Volatile Organic Compounds in Disposable Diapers and Baby Wipes in the US: A Survey of Products and Health Risks

Abstract

Many thousands of diapers are worn by young children and the elderly, who have thin and sensitive skin that is highly vulnerable to chemicals, including volatile organic compounds (VOCs) that may be ingredients of these products or present as inadvertent or residual components. The levels and potential health risks of VOCs in diapers have not been reported previously. In this study, we collected 31 disposable hygiene products in the US market based on market share and analyzed 98 target VOCs using purge and trap sampling and thermal desorption/gas chromatography/mass spectrometer analysis. Exposures and risks were modeled using reasonable upper level exposure scenarios. Adult diapers contained the highest total target VOC concentration (median level of 23.5 μg/g), and the predominant VOCs were alkanes. In some diapers, the estimated noncancer risk from these VOCs was sometimes very large (hazard quotient of 1609) due to n-heptane. Baby diapers contained several known or suspected carcinogens, including benzene and 1,4-dioxane, and the lifetime cancer risk from some diapers approached 1 per million under a worst-case scenario. Store-brand products had higher levels of VOCs than generic brands, and products labeled “organic” or “for sensitive skin” did not necessarily have lower levels. Our results show that toxic VOCs were found in all tested disposable diapers and wipes at trace levels, and risks from using some diapers in high use exposure scenarios are high enough to warrant additional attention and possibly corrective measures. We recommend eliminating and monitoring toxic ingredients and disclosing all chemicals that may be in these products.

Graphical Abstract

1. INTRODUCTION

In the United States (US), 27.4 billion disposable diapers are used each year¹ at a cost that exceeds $9 billion annually in North America.² The average child will use approximately 7000 disposable diapers before they are toilet trained.³ Compared to the disposable diapers introduced in the 1940s, today’s disposable diaper is highly effective with regard to moisture absorption and dryness due to the use of super-absorbent polymers that were introduced in the mid-1980s, resulting in healthier skin for the user. Diapers are used by young children before they are toilet trained, by a subset of children who continue use after being toilet trained, and by seniors and others suffering from incontinence, which includes over 20 million women in the United States.⁴

Diapers are in immediate contact with skin. For healthy individuals, the skin forms a strong protective barrier. However, repeated exposure to urine, feces, and excessive moisture may weaken the skin barrier. Some diaper users have growing, damaged, or deteriorated skin, conditions which can increase the risk of dermal complications such as diaper dermatitis. Diapers also contact external genitalia and tissues, such as the vulva, that have high chemical permeability.^5,6 Baby wipes are another widely used infant product. Baby wipes are used to clean the skin, and are comparable in terms of mildness to the skin as warm water and cotton wool.⁷ Baby wipes decrease skin pH (skin pH = 4.5–6.0, increased by excretion and probably restored by the pH buffering in baby wipes), which may be helpful for skincare.^8,9

Skin in the diaper area can be vulnerable to irritants, particularly for infants and the elderly. Infant (3–24 months old) epidermis and stratum corneum are 20% and 30% thinner, respectively, than adult skin.¹⁰ The thinner stratum corneum in combination with the much larger (up to 5-fold) body-surface-to-weight ratio of the newborn can increase the risk of skin damage, infection, percutaneous absorption, and toxicity from surface exposures.¹¹ These concerns can be exacerbated by diaper dermatitis, which while decreasing with the use of modern diapers,^12–14 remains common among babies, especially in their first year. In the elderly, the epidermis is thinned by decreased mitotic activity and slow replacement of barrier lipids, which also deteriorates many skin functions.¹⁵

A number of toxic chemicals, including volatile organic compounds (VOCs), have been found in disposable diapers. In 1999, by monitoring the respiration of mice exposed to disposable diapers and cloth diapers, Anderson reported that some disposable diapers contained m-xylene, toluene, and styrene and considered these respiratory toxicants to be a risk factor for asthma onset and exacerbation.¹⁶ In 2019, toluene and xylene, which are eye, skin, and respiratory irritants and linked to dermatitis, were detected in the plastic packaging of baby diapers.^17–19 Phthalates and the metal tin also have been reported in baby diapers.^20,21 In the European Union, naphthalene, styrene, toluene, dichlorobenzenes, xylenes, and chlorobenzenes were detected in diapers, which motivated a French proposal to restrict chemicals in baby diapers in 2021.²²

VOCs in diapers and wipes include components of fragrances, adsorbents, moisture barriers, adhesives, and binders; e.g., limonene is a common fragrance, and n-heptane is used in binder formulations. Other VOCs may be inadvertent contaminants or manufacturing byproducts, e.g., chloroform and 1,4-dioxane. VOCs can cause a wide range of health effects, e.g., irritation and toxicity to the eyes, skin, nose, respiratory system, liver, kidney, and reproductive system, as well as cancer.^23–25 Both acute exposure to high concentrations and long-term or chronic exposure to lower concentrations can pose health risks.^26–28

Disclosure of the ingredients in baby care products is not required in the US,²⁹ and information on the types and levels of VOCs in diapers and wipes is scant. Industry guidelines designed to ensure the safety of the ingredients for a number of consumer products have been published,³⁰ but these focus on liquids, creams, aerosols, detergents, mouthwash, toothpaste, and some other products; diapers have not been addressed. Overall, the qualitative nature of prior studies, the potential health concerns associated with diaper use, and the reported detection of several toxic VOCs in diapers and baby wipes suggest that a more comprehensive assessment is warranted.

