Do sunscreen sprays actually work? The science

Sprays seem like the perfect way to reapply sunscreen without messing up your makeup. But how effective are they? I did some experiments to find out – the results were very eye-opening! 

The video with footage of the experiments is here, keep scrolling for the text version…

A while ago, I looked at the protection you get from sunscreen sticks in this post – unfortunately, they gave a lot less protection than I expected with normal levels of use.

Related post: How Much of a Sunscreen Stick Should You Apply?

A lot of people asked me if I could look at sunscreen sprays – how much protection are we actually getting?

Why sprays are tricky

Sprays are a lot harder than sticks. 

For all sunscreen products, the amount of product on your skin is the most important factor that determines the level of sun protection.

The labelled SPF comes from the ISO 24444 test (or the very similar FDA test), where a certified lab measures the protection after 2 milligrams per square centimetre of product has been applied to skin evenly (yes, the FDA uses metric for this!).

So if I get 0.76 grams of sunscreen onto my 381 cm² face perfectly evenly – basically, replicating the SPF test –  I should get the SPF on the label. 

Related post: How Much Sunscreen Do You Need For Your Face?

If there’s less product, there’s less protection. On average, studies have found that the relationship is roughly linear – if you apply half the amount used in testing (1 mg/cm²), you get about half of the labelled SPF.

Related post: How SPF Changes With How Much Sunscreen You Use

So checking the mass of product on your skin lets you estimate the level of protection achieved.

Sticks are at the very easy end of the scale. You touch the stick to your face, so thanks to the law of conservation of mass (science!), you just weigh the stick before and after application – the missing weight is on your face.

Standard cream and lotion sunscreens are a bit harder. You can still work out the weight dispensed by weighing the tube before and after (you can also estimate this with a measuring spoon), but it doesn’t all go on your face. It’s also on your fingers, or whatever tool you applied with – so it’s a good idea to use a bit extra.

But sprays are on a different level.

In the official SPF test, products aren’t applied to skin like in real life. For most products, little droplets are squeezed from a syringe, then spread out. This is to ensure testing staff in different labs run the test similarly, so SPF acts as a relatively fair comparison between different products.

Related post: Purito Sunscreen and All About SPF Testing

For lotions and sticks, this isn’t that different from how you’d apply them to your skin. But for sprays – in real life, the product that leaves the spray doesn’t all end up on your skin, unlike in the dollopy syringe testing situation. It goes freaking everywhere.

Disclaimer: These experiments involve a lot of products being used in ways they’re not meant to be used. I have a chemistry PhD – I’ve been trained to handle extremely dangerous chemicals. Do not try these experiments yourself without consulting someone who can help you with a proper safety assessment and appropriate precautions. Follow the instructions on the label.

Test 1: Where does the spray go?

My original plan was to collect the sunscreen spray that didn’t land on a volunteer’s face and weigh it – the remainder would’ve landed on their face. This way, the volunteer could feel how much spray landed on their face, and decide when they felt they applied enough.

So the first thing I needed to find out was: where does the spray actually go?

I stuck up a bunch of paper and got my beautiful assistant – my husband Omar – to spray sunscreen at his face.

Then we looked through a UV camera. These can’t tell you how much protection you’re getting, but you can see where the sunscreen went.

Aside from his face, the sunscreen mostly landed on the paper bib, on his right side (his spraying arm). Some also hit the wall next to his head. There wasn’t much on his left side, and there were only a couple of spots on the ground.

Pilot test: Catching stray spray

Next I tried catching the spray with paper around Omar’s face – essentially, a toilet seat-looking rig. But the weight of the paper didn’t change enough for the scales to register any change after 20 sprays.

The paper was also really floppy, and I couldn’t make a rig to catch all the spray dropping down without interfering with the spraying arm.

Instead, I decided to estimate the spray landing on the face with, essentially, a paper sheet mask.

Volatile components

But before measuring how much spray lands on the mask, there was an extra layer of complexity to solve. 

Sprays contain both volatile and non-volatile components:

• Non-volatile components land on skin and stay there. This includes ingredients like the active sunscreen ingredients and moisturisers.
• Volatile components evaporate as the spray flies through the air, and as it dries on your skin (or paper). This includes water and alcohol, and they make up most of the spray. 

This means that even if all of the spray landed on the mask, the extra weight wouldn’t be the same as the weight that came out of the spray. 

So I needed to work out the dried weight on the mask if all (100%) of the spray landed on it. Then I could do the mask experiment, and work out what fraction of the total landed on the mask.

