NIOSH Hazard Review: Carbonless Copy Paper/Exposure

by Centers for Disease Control and Prevention

3 Exposure

3.1 Introduction

Workers may be exposed to CCP or its components during handling or manufacturing. This chapter summarizes exposures from CCP handling reported in the literature (Table 3–1). Little consistency exists among these reports: they vary depending on the chemical composition of the CCP, the method of manufacturing during the study period, and the number of forms handled during the industrial hygiene survey. For workers who handle CCP, the most common exposures are to formaldehyde and kerosene or its components. Formaldehyde is used in some microcapsule manufacturing processes as part of the mixture that forms the shell for the microcapsules; it is also used in the manufacture of other paper products such as plain bond paper. Kerosene is one of the principal solvents used to solubilize the precursor dyes contained in the microcapsules.

3.2 Exposure Data

3.2.1 Published Studies

Some studies listed in this section are described in another section of this review.

Jujo Paper Company, Ltd. 1979. One of the earliest reports with CCP exposure data came from the Jujo Paper Company, Ltd. [1979]. They reported that the maximum concentration of CCP solvent (unspecified) in a finishing room where 100 tons of CCP were handled each day was 0.3 mg/m³. The company also reported an average CCP solvent retention of 180 μg on the fingers of women who handle, sort, and count 50,000 to 70,000 sheets each day. Blood samples obtained 15 to 16 hr after work revealed no detectable concentrations of the solvent. Further biochemical tests of the blood and urine, skin tests (types unspecified), examinations, and interviews of 135 exposed workers and 84 comparison workers revealed no differences between the two groups. The company indicated that no skin disorders had been reported by any worker since the CCP mill came into operation. No independent survey of worker complaints was performed.

Mølhave and Grunnet 1981. In an addendum to the telephone company report by Menné et al. [1981], Mølhave and Grunnet [1981] reported on a headspace analysis (sampling of the gaseous phase of a sample heated to 50 °C) of the CCP in use at the time of the study. They used one paper sample received from the factory where the problem was investigated and one sample from the manufacturer of the paper. The authors reported that more than 42 chemicals degassed from the paper samples, and concentrations were seven times greater in the paper from the facility than in those of the manufacturer's sample. About 90% of the emission was alkanes or alkenes (C₅–C₁₄). Another analysis was performed on the Santosol oil content of both CCP samples. The CCP that (according to the authors) caused the original skin problems contained up to 150 times the amount of Santosol oil contained by the manufacturer's sample of CCP. According to the authors, the Santosol oil consists

Table 3–1. Exposures from CCP handling

Reference and country		Occupation or exposure scenario		Sample type		Airborne concentration
Jujo Paper Company, Ltd. [1979], Japan		Paper finishing		CCP solvent (unspecified)		0.3 mg/m3
Mølhave and Grunnet [1981], Denmark	Experimental laboratory conditions	Alkanes or alkenes (C5–C14)
		Santosol	Levels not reported
Göthe et al. [1981] and Norbäck et al. [1983b], Sweden	Printing offices	Kerosene	7.0 mg/m3
		MIPB[1]	0.2 mg/m3
		Diarylethane	0.2 mg/m3
		Hydrogenated terphenyl and diisopropylnaphthalene	<0.01 mg/m3
	Ordinary offices	Kerosene	0.7 mg/m3
		MIPB	0.2 mg/m3
		Diarylethane	0.02 mg/m3
		Hydrogenated terphenyl and diisopropylnaphthalene	<0.01 mg/m3
Gockel et al. [1981], United States	Office workers	Formaldehyde	<0.51 ppm
Chrostek and Moshell [1982], United States	Telephone workers	Total dust	0.06–0.2 mg/m3
		Formaldehyde	0.22 mg/m3[2]
		Glove analyses:
		Dibutyl phthalate	Detected
		Diethyl phthalate	Detected
		Dioctyl adipate	Detected
Norbäck [1983b], Sweden	Experimental laboratory conditions	Kerosene	0.35–15.5 mg/m2 per hr
		MIPB	0.33–0.54 mg/m2 per hr
		Formaldehyde	0.1–0.3 mg/m3
Norbäck and Göthe [1983], Sweden	Offices and print shops	Total dust	0.02–0.05 mg/m3
		Kerosene	0.7–0.81 mg/m3
		MIBP	0.06 mg/m3
		Diarylethane	0.03 mg/m3
		Hydrogenated terphenyls	<0.01 mg/m3
Winfield [1983], United States	Purchasing office	Formaldehyde	ND—0.04 ppm
Hazelton Laboratories [1985], United States	Experimental laboratory conditions	Formaldehyde	0.033 ppm for separating 30 4-ply forms/hr for 8 hr
Olsen and Mørck [1985], Denmark	Office workers	Total dust	0.11–0.21 mg/m3
		Kerosene	1.9 mg/m3
		Hydrogenated terphenyls	ND
Apol and Thoburn [1986], United States	CCP production	HMDI	<0.7–14.0 μg/m3
		DETA	<0.01–<0.35 ppm
		Phenol	<0.02–0.15 ppm
		Formaldehyde	ND[3]
		Biphenyl	0.003–<0.02 ppm
		Butyl biphenyl	0.12–0.29 ppm
		Petroleum solvents	0.7–12 mg/m3
		Total particulate	2.70 mg/m3
Chovil et al. [1986], United States	University office	Formaldehyde	0.015–0.022 ppm
Burton and Malkin [1993], United States	Printing shop	Isopropanol	53–132 ppm
		Isobutanol	0.15–0.91 ppm
		1,1,1-trichloroethane	0.11–0.23 ppm
		Toluene	1.09–5.03 ppm
		Beryllium, calcium, copper, iron, magnesium, and zinc	0.02–1.05 μg/m3
Omland et al. [1993], Denmark	Office workers	Formaldeyde	0.1–0.62 mg/m3
		Total dust	0.28–0.34 mg/m3
Zimmer and Hadwen [1993], United States	Federal records storage center	Acetic acid	<25 mg/m3 (REL)
		Cyclohexene	<1,050 mg/m3 (REL)
		Formaldehyde	0.023–0.034 mg/m3
Thompson [1996], United States	Office workers	Decane	1.0–1.1 ppb
		Undecane	0.3 ppb
		Dodecane	0.6 ppb
		meta-, para-Xylene	0.6–1.2 ppb
		ortho-Xylene	0.2–0.4 ppb
		Toluene	0.5–1.3 ppb
		Ethyl benzene	0.3–0.5 ppb

1. Abbreviations: DETA=diethylene diamine tetracetic acid; HMDI=hexamethylene diisoeyanate; MIPB=monoisopropyl biphenyl; ND=none detected; REL=NIOSH recommended exposure limit.
2. Attributed to cigarette smoking.
3. Limits of detection varied from 0.04 to 0.08 ppm.

mainly of hydrogenated terphenyl,^[1] which is known to produce eye, skin, and respiratory irritation and possibly sensitization in experimental animals [Haley et al. 1959]. At the telephone company that reported the problem, workers exposed to CCP dust and vapors emitted from the paper experienced marked irritation at air concentrations exceeding 10 mg/m³ (data not given). Molhave and Grunnet [1981] believe that the terphenyls act as primary irritants, particularly when workers are wearing protective gloves that trap moisture and exposures next to the skin.

Göthe et al. 1981 and Norbiick et al. 1983b. Gothe et al. [1981] and Norback et al. [1983b] reported on an investigation of climatic and airborne concentrations of microcapsule solvents found in printing offices and ordinary offices that used the same type of CCP. The authors noted that very few complaints were related to CCP in the printing offices compared with ordinary offices. Temperature and relative humidity were, on the average, about the same in the two environments. The highest concentrations of microcapsule solvents were observed in the printing offices (Table 3-1). This finding suggests that no simple correlation exists between solvent vapor concentrations and the occurrence of complaints; or it may indicate that skin contact is the important factor.

Gockel et al. 1981. Gockel et al. [1981] reported on formaldehyde released to the air from CCP forms that were suspected of causing eye, skin, and respiratory irritation among office workers. Water extraction of CB white sheets of CCP yielded 0.18 to 1.89 mg formaldehyde per 8.5-×11-in. top sheet of CCP. The authors felt that water extraction might have enhanced the formaldehyde concentrations, so they adopted a sampling procedure that collected the formaldehyde released into 15 L of air (1 L/min for 15 min). Formaldehyde concentrations ranged from 33.6 to 858 μg/kg of forms sampled and from 0.02 to 0.96 ppm in the 15-L air samples using 8 different CCP forms. A modification of the procedure ensured adequate air flow past all parts of each form in the sampling apparatus. Standardized testing of four sheets of equivalent area for each type of five different forms resulted in formaldehyde concentrations ranging from 0.45 to 16.8 μg/kg, demonstrating a 37-fold difference in formaldehyde emissions. The authors demonstrated that these air sample analyses using a standardized testing area produced results that varied by a factor of 0.83 to 1.42 compared with sampling of a full form. The authors also provided evidence that the residual formaldehyde is dissipated into the air as a result of handling and storage. Air concentrations of formaldehyde were as high as 0.51 ppm in filing cabinet drawers where the forms had been separated and stored for more than 6 months.