The objectives of this study are to screen disposable diapers (including adult and baby diapers) and baby wipes sold in the US for VOCs and to provide initial estimates of exposures and health risks from the use of these products. We collect and test commercially available baby wipes, baby diapers, and adult diapers and assess exposures and noncancer and cancer risks using high-end but plausible exposure scenarios.

2. MATERIALS AND METHODS

2.1. Products Selection.

A broad range of baby and adult diapers and baby wipes were selected that included popular brands and products based on market data from Statista.com and Amazon.com. We sampled products of at least two of the best-selling brands in the US. We also collected the bestselling products from several store brands and at least one product labeled “organic” or “natural” for baby wipes and baby diapers. No store brands, “organic” or “natural” products for adult diapers were identified in our market area. All products were purchased from stores in Michigan (US) or official company websites. In all, we obtained 31 products (Table 1): 9 types of baby wipes (coded BW1–BW9), 12 types of baby diapers (BD1–BD12), and 10 types of adult diapers (AD1–AD10).

Table 1.

Characteristics of Products in the Study

	Number of Products
Product type	Tested	Store brand	“Organic”	Date Labeleda	Place labeledb	Mass (g) /unitc	Total number of target VOCs detected
Baby wipe	9	1	1	7	9	5.7 ± 0.7	50
Baby diaper	12	1	1	1	12	25 ± 5.7	43
Adult diaper	10	0	0	0	9	95 ± 43	34

^aProducts with identified manufacturing or expiration dates.

^bProducts with identified manufacturing sites.

^cMean ± standard deviation.

2.2. VOC Sampling: Purge and Trap Method.

Purge and trap methods were used to sample VOCs.³¹ For baby wipes, a whole wipe was weighed and inserted into a 40 mL vial, which was sealed using a Teflon septum and a screw-on cap. After heating to 40 °C for 10 min in a dry bath, the vial was purged with pure N₂ that was injected into the bottom of the vial using a long needle that pierced the cap’s septum; flow exiting the vial passed through a 10 cm long stainless-steel adsorbent sampling tube (Scientific Instrument Services, Inc., Ringoes, NJ, US) equipped with a needle inlet that also pierced the septum. The sampling tubes were packed with 150 mg of anhydrous sodium sulfate (Fisher Scientific, Fair Lawn, NJ, US) to remove water vapor, followed by 160 mg of 60/80 mesh Tenax-GR (Scientific Instrument Services, Inc., Palmer, MA, US) to trap VOCs. The total purge volume was 600 mL of N₂ over a 30 min period during which the vial temperature was held at 40 °C. For diapers, a small section from the middle of the diaper was cut and weighed (~1 g), then placed in a 40 mL vial to which 5 mL of LC–MS grade deionized water (MilliporeSigma, Burlington, MA, US) was added to simulate the absorption of urine by the diaper. The vial was capped, heated to 40 °C for 10 min, and then purged with 400 mL of N₂ for 20 min to collect VOCs, as described previously for wipes. These protocols, including the duration, volume, and temperature of the purge, were developed to capture 90% or more of VOCs in the products, based on repeated tests of the same sample (i.e., the second test contained <10% of the concentration of the first). The temperature of 40 °C was selected as it is near body temperature.

2.3. VOC Analysis.

Sample analyses followed well-developed protocols.^32,33 Each adsorbent tube was injected with internal standards (fluorobenzene, p-bromofluorobenzene, 1,2-dichlorobenzene-d₄) and analyzed using an automated short-path thermal desorption system (ATD, Scientific Instrument Services, Inc., Ringoes, NJ, US). The system was coupled to a gas chromatography–mass spectrometer (GC–MS, Model 6890/5973, Agilent Technologies, Santa Clara, CA, US) equipped with a cryotrap/focuser (−140 °C to focus, 250 °C to inject). Chromatographic separation was obtained using a DB-VRX capillary column (60 m × 0.25 mm, 1.4 μm film thickness) with helium as the carrier gas. The temperature program started at 45 °C (10 min hold), ramped at 8 °C/min to 140 °C (10 min hold), and finally ramped at 30 °C/min to 225 °C (hold for 13 min). The temperatures of the MS detector, transfer line, EI ion source, and quadrupole were 250, 300, 230, and 150 °C, respectively. The MS was operated in full scan mode from 27–270 atomic mass units (AMU). Peak areas were extracted by a ChemStation macro program (G1701BA Version B.01.00, Agilent, Santa Clara, US), adjusted for internal standards, and transferred to a spread-sheet.

2.4. VOC Calibration and Quality Control.

Samples were analyzed for 98 target VOCs. Multipoint calibrations (1, 3, 10, 30, and 100 ng) were performed for each target VOC using authentic standards in mixtures (using four mixture standards for 60 target VOCs and one mixture for three internal standards) or individual compounds (28 target VOCs; MilliporeSigma, Burlington, Massachusetts, US) by injecting the calibration mass into clean sampling tubes for analysis. Calibration ranges were extended for several VOCs that were very prevalent in the analyzed products, specifically, to 400 ng for isopropyl acetate, hexanal, heptanal, p-isopropyltoluene, n-dodecane, and n-tetradecane; 1000 ng for 2-butanone and n-octane; 2000 ng for hexane and limonene; and 100,000 ng for n-heptane. Piecewise linear calibrations were used, and the correlation coefficients of calibration curves for each segment exceeded 0.99. Recovery rates for most compounds ranged between 80% and 120%. Method detection limits (MDLs) for the target VOCs, determined as the standard deviation of seven replicate low concentration injections multiplied by 3.14,³⁴ ranged from 0.01 to 2.5 ng. Measurements below MDL were set to 0 and are shown as “< MDL”. The 98 target VOCs were divided into eight chemical groups: 6 aldehydes, 12 alkanes, 19 aromatics, 40 halohydrocarbons, 2 terpenes, 4 ketones, 7 esters, and 8 others. In the Supporting Information, Table S1 shows the target VOCs, internal standards, chemical groups, target ions, qualifier ions, and MDLs.

The sum of the 98 target VOCs is designated as the total target VOCs (TTVOCs). For a subset of products, we examined the larger nontarget peaks, using the MS fragmentation pattern and elution time, to provide a tentative identification if the match quality value with the NIST 98 spectral library³⁵ exceeded 90%.

Quality assurance activities included preparation and analysis of blanks and duplicates (47% of samples). The coefficient of variation (COV) of true duplicates averaged 40%. Duplicates were averaged in the data analysis. A freshly loaded adsorbent tube injected with 10 ng of standards was analyzed daily. Differences between daily checks and calibration results were within 30%. Trace level contamination (<4 ng) was detected in blanks for 10 compounds (hexane, benzene, toluene, hexanal, ethylbenzene, styrene, n-undecane, nonanal, naphthalene, and n-tridecane); blank-corrected results were used for these cases.

2.5. Indoor Air Quality Impacts.

Given the way the baby wipes are used, we assumed that VOCs remaining in baby wipes after use (e.g., placed in a trash bin) would evaporate into the indoor air. The evaporation of VOCs in baby wipes induces a concentration increase in indoor air.

VOCs in the baby products were assumed to disperse in indoor air in three stages or zones for a reasonable worst-case scenario used to estimate concentrations and exposure of the baby and caregiver: (1) the nearfield case assumed that for the first 2 min after use, VOCs distributed within a 2 m³ volume around the user (equivalent to a radius of 0.8 m around the user) with an air change rate (ACR) of 360 h⁻¹;^36,37 (2) the midfield case modeled the subsequent 4 min (90th percentile of time spent in bathroom = 60 min per day in the US³⁸ and 10 wipes used by a baby per day, i.e., 6 min in bathroom each time) when VOCs dispersed within a bathroom with volume of 10 m³³⁶ and ACR of 0.18 h⁻¹ (10th percentile value in US residences);³⁸ and (3) the far-field case when VOCs distributed throughout the residence, which has an interior volume of 154 m³ and ACR of 0.18 h⁻¹ (again, 10th percentile value).³⁸ The three stages used fully mixed box models to estimate concentrations.³⁹ VOC loss through the building envelope occurs due to air change with outside air during the third stage

$$C_t=\left(C_{\text{s}}-C_{\text{R}}\right)\times e^{\left(-A\times t\right)}+C_{\text{R}}$$ (1)

$$C_{\text{s}}=\frac{C_{\text{P}}\times M_{\text{R}}\times {10}^{-3}}{V}$$ (2)

where $C_t$ is the concentration (μg/m³) in air at time $t\left(\text{h}\right)$; $C_{\text{s}}$ is the initial VOC concentration (μg/m ) in air, herein, we assume the beginning concentration of all remaining VOC in wipes evaporating into the dispersal volume, as in eq 2; $C_{\text{R}}$ is the VOC concentration (μg/m³) in the zone at the end of this stage, herein, the beginning concentration in the next stage; $e$ is the natural constant; $A$ is the ACR (h⁻¹); $C_{\text{P}}$ is the VOC concentration (ng/g) in product; $M_{\text{R}}$ is the remaining mass (g) of product after use; $V$ is the dispersal volume for each stage (2 m³ in nearfield; 10 m³ in midfield; and 154 m³ in farfield); and 10⁻³ is the conversion from ng to μg. All parameters are shown in Table S2.

The average increment of the concentration in indoor air is

$$C_{\text{I}}=\frac{{\int }_0^T{C}_t\text{d}t}{T}$$ (3)

where $C_{\text{I}}$ is the average increment over period $T$; $t$ is the time in the zone; and $T$ is the total time.

2.6. Exposure and Risk Assessment.

Exposures were calculated using reasonable upper level exposure scenarios that assumed dermal exposure for disposable diapers and dermal and inhalation exposure pathways for baby wipes.^22,31 For diapers, we assumed that all of the VOCs in the product were absorbed via dermal exposure. For wipes, the contact times are too short to absorb all VOCs into the skin; thus, we conservatively assumed that unabsorbed VOCs evaporated into indoor air with inhalation exposure occurring using concentrations calculated using the 3-zone indoor air quality model described previously.⁴⁰ Risks were calculated as the sum of dermal and inhalation exposures for baby wipes.

