Avobenzone

Photostabilization of Butyl methoxydibenzoylmethane (Avobenzone) and Ethylhexyl methoxycinnamate by Bis-ethylhexyloxyphenol methoxyphenyl triazine (Tinosorb S), a New UV Broadband Filter(para)

ABSTRACT

It is now well documented that chronic UVA exposure induces damage to human skin. Therefore, modern sunscreens should not only provide protection from both UVB and UVA radiation but also maintain this protection during the entire period of exposure to the sun. UVA filters, however, are rare and not sufficiently photostable. We investigated the effect of the introduction of a new UV filter, bis-ethylhexyloxyphenol methoxyphenyl triazine (Tinosorb S), in oil in water sunscreen formulations on the photostability of butyl methoxydibenzoylmethane (Avobenzone [AVB) after irradiation with an optically filtered Xenon arc source (UV irradiance adjusted at 1 mean effective dose [MED]/min). With spectrophotometrical methods to assess the sun protection factor (SPF) and UVA ratio and chromatographical methods to determine the amount of UV filters recovered after irradiation we showed that Tinosorb S prevented the photodegradation of AVB in a concentration-dependent way, leading to a sustained SPF and UVA ratio even after irradiation with doses of up to 30 MED. Since AVB was shown to destabilize ethylhexyl methoxycinnamate (EHM) we tested the effect of Tinosorb S in sunscreens containing this UV filter combination. Here too Tinosorb S showed photoprotective properties toward both UV filters. Thus, Tinosorb S can be used successfully to improve the photostability and efficiency of sunscreens containing AVB and EHM.

^Abbreviations: AVB, butyl methoxydibenzoylmethane (avobenzone); EHM, ethylhexyl methoxycinnamate; HPLC, high-performance liquid chromatography; MED, mean effective dose; SPF, sun protection factor; Tinosorb S, bis-ethylhexyloxyphenol methoxyphenyl triazine.

Received 7 February 2001; accepted 7 June 2001

INTRODUCTION

There is an increasing need for good topical sunscreens to prevent the well-documented damaging effects of UV light on human skin. Sun protection factor (SPF^) grading, based on the erythemal reaction, mainly refers to the acute UVB (290-320 nm) effects and only marginally to those of UVA (320-400 nm). Although a majority of sunscreens provide excellent protection against UVB radiation adequate protection against UVA radiation is increasingly required. Indeed, numerous studies in recent years have highlighted the significant role played by UVA radiation in the sunlight-induced damage and alterations of the skin (1,2), namely, immunosuppression (3), DNA mutations, lipid and protein oxidative damage, enhanced epidermal ferritin expression (4,5), photoaging of the skin and decreased skin hydration and elasticity (6). Through the generation of singlet oxygen, 102, UVA can also induce the expression of matrix metalloproteinases that destroy the connective tissue. In this context the increasing availability of high-SPF sunscreens is a cause for concern because consumers may stay in the sun for a longer period of time. Concomitantly, they are more exposed to UVA radiation. Although there is a large choice of UVB filters good UVA absorbers are rare; they are either poor performing or not sufficiently photostable.

The ideal sunscreen should be such that no photochemical or photosensitizing transformation of its components occurs within the formulation or on the skin. Photochemical stability is indeed the most important characteristic of an effective UV filter since the light-induced decomposition of the sunscreen agent not only reduces its photoprotective power (decomposition into non-UV absorbing molecules) but can also promote phototoxic or photoallergic contact dermatitis (interaction of photodegradation products with sunscreen excipient or skin components like sebum or others, formation of new molecules with unknown toxicological properties, etc.) (7-9). Photoinstability of the sunscreen can also result in the formation of singlet oxygen leading to alterations of the skin (10,11). Upon absorption of damaging UV radiation there is a switch of the UV filter molecule from a ground state (before absorption) to either a singlet (often short lived) or a triplet excited state (longer lived). The excited molecule may reach a stable equilibrium through reversible isomerization or, under certain conditions, return to its original form (ground state), the energy of both excited states can eventually be dissipated as heat, light (fluorescence from the singlet state or phosphorescence from the triplet state) or both or transferred to the surrounding molecules. If the excitation energy cannot be disposed off by energy transfer or by emission of light chemical bonds of the absorber molecule may break or rearrange (photoaddition/substitution, cycloaddition photofragmentation).

UVA filters that fulfill both efficacy and photostability requirements are rare. Benzophenones are very stable molecules but their UVA absorption ability is poor. Butyl methoxydibenzoylmethane (Avobenzone [AVB]) has a strong absorption in the UVA range at 360 nm but is rapidly destroyed after irradiation (12). Following this photoreaction AVB can react with ethylhexyl methoxycinnamate (EHM) to form cycloaddition products thus destabilizing the otherwise photostable cinnamate molecule (13). Since some filters affect the stability of others the resulting stability may change in different filter combinations. A good example is the stabilization of AVB by octocrylene or methylbenzylidene camphor (14). On the other hand, this combination becomes less stable when it is combined with EHM. Hence, in order to enhance the effectiveness and safety of this sunscreen agent there is a need for new systems exhibiting improved AVB photostability. Other chemicals are being used in sunscreens to try and photostabilize AVB. Bonda and Steinberg (15) successfully used diethylhexyl 2,6-naphtalate, a chemical that stabilizes AVB. The inclusion of AVB in hydroxypropyl-beta-cyclodextrin both in solution and in the solid state enhanced the photostability of the sunscreen agent and reduced its photodecomposition (16).

The aim of our work was, therefore, to look at the properties of a new broadband UV filter, bis-ethylhexyloxyphenol methoxyphenyl triazine (Tinosorb S), with regard to its possible photostabilizing effect on AVB and the combination AVB/EHM. For this purpose numerous combinations of AVB and other UV filters in formulated model sunscreens have been examined.

MATERIALS AND METHODS

Materials. Tinosorb S was obtained from CIBA Specialty Chemicals Inc. (Basel, Switzerland), AVB (Parsol' 1789) and EHM (Parsol MCX) from F. Hoffmann-La Roche Ltd. (Basel, Switzerland). Formulations with varying amounts of the above mentioned UV filters in a oil in water emulsion were used during this study. The vehicle consisted of water, carbomer, trometamol, sorbitol 70%, polysorbate 20, sorbitan laurate, cetearyl alcohol, dicapryllyl maleate, dimeticone 350 PHE, C12-15 alkyl benzoate, methyl and propyl parahydroxybenzoate; the amount of water was adjusted according to the quantity of UV filters.

Sunscreen photostability assessment. The method used to assess the photostability of the different sunscreen formulations was originally described by Diffey et al. (17) and later used by Sayre and Dowdy (12). Briefly, the sunscreen to be investigated was applied onto two quartz plates with roughened surfaces at a dose of 2 mg/ cm^sup 2^. Fifteen minutes after application one plate was irradiated with UV light from a Xenon arc source (maximal total UV output of 55 mW/cm2) optically filtered to achieve a spectrum simulating summer sunlight (Multiport Solar UV simulator model 501, Solarlight, Philadelphia, PA). The other plate remained nonirradiated and was kept inside a dark oven set at 30 deg C. Four liquid light guides bound together were used to irradiate a total area of 4.5 cm^sup 2^. The UV radiation emitted from the liquid light guides of the Multiport was measured with a radiometer (Model DCS-600, Solarlight) calibrated to read the erythemally weighted irradiance directly and adjusted at I MED/min. After a total irradiation equivalent to 30 MED (1 MED = 25 mJ/cm^sup 2^)-this dose was chosen on the basis of a SPF of 25 for the sunscreen product containing the three UV filters at their maximal concentration as measured in preliminary experiments. The spectral transmission of both irradiated and nonirradiated samples was measured using a spectrometer SPEX1681 (GlenSpectra, Middlesex, UK); the SPF of the different sunscreen formulations was calculated as described by Diffey and Robson (18). This in vitro measurement was completed by the determination of the UVA ratio (19). The method required by the Australian authorities was also performed to assess if the tested sunscreens met the `AS/NZS 2604: 1997' criteria (20). To determine the amount of UV filters degraded after the irradiation period (30 min) the sunscreens were solubilized from the control and irradiated quartz plates, and the amount of UV filters remaining was quantified by high-performance liquid chromatography (HPLC). Each measurement was performed in triplicate.

