the irradiation dose, However, Figure 3 shows that the intensity of the absorption band depends much more on Br concentration than for N2Osaturated solutions. The absorbance at 266 nm is reported a function of absorbed dose for different O2-saturated solutions in Figure 2. The yield of Br3 is almost doubled on increasing the concentration of Br from 2 to 6 M. But at pH 6 (sample g) even with 6 M Br, the yield is low (1.61  107 mol J1). H2 Measurement. H• atom chemistry is very important in these systems because the high acid content rapidly converts hydrated electrons to H•. atoms. In order to probe the mechanism involving 4328

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A

Figure 4. Model simulation results for O2-saturated solutions of 2, 4, and 6 M Br. Only water radiolysis is considered using the reaction scheme listed in Table 2.

the H• atom chemistry, the yield of H2 was determined for each of the systems given in Table 1. The resulting yields are 0.33  107, 0.37  107, and 0.49  107 mol J1 for O2saturated solutions (pH = 1) and 0.44  107, 0.50  107, and

ARTICLE

Figure 5. Model simulation results for N2O-saturated solutions of 2, 4, and 6 M Br. Only water radiolysis is considered using the reaction scheme listed in Table 2.

0.59  107 mol J1 for N2O-saturated solutions (pH = 6). These yields are very near the value of 0.47  107 mol J1 expected for water alone. 4329

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A Model Calculations. The underlying chemistry of the systems

examined here can be elucidated by comparing measured results with the predictions of model calculations. Chemistry within an average spur produced in γ-radiolysis was simulated using a deterministic model that was parametrized to give similar results as a fully simulated Monte Carlo track model.16 This model assumes an initial picosecond yield of water decomposition products with Gaussian spatial distributions. The subsequent kinetics is followed using a nonhomogeneous deterministic model in which the coupled differential equations for the various reactions were stepped in time using FACSIMILE.17 Table 2 presents the main reactions in this system with the rate constants reported as in the literature, unless otherwise noted. The spur simulations of the six different systems were performed with initial or picosecond radiolytic yields of OH• radicals, hydrated electrons, and H• atoms of 5.3  107, 4.4  107, and 0.62  107 mol J1, respectively. Figures 4 and 5 show the kinetics in the spur for several species with the conditions for the systems studied here. The time evolutions of the hydrated electron, H•, OH•, H2, Br•, Br2•, and Br3 are shown for each irradiated system. In these simulations, the yields are only representative of the radiolysis of water, and the direct effect due to the decomposition of Br is not taken into account. Only three stable species, Br3, H2, and H2O2, are produced on the long time scale. The last product was not examined in this study but will be in a future work.

’ DISCUSSION Mechanism of Oxidation of Br in Acidic O2-Saturated Solutions. Three reactive species are present in the solution at

the beginning of the spur lifetime, OH•, H•, and hydrated electrons. Hydrated electrons are converted to H• atoms within 100 ps (Figure 4). On about the same time scale, OH• radicals are scavenged very quickly by Br to form BrOH•-. This species reacts with the high concentration of acid to give Br•. The resulting Br• is very transient and is quickly converted to Br2• by reaction with Br. After a few nanoseconds the H• begins to react with Br to give HBr•. There is a competition between O2 and Br for H•. The reaction of H• with O2 is fast and that with Br is relatively slow, k = 1.7  106 dm3 mol1 s1, but the Br concentration is very high. The yield of Br2 decreases slightly on the nanosecond time scale due to the disproportionation reaction of Br2 to give Br3 and Br. At about 10 ns the H• is rapidly depleted by reaction with O2. On the microsecond time scale, the HBr• is oxidized by O2 to give Br•, which is rapidly converted by reaction with Br to Br2•. In N2-saturated solutions at pH = 1 and pH = 6 (solutions i and j, Table 1), the yield of Br oxidation is 3.44  107 and 1.48  107 mol J1. In fact, at neutral pH, the hydrated electron acts fully as a reducing species to decrease the yield of oxidation. For the solution, with 6 M Br at pH 1, the hydrated electron is quickly converted to H• atom, and H• is acting only partly as a reducing species showing that in the absence of O2 it is not possible to convert all HBr• to Br• (Figure 4). By increase of the concentration of Br from 2 to 6 M, the solubility of O2 is decreased and the reaction of H• with Br is favored. The significant change in the yield of oxidizing species from 5  107 to 10  107 mol J1, when Br concentration is increased from 2 to 6 M (samples e and g), is mostly due to this competition. Therefore, for the solutions containing 6 M Br, even in the presence of oxygen, almost all radicals (hydrated electrons, OH•, and even H•) are scavenged and transformed

