17 minutes reading time (3330 words)

    Sunscreens: Molecules That Absorb Ultraviolet Radiation

    Brian Wilkins

    Wearing hats and sitting under trees does not protect us from the significant amounts of ultraviolet radiation (UVR) reflected from most surfaces. We need extra protection in many situations. Chemistry has a role to play in providing harmless molecules that absorb UVR and dissipate its potentially dangerous energy into harmless longer wavelengths, the warmth of infrared.

    Using human skin to find the amount of UVR that may reach it through a sunscreen film would seem like a rough and risky procedure. Yet this is the most highly regarded method for testing sunscreens. The test depends on the presence or absence of a mild burn (erythema), and the amount of UVR needed to cause that burn. To measure this minimal erythemal dose (MED) of UV, adjacent areas of skin are exposed to successively longer periods of irradiation from a xenon lamp, each period usually being 1.26 times the previous period. The ratio of the irradiation time required to produce erythema on the skin covered with a 2mg per sq cm film of sunscreen, to the time for an equivalent reddening of untreated skin, is defined as the sun protection factor (SPF), or more correctly, as the initial SPF.

    Because burning is only one harmful effect among many, such as accelerated skin ageing, troublesome or fatal cancer, and suppression of the immune system, we would like to know whether or not sunscreen tests based on burning give us a true picture of the protection against all these other dangers. The evidence so far is that they don't, and that the early stages of cancer can be seen in cells exposed to amounts of UVR much less than those needed to cause even the mildest burns. However the sunburn region of the spectrum does coincide pretty well with the cancer-producing part of the spectrum. What we would like to know more about is how much protection we need from those wavelengths.

    UV absorbers employed in sunscreensIrradiation of human skin with narrow bands of UV shows that their ability to produce erythema rises sharply as the range from 320nm to 290nm is traversed. This, the main sunburn region, and the region of most damage, is called the UVB region (figure 1).

    Shorter wavelengths are present outside the atmosphere but are screened out, mainly by ozone and oxygen. Combining the erythemal sensitivity of skin and the intensity of this radiation at the surface of the earth, we can attribute most erythema to a fairly narrow band between 295nm and 320nm, peaking at about 307nm.

    The erythemal sensitivity of human skin at longer wavelengths, 320nm to 400nm, peaks at about 380nm, but is less than that in the UVB region by a factor of one thousand to several thousand.

    In figure 2 are drawn some types of structures that give UV absorbance. All have alternating double and single bonds. These are called conjugated systems and have overlapping Π orbitals. Π electrons are promoted to a higher energy level during the UV absorption process, which is called a Π→Π* transition. 

     The hydrocarbon chain on the right of compound A is not part of the conjugated system but is designed to make it more soluble in the oily components of the sunscreen.

    We compare UV-blocking ability by dissolving each compound in a solvent (which itself does not absorb UV) and placing the solution in a small cell (called a cuvette) made of a material (normally silica) that allows UV to pass through it.

    So that we can compare different compounds we need to be able to express UV-blocking as a number, the most common unit being the absorbance (A) which is related to the intensity of the radiation entering the solution (IO) and the intensity leaving the solution (I).

    A = log10(IO / I)

     Measurement of the absorbance is carried out in a instrument called a spectrophotometer, which allows us to select a series of narrow bands of wavelengths and pass them through the solution. The absorbance of most dilute solutions (where there are no complications such as changing ionization) is directly proportional to concentration, a relationship referred to as Beer's Law.

    In practical terms, A is often expressed as the standard absorbance, calculated for one percent solutions measured in a cuvette where the beam passes through 1cm of solution. Alternatively, when the molecular mass is known, we can use the term molar absorptivity (ε)

    ε = A / cl

    where c is the concentration in moles per litre, and l the path length of the beam through the solution in cm. Since the curves in Figure 1 were all measured at the same concentration we can compare their effect at a glance.

    Look at the curves in Figure 1 and decide which compounds would protect better in the UVA region and which would be better in the UVB. To obtain protection in both regions and produce a so called "broad spectrum" sunscreen we need a combination of two or more compounds. However, those with the highest curves are not necessarily the best to use in sunscreens. Maybe the price is too high, and it is better to use more of a cheaper compound. Or perhaps the compound is not soluble enough in the sunscreen base ingredients.