UV filters determination by HPLC. For AVB and EHM quantification the sunscreens were solubilized in methanol and analyzed using a Nucleosil C18HD-3 column, at a temperature of 25 deg C, and methanol/H^sub 2^O/CH^sub 3^COOH (83/17/0.01) as a mobile phase with a flow rate of 0.2 mL,/min. Ten microliters of the sample was injected and detection occurred at 300 nm using a diode-array detector. For Tinosorb S quantification the sunscreens were solubilized after irradiation in dioxane and analyzed using a Hypersil 120-5 ODS column at a temperature of 40 deg C, using a 80% dioxan/20% sodium acetate, pH 4.6, mobile phase at a flow rate of 1 mL/min. Ten microliters of the sample was injected and detection occurred at 320 nm using a diode-array detector. Calibration curves were established using peak areas.

RESULTS

Sun protection factor

SPF, in vitro, was measured for all sunscreens before irradiation. For sunscreens containing only AVB (2.5 or 5%) the SPF was very low ranging between 2.5 and 4. As expected, addition of either EHM or Tinosorb S or both boosted the SPF values to 20-50 on an average (data not shown).

After irradiation the SPF values measured for the formulations containing no Tinosorb S were clearly lower than the control values (Fig. 1). A sharp decrease of the SPF value after irradiation was observed which was more important when EHM was present in the sunscreen. When Tinosorb S (5 or 10%) was included in the formulations no differences between the control SPF values and SPF values after irradiation of the sunscreens were observed (data not shown). The SPF remained stable even after 30 MED irradiation.

UVA ratio

As for the SPF the UVA ratio was determined for all the sunscreens. All sunscreens tested, with the exception of formulations containing only 2.5% Avobenzone, met the AS/ NZS criteria. For formulations containing AVB the UVA ratio was clearly higher than 1, ranging between 1.6 and 1.8, reflecting the fact that AVB is a UVA filter with very small UVB filtering capacity. The addition of the UVB filter, EHM, in the formulations reduced the UVA ratio to 0.8-0.9. The subsequent addition of Tinosorb S enhanced again the UVA ratio to values close to 1.0 (0.95-1.00).

After irradiation with 30 MED the UVA ratio of the formulations containing only Avobenzone were decreased to 30-40% of the control value (Fig. 2). In preliminary studies we have observed a dose-dependent decrease of the UVA ratio (data not shown). For sunscreens containing the combination AVB/EHM, a decrease of 20-40% of the UVA ratio compared to the control value was observed. However, as soon as Tinosorb S was included in the formulations the UVA ratio remained unchanged after irradiation in comparison to the control plates (Fig. 2).

Recovery of the UV filters after irradiation

As shown in Fig. 3 irradiation of the sunscreens containing only AVB led to a great decrease in the amount of UV filters originally applied; between 56 and 70% of AVB was lost following irradiation. When EHM was present in the formulations the amount of AVB recovered was comparable to the formulations with AVB alone. In all the cases, the addition of Tinosorb S in the sunscreens prevented or greatly diminished the photodegradation of AVB. This prevention was higher in sunscreens containing only AVB (5-15% degradation) compared to sunscreens containing the combination AVB/EHM (17-35% degradation). The prevention of the photodegradation of AVB, in formulations containing only AVB, was already maximal at a concentration of 5% Tinosorb S in the formulation. An increase to 10% Tinosorb S in the sunscreen induced only a marginal increase of the photoprotection of AVB (Fig. 4). With regard to the AVB concentration the photodegradation prevention by Tinosorb S was similar in formulations containing either 2.5 or 5% AVB (Fig. 3).

