Abstract

Thermal paper has transitioned away from bisphenol A (BPA) towards alternative color developers such as bisphenol S (BPS), D-8 and Pergafast 201 (PF201), primarily in response to regulatory action on BPA. However, quantitative data on the dermal absorption of these substitutes remain limited, even though skin contact with receipts is a major route of exposure for cashiers and frequent users. This work synthesizes ex-vivo human skin absorption experiments conducted in flow-through diffusion cells to compare permeation of BPA, BPS, D-8 and PF201 under standardized conditions and to explain discrepancies between previously published BPA skin data. Using validated LC-MS/MS analysis, absorbed doses in receptor fluid after 24 h were approximately 25% of the applied dose for BPA, 17% for D-8, 0.4% for BPS and below the limit of quantification for PF201, with additional fractions retained in the skin (≈15–16% for BPA and D-8, ≈11% for BPS). A systematic exploration of experimental parameters demonstrated that skin thickness, freezing, donor volume and concentration can alter apparent BPA flux by factors of three- to ten-fold, reconciling inconsistent literature values. Negligible dermal metabolism of BPA to BPA-glucuronide was observed, highlighting the relatively high systemic bioavailability of free BPA after skin exposure. Overall, BPA shows the highest dermal permeability, D-8 is only moderately lower, BPS permeates slowly but accumulates in skin, and PF201 appears to have minimal permeation under the tested conditions. These findings underscore that substitution in thermal paper must be guided by robust dermal absorption and toxicity data rather than the absence of BPA alone.

Keywords

Dermal absorption of BPA; thermal paper; BPS; D-8; Pergafast 201; ex vivo human skin; diffusion cells; permeability coefficient; occupational exposure; regrettable substitution.

1. Introduction

Bisphenol A (BPA) has been one of the most intensively studied endocrine-disrupting chemicals of the last two decades. Beyond its well-known use in polycarbonate plastics and epoxy resins, BPA has been widely employed as a colour developer in thermal paper, including receipts, tickets and labels. In this application, BPA is present in its free, non-polymerised form, making it available for direct transfer to human skin during handling and potentially for systemic absorption.

Epidemiological and biomonitoring studies have repeatedly reported elevated urinary BPA levels in cashiers and workers who handle receipts compared with the general population, indicating that dermal exposure is a relevant contributor to total BPA body burden. At the same time, regulatory agencies have progressively restricted BPA in thermal paper; in the European Union, a ban on BPA in thermal papers entered into force in 2020. As a result, manufacturers have shifted to alternative colour developers such as bisphenol S (BPS), D-8 (WinCon 8) and Pergafast 201 (PF201).

However, replacing BPA with new developers raises two critical questions:

1. How efficiently do these alternatives permeate human skin relative to BPA?
2. To what extent can variations in experimental design explain discrepancies in published BPA permeation data?

Addressing these questions is essential for realistic dermal exposure assessment and for avoiding regrettable substitution, where a banned substance is replaced by alternatives with equal or greater health risks.

This paper synthesizes a series of systematic ex-vivo human skin experiments that:

• Develop and validate a high-sensitivity LC-MS/MS method for BPA, BPA-glucuronide (BPA-G), BPS, D-8 and PF201 in receptor fluid.
• Quantify and compare the dermal permeation of these compounds through human skin in flow-through diffusion cells.
• Investigate how skin thickness, viability (fresh vs frozen), donor volume and concentration influence apparent BPA permeation, thereby explaining inconsistencies in the literature.

2. Materials and Methods

2.1 Analytical Method for Quantifying Bisphenols

An LC-MS/MS method was established to quantify BPA, its major conjugate BPA-glucuronide (BPA-G), and three thermal-paper alternatives (BPS, D-8, PF201) in receptor fluids obtained from in-vitro skin absorption experiments.

Chromatographic and detection conditions

• Instrumentation: LC-MS/MS with electrospray ionization in negative mode.
• Column: reversed-phase C18.
• Matrix: saline that had been in contact with ex-vivo human skin for up to 24 h; calibration performed in saline and validated as a surrogate matrix for “salineskin”.