We used eq 4 to calculate daily dermal exposure dose (mg/kg-day) for the tested products, eq 5 to estimate noncancer risks (hazard ratio for adverse nonneoplastic health effects) of dermal exposure, and eq 6 to estimate cancer risks of dermal exposure

$$\text{daily}\text{dose}=\frac{C_{\text{i}}\times M\times \text{Freq}\times {10}^{-6}\times 365\times D_{\text{E}}}{\text{body}\text{weight}\times 365\times D_{\text{E}}}$$ (4)

$$\text{hazard}\text{ratio}\left(\text{HR}\right)\text{of}\text{dermal}\text{exposure}\text{=}\frac{\text{daily}\text{dose}}{\text{RfD}}$$ (5)

$$\text{cancer}\text{risk}\left(\text{CR}\right)\text{of}\text{dermal}\text{exposure}=\frac{\text{daily}\text{dose}\times D_{\text{E}}}{\text{life}\text{expectancy}}\times \text{CSF}$$ (6)

where $C_i$ is the concentration (ng/g) of the $i\text{th}$ VOC in the product; $M$ is the residual mass (g) of product on skin each time for wipes, or the measured mass of each unit for diapers (Table S2); $\text{Freq}$ is the number of times the product is used per day (Table S2); 10⁻⁶ is the conversion from ng to mg; 365 is for 365 day/year; $D_{\text{E}}$ is the number of years of using the products; RfD is the reference dose (mg/kg-day), an estimate of a daily exposure that is likely to be without an appreciable risk of adverse noncancer health effects;⁴¹ life expectancy is 80 years assumed in this study; and CSF is the cancer slope factor (kg-day/mg), which is the upper bound (approximately a 95% confidence limit) for the increased cancer risk from a lifetime exposure expressed as dose (mg/kg-day).⁴¹ Potential risks associated with dermal doses were evaluated using RfDs (Table S3) and CSFs (Table S3) from CalTOX⁴² or calculated using eqs 7 and 8 based on oral exposure and gastrointestinal absorption factors (GIABS)⁴³

$${\mathrm{RfD}}_{\text{dermal}}=\text{Rf}{\text{D}}_{\text{oral}}\times \text{GIA}BS$$ (7)

$${\text{CSF}}_{\text{dermal}}=\frac{\text{Rf}{\text{D}}_{\text{oral}}}{\text{GIABS}}$$ (8)

where ${\mathrm{RfD}}_{\text{dermal}}$ is the reference dose (mg/kg-day) of dermal exposure; $\text{Rf}{\text{D}}_{\text{oral}}$ is the reference dose (mg/kg-day) of oral exposure; and $\text{GIABS}$ is the gastrointestinal absorption factor (dimensionless).⁴³

For inhalation exposures, eq 9 was used to calculate inhalation exposure concentration (μg/m³)

$$\text{inhalation}\text{concentration}=\frac{C_{\text{I}}\times \text{Freq}\times {10}^{-3}\times T_{\text{indoor}}\times 365\times D_{\text{E}}}{24\times 365\times D_{\text{E}}}$$ (9)

where $C^{\text{I}}$ is the average increment of the indoor air concentration over period $T_{\text{indoor}}$; $\text{Freq}$ is the times of the product used per day (Table S2); $T_{\text{indoor}}$ is the time period spent in house per day (h/day); $D_{\text{E}}$ is the duration of using the products (years); 10⁻³ is the conversion from ng to μg; 24 is for 24 h/day; and 365 is for 365 day/year.

The noncancer risk (hazard ratio) and the cancer risk, both from inhalation exposures, were calculated using eqs 10 and 11

$$\text{hazard}\text{ratio}\left(\text{HR}\right)\text{of}\text{inhalation}\text{exposure}\text{=}\frac{\text{inhalation}\text{concentration}}{\text{RfC}}$$ (10)

$$\mathrm{cancer}\text{risk}\left(\text{CR}\right)\text{of}\text{inhalation}\text{exposure}=\frac{\text{inhalation}\text{concentration}\times D_{\text{E}}}{\text{life}\text{expectancy}}\times \text{UR}$$ (11)

where RfC is the reference concentration (μg/m³), which means an upper estimate of a continuous inhalation exposure that is likely to be without an appreciable risk of adverse noncancer health effects over a specified duration of exposure;⁴¹ life expectancy is 80 years assumed in this study; and UR is the unit risk (m³/μg), which means the upper bound, approximating a 95% confidence limit, on the increased cancer risk from continuous inhalation exposure to a chemical at a concentration of 1 μg/m³ for a lifetime.⁴¹ The RfCs (Table S3) and URs (Table S3) for inhalation exposure were obtained from the US EPA⁴⁴ and the State of Michigan.⁴⁵

We considered risks for baby caregivers, including parents and staff in child care centers who contact baby wipes and inhale the air contaminated with baby wipes. The latter were assumed to have the most exposure. Because caregivers usually wash their hands after using wipes, dermal exposures were estimated using the “water solution” scenario in IH SkinPerm v2.0,⁴⁶ which simulates evaporation, uptake in the stratum corneum, and permeation through the skin (parameters in Table S2). As a reasonable upper level exposure scenario, we assumed a caregiver responsible for 5 babies,⁴⁷ worked in a 178 m³ volume space with an attached bathroom of 10 m³,⁴⁸ and worked 40 h per week for 30 years (aged 25–54). The inhalation concentration was calculated by eq 12

$$\text{inhalation}\text{concentration}\text{=}\frac{C_{\text{I}}\times \text{Freq}\times {10}^{-3}\times 40}{168}$$ (12)

where $C_{\text{I}}$ is the average concentration increment in indoor air during the working day.