If one looked at the amount of EHM recovered after irradiation only 35% of the applied dose was found. However, in the presence of Tinosorb S 65% of EHM were recovered after irradiation. When both AVB and EHM were present in the formulations Tinosorb S showed a clear concentrationdependent protective effect with regard to the amount of AVB and EHM recovered after irradiation of the sunscreens (see Fig. 5 for sunscreens containing 5% AVB and 5% EHM). The amount of EHM recovered after irradiation in the presence of Tinosorb S was always higher when AVB was not present in the sunscreens-70-78% in comparison with 55-68%-and there was also a clear concentration-dependent effect of Tinosorb S (Fig. 6).

DISCUSSION

We have shown that sunscreens containing AVB experience photodegradation from UV exposure leading to decreased SPF and UVA protection as measured by the UVA ratio. The latter decreased to values less than 1 showing that the spectral characterization of the products has been modified toward a clear decrease of the UVA absorbance. This was expected and confirms results from others (17). Indeed, it is now well documented that dibenzoylmethanes lose much of their UV protective capacity after UV irradiation through tautomerization, fragmention and formation of new products with distinctly altered UV absorption characteristics (21,22). Photodimerization of dibenzoylmethanes has been observed by Dubois et al. (23) and photooxidation was proposed to be a major photodegradation route for AVB (24). Similar to the amount of AVB we recovered after irradiation by 30 MED, Deflandre and Lang (27) have observed photodegration rates of AVB of up to 50% after only I h of irradiation albeit at very high energy (25). Although AVB has been shown to be relatively photostable in the polar protic solvent isopropanol it was relatively photolabile in the nonpolar solvent cyclohexane (21,26). Moreover, it was shown that the photodegradation of AVB strongly depends on the presence of the 1,3-dik-eto forms. In contrast to benzylidene camphor and some other dibenzoylmethane derivatives where a photostationary equilibrium mixture of Z/E isomers in these classes of sunscreen exhibit an excellent photostability under solar simulated irradiation (25,27), enol-keto tautomerization seems to be the primary mechanism for the lability of AVB.

When we incorporated EHM, normally known as a photostable UVB filter, together with AVB we observed a similar decrease of the SPF and UVA ratio as well as a low recovery of the UV filters after irradiation. Following irradiation the relationship between UVA and UVB absorbance was modified toward less UVA or more UVB. Since the SPF concomitantly decreased a breakdown of the of UVA filtration is more likely. In this case, the loss of absorption was proportional to the amount of UV filters recovered. EHM alone undergoes cis/trans isomerization (28) mechanism that cannot really be considered as photoinstability but rather a very efficient way of dispersing the absorbed energy. Therefore, the loss of EHM due to photoisomerization is generally low, and its effect upon the SPF is expected to be minimal as compared to a more photostable UVB filter. However, the cis isomer, although absorbing at the same wavelength, has a reduced extinction, thus giving lower spectrophotometric values. This absorption loss was reported to range from 4.5 to 50% (28). It was also reported that when the material is exposed to radiation of wavelength greater than 300 nm trans-2-EHMC not only photoisomerizes to the cis isomer but also undergoes self dimerization by means of a [2 + 2] cycloaddition reaction across the ethylenic double bond (21,29). Moreover, following irradiation AVB and EHM react with each other to form cycloaddition products and perhaps other photoadducts. All these considerations would explain the lower amount of EHM we recovered after irradiation. Comparable loss in the efficacy of sunscreens was reported by Diffey et al. (17) who showed that UV filtering systems containing AVB/EHM suffered protection efficacy losses of up to 50-60% of their initial value. In their case, the addition of known photostabilizing UV filtering systems did not prevent these reductions in protection efficacy.