Sample preparation

• Receptor fluid samples were processed using C18 solid-phase extraction to pre-concentrate analytes and reduce matrix interferences.
• Isotope-labelled internal standards were used where available to correct for losses and ion suppression.

Validation

• Lower limit of quantification (LLOQ): 0.2 μg/L for all analytes; upper limits: 200 μg/L for BPA and 20 μg/L for BPA-G, BPS, D-8 and PF201.
• Linearity: R² ≥ 0.998 across the calibration range.
• Accuracy: within ±8% for low and high quality controls.
• Precision: generally <10% relative standard deviation.
• Matrix effects: negligible for D-8 and PF201; more pronounced for BPA, BPA-G and BPS, yet overall process efficiency remained acceptable (~71–113%) with internal standard correction.
• Stability: all analytes were stable in aqueous solutions and receptor matrix over the study period; PF201 in methanol stock showed marked degradation (~90% loss over six months at −20 °C), requiring frequent stock renewal.

This method allowed robust quantification of very low dermal absorption levels under OECD-compliant experimental conditions.

2.2 Ex-Vivo Human Skin Model

Human skin was obtained from cosmetic or reconstructive surgery with appropriate ethical approval and informed consent. Skin samples were transported on ice and prepared within approximately two hours of excision.

Skin preparation

• Two thicknesses were prepared by dermatome:
  • 200 µm viable skin for permeation kinetics under infinite dose.
  • 800 µm skin, previously frozen at −20 °C, for mass balance experiments.

Skin discs were mounted on flow-through diffusion cells consistent with OECD Test Guideline 428. The receptor compartment was perfused with physiological saline at 32 °C to maintain skin temperature near physiological conditions. Skin integrity was verified (e.g. transepidermal water loss or electrical resistance), and compromised samples were excluded.

2.3 Exposure Design and Test Chemicals

The focus was on colour developers used in thermal papers:

• BPA
• BPS
• D-8
• PF201

In addition, BPA-G was monitored as a possible dermal metabolite of BPA.

Infinite dose experiments (kinetics)

• Donor phase: aqueous solutions of each test compound, applied at 100 µL/cm² to 200 µm viable skin.
• Exposure duration: 24 h, with periodic sampling of receptor fluid.
• Objectives: derivation of flux (J), permeability coefficient (Kp) and lag time (tlag) where steady state could be achieved.

Mass balance experiments

• Donor phase: 100 µL/cm² aqueous solutions applied to 800 µm frozen skin for 24 h.
• After exposure, analytes were quantified in:
  • Donor solution,
  • Surface swabs (cotton),
  • Skin (epidermis + dermis),
  • Receptor fluid.
• The sum of these fractions provided overall recovery and distribution.

2.4 Parameter Study for BPA Permeation

To quantify how methodological choices affect dermal permeation estimates, additional experiments systematically varied key experimental parameters for BPA:

• Skin thickness: 200, 400 and 800 µm.
• Skin state: fresh viable vs previously frozen.
• Donor volume: 10, 100 and 1000 µL/cm².
• BPA concentration: 250, 164 and 18 mg/L.

All other conditions, including diffusion cell type, receptor fluid and sampling schedule, were kept consistent with a prior reference study to enable meaningful comparison.

Permeation data were analysed using a linear mixed-effects model with skin donor as a random effect, isolating the influence of each experimental factor on flux and Kp.

3. Results

3.1 Relative Dermal Absorption of BPA and Alternatives

After 24 h exposure under standardized conditions (aqueous donor, 100 µL/cm², 200 µm viable skin), the fractions of the applied dose recovered in receptor fluid (systemically available fraction) were:

• BPA: ~25%
• D-8: ~17%
• BPS: ~0.4%
• PF201: below the limit of quantification

Mass balance experiments (800 µm skin) revealed additional retention within the skin compartment:

• BPA: ≈15% of applied dose in skin
• D-8: ≈16% in skin
• BPS: ≈11% in skin
• PF201: not reliably quantifiable in skin or receptor fluid

Thus, approximate total entry into skin (skin + receptor) was:

• BPA: ~40%
• D-8: ~33%
• BPS: ~11.4%
• PF201: negligible under the tested conditions

In terms of systemic availability via dermal route (receptor fluid), the ranking was:

BPA > D-8 >> BPS > PF201

These differences are consistent with the physicochemical profiles of the compounds, where higher lipophilicity and appropriate molecular size favour permeation.