ARTICLE

into Br3. Only a small amount of H• is converted to HO2• which reduces the radiolytic yield of Br3. Therefore, for air-saturated solutions the yield of oxidizing species resulting from decomposition of water is obtained by addition of the concentration of Br• at long time (millisecond range):  GðOXÞ ¼ 2GðBr 3 Þ þ GðBr2 Þ

ð1Þ

According to the simulations, for samples containing 6 M Br, the radiolytic yield of oxidizing species issued from water radiolysis is around 8.0  107 mol J1. But the experimental value, if we considered that the dose is absorbed only by water molecules (no direct effect), is 10  107 mol J1. For highly concentrated aqueous solutions, it is important to note that the dose given by Fricke dosimetry in water should be multiplied by the factor F: F ¼ dsol ðZNaBr p=ANaBr þ Zwater ð100  pÞ=Awater Þ ðZwater 100=Awater Þ1

ð2Þ

dsol is the density of the solution, Z is the number of electrons, A is the mass number, and p is the weight fraction of the solute. The dose absorbed by solution is dsol ¼ F  DFricke

ð3Þ

The value of F is 1.09, 1.2, and 1.29 for the solution containing 2, 4, and 6 M Br, respectively. For example, correcting the yields by taking into account the part of the dose absorbed by solutes reduces the yield of oxidation for solutions of 6 M Br to Gexp/ corr = 7.8  107 mol J1. The difference between the two O2 values for oxidation yield is mainly from the direct effect. Therefore, the direct ionization of Br (directly or through an excited state) can be considered as follows: Br f Br þ e

ð4Þ



The direct decomposition of Br will be more thoroughly discussed below. Mechanism of Oxidation of Br in N2O-Saturated Solutions. The kinetics of the decomposition of N2O-saturated solutions of Br are shown in Figure 5. The fast kinetics is essentially the same as in O2-saturated solutions except the hydrated electron is converted to OH• radicals instead of H• atoms. This conversion takes up to several tens of nanoseconds instead of the subnanosecond time scale in very acidic solutions. Also, the BrOH• is converted to Br• at much longer times because of the slow reaction of BrOH• with Hþ at pH 6. The result is that Br• never builds up to any noticeable concentration, but Br2 is formed at about a nanosecond. The second maximum in Br2 at about 10 ms follows the conversion of the hydrated electron to OH• radicals. The long time regime shows the disproportionation reaction of the remaining Br2• to Br3. The radiolytic yield of oxidizing species depends much less on the concentration of Br in N2O-saturated solutions than in O2saturated solutions. But it is important to note that the yield is decreased for the sample d at pH 1. In fact, in that case H• atoms are not fully scavenged by N2O and a low amount of H• atoms are acting as reducing species and contribute to decrease the yield. The noncorrected yield increases from 8.24  107 to 9.63  107 mol J1 for 2 and 6 M, respectively. Model calculations also show relatively little change in the Br3 yield with Br concentration. After correction by the F factor, the experimental values are reduced to 7.5  107, 7.7  107, and 7.5  107 mol J1, for the samples a, b, and c, respectively. In N2O-saturated solutions, the 4330

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A

ARTICLE

Table 3. Oxidation Radiolytic Yield of Different Solutions before and after the Correction of the Absorbed Dosea [Br]

fs

corr fw Gexp/air Gexp/N2O F Gcorr Gw/N2O Gw/air exp/air Gexp/N2O

2

0.15 0.85

5.02

8.24

1.1

4.19

7.5

7.4

6.1

4 6

0.28 0.72 7.12 0.38 0.62 10.04

9.23 9.63

1.2 1.3

5.94 7.79

7.7 7.5

7.45 7.7

7.2 7.8

The values of G are given in the 107 mol J1 unit. Gexp/air and Gexp/N2O are obtained from Figure 2. F is calculated according to eq 2. Gcorr exp/air and Gexp/N2corr O are calculated by dividing the experimental value by the F factor. Gw/N2O and Gw/air are deduced from the simulations reported in Figures 4 and 5. a

Figure 6. Radiolytic yield of oxidizing species as a function of the scavenging capacity for hydroxyl radicals: (blue 9) this work uncorrected for direct effects; (orange b) this work corrected for direct effects; data from ref 9 (green 9) uncorrected dose and (black b) with corrected dose.

