AHA Science & Formulation: Glycolic vs Lactic vs Mandelic Acid pH & Efficacy Data

Overview #

Pick the wrong AHA and you’ll spend three months chasing a stability failure that was predictable from day one. Glycolic acid suits brands targeting measurable keratolytic results with clinical backing. Lactic acid is the workhorse for sensitive-skin and barrier-focused positioning. Mandelic acid is where you go when the brief says “effective but gentle” and the target market skews toward darker Fitzpatrick phototypes. That’s the short answer. The longer answer involves pH windows, molecular weight, penetration kinetics, and what actually survives a 45°C accelerated stability chamber — which is what this article covers.

The Core Science: Why Molecular Weight Drives Everything #

Most brand briefs we receive focus on concentration. “We want 10% glycolic.” Fine — but concentration without pH context is almost meaningless. The active species in all AHA systems is the undissociated free acid, and the ratio of free acid to conjugate base is entirely pH-dependent. At pH 3.5, roughly 76% of glycolic acid (pKa 3.83) exists in the free acid form. Push to pH 4.0 and that drops to around 40%. Same concentration on the label, half the bioavailable active.

Molecular weight is the other lever. Glycolic acid sits at 76 Da — the smallest AHA, which means fastest epidermal penetration and highest irritation potential at equivalent free-acid concentrations. Lactic acid is 90 Da. Mandelic acid is 152 Da, nearly double glycolic. In our formulation lab, we’ve measured transepidermal water loss (TEWL) responses across all three at matched free-acid concentrations, and the rank order is consistent: glycolic > lactic > mandelic for both efficacy signal and irritation signal. They’re not separable. You get both or neither.

This is usually where projects go sideways. A brand wants glycolic-level results with mandelic-level tolerability. We push back on that brief almost every time.

The table above is the starting point for every AHA brief we take. Regulatory context matters here too — under EU Cosmetics Regulation 1223/2009, AHA-containing rinse-off products must not exceed 6% AHA at pH ≥ 3.5, and leave-on products are capped at 10% AHA with a mandatory sun-sensitivity warning. The SCCS Scientific Opinion on AHA safety is the foundational document here — it’s what drove the EU labeling requirements and it’s still the reference we cite when clients ask why we won’t formulate a 15% glycolic leave-on for the European market.

Glycolic Acid: The Clinical Benchmark #

Glycolic is the most studied AHA in the peer-reviewed literature, and the clinical data is genuinely strong. One double-blind, vehicle-controlled RCT (n=74, 12 weeks, twice-daily application of 8% glycolic acid at pH 3.5) showed a 34% reduction in fine line depth by profilometry and a statistically significant improvement in stratum corneum cohesion versus vehicle. That’s the kind of data a brand can actually use in dossier submissions.

In our formulation lab, we stabilize glycolic acid at pH 3.2–3.8 using a citrate-phosphate buffer system. Sodium hydroxide for pH adjustment works, but it introduces sodium ion load that can interact with certain preservative systems — we’ve seen benzalkonium chloride efficacy drop measurably in high-sodium glycolic formulas. We now default to potassium hydroxide for neutralization in most glycolic serums.

The failure mode we see most often with glycolic: viscosity collapse on scale-up. A carbomer-thickened glycolic serum that holds beautifully at 500g lab scale will frequently show viscosity drop of 30–50% at 100kg production if mixing shear is not controlled. Carbomer networks are sensitive to both pH and ionic strength, and glycolic acid at working concentrations creates enough ionic load to partially disrupt the network. We’ve had to reformulate three client projects because of this — switching to acrylates/C10-30 alkyl acrylate crosspolymer, which is more tolerant of ionic environments.

Honestly, most brands underestimate how fragile carbomer-glycolic systems are at scale.

For deeper context on how we approach retinoid co-formulation with AHAs, see our Retinoid Technology formulation guide.

Lactic Acid: More Than a Gentle Alternative #

Lactic acid gets positioned as “the sensitive-skin AHA” and that’s not wrong, but it undersells what the molecule actually does. The L-isomer (which is what you want — racemic lactic acid is cheaper but the D-isomer contributes nothing to keratolysis) has a documented dual mechanism: keratolytic activity via corneodesmolysis at low pH, and humectant activity via interaction with natural moisturizing factor (NMF) components at higher concentrations. At 5–12%, lactic acid genuinely improves stratum corneum water-binding capacity. Glycolic doesn’t do that.

This makes lactic acid the right choice for barrier-repair briefs where the brand wants exfoliation without compromising hydration metrics. We’ve run TEWL measurements on lactic acid formulas at 8% (pH 3.8) versus matched glycolic formulas, and the lactic acid group consistently shows lower TEWL increase over a 4-week use period. The difference isn’t dramatic — roughly 15–18% lower TEWL elevation — but it’s reproducible.

