Ingredients

Vitamin C serums: forms, concentration and stability

L-ascorbic acid is the only form of vitamin C that skin uses without converting it first, and it crosses the stratum corneum only when the formula sits below pH 3.5. Percutaneous absorption work on pig skin puts the ceiling at 20 percent, and a product is generally held to need more than 8 percent to do anything measurable, which is why reputable serums sit between 10 and 20 percent. Everything else on the label, the derivatives, the antioxidant partners and the packaging, is an argument about stability rather than about strength.

Why the pH decides whether anything gets in

Skin already holds vitamin C at levels comparable to other body tissues. Measured levels run from 6 to 64 mg per 100 g of wet tissue in the epidermis and from 3 to 13 mg per 100 g in the dermis, a 10-fold spread across independent studies. Humans cannot make the vitamin at all, so whatever sits in the skin arrived through the blood or through the surface. Once blood levels are saturated, more vitamin C in the diet does not appear to raise them further, which leaves whatever you put on the surface to do the work.

L-ascorbic acid is water-loving and carries a charge at the pH of an ordinary cream. The stratum corneum is lipid rich and turns that molecule away. Only when a formula drops below about pH 4, and in practice below pH 3.5, does enough of the uncharged form exist for measurable penetration. A vitamin C cream buffered to pH 5.5 for comfort has traded away the property that made the ingredient work in the first place.

The absorption work behind those numbers applied L-ascorbic acid to pig skin and measured what arrived. Below pH 3.5 it entered. The maximal concentration for optimal uptake was 20 percent. Tissue levels were saturated after 3 daily applications, and the half-life of disappearance from tissue was about 4 days. Two practical points follow: applying a serum twice a day does not double the reservoir, and missing a day does not empty it. All of that was measured in an animal model, not in a face.

A product generally needs more than 8 percent to carry biological significance, and above 20 percent there is no further gain, only a greater chance of stinging. That is why the sensible commercial range sits at 10 to 20 percent, and why the acidity that makes a serum work is also the reason it can burn on thin or inflamed skin.

What the derivatives actually buy you

L-ascorbic acid in water oxidises. It gives up electrons to dissolved oxygen, turns first into dehydroascorbic acid and then into compounds with no vitamin activity, and the liquid darkens. That single weakness is the reason the shelf carries a dozen derivatives, each with a protective group parked on the reactive part of the molecule.

Phosphate types are the oldest answer. Sodium ascorbyl phosphate and magnesium ascorbyl phosphate are stable at neutral pH and far kinder to reactive skin. The phosphate group does confer greater stability, and these forms can be converted back to ascorbic acid in the body, although slowly, and they are poorly absorbed through the skin. In the 2001 pig skin absorption study, daily application of magnesium ascorbyl phosphate, ascorbyl-6-palmitate and dehydroascorbic acid did not raise L-ascorbic acid levels in skin at all.

Ascorbyl glucoside scores better on one count: it is stable and it can penetrate. What nobody has measured is how fast enzymes in living skin cut the glucose off and release ascorbic acid, so the delivered dose cannot be stated. Tetrahexyldecyl ascorbate, also written tetraisopalmitoyl ascorbic acid, is the oil-soluble option and sits among the stable esters. Lipid-soluble derivatives do show higher uptake in animal work, but there is doubt about whether they are converted efficiently once inside, and they do not automatically bring better stability with them.

Read a derivative percentage as a different unit. A serum with 10 percent sodium ascorbyl phosphate is not a 10 percent L-ascorbic acid serum, because a conversion step of unknown yield sits between the two. Choose a derivative for tolerance, at a near neutral pH, when acid stings. Choose L-ascorbic acid when you want the form that carries the evidence.

What the 12 week trials measure

The clinical record is thinner than the category's confidence. A double-blind, placebo-controlled study in 10 subjects, using 10 percent topical vitamin C over 12 weeks, found a statistically significant reduction in photoageing scores and improvement in wrinkling against vehicle. A second double-blind study, 20 subjects on 5 percent for 6 months, found improvement in furrows on skin histology and in clinical appearance. Daily 3 percent vitamin C over 4 months increased the density of dermal papillae, and a 25 percent formulation with a chemical penetration enhancer reduced melasma pigmentation after 16 weeks.

