Kinetics and Mechanism of Nitrite Oxidation by Hypochlorous Acid in the Aqueous Phase

(1)

Kinetics and mechanism of nitrite oxidation by hypochlorous acid in the aqueous phase

Nazafarin Lahoutifard

^*

, Philippe Lagrange, Janine Lagrange

Laboratoire de Cineetique et Analyse, ECPM, UniversiteeLouis Pasteur de Strasbourg, UMR 7512 au CNRS, 25 rue Becquerel, 67087 Strasbourg, France

Received 7November 2001; received in revised form 17September 2002; accepted 25 October 2002

Abstract

The rate coeﬃcient for the reaction of nitrite with hypochlorite and hypochlorous acid has been studied using spec- trophotometric measurements. The reaction rate has been determined in a wide range of H^þ concentration (56 log½H^þ611). The kinetics were carried out as a function of NO₂, H^þ and total hypochlorite ð½HOCl_total¼

½HOCl þ ½ClO þ ½ClNO2Þconcentrations. The observed overall rate law is described by:

d½HClO_T

dt ¼fa½NO₂²þb½NO₂g½H^þ²

cþd½H^þ þe½NO₂½H^þ² ½HOCl_total

AtT ¼298 K and in Na2SO4 at an ionic strength (I¼1:00 M), we obtained using a nonlinear ﬁtting procedure:

a¼ ð1:830:36Þ 10⁷ s¹,b¼ ð1:140:23Þ 10⁵ M s¹,c¼ ð1:120:17Þ 10¹³M,d ¼ ð1:430:29Þ 10⁶ M² ande¼ ð1:410:28Þ 10³M where the errors represent 2r. According to the overall rate law,a=b¼k1=k3,b=e¼k3, c¼Kw,d=c¼Ka,d¼KaKw ande¼K1Ka. In Na₂SO₄at an ionic strength (I¼1:00 M), the values ofK1 andKaare ð1:10:1Þ 10⁴ and 1:2810⁷M¹, respectively. A mechanism is proposed for the NO₂ oxidation which involves the reversible initial step: NO₂ þHOClClNO₂þOH(K₁), while ClNO₂ undergoes the two parallel reactions: attack by NO₂ (k1) and hydrolysis (k3). ClNO2 and N2O4 are proposed as important intermediates as they control the mechanism. The rate coefficientsk1andk3have been determined at different ionic strengths in NaCl and Na2SO4. The influence of the ionic strength and ionic environment has been studied in this work.

Keywords:Rate coeﬃcient; Ionic strength; NO₂; HOCl/ClO; Atmospheric water

1. Introduction

Kinetic studies of reactions involving nitrite have been pursued as a result of their atmospheric importance. Dew and fog nitrite concentrations are found to

be higher than in rain and snow nitrite levels in the atmosphere (Harris et al., 1982; Brauer et al., 1991; Simon and Dasgupta, 1995). Nitrite concentrations have been reported to range from small levels in rain and snow and up to several hundredlmol dm³ in dew and fog (Sigg et al., 1987; Fuzzi et al., 1988; Miller et al., 1991;

Takenaka et al., 1998, and references therein). As a result, the reactions of nitrite ions with diﬀerent oxidants such as hypohalogenite ions and hypohalogeneous acid are of atmospheric importance.

The presence of nitrate ions in tropospheric water results mainly from the dissolution of NO_x(NOþNO₂) www.elsevier.com/locate/chemosphere

*Corresponding author. Address: Department of Chemistry, University of Ottawa, 10 rue Marie Curie, Ottawa, Canada K1N 6N5. Tel.: +1-613-562-5800; fax: +1-613-562-5170.

E-mail address: [email protected] (N. Lahoutifard).

PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 76 5 - 8

(2)

formed in the gas phase, but oxidation of nitrite ions by strong oxidant in the aqueous phase must be also taken into account. Once nitrate ions are formed, they will be removed from the atmosphere by dry deposition or wet removal processes. Nitric acid is one of the most water soluble atmospheric gases with a HenryÕs law constant (at 298 K) of 2:110⁵ M atm¹ (Seinfeld, 1986).

Thereby, HNO₃ will contribute to the acid rain issue.