For example:

• If I sprayed out 10 grams of product onto a sheet of paper, and it weighed an extra 2 grams after drying, this would represent 100% of the spray landing.
• If the mask experiment was performed with 10 grams and the mask only weighed an extra 1 gram, only half of the spray landed on the mask (1/2 x 100 = 50%).

Test 2: Measuring the volatile fraction

I tested 5 face sprays to see which would be best for the mask experiment. These are all meant to be sprayed directly on the face, with instructions like “on the go”, “sets makeup”, and “for reapplication”. I also tested a body sunscreen spray out of curiosity.

• I sprayed ~5 grams of each into separate funnels (to catch more spray) made from A4 printer paper over my stove, with the fan on and the windows open.

• Each bottle was weighed before and after spraying to calculate the mass dispensed.
• The funnels were weighed before spraying, and 24 hours after to allow volatiles to evaporate completely.

That’s the nice summarised version, but in reality there were problems (of course, because sprays are just problems):

• The spray didn’t all land on the funnel – some mist was visibly floating off. But this should be a negligible amount, and any lost mist here would overestimate the spray landing on the face.
• I managed to hang the funnels without them touching anything, but some were really wet and dripped onto the ground, so I had to redo those.

Results

Spray	Protection	% Non-volatile
A	SPF 50+	27.3
B	SPF 50+	31.9
C	SPF 50	26.2
D	SPF 30	27.9
E	SPF 30	18.9
F – body	SPF 50+ 4 Hr WR	34.1

The face sprays all roughly contained 25-30% non-volatiles.

Spray E is the lowest, which makes sense – it’s SPF 30, and it’s a makeup setting spray which feels really lightweight. The other SPF 30 product is mostly meant to moisturise, so there’s probably more non-volatile ingredients that stay on the skin.

I decided to do the mask test with Products A and B, and redid the funnel measurements to get a more reliable number and check that I didn’t massively mess this up (I did not).

Spray	A	B
Run 1	27.3	31.9
Run 2	26.6	29.9
Run 3	28.5	30.3
Average	27.5	30.7

Test 3: How much per pump?

I also measured how much that comes out per pump of each spray. I added an extra face spray, and another bottle of Spray A in slightly different pump packaging.

Results

Spray	Protection	Mass per spray (g)
A	SPF 50+	0.088
A(2)	SPF 50+	0.065
B	SPF 50+	0.089
C	SPF 50	0.064
D	SPF 30	0.144
E	SPF 30	0.213
F – body	SPF 50+ 4 Hr WR	0.195
G	SPF 50+	0.114

The numbers were all over the place. The top product (Product E) dispensed about 3.5 times more per pump than the bottom (Product C). It even sprayed out more than the body spray (Product F).

I was going to collect this data during the funnel test, but I did this later with 10 nice, solid sprays because:

1. The sprays didn’t come out consistently – sometimes there would be big drips, or some air in the tube. One product needed a really long, solid press to get to the bottom, but pressing halfway felt more natural when I was spraying it quickly. So these numbers are the best case scenario.

2. I’m really bad at counting. I lose count after 10 all the time. It’s just part of who I am as a person.

Test 4: How much lands on your face?

Onto the mask test! I cut out masks from A4 printer paper, with little eye holes so I could see where the spray was pointing.

I clipped each mask to my hair. They don’t hug my face completely, but this is going to be a ballpark estimate anyway (the variation in spraying technique will probably make more of a difference).

I sprayed about 4.5 grams onto my face for each mask, following the instructions for one of the products, in an X and a T to make an asterisk (like a makeup setting spray).

I naturally hold the spray about 20 cm (8 inches) away from my face, which is what most products recommend. 3 trials were performed per product.

This whole process was really unpleasant. I was aware of the inhalation risk, so the window was wide open, giving good ventilation with no breeze (average of ~500 ppm CO2 on my Aranet4 meter). While I was using a LOT more sunscreen than normal, the masks protected my nose somewhat, and I timed my breathing so the spray had time to settle. But the fragrance overload was not nice. Sunscreen ended up on everything, and the clips pulled on my hair. Science is pain.

The masks were allowed to dry for 24 hours before weighing.

Results

Spray	Run	% on mask
A	1	16.9
A	2	17.9
A	3	15.5
A	Average	16.8
B	1	11.7
B	2	13.8
B	3	13.2
A	Average	12.9

I was really surprised by how little spray landed on the mask: 15-18% for Spray A, and 11-14% for Spray B.