Chrostek and Moshell 1982. See Section 4.2.1 for a description of this study.

Norbäck 1983b. Norback [1983b] studied the chemical emissions from entirely unused paper and from paper in which approximately 1% of the microcapsules had been crushed by standard writing. Most of the CCPs studied were handled by workers who had experienced work-related respiratory irritation symptoms when handling CCP. In light of the observed emissions of formaldehyde from CCP over time [Gockel et al. 1981], the 1- to 2-year-old paper was replaced with fresh CCPs of various types collected from three different printing shops. Most measurements were performed at an ambient temperature of 22 °C and 20% to 30% relative humidity. Several tests were also performed at 27 °C. CCP was cut, weighed, and measured for surface area. It was then placed into wash bottles (0.25-L), and air was passed through them at the rate of 0.1 L/min. Charcoal (for solvent analysis) or Amberlite XAD (coated with 2,4-dinitrophenyl hydrazine for aldehyde analyses) was used to collect the emissions for 30 to 60 min. Solvent concentrations were measured using gas chromatography (GC), and aldehyde analyses were performed with liquid chromatography. The relative solvent emissions calculated were based on measurement times, surface area, and amounts of solvent/aldehyde released. Mann-Whitney's rank sum test was used for testing the statistical significance of paired t-values. Norback [l983b] found small but measurable amounts of formaldehyde (0.1 to 0.3 mg/m³; detection level=0.3 mg/kg per hr) in 3 of 4 fresh CCP samples. No glutaraldehyde was detected (detection level=0.l mg/kg per hr). No aldehyde emissions were detected from any of the papers that were 1 to 2 years old. One week after the microcapsules had been crushed, four of the five solvents studied were still being released in measurable quantities, including monoisopropyl biphenyl (MIPB), kerosene, phenylxylylethane, and diisopropylbiphenyl—but not hydrogenated terphenyl. The kerosene emissions ranged from 5 to 60 mg/m3, with the two CCP samples not linked to work-related respiratory tract symptoms yielding the lowest kerosene emissions. On the basis of this observation, the author tested three different groups of kerosene-containing CCP, some of which had observed links with work-related respiratory tract symptoms. He demonstrated in this study that no links existed between mucous membrane symptoms and kerosene emissions. He also showed that there were no statistically demonstrable trends toward a link between work-related respiratory tract symptoms and high kerosene emissions—even where all CCPs associated with respiratory symptoms were combined, and regardless of the solvent content. This difference was attributed to the difference in encapsulation processes (MIPB used "poly- mer," and hydrogenated terphenyl used gelatin). The author noted that the kerosene concentrations in the wash bottles were 10 to 100 times higher than those measured in the breathing zones of workers involved in intensive manual handling of CCP. The author also concluded that aldehyde emissions from CCP were not likely to explain the irritative mucous membrane symptoms among workers who handle such paper. Table 3-2 demonstrates how writing on CCP (and thereby crushing the microcapsules) affects the solvent emissions from the paper.

Norbäck and Göthe 1983. In a Swedish study, Norback and Gothe [1983] collected personal and area samples in Stockholm at ll offices where large quantities of CCP were handled and at five printing shops where form (manifold) sets of CCP were produced. The measurements were made from January 1980 to November 1981, mainly during the winter half of the year (the period in which problems

Table 3-2. Solvent emissions from CCP with intact microcapules and 1% crushed microcapsules (mg/m2 per hour)

CCP form treatment	Kerosene emissions	MIPB emissions
Unused	10.35	0.33
Crushed (fresh writing)	15.50	0.54
Week—old writing	13.70	0.24

Source: Norbäck [1983b].

were reported). Workers from 10 of the offices studied had been referred to the clinic of Occupational Medicine at Southern Hospital because of health problems associated with handling CCP. The authors measured ambient temperatures, relative humidities, and ventilation efficiencies. As a measure of the chemical emissions from CCP, airborne concentrations of the solvents from microcapsules were analyzed using activated charcoal tubes. The carbon-disulfide-desorbed solvents were analyzed by GC, and detection limits varied between 0.001 and 0.02 mg/m³.

Airborne concentrations of total dust, dust-bound solvent, solvent in the vapor phase, and formaldehyde were also determined in a laboratory situation using a 34-m³ room with an air-exchange rate of 0.8 times/hr. Thirty sheets of each type of paper were handled in a standardized procedure for 60 min. Table 3-3 shows that airborne solvent concentrations are generally low, and they are considerably lower in the office environment than in the printing shop. Area samples were also consistently lower than personal samples, suggesting that manual handling generates airborne solvent. For example, kerosene (which is relatively volatile) had the highest airborne concentrations, whereas the hydrogenated terphenyls (whose volatility is low) produced unmeasurable concentrations. The data indicate that various paper types generated similar concentrations of dust during standardized paper handling in the laboratory.