    A natural constituent of human cells, with a UV absorption peak coinciding with the erythemal sensitivity peak, and possessing a high standard absorbance value, would seem to be the ideal sunscreening agent. Therefore when it was realised that para-aminobenzoic acid (PABA) met these requirements the place of PABA was assured in sunscreens for many years. Chemical sunscreens have been in use for 60 years and PABA has been popular for about 40 years, but it has now disappeared almost entirely from use. It is useful to look at the reasons for this because they tell us that UV absorption is not the only requirement for acceptability.

    PABA is oxidisable under the conditions of use. It forms a brown compound that stains cotton fabrics. It is also relatively insoluble in the lipophilic (oil-loving) components of emulsions and cosmetics, and when administered in the rather inconvenient vehicle of an ethanolic solution, is likely to crystallise on the skin.

    When the related dimethylaminobenzoate ester (A of Figure 2) is employed thesef problems are overcome and it is now in widespread use. However A is not accepted by all manufacturers because it irritates the skin of some users.

    Chemists have not been slow in drawing on existing compounds and making new ones. Nearly all of the compounds in use today have been employed in sunscreens for 25 years or more. All have been tested for toxicity.

    The UVB is well catered for, but because of the relatively late attention being given to the possible dangers from UVA radiation, and the continuing lack of agreement concerning the extent of that danger, the number of UVA blockers is very limited. The dibenzoyl methane derivatives (eg. C) are the most effective on a weight for weight basis, but C (trade name Parsol 1789) is likely to face competition from the less costly microfine titanium dioxide (TiO2). C has not been approved for general use in the USA.

    The benzophenones such as D (generic name oxybenzone) were the original "broad spectrum" UVA absorbers, but their standard absorption in the region is only moderate (Figure 2) and they no longer dominate the field. They are used extensively to slow down the deterioration of paint.

    How does UVR cause cancer? UV damage to thymine, one of the base pairs in DNA via thymine dimerization (two thymine structures joining together) is well documented as a potential precursor to skin cancer. UVR is also known to produce potentially damaging free radicals (short lived reactive structures with unpaired electrons) in the skin. Much remains to be discovered in this area.

    The particulate UV blockers
    Sunscreens based on TiO2 and zinc oxide (ZnO) have retained, until recent years. an aura of total impenetrability; the proverbial brick wall. If users were prepared to look like white-painted circus clowns it was believed that they could achieve total protection. But the reality was not as clear cut. If used in the same concentration and film thickness as the visibly transparent sunscreens, these so-called opaque agents were not nearly as effective as organic-based products. Even when plastered thickly on the skin, they frequently melt away rapidly in the heat.

    The pigment grade TiO2 and ZnO used in these products normally had a particle size of at least 0.25 microns. Theory predicts a different cause of scattering for particles of this size compared to finer particles. Particles in the 0.02 - 0.05 micron range display a critical difference in behaviour in that they act by UV absorption as well as UV scattering and are relatively transparent in the visible range. Absorption curves of the fine TiO2 (E) compared to the coarser TiO2 (H) in Figure 1 show the advantage of smaller size. TiO2 is a semi-conductor, and UV radiation excites an electron from the valence electronic band to the conduction band.

    Several brands of microfine TiO2 and ZnO are now on the market. The near transparency in the visible region eliminates the undesirable whiteness on the skin- except when high percentages are used. The UV absorbance of microfine TiO2 is better in the UVB region than in the UVA, but not as effective on a weight-for-weight basis as is the organic UVA blocker C. Microfine TiO2 on its own can provide very adequate UVA protection.

    Nephelolometry, where insoluble suspended particles are shown to exhibit Beer's Law-type€ behaviour in the spectrophotometer, has a useful part to play in sunscreen development.

    Spectrophotometric analysis
    Routine spectrophotometry in the range 295-400 nm is the principal method of quality control. of great interest is the question of whether those results enable us to predict the performance of a sunscreen in use. But first we need to consider the behaviour of sunscreens on the skin.

    Sunscreens vary in their ability to maintain an intact protective layer on, or partly in, the skin. This property is called substantivity and depends on a totality of properties related to all the constituents of the formulation. Substantivity is not related to UV absorption except insofar as some of the UV-active compounds may influence substantivity through other properties, (eg. water repellency) that they may possess.