2.7. Data Analysis.

Descriptive statistics noted in the text include median VOC concentrations measured for each product; the tables provide additional statistics, e.g., mean, standard deviation (SD), and range. Differences between product groups were evaluated using Mann–Whitney U (two samples) and Kruskal–Wallis (three or more samples) tests. The potential influence of different factors on VOC composition, including product type, manufacturer, “organic” (one product “for sensitive skin” coded as “organic”), store-brand, “scented,” and interaction terms between product type and the other four individual factors, were evaluated using permutational multivariate analysis of variance (PERMANOVA) with 9999 permutations.⁴⁹ Two-sided tests and a type-I error rate of 0.05 were used. Analyses used SPSS (SPSS, Inc., Chicago, Illinois, US) and R version 3.6.0.⁵⁰

3. RESULTS AND DISCUSSION

3.1. Product Characteristics.

The 31 tested products are listed in Table 1. Baby products (wipes and diapers) included both store and nonstore brands and some products labeled “organic”. Adult diapers were obtained from only “big name” manufacturers (no store-brands), and no “organic” adult diapers were found or tested. Only 26% of the products indicated the manufacturing or expiration date (mostly baby wipes). Most products (97%) listed their manufacturing location. The “organic” products were produced in European countries. All other products were manufactured in the US except for one product from Canada. Diaper size and weight depended on the intended user: the tested baby diapers averaged 25 ± 5.7 g/unit (N = 12), and adult diapers averaged 95 ± 43 g/unit (N = 10). Baby wipes had similar weights, averaging 5.7 ± 0.7 g/unit (N = 9), a little heavier than a type of similar product, feminine wipe, as 4.9 ± 1.2 g/unit.³¹

3.2. VOCs in Different Products.

Across the products, we identified from 11 to 37 target VOCs, which represented most (50–70%) of the VOC mass estimated for the 20 to 130 distinct peaks seen on chromatograms. For baby wipes, most of the larger nontarget peaks were identified as fragrances, e.g., dihydromyrcenol (Table S4). For diapers, several nontarget peaks were identified as components of adhesives, e.g., 2-methylhexane and methylcyclohexane, likely used for binding the multiple layers together.⁵¹

Several VOCs that are known toxins were found in many products but generally at low concentrations. Chloroform, considered a likely human carcinogen, was found in 56% of baby wipes, 67% of baby diapers, and 50% of adult diapers; the highest concentration was 12 ng/g in a baby diaper (product BD9). Benzene, a known carcinogen, was found in 90% of baby products, and the highest level was 11 ng/g in a baby diaper (BD2). Bromodichloromethane, a probable carcinogen, was found in only baby wipes (44%); the highest concentration was 1.1 ng/g (BW7). 1,4-Dioxane, classified as a likely carcinogen,⁵² was found in 67% baby products (78% of baby wipes and 58% of the baby diapers); the highest level was 3.6 ng/g in a baby diaper (BD10). Toluene was present in all analyzed products; an adult diaper (AD3) had the highest concentration, 14 ng/g. Limonene, coded with a yellow triangle by the US EPA to designate potential hazard profile issues,⁵¹ was in 94% of products with the highest level of 307 ng/g in a baby wipe (BW7). Naphthalene was found in 87% of products at very low concentrations (<1.0 ng/g). Other known or suspected carcinogens found in only baby wipes and at low concentrations included bromoform (0.3 ng/g) and hexachlorobutadiene (0.05 ng/g).

VOC compositions and levels differed by product type (Figures 1 and 2 and Table S5). Adult diapers had the highest TTVOC levels (median of 15,603 ng/g; p = 0.07) and baby diapers the lowest (81 ng/g; p = 0.1). The dominant compound classes were terpenoids (44% of TTVOC) in baby wipes, aldehydes (33%) and alkanes (29%) in baby diapers, and alkanes (>99%) in adult diapers. Baby wipes contained a variety of VOCs, including aldehydes (e.g., butanal, hexanal, heptanal), esters (ethyl acetate, n-propyl acetate), aromatics (isopropyl benzene, 1,3,5-trmethylbenzene), ketones (2-hexanone), alkanes (n-octane, n-undecane), and terpenes (α-pinene). Baby diapers had relatively complex ingredients and included 2-butanone, benzene, nonanal, and n-pentadecane. In comparison, adult diapers had few VOCs with the predominant compounds being aromatics (e.g., toluene, ethylbenzene) and alkanes (hexane, n-dodecane, n-tridecane, n-tetradecane). We speculate that adult diapers prioritize absorbency with few requirements for visual aesthetics and fragrance that require VOCs or other chemicals. In contrast, baby diapers may utilize more VOCs and complex production processes to impart a pleasant appearance and scent.

Figure 1.

VOC composition by compound group (mass proportion) for the three types of products.

Figure 2.

Boxplots showing concentrations of individual VOCs (A) and VOC groups (B) in different products.

The VOC content in baby products is much lower than the levels in many other types of household and personal care products. For example, TTVOCs in baby wipes and diapers averaged 178 and 94 ng/g, respectively, compared to 2067 and 8473 ng/g in feminine wipes and menstrual pads, respectively.³¹ Products such as shampoo and sprays have much higher levels of terpenes, likely used as fragrance or masking scents.^31,53 VOC emissions in indoor spaces occur from numerous other household and personal care products, including cleaning products, disinfectants, waxes, deodorants, antiperspirants, aerosol sprays, adhesives, pesticides, and lubricants, as well as building materials and furnishings.⁵⁴ While the VOCs emitted by baby products add only incrementally to indoor concentrations, they contribute to total VOC exposure and risk.