3.2 Permeation Kinetics and Steady-State Behaviour

In the kinetic experiments:

• BPS reached steady state within the 24 h exposure, with:
  • Jmax ≈ 0.006 µg/cm²/h
  • tlag ≈ 6.4 h
• D-8 showed:
  • Jmax ≈ 0.032 µg/cm²/h (around five times higher than BPS)
  • tlag ≈ 3.9 h
• For BPA, the applied conditions (100 µL/cm²) were insufficient to achieve a clear steady state within 24 h, indicating that the donor phase behaved closer to a declining or finite dose in this configuration.

Separate experiments with very high donor volume (1000 µL/cm² of 250 mg/L BPA) finally achieved steady state through water, permitting derivation of robust Kp values in the range of 1.6–5.2 × 10⁻³ cm/h, depending on skin thickness. These values are among the most reliable permeability coefficients for BPA in water reported to date and provide a benchmark for future exposure modelling.

For PF201, concentrations in receptor fluid remained largely below quantification, precluding meaningful kinetic analysis.

3.3 Dermal Metabolism of BPA

A critical aspect of dermal exposure is the possibility of local metabolism in the skin before systemic entry. In this work, BPA-glucuronide (BPA-G) in receptor fluid was consistently:

• Detected at very low levels or
• Below the quantification limit (typically <2.5% of BPA levels).

This indicates that first-pass metabolism in skin is limited under the studied conditions. In contrast to the oral route, where a large fraction of BPA undergoes rapid conjugation in the liver, dermal exposure appears to deliver a relatively high proportion of unconjugated, bioactive BPA into the systemic circulation.

3.4 Influence of Experimental Parameters on BPA Permeation

The parameter study demonstrated that experimental design has a profound influence on apparent BPA skin permeation:

• Skin thickness
  • Increasing thickness from 200 to 400 or 800 µm reduced flux (J) approximately three-fold.
  • Thicker skin provides a longer diffusion path and additional barrier, lowering Kp.
• Skin state (viable vs frozen)
  • Fresh, viable skin yielded about 3.4-fold lower flux than frozen skin.
  • Freezing likely alters the stratum corneum structure and barrier properties, artificially enhancing permeation.
• Donor volume
  • Increasing aqueous donor volume from 100 to 1000 µL/cm² increased flux by roughly two-fold, primarily by hydrating the stratum corneum and maintaining a more constant concentration (closer to infinite dose).
  • Smaller volumes (10 µL/cm²) behaved as finite doses, reducing flux.
• BPA concentration
  • A 14-fold decrease in donor concentration produced an approximately 11-fold reduction in flux, indicating that permeation remained approximately proportional to concentration in the tested range.

By adjusting for these parameters, historical BPA permeation studies—whose results initially appeared inconsistent—can be reconciled. Differences of an order of magnitude in reported flux values often reflect variations in skin thickness, freezing protocols, donor volumes and concentrations, rather than fundamental discrepancies in BPA’s intrinsic permeability.

4. Discussion

4.1 Comparing BPA with Its Replacements

The dermal absorption data obtained in ex-vivo human skin show that:

• BPA has the highest systemic penetration, with roughly one quarter of the applied dose crossing the skin within 24 h and an additional fraction retained. Combined with limited dermal metabolism, this supports concerns about occupational exposure for cashiers and workers with frequent receipt handling.
• D-8 shows only moderately lower permeation than BPA. Although the absorbed fraction is somewhat reduced, D-8 reaches the receptor fluid in substantial amounts and accumulates in skin to a similar extent. Given emerging data on its estrogenic potential, D-8 cannot be considered a benign replacement without further toxicological scrutiny.
• BPS exhibits significantly lower penetration into receptor fluid (≈0.4% of the applied dose), but still accumulates in the skin (~11%). This suggests that skin may act as a reservoir, slowly releasing BPS over time. While systemic exposure via dermal route may be reduced relative to BPA, the compound’s intrinsic endocrine activity remains a concern.
• PF201 appears to have very low dermal permeation, but the conclusion is tempered by analytical challenges and potential degradation. Its low solubility and instability in some solvents complicate interpretation. PF201 is therefore promising from a dermal absorption standpoint, but requires further evaluation of its degradation products and systemic toxicity.