    Substantivity and SPF measurement are inevitably linked to some extent because, other factors being equal, the longer the irradiation time, whether it be seconds under a lamp or hours in field testing, the lower will be the recorded SPF because of the progressive deterioration of the protective film. Any information we obtain from spectrophotometry can therefore relate only to the initial SPF that has been determined under artificial lamps with short exposure times.

    Absorbance expressed as the absorbance of a film of sunscreen 10 microns in thickness is an appropriate parameter to relate to SPF. Correlations, useful in sunscreen design up to about SPF 36, have been published most recently by Wilkins (Figure 3). The literature contains mistakes to be avoided. Sandwich techniques, where a very thin film of sunscreen is held between two quartz plates, involve such a small amount of material being held in the beam that the sampling is inadequate and the results on emulsions, which by their nature are not homogeneous, are inconsistent. Furthermore in the case of an emulsion, the unbroken emulsion is not the form in which the sunscreen resides after being spread on the skin. 

    The following method, using normal solution spectrophotometry, allows the calculation of the absorbance of a neat film of the sunscreen. Ten microns is the chosen thickness. Twenty microns is the thickness applied to skin for in vivo testing, but experience indicates that users generally apply less than that. Skin being a highly uneven surface, has a proportion of area covered by even less than 10 microns.

    Method: About 50 mg of sunscreen is weighed into a 50 mL volumetric flask and dissolved in solvent (Universal Sunscreen Solvent). Ultrasonication, or at least very thorough shaking, is essential when particulate sunscreening agents are present. 3mL is transferred to a 25mL volumetric flask and diluted to volume with USS. The absorbance (A) is measured in a 1cm cell.

    A (calculated for 10 microns film) = A (Found) / Final concentration in mg/mL

    The most useful wavelength at which to apply this method is 307 nm, because it lies at the centre of the erythema-producing UVB region. The peak height of most UV-active compounds in the UVB is roughly proportional to peak area, but not so for the flattish curves of the particulates.

    The limitations of SPF testingThe spectrophotometer reacts to the same radiation that burns the skin. We would expect the results to relate to each other, and they do. Where substantivity is good, spectrophotometry also relates well to field tests where users display themselves to the sun with weighed amounts of sunscreen carefully applied to measured areas of skin.

    Although the xenon lamps employed in in vivo indoor testing imitate the spectral range and energy of the sun, the following results of Pathak ring the alarm bells.

    We can understand this rapid fall-off in protection to half or less during use only by analysing the time factor as it applies to artificial, as compared to natural, methods of testing.

    Contrary to much of the advertising copy we read, the sun protection Factors (SPF) does not tell us how long we can stay out in the sun without burning. This is because the tests to measure SPF are carried out with UV lamps about a hundred times the intensity of sunlight and are completed in such a short time that the all-important time factor in the protection afforded by the sunscreen is ignored.

    I can illustrate this point with typical data from a recent test with which I was associated. One of the subjects in the testing panel required 10 seconds exposure to the xenon lamp to receive a detectable burn. The same subject, with the standard amount (2 mg/sq cm) of sunscreen on the skin, required 240 seconds of exposure to produce the same redness. The SPF is the ratio of these two times, 240:10 = 24.

    Since the exposure time is a mere four minutes this type of test could tell us only about the protection we get beyond that time if the protective film remains fully intact throughout the longer period. In fact it does not, and it has been known for twenty years that the protection afforded by sunscreens falls off with time.

    The SPF therefore tells us only what we start with. A more useful application of the xenon-lamp burn technique is to find the SPF at the end of a specified period rather that at the beginning. In this way we can check on the all-important lasting qualities, which vary tremendously among the different products on the market. Very few commercial sunscreens carry information on this final SPF.

    The latest blow to the once sacred dictum that SPF 15 represents maximum protection and that suberythemal doses of UV are safe, comes from research on immunosuppression. A rapidly growing body of work now attests to the ability of UVB radiation to suppress vital immune reaction in animals and in man. Van Praag et al. state that their "results suggested that the minimal erythemal dose is not an accurate method to determine protection against UV-induced immunologic alteration". The mediocre performance of certain commercial sunscreens as reported in work in this area is also a cause for concern.

