Environ. Sci. Technol. 2003, 37, 379-385
Electron-Beam Treatment of Aromatic Hydrocarbons That Can Be Air-Stripped from Contaminated Groundwater. 2. Gas-Phase Studies LUTZ PRAGER,† GERTRAUD MARK,‡ H A R T M U T M A¨ T Z I N G , § HANNS-RUDOLF PAUR,§ JU ¨ RGEN SCHUBERT,| FRITZ H. FRIMMEL,⊥ SEBASTIAN HESSE,⊥ HEINZ-PETER SCHUCHMANN,‡ MAN NIEN SCHUCHMANN,‡ AND C L E M E N S V O N S O N N T A G * ,‡ Max-Planck-Institut fu ¨ r Strahlenchemie, Stiftstrasse 34-36, P.O. Box 101365, 45413 Mu ¨ lheim, Germany, Institut fu ¨ r Oberfla¨chenmodifizierung (IOM), Permoserstrasse 15, 04318 Leipzig, Germany, Forschungszentrum Karlsruhe, Institut fu ¨ r Technische Chemie, P.O.Box 3640, 76021 Karlsruhe, Germany, Stadtwerke Du ¨ sseldorf, Luisenstrasse 15, 40215 Du ¨ sseldorf, Germany and Universta¨t Karlsruhe, Engler-Bunte-Institut, Engler-Bunte-Ring 9, 76131 Karlsruhe, Germany
The electron-beam (EB) degradation of volatile aromatics (benzene, toluene, ethylbenzene, xylenes: BTEX) in groundwater strip gas, which in the present work has been modeled by the introduction of the desired aromatic(s) to a stream of air or another gas, such as oxygen, is initiated essentially by the addition of •OH radicals to the aromatic ring, giving rise to hydroxycyclohexadienyl radicals, which form the corresponding peroxyl radicals upon addition of oxygen. As studied in some detail with benzene as a BTEX representative, various reactions of these lead to numerous oxidation products in a cascade of reactions, including the decomposition of products under the prevailing conditions of high turnover of the initial aromatic. Importantly, hydroxycyclohexadienylperoxyl radical formation is partly reversible, and the reactions of the hydroxycyclohexadienyl radicals, which thus have a significant presence in these systems, must therefore also be taken into consideration. In the gas phase, in contrast to the aqueous phase (see Part 1), the reactions of the hydroxycyclohexadienyl radicals lead to oligomeric products that appear to contribute, in addition to ionic clusters, to nucleation for the aerosols observed. Various nitrated products, among them nitrophenols, are observed when air is used for the stripping. However, these studies did not clear the pilot plant stage, since BTEX degradation using a bioreactor carried out in parallel was so successful that the EB technology was judged to be noncompetitive. As for the latter, expensive equipment consisting of a stripper, the EB * Corresponding author fax: +49-208-306-3951; e-mail: Clemens@ vonSonntag.de. † Institut fu ¨ r Oberfla¨chenmodifizierung. ‡ Max-Planck-Institut fu ¨ r Strahlenchemie. § Forschungszentrum Karlsruhe. | Stadtwerke Du ¨ sseldorf. ⊥ Universta ¨ t Karlsruhe. 10.1021/es020930d CCC: $25.00 Published on Web 12/07/2002
2003 American Chemical Society
machine, and an aerosol precipitator would be required. The condensed aerosols are biorefractory and would require further treatment for detoxification.
Introduction As mentioned in Part 1 of this study, the incentive to investigate the destruction of benzene and its alkylated derivatives (toluene, ethylbenzene, xylenes), BTEX, by stripping and electron-beam (EB) irradiation was a necessity to ameliorate a large area at a former coking site. Parallel to trying this technology, we also attempted to degrade the BTEX using a bioreactor. The latter study is reported elsewhere (1). In Part 1, we presented a detailed study on the reactions of •OH with benzene, a model of the BTEX group, in aqueous solution in the absence and the presence of another radical, • NO2. In the radiolysis of humid air, •OH and •NO2 are formed (see below). Although only •OH is sufficiently reactive to attack the aromatics (Scheme 1, reaction 1), it will be shown that •NO plays a major role in the ensuing free-radical reactions 2 and also gets incorporated into the products. In the presence of O2, these hydroxycyclohexadienyl radicals 1 are converted into the corresponding peroxyl radicals 2 and 3 (reactions 3 and 4); these reactions are reversible both in aqueous solution (2-4) and in the gas phase (5-8). As shown below, this reversibility is of considerable importance under electron-beam irradiation in the gas phase in that it may inhibit the oxidation process (e.g., reactions 5 and 6), and contributes via reactions 2, 7, and 8 to the formation of aerosols. The •NO2-derived products also have an impact on the prospects of the electron-beam process, since nitrated compounds may represent an enhanced environmental hazard, because they are usually refractory toward biological degradation.