Under our experimental conditions, however, we found that Tinosorb S abolished or greatly reduced the photochemical decomposition of AVB alone or in combination with EHM in the formulations, and ensured that almost no change of the in vitro protective properties (SPF and UVA ratio) of the test sunscreens occurred. Further, it prevented the destabilization of EHM by AVB following irradiation. Protection of AVB photochemical decomposition by Tinosorb S was concentration dependent. The weight ratio or molar ratio between the stabilizing chromophore and AVB has been shown to be crucial (25). As shown in Fig. 4 a, minimal concentration of the stabilizer is required for an optimal result and increasing its concentration does not improve photostability. This is in agreement with the mechanism of triplet-triplet energy transfer where the determining concentration is the concentration of the acceptor. AVB has a triplet energy of 59.5 kcal/mol (30) and a triplet life time of the order of 0.4 (mu)s, long enough for efficient quenching by a suitable acceptor. Many organic molecules do have an adequate triplet energy of the order of 55-59 kcal/mol but not all of them can be considered as candidates for stabilizing AVB due to the lack of photostability per se. Only 4-methylbenzylidene camphor and octocrylene fulfil these criteria (among registered UV filters). Here we showed that Tinosorb S might be another potential candidate. Indeed, Tinosorb S has been shown to be a very photostable UV filter, its recovery (7% active ingredient in aqueous suspension, 2 liL/cmz on a rough quartz plate) was higher than 99% after irradiation by 10 MED and higher than 95% after irradiation by 50 MED (31). Given its molecular symmetry, the presence of aromatic rings conjugated with carbonyl groups and electron-- releasing group such as hydroxyl group substituted on the aromatic rings Tinosorb S possesses an optimal structure for energy dissipative processes allowing electron resonance delocalization upon absorption of a photon, and is most probably able to deactivate sensitizers through energy transfer (triplet-triplet energy transfer) leading to the isomerization of the acceptor (reversible photoisomerization and deactivating capacity). To return to its ground state it can efficiently dissipate the accepted energy through intramolecular hydrogen transfer in the excited state followed by internal conversion and thermal deactivation (32). Another possibility is that Tinosorb S, as a broadband UV filter, will absorb at any wavelength a substantial part of the incident monochromatic energy proportional to its contribution to the total absorbance at this wavelength. Thus, all the photons absorbed by Tinosorb S are not available for absorption by AVB or EHM, and consequently AVB appears more stable because it is partially shielded by Tinosorb S without any interactions at a molecular level. AVB being thus stabilized it is not able to affect EHM anymore. This would be a filter effect or energy sharing. The exact mechanism by which Tinosorb S protects AVB and EHM from photodegradation remains to be elucidated.

In conclusion, commercial sunscreens containing the combination of AVB and EHM UV filters have been shown to be photolabile to various degrees. A common feature of these sunscreens is that they begin with SPF higher than that claimed on the label; however, as their exposure to sunlight increases the UV absorbance of these sunscreens decline. Since during SPF measurements in vivo, photochemical changes during exposure are automatically taken into account, one cannot differentiate between a constant SPF and a decreasing SPF over irradiation time. A way to avoid these problems is to incorporate photostable UV filters in sunscreens. We have shown that the incorporation of Tinosorb S in sunscreens because of its high photostability, its broadband filtering capacity and its photostabilizing effect toward AVB and EHM is a powerful means to provide optimal sunscreens.

Acknowledgements-We thank the Pharmaceutical Technology Department of Spirig Pharma Ltd for providing us with the different formulations tested and Ms. C. Kissling, Ms. S. Ludi and Ms. E. Bieli for their excellent technical assistance.

(para)Posted on the website on 26 June 2001.