    Designing a sunscreen
    Good sunscreen design requires that we put enough UV absorber on the skin and ensure that it stays there during the period of use. The focus on initial SPF and the relative lack of attention to substantivity has resulted in many inadequate products.

    Thirty years ago it was thought that because a day's solar radiation equated to about 15 MED then an SPF of 15 provided maximum protection. Since SPFs drop in value during use, then if we continue to use formulations of doubtful substantivity it would be prudent to begin with SPFs greater than 15. Better substantivity also ensures a more efficient use of the UV-actives.

    Because most sunscreens are emulsions, formulators are faced with the limitation that the emulsifying surfactants (soap-like agents) ending up on the skin will act against good substantivity in the presence of water, including sweat.

    Designing for good substantivity has involved formulating emulsions containing water-repellent polymers or avoiding emulsions altogether.

    The total solar energy (irradiance) in the UVA region reaching the surface of the Earth is about three times that in the UVB. Why then is there a lack of agreement on how much protection we need against it, or in fact whether we need to be protected from it at all?

    For the chemist, the Grotthus-Draper Law is a good place to start. This tells us that only radiation absorbed by a system is effective in producing photochemical change. We also know that absorption of radiation depends on molecular structure. The total quantity of radiation present per se is therefore irrelevant.

    We know that the biological effects of UVA are relatively mild compared to those of UVB. Rather than demonstrating the danger from UVA, the deeper penetration of UVA shows that it is not being absorbed very strongly by the skin tissue. If it is not absorbed it is having no effect; no more effect than when we shine a torch light through our hand. UVB and UVC radiation penetrates less because it is absorbed by some of the cell constituents to produce many harmful biochemical effects.

    Nevertheless UVA is absorbed to some extent and does have a number of effects including erythema production, pigmentation, dermal connective tissue alteration, and tumour production in animals. Phototoxic and photoallergic reactions have also been reported. But the weakness of these effects compared to those associated with UVB has prevented agreement on the amount of protection, if any, we need against UVA.

    An extreme view is that taken by the Garland brothers, who claim that by using UVB-blocking sunscreens with poor UVA-blocking ability we are encouraged to stay longer in the sun and therefore suffer harm from UVA. In support of this argument they quote the rising rates of melanoma in countries where sunscreens have been in use for many years. A more mundane explanation does not seem to have occurred to them. This is simply that the figures may merely be demonstrating the inadequacy of most commercial sunscreens.

    A more extended treatment of some of the issues discussed in this paper are available in the paper "Sunscreens and the Human Photometer" published in Chemistry in New Zealand, November, 1993 pp7-12.


    1. Pathak, M., Fitzpatrick, T., Parrish, J., Mosher, D., and Greiter, F., Sunscreens: basic and clinical aspects of photo-protection in health and disease. 38th Meeting of the American Academy of Dermatology, Chicago, 1979.
    2. Garland, C.F., Garland, F.C., and Gorman, E.D., Could sunscreens increase melanoma risk? Am. J. of Public Health, 82,549,1992.
    3. Wilkins, B.J., Sunscreen Testing, 8th Annual Conference of the New Zealand Society of Cosmetic Chemists, Wairakei, 1986.
    4. van Praag, M.C.G., Out-Luyting, C., Class, F.H.J., Vermeer, B.J., and Mommaas, A.M., Effect of topical sunscreens on the UV-radiation-induced suppression of the alloactivating capacity of human skin in vivo. J. Invest. Dermatol., 97, 629, 1991.

    About the author

    With a BSc from Otago, Brian Wilkins began work analysing the quality of drug products for a manufacturer. Part time study later gained him a Masters degree, after which he taught science in a NZ high school for eight years. He later joined the NZ School of Pharmacy as a lecturer in pharmaceutical chemistry, spending most of his career in the position. Early in that period he spent three years getting a PhD in the London School of Pharmacy. Now as an Honorary Research Associate in the Chemistry Department of Victoria University of Wellington, he develops new technology for the manufacture of sunscreens, and hopes to interest industry in his improvements. 

    A New Look at the Old Drug - Aspirin


    No comments made yet. Be the first to submit a comment
    Already Registered? Login Here
    Thursday, 28 October 2021

    By accepting you will be accessing a service provided by a third-party external to https://racichemedcentral.com.au/

    My Account

    Get Social