The major acids believed to contribute to acid deposition in the troposphere have been sulphuric and nitric acids, formed by the oxidation by SO2 and oxides of nitrogen, respectively. In addition, since nitrous acid is formed whenever NO_x and water are present, its contribution to the total acidity has become of interest and concern. The relative contribution of these acids to the total acid deposition depends on the particular emission sources. For example, the ratio of H₂SO₄ to HNO₃ is typically 2:1 on the east coast of the US, where there are signiﬁcant sources of SO2, but about 1:2 on the west coast, where NOx emissions predominate (Finlayson- Pitts and Pitts, 2000). To correctly model and predict atmospheric nitrite reactions, it is necessary to perform controlled laboratory experiments. In aqueous solutions, nitrite ion reacts with HOCl (reaction (1)). This reaction proceeds according to:

NO₂ þHOCl!NO₃ þClþH^þ ð1Þ Under atmospheric conditions, high ionic strengths may be observed particularly in aerosols during their condensation or evaporation phases, whereas low ionic strengths may be representative of cloud droplets. Aero- sols in polluted urban areas can be highly concentrated solutions with ionic strengths in the range of 8–19 M (Stelson and Seinfeld, 1981). On the other hand, cloud water and rain water in clean areas contain much lower solute concentration; for example, from the ionic composition of precipitation samples in the maritime area of Cape Grim, Australia (Ayers, 1982), the ionic strength can be calculated to be 10³ M (Finlayson-Pitts and Pitts, 2000). However, there is a lack of information about the inﬂuence of ionic strength and ionic environment on the oxidation of nitrite by HOCl/ClO. Labo- ratory studies of this reaction were essentially at constant ionic strength (Anbar and Taube, 1958; Lister and Rosenblum, 1961; Pendlebury and Smith, 1973;

Cachaza et al., 1976; Johnson and Margerum, 1991;

Behnke et al., 1997; Frenzel et al., 1998). Reaction (1) in alkaline solutions has been thought one of the ﬁrst ex- amples of oxygen atom transfer (Anbar and Taube, 1958). These authors reported that¹⁸O was completely transferred from¹⁸OCl to give labeled nitrate. Later, other laboratory studies of reaction (1) (Lister and Ro- senblum, 1961; Cachaza et al., 1976; Johnson and Margerum, 1991; Behnke et al., 1997; Frenzel et al.,

1998) or of the oxidation of NO₂ by aqueous chlorine, Cl2, (Pendlebury and Smith, 1973), did not mention any oxygen atom transfer. Margerum and coworkers (Johnson and Margerum, 1991) used¹⁵N NMR methods to distinguish¹⁸O from¹⁶O in the nitrate ion that is the ﬁnal reaction product. These results contradict An- barÕs results (Anbar and Taube, 1958) and show no evi- dence for direct O atom transfer.

In the present work, we studied the influence of ionic strength and ionic environment on this oxidation process. We studied the oxidation of nitrite by hypochlorite and hypochlorous acid at different ionic strengths using different salts over a wide range of acidity (from pH¼5 up to 11). We have previously shown the importance of the ionic environment on rate constants for the oxidation of SO2by H2O2or O3(Lagrange et al., 1993, 1994, 1996). Two kind of electrolytes were used (NaCl and Na2SO4) at ionic strength ranging from 0.1 to 1.0 M.

Both kind of electrolytes used in this work are present in the aqueous phase of clouds. Na2SO4may be representative of continental and polluted clouds whereas NaCl may be representative of marine clouds. The kinetics were carried out as a function of NO₂, H^þ and total hypochloriteð½HOCl_total¼ ½HOCl þ ½ClO þ ½ClNO2Þ concentrations. Finally we will propose a mechanism describing the observed rate law. Our work gives new information about this reaction. As indicated in Eq.

(2), ClNO2 is a reaction intermediate, formed by Cl^þ transfer from HOCl to NO₂:

NO₂ þHOCl!OHþClNO2 ð2Þ

2. Experimental 2.1. Reagents

Sodium hypochlorite solutions were freshly prepared from commercial solutions with about 3.5% active chlorine (Prolabo, Rectapur). Nitrite solutions were prepared from reagent grade NaNO₂(Carlo Erba, p.a.) and were standardised with MnO₄ solutions (Charlot, 1974). Sodium hydroxide, hydrochloric and sulphuric acids solutions were prepared from Carlo Erba, Nor- mex. Buﬀer solutions (CH₃COONa, Na₂B₄O₄10H₂O and Na2HPO4) were prepared from Merck, p.a. The reagents were used without any further puriﬁcation.