Photoprotection is where the numbers are firmest. In pig skin, 10 percent topical vitamin C reduced UVB induced erythema by 52 percent and apoptotic sunburn cell formation by 40 to 60 percent. In pig skin again, adding 0.5 percent ferulic acid to a solution of 15 percent L-ascorbic acid and 1 percent alpha-tocopherol improved the chemical stability of both vitamins and doubled photoprotection against solar-simulated light, from about 4-fold to roughly 8-fold. The doubling belongs to the ferulic acid: 4-fold is what vitamin C and E managed alone. Read all of it as antioxidant support measured in animal skin, not as a sun protection factor, and none of it replaces a sunscreen.

A 12 week split-face trial published in 2026 shows both the shape of the evidence and its limits. 31 adults, 28 of them women, mean age 49.5 years and a range of 31 to 64, had microneedling on both sides of the face at baseline and again at week 4, then applied a serum containing 15 percent vitamin C, 1 percent vitamin E and 0.5 percent ferulic acid to one randomised side and a matching vehicle to the other. At week 12 the treated side had improved 23.9 percent on a photodamage score against 6.8 percent for vehicle, and 31.2 percent on a melasma severity score against 5.1 percent.

Then come the caveats, which the authors list themselves. It was a single centre study in 31 Korean adults, all of phototype III or IV, so it may not transfer to other groups. 12 weeks is short for collagen remodelling. 3 of the 31 stopped after week 4, two with allergic contact dermatitis from the needling device and one with a photoallergic reaction to a sunscreen. Two of the authors were employed by the company that makes the serum, and one of the two university grants behind the work named its parent company. The serum was also used after a procedure, so those percentages do not describe what a bottle does on its own.

How to buy, store and use one

Packaging is part of the formula. A water-based L-ascorbic acid serum belongs in opaque glass or an air-limiting pump, kept cool and out of the light. Colour is the honest indicator: colourless to pale straw is fine, deep yellow means oxidation is under way, brown means most of the active has gone. Nothing reverses that, and no date on the carton overrides what you can see. The period after opening symbol is the only shelf life the maker stands behind, and the Shelf-Life Tracker is a simple way to keep count.

Morning use is conventional because the photoprotection data concern UV exposure, though the 4 day tissue half-life makes the clock forgiving. Apply 3 to 4 drops to dry skin, wait a minute, then moisturiser and sunscreen. Vitamin C sits under a sunscreen without changing its SPF figure: it is an antioxidant layer beneath the filter, not a substitute for one. For the rest of the order, see the layering rules, and the Ingredient Pairing Check for what to keep apart.

Tolerance decides the form. A pH of 3.5 or below stings on skin that is already inflamed, so if you have eczema, rosacea or a barrier that has been over-exfoliated, start with a near neutral derivative, or use L-ascorbic acid 2 or 3 times a week and build up slowly. Patch test on the inner forearm for 3 to 5 days if acids usually upset you. Allergic reactions to topical actives are uncommon but real, and in the 12 week trial above 2 participants left with allergic contact dermatitis from the needling device.

Stop and see a doctor if you get persistent burning, swelling, blistering or a spreading rash, and do not keep using something that stings every time in the hope that skin will adapt. Topical vitamin C is not on the usual restriction lists for pregnancy or breastfeeding, but other ingredients in the same formula may be, so check with your midwife or doctor. Expect changes in tone and texture over 8 to 12 weeks, and read any single before and after photograph as a photograph rather than as evidence.

Before you buy a vitamin C serum

Frequently asked

Is a 20 percent vitamin C serum stronger than a 10 percent one?

In pig skin absorption studies uptake rises with concentration only as far as 20 percent, and beyond that there is no further biological gain, just a greater chance of irritation. Most well made serums sit between 10 and 20 percent, and about 8 percent is the floor for a measurable effect.

Do vitamin C derivatives work as well as L-ascorbic acid?

They are more stable and gentler, but the delivered dose is not established for most of them. In the 2001 pig skin absorption study, daily magnesium ascorbyl phosphate and ascorbyl-6-palmitate did not raise L-ascorbic acid levels in skin, and the rate at which ascorbyl glucoside is converted in living skin has not been measured.