From a substitution policy perspective, these results support a nuanced view: moving away from BPA is justified, but some alternatives, particularly D-8, may not yield the magnitude of exposure reduction that decision-makers expect.

4.2 Dermal vs Oral Exposure: Bioavailability of Active BPA

A key insight is the contrast between dermal and oral toxicokinetics:

• Oral intake of BPA undergoes extensive first-pass metabolism in the liver, with only a small fraction of free BPA entering systemic circulation.
• Dermal exposure, by comparison, shows minimal conversion to BPA-glucuronide in skin and delivers a higher proportion of unconjugated BPA to the receptor side, which mimics systemic availability.

Even if total dermal doses (in µg/kg body weight) are lower than typical dietary intakes, the bioactive fraction from dermal exposure may be disproportionately large. For workers handling receipts for extended periods, this route cannot be neglected in cumulative exposure assessments.

4.3 Experimental Standardisation and Literature Consistency

The systematic parameter study also addresses a persistent problem in the BPA literature: widely divergent estimates of skin absorption, sometimes ranging from low single-digit percentages to nearly half of the applied dose.

This work shows that:

• Thinner or frozen skin, combined with large donor volumes and high concentrations, will logically produce higher apparent flux.
• Thicker, viable skin with finite volumes yields lower flux and may not reach steady state, leading to reduced Kp estimates.

When these parameters are accounted for, many apparently conflicting studies converge within predictable ranges. For regulators and risk assessors, this underlines the importance of reporting and harmonising key experimental conditions in dermal absorption studies.

4.4 Limitations and Real-World Relevance

Several limitations must be recognised:

• The donor solutions were aqueous, whereas real exposure involves dry or slightly moist powders embedded in thermal coatings. High volumes (100–1000 µL/cm²) likely overestimate absorption for dry hands but may approximate conditions in a humid or sweaty work environment.
• The receptor fluid was physiological saline rather than complete culture medium, which may under-represent metabolic activity but was necessary for analytical robustness.
• The study focused on single compounds, not full thermal coating mixtures. In real receipts, interactions between developers, dyes, binders and other additives could alter permeability.
• For PF201 and the alternatives, metabolic pathways in skin remain poorly characterised; only parent compounds were quantified.

Despite these constraints, the ex-vivo model and parameter analysis provide a solid quantitative basis for comparing dermal permeation potential and for informing exposure models.

5. Conclusion

Ex-vivo experiments with human skin demonstrate that:

1. BPA exhibits substantial dermal absorption from aqueous donors, with ≈25% of the applied dose reaching the receptor compartment within 24 h and minimal dermal metabolism to BPA-glucuronide.
2. D-8 is only moderately less permeable than BPA, suggesting that its use as a thermal paper developer may not dramatically reduce dermal systemic exposure.
3. BPS permeates more slowly, delivering a low systemic dose but accumulating in skin, which may act as a reservoir. Its intrinsic endocrine potency must still be considered.
4. PF201 shows very low detectable permeation under the tested conditions, but further work is required to understand its stability, metabolites and toxicity profile.
5. Experimental conditions—particularly skin thickness, freezing, donor volume and concentration—can change measured BPA flux by factors of three to ten, explaining many of the discrepancies in published data and highlighting the need for methodological transparency and standardisation.

For industry, regulators and occupational health professionals, these findings make one point clear: a “BPA-free” label on thermal paper does not automatically mean low dermal exposure or low health risk. Substitution must be guided by robust dermal permeation data and toxicological evaluation of both the parent compounds and their metabolites. Only then can the transition away from BPA be considered a genuine improvement in worker and consumer protection.

Last Updated on 12/12/2025 by Jota Machinery