Mineral oil or petroleum jelly: which is better for dry skin?
Same refinery, different seal. Swapping petrolatum for mineral oil in a test cream took it from occlusive to near bare skin. Film thickness, FDA bands, which to use.
Petroleum jelly is the better choice for dry skin, by a wide margin: it seals in water, and mineral oil, on the evidence, only slows its escape about as much as a plant oil does. The two come from the same refinery and share a reputation, but the measurements separate them sharply. In a 1988 study that held the oil phase of a test cream at 50 percent and swapped petrolatum for mineral oil ten points at a time, every version with 20 percent petrolatum or more brought water loss below the authors' occlusion threshold within 30 minutes, while the all-mineral-oil version stayed close to bare skin. The authors' own summary: "petrolatum provides higher occlusivity to vehicles than does mineral oil." A 2012 study on volunteers found paraffin oil, another name for mineral oil, no better than most vegetable oils, and petrolatum the "most effective" of everything tested. Choose petroleum jelly when the job is stopping water loss; choose mineral oil when you need a liquid that spreads, glides and wipes off, and do not expect it to do petroleum jelly's job.
For dry, cracked or chapped skin, use petroleum jelly. A film 0.1 mm thick cut water loss by 77.5 to 93.0 percent in four people, and in a direct swap petrolatum, not mineral oil, was the ingredient that made a cream occlusive. Mineral oil gives what one study calls "semi-occlusion", in the same range as most vegetable oils. The US monograph reflects the gap: petrolatum counts as a skin protectant from 30 percent, mineral oil only from 50 percent, or 30 to 35 percent alongside colloidal oatmeal.
The two materials, measured side by side
The first page of results for this comparison is ingredient listings for products that contain one or the other, a brand blog about mineral oil, a hair forum thread and supplier question pages. None of them puts the two side by side with numbers. The table below assembles the figures that exist, from two in vivo papers in the Journal of the Society of Cosmetic Chemists, two volunteer studies that tested paraffin oil against vegetable oils with petrolatum as the benchmark, a dermatology review, and the US skin protectant monograph.
| Property | Mineral oil | Petroleum jelly |
|---|---|---|
| What it is | Liquid paraffin, "also called mineral oil and paraffinum liquidum" (Sethi) | Semi-solid; the samples in Berube's study were 52.8 to 74.6% mineral oil by content |
| Water loss under a 0.1 mm film | Not tested | 77.5 to 93.0% lower, four subjects (Berube) |
| Film needed to stop water loss, extrapolated | Not tested | 0.38 to 0.40 mm, white petrolatum (Berube) |
| As the only oil in a cream, 50% oil phase | "Close to the value of the bare skin" (Wang) | Below the occlusion threshold 30 minutes after application (Wang) |
| Against vegetable oils, volunteers | "Similar occlusion" to the vegetable oils except jojoba (Patzelt) | "The most effective occlusion" (Patzelt) |
| Stratum corneum swelling, a measure of occlusion | 10 to 20%, "modest" (Stamatas) | 40 to 60%, "moderate" (Stamatas) |
| Reduction in water loss, review figure | 20 to 30%, grouped with lanolin and silicones (Sethi) | More than 98% at a minimum of 5% (Sethi) |
| US skin protectant active, 21 CFR 347.10 | 50 to 100%; 30 to 35% only with colloidal oatmeal | 30 to 100% (petrolatum and white petrolatum) |
| Depth in skin | "Only into the first upper layers of the stratum corneum" (Patzelt) | Not measured in these sources |
Every occlusion row points the same way, and they were measured four different ways: water vapour leaving the skin under a film of known thickness, the same measurement over a series of creams, the same again on six volunteers, and the swelling of the outer skin layer as it holds back water. No source here found mineral oil matching petrolatum on any of them. The fuller material notes, including refining grades and the safety questions, are on the petrolatum and mineral oil ingredient page, which treats the two as one class because on chemistry, safety and sourcing they are one. On occlusion they are not.
Same refinery, two different physical forms
Sethi and colleagues' review describes the two together: "The two most important materials are liquid paraffin (also called mineral oil and paraffinum liquidum) and petrolatum, consisting of a complex combination of hydrocarbons. Depending on the distribution of molecular weight, materials with different viscosity are obtained." Both are refined, saturated hydrocarbons, and the review adds that the refining "hydrogenated" the material "to create oxidation resistant molecules throughout from the liquid to the solid waxes". That shared origin is the root of the idea that they are interchangeable.