Experimental Section Electron-beam irradiation of “synthetic” strip gas (i.e., a stream of gas to which the aromatic was added to the desired concentration) was carried out in Karlsruhe (ambient air) and Leipzig (humidified O2). Samples of irradiation products separated from the air stream (aerosol, aqueous solutions from washing bottles) were collected in Karlsruhe and analyzed in Mu ¨ lheim. γ-Radiolysis of benzene in aqueous solutions and in glass vessels filled with humidified air was performed in Mu ¨ lheim. In the Karlsruhe EB pilot-plant-sized flow reactor (9) (cross section 0.3 × 0.6 m2, length 0.8 m), 500 m3 h-1 (volume refers to standard conditions of 0 °C and 1 bar) ambient air with a water partial pressure below saturation and BTEX vaporladen by injection (e.g., 200 mg m-3) was irradiated to a dose of up to 10 kGy (residence time in the reactor, ∼1 s). For pilot plant process studies, the irradiated gas with the aerosol produced was passed through a washer column and a custommade wet electrostatic precipitator (Noell-KRC, Bensberg). To collect aerosol samples for chemical analysis, 1.7 m3 of the gas was passed through membrane filters (Millipore, Fluoropore membrane type FG, pore size 0.2 µm). Separately, ∼0.2 or 0.4 m3 of the filtered gas was passed through two washing bottles in series, each filled with 75 mL of water. Further EB irradiations (Linac Elektronika 10 MeV, 16.5-Gy pulses of 5-µs duration, 50 pulses/s, beam moved back and forth along the water-cooled reactor at a sweep rate of 1 Hz) were carried out in Leipzig. The cooling water, kept at 40 °C, prevented condensation at the reactor surface and also served as a medium for the generation of secondary electrons, which are absorbed by the gas with high efficiency. Dosimetry was VOL. 37, NO. 2, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
379
SCHEME 1
performed by determining the ozone formed in synthetic air (10). The aromatics were added to the gas stream (120 L/h) by means of a syringe pump (Bioblock, Razel Scientific Instruments). The gas was humified by passing it over a surface of water kept at constant temperature. The water vapor content could be varied by changing the level of the water surface in the humidification vessel. It was determined by FT-IR (Vector 22, Bruker Analytic) together with the gaseous products (detection limit ∼2 ppm; for comparison, the initial concentration of the aromatics was on the order of 102 ppm). γ-Radiolysis of humidified air was carried out (in Mu¨lheim) in glass-stoppered 830-mL vessels filled with synthetic air passing through a water-filled washing bottle. Benzene (50 µL) was then injected, and the stopper was put on. After irradiation, 10 mL of water was introduced to wash out the products for analysis. The dose rate was determined by Fricke dosimetry. This approximation seemed justifiable considering the irradiation vessel in the light of cavity-chamber theory (11), under the assumption that the dose rate in a cavity with graphite walls for practical purposes equals that in a volume of water, whereas the dose rate in a glass-walled cavity is only a few percent higher (cf. Figure 1 in 12, cf. also Figure 3 in ref 11). Irradiated samples were analyzed by HPLC (precolumn 40 mm, column 12.5 cm Nucleosil 5 C18, optical detection by photodiode array); eluent: methanol/water/acetic acid 75:25:1 for biphenyl, 30:70:1 for phenol, 7:30:1 for the nitrated products. Acids were determined by HPIC (Dionex DX-100, precolumn AG9-SC 4 × 50 mm column AS9-SC, 4 × 250 mm; eluent, 1.8 mM CO32-/1.7 mM HCO3- or 3.5 mM CO32-/1.0 mM HCO3-). Aerosol particle size distributions were determined by a scanning mobility particle sizer SMPS (TSI, Aachen, Germany). An aqueous solution of the aerosol was subjected to gelpermeation chromatography (TSK HW 40S, 40 µm, column 250 × 20 mm, eluent 0.03 M phosphate buffer pH 6.6 at 1.0 mL/min) with optical (254 nm) and DOC detection (Gra¨ntzel, Karlsruhe). The stationary phase TSK HW 40S (Toyopearl; column packed by Grom, Herrenberg, Germany) is a terpolymer of oligoethylene glycol, glycidyl methacrylate, and pentaerythritol dimethacrylate. Poly(ethylene glycol) (Fluka; MW 4000, 1550, and 1000), raffinose, maltose, glucose, and glycerol were used for the molecular weight calibration.