This means with my spraying technique (which is commonly recommended), I’d need to spray 5 to 9 times more sunscreen than I’d actually get on my face.

This translates to 48-76 sprays for 0.76 g (a full application for my face size). You’d need 80 to 120 sprays for 1/4 teaspoon (recommended if you don’t know your face size).

I was honestly pretty shocked. I thought maybe one-third or one-half  (33-50%) would land on my face. I’ve recommended sprays before for reapplication, but this was pretty eye opening.

A different spraying technique?

I thought that would give a definitive answer, but I wasn’t entirely satisfied. Why were the numbers so low? And was there a way to spray better?

During the experiment, I noticed a few things:

• I could feel lots of mist landing on my arms, neck and chest. A lot landed on my scalp, and some even dripped into my eyes.
• The UV camera showed much better coverage near the hairline on the masks:

• While reviewing the footage, I saw a lot of downward strokes (going up felt weird).

To me, this suggested that the bulk of the spray might come out at the start of each pump, and so the spray could mostly be spraying above my face/the mask.

I confirmed this by spraying different sunscreens at paper from 20 cm away – even when moving the spray in a downward stroke, the sunscreen was concentrated at the start, with more of an oval spray pattern than a line.

I also recorded the sprays in slow motion and played them in reverse to see when the densest mist came out – during the first 10% of each press down.

So while my technique reflects what people are doing in practice, and might even be better than average (there are lots of videos with people spraying in huge zigzags and circles – at least the X and T cross the middle of the face), the coverage could improve by spraying directly at the centre of the face. This isn’t very safe – it’s the worst method if you want to avoid inhaling the sunscreen – but I wanted to know how much of an improvement that would give.

Test 5: What if you try really hard?

I was really sick of getting sunscreen all over myself, but I needed to know the answer.

So I made a dummy head: Plastic Michelle, roughly the same size as my head, with tape loops for clipping the mask.

I also checked the spray pattern (or blast radius) of the sprays – it’s about 4 cm from 20 cm away.

This meant aiming within this circle should keep the spray mostly on the mask.

This setup also allowed me to aim the spray more accurately. I used a ruler to keep spraying from a distance of 20 cm. I also did this outside because I was sick of the fragrance (they’re pretty nice fragrances, but not at this volume, and not when you’ve been smelling it for the last month).

But during the experiment, I noticed yet another problem. It wasn’t windy – the leaves on the trees were barely moving – but the slightest breeze made a whole bunch of spray drift sideways.

Again, this made sense in hindsight – mist particles are far lighter than leaves. Even the movement of my arm seemed to pull some spray away.

So I tried not to spray when there was literally any wind. I did 2 runs per spray, and wiped down Plastic Michelle’s face with alcohol in between. The masks were weighed again after 24 hours.

Results

Spray	Run	% on mask	Order performed
A	1	35.5	2
A	2	34.8	3
A	Average	35.2
B	1	27.9	4
B	2	42.6	1
B	Average	35.3

The results are better – on average 35% (about a third) of the spray landed on the mask. It’s about 2.5 times more than the previous experiment.

But there was a pretty big difference between the Spray B results. I triple checked if I mixed up the masks – I didn’t – so I was confused until I remembered these were performed in a randomised order (see above).

So it could be that it got windier during the experiment, or maybe I tried harder at the start, or both. But this was always going to be variable, in real life people don’t spray as well as I was spraying, and wind exists.

Test 6: What if you try even harder?

I was meant to be done at this point, but the wind bit bothered me.

I got 1.5 times more spray on the mask in the first Spray B test. I didn’t have this big a difference in the first experiment, when I could barely see where I was aiming. I couldn’t confidently say 35% definitely represents the best case scenario.

So at midnight, because this was eating away at me, I did one absolutely final mask experiment – the same test but indoors, with good ventilation but no breeze. I didn’t even move the bottle while spraying, I just sprayed straight towards the middle of the mask.

Results

Spray	Run	% on mask	Sprays for 1.25 g	Sprays for 1.25 g
A	1	49.0	29 or 39	29 or 39
A	2	46.1	31 or 42	31 or 42
A	Avg	47.6	30 or 40	30 or 40
B	1	53.1	26	26
B	2	61.8	23	23
B	Avg	57.4	25	25

These are the actual best case scenario numbers: spraying directly at the mask from 20 cm away, within the borders, indoors, with no wind from the environment or my arm moving.