The alkanes in adult diapers included compounds typically seen in solvents and lubricants,⁵⁵ e.g., n-hexane, n-heptane, n-nonane, n-decane, n-undecane, n-tridecane, and n-tetradecane. Several nontarget alkanes were also identified, e.g., 3-methylhexane, 2-methylhexane and methylcyclohexane. While used as solvents, these VOCs are probably constituents of adhesives in these products,⁵⁶ e.g., binding multiple layers together in diapers.²² The same VOCs have been found in feminine hygiene products (menstrual pads), probably to help stick the pad onto underwear.³¹

The products included chloroform (58% of the measured products) and other halohydrocarbons. Baby wipes contained six additional halohydrocarbons, including butyl chloride, bromodichloromethane, dibromochloromethane, and tetrachloroethene. These VOCs may be byproducts and/or residuals of bleaching and disinfection.^57,58 Although baby wipes are designed mainly for mild cleansing (without adding disinfecting irritants), the raw material, nonwoven fabric, is bleached during manufacturing.^59,60 For diapers, the major raw material is wood pulp, which is bleached and processed to make the fibers soft, fluffy, and absorbent. Bleaching using elemental chlorine gas was eliminated in the 1990s as trace amounts of toxic byproducts like dioxins could be formed; today, pulp is purified using elemental chlorine-free (ECF), enhanced elemental chlorine-free (EECF), and total chlorine-free (TCF) methods that have virtually eliminated many chlorinated chemicals, including dioxins.⁶¹ The chlorine dioxide used in ECF processes is commonly used to disinfect water.⁶² Note that “chlorine-free” bleaching does not mean that the product has no chlorine atoms as they can be introduced through municipal water sources.⁶³ The limit for total trihalomethanes (THM, including chloroform, bromodichloromethane, dibromochloromethane, and bromoform) in municipal drinking water is 80 μg/L in the US,⁶⁴ and the mean concentration of total THM in large water systems in the US was reported as 30.5 μg/L with a 95th percentile level of 71.2 μg/L.⁶⁵ Several reports have shown that municipal water used in manufacturing processes has been further disinfected to ensure safety, resulting in elevated levels of halogenated disinfection byproducts.^66,67

3.3. VOC Differences by Product Characteristics.

While our sample size is limited, several trends in VOC levels and compositions were observed. First, several baby wipes and baby diapers labeled “scented” had higher levels of esters compared to the other products (Figure S1A); e.g., scented wipes contained significantly higher levels of n-propyl acetate, and scented baby diapers had significantly higher isopropyl acetate. Second, product labeling was not a consistent indicator of the VOC content (Figure S1B). Products labeled “organic” were obtained for wipes and baby diapers; an additional wipe sample was labeled “for sensitive skin”. While having lower levels of terpenes (0.4 ng/g, especially limonene at 0.2 ng/g and ketones at 2.0 ng/g) compared to conventional products (10 ng/g for terpenes, 8.9 ng/g for ketones in baby wipes; p = 0.06), the “organic” baby wipes had the highest levels of several alkanes (n-dodecane, n-tridecane, n-tetradecane, and n-hexadecane), though they did have the lowest levels of several other compounds (2-hexanone, n-octane, styrene, and limonene). The “organic” baby diapers contained several VOCs at higher concentrations (toluene, 1,2,4-trimethylbenzene, naphthalene, and n-tetradecane). Wipes labeled “for sensitive skin” contained the highest levels of heptanal (34 ng/g), n-decane (2.4 ng/g), and octanal (6.8 ng/g) but the lowest levels of 2-butanone (0.6 ng/g) and α-pinene (0.1 ng/g).

The reasons for the higher VOC levels in the “organic” and some other products are unknown. As mentioned, the “organic” products were from Europe, while the other products were from North America. One possibility is that the VOC content may reflect regional differences in the feedstocks, manufacturing practices, and chemicals, e.g., greater use of plant-derived “natural” solvents in Europe. A second possibility is contamination during shipping and storage, which may be more likely with long-distance transport. A third speculation is differences in packaging; well-sealed and low-diffusivity packaging can maintain the VOC content of the product, and the packaging itself may also emit VOCs. Finally, VOC formulations may be adjusted to meet consumer expectations, product performance requirements, and local regulations.

Among the public, labels such as “organic” and “for sensitive skin” suggest that the product is “cleaner” and “healthier” than conventional (unlabeled) products. However, unlike labels on foods and other products that are regulated and based on objective criteria, e.g., the US Department of Agriculture’s (USDA) National Organic Program standards, these terms are not well-defined or regulated for personal care products.⁶⁸ (A few baby products displayed the USDA organic seal.) Such labels on diapers and other baby care products were not informative about VOC emissions.

Store-brand products tended to have higher VOC levels than nonstore brands (Figure S1C). For baby wipes, the store-brand sample had higher levels of n-propyl acetate, 2-butanone, chloroform, toluene, and styrene (but lower levels of benzene, n-heptane, hexanal, heptanal, and nonanal); the store-brand baby wipes were the only products tested that contained tert-butylbenzene and hexachlorobutadiene. For baby diapers, the store-brand sample had higher levels of aldehydes (e.g., butanal, pentanal, octanal, and nonanal) and several alkanes (hexane, n-decane, n-dodecane, n-tridecane, and n-pentadecane). Store-brand baby wipes and baby diapers had higher levels of several VOCs compared to those of nonstore brands. Again, reasons for differences are unknown, but they may reflect different product specifications, ingredients, and manufacturing processes.

