American Journal of Industrial Medicine 1:427-434 (1980)

SO,-Particulate Interactions: Recent Observations

Robert Frank, MD

The toxicology of sulfur oxides is discussed briefly. The effects of sulfur dioxide (SO,) are largely confined to the upper airways except during exercise or if the gas is taken up by a carrier aerosol. SOz may be adsorbed as a monomolecular layer on dry particles, such as elemental carbon, or dissolved in aqueous droplets. Hydrated SO2 forms bisulfite and sulfite ions, which are rapidly oxidized (detoxified) by sulfite oxidase, an enzyme, to form sulfate. SO, in carrier aerosols (dry or aqueous solutions) may be oxidized to sulfuric acid. The mixture of SO2 (1 ppm) and a droplet of sodium chloride (1 mg/m3) has been shown to be synergistic in guinea pigs. In healthy adults, the same gas-aerosol mixture caused no func- tional lung changes at rest (two separate studies), but did cause significant changes following moderate exercise. Shortness of breath and wheezing were experienced by about half of the subjects.

Key words: SO2, particulates, interactions

INTRODUCTION

This paper concerns the interactions between sulfur oxides and other airborne matter that have implications for health. Such reactions have been postulated and tested in clinical and toxicological studies for years, and yet their significance remains largely unknown. This brief review will cover only one type of interaction, namely, the absorption of sulfur dioxide (SOz) by an aqueous droplet. The “irritant” droplet that forms may impair the function of the lung during acute exposure.

A second type of interaction involves the neutralization of sulfuric acid (H2S04) by trace amounts of ammonia (NH,); this interaction would appear to be protective. Neutralization may occur not only in ambient air but also-which is intriguing-within the bronchial airways. There has been considerable interest in whether stable metallic sulfites occur in ambient air and at what levels, especially in relation to smelter activities. Efforts are underway at the University of California at Davis and at New York University

Department of Environmental Health, University of Washington School of Public Health, Seattle, WA 98195 Accepted for publication March 28, 1981

0271-3586/80/0103-040427 $02.50 0 1980 Alan R. Liss, Inc.

428 Frank

TABLE I. Ambient Aerosols: Sulfur Species ~

Reduced states S(-II): hydrogen sulfide

S(0): elemental sulfur

S(1V): bisulfite, sulfite Oxidized states

stable metallic complexes* S(V): dithionate* SWI) stable metallic sulfates

sulfuric acid, ammoniated salts organosulfates

to develop methods for generating these compounds free of contamination from either SOz or sulfate; these sulfites could then be tested biologically. To generate contaminant- free sulfite is technically demanding.

Classes of sulfur-containing compounds that may be found in ambient air are shown in Table I. Most of them occur in association with smelting. The list is probably not complete, and the existence of those sulfur species that are starred is still an open question. Generally, the oxidized sulfur states pose a greater risk to public health than do the relatively reduced states of hydrogen sulfide and elemental sulfur. The focus of this paper will be on sulfur IV and sulfur VI species.

SULFUR DIOXIDE

It is appropriate to begin with some remarks about the toxicology of S02-as a preamble to the discussion of gas-particle interactions.

Two factors appear to be of overriding importance in shaping the toxicological effects of SOz: One is the relatively high solubility of the gas in the liquid lining of the respiratory tract; the other is the ability of the body, once the gas has been absorbed and hydrated, to oxidize the newly formed bisulfite (HSO,) and sulfite (SO,’) ions to sulfate (SO;) ions. The oxidation is essentially a detoxifying process. It occurs rapidly, being driven chiefly by sulfite oxidase, an enzyme found in mitochondria. Sulfite oxidase would appear to be chiefly responsible for confining the effects of SOz to the tissues that are directly exposed, and for curbing the intensity and duration of these effects. Deficiency of this enzyme is rare [Mudd et al, 19671.

Because of the high solubility of SO2 in tissue liquids, individuals breathing quietly by nose are likely to absorb over 95% of the inhaled gas in the airways above the larynx. This has been demonstrated in humans for concentrations ranging from 1 to 25 ppm [Speizer and Frank, 19751, and in dogs for concentrations up to 50 ppm [Frank et al. 19691. As a consequence, exposure of the tracheobronchial tree and parenchyma to SO2 is minimized. (The absorptive sites in the upper airways are themselves subject to damage: for example, SO2 depresses nasal mucous clearance and may possibly reduce local re- sistance to infection [Anderson et al, 19741.)