The numbers are better, but not by that much – only about half the spray (average of 52.5%) is landing on the mask. But this is as good as it gets.

Experimental conclusions

To me, these results show a fundamental physics conundrum with sprays. They’re designed to produce a fine mist with small drops, so they don’t drip and mess up your makeup when they land on your face.

But if the drops are too small, several things can happen (and remember, the drops shrink as they travel through the air):

1. They stop and drop down before getting to your face, as seen in the slow motion videos. Sprays don’t produce a single droplet size, so some smaller particles fall short of 20 cm.

2. Smaller drops get pushed around easily by even tiny air currents, such as from arm movements – that’s all the mist you see flying around in spray sunscreen videos. These are also easier to breathe in.

3. Even smaller particles turn airborne and float around, becoming too small to see – in the slow motion videos, some mist particles just disappear without dropping to the ground.

It’s possible that a spray could be engineered with very uniform particles. These could just be big enough to land 20 cm away without dripping, and maximise the product landing on your skin. But there’s also the problem of even coverage.

These last two mask experiments were the best case scenario for maximising the amount of product landing on your face. But this spraying technique sacrificed how even the layer was.

The sunscreen is concentrated in the middle of the face, but because the spray pattern is spotty, there’s not enough overlap near the edges to give continuous coverage. Spraying closer to the edges could even it out, but more sunscreen would miss the face.

Rubbing the spray in afterwards is the best approach, and a lot of guidelines recommend this. You can even spray it into your hands and apply it like a lotion or cream. But then you don’t get the most lauded benefit of sprays: easy reapplication, without messing up your makeup or getting your hands dirty.

So even though getting 1/2 or 1/3 of the labelled SPF in these “best case scenario” experiments sounds pretty good, you’re not actually getting that much because it’s so uneven. 

This means you’re not getting the SPF on the label even if you spray 14 to 25 sprays for my face size, or 23 to 42 sprays if you’re aiming for 1/4 teaspoon. (Realistically, I don’t think most people are doing even 10 sprays.)

Aerosol sprays are… worse

The other type of sprays are aerosol sprays. These come in cans that spray continuously when you hold down the nozzle, because the product is pushed out by a gas. Most have a propellant like butane mixed with the sunscreen.

These have a few upsides compared to pumps (more consistent spray, less effort to apply), but there are probably more downsides.

Paying for less protection

The propellant is counted as part of the weight of the product on the label. But SPF testing is conducted after getting rid of the propellant.

This blew my mind at first, but it makes sense – the propellant is a gas once it’s outside the container, so you can’t weigh it accurately during testing. This way, all aerosol sunscreens are measured relative to each other.

But the propellant takes up a pretty big portion of the product. One Australian scientist (Dr Elke Hacker) released the propellant from some aerosol sunscreen sprays, and found that 27 to over 61% of these products was propellant. This means a 100 g can only contains 39 to 73 g of a SPF 50 product, and a standard 150 ml (3.2 fl oz) aerosol can has about 3 adult full body applications.

(If you’re wondering, you also get less protection with aerosol mousse or foam sunscreens, but they usually only have 5-10% propellant, and you can actually get all of the other 90-95% onto your skin.)

Smaller droplets

Droplets from aerosol sprays are usually even smaller than pump droplets, so they’re even more easily messed up by wind.

Dr Elke Hacker also tested this by spraying aerosol sunscreens across a 12.5 cm gap (a bit less than 5 inches for imperial heathens) into an 18 cm (7 inch) funnel, with different levels of wind blowing sideways through the gap.

• With 10 kph wind (leaves rustling, can feel wind on your face), 32-79% (average of ~50%) more is lost – this matches my outdoor experiment issues
• With 20 kph wind (loose paper flies around), 28-93% (average of ~70%) more is lost

At popular Australian beaches, there’s over 10 kph wind 90% of the time, and there are lots of random gusts that are over 10 kph the rest of the time.

This is doesn’t take into account the other issues highlighted in my experiments, like the spray going past your skin. Her setup was also a best case scenario situation – an 18 cm funnel will catch a lot more sunscreen than, say, your arm.

Safety

There’s also the issue of safety. Very small particles can get inhaled deep into your lungs and stay there, leading to respiratory issues like asthma, emphysema, brochospasm and COPD. There’s also possible airway irritation.