Distilled water was further puriﬁed by passing through a mix bed of ion-exchanger (Millipore) and satu- rated with argon. All solutions were freshly prepared to avoid any complication due to decomposition of hypochlorite or hypochlorous acid. Oxygen free solutions were obtained by bubbling argon through the solution before analytical and kinetic measurements were taken.

(3)

2.2. H^þconcentration andpK_a measurements

For determination the rate law, the concentration of each species must be measured and not their activity.

The activity is related to the pH by:

pH¼ logaHþ and aHþ¼c_Hþ½H^þ

wherea_Hþ is the activity andc_Hþ is the activity coeﬃ- cient. In this work the activity coeﬃcient (c_Hþ) was maintained constant in the presence of high concentration of electrolyte in the solution. H^þ concentration measurements were made with a Tacussel TB/HA glass electrode and (Ag/AgCl/0.1 M Cl) reference electrode.

In NaCl, the glass electrode was calibrated against strong acid solutions (10² M HCl in NaCl), for each ionic strength and both electrolytes. For these standard solutions we adoptedlog½H^þ ¼2:00. In Na₂SO₄, the standardization of glass electrode was performed in basic medium. In Na₂SO₄ (for example: I¼0:7M), a solution of 0.01 M NaOHþ0.23 M Na2SO4, was used as a standard. With regards to the apparent constant Kw¼ ½H^þ½OH and pKw¼13:03 (Table 1), we used log½H^þ ¼11:03 for this standard at 298 K. Thus, in our study for a given ionic medium, the logarithm of the proton concentration was measured and not the activity (Lagrange et al., 1993, 1994, 1996, 1999).

Prior to this study, the acid dissociation constant (pK_a) of HOCl was not known at diﬀerent ionic strengths. Therefore hypochlorite solutions were titrated by acidimetric methods (the calibrated electrode was used to measure the change in pH when HCl or H2SO4

was added to a alkaline ClOsolutions). These titrations were realized in NaCl or Na₂SO₄ solutions at constant ionic strength between 0.1 and 2.0 M at 298 K. The results were analyzed using the Miniquad commercial software (Sabatini et al., 1974). Changes in the pKa are shown in Fig. 1 as a function of ionic strength. Ac-

cording to the extended Debye–H€uuckel law which shows the deviation in higher ionic strength compared with limiting Debye–H€uuckel law (Laidler and Meiser, 1999), the acidity constant can be related to the ionic strength by:

logK¼ A⁰ ffiffi pI 1þ ffiffi

p þI B⁰IþlogK0 ð3Þ where A⁰ and B⁰ are adjustable parameters which are function of temperature, kind of electrolyte and ionic strength. I is the ionic strength, and K0 is the acidity constant at inﬁnite dilution.

The parameters in Eq. (3) were calculated using a nonlinear ﬁtting software. The values of K0, extrapo- lated at zero ionic strength can be expressed as Eqs. (4) and (5) (in two kind of electrolytes), which is in good agreement with pK_a of HOCl, 7.53, reported previously (Weast, 1988):

logKHOCl=ClO¼ 2:76 ffiffi pI 1þ ffiffi

p þI 1:13Iþ7:55

in NaCl at 298 K ð4Þ

logKHOCl=ClO¼ 2:87 ffiffi pI 1þ ffiffi

p þI 1:26Iþ7:55

in Na2SO4 at 298 K ð5Þ

2.3. Kinetic measurements

Nitrite oxidation by ClOwas followed using a UV/

VIS spectrophotometer (Shimadzu Model 1601 with a 10 mm quartz cell) interfaced to a PC. Data sets, which were averaged over at least three replicates, were Table 1

Values of pKw determined by titration for diﬀerent ionic strengths in NaCl and Na2SO4at 298 K

I(M) NaCl Na2SO4

0.05 13:590:01

0.10 13:740:01 13:470:06

0.30 13:630:05 13:230:05

0.50 13:680:03 13:110:07

0.70 13:030:05

0.90 12:960:02

1.00 13:690:03 12:950:06

1.10 12:940:01

1.30 12:880:06

1.50 13:790:06 12:850:04

1.70 12:860:05

2.00 13:820:08 12:840:05

3.00 14:090:11

4.00 14:350:09

Fig. 1. Inﬂuence of the ionic strength on the acidity constant of HOCl/ClOat 298 K.