The penetration of SO2 beyond the upper airways, expressed as the percentage of

SOz Particulate Interactions 429

the inspired concentration that reaches the trachea, may increase under two circumstances, exercise and the introduction of a carrier aerosol.

With exercise, if the ventilatory flow rate rises sufficiently, there is an obligatory shift from chiefly nose breathing to chiefly mouth breathing. In dogs the mouth is less efficient than the nose as a scrubber of SO, at elevated flow rates. Similar findings may be expected in humans (such measurements have not been reported) in whom the oral passage is probably less efficient than the nasal passage as a scrubbing surface, owing to a lower ratio of surface area to volume. Consequently, moderate to vigorous exercise is likely to be associated with significant increments in dose to the lower airways, owing to changes in both rate of penetration by the gas and minute volume. Presumably, the hazard posed to the lower airways increases commensurably.

SO,-PARTICULATE MIXTURES

Carrier aerosols provide the alternative means of increasing the delivery of SO2 and other soluble gases to the periphery of the lung. A schematic of the physical-chemical properties of carrier aerosols is shown in Table 11. Dry particles adsorb a monomolecular layer of gas. If the contour of the particle is irregular or fissured, the adsorptive surface increases proportionally. Because small particles have higher surface-area-to-volume ra- tios than do large particles for a specified mass or volume, they are more effective carriers. Wet particles, whose absorptive capacity is limited by the volume of liquid they can contain, are in turn more efficient carriers than dry particles.

The oxidation of SO, to sulfuric acid (H2S04) may occur in either the dry or wet state, depending on the chemical composition of the particle. The rate of oxidation may be accelerated by elemental carbon [Novakov et al, 19741 or by metallic catalysts.

Hygroscopic particles, of which H2S04 is an example, absorb moisture as a con- tinuous function of relative humidity (RH). The uptake of moisture by deliquescent particles is discontinuous. Thus sodium chloride (NaC1) is a dry crystal at all RHs below about 68%, at which juncture it deliquesces to become a droplet (Fig. 1). The carrier capacity of such particles may be expected to differ significantly for these two physical states, which in turn should influence their imtant potential when mixed with soluble gases.

To test this hypothesis, McJilton and co-workers [1973] exposed groups of lightly anesthetized guinea pigs to one of six experimental modes: SO2 alone, NaCl aerosol alone, or SO2 plus NaCl aerosol, at low (< 40%) and high RH (> 80%); the respective

TABLE 11. Carrier Aerosols

Dry Insoluble: fused aluminosilicates (ash) elemental carbon metallic oxides

deliquescent salts (below deliquescent point) Soluble:

Wet Droplets, liquid films

430 Frank

b ! 0 20 40 a Y)

R ~ ~ I I U humidicy (xi

Fig. 1 . The light-scattering ratio (b& measured with a nephelometer is shown on the ordinate against RH on the abscissa. The ratio is an index of aerosol volume. Deliquescence occurs at an RH of about 68% [McJilton et al, 19761.

4 lb f 0

1 - a

z

I

Fig. 2. P < 0.05 (SO*, high RH) or P < 0.01 (remaining four modes) [McJilton et al, 19731.

The increase in RL in response to S02-NaCI at high RH exceeded changes in all other modes by either

concentrations were 1 ppm and 1 mg/m3 (peak particle count = 0.1 pm), with exposures lasting 1 hour. Respiratory mechanical function was monitored every 5 minutes. The result was that pulmonary flow resistance (RL) increased significantly only when the gas- aerosol mixture was administered at a high RH (Fig. 2).

The amount of SO, estimated to dissolve in the droplet aerosol is quite small, ranging from about 1% at pH 6 down to about O.OOOl% at pH 1 (Fig. 3). This inverse relationship between acidity and the solubility of SOz might be expected to limit inter- actions between the gas and strong acids. Thus, any H2S04 that formed in the droplet would tend to drive SO, out of solution. Amdur, however, has observed empirically that a number of metallic salts potentiate the irritative effects of SO,, as measured by exag- gerated changes in RL [1973, 19741. She has attributed this potentiation to the catalytic

SO2 Particulate Interactions 431

16‘

10-J

,0.6 10-5 lo-+ 10-3 10-2 10-1

I y d r o i o 1- Cannmrr.tiom

(m1.dI.i tar)

Fig. 3 . Larson). HzS04 has pH - 1; NaCI, in equilibrium with ambient CO1, has pH - 6.