The difference is form. Mineral oil is a liquid at room and skin temperature, so it spreads into a thin layer, runs into skin lines and is easily wiped or rubbed away. Petroleum jelly is a semi-solid: liquid oil held in a network of solid hydrocarbon, the same heavy fraction that supplies the microcrystalline wax described on mineral waxes. The 1971 study below gives a sense of the proportions. Its six white petrolatum samples ranged from 52.8 to 74.6 percent oil, and all but one met the US Pharmacopeia standard for white petrolatum. In other words, a jar of petroleum jelly is mostly mineral oil by content. What it adds is the structure that holds the oil in place as a film.
"Baby oil" is usually mineral oil with fragrance, and "paraffin oil", "liquid paraffin" and "liquid petrolatum" are older or regional names for the same liquid. The studies on this page use "paraffin oil" (Patzelt, Stamatas) and "mineral oil" (Wang, Berube) for the same class of material.
The direct swap: what happened when mineral oil replaced petrolatum
The cleanest comparison comes from a pharmaceutical study that was not about moisturising at all. Wang, Winarna, Lichtin and Patel, working at Bristol-Myers and the University of Cincinnati, published it in 1988. They were designing a cream for a corticosteroid, hydrocortisone 17-valerate at 0.2 percent, and wanted to know how its oil phase affected water loss and drug uptake. They built 36 oil-in-water formulas in which petrolatum, mineral oil and water always added up to 78.9 percent, using USP grade white petrolatum and mineral oil, and measured the evaporation of water from the wrist and forearm of volunteers with an evaporimeter, every 15 minutes for an hour and a half, at 23 C and 40 percent relative humidity.
Bare skin in their volunteers lost a mean of 7.7 grams of water per square metre per hour, with a standard deviation of 2.1. The authors took anything under 5.6, the lower margin of that range, as occlusive. One series is the experiment this page needs: six formulas with the oil phase fixed at 50 percent and water at 28.9 percent, in which petrolatum replaced mineral oil ten points at a time.
| Formula | Petrolatum | Mineral oil | Water loss after application |
|---|---|---|---|
| #6 | 0% | 50% | "Close to the value of the bare skin": "little occlusivity" |
| #13 | 10% | 40% | "Close to the value of the bare skin": "little occlusivity" |
| #19 | 20% | 30% | Below 5.6 g/m2h from 30 minutes |
| #24 | 30% | 20% | Below 5.6 g/m2h from 30 minutes |
| #28 | 40% | 10% | Below 5.6 g/m2h from 30 minutes |
| #31 | 50% | 0% | Below 5.6 g/m2h from 30 minutes |
Half a cream's weight in mineral oil, with no petrolatum, left the skin losing water at close to the bare skin rate. Twenty percent petrolatum in the same oil phase was enough to cross the line. The authors explain the 30 minute delay: "occlusivity occurred after most of the water evaporated", which is how an oil-in-water cream works, leaving its oil behind as a layer once the water has gone. The most occlusive product in the study was formula #36, 70 percent petrolatum with no mineral oil and 8.9 percent water, which "provided occlusivity at the time of the application".
The other direction was tested too. "Increasing the amount of mineral oil from 0% to 10% at a constant percentage of petrolatum did not materially affect the TEWL value. On the other hand, increasing the concentration of petrolatum at a constant level of mineral oil decreased the TEWL values." TEWL is transepidermal water loss, the water that leaves through the skin's surface. The study's abstract puts the result in one line: "The major ingredient affecting occlusivity of the o/w emulsion systems is petrolatum, not mineral oil."
There is a cost, and the same paper measured it. When the twelve creams were scored for odour, texture, spreadability, greasiness and homogeneity, "formulas #31 and #36 were the least cosmetically acceptable": the two richest in petrolatum, and among the most occlusive. Taking all their results together, the authors judged formula #28, 40 percent petrolatum and 10 percent mineral oil, "the most desirable" for their purpose. That trade between sealing and feel runs through every choice below, and it is the subject of occlusive, emollient and humectant.
These were creams with water, a steroid, an emulsifier and a thickener, not anhydrous balms or neat materials, and the paper does not state how many volunteers took part in the water loss measurement or how much of each formula was applied. It answers which of the two oils makes a cream occlusive, at equal weight in the same base. It does not give a percentage reduction for either material on its own.
How much a film of petroleum jelly cuts water loss
For petroleum jelly on its own, the precise data come from a 1971 paper by Berube, Messinger and Berdick of the Chesebrough-Pond's research laboratories. They built a dew point instrument that swept the water from a 4 by 4 cm patch of forearm into a stream of dry air, kept subjects in a room at 19.4 to 20.0 C to stop sweating from spoiling the reading, and waited 15 minutes for the outer skin layer to stop giving up its own stored water before reading the rate. Product was weighed onto a 6 by 6 cm area so that the thickness of the film was known.
Fourteen people aged 19 to 48 lost between 0.2 and 0.46 mg of water per square centimetre per hour from untreated forearm skin. Four of them then had films of white petrolatum at five thicknesses applied, and the result for the thinnest commonly quoted case, a film 0.1 mm thick, was this:
| Subject | Base rate, mg/cm2/hr | Reduction by 0.1 mm film | Occlusive thickness |
|---|---|---|---|
| A | 0.39 | 77.5% | 0.38 mm |
| B | 0.24 | 90.2% | 0.38 mm |
| C | 0.32 | 83.0% | 0.40 mm |
| D | 0.19 | 93.0% | 0.39 mm |
Read across the rows and a pattern appears that explains why published percentages for petrolatum disagree. The same film cut water loss by 93.0 percent on the person who lost least water to begin with and by 77.5 percent on the person who lost most. The authors put it plainly: "For the same thickness of film widely different per cent reductions were obtained." A percentage reduction is a property of the product and the person together.
Their fix was the occlusive thickness. By plotting film thickness against the fraction of water loss removed and extending the line, they found the thickness at which water loss would theoretically fall to zero: "the theoretical thickness of products necessary to reduce the rate of transepidermal moisture loss to zero". For white petrolatum it came out at 0.38 to 0.40 mm in all four subjects, "despite the variation in base rate". That is a number for the material, not the person, and it is the most precise figure for petroleum jelly's seal that any source on this page gives.
It also puts the popular "98 percent" in context. The Sethi review states that "Petrolatum in a minimum concentration of 5% reduces TEWL by more than 98%", citing earlier work, with no film thickness, dose or skin site attached. Berube's measured figures at 0.1 mm are lower, 77.5 to 93.0 percent, and the occlusive thickness says why a single percentage cannot be the whole story: the seal depends on how thick the film is, and approaches complete only as the film approaches about 0.4 mm. By the paper's own formula, thickness in millimetres is ten times the weight in grams divided by the specific gravity and the area in square centimetres, so a 0.1 mm film is roughly 8 to 10 mg of petroleum jelly per square centimetre. That is a visibly greasy layer. How much people actually put on in daily use is covered in how much balm to apply.
The surprise: oilier petroleum jelly sealed better
The same 1971 paper contains the result that most directly addresses this comparison, and it runs against intuition. The authors prepared six white petrolatums of different mineral oil content and measured the occlusive thickness of each, four film thicknesses per sample on each of three subjects.
| Sample | Oil content | Occlusive thickness |
|---|---|---|
| A | 52.8% | 2.31 mm |
| B | 57.0% | 1.35 mm |
| C | 61.7% | 0.74 mm |
| D | 66.1% | 0.39 mm |
| E | 70.4% | 0.22 mm |
| F | 74.6% | 0.13 mm |
"The occlusive thickness decreases logarithmically with an increase in mineral oil content." Across the range tested, the oiliest petroleum jelly needed a film about one eighteenth as thick as the hardest one to make the same theoretical seal. Within petroleum jelly, more mineral oil meant a better seal per millimetre, not a worse one.
That looks like a contradiction of everything above, and it is not. Every sample in the table was still a petrolatum, with its oil held in a solid network, and all but one were within the pharmacopoeial standard. The trend cannot be followed out to 100 percent oil: the paper did not test liquid mineral oil, and the studies that did, Wang's creams and the two volunteer studies below, found it well short of petrolatum. The reasonable reading is that the oil in petroleum jelly is doing much of the sealing, and the solid part is what holds it on the skin as a continuous layer instead of letting it spread thin. Neither source states a mechanism, so this page does not claim one.