The airborne formaldehyde concentrations were below the limit of detection (<0.05 mg/m³). This finding does not support formaldehyde as the cause of the health effects. The particle-bound solvents were also consistently below the detection limit (<0.0002 mg/m3), which corresponds to a dust solvent content of less than 1% by weight. Norback and Gothe [1983] concluded that no obvious climatic differences were evident between the two environments, even though health problems occurred in the offices and not in the printing shops. The author observed that these health problems occur in offices with both high and low levels of

Table 3-3. Airborne concentrations of total dust and solvents produced with standardized paper handling in the laboratory (mg/m³)

		Solvent
Paper type	Total dust	Kerosene	MIBP	Diarylethane	Hydrated terphenyl
Paper containing MIBP + kerosene	0.05	0.81	0.06	—[2]	—
Paper containing diarylethane	0.02	—	—	0.03	—
Paper containing hydrogenated terphenyl + kerosene	0.05	0.70	—	—	<0.01
Ordinary paper	0.05	—	—	—	—
Control—without paper	0.02	—	—	—	—

Source: Norbäck and Göthe [1983].

1. This description differs from other descriptions of Santosol components, which do not refer to terphenyls but to diphenylmethanes.
2. Dash indicates that no measurement was performed.

ventilation. The solvent concentrations were relatively higher in printing shops than in offices, but the number of health problems in the printing shops was low. The authors cited a study by Hasegawa et al. [1973] that found a diisopropylnaphthalene concentration of 0.3 mg/m³ in the air at a sorting department in which each worker daily handled 50,000 to 70,000 CCP sheets containing the solvent. They also cited an unpublished report by Dodds [1980] who found hydrated terphenyl concentrations in the ppb range during the production of microcapsules containing color former dissolved in hydrogenated terphenyls. Norbäck and Göthe [1983] conclude that the measured dust concentrations did not contain solvents in sufficient quantity to be associated with primary irritation. This study is unclear as to whether encapsulated CCP solvent attached to airborne fibers is extractable by carbon disulfide and is thus included in measurements of dust-bound solvent. This study did not consider the effect of high local concentrations of solvent on the epidermis when a microcapsule fractures. Also unresolved are the relative skin exposures for workers in offices and printing plants. Although printing plant workers process a far greater tonnage of paper than office workers, its not clear whether printing plant

workers have more or even as much skin contact as CCP users in offices.

Olsen and Mørck 1985. Olsen and Mørck [1985] extensively studied a brand of CCP that was dominant in the Scandinavian countries at that time. They performed gas chromatography/mass spectrometry (GC/MS) analysis, finger analysis of the residual CCP components on the skin, analysis of keyboard surfaces of computers and typewriters, microbiological analysis of the microcapsules, analysis of the mucous membranes of the nose, electron microscope studies of the skin using tape before and after handling CCP, dust measurements, and headspace analysis of CCP emissions. The authors found that hydrogenated terphenyls are transferred to the skin (120 μg per sorting finger) along with their impurities of bi-, tetra-, and pentaphenyls, but they did not find kerosene in detectable amounts owing to its volatile nature. The ratio of hydrogenated terphenyls to kerosene in the microcapsules was 1:3; but after rupture, analysis of the CF layer revealed that more than half of the kerosene had evaporated. Analysis of exposed workers' mucous membrane secretions failed to reveal any CCP components. Headspace analysis demonstrated that kerosene evaporated from the CCP without mechanical rupture of the microcapsules (value not given). The amount evaporating increased after rupture (the highest concentration found in room air was 1.9 mg/m³ [0.3 ppm]), but hydrogenated terphenyls were not released into the air as vapor. Analysis of keyboards revealed concentrations of hydrogenated terphenyls and transfer of this compound to telephones, table tops, etc. in the office. Measurements of total dust ranged from 0.11 to 0.21 mg/m³, and no chemical components of the CCP were associated with it. No growth of fungi or bacteria resulted from the incubation of microcapsules, but one base paper sample (not CCP) supported the growth of actinomycetes at 50 °C. Electron microscopy did not show transfer of the clay/kaolin components to the hands after 3 hr of handling CCP.

Apol and Thoburn 1986, Chovil et al. 1986, and Burton and Malkin 1993. See Section 4.2.1 for a discussion of these studies.

