photo of woman with bad sunburn

© Walters


See also: UV Index


People and ultraviolet light

graph of solar spectrum,, energy content against wavelength

Solar spectrum at the Earth's surface. Height is energy at that wavelength. The black area to the right of the violet is ultraviolet. Courtesy Lawrence Berkeley Laboratory.

Tanning, sunburn and the skin damage that can eventually lead to skin cancer are caused by the invisible ultraviolet light in sunlight. The range of ultraviolet wavelengths is divided into three bands: UVA, UVB and UVC. UVC is the germicidal light used, for example, in hospitals. The Earth's atmosphere filters out the UVC in sunlight, so sunscreens don't deal with UVC.

UVB is the band that causes sunburn. The wavelengths with the biggest effect are those from 300 to 307 nanometers. (Testing with artificial ultraviolet sources showed that 297 nm and 254 nm had the greatest reddening effect.)

3D cross section of skin

UVA is less energetic than UVB but is more penetrating. UVA is generally considered nonreddening, although long exposures to UVA (for example, 2 hours of midday sun on a fair-skinned person) can cause sunburn. In fair-skinned people, 10 to 15 per cent of UVA gets through the epidermis to the dermis, and in dark-skinned persons 5 to 10%. UVA, to a greater extent than UVB, is responsible for aging effects. It does cause degenerative skin changes.

Rating protection against UVB: SPF

The first system used to rate the effectiveness of sunscreens used SPF numbers (SPF stands for Sun Protection Factor). It is based on tests of how long it takes a person's skin to redden when that sunscreen has been applied, compared to how long it takes without the sunscreen. In theory, if a person's skin would redden after 10 minutes' exposure to the sun under a particular set of conditions, protected by a suntan lotion with an SPF of 15 it would not redden until it had been exposed for (10 minutes times SPF 15 =) 150 minutes. SPF numbers for products are determined experimentally, by exposing patches on people's arms to ultraviolet light from sunlamps with and without the sunscreen, which is applied in a rigorously controlled way. Because reddening is caused by UVB, not UVA, the SPF test says nothing about the sunscreen's ability to block UVA radiation.

The SPF method of rating sunscreens was chosen because consumers were likely to find SPF numbers easier to interpret than transmittance/absorbance percentages. For comparison, SPF 2 sunscreen blocks 50% of UVB; SPF 15 blocks 93%, and SPF 30 blocks 97%. As you can see, SPF 30 does not provide twice as much protection as SPF 15, which is what consumers are likely to assume. Products with ratings above 30 do not provide significantly more protection than SPF 30 lotions. To prevent a misleading arms race in SPF numbers, Australia and the United States (since 1999) don't allow manufacturers to label sunscreens with SPF numbers above “30+”. In August 2007, the US proposed raising this limit to "50+."

For many years there was no international agreement on how to do an SPF test. In 2006 representatives of the European, American, Japanese and South African sunscreen makers agreed upon an International Sun Protection Factor Test Method (2006).¹ The relevant trade association in the U.S., the Cosmetic, Toiletries and Fragrances Association, participated in the development of the test protocol. The FDA, however, proposes to adopt a slightly different test.

The Australians, who are by far the nation most active in sun protection, have not adopted the International SPF Test. Their SPF test² is unique in that they apply the lotion, then immerse the body part in clean turbulent water, and then do the test.  In other words, the Australians assume that many people who go out in the sun are likely to get wet. In Europe, they apply the lotion and do the test. In the real world, then, two lotions with the same SPF rating, one Australian and one European, might perform very differently.

While the current test is very useful, it is not ideal. For one thing, it suffers from low reproducibility: different laboratories testing the same lotion get slightly different results. In fact, for SPF's over 30, results may differ by 30 to 40% of two standard deviations for the SPF value measured.3 The European Commission is also concerned that running potentially harmful tests on people is unethical, and would prefer an in vitro test that didn't require living things. Work on developing new tests is ongoing.

Rating protection against UVA

For many years the public was oblivious to the danger from UVA, well after the introduction of SPF ratings. As awareness grew, manufacturers added UVA protection to their products, and general terms like “broadband” or claims like “protects against UVA” began to appear on product labels. Such labeling was potentially misleading and did not allow the consumer to compare products' effectiveness.

In 1996 a test for protection against UVA was adopted by the cosmetics industry in Japan: the persistent pigment darkening test. It measures the ratio of the minimum UVA dose needed to produce persistent pigment darkening of skin protected by a sunscreen, to the minimum dose needed to produce darkening in the same unprotected skin. A government agency in France refined the test and that version has been nominated as a standard by the European Commission. The numerical result of the persistent pigment darkening test is called the “UVA protection factor.”

How should a product's label indicate its effectiveness against UVA? Many years' dependence on SPF ratings has trained the public to believe that the higher the SPF number the better the sunscreen. The prevalence of this belief suggests that in compounding products the policy should be that the more a product protects against UVB (that is, the higher its SPF rating), the more it should protect against UVA. Manufacturers generally follow this policy.

The United Kingdom, while coupling UVA with UVB protection, also has a labeling system that allows consumers to compare two products with the same SPF rating but different levels of UVA protection. At any SPF level, the products with the greatest UVA protection get 5 stars, those with somewhat less, 4 stars, and so on. Some consumers, however, mistakenly interpret the star rating as independent of the SPF. They think that any 5-star lotion will provide more UVA protection than any 4-star lotion. In fact, an SPF 30 lotion with 3 stars gives more UVA protection than an SPF 10 lotion with 5 stars.

Seeking to avoid the British problem, the European Commission has recommended simply fixing the ratio between a product's SPF rating and its UVA protection factor at one-third. One-third was chosen to make a product's transmittance in the UVA band approximately the same as its transmittance in the UVB band. The Commission would also like to move the public away from reliance on SPF numbers to a system of four categories, as described in the table below.

UVA symbol

on the label
SPF rating
on the label
Actual measurement
by the International SPF test
falls in this range
Minimum UVA 
protection factor
Low 6 6 – 9.9 The UVA protection factor measured by the persistent pigment darkening test must be no less than 1/3 of the labeled SPF rating (the number shown in column 2)
10 10 – 14.9
Medium 15 15 – 19.9
20 20 – 24.9
25 25 – 29.9
High 30 30 – 49.9
50 50 – 59.9
Very High 50+ ≥60

Only products that meet the above specification are entitled to display on their label the UVA protection symbol shown at right. The symbol is to be phased in gradually beginning in the summer of 2007.

European "use-by" labeling

As a cosmetic, in the European Union since 11 March 2005 sunscreens that the maker believes have a shelf life of more than 30 months must have a “use-by” symbol on the label. The symbol looks like a squat jar of cold cream, with a number above it. If an “M” also appears, the number is the number of months the product will be usable once it has been opened. If there is no “M”, the number is a similar number of years. If the product has a shelf life of 30 months or less, an expiration date must be given.

Practical Implementation of Article 6(1)(c) of the Cosmetics Directive (76/768/EEC): labelling of product durability: period of time after opening. Adopted 19 April 2004.

Proposed new US labeling requirements

proposed FD label

Source: FDA/Zebulon Rogerson

On 23 August 2007, the U.S. F.D.A. proposed major changes in its sunscreen regulations. UVA and UVB protection will now be given equal prominence on labels. Like the Europeans, the FDA has introduced four defined categories, but the highest is called “highest” instead of “very high.” 

The SPF test is slightly modified. The SPF ratings are translated into categories by: low = SPF 2 to <SPF 15; medium = SPF 15 to <SPF 30; high = SPF 30 to SPF 50; highest, greater than SPF 50.

Two UVA tests are required, one a modified version of the PPD test described above, and the other a test in which the product is applied to quartz plates and the coated plate's absorbance is measured across the whole UVA band. Both tests are scored by placing the product in one of the four categories, and the product's label is to show the lowest of the two results.  The stars directly correspond to the categories, “highest” (4 stars) down to “low” (1 star). A product that hasn't been show to provide UVA protection must say “No UVA protection” near the SPF value. 

Unlike the star system used in Britain, the UVA stars on U.S.-labeled sunscreens are completely independent of the SPF rating. A lotion will provide more UVA protection than any lotion with a smaller number of stars. 