The percentage of ambient SOz dissolved in droplet aerosol as function of pH (calculated by Dr. T.

oxidation of the gas to H2S04. Her observation and its implications for health are of sufficient importance to warrant additional efforts at confirmation.

Studies employing identical concentrations of SO2 (1 ppm) and NaCl aerosol (1 mg/m3) have since been carried out on healthy human adults [Morgan et al, 19771 and adolescents with asthma [Koenig et al, 19801.

The healthy adults were exposed for 2 hours. To streamline the procedure, only two experimental modes were used: clean air and the S02-NaC1 aerosol mixture, both at a high RH. The results when the subjects were exposed entirely at rest can be sum- marized briefly: There were no significant changes in pulmonary function measured every 30 minutes compared with baseline values.

Next, intermittent exercise was added as follows: 30 minutes of exposure at rest was followed by 10 minutes of exposure during exercise, repeated twice for a total of 120 minutes. The subjects walked on an inclined treadmill at a speed sufficient to increase minute ventilation five- to sixfold.

Several arresting features of the response are illustrated in Figure 4: Mean flow resistance did not increase after exposure at rest; indeed, all average values at these junctures were below baseline. Because mean flow resistance rose after exercise (relative to baseline), it follows that functional recovery or improvement must have occurred during each subsequent resting period, even though SO1 was being inhaled.

By contrast, mean flow resistance increased following exposure that was combined with exercise. With repeated exercise, however, the magnitude of this change declined progressively, as if some form of “adaptation” were occurring.

Exercise on clean air alone evoked a smaller increase in flow resistance. This effect was no longer statistically significant after the third episode of exercise. Correction for the effect of the exercise itself was necessary in these subjects to solve for the contribution of the S02-aerosol mixture.

Finally, we come to observations on a population generally considered to be un- usually susceptible to inhaled irritants: patients with asthma. Koenig and co-workers have just completed a study on eight adolescents, 14-18 years old, who had extrinsic asthma

432 Frank

30 60 90 120 0 30 60 90 I20 Exposure, mn.

Fig. 4. exercise [Morgan et al, 1977; National Research Council, 19781.

Healthy adults: Effect of SO2 plus NaCl aerosol on pulmonary flow resistance at rest and following

Asthmatic Adolescents

"mx 50 wan. s.e.

n.8

3.01

lab------ 2.5

2.0 -

Fig. 5. value following exercise differs from baseline at P < 0.001 [Koenig et al, 19801.

Asthmatic adolescents: Effect of S G plus NaCl-aerosol on VmaxM at rest and exercise (Ex). Mean

SO2 Particulate Interactions 433

that required medical therapy [ 19801. The subjects received no medication on the mornings of the exposure. Three experimental modes were used in random sequence: clean air, NaCl aerosol, and the SO2-NaC1 aerosol mixture, all at RH = 75%.

After baseline measurements of pulmonary function, the asthmatic subjects were exposed by mouth for 30 minutes at rest and for 10 minutes while exercising on an inclined treadmill that increased minute ventilation sixfold. Pulmonary function meas- urements were obtained after the resting exposure and again after exercise, the latter over a period lasting up to about 20 minutes. The principal measurements were partial flow-volume curves from which maximal flow rates at 50% (Vmaxso) and 75% (Vmax 7 5 ) of expired vital capacity were determined, and total respiratory resistance (RT) meas- ured with an oscillatory pressure method. RT is considered to be dominated by the larger central airways; Vmaxso and V m a ~ ~ ~ reflect principally the caliber of the peripheral airways.

As with the healthy adults, there were no significant changes in pulmonary function following exposure at rest (Figs. 5 and 6) in any mode. (In an earlier study carried out on nearly the same group of subjects, exposure to the gas-aerosol mixture at rest for 1 hour was associated with only slight but statistically significant function changes that were consistent with a narrowing of the peripheral airways [Koenig et al, 19801 .) However, after the exposure to S02-NaCI aerosol during exercise, Vmaxso (Fig. 5) and V m a ~ ~ ~ (not shown) fell sharply. The maximal reductions were 44% and 50%. RT increased by 67% (Fig. 6).