Omland et al. 1993. See Section 4.2.3.2 for a discussion of this study.

Zimmer and Hadwen 1993. In response to a request from the management of the Federal Records Center in Dayton, Ohio, Zimmer and Hadwen [1993] investigated six worker complaints of an overpowering, irritating odor in the archives area where Federal tax records and X-ray films were stored. Acetic acid was the apparent source. Concentrations of acetic acid and cyclohexane were below the NIOSH RELs of 25 and 1,050 mg/m³, respectively. Formaldehyde concentrations were 0.023, 0.024, and exceeding the NIOSH REL of0.02 mg/m³. The most likely source of the formaldehyde was the CCP records located throughout the center.

Thompson 1996. Thompson [1996] reported measurements of indoor air quality in an unpublished U.S. study of 75 workers who continuously handled CCP in the finance and accounting building of a university. This building had a history of indoor air quality problems and medical complaints from workers dating from 1992. The relative humidity, temperature, and mold and fungus counts were within the American Society of Heating, Refrigerating, and Air-Conditioning (ASHRAE) limits of 40% to 60% relative humidity, 74 to 78 °F temperature, and low indoor spore counts (relative to outside counts for mold and fungus) [ASHRAE 1981]. The carbon dioxide concentration was 1,000 ppm, which exceeded the

ASHRAE standard [ASHRAE 1989]. Corrective actions to the ventilation system included repair of the heating, ventilating, and air-conditioning system, increased outside fresh air supply, earlier air-handling startup times, and increased air circulation (which decreased carbon dioxide concentrations to 400 to 700 ppm).

Area air samples were collected in two locations of the CCP building and compared with an air sample from another building that had 60 occupants, no history of medical complaints, and minimal use of CCP. GC/MS standards were prepared from the CCP forms (all three sheets, top sheets alone, and bottom sheets alone) and a sample of SurSol 290 (a solvent carrier for dyes used in the production of the microcapsules). Table 3-4 lists the con- centrations of chemicals found in these samples. The sample from the comparison building showed concentrations that were about three orders of magnitude less than those found in the CCP building. Of the chemicals for which

Table 3-4. GC/MS analysis of CCP samples, SurSol 290 solvent, and area air samples (ppb)

					Area air samples
	CCP samples				CCP building
Chemical measured	Top sheet	Bottom sheet	Three sheets	SurSol 290 solvent	Sample 1	Sample 2	Comparison building
Decane	13.2	0.001	7.9	—	1.0	1.1	0
Undecane	0.02	<0.001	12.7	—	0.3	0.3	0.004
Dodecane	2.6	<0.001	4.8	0.027	0.6	0.6	0.006
meta-, para- Xylene	0.1	0	0	6.2	1.2	0.6	0
ortho-Xylene	0.0	0	0	—	0.4	0.2	0
Toluene	2.9	0.3	0.1	—	1.3	0.5	<0.001
Ethyl benzene	0.4	0	0	8.9	0.5	0.3	0

Adapted from Thompson [1996].

occupational safety and health standards exist, the concentrations were four to six orders of magnitude lower than the standards.

3.2.2 NIOSH Docket Submissions

3.2.2.1 Winfield 1983

Winfield [1983] performed an industrial hygiene survey in a purchasing office at the University of Texas in response to worker complaints of headaches, skin eruptions, upper airways irritation, and other symptoms. The number of workers who reported symptoms was not given, but the report stated that the incidence of symptoms was higher among the 22 employed in the purchasing section than in the 16 employed in the vouchers section. Several former employees reported that their symptoms ceased when they terminated employment. Formaldehyde was measured inside a closed cabinet containing CCP forms, and the level was found to be just above the limit of detection. Other measurements were obtained for hydrocarbons linked to toner solvent from a copy machine and for chlorinated solvents linked with correction fluid, waxes, glues, etc. Interviews were conducted at four other offices where workers handled CCP forms. Workers reported no symptoms in the press office, where forms were handled for printing and gluing. In the personnel office, where forms were handled but not typed on, one worker reported transient skin irritation. Two of four workers in the mail and supply office reported skin irritation. In the central receiving office, two workers complained of odor and headaches when using continuous-roll copy paper; however, the report did not note whether this paper was CCP. Alterations in the air-handling system (which were engineered to exceed the minimum rate for office spaces) did not reduce the reported symptoms. The author stated that the reported symptoms were probably caused by CCP based on the available scientific literature, but she offered no definitive scientific evidence in support of this conclusion. Recommendations from the available literature were suggested to improve the comfort and health of the workers, but no followup survey was reported.