The practical point for anyone making an ointment base survives without a mechanism. Softening a petrolatum base with a modest amount of mineral oil is not, on this evidence, a trade against occlusion: Wang found 0 to 10 percent added mineral oil "did not materially affect" water loss, and Berube found softer, oilier petrolatums more occlusive per millimetre. What does cost occlusion is replacing the petrolatum with mineral oil.
Mineral oil against plant oils: the same league
If mineral oil is not petroleum jelly's equal, where does it sit? Two volunteer studies that tested paraffin oil alongside vegetable oils, with petrolatum as the benchmark, give the same answer: among the oils, not above them.
Patzelt and colleagues, at the Charité in Berlin, applied four vegetable oils and paraffin oil to six healthy volunteers and measured water loss, with petrolatum "as a positive control for skin occlusion". Their result: "the application of the vegetable oils (except jojoba oil) as well as paraffin oil, led to a similar occlusion of the skin surface. The most effective occlusion was found for petrolatum." They describe what the oils achieved as "a semi-occlusion of the skin surface". Jojoba was the exception among the vegetable oils; the abstract does not say in which direction, a gap discussed on jojoba oil against sweet almond oil.
Stamatas and colleagues used confocal Raman microscopy on the forearms of nine adults and seven infants, measuring how much the outer skin layer, the stratum corneum, swelled with retained water at 30 and 90 minutes after paraffin oil, two vegetable oils or petrolatum. The three oils gave "modest" swelling of 10 to 20 percent against "moderate" swelling of 40 to 60 percent for petrolatum, and the authors concluded that "there is no statistical difference between the paraffin oil and vegetable oils in terms of skin penetration and skin occlusion".
On depth, both studies place the tested oils, paraffin oil included, in the upper stratum corneum. Patzelt found paraffin oil and the vegetable oils "penetrated only into the first upper layers of the stratum corneum"; Stamatas found they "penetrate the top layers of the SC with similar concentration profiles", in adult and infant skin alike. The full set of depth studies, and why claims that one oil goes deeper than another rarely survive them, are on how far oils penetrate skin.
The Sethi review's ranking is consistent with both: after petrolatum come "lanolin, mineral oil, and silicones which only reduce TEWL by 20–30%". As the lanolin against petroleum jelly page notes, that sentence comes without a method, so the range is a guide to the ranking rather than a measured gap. The measured results above support the same order without needing it.
Why "they are the same thing" persists
The belief that mineral oil and petroleum jelly are interchangeable is not careless; it comes from the way the two are argued about. Almost everything written about them is about safety and origin, and on those questions they really are one class. Both are refined, saturated hydrocarbons from crude oil, both are stable because they have no double bonds to oxidise, both are subject to the same refining and contamination debate, and both are judged together in the arguments over fossil sourcing. The answers to those questions, including what the cancer classification actually covers, are in is petroleum jelly bad for you, and they apply to both materials equally.
The second source of the confusion is the share of oil in the jar. Berube's samples were more than half mineral oil, and his best-sealing petrolatum was three quarters oil. It is easy to go from "petroleum jelly is mostly mineral oil" to "mineral oil is petroleum jelly without the stiffness". The swap study is the answer to that step: take the structure away, keep the same weight of oil in the same base, and the occlusion largely goes with it.
The third is that the debate over petroleum against natural ingredients treats "petroleum" as one side. A comparison that pits a plant balm against "petroleum" usually means petroleum jelly, and the case it makes for the plant oils, that they cannot match the seal, does not carry over to mineral oil, which on the evidence above matches most plant oils rather than beating them. The broader comparison is scored on natural balm against petroleum jelly.