In contrast to the situation in a dry mixture, much of the energy deposited in the humid gas is channeled into the production of •OH via a cascade that may be summarily represented by reactions 9-11, given the large excess of H2O over the organics.
N2(O2 ) f N2•+(O2•+) + e-
(9)
N2•+(O2•+) + H2O f H2O•+ + N2(O2)
(10)
H2O•+ + nH2O f •OH + H+(H2O)n
(11)
e- + O2 f O2•-
(12)
O2•- + nH2O f O2•-(H2O)n
(13)
Neutralization and termination reactions lead to the removal of O2•-/HO2• (reactions 14 and 15).
H+(H2O)n + O2•-(H2O)n f HO2• + 2nH2O
(14)
2HO2• f H2O2 + O2
(15)
At these relatively low concentrations of the organic component with respect to the water-vapor content, •OH is the main primary reactive species; it reacts with organic components by H-atom abstraction or by addition to a C-C double bond, that is, in the case of aromatics, by addition to the ring. Gas-phase reactions of •OH with volatile aromatics (6, 17-22) are of interest regarding pollution-abatement and atmospheric chemistry research alike. Obviously, as the destruction of the initial pollutant proceeds, the products are progressively oxidized in their turn (23). In its initial stage, the radiation chemistry of organics in humid gases, such as air or its elements, shows considerable resemblance to their aqueous solution radiation chemistry (e.g., generation of •OH). Differences then appear as hydrolytic reactions, and other reactions that involve the emergence and separation of positive (proton) and negative species (especially radical anions that are produced as the cascade of reactions unfolds) are strongly disfavored in the gas phase. Moreover, thermal equilibration of any vibrationally excited species is somewhat slower than in the liquid phase because of the lower density of the heat bath. In air, the formation of nitrogen oxides gives rise to nitro compounds among the products. Reactions 16-22 have been put forward (15) that lead to the formation of •NO2.
O2•+ + e- f 2O
(16)
N2•+ + e- f 2N
(17)
O2•+ + O2•- f 2O + O2
(18)
N2•+ + O2•- f 2N + O2
(19)
N + O2 f •NO + O
(20)
O + O2 f O3
(21)
Results and Discussion
Generation of •OH and •NO2 upon Radiolysis of Humid Air. The radiolytic decomposition of gases has attracted interest since the early years of radiation chemistry (for reviews, see refs 13-16). Air-based strip gas is a mixture (typically 3.1 × 10-2 M N2, 8 × 10-3 M O2, ∼5 × 10-4 M H2O at a water temperature of 10 °C, and up to a few times 10-6 M organics) detailed radiation chemistry of which is complex (10), not the least of which is due to the involvement of peroxyl radicals. 380
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 2, 2003
•
NO + O3 f •NO2 + O2
(22)
Further reactions that give rise to •NO2 are conceivable in the presence of organic additives (reactions 23, ref 24; reactions 24 and 25, ref 25). •
NO + O2 f ONOO•
(23)
ONOO• + RO2• f •NO2 + RO• + O2
(24)
•
NO + RO2• f •NO2 + RO•
(25)
The introduction of the nitro function into the organic material might at the same time also occur by the recombination of free radicals with •NO, and subsequent oxidation of the nitroso to the nitro function by peroxyl radicals (26, 27) (reactions 26 and 27). •
NO + R• f RNO
(26)
RNO + RO2• f RNO2 + RO•
(27)
Much of the radiolytically produced ozone is expected to react with the olefinic products, but some also react with carbon-centered radicals, such as hydroxycyclohexadienyl 1 (reaction 28). This reaction is fast, and in contrast to that with O2 (see below), is irreversible. •
C6H5OH 1 + O3 f C6H5(OH)O• + O2
(28)