REFERENCES

1. Scharfetter, K., M. Wlaschek, A. Hogg, K. Bolsen, A. Schothorst, G. Goerz, T. Krieg and G. Plewig (1991) UVA irradiation induces collagenase in human dermal fibroblasts in vitro and in vivo. Arch. Dermatol. Res. 283, 506-511.

2. Seite, S., D. Moyal, S. Richard, J. de Rigal, J. L. Leveque, C. Hourseau and A. Fourtanier (1998) Mexoryl(R) SX: a broad absorption UVA filter protects human skin from the effects of repeated suberythemal doses of UVA. J. Photochem. Photobiol. B: Biol. 44, 69-76.

3. LeVee, G. J., L. Oberhelman, T. Anderson, H. Koren and K. D. Cooper (1997) UVA II exposure of human skin results in decreased immunization capacity, increased induction of tolerance and a unique pattern of epidermal antigen-presenting cell alteration. Photochem. PhotobioL 65, 622-629.

4. Kielbassa, C., L. Roza and B. Epe (1997) Wavelength dependence of oxidative DNA damage induced by UV and visible light. Carcinogenesis 18, 811-816.

5. Stary, A., C. Robert and A. Sarasin (1997) Deleterious effects of ultraviolet A radiation in human cells. Mutat. Res. 383, 1-8.

6. Lavker, R. M., D. Veres, C. J. Irwin and K. H. Kaidbey (1995) Quantitative assessment of cumulative damage from repetitive exposures to suberythemogenic doses of UVA in human skin. Photochem. Photobiol. 62, 348-352.

7. DeLeo, V. A., S. M. Suarez and M. J. Maso (1992) Photoallergic contact dermatitis. Arch. Dermatol. 128, 113-118.

8. Rieger, M. M. (1997) Photostability of cosmetic ingredients on the skin. Cosmet. Toil. 112, 65-72.

9. Schrader, A., J. Jakupovic and W. Baltes (1994) Photochemical studies on trans-3-methylbutyl 4-methoxycinnamate. J. Soc. Cosmet. Chem. 45, 43-52.

10. Allen, J. M., C. J. Gosset and S. F. Allen (1996) Photochemical formation of singlet molecular oxygen (^sup 1^O^sub 2^) in illuminated aqueous solutions of p-aminobenzoic acid (PABA). J. Photochem. Photobiol. B: Biol. 32, 33-37.

11. Allen, J. M., C. J. Gosset and S. F. Allen (1996) Photochemical formation of singlet molecular oxygen (^sup 1^O^sub 2^) in illuminated aqueous solutions of several commercially available sunscreen active ingredients. Chem. Res. Toxicol. 9, 605-609.

12. Sayre, R. M. and J. C. Dowdy (1999) Photostability testing of Avobenzone. Cosmet. Toil. 114, 85-91.

13. Bonda, C. and P. Marinelli (1999) The photochemistry of sunscreen photostability, pp. 46-51. Step Publishing Ltd., UK, European UV Sunfilters International Conference Proceedings, 34 November 1999, Paris.

14. Wunsch, T. and H. Westenfelder (1998) New aspects in sunscreens, pp. 56-60. Step Publishing Ltd., UK, European UV Sunfilters International Conference Proceedings, 17-18 November 1998, Paris.

15. Bonda, C. and D. C. Steinberg (2000) A new photostabilizer for full spectrum sunscreens. Cosmet. Toil. 115, 37-45.

16. Scalia, S., S. Villani, A. Scatturin, M. A. Vandelli and F. Forni (1998) Complexation of the sunscreen agent, butyl-methoxydi

benzoylmethane, with hydroxypropyl-beta-cyclodextrin. Int. J. Pharm. 175, 205-213.

17. Diffey, B., R. P. Stokes, S. Forestier, C. Mazilier and A. Rougier (1997) Suncare product photostability: a key parameter for a more realistic in vitro efficacy evaluation. Eur. J. Dermatol. 7, 226-228.