The US monograph: 30 percent against 50
The regulation reflects the difference. In the United States, a product that claims to protect skin is an over-the-counter drug, and 21 CFR 347.10 lists the active ingredients that may carry the claim, "within the concentration specified for each ingredient". Petrolatum is listed at 30 to 100 percent, and white petrolatum separately at the same 30 to 100 percent. Mineral oil is listed at "50 to 100 percent; 30 to 35 percent in combination with colloidal oatmeal in accordance with § 347.20(a)(4)". Petrolatum qualifies at a lower floor than mineral oil, and mineral oil reaches the lower level only with a second active.
| Formula | Petrolatum active? | Mineral oil active? | Skin protectant claim possible? |
|---|---|---|---|
| 100% petroleum jelly | Yes | n/a | Yes |
| 100% mineral oil | n/a | Yes | Yes |
| 40% petrolatum with 10% mineral oil | Yes | No, below 50% | Yes, on petrolatum; mineral oil is then inactive |
| 40% mineral oil, no petrolatum or oatmeal | n/a | No, below 50% | No |
| 30% mineral oil with colloidal oatmeal at 0.003% or more | n/a | Yes, in combination | Yes, as the 347.20(a)(4) combination |
| 50% mineral oil with 30% petrolatum | Yes | Yes | Yes, as a combination under 347.20(a)(1) or (2) |
The oatmeal route is the one place where mineral oil has a use of its own in the monograph. Colloidal oatmeal is an active at "0.007 percent minimum; 0.003 percent minimum in combination with mineral oil", and the combination carries its own labelled use under 347.50(b)(7): "temporarily protects and helps relieve minor skin irritation and itching due to" rashes or eczema. The oatmeal side of that is covered on colloidal oatmeal. Otherwise, the two can be combined as actives under 347.20(a), "provided each ingredient in the combination is within the concentration specified", which with mineral oil at 50 percent and petrolatum at 30 percent leaves at most 20 percent for everything else.
Most small makers do not want a drug product, and the way to stay a cosmetic is to make no protectant claim, whatever the levels. Where that line falls is set out in cosmetic against drug claims.
Which to choose for which job
| Job | Choose | Why |
|---|---|---|
| Cracked hands, heels or knuckles overnight | Petroleum jelly | The only one of the two shown to make a product occlusive |
| Chapped lips | Petroleum jelly, or a balm built on it | Seal, and a US lip protectant claim from 30% |
| Softening a petrolatum ointment | Petroleum jelly with a little mineral oil | 0 to 10% added mineral oil "did not materially affect" water loss (Wang) |
| Massage or large areas where glide matters | Mineral oil | Spreads as a liquid; occlusion is not the job |
| Cleansing balm or makeup removal | Mineral oil | The product is meant to come off, not stay on |
| US anti-itch claim for rashes or eczema | Mineral oil 30 to 35% with colloidal oatmeal | The only monograph route that names mineral oil at under 50% |
| Replacing a plant oil to avoid oxidation | Mineral oil | Similar occlusion to most plant oils, no double bonds to go rancid |
| Replacing petroleum jelly to get a lighter feel | Neither swap works on occlusion | A lighter film seals less; use less petroleum jelly instead |
The pattern behind the table is simple. Petroleum jelly is a sealing material, so it belongs wherever water loss is the problem: hands cracked from washing, worked through in the hand balm for frequent handwashing, lips, heels and skin around a nappy rash. Mineral oil is a liquid emollient that does not go off, so it belongs where a plant oil would otherwise go and the concern is shelf life, cost or allergy to plant materials: a baby massage balm, or a cleansing balm that is wiped away. The ingredients for chapped lips page covers the sealing job on lips.
For skin prone to eczema, the case for petroleum jelly is the seal, and the case for mineral oil is the monograph's oatmeal combination; whether either does more than hold water in is reviewed in balm for eczema: the evidence.
If petroleum jelly feels too heavy, use less of it rather than switching to mineral oil. Berube's data show that even a 0.1 mm film cut water loss by 77.5 to 93.0 percent. Swapping to mineral oil keeps the greasy feel at lower intensity and loses most of the seal.
What these numbers do not settle
No source here measured mineral oil and petroleum jelly as neat materials, side by side, under the same film thickness on the same people. The direct swap is in creams, with water, emulsifier and a drug present, and its outcome is reported against a threshold rather than as a percentage. The two volunteer studies used paraffin oil against plant oils with petrolatum as a benchmark, and both are available to us as abstracts only, which do not give doses or individual values. What the sources agree on is the direction, petrolatum well above mineral oil, and that mineral oil sits with most plant oils. They do not support a single number for the size of the gap, so this page does not give one.