Three subjects complained of shortness of breath, and five experienced wheezing. The changes after exposure to the other two modes, clean air and NaCl aerosol, alone were random.

kthmtlc Adolescents

F+ 3 H t

mann, 5 . e .

n-8

5.0

4.0

3.0

; 2.0 I - 2 a F I

O r i Q i UfMlt.l

Fig. 6. exercise. Mean value following exercise differs from baseline at P < 0.01 [Koenig et al, 19801.

Asthmatic adolescents: Effect of SO2 plus NaC1-aerosol on RT measured at 3 Hz at rest and following

434 Frank

Until the effect of SO2 alone on these subjects is determined, a sfep now underway, it remains uncertain whether these functional and clinical changes are attributable to the irritant aerosol, the gas, or the combination. If SOz alone in these subjects is also found to cause functional abnormalities, it will be of interest to compare the distribution of such changes within the airways (central versus peripheral) with those associated with the gas- aerosol mixture. (That low concentrations of SO2 in the absence of a droplet aerosol and combined with exercise may evoke bronchcconstriction in asthmatic subjects has recently been shown by Sheppard et a1 [1980]. Specific airway resistance (SRaw) more than doubled in response to 0.5 pprn of SO2 combined with exercise; the response was less but still statistically significant at 0.25 ppm. Exercise alone and 0.5 ppm of SO2 at rest had no effect.)

With respect to short-term exposures, these findings support the hypothesis that individuals with asthma, who presumably have hyper-reactive airways, are more sus- ceptible to sulfur oxides than is the general population. This increased susceptibility is demonstrated conclusively if exercise and its associated hyperpnea are imposed on the exposure.

ACKNOWLEDGMENTS

This research was supported in part by the U.S. Environmental Protection Agency (grant R-805378010) and the National Institute of Environmental Health Sciences (grant 5-POIES01478)03.

Amdur MO (1973): Animal studies. In “Proceedings of the Conference on Health Effects of Air Pollutants, Washington, DC, October 3-5, 1973. Report prepared for the Committee on Public Works, United States Senate (Serial No. 93-15). Washington DC: U.S. Government Printing Office, pp 175-205.

Amdur MO (1974): The long road from Donova. 1974 Cummings Memorial Lecture. J Am Ind Hyg Assoc 35589-597.

Anderson I, Lundquist GR, Jensen PL, Proctor DF (1974): Human response to controlled levels of sulfur dioxide. Arch Environ Health 28:31-39.

Frank NR, Yoder RE, Brian JD, Yokoyama E (1969): SO1 (%labeled) absorption by the nose and mouth under conditions of varying concentrations and flow. Arch Environ Health 18:315-322.

Koenig JQ, Pierson WE, Frank R (1980): Acute effects of inhaled SO2 plus NaCl droplet aerosol on pulmonary function in asthematic adolescents. Environ Res 22: 145-153.

Koenig JQ, Pierson WE, Horike M, Frank R: Effects of SO2 plus NaCl aerosol combined with moderate exercise on pulmonary function in asthmatic adolescents. Submitted to Environ Res.

McJilton CE, Frank R, Charlson R (1973): Role of relative humidity in the synergistic effect of a sulfur dioxide- aerosol mixture on the lung. Science 182503-504.

McJilton CE, Frank R, Charlson RJ (1976): Influence of relative humidity on functional effects of an inhaled SO*-aerosol mixture. Am Rev Resp Dis 113:163-169.

Morgan MS, Koenig JQ, Horike N, Covert DS, Frank R (1977): Acute effects of inhaled SO2 combined with hygroscopic aerosol in healthy man. Am Rev Resp Dis 155(4) Part 2:231 (abstract).

Mudd SH, Irreverre F, Laster L (1967): Sulfite oxidise deficiency in man: Demonstration of the enzymatic defect. Science 156: 1599-1602.

National Research Council (1978): Sulfur oxides. In “National Academy of Sciences.” Chapter 7. Washington DC: National Research Council, pp 13Cb179.

Novakov T, Chang SG, Harker AB (1974): Sulfates as pollution particulates: Catalytic formation on carbon (soot) particles. Science 186:259-261.

Sheppard D, Nadel JA, Boushey HA (1980): Exercise increases sulfur dioxide-induced bronchcconstriction in subjects with asthma. Physiologist 23(4):3 (abstract).

Speizer FE, Frank NR (1975): The uptake and release of SOz by the human nose. Arch Environ Health 12725-728.