Weschler and co-workers (Weschler et al., 1992a, 1994) suggested that the formation of N03 indoors,

followed by its well-known secondary chemistry to form hno3,

may be an important indoor source of HN03. Indirect evidence for an indoor source is the measurement of the indoor-to-outdoor ratio (I/O) of HN03 and S02 in residences in several locations on the east coast of the United States during the summer months. The I/O

ratio for HN03 was about the same as that for S02 at one location and larger at a second site. This is not expected, since HN03 is much more rapidly lost to surfaces than is S02. In a third location, the I/O ratio for HN03 was smaller than that of S02, but not by as much as would be expected from its rapid wall loss. Such data are indicative of an indoor source of HN03, and Weschler et al. (1992a) suggest reactions (3)-(6).

Evidence for such indoor chemistry has also been obtained by measuring NO, NO , and 03 indoors and outdoors in an office building (Weschler et al., 1994). For example, as 03 rises and NO decreases during the morning hours, the 03 reaction with NO indoors leads to faster decay of NO than otherwise expected and a slower rate of increase of 03. On the other hand, when NO is rising and 03 falling in the early evening, this reaction speeds up the decay of 03 and slows the increase in NO.

Weschler and Shields (1997b) suggest that with the higher concentrations of VOCs indoors, their reactions with 03, N03, and OH may be important. There is some experimental evidence that this is indeed the case. For example, carpet exposed to 03 in a chamber generated HCHO, benzaldehyde, benzoic acid, and acetophenone, all expected products from the reaction of 03 with styrene (Zhang et al., f994b) emitted from the latex adhesive used to bind the backing to the carpet; styrene decreased simultaneously (Weschler et al., 1992b). The formation of a series of C5-C10 aldehydes was observed, which appeared to be from the reaction of 03 with nonvolatile organics associated with the carpet fibers. Concentrations of HCHO increased by up to a factor of 3 and CH3CHO by up to a factor of 20 in the presence of 03. Interestingly, no additional effect was observed when N02 was also present, suggesting that the nitrate radical was not a significant contributor to the formation of these aldehydes and ketones.

Salthammer et al. (1999) examined emissions from commonly available coatings used on furniture and identified numerous oxidation products. These were observed without the addition of oxidants such as ozone, indicating that oxidation in air (perhaps including pho-todecomposition for some compounds) under typical conditions is sufficient to generate such products. For example, emissions of 2-ethylhexanol were identified from di-2-ethylhexyl phthalate, used as a plasticizer in many coatings.

Reiss et al. (1995b) exposed latex paint to 03 and observed the production of HCHO as well as CH3CHO and CH3COCH3 for some paint samples. They proposed that these were formed by the reactions of 03 with some remaining double bonds that were not fully reacted during the process in which the CH2=CHR

was polymerized to form the latex paint. Similarly, Chang and Guo (1998) and Fortmann et al. (1998) report emissions of hexanal during the drying of an alkyd paint; since hexanal was not a component of the paint itself, they proposed that it was formed by the oxidation of unsaturated fatty acid esters in the alkyd resin.

A correlation between indoor ozone and the concentrations of carbonyl compounds and organic acids in homes has been reported in several studies and attributed at least in part to indoor 03 reactions (e.g., Reiss et al., f995a; Zhang and Lioy, f994; Zhang et al., 1994a-c). Not only HCHO and CH3CHO but also larger aldehydes have been measured indoors, with indoor concentrations for all but possibly propionalde-hyde being much larger than those outdoors (Fig. 15.12). The same is true for formic and acetic acids, which can be formed by ozone reactions from reaction of the Criegee biradicals with water vapor (see Chapter 6.E.2). Zhang et al. (f994c) report that indoor formic acid concentrations increased with the indoor concentration of 03 and with relative humidity, as expected if the reaction of the HCHOO Criegee biradical with water vapor was a significant indoor source.

However, as discussed by Reiss et al. (1995a), separating the contribution of ozone reactions from other factors such as temperature and relative humidity, which also affect direct emissions, is difficult. For example, while the production rate of oxygenated organ-ics is correlated with the ozone removal rate, the latter is also correlated with temperature. As a result, both reaction and increased direct emission rates due to higher temperatures may be contributing to these enhanced indoor levels.

Ozone can also react with components found in air ducts. For example, Morrison et al. (f998) reported that the sealant and neoprene gaskets used in the ducts emitted VOCs into the airstream, but at relatively low levels compared to the typical concentrations found indoors. However, reaction with 03 led to increased emissions of aldehydes, particularly the C5-C H) aldehydes.

The mechanism of the reaction of 03 with alkenes was discussed in detail in Chapter 6.E.2. It was seen there that these reactions serve as an indoor source of OH through decomposition of the Criegee biradical. Weschler and Shields (1996b) proposed that the indoor reaction of 03 with alkenes could serve as a source of OH and calculated that at 20 ppb 03 and average indoor alkene concentrations, a steady-state OH concentration of about 2 X 105 molecules cm"3 might be expected.

Subsequently, they measured OH concentrations in an office building using the rate of decay of 1,3,5-tri-

methylbenzene, which reacts with OH but not 03 (Weschler and Shields, 1997a). Although 03 and d-limonene were injected during the experiment, their concentrations were chosen to be similar to that measured under normal operating conditions. An average OH concentration of 7 X 105 molecules cm~3 was obtained, which is about an order of magnitude smaller than daytime peak OH concentrations outdoors but also more than an order of magnitude larger than those outdoors at night (see Chapter 11.A.40.

As discussed in Chapter 9.C.2, some of the larger alkenes such as terpenes form particles containing low-volatility organics on oxidation with ozone. Hence particle formation might be expected indoors in the presence of such compounds, and indeed this has been observed (Weschler and Shields, 1999).

In short, much of the chemistry that has been observed outdoors also occurs indoors. However, the relative importance of various reactions may be somewhat different due to the different absolute and relative concentrations of the reactants, the lower photolysis rates, the exchange of air with outdoor air, and the presence of relatively large surface areas, which can both remove various species and act as substrates for heterogeneous reactions.

0 0

Post a comment