(4)

acquired using a commercial software (Biologic, 1990).

Faster reactions (nitrite oxidation by HOCl) were followed using a stopped-flow spectrophotometer (Applied Photophysics Model SX 18 MV). This instrument can measure fast kinetics (with total reaction times between 5 ms and 20 min). In this case, the absorption cell had a 2 mm path length. Two syringes rapidly injected equal volumes (0.25 ml) of reactants through a mixer and an observation cell and into another syringe. When the latter was completely filled, the flow stopped and a trigger activated the data acquisition. A 200 W xenon lamp connected to a monochromator was used as a light source. The photomultiplier output was recorded between 260 and 800 nm. In this setup, mixing is complete within 1–4 ms. This stopped-flow spectrophotometer can also be run as a function of time in a multichannel mode by means of a diode array with a time resolution of 10 ms.

For both kinetic measurements, nitrite concentrations were ranging from 5:010³ to 2:510² M (at constant H^þconcentration) were mixed with 1:510³ M hypochlorite solutions in a given electrolyte. The nitrite ions concentrations were in suﬃcient excess (at least 10-fold excess over [HOCl]total) to ensure that the observed reaction was always pseudo-ﬁrst-order with respect to [HOCl] or [ClO]. The measurements were carried out at 2981 K and the reactions were observed at 292 nm by following the loss of [HOCl]_total (Fig. 2).

3. Results and discussion

3.1. Determination of the rate law

Experiments were performed in the pH range from 5 to 11, where the predominating N(III) species was NO₂.

In this range, HOCl as well as its conjugate base ClO, had to be taken into account. The ionic strength was maintained constant at 1.0 or 0.1 M using a kind of electrolyte. In more acidic solution, the formation of Cl2

and Cl₃ prevent the determination of the rate law and thus these experimental conditions were not chosen.

Inﬂuence of HOCl/ClOon the oxidation rate: When [H^þ] and [NO₂] were kept constant, the plot of absor- bance versus time was an exponential (Fig. 3). These results show ﬁrst-order kinetics with respect to total hypochlorite concentrationð½HOCl_total¼ ½HOCl þ ½ClO þ Pall other kind of ClðþIÞsuch as ClNO₂Þ.

Therefore, the rate law can be written as:

d½HOCl_total

dt ¼kobs½HOCl_total wherekobsis function

of ½H^þand½NO₂ ð6Þ

Inﬂuence of NO₂ concentration on the oxidation rate:

Table 2 summarises the effect of a variation of the NO₂ concentration on the rate coefficient. Thekobs obtained at constant H^þconcentration, for various initial nitrite concentrations (from 0:2510² up to 2:510² M) and for different ionic strengths in NaCl and Na₂SO₄as a kind of electrolyte. The following function has been adjusted to our experimental data which were obtained by classical UV/VIS spectrophotometry in alkaline solution (Fig. 4a):

kobs¼A½NO₂ þB½NO₂² at pH¼10:66 ð7Þ At 298 K and in Na2SO4 at an ionic strength (I ¼1:00 M) the values for A and B are ð9:431:41Þ 10¹ M¹s¹ and ð1:080:16Þ 10⁴ M²s¹, respectively.

The results for more acidic medium were not the same as alkaline solution. The results were developed in more Fig. 2. UV/VIS spectra of Cl(I) in 1.0 M NaCl: (a) pH¼4:00

and (b) pH¼9:61. Fig. 3. Kinetic proﬁle obtained from stopped-ﬂow spectro-

photometer: ½NO₂ ¼110² M; ½HOCl_total¼110³ M;

1.0 M NaCl; pH¼6:52 at 298 K;½phosphate ¼0:02 M.

(5)

acidic medium using a phosphate buﬀer (0.02 M). With regards to the mechanism developed further (9)–(15) and the overall rate law (Eq. (19)), the following function has been adjusted to our experimental data for a larger scale of nitrite concentrations in the above mentioned conditions (Fig. 4b):

d HOCl½ _total

dt ¼A⁰½NO₂²þB⁰½NO₂

1þC⁰½NO₂ at pH¼6:26 ð8Þ

At 298 K and in Na2SO4 at an ionic strength (I¼1:00 M), the values forA⁰,B⁰andC⁰areð2:440:48Þ 10⁸, ð1:570:32Þ 10⁶ andð2:530:51Þ 10⁴, respectively.