The film thickness data come from four people, and the oil content series from three, in a 1971 laboratory with its own instrument, and the occlusive thickness is an extrapolation to zero water loss, not a measured film. Both the 1971 and 1988 studies had authors at companies that made or used these materials, Chesebrough-Pond's and Bristol-Myers, though neither was comparing products for sale.
Mineral oils differ in viscosity, as the Sethi review notes, and petroleum jellies differ in oil content, as Berube's table shows. None of the sources compares a light and a heavy mineral oil for occlusion, so this page does not say whether a heavier grade closes any of the gap. Finally, the monograph bands are US law only. A balm sold in the UK or EU is judged under those markets' own cosmetic rules, including a safety assessment, whatever its levels of either.
Frequently asked questions
Is mineral oil the same as petroleum jelly?
They come from the same refining and are both saturated hydrocarbons, but they are not interchangeable. Mineral oil is a liquid; petroleum jelly is mineral oil held in a semi-solid network, and samples in one study were 52.8 to 74.6 percent oil. In a test that swapped one for the other at equal weight in a cream, petrolatum made the cream occlusive and mineral oil left water loss close to bare skin.
Which is better for dry skin, mineral oil or petroleum jelly?
Petroleum jelly. A 0.1 mm film of white petrolatum cut water loss by 77.5 to 93.0 percent in four people, and studies that tested paraffin oil found it gave similar occlusion to most vegetable oils, with petrolatum the most effective. Mineral oil is a good emollient that never goes rancid, but it is not a strong seal.
Is baby oil as good as Vaseline for dry skin?
Not for holding water in. Baby oil is usually mineral oil with fragrance, and the studies here place mineral oil with most plant oils for occlusion, well below petroleum jelly. Baby oil spreads and glides better, which suits massage, but for cracked or very dry skin petroleum jelly is the stronger choice.
Does adding mineral oil to petroleum jelly make it less effective?
Not in modest amounts, on the evidence available. A 1988 study found that raising mineral oil from 0 to 10 percent at a constant level of petrolatum did not materially affect water loss, and a 1971 study found that white petrolatums with more oil content needed a thinner film for the same seal. Replacing the petrolatum with mineral oil is what loses occlusion.
How much mineral oil or petrolatum does a skin protectant need?
Under the US monograph, 21 CFR 347.10, petrolatum and white petrolatum count as skin protectant actives at 30 to 100 percent. Mineral oil needs 50 to 100 percent, or 30 to 35 percent when combined with colloidal oatmeal at 0.003 percent or more, a combination labelled for itching from rashes or eczema.
Do mineral oil and petroleum jelly penetrate the skin?
Studies that measured paraffin oil found it in the upper stratum corneum, at the same depth as the vegetable oils tested, with deeper penetration excluded. Petrolatum was used in those studies as the benchmark for occlusion, and its depth was not reported there.
Sources and further reading
- Berube GR, Messinger M and Berdick M, Measurement in vivo of transepidermal moisture loss, Journal of the Society of Cosmetic Chemists 22(6): 361-368, 1971. Tables III and IV.
- Wang JCT, Winarna SR, Lichtin JL and Patel BG, Effect of phase-volume ratio of o/w emulsion vehicles on the activity of a topically applied vasoconstrictor, Journal of the Society of Cosmetic Chemists 39(2): 139-157, 1988.
- Patzelt A, Lademann J, Richter H and colleagues, In vivo investigations on the penetration of various oils and their influence on the skin barrier, Skin Research and Technology 18(3): 364-369, 2012.
- Stamatas GN, de Sterke J, Hauser M, von Stetten O and van der Pol A, Lipid uptake and skin occlusion following topical application of oils on adult and infant skin, Journal of Dermatological Science 50(2): 135-142, 2008.
- Sethi A, Kaur T, Malhotra SK and Gambhir ML, Moisturizers: the slippery road, Indian Journal of Dermatology 61(3): 279-287, 2016.
- US Food and Drug Administration, 21 CFR Part 347, Skin protectant drug products for over-the-counter human use, sections 347.10, 347.20 and 347.50, eCFR, read October 2026.
Reviewed and updated 12 October 2026. Spotted an error? Tell us and we will fix and log it.