18. Diffey, B. and J. Robson (1989) A new substrate to measure sunscreen protection factors throughout the ultraviolet spectrum. J. Soc. Cosmet. Chem. 40, 127-133.

19. Stokes, R. (2000) Project sunscreen protection. In Dermatopharmacology of topical preparations: a product-development oriented approach (Edited by Gabard, Elsner, Surber, Treffel), pp. 365-382. Springer, Berlin.

20. Australian/New Zealand Standard (1993) 15/NZS 2604.

21. Schwack, W. and T. Rudolph (1995) Photochemistry of dibenzoyl methane UVA filters Part 1. J. Photochem Photobiol. B: Biol. 28, 229-234.

22. Tarras-Wahlberg, N., G. Stenhagen, O. Larko, A. Rosen, A.-M. Wennberg and O. Wennerstrom (1999) Changes in ultraviolet absorption of sunscreens after ultraviolet irradiation. J. Int. Dermatol. 113, 547-553.

23. Dubois, M., P. Gilard, P. Tiercet, A. Deflandre and M. A. Lefebvre (1998) Photoisomerisation of the sunscreen filter PARSOL(R) 1789. J. Chim. Phys. 95, 388-394.

24. Roscher, N. M., M. K. O. Lindemann, S. B. Kong, C. G. Cho and P. Jiang (1994) Photodecomposition of several compounds commonly used as sunscreen agents. J. Photochem. Photobiol. A: Chem. 80, 417-421.

25. Deflandre, A. and G. Lang (1988) Photostability assessment of sunscreens. Benzylidene camphor and dibenzoylmethane derivatives. Int. J. Cosmet. Sci. 10, 53-62.

26. Andrae, 1. (1997) A UVA filter (4-ter-butyl-4'-methoxydibenzoylmethane): photoprotection reflects photophysical properties. J. Photochem. Photobiol. B: Biol. 37, 147-150.

27. Deflandre, A. and G. Lang (1988) Photoisomerization of benzylidene camphor and derivatives. Cosmet. Toil. 103, 69-75.

28. Morliere, P., O. Avice, T. S. Melo, L. Dubertret, M. Giraud and R. Santus (1982) A study of the photochemical properties of some cinnamate sunscreens by steady-state and laser photolysis. Photochem. Photobiol. 36, 395-399.

29. Broadbent, J. K., B. S. Martincigh and M. W. Raynor (1996) Capillary supercritical fluid chromatography combined with atmospheric pressure chemical ionization mass spectrometry for the investigation of photoproduct formation in the sunscreen absorber 2-ethyl-p-methoxycinnamate. J. Chromatogr. A 732, 101-110.

30. Gonzenbach, H., T. J. Hill and T. G. Truscott (1992) The triplet energy levels of UVA and UVB sunscreens. J. Photochem. Photobiol. B: Biol. 16, 377-379.

31. Tinosorb S a UV broadband absorber for personal care products, Technical Bulletin, CIBA Specialty Chemical Inc., Basel, Switzerland, 1999.

32. McGarry, P. F., S. Jockusch, Y. Fujiwara, K. Kaprinidis and N. J. Turro (1997) DMSO solvent induced photochemistry in highly photostable compounds. The role of intermolecular hydrogen bonding. J. Phys. Chem. A 101, 764-767.

Eric Chatelain* and Bernard Gabard

Department of Biopharmacy, Spirig Pharma Ltd, Egerkingen, Switzerland

*To whom correspondence should be addressed at: Department of Biopharmacy, Spirig Pharma Ltd, CH-4622 Egerkingen, Switzerland. Fax: 41-62-387-8790; e-mail: eric.chatelain@spiri.ch

Copyright American Society of Photobiology Sep 2001
Provided by ProQuest Information and Learning Company. All rights Reserved