Our results suggest some brand or manufacturer effects on VOC levels. Adult diapers were obtained from five manufacturers. Diapers from manufacturer PB had highest VOC levels (median TTVOC concentration of 44,827 ng/g compared to 89 ng/g for the other four manufacturers; Figure S1D). Diapers from manufacturer PB also had higher levels of toluene (p = 0.02) and hexane (p = 0.02), although o-xylene and n-pentadecane were lower (p < 0.05). “Conventional” baby diapers were obtained from two nonstore brands. (The sample did not include store-brand or “organic” baby diapers.) Baby diapers from manufacturer P had higher levels of isopropyl acetate, 1,4-dioxane, toluene, ethylbenzene, and xylene but lower levels of aldehydes (especially heptanal and nonanal) and several alkanes (i.e., n-undecane, n-dodecane, and n-pentadecane) than manufacturer K (p < 0.05, Figure S1D). For baby wipes, most differences resulted between store-brand and “organic” manufacturers. No significant differences were seen for products from the other two manufacturers. We note that most products were produced in the US (90%) and Canada (3%), while the “organic” products were produced in Europe (7%). As noted earlier, location may affect VOC levels through differences in regulations, feedstocks, manufacturing process, and consumer expectations.^69,70

Overall, the test results suggest that product type and manufacturer affect the VOC composition. Based on PERMANOVA results, the most important factors are product type and manufacturer (p < 0.01; see R² in Table 2). In addition, the significant interaction term between the product type and manufacturer indicates that the manufacturer effect depends on the product type: VOC differences by manufacturer were found for baby and adult diapers but not for baby wipes. However, these findings are considered tentative due to our limited sample size and the need to test multiple samples from different lots of the same product to ensure representative results.

Table 2.

PERMANOVA Comparing the VOC Compositions by Product Characteristics

Factor	df	Sums of Sqs	Mean Sqs	F. Model	R 2	p value
Product type	2	1.6620	0.83098	6.2079	0.22782	0.0003b
Manufacturer	4	1.6087	0.40218	3.0045	0.22052	0.0029a
“Organic” or “For sensitive skin”	1	0.1723	0.17232	1.2874	0.02362	0.1745
Store-brand	1	0.2112	0.21118	1.5776	0.02895	0.1088
“Scented”	1	0.0448	0.04484	0.3350	0.00615	0.9047
Product type × Manufacturer	2	1.1042	0.55212	4.1246	0.15137	0.0043a
Product type × “Organic”	1	0.0679	0.06789	0.5072	0.00931	0.7381
Product type × Store-brand	1	0.1366	0.13664	1.0208	0.01873	0.3841
Product type × “Scented”	1	0.1457	0.14567	1.0882	0.01997	0.3365
Residuals	16	2.1417	0.13386		0.29358
Total	30	7.2952			1.00000

^ap < 0.01.

^bp < 0.001.

The types of VOC measurements used in this study can establish the presence of unnecessary and undesirable VOCs in personal care and hygiene products. This information can be used by consumers to select low-emitting products and by manufacturers to identify contamination sources and to change product formulations. While we sampled a relatively broad range of products, repeat sampling is needed to develop mean values and confidence intervals and to confirm that brand differences are consistent.

3.4. Environmental Impact and Health Risks.

Based on the 3-stage indoor air quality model, the residual VOCs in a discarded baby wipe would produce a concentration increment of up to 1.3, 0.3, and 0.02 μg/m³, in near-, mid-, and far-field zones, respectively (Figure 3). Concentrations decreased rapidly, e.g., by 99% in the first 4 h due to dilution, and the first 4 h of inhalation exposure accounted for 63% of the total inhalation exposure by using a baby wipe. Because these products are used periodically, typically every 4 h or more frequently, indoor VOC concentrations remain elevated. Still, inhalation risks were much lower than dermal exposure for babies (Figure S2). The indoor air quality model reflected dilution and exchange with outdoor air. In addition, VOCs can undergo chemical and physical transformations driven by oxidation, photolysis, and other processes⁷¹ that can form oxygenated species and secondary organic aerosols (SOA), some of which may cause health impacts.⁷² These processes, as well as deposition, hydrolysis, and ventilation, will also remove VOCs from indoor air.⁷³ These secondary processes, which depend on a number of indoor air parameters, were not evaluated in this study.

Figure 3.

TTVOC trends (or increments) in indoor air and the inhalation exposure proportion over time due to the use of a single baby wipe at time = 0. Different colors represent different brands of baby wipes. Inset plots show concentrations in near-, mid-, and far-field exposure zones.

The potential for chronic noncancer health effects, expressed as a hazard ratio (HR), is shown in Figure 4A. The highest HR (1609) was associated with adult diapers and was largely due to dermal exposure to n-heptane (>99% of the risk), which can irritate the skin and the respiratory tract and cause central nervous system effects.^74,75 The daily dermal exposure dose of n-heptane from one sample of an adult diaper (AD3) was 0.5 mg/kg-day, compared to the reference dose of 0.0003 mg/kg-day. Notably, all four products from the manufacturer had extremely high and similar levels of n-heptane. Baby diapers had HRs below 1, indicating negligible potential for noncancer health effects, with one exception (BD10) that had an HR = 4.3, also due to n-heptane (101 ng/g). Baby wipes had HRs below 0.5; most of the risk was due to dermal exposure to n-nonane (HR = 0.2), which is a skin and respiratory tract irritant.^74,75 Overall, the noncancer risks were driven by alkanes, a VOC class with relatively low toxicity and little if any cancer risk. Still, long-term exposure can irritate skin, mucous membranes, and the respiratory tract,⁷⁶ and some alkanes have been associated with reproductive toxicity.⁷⁷ The nontarget alkanes found in the products, e.g., 2-methylhexane and methylcyclohexane (mentioned above), also may have potential for similar adverse health impacts.^51,56

Figure 4.

Boxplots showing hazard ratios (A) and cancer risks (B) for using different products including baby caregivers.