hydrated electron and part of the H• atoms are scavenged by N2O to give OH• radicals, which ultimately oxidize the Br. In O2saturated solutions, the H• atoms oxidize Br in reactions competing with scavenging by O2. Therefore, the resulting oxidation of Br is very dependent on the Br concentration. For N2O-saturated solution at pH = 1 (Table 1, solution d), the yield of oxidation is only 7.9  107 mol J1 showing that the H• atoms are also reducing species. In fact, the H• atoms are scavenged very slowly by N2O, and an important fraction of H• atoms can reduce Br2• and Br3 but also Br2. H2 Production. H2 was measured for each of the experimental conditions in order to clarify the fate of the H• atom. In airsaturated solutions the H• atom adds to the Br to give HBr•. One possibility for this latter species to continue the oxidation process is to react with Hþ to give H2 þ Br•. However, this reaction would produce a considerable amount of H2. Model calculations predict an H2 yield of 3.88  107 mol J1 in 6 M Br solutions at pH 1. This yield is more than an order of magnitude greater than the observed value of 0.49  107 mol J1, which is close to that expected for water alone. Clearly, such a process cannot occur. Higher order processes can be envisioned for the oxidation of HBr•, but the simplest scheme is to assume that O2 oxidizes HBr• directly to give HO2 and Br•. The model calculations using these schemes agree well with the observed yields of H2. The neutralization of HO2 may lead to a noticeable change in the yield of H2O2, and further measurements will be performed to confirm the mechanism. Direct Ionization of Br. Several studies have been performed to measure the yield of oxidizing species in N2O-saturated aqueous solutions. The results are usually analyzed by examining the product yield as a function of the scavenging capacity of the medium for OH• radicals. Scavenging capacity for a particular solute is defined as the product of the solute concentration and the rate constant for the scavenging reaction. The yield of products due to the scavenging of OH• radicals in N2O-saturated solutions is shown in Figure 6 as a function of scavenging power. The data from the present work is shown with the data of Schuler

Figure 7. Derivation of the optimal value of Gs according to eq 6.

et al.3 Yields from the N2O-saturated solutions in this work are reported by calculating the yield of oxidation with the dose as given by Fricke dosimeter and by correcting the value of the dose in consideration of the absorption of energy by the solute. A general equation for the scavenging of OH• radicals in N2Osaturated solutions is given by the following equation3 GðPÞ ¼ 5:2 þ 3:0

ðkp ðPÞ=λÞ1=2 1 þ ðkp ðPÞ=λÞ1=2

ð5Þ

with λ = 4.7  108 s1. At the high scavenging capacities used in this work, the uncorrected yields are far greater than expected from eq 5. A scavenging power of 1011 s1 obtained for 6 M Br solutions has an observed oxidation yield of 10.05  107 mol J1. By considering that an important amount of the dose is absorbed by the 1 solute, this yield is decreased to Gexp/N2corr O mol J . The latter value is much more in line with the predictions of eq 5 assuming that the ionization of solvent and Br have both the same chemical effect. By assuming an additivity between direct and indirect effects of ionizing radiation, the radiolytic yield for a given transformation can be expressed by the following relation G ¼ f s Gs þ ð1  f s ÞGw

ð6Þ

where fs is the ratio of the energy directly absorbed by the solute S to the total energy absorbed by the solution. Energy loss by γ-rays due to the Compton effect is the major source of ionization in the medium, and to a first approximation fs can 4331

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A

ARTICLE

Figure 8. Scheme of Br oxidation mechanism in highly concentrated N2O-saturated aqueous solutions. Br is oxidized directly by ionizing radiation and indirectly through the reactions with radicals from water decomposition. The combination reactions between OH• radicals (or Br2) with H• atoms and hydrated electrons are not fully avoided. These reactions and the dimerization reactions of radicals are not reported in the figure.

be evaluated by considering the electron fraction of the solute. Gs is the radiolytic yield due to the direct effect. Gw is the yield of the indirect effect and expresses the yield of the radiolytic species formed in water that participate in the oxidization of Br. Values of fs, fw, and Gw obtained for the three samples a, b, and c are given in Table 3. The value of Gs can be determined using the data for the similar systems irradiated under N2O (solutions a, b, c) by minimizing χ2 (mean square deviation) with following equation for the various values of Gs from 2.5 to 10  107 mol J1: X2 ¼

1 3 ½Gcorr  Rðf s G0s þ ð1  f s ÞG0w Þ2 3 i ¼ 1 exp



ð7Þ

Gcorr exp stands for the corrected value of Gexp by taking into account the absorption of solute using the F factor. G0w is obtained by simulations of water radiolysis from Figure 5 and stands for the initial yield of radicals (hydrated electron, H•, and OH• radicals) at around 1 ps (9.65 107 mol J1) . As the radicals produced by direct ionization and water ionization are almost similar, by estimating the amount of the radicals contributing to the oxidation of Br, we can find the initial ionization yield (at 1 ps) of Br. In fact, according to the simulations (Figure 5) all radicals produced at 1 ps do not contribute to the oxidation of Br. For example, for 6 M Br solution, only 80% of the initial amount of radical issued from water radiolysis contributes to the oxidation of Br. The value of R depends on the Br concentration, and it is 0.76, 0.78, and 0.80, for 2, 4, and 6 M Br solutions, respectively. The results of the compilation according to the eq 7 are shown in Figure 7. The optimal value for the initial yield of the radicals produced by direct radiolytic Br ionization is found to be (9.6 ( 0.5)  107 mol J1. Each direct ionization of Br leads to the production of Br• and an electron, which is quickly solvated, and then the radiolytic yield of Br ionization is (4.8 ( 0.5)  107 mol J1. Such a direct effect was already reported by Schuler et al.3 Ferradini et al.7,18 performed also a systematic investigation of highly concentrated Cl and I

aqueous solutions saturated in N2O and N2 to probe the direct effect. In the case of Cl solutions, they could not observe a direct effect. However, for iodide solutions they found Gs and Gw oxidation yield of 7.6  107 and 4.9  107 mol J1, respectively. Interestingly, the initial yield for the oxidization of Br is very close to that of water. Three different observations could be considered: (1) The ionization potential of Br in water is different than that of H2O; (2) The Cerenkov light induced by irradiation in solution can also partly ionize Br without ionizing water molecules; (3) The geminate recombination between Br• and the precursor of the solvated electron is slow. In our calculation the same value of R is assumed for both direct and indirect effect. That means the geminate reactions are considered to have the same efficiency for decreasing the yield of oxidizing species. Therefore, the two first points could not have an important effect on the value of the initial ionization yield of Br compared to that of water. The compensation of different effects can also be the reason to obtain almost the same value for Br ionization yield as that for water. Finally, at high concentration, each Br ion has almost each water molecule in contact with one Br. Even if the water molecules are ionized by radiation, Br ions can reduce H2Oþ to form Br•. Therefore, the mechanism of Br oxidation is changed, and water radiolysis to give OH• radicals is replaced by Br oxidation with H2O•þ. The result at very high Br concentrations is that almost no OH• radical is formed, and it obviously cannot contribute significantly to oxidation processes in the system. Figure 8 shows a summary for the mechanism of oxidation of Br for 6 M N2O-saturated solution. Note that the total yield of radicals at 1 ps is 20% higher than the total yield of oxidized Br found at the end of irradiation, which is equal approximately to two times the concentration of Br3. The neutralization reactions between OH• (or Br2•) with H• atoms and hydrated electron are not fully avoided. These reactions and the dimerization reactions of radicals are not reported in the figure. 4332

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333

The Journal of Physical Chemistry A

’ CONCLUSIONS The radiolytic yield of the stable Br3 species has been carefully measured in a variety of concentrations of Br aqueous solutions. The results strongly suggest that a simple inorganic system can be used in which all reactive water decomposition products—hydrated electron, OH• radical, and H• atom—are scavenged on the picosecond time regime to ultimately form only one species, Br3, which is stable with time and displays a strong absorption band. Thanks to distinguishing between direct effect and scavenging efficiency, we demonstrate the direct effect of ionizing radiation. For solutions with Br concentration greater than 2 M, the direct ionization of the solutes is shown to be definitively non-negligible. Air-saturated aqueous solutions of 6 M Br are found to have an apparent yield of oxidizing species of about (10 ( 0.1)  107 mol J1, showing that this system with a stable product having a strong extinction coefficient could be used easily as a dosimeter for low doses. We note that the direct effect depends not only on the concentration but also on the electron density of the solutes. For some solutes with high electron density, the direct effect could happen even at lower concentrations than decimolar. This work on Br is the continuation of that we recently reported,14 but I3 constitutes also an adequate probe of the direct effect of ionizing radiation because it presents a higher stability constant and a higher extinction coefficient than those of Br3. The work on this system is in progress. The precursors of Br3 are BrOH• and Br2, and both of these species have strong absorption bands. Picosecond pulse radiolysis measurements are in progress to more fully understand the effect of the direct ionization of Br in the radiolysis of highly concentrated aqueous solutions at short times, and in particular to monitor the reaction between H2Oþ and Br.