Inﬂuence of H^þ concentration on the oxidation rate:

To determine the role of [H^þ] in these experiments, measurements were carried out at different pH (between 5.00 and 11.0) with the nitrite concentration being constant at 1:010²M. The reaction rate appeared to be a hyperbolic function of [H^þ] for different kind of electrolytes and different ionic strengths (as shown in Fig. 5). In acidic solutions, we used different buffers Table 2

Pseudo-ﬁrst-order rate constants for the oxidation of nitrite by HOCl/ClOas a function of pH

log½H^þ ½NO₂ 10²(M) kobsmean (h¹) log½H^þ ½NO₂ 10²(M) kobsmean (s¹) A

0.50 0:660:03

1.00 1:850:15 0.40 96.73.0

10.0 1.25 2:750:10 6.26 0.50 1114

1.50 3:950:20 0.75 1376

2.00 6:500:25 1.00 1555

2.50 8:900:40

B

0.50 0:260:01 0.25 24.30.7

1.00 1:280:03 0.40 35.90.8

10.71.25 2:020:06 6.50 0.45 40.10.9

1.50 2:710:070.50 44.81.0

2.00 4:390:20 1.00 84.02.0

2.50 7:400:20 1.50 1266

C

0.25 0:350:02 0.4 783

1.00 4:000:08 0.5 874

10.5 1.50 8:300:30 6.50 0.75 1056

2.00 16:10:6 1.00 1188

Experimental conditions:½HOCl ¼1:0510³M at 298 K. A: Na2SO4,I¼1:0 M, B: NaCl,I¼1:0 M and C: NaCl,I¼0:1 M.

– –

Fig. 4. Dependence ofkobson nitrite concentration at 298 K in Na2SO4, I¼1:00 M, in various [H^þ]: (a)log½H^þ ¼10:66 and (b)log½H^þ ¼6:26. Solid lines are ﬁts to Eqs. (7) and (8) (see text).

(6)

(phosphate, borate and mix of them at a concentration 0.02 M).

3.2. Proposed mechanism

Kinetic and mechanistic studies of the nitrite oxidation by HOCl/ClO (or Cl2) have been reported by Pendlebury and Smith (1973), Cachaza et al. (1976), Johnson and Margerum (1991), Behnke et al. (1997) and Frenzel et al. (1998). These investigators reported that the reaction of NO₂ with HOCl or Cl₂ proceeds via the formation of nitryl chloride (ClNO2):

Cl₂þNO₂ ClNO₂þCl ðk₁⁰=k₁⁰ Þ HOClþNO₂ClNO₂þOH ðk⁰₂=k⁰₂Þ

In our proposed mechanism, the formation of nitryl chloride is also supposed (Eq. (10)). In water, ClNO2

dissociates to produce a very short-lived intermediate NO^þ₂ (Behnke et al., 1997). Given its rather short life- time, our proposed mechanism (9)–(15) assumes steady state for NO^þ₂ and thus summarises k⁰₃=k₃⁰ and k⁰₄ as reaction (13).

ClNO2NO^þ₂ þCl ðk⁰₃=k⁰₃Þ NO^þ₂ þH2O!NO₃ þ2H^þ ðk₄⁰Þ

Behnke et al. (1997) suggested that the heterogeneous loss of ClNO2in chloride solutions is much slower than in water, indicating that the main loss process, k⁰₃ followed byk₄⁰, is suppressed byk₃⁰ in the presence of Cl and that the reactionk₁⁰ must be ineﬃcient, even at high chloride concentrations. The reaction of ClNO₂ in nitrite solutions and subsequent formation of N2O4, has been suggested by Pendlebury and Smith (1973). Thus, in this work, we agree with the mechanism given by previous work (Cachaza et al., 1976; Johnson and Margerum, 1991) where HOCl reacts with NO₂ by Cl^þ transfer to give nitryl chloride (Eq. (10)). We also agree that ClNO₂ decomposes by two paths (Eqs. (11) and (13)) the ﬁrst one being attack by NO₂ to form N2O4

(Eq. (11)), and the second step of this mechanism is the hydrolysis of ClNO2and formation of nitrate (Eq. (13)):

very fast ClOþH^þ^K^aHOCl ð9Þ fast HOClþNO₂^K¹ClNO2þOH ð10Þ slow ClNO2þNO₂!^k¹ N2O4þCl ð11Þ fast N2O4þOH!^k² NO₃ þNO₂ þH^þ ð12Þ slow ClNO2þH2O!^k³ NO₃ þClþ2H^þ ð13Þ very fast HNO₃

K_HNO3

NO₃ þH^þ ð14Þ

very fast H2O^K^wOHþH^þ ð15Þ The equilibrium constants for (9) and (10) are:

Ka¼ ½HOCl

½H^þ½ClO and K1¼½ClNO2½OH

½HOCl½NO₂ ð16Þ Since we have pseudo-ﬁrst-order conditions, the rate law can be written as:

d½HOCl_total

dt ¼kobs½Hypochlorite_total ð17Þ where½HOCl_total¼ ½HOCl þ ½ClO þ ½ClNO2.

As the rate law is the sum of two determining (slow) steps, we have:

d½HOCl_total

dt ¼k1½ClNO2½NO₂ þk3½ClNO2 ð18Þ With these assumptions, the expression for the rate law is given in Eq. (19) on the basis of a steady-state treat- ment for the ClNO2.

d½HOCl_total

dt ¼

k1K1Ka½NO₂²þk3K1Ka½NO₂

n o

½H^þ²

KwþKaKw½H^þ þKaK1½NO₂½H^þ² ½HOCl_total

ð19Þ Fig. 5. Pseudo-ﬁrst-order rate constant (kobs) of nitrite oxida-

tion as a function of [H^þ] at 298 K in various ionic media. At t¼0,½HOCl_total¼1:0510³M and½NO₂ ¼1:0010²M.

(d) Measured by classical spectroscopy and without using any buffer; (.) measured by stopped-flow and using phosphate (0.02 M) as a buffer; () measured by stopped-flow and using borate (0.02 M) as a buffer: (a) NaCl,I¼1:0 M; (b) NaCl,I¼0:1 M;

(c) Na2SO4,I¼1:0 M. Solid lines are ﬁts to Eq. (19) (see text).

(7)

The parameters for this equation are adjusted by a generalized nonlinear least-squares ﬁtting procedure.

The results at 298 K are summarized in Table 3 as a function of kind of electrolyte and ionic strength. In NaCl, we observed a sharp decrease of the rate constant of reaction (13) with increasing NaCl, which is very good agreement with results obtained by Zetzsch and coworkers (Behnke et al., 1997), who assigned this eﬀect tok₃⁰ .

Lister and Rosenblum (1961) studied this reaction in alkaline solutions, i.e. 13<pH<13:6 in NaCl (1:5<

I<2:5 M). They proposed a direct reaction of nitrite with HOCl and they did not report any second order component for the rate law. We agree with the mechanism given by Cachaza et al. (1976) but in the case of ClNO₂dissociation, our proposed mechanism is slightly different. As discussed above, the second rate determining step is the hydrolysis of ClNO₂. These authors studied the oxidation of nitrite by ClOin a very narrow acidity range (i.e. 10:8<pH<11:6) at different tem- peratures (292–303 K). From their mechanism at 298 K in 0.4 M NaNO3 and NaCl solution, they reported a complex rate law and had to assume many simplifica-

tions and ﬁnally reported a reduced rate law. Johnson and Margerum (1991) studied this reaction in a wider range of acidity (8<pH<13) in NaClO4(I ¼0:5 M) at 298 K by spectroscopic measurements. The results obtained by Johnson and Margerum (1991) are consistent with the work by Cachaza et al. (1976). Thus, in their mechanism ClO, HOCl and ClNO₂are preequilibrium species and the rate law must be calculated by consid- ering the sum of all preequilibrium species.

The rate coeﬃcient for the reaction of ClNO2 with NO₂ (Eq. (11)) and the reaction of hydrolysis of ClNO₂ (Eq. (13)) reported in this work (see Table 3) in NaCl (I¼1:00 M), are in very good agreement, with the values obtained by Frenzel et al. (1998) using a wetted- wall ﬂowtube (k1¼7:9810³ M¹s¹,k⁰₃¼90 s¹ and k₄⁰ ¼8:9010³ M¹s¹).