If after 40 minutes immersion in water a sunscreen retains the level of SPF and UVA protection stated on its label, it may be labeled “water resistant,” “water/sweat resistant” or “water/perspiration resistant.”  If it retains that level of protection after 80 minutes, the word  “very” may be added.

Clothing as sun protection

Authorities differ on the SPF value of clothing.  Some experts4 say a dry white cotton T-shirt has an SPF of 7.4, and that wet it is SPF 5.5, while Dr. Diane Berson5 has been quoted as saying the dry shirt is about SPF 5 and the wet shirt about SPF 1. This difference may simply reflect a difference in the “heaviness” of the shirts being tested. The essential point is that T-shirts are not sufficient protection for children, or anyone else. Heavier, darker garments can of course provide more protection than a T-shirt.

The ability of clothing to block ultraviolet light may increase with washing, because many detergents contain “brightening agents” that permeate the cloth. They convert ultraviolet light to visible light, which is the source of the brightening effect.

Getting the most from sunscreens

People don't apply as much lotion as the testers assume

icon for use enough sunscreen

Numerous studies have shown that people do not apply as much lotion as is applied in the SPF test, and so are not getting the level of protection indicated by the SPF rating. The European Commission recommendations note, “If the quantity applied is reduced by half, protection may fall by as much as two-thirds.”6 The EU has adopted the icon shown at the right as part of a campaign to get people to use enough sunscreen.

The SPF test calls for the application of 2 milligrams per square centimeter of skin. This amounts to about 36 grams for an adult body (6 teaspoons)

The Australians ;recommend 5 milliliters per body part: each arm, each leg, front, back, and face (including neck and ears). That makes 7 parts or 35 milliliters for the whole body. Five milliliters is a teaspoonful; 35 milliliters is about a shotglassful. A child requires about half this amount.

The next time you buy sunscreen, notice how many milliliters the package contains. Divide that number by 35 to get the number of adult applications the package can provide. If you get more applications than that, you are not applying enough sunscreen.

Sunscreen should not be rubbed in.

Sunscreen should be applied about 15 minutes before exposure.

People don't apply lotion as often as the testers assume

woman applying sunscreen to child's nose


The tests also assume people will reapply lotion, especially on emerging from the water, but many don't. The Australians specify that lotion must be reapplied at least every 2 hours, even if the consumer has not been swimming or engaged in other activities that might dilute the sunscreen.

Sunscreen products have a limited shelf life

Sunscreen doesn't last forever. Kept in a cool place, it may be good for a couple of years, but much less if stored in a hot car. If it's 5 years old, throw it out.

No sunscreen gives 100% protection

At one time U.S. regulations allowed a product with an SPF of 8 to say it provided “maximal protection.”  Now that is not only laughable, but not permitted. In September 2006 the European Commission recommended that terms like “sunblock” and “gives 100 per cent protection” not be used on labels. Special care must be taken with very young children; some argue they ought not to be exposed to direct sunlight at all.

Another cautionary note is that a sunscreen, even one with a very high SPF number, may not prevent serious damage from UVB itself. Some researchers have found that sunlight suppresses the immune system, and speculate that this may be the result of the action of UVB on a substance found in the uppermost surface of the skin, while sunscreens tend to be absorbed into a lower layer.

1. The protocol can be purchased from

2. AS/NSZ 2604, 1997.

3. M. W. Brown.
The Sun Protection Factor: Test Methods and Legal Factors.
SÖFW-Journal, vol. 128, pages 10- 12, (2002).

4. Hearing

Page 291.

5. Mary Tannen.
After all those days in the sun, years of damage control.
New York Times, June 23, 2002, page 8.

6. Commission Recommendation of 22 September 2006 on the efficacy of sunscreen products and the claims made relating thereto.
Document number C(2006) 4089.
Official Journal of the European Union, 26 September 2006. L 256/39.

Paragraph 14.

for further reading

Lim, H. W. et al.
American Academy of Dermatology Consensus Conference on UVA Protection of Sunscreens: Summary and Recommendations.
Journal of the American Academy of Dermatology, volume 44, pages 505-508, 2001.

home | people index | search |  contact drawing of envelope |  contributors | 
help | privacy | terms of use