3.2.2.2 Hazelton Laboratories 1985

A NIOSH docket submission by Hazelton Laboratories [1985] (Final Report, March 11, 1985: A Study to Determine the Potential Emanation of Formaldehyde Vapor from Carbonless Copy Paper) describes an investigation performed for a member of the U.S. CCP industry to determine the potential emanation of formaldehyde vapor from CCP.

The experiments were performed in a glove box to measure the following: (1) the maximum formaldehyde air concentration (collected with impinger and measured using NIOSH Method 125 [NIOSH 1994]) produced by a set number of sheets of CCP and (2) the effects of marking and separating four-ply CCP forms on the emission of formaldehyde. The experiments also evaluated the effects of ventilation on the formaldehyde concentrations from various types of CCP. The formaldehyde concentration in the glove box ranged from 0 to 0.7 ppm for the CF and the "self-contained black" paper, respectively. Those products containing black ink produced substantially higher formaldehyde concentrations than those containing blue ink. A model was developed from the kinetic experiments to predict air concentrations of formaldehyde attributable to handling CCP in the office environment.

Product test methods. An aluminum pouch containing the papers was placed in the 285-L chamber for testing. Table 3-5 presents the data for turning 2 or 6 sheets/ min using a varying number of total sheets turned. Chamber concentration of formaldehyde increased as

Table 3-5. Formaldehyde concentration after repeated turning of CCP sheets in a test chamber[1]
Rate of turning and total number of sheets in chamber	Average formaldehyde concentration (ppm)
2 sheets/min:
24	0.089
48	0.165
72	0.171
6 sheets/min:
72	0.081
144	0.102
216	0.212
288	0.501

Adapted from Hazelton Laboratories [1985].

1. The indicated number of sheets placed in the chamber, turned at the stated rate, and repackaged. Air samples were then collected from the chamber.

the number of exposed sheets increased. Another test (Table 3-6) was performed to determine whether the rate of turning would affect the final concentration of formaldehyde in the test chamber immediately after turning and 60 and 90 min after turning. When measured immediately after turning, concentrations decreased as the turning rate increased. But concentrations varied little when measured 60 and 90 min after the tests. This result indicates that the rate-limiting factor for total formaldehyde released from CCP is the amount of time spent equilibrating with the environment.

The final test method evaluated was an emission rate study. In this study, 120 sheets of paper were placed in the chamber and turned at a rate of 4 sheets/min. Short-interval sampling began with the initiation of the page turning and continued for 90 min (Table 3-7). The chamber air achieved a constant formaldehyde concentration in less than 30 min. The initial rate of formaldehyde release was 0.098 Mg/sheet per min. This rate was calculated from the first sample by considering the 0.310-μg/L concentration as the midpoint concentration between 0 μg/L and equilibrium, and by assuming an approximately linear increase in the airborne concentration of formaldehyde over the 15-min sampling period.

Product testing. Three replicate sets of eight types of CCP were tested by placing 60 sheets of CCP in the chamber and turning them at a rate of 4 sheets/min. They remained stacked in the chamber for 15 min and were then returned to the foil packages for the duration of the air sampling, which was conducted for 20 min at a rate of approximately 0.5 L/min. The airborne concentrations of formaldehyde in the test chamber averaged from 0.009 to 0.693 ppm (Table 3-8). Little formaldehyde would be expected from the CF since it contains no microcapsules. All types of the CB and CFB with black ink produced higher average formaldehyde concentrations than did the blue ink counterpart. The self-contained samples yielded the highest formaldehyde

Table 3-6. Formaldehyde concentrations in the test chamber at Various points after turning (total of 60 sheets for each condition)

	Average formaldehyde concentration (ppm)
Turning rate (sheets/min)[1]	Immediately after turning[2]	60 min after turning[3]	90 min after turning[4]
6	0.150	0.307	0.456
5	0.184	0.318	—
4	0.210	0.308	0.387
3	0.316	0.318	0.386
2	0.334	0.367	0.429
0	—	0.260	0.427

Adapted from Hazelton Laboratories [1985].

1. Sixty sheets were placed in the chamber and turned at the indicated rate.
2. After turning was completed, the sheets were repackaged in the foil pouch and an air sample was collected.
3. Sixty minutes from the start of the turning, the sheets were repackaged in the foil pouch and an air sample was collected.
4. Ninety minutes from the start of the turning, the sheets were repackaged in the foil pouch and an air sample was collected.

Table 3-7. Formaldehyde concentrations in the test chamber during a 90-min period^[1]

Sampling interval (min)	Formaldehyde concentration (ppm)
0—15	0.252
7.5—22.5	0.317
20.8—30.5	0.413
24—39.5	0.454
32—46	0.420
41—56.5	0.456
48.5—64	0.406
60.5—75	0.479
66—84.5	0.415
76.5—91	0.441

Adapted from Hazelton Laboratories [1985].

1. One hundred twenty sheets were placed in the chamber and turned at 4 sheets/min. Air sampling began when turning began and continued for 60 min after turning was completed.

Table 3–8. Formaldehyde concentrations in a test chamber containing eight CCP products^[1]

Product	Average formaldehyde concentration for 3 replicates (ppm)
CF	0.009
CFB-blue	0.108
CFB-black	0.209
CB-blue	0.258
CB-black	0.291
SC[2]-blue	0.355
SC-black	0.693
Four-part form[3]	0.178

Adapted from Hazelton Laboratories [1985].

1. Sixty sheets were placed in the chamber, turned at the rate of 4 sheets/min, and left stacked for 15 min before they were repackaged in aluminum foil pouches. Air samples were then collected.
2. SC=self-contained.
3. The four-part form consisted of a CB sheet, two CBF sheets, and a CF sheet.

concentrations. The study director stated that the total formaldehyde release for the four-part form could be predicted from the sum of its parts.

Office activities. Experiments were performed to examine the effects of office activities on formaldehyde emissions from CCP. Four-ply CCP forms were manipulated by marking, separating marked forms, and separating unmarked forms.

• Marking forms: Four-ply forms were used to examine the effects of marking on the emission of formaldehyde vapor. Thirty forms (120 sheets) were placed inside the chamber for each test. A template was used to achieve consistent pencil lines. The desired rate of marking was 40 lines/minute, 20 lines/form, repeated four times throughout the 1-hr sampling period. This rate was achieved on the second test; the first test averaged a rate of approximately 28.7 lines/min. The sampling flow rate was approximately 0.5 L/min. The maximum average formaldehyde concentration for two replicates was 0.402 ppm after 1 hr.

• Marking and separating forms: Four-ply forms were used to examine the effects of marking and separating pages on the emission of formaldehyde vapor. Thirty forms (120 sheets) were placed inside the chamber for each test and a template was used to achieve consistent pencil lines. Each form was marked with 20 lines and separated in 1 min. After the 30-min marking and separating period, the forms were left exposed in the chamber the rest of the 1-hr sampling process.

The sampling rate was approximately 0.5 L/min. The maximum average formaldehyde concentration for 2 replicates was 0.402 ppm after 1 hr.

• Separating unmarked forms: Four-ply forms were used to examine the effects of separating unmarked forms on the emission of formaldehyde vapor. Thirty forms (120 sheets) were placed inside the chamber for each test. Sheets were separated at the rate of 1 form or 4 sheets/min. After the 30-min separating procedure, the sheets were left exposed in the chamber for the rest of the 1-hr sampling process. The sampling flow rate was approximately 0.5 L/min. The maximum average formaldehyde concentration for 3 replicates was 0.37 ppm after 1 hr.
• Ventilation studies: Four types of paper were used to examine the effect of ventilation on the concentration of formaldehyde in the chamber air. For each test, 120 sheets of paper were placed inside the chamber. Page turning, ventilation, and sampling all began at time zero. Pages were turned at the rate of 4 sheets/min for 30 min and were left exposed in the chamber for the final 30 mins. Ventilation and sampling were continuous for the full hr. Ventilation was simulated by forcing compressed air into the chamber and allowing the air to flow out through a hole in the rear of the chamber. The ventilation rate was approximately 0.5 air change/hr for CB–15 blue, CB–15 black, and self-contained-17 black. This rate was obtained by using a flow rate of 2.6 to 2.9 L/min. Ventilation for SC–14 black was approximately 1 air change/hr, obtained by using a flow rate of 5.1 L/min. The sampling flow rate was approximately 0.5 L/min. The results are shown in Table 3–9.

The release of formaldehyde for the combined marking and separating activity demonstrated a value between the maximum concentrations for either activity measured alone. The maximum average formaldehyde concentration was 0.402 ppm after 1 hr for 2 replicates for marking and separating. Marking the forms (maximum average formaldehyde concentration was 0.497 ppm after 1 hr for two replicates) had a greater impact on the release of formaldehyde than did separating them (maximum average formaldehyde concentration was 0.37 ppm after 1 hr for 3 replicates). The results permitted the investigator to develop a formula for predicting formaldehyde release in the office environment. Using the rate constants developed (the assumptions and calculations used were not provided), the investigator predicted a formaldehyde concentration of 0.033 ppm for a worker confined to a 1,000 ft3 room with no ventilation while marking and separating 30 four-ply forms/hr for 8 hr. This value is between the NIOSH recommended exposure limit (REL) of 0.016 ppm as an 8-hr time-weighted average (TWA) (with a 15-min ceiling limit of 0.1 ppm) and the OSHA permissible exposure limit (PEL) of 0.75 ppm as an 8-hr TWA (with a 2-ppm short-term exposure limit [STEL]).

3.3 Conclusions

Little consistency has been found in the literature when various investigators elected to perform air sampling analyses to assess potential exposure to CCP and its components as summarized in Table 3–1. The most frequently chosen analyte was formaldehyde. Of the seven studies reporting formaldehyde concentrations (summarized in Table 3–10), nearly all measurements exceeded the NIOSH REL of 0.016 ppm as an 8-hr TWA with a 15-min ceiling limit of 0.1 ppm [NIOSH 1981]; however, none exceeded the OSHA PEL of 0.75 ppm as an 8-hr TWA with a short-term exposure limit of 2 ppm [29 CFR 1910.1048]. Short-term exposures to this strong-smelling gas cause eye, nose, and throat irritation in some persons at concentrations of <1 ppm. At 5 to 30 ppm, formaldehyde causes cough, chest tightness, unusual heartbeat, and lower airway and chronic pulmonary obstruction [NIOSH 1996, 1998; NRC 1981]. The OSHA formaldehyde standard [29 CFR 1910.1048]

Table 3–9. Effect of ventilation on formaldehyde concentrations in test chambers containing CCP

	Average formaldehyde concentration in test chamber (ppm)
Exposure time (min)	CB–15 blue	CB–15 black	SC–17 black	SC–14 black
0–10	0.050	0.080	0.166	0.065
5–15	0.049	0.089	0.216	0.072
10–20	0.071	0.099	0.214	0.072
15–25	0.060	0.089	0.219	0.074
20–30	0.055	0.093	0.227	0.068
25–35	0.039	0.064	0.173	0.023
30–40	0.028	0.042	0.113	0.010
40–50	0.007	0.032	0.053	0.001
50–60	0.0	0.010	0.059	0.004

Adapted from Hazelton Laboratories [1985].

Table 3–10. Summary of formaldehyde concentrations reported in CCP studies^[1]

Reference	Concentration (ppm)[2]
Gockel et al. [1981]	<0.51
Norbäck [1983b]	0.08–0.24
Hazelton Laboratories [1985]	0.033[3]
Apol and Thoburn [1986]	ND[4]
Chovil et al. [1986]	0.015–0.022
Omland et al. [1993]	0.08–0.5
Zimmer and Hadwen [1993]	0.019–0.028

1. The NIOSH REL is 0.016 ppm as an 8-hr TWA with a 15-min ceiling limit of 0.1 ppm. The OSHA permissible exposure limit is 0.75 ppm as an 8-hr TWA with a short-term exposure limit of 2 ppm.
2. 1 ppm=1.23 mg/m³.
3. For marking and separating 30 four-ply forms/hr for 8 hr (range 0.009–0.693).
4. Limits of detection varied from 0.04 to 0.08 ppm.

is based on a number of adverse health effects ranging from irritation to cancer [57 Fed. Reg. 22290 (1992)]. A full discussion of the health effects of formaldehyde is beyond the scope of this review.

Reported measurements for kerosene and total dust were far below the occupational exposure limits. The NIOSH REL for kerosene is 100 mg/m³ as a 10-hr TWA during a 40-hr workweek [NIOSH 1977]. No NIOSH REL has been established for total dust. OSHA has a PEL of 5 mg/m³ for the respirable fraction of particulates not otherwise regulated [29 CFR 1910.1000(z)(1)]. Kerosene is defined as Fuel Oil No. 1, Range oil (note: a refined petroleum solvent [predominantly C9–C16] that is typically 25% normal paraffins, 11% branched paraffins, 30% monocycloparaffins, 12% dicycloparaffins, 1% tricycloparaffins, 16% mononuclear aromatics, and 5% dinuclar aromatics) [NIOSH 1997]. Santosol, SurSol, and odorless kerosene are similar in chemical composition to kerosene. Symptoms of kerosene exposure include eye, skin, nose, and throat irritation; burning sensation in the chest; headache; nausea; weakness; restlessness; incoordination; confusion, drowsiness; vomiting, diarrhea; dermatitis; and chemical pneumonia (if liquid kerosene is aspirated). Airborne exposures at concentrations cited in the CCP studies are not likely to lead to eye or upper respiratory irritation. Quantitation of skin exposure to kerosene from CCP has not been reported. However, skin contact with CCP containing kerosene or its components could result in skin irritation.