4. Conclusions

Very high ionic strengths, as high as 20 M, may be seen in highly concentrated aerosols (Stelson and Seinfeld, 1981). In this work, the rate coeﬃcient for the Table 3

Kinetic and thermodynamic parameters^afor the oxidation of nitrite by HOCl/ClOat 298 K

Na2SO4(I¼1:00) M NaCl (I¼1:00) M NaCl (I¼1:010) M

Ka(M) 1.2810⁷ 1.9910⁷ 0.8910⁷

k1(l mol¹s¹) ð131Þ 10³ ð8:00:7Þ 10³ ð2:90:2Þ 10³

k3(s¹) 8174.80.4 797

K1 ð1:10:1Þ 10⁴ ð8:80:8Þ 10⁴ ð2:00:2Þ 10⁴

pKw 12.95 13.68 13.75

aDerived from the ﬁt to Eq. (19). Errors represent only the statistical error at the 2rlevel.

Fig. 6. Global diagram of proposed mechanism of the reaction of nitrite oxidation by HOCl/ClO.

(8)

oxidation of nitrite by Cl(þI) has been studied at three different ionic strengths at two ionic environments which are important parameters since they can lead to important variations of reaction rates. Our results show that the rate is influenced by both ionic strength and ionic environment. For the rate determining step (i.e. reaction of nitrite by ClNO2, Eq. (11)), we observed that the rate coefficient in Na₂SO₄ is about 1.5 times faster than in NaCl (at the same ionic strength). In NaCl, by a factor of one order of magnitude increase of ionic strength, the rate coefficient is 3 times faster. In the case of hydrolysis of ClNO2(Eq. (13)), we observed the same effect of kind of electrolyte. The reaction in Na₂SO₄is about 2 orders of magnitude faster than in NaCl. For the latter reaction (Eq. (13)), we observed the opposite effect in NaCl solution. A sharp decrease for the rate constant of the reaction (11) with increasing NaCl concentration was observed which this effect has been also observed by other investigators using another method, wetted-wall flowtube (Behnke et al., 1997). Our mechanistic data confirm previous work (Johnson and Margerum, 1991) that this reaction is not an oxygen atom transfer process (Anbar and Taube, 1958). The experimental results show that the reaction leading to N₂O₄ formation and the hydrolysis of ClNO2 are the rate determining steps and N₂O₄and ClNO₂ are intermediate species.

In conclusion, a reaction mechanism for the reaction of NO₂ with HOCl/ClOis shown in Fig. 6.

Acknowledgements

The authors are extremely grateful to Christian GEORGE, University of Lyon 1, for valuable discus- sions and access to laboratory facilities. Nazafarin LA- HOUTIFARD wishes to thank Prof. C. ZETZSCH, FHG-ITA at Hannover, for valuable discussion. Sup- port of this study by the PNCA (Programme National de Chimie Atmospheerique) is gratefully acknowledged.

In addition, this work was a contribution to project EUROTRAC, subproject CMD.

References

Anbar, M., Taube, H., 1958. The exchange of hypochlorite and of hypobromite ions with water. J. Am. Chem. Soc. 80, 1073–1077.

Ayers, G.P., 1982. The chemical composition of precipitation:

a southern hemisphere perspective. In: Goldberg, (Ed.), Atmospheric Chemistry. Springer-Verlag, New York, pp.

41–56.

Behnke, W., George, Ch., Scheer, V., Zetzsch, C., 1997.

Production and decay of ClNO2 from the reaction of gaseous N2O5 with NaCl solution: bulk and aerosol experiments. J. Geophys. Res. 102 (D), 3795–3804.

Biologic Biokine, 1990. UserÕs Manual. Biologic Company, Echirolles.

Brauer, M., Koutrakis, P., Keeler, G.J., Spengler, J.D., 1991.

Indoor and outdoor concentrations of inorganic acidic aerosols and gases. J. Air Waste Manage. Assoc. 41, 171–

181.