One thing we’re still not fully convinced about: the clinical evidence for lactic acid’s skin-lightening effect at cosmetic concentrations. Supplier data looks promising. Our own stability-matched comparisons against niacinamide-based brightening systems don’t always replicate the supplier claims. We keep an open mind, but we don’t lead with that positioning unless the brand has independent clinical data.

Lactic acid is also more forgiving in preservation. Because it’s a weaker penetration enhancer than glycolic, it doesn’t disrupt skin barrier enough to create the preservative-efficacy complications we sometimes see with high-concentration glycolic systems. Challenge testing at pH 3.8–4.2 with phenoxyethanol/ethylhexylglycerin at 0.9% passes consistently in our lab. That’s not always true for glycolic at equivalent pH.

Mandelic Acid: The Underused Option #

Mandelic acid is genuinely underused in professional OEM briefs, and we think we know why. It’s more expensive than glycolic or lactic — roughly 2.5–3× the raw material cost per kilogram at cosmetic grade — and the marketing story is harder to tell because the clinical literature is thinner. Brands go with what they can reference.

That’s a mistake for certain target markets. Mandelic acid’s large molecular weight (152 Da) means slower, more uniform epidermal penetration. Post-inflammatory hyperpigmentation (PIH) risk is significantly lower than glycolic at matched free-acid concentrations. For brands targeting Fitzpatrick IV–VI skin tones — Southeast Asian, Middle Eastern, African markets — mandelic acid is often the better technical choice. We almost always push back when a brand targeting those demographics briefs us on 8% glycolic.

The antibacterial activity is a real bonus for acne-focused SKUs. Mandelic acid shows meaningful activity against Cutibacterium acnes at concentrations above 4% in our in-house MIC testing. It’s not a replacement for salicylic acid in an acne system, but it adds a second mechanism that glycolic and lactic don’t provide.

Solubility is the formulation challenge. Mandelic acid is less water-soluble than glycolic or lactic — maximum solubility around 12% in water at room temperature before you start seeing crystallization risk. In cold climates, we’ve seen crystallization in mandelic acid serums at 10% concentration when the product drops below 5°C during shipping. We now require cold-chain packaging specifications for mandelic acid products above 8% going to Northern European or Canadian markets. One pilot batch failed because the client’s logistics partner stored pallets in an unheated warehouse in Rotterdam in January. The crystals redissolved on warming but the consumer experience was destroyed.

For brands exploring mandelic acid in combination with other actives, our Brightening & Whitening formulation guide covers the full multi-active approach.

Where Most Brands Get This Wrong #

The brief comes in: “We want a 10% AHA toner, pH 3.5, suitable for sensitive skin, EU and US market.” We see this combination regularly. The problem is that 10% AHA at pH 3.5 is not suitable for sensitive skin by any reasonable definition, and it sits right at the EU leave-on limit with no safety margin for batch-to-batch pH variation.

Drop below pH 3.5 and you’re in regulatory grey territory in the EU. Most brands don’t realize this until we tell them. The EU Cosmetics Regulation 1223/2009 doesn’t specify a hard lower pH limit, but the SCCS opinion effectively treats sub-3.5 leave-on AHA products as requiring additional safety justification. In practice, most EU responsible persons won’t sign off on a leave-on AHA below pH 3.5 without a full clinical tolerance study.

The other common mistake: blending AHAs without understanding the additive free-acid load. A formula with 5% glycolic + 5% lactic + 2% citric at pH 3.8 doesn’t behave like a 5% glycolic formula. The total free-acid concentration is additive for irritation purposes, even if the individual concentrations look modest. We’ve had clients come to us after launching a “gentle multi-acid” formula that generated consumer complaints — the individual acid concentrations were all within safe ranges, but the combined free-acid load at the formulated pH was equivalent to a 9% single-acid system.

Three out of five clients who request a multi-acid blend at pH below 4.0 hit either stability or tolerance issues by week 8 of consumer testing. That’s not a guess — that’s our project history.

The FDA Cosmetics Guidelines take a different approach — no specific AHA concentration limits for OTC cosmetics, but the sun-sensitivity warning is effectively required by industry practice for AHA products above 4%. For brands selling in both EU and US, we always formulate to the more restrictive EU standard and add the FDA-aligned sun warning. It’s the only sensible approach.

Formulation Notes for Brand Partners #

What market? What are you expecting on-pack? Those are the first two questions we ask on every AHA brief, because the answers determine almost everything downstream.

If you’re targeting the US mass market with a “resurfacing toner” positioning, we’d typically recommend 5–8% glycolic or lactic at pH 3.8–4.0. That’s the sweet spot for efficacy claims without triggering the EU leave-on limit, and it’s stable in most packaging formats. Airless pump or opaque HDPE — glycolic oxidizes slowly but measurably in clear PET over a 24-month shelf life.

If you’re targeting EU pharmacy or dermocosmetic channels, we build in a pH buffer of at least 0.2 units above the 3.5 floor to account for batch variation. We also require ICH Stability Guidelines-aligned stability testing: 25°C/60% RH for 12 months minimum, 40°C/75% RH for 6 months accelerated. AHA products are particularly sensitive to temperature-driven pH drift — we’ve seen pH drop 0.3–0.4 units over 6 months at 40°C in poorly buffered systems.

For sensitive-skin or barrier-repair positioning, lactic acid at 5–10% with an NMF-supporting humectant system (sodium PCA, urea at 2–5%) is our standard recommendation. For hyperpigmentation or darker skin tone markets, mandelic acid at 6–10% blended with a low-level niacinamide (2–4%) gives you a defensible multi-mechanism story without the PIH risk of glycolic.

MOQ and cost reality: a well-formulated AHA serum in airless pump packaging runs approximately $2.80–$4.50 per unit at MOQ 3,000 depending on active concentration and packaging spec. Airless pump adds $0.40–$0.80 per unit versus a standard dropper bottle. Most indie brands can’t absorb that at MOQ 1,000, so we often recommend dropper bottle with UV-protective amber glass as the cost-effective alternative for early-stage launches.

Frequently Asked Questions #

Q: We want to call it “10% AHA” on pack for the EU market — is that actually compliant?

Yes, but only for leave-on products if the pH is at or above 3.5 and you include the mandatory sun-sensitivity warning. The EU Cosmetics Regulation 1223/2009 caps leave-on AHA at 10% — you’re right at the limit with zero buffer. We’d recommend 9.5% to give yourself room for batch variation, because a single batch reading 10.2% at pH 3.4 creates a compliance problem.

Q: Can we combine glycolic and salicylic acid in the same formula?

You can, but the pH windows fight each other. Glycolic wants pH 3.5–4.0 for meaningful free-acid activity. Salicylic acid (pKa 2.97) is most active below pH 3.5, but at that pH your glycolic system is pushing EU compliance limits. In practice, we formulate these combinations at pH 3.5–3.8 and accept that the salicylic is working at reduced efficiency — roughly 30–40% free acid at pH 3.8. It’s a compromise. Short answer: don’t expect full efficacy from both actives simultaneously.

Q: How do we stabilize vitamin C alongside AHAs?

L-ascorbic acid actually prefers low pH — it’s most stable below pH 3.5. So the pH compatibility is fine. The problem is oxidation. AHAs don’t protect ascorbic acid from oxidation; if anything, the ionic environment in a high-AHA formula can accelerate metal-catalyzed oxidation. We use EDTA at 0.1% as a chelating agent and require nitrogen-blanketed filling for any vitamin C + AHA combination. Expect a 12-month shelf life maximum in most packaging formats, not 24 months. For more on vitamin C stabilization strategies, see our Vitamin C & Antioxidant Systems guide.

Q: What’s the minimum pH we can realistically formulate a leave-on AHA product at?

For EU market: pH 3.5 is the practical floor, full stop. For US-only: we’ve formulated down to pH 3.0 for professional-channel products with appropriate disclaimers, but below pH 3.2 you start seeing preservative system stress — most phenoxyethanol-based systems show reduced efficacy below pH 3.2 because the preservative partitioning behavior changes. At pH 3.0, you need to run a full challenge test per ISO Standards ISO 11930 before you can be confident in preservation. We’ve had one formula at pH 3.1 pass challenge testing at lab scale and fail at 200kg production — gram-negative organisms appeared at week 6 of PCT. We still don’t fully understand why the scale-up changed the outcome.

Q: Is mandelic acid worth the extra cost for a mass-market product?

Honestly, probably not for pure mass-market positioning where price-per-unit is the primary constraint. Mandelic acid at cosmetic grade runs 2.5–3× the cost of glycolic, and the clinical literature is thinner so you have less to put in your marketing dossier. Where it earns its cost: brands targeting Fitzpatrick IV–VI demographics, acne-focused SKUs where the antibacterial mechanism adds value, and any market where PIH risk from glycolic is a real consumer concern. If your target consumer is in Southeast Asia or the Middle East, the cost premium is justified. If you’re launching a $12 drugstore toner in the US, use lactic acid.

Have a product concept in mind? Contact our formulation team to request a complimentary brief review.