Because of Dr Hacker’s testing on aerosol spray effectiveness and potential safety concerns, the updated Australian sunscreen standard requires aerosol and pump sprays to have these instructions on the label from July 2024 (similar rules have been proposed in the US too):

• Hold the container 10 to 15 cm away from the body and apply liberally and evenly until the product looks and feels wet on the skin.
• Do not spray directly onto the face. Spray onto hands and then apply to the face.
• Do not apply the product in windy conditions.
• Use in a well-ventilated area and avoid inhalation.

What we learned

So after all this, what have we learned?

SPF face sprays aren’t that effective for (re)applying sunscreen

Even if you don’t care about safety, you need to spray a lot more than you’d think – with standard application, I’d estimate that only ⅕ to 1/15 (or less) of the product actually ends up on your face. That means you’re wasting at least 80% of the product.

Even if you spray right at the middle of your face (the least safe place to spray), you’re only getting good coverage in the centre, and you’re still losing ½ to ⅔ of the product.

If there’s a slight breeze, you’ll probably need to double that amount again – wind is extra bad for sprays!

Useful situations for sprays

I do think sprays can be useful in some situations (I’m probably missing some):

They allow you to apply body sunscreen quickly, and they’re easier to spread. This might be handy if you have impatient kids, for example.

For areas with body hair, sunscreen sprays are usually clear and don’t need to be rubbed in as much. If you have a beard, spraying it into your hands and then applying is a good way to avoid the whiteness from regular sunscreens.

Sprays are also probably the best way to get sunscreen onto your scalp if you have hair. They’re probably far more effective than powders (my least favourite type of sunscreen).

Related post: Do They Work? Colorescience SPF Powder, Skinnies “Pea-Sized Amount”

How to reapply over makeup

I’d recommend patting on a regular sunscreen with a sponge (you could also pat on a sunscreen spray you’ve sprayed into your hand), or using a sunscreen stick (if you apply enough).

Related post:  How to Reapply Sunscreen Over Makeup

If you’re getting lots of sun, I’d give up on saving the makeup – to get a solid reapplication, it’s best to apply your regular sunscreen normally and redo the makeup on top.

And of course, you should layer sunscreen with other types of sun protection (use shade, stay out of the sun during peak UV times, wear hats and protective clothing and sunglasses).

Additional notes

There are two other studies I found on spray sunscreens, but I don’t think they’re very relevant:

• Broussard and coworkers (2020) collected aerosol sunscreen sprayed on arms indoors from 10-30 cm away with an absorbent towel. They compared the weight lost from the bottle with the weight of the absorbent towel, measured within 30 seconds, and found that 56.9–60.94% of the mass was lost to the environment. However, they didn’t take the weight of the volatile components into account – the propellant and alcohol would have likely evaporated within 30 seconds.
• Novick and coworkers (2015) asked people to apply stick and lotion sunscreens on themselves, and spray paper towel on a silhouette of a back with an aerosol sunscreen. They estimated that 25% of the spray was lost “based on observation”, but there aren’t any details about what that means.

References

International Organization for Standardization. Cosmetics – sun protection test methods – in vivo determination of the sun protection factor (SPF), ISO 24444:2019.

Ou-Yang H, Shyr T. Dose-response of SPF values: linear or exponential?. Photodermatol Photoimmunol Photomed. 2016;32(1):52-54. doi:10.1111/phpp.12210

Petersen B, Wulf HC. Application of sunscreen–theory and reality. Photodermatol Photoimmunol Photomed. 2014;30(2-3):96-101. doi:10.1111/phpp.12099

Novick R, Anderson G, Miller E, Allgeier D, Unice K. Factors that influence sunscreen application thickness and potential preservative exposure. Photodermatol Photoimmunol Photomed. 2015;31(4):212-223. doi:10.1111/phpp.12171

Broussard L, Hirner S, Dellavalle RP, Dunnick CA, Hugh J. Spray sunscreen: Characterizing application area density and implications for sun protection. J Am Acad Dermatol. 2020;82(3):749-751. doi:10.1016/j.jaad.2019.07.094

Hacker E. Testing and evaluating aerosol sunscreens. November 2020. 

Hacker E. Testing aerosol sunscreen products: exploring the impact of wind on the application of sunscreen. November 2021.

Department of Health and Human Services, Food and Drug Administration. Sunscreen Drug Products for Over-the-Counter Human Use, Proposed Rule, FDA-1978-N-0018. 26 February 2019. 

Therapeutic Goods Administration. Proposed adoption of the Australian/New Zealand Sunscreen Standard AS/NZS 2604:2021: Impact Analysis. January 2024.