Excess cancer risk (CR) estimates are shown in Figure 4B. Most (67%) baby diapers had CRs in the 10⁻⁷–10⁻⁶ range, and none exceeded 10⁻⁶, a conservative benchmark. The highest CR was 7.9 × 10⁻⁷, largely due to benzene (CR = 6.7 × 10⁻⁷) and 1,4-dioxane (CR = 1.1 × 10⁻⁷). Most (89%) of the baby wipes also fell in the 10⁻⁷–10⁻⁶ range, and the highest was 4.9 × 10⁻⁷, due mostly to dermal exposure of bromodichloromethane (44% of the CR), dibromochloromethane (28%), chloroform (13%), and benzene (12%). CRs were lower for adult diapers, and only 40% of the samples had CRs exceeding 10⁻⁷; the CR was mainly (>90%) due to chloroform. The highest CR was produced by aromatic and halogenated VOCs, including benzene, chloroform, and bromodichloromethane. The halomethanes and benzene are known or suspected carcinogens;⁵⁶ dibromochloromethane and benzene have been associated with genetic defects; chloroform has been linked to fetal toxicity;⁵⁶ and 1,4-dioxane is a likely carcinogen⁷⁸ with effects on kidneys and liver.^79–81 Overall, while known, likely, and suspected carcinogenic VOCs were detected in the tested products, levels were low and unlikely to pose a meaningful cancer risk.

All building occupants can experience inhalation exposures from the VOC emissions from diapers and wipes, although the child and caregiver(s) will have the highest exposures. For the upper level exposure scenario (caregiver is responsible for 5 babies in a 178 m³ child care classroom), the total (dermal and inhalation) exposure produced only low risks: the hazard ratio was 0.04 and the cancer risk was 1.4 × 10⁻⁹ (Figure 4). Dermal exposures provided nearly all of the noncancer risk, while inhalational exposures dominated the cancer risk (Figure S3).

We used reasonable upper level estimates to provide conservative exposure and health risk estimates. As examples: we assumed 7 baby diapers are used every day for a child from 0 to 3 years of age (7665 diapers), compared to the average of 5 diapers per day (total of ~7000 diapers),^3,22 and 10 baby wipes per day (total of ~10000 wipes). Still, risks may be underestimated for several reasons: diapers and wipes are used by babies and elderly people who may be more sensitive and vulnerable to chemical exposure than healthy adults; risks were quantified for only target VOCs, although several nontarget VOCs were identified and chemicals may be present, e.g., phthalates and tin;^20,21 potentially more than 5 infants may be in a large facility; and effects of mixtures or the simultaneous use of several products were not considered although these products are often used in combination (e.g., baby diapers and wipes). Conversely, if considering the risks to babies, baby diapers and wipes are typically not used for a 7-year or longer period that is usually considered as chronic exposure to carcinogens,⁸² but the exposure period for adult diapers could be very long, e.g., for individuals who are paralyzed or suffering from incontinence.

3.5. Study Strengths and Limitations.

Study strengths include the analysis of 31 different products that included a wide range of the bestselling products and brands and the measurement of a wide range of VOCs. To the best of our knowledge, this is the first such survey of VOCs in such consumer products, and it both fills a data gap and paints an overall picture of VOC content. We utilized sensitive and validated methods, and the target VOCs included many toxic compounds. Exposures and health risks were calculated using reasonable upper level exposure scenarios, and the results provided conservative estimates.

We recognize several limitations in the study. Sample sizes were small, especially for subgroups of products, those with “organic” labels. We did not test multiple lots or assess the variability within a product type or brand. Extraction conditions (e.g., temperature and water addition) for the samples may not match actual conditions in practice. Our analysis considered only the direct emissions of VOCs from selected products, and we did not address issues of VOC fate, the formation of oxygenated species and secondary organic aerosols that may have additional health impacts. Effects of product storage and transport (e.g., time and temperature at the manufacturer, store, and user’s home) were not evaluated. Long storage times in clean environments might allow VOCs to disperse, although most products were wrapped in near airtight plastic that would limit dispersion. The exposure scenarios were simplified and do not apply to all situations, e.g., assumptions regarding exposure durations and locations. Chronic health risks were estimated for baby products that have typically been used for only a few years. Exposures and risks may be underestimated for several reasons: dermal permeation parameters and concentration–effect relationships for infants and the elderly were not available, and these groups may be more vulnerable than adults; some extreme exposure scenarios could be worse than our assumption; and while a broad range of compounds were quantified, other VOCs and exposures should be considered. Despite these limitations, the study reveals that diapers and baby wipes contain undesirable VOCs. More comprehensive follow-up studies, including products used in other countries, are warranted.

3.6. Implications of the Study.

All 31 disposable hygiene products contained VOCs, including toxic compounds, but at mostly low levels. The composition and content differed by product type and manufacturer; e.g., adult diapers had simpler ingredients than baby diapers and baby wipes. In general, product labeling as “organic” or “sensitive skin” was not informative with respect to VOC composition and concentration. Based on conservative but realistic exposure assumptions, cancer risks associated with VOCs in disposable diapers and wipes were negligible. However, some adult diapers and some baby diapers contained n-heptane at levels that exceeded protective reference levels for noncancer risks; in those cases, exposures may be sufficiently high to irritate sensitive tissues.

We conclude with several recommendations. For these (and other consumer) products, ingredients and production dates should be disclosed, and labeling as “organic” does not serve as a proxy for meaning that a product is harmless or beneficial. Toxic constituents, such as n-heptane, benzene, and 1,4-dioxane, should be eliminated. Finally, uniform and protective testing and assessment approaches should be implemented to ensure the safety of products.