ARTICLE

(6) Matthews, R. W.; Mahlman, H. A.; Sworski, T. J. J. Phys. Chem. 1978, 76, 2680. Daniels, M. J. Phys. Chem. 1969, 73, 3710. KozlowskaMilner, E.; Broskiewis, R. K. Radiat. Phys. Chem. 1978, 11, 253. Pikaev, A. k.; Glazunov, P. Ya.; Yakobovich, A. A. Dikl. Akad. Nauk SSSR 1974, 215, 645. Lesigne, B.; Ferradini, C.; Pucheault, J. J. Phys. Chem. 1973, 17, 2156. Katsumura, Y.; Jiang, P. K. Y.; Nagaishi, R.; Oishi, T.; Ishigure, K.; Yoshida, Y. J. Phys. Chem. 1991, 95, 4435. (7) Pucheault, J.; Ferradini, C.; Julien, R.; Deysine, A.; Gilles, L.; Moreau, M. J. Phys. Chem. 1979, 1979 (83), 330–336. Hadjadj, A.; Jullen, R.; Pucheault, J.; Ferradini, C. J. Phys. Chem. 1982, 86, 4630–4634. Woods, R. J.; Lesigne, B.; Gilles, L.; Ferradini, C.; Puchealt, J. J. Phys. Chem. 1975, 79, 2700. Grigor’eva, A. E.; Makarov, I. E.; Pikaev, A. K. High Energy Chem. 1991, 25, 172. (8) Rafi, A.; Sutton, H. C. Trans. Faraday. Soc. 1965, 61 (509), 877. (9) Matheson, M. S.; Mulac, W. A.; Weeks, J. L.; Rabani, J. J. Phys. Chem. 1966, 70, 2092–2099. (10) Zehavi, D.; Rabani, J. J. Phys. Chem. 1972, 76, 312. (11) Mamou, A.; Rabani, J.; Behar, D. J. Phys. Chem. 1977, 81, 1447. (12) D’aneglantonio, M.; Venturita, M.; Mulazzani, Q. G. Radiat. Phys. Chem. 1988, 32, 319–324. (13) Wang, T. X.; Kelley, M. D.; Cooper, J. N.; Beckwith, R. C.; Margerum, D. W. Inorg. Chem. 1994, 33, 5872–5878. Beckwith, R. C.; Wang, T. X.; Margerum, D. W. Inorg. Chem. 1996, 35, 995–1000. (14) Mirdamadi-Esfahani, M.; Lampre, I.; Marignier, J.-L.; Waele, V. d.; Mostafavi, M. Radiat. Phys. Chem. 2009, 78 (2), 106–111. (15) Schumpe, A. Chem. Eng. Sci. 1993, 48, 153. (16) Pimblott, S. M.; LaVerne, J. A. J. Phys. Chem. A 1997, 101, 5828. (17) Chance, E. M.; Curtis, A. R.; Jones, I. P.; Kirby, C. R. Report AERE-R 8775; AERE: Harwell, U.K., 1977. (18) Hadjadj, A.; Jullen, R.; Pucheault, J.; Ferradini, C.; Hickel, B. J. Phys. Chem. 1982, 86, 4630.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]

’ ACKNOWLEDGMENT The ELYSE team was supported by RTRA Tringle de Physique. The research of J.L. described herein was supported by the Office of Basic Energy Sciences of the U.S. Department of Energy. This contribution is NDRL-4879 from the Notre Dame Radiation Laboratory. ’ REFERENCES (1) Buxton, G. W. . The radiation chemistry of liquid water: Principles and applications. In Charged particle and proton interaction with matter; Mozumder, A., Hatano, Y., Eds.; Marcel Dekker: New York, 2004 . (2) Bartels, D. M.; Gosztola, D.; Jonah, C.D. J. Phys. Chem. A 2001, 105 (34), 8069–8072. Muroya, Y.; Lin, M. Z.; Wu, G. Z.; Iijima, H.; Yoshi, K.; Ueda, T.; Kudo, H.; Katsumura, Y. Radiat. Phys. Chem. 2005, 72 (23), 169–172. (3) Schuler, R. H.; Hartzell, A. L.; Behar, B. J. Phys. Chem. 1981, 85, 192–199. (4) Atinault, E.; De Waele, V.; Schmidhammer, U.; Fattahi, M.; Mostafavi, M. Chem. Phys. Lett. 2008, 460 (46), 461–465. (5) Katsumura, Y. In Radiation Chemistry: Present Status and Future Trends; Jonah, C. D., Rao, B. S. M., Eds.; Elsevier: Amsterdam, The Netherlands, 2001; p 163. 4333

dx.doi.org/10.1021/jp2012528 |J. Phys. Chem. A 2011, 115, 4326–4333