The transformations in the gas phase of hydroxycyclohexadienyl-type radicals (e.g., 1), which are the main precursors of the multitude of radiolysis products that are observed under these conditions, have been extensively studied (6, 17-22, 28-33). Scheme 1 describes the initial stages of the OH-radical-induced oxidation of the prototype benzene in oxygenated aqueous solution as well as in air. However, the O2•--elimination pathways from the peroxyl radicals 2 and 3 (see part 1) are much less accessible in the gas phase, in contrast to the situation in aqueous solution. As a consequence, the reversibility of the formation of these dienyl-type peroxyl radicals (3-6, 34) becomes mechanistically more important, since this opens the way for the recombination of hydroxycyclohexadienyl radicals 1 with their corresponding peroxyl radicals 2, 3 as well as with each other (reactions 5-8). The recombination products that may be formed under these conditions and possibly others are hard to determine quantitatively, for example, in an atmospheric-chemistry reaction chamber, because of their low volatility at ambient temperature. Product studies of the radiolysis of aromatic hydrocarbons generally suffer from the difficulty of analyzing products of very low volatility that may stick to the walls or form an aerosol (35). In the present case, the operation of the electronbeam oxidation as a flow system and the use of filters has allowed the collection of such an aerosol. Aerosol formation has also been reported during the EB treatment of humidified chlorobenzene vapor (36). γ-Radiolysis. Nitrogen oxides are unavoidably produced in the radiolysis of air (15, 16). Thus, in the radiolysis of benzene-laden humidified air, one observes the formation of several nitro compounds (Figure 1 and Table 1; for EB radiolysis, see below). The results in Table 1 show that under these conditions, G(•NO2) has a value of not less than 1 × 10-7 mol J-1. In this case, any oligomeric material that is expected to have been produced under these gas-phase conditions as shown below has not been investigated. The mechanism regarding the
FIGURE 1. Formation of nitrobenzene (b), 2-nitrophenol (4), and 4-nitrophenol (2) in the γ-radiolysis of humidified air containing 4 × 10-4 mol dm-3 benzene.
TABLE 1. γ-Radiolysis of Benzene (4 × 10-4 M) in Humidified Air in a 830-mL Glass Container at 0.5 Gy s-1a G, 10-7 mol J-1
product phenol nitrate nitrobenzene o-nitrophenol p-nitrophenol nitrocatechol glyoxal sum of formic, acetic, glycolic, and lactic acids oxalic acid malonic acid succinic acid tartronic acid maleic acid a
Products with G values. Total dose, 2-6 kGy.
b
2.0b 0.5 0.03 0.2 0.2 0.02 trace ∼2 ∼0.2 ∼0.02 trace trace ∼0.06
For doses up to 720
Gy.
FIGURE 2. γ-Radiolysis, in a 830-mL glass container, of an equimolar mixture of benzene (b), toluene (4), ethylbenzene (O), and m-xylene (2) (each 10-4 M) in humidified air at 1.9 kGy/h. reactions of •NO2 is assumed to resemble that shown in Scheme 5 of Part 1, except for the absence of the hydrolysis of N2O4. In an equimolar mixture of m-xylene, ethylbenzene, toluene, and benzene, the rates of degradation exceed one another in that order (Figure 2), following the same trend as the rate constants for the reactions of these aromatics with •OH (37). From Figure 2, one arrives at a very rough estimate of an initial G value of below 2 × 10-6 mol J-1 for the disappearance VOL. 37, NO. 2, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
381
TABLE 2. EB Radiolysis, in a Flow Reactor, of a Mixture of Benzene (c0 64 ppm ) 1.0 µM) and p-Xylene (c0 46 ppm ≈ 0.5 µM) in Humidified Oxygena
a
dose, kGy
[benzene]/[benzene]0
σn-1
[xylene]/[xylene]0
σn-1
[xylene]/[benzene]
0 7.2 10.8 16.5
1 0.96 0.97 0.83
0.024 0.039 0.025 0.012
1 0.28 0.16 0.11
0.030 0.027 0.032 0.039
0.72 0.21 0.12 0.09
Train (50 Hz) of 16.5 Gy pulses of 5 µs duration, sweep rate 1 hz. Standard deviation, σn-1, with five experimental values.
TABLE 3. Analysis of the Aerosol from the EB-Degradation of Benzene in Aira dose, kGy; ([benzene], mg m-3)
benzene consumptionb G(-benzene)/10-7 mol J-1 aerosol aerosol, benzene equivalentc nitric acid oxalic acid malic acid meso-tartaric acid/malonic acid D,L-tartaric acid/maleic acid tartronic acid phenol catechol hydroquinone o-nitrophenol p-nitrophenol glyoxal a
Gas volume sampled, 1.7 m3. Amounts in mg m-3
b
1.8; (200)
1.8; (400)
5; (200)
10; (200)
36 2.6 26 11.3 0.08 0.13 0.05 0.05 0.24 0.01