Why has my serum turned orange or brown?

That is oxidation: L-ascorbic acid in water loses electrons to oxygen and becomes dehydroascorbic acid and then compounds with no vitamin activity. A deep yellow serum has lost part of its activity, a brown one has lost most of it, and nothing restores it.

Can I use vitamin C with a retinoid or with niacinamide?

Yes. Many people find it easier to use vitamin C in the morning and a retinoid at night, since both can sting on the same evening, and niacinamide sits comfortably in either routine. Separating them is a tolerance decision rather than a chemistry problem.

If my serum contains ferulic acid, do I still need sunscreen?

Yes. Adding 0.5 percent ferulic acid to a solution of 15 percent L-ascorbic acid and 1 percent vitamin E doubled measured photoprotection from about 4-fold to about 8-fold in pig skin, but antioxidant support is not a sun protection factor and it does not filter ultraviolet light.

How long should I give a vitamin C serum before judging it?

The published trials ran from 12 weeks to 6 months. A 12 week study at 10 percent found improvements in photoageing scores and a 6 month study at 5 percent found changes in skin furrows, so months is the right unit, not days.

Sources

• Al-Niaimi F, Chiang NYZ. Topical Vitamin C and the Skin: Mechanisms of Action and Clinical Applications. J Clin Aesthet Dermatol. 2017;10(7):14 to 17 · Peer reviewed dermatology review, read in full. Source of the 8 percent floor and 20 percent ceiling, the pH 3.5 threshold, the 10 to 20 percent commercial range, the 10 subject 12 week trial at 10 percent, the 20 subject 6 month trial at 5 percent, the 4 month result at 3 percent, the 16 week melasma result at 25 percent, and the 52 percent erythema and 40 to 60 percent sunburn cell figures, which the review takes from porcine work.
• Pinnell SR, Yang H, Omar M, et al. Topical L-ascorbic acid: percutaneous absorption studies. Dermatol Surg. 2001;27(2):137 to 142 · The absorption study that set the formulation rules, abstract read through the National Library of Medicine record. Source of the pH below 3.5 requirement, the 20 percent maximum, saturation after 3 daily applications, the 4 day half-life, and the finding that magnesium ascorbyl phosphate, ascorbyl-6-palmitate and dehydroascorbic acid did not raise skin levels. The work was done on pig skin, so it guides formulation rather than describing a human dose.
• Pullar JM, Carr AC, Vissers MCM. The Roles of Vitamin C in Skin Health. Nutrients. 2017;9(8):866 · Open access review, read in full. Source of the vitamin C content of epidermis (6 to 64 mg per 100 g) and dermis (3 to 13 mg per 100 g), the 10-fold spread between studies, the pH below 4 penetration point, the observation that dietary vitamin C stops raising skin levels once blood levels are saturated, and the assessment of phosphate, glucoside and lipid soluble derivatives.
• Lin FH, Lin JY, Gupta RD, et al. Ferulic acid stabilizes a solution of vitamins C and E and doubles its photoprotection of skin. J Invest Dermatol. 2005;125(4):826 to 832 · Abstract read through the National Library of Medicine record. Source of the 15 percent L-ascorbic acid, 1 percent alpha-tocopherol and 0.5 percent ferulic acid combination and the rise in photoprotection against solar-simulated light from 4-fold to roughly 8-fold. Ferulic acid was the addition that doubled the figure, and the experiments were run on pig skin.
• Liu C, Boo J, Kim H, et al. A Double-Blinded, Split-Face Clinical Trial Evaluating the Effects of a Vitamin C, E, and Ferulic Acid Serum Combined with Microneedling on Facial Photoaging. Clin Cosmet Investig Dermatol. 2026;19 · 12 week randomised split-face trial in 31 adults, read in full. Source of the week 12 figures (23.9 against 6.8 percent, 31.2 against 5.1 percent), the demographics, the 3 dropouts and the authors' own limitations. Two authors were employed by the company that makes the serum and one of the two university grants named its parent company, so the result is industry involved, and the serum was used after a procedure.

Read on

Tools that fit

This guide is for general information. It does not replace advice from a dermatologist, physician or qualified practitioner, or an examination.