Cachaza, J.M., Casado, J., Castro, A., Lopez Quintela, M.A., 1976. Kinetics of oxidation of nitrite by hypochlorite ions in aqueous basic solution. Can. J. Chem. 54, 3401–3406.

Charlot, G., 1974. Chimie Analytique Quantitative. Masson and Cie, Paris.

Finlayson-Pitts, B.J., Pitts, J.N., 2000. Chemistry of the Upper and Lower Atmosphere, Theory, Experiments and Appli- cations. Academic Press, San Diego.

Frenzel, A., Scheer, A., Sikorski, R., George, Ch., Behnke, W., Zetzsch, C., 1998. Heterogeneous interconversion reactions of BrNO2, ClNO2, Br2, and Cl2. J. Phys. Chem. (A) 102, 1329–1337.

Fuzzi, S., Orsi, G., Nardini, G., Facchini, M.C., McLaren, S., McLaren, E., Mariotti, M., 1988. Heterogeneous processes in the Po valley radiation fog. J. Geophys. Res. (D) 93, 11141–11151.

Harris, G.W., Carter, W.P.L., Winer, A.M., Pitts, J.N., Platt, U., Perner, D., 1982. Observations of nitrous acid in the Los Angeles atmosphere and implications for predictions of ozone-precursor relationships. Environ. Sci. Technol. 16, 414–419.

Johnson, D.W., Margerum, D.W., 1991. Non-metal redox kinetics: a re-examination of the reaction between hypochlorite and nitrite ions. Inorg. Chem. 30, 4845–4851.

Lagrange, J., Pallares, C., Wenger, G., Lagrange, Ph., 1993.

Electrolyte eﬀects on aqueous atmospheric oxidation of sulphur dioxide by hydrogen peroxide. Atmos. Environ. 27, 129–137.

Lagrange, J., Pallares, C., Lagrange, Ph., 1994. Electrolyte eﬀects on aqueous atmospheric oxidation of sulphur dioxide by ozone. J. Geophys. Res. (D) 99, 595–600.

Lagrange, J., Pallares, C., Wenger, G., Lagrange, Ph., 1996.

Kinetics of sulphur(IV) oxidation by hydrogen peroxide in basic aqueous solution. Atmos. Environ. 30, 1013–1018.

Lagrange, C., Wenger, G., Lagrange, Ph., 1999. Kinetic study of HMSA formation and decomposition: tropospheric relevance. J. Chim. Phys. 96, 610–633.

Laidler, K.J., Meiser, J.H., 1999. Physical Chemistry, third ed.

Houghton Miﬄin Company, Boston.

Lister, M.W., Rosenblum, P., 1961. The oxidation of nitrite and iodate ions by hypochlorite ions. Can. J. Chem. 39, 1645–

1651.

Miller, D.R., Byrd, J.E., Perona, M.J., 1991. The source of Pb, Cu and Zn in fog water. Water Air Soil Pollut. 32, 329–340.

Pendlebury, J.N., Smith, R.H., 1973. Kinetics of oxidation of nitrite by aqueous chlorine. Aust. J. Chem. 26, 1857–

1861.

Sabatini, A., Vacca, A., Gans, P., 1974. MINIQUAD: a gen- eral computer program for the computation of formation constants from potentiometric data. Talanta 21, 53–77.

Seinfeld, J.H., 1986. Atmospheric Chemistry and Physics of Air Pollution. Wiley Interscience Publication, New York.

Sigg, L., Stumm, W., Zobrist, J., Zurcher, F., 1987. The chemistry of fog: factors regulating its composition. Chimia 41, 159–165.

(9)

Simon, P.K., Dasgupta, P.K., 1995. Continuous automated measurement of gaseous nitrous and nitric acids and particulate nitrite and nitrate. Environ. Sci. Technol. 29, 1534–1541.

Stelson, A.W., Seinfeld, J.H., 1981. Chemical mass accounting of urban aerosol. Environ. Sci. Technol. 15, 671–679.

Takenaka, N., Daimon, T., Ueda, A., Sato, K., Kitano, M., Bandow, H., Maeda, Y., 1998. Fast oxidation reaction of nitrite by dissolved oxygen in freezing process in the tropospheric aqueous phase. J. Atmos. Chem. 29, 135–150.

Weast, R.C., 1988. CRC Handbook of Chemistry and Physics.

CRC Press, Boca Raton, FL.