Is silicon dioxide bad for you? What silica in supplements does, in real numbers

Share
Summarize with AI

The silica with a bad name is crystalline dust breathed deep into the lungs in quarries, on building sites and in stone benchtop workshops. The silica in a supplement is a fluffy, amorphous powder that you swallow a few milligrams at a time, and your gut absorbs about 1% of it 4. Those two facts settle most of the worry about silicon dioxide. The one open question, about nano-sized particles, is real but small, and the preclinical doses behind it are thousands of times higher than any capsule.

What silicon dioxide is, and the names it goes by

Silicon dioxide is a compound of one silicon atom and two oxygen atoms, written SiO2 and usually called silica. Quartz and sand are crystalline silica, meaning the atoms sit in a regular, repeating lattice. The additive in food and supplements is synthetic amorphous silica: amorphous means the atoms have no repeating order. Manufacturers make it either by burning a silicon compound in a flame (fumed silica) or by settling it out of a solution (precipitated silica). The formula is the same as sand; the structure is different, and the structure decides how the particles behave in the body.

What the label saysWhat it means
Silicon dioxideThe standard name for the additive
Colloidal silicon dioxideVery fine amorphous silica, the usual grade in capsules and tablets
SilicaThe everyday name for silicon dioxide
Fumed silica, precipitated silicaThe two ways synthetic amorphous silica is made
E551The EU food-additive number for silicon dioxide

Two look-alike names cause most of the confusion. Silicone is a rubbery polymer used in bakeware and medical implants, and it is not an ingredient in capsules. Titanium dioxide (E171) is a white colouring, and it is the additive the EU withdrew from food in 2022 under Commission Regulation (EU) 2022/63, after EFSA said it could not rule out genotoxicity (damage to DNA). Silicon dioxide was not part of that decision and remains permitted in the EU and the US.

Why silicon dioxide is in your capsules

Silicon dioxide does two jobs in a capsule. As an anti-caking agent, it stops powders from soaking up moisture from the air and clumping; powdered herbal extracts, amino acids and many minerals are hygroscopic, meaning they pull water out of the air. As a glidant, a term for an ingredient that helps powder flow, it coats each particle so the powder pours evenly into the filling machine. Capsule machines fill by volume, so powder that sticks or bridges gives some capsules more of the active ingredient and some less.

The Handbook of Pharmaceutical Excipients, the standard reference for inactive ingredients, lists colloidal silicon dioxide as a glidant at about 0.1 to 1% of a formulation's weight. In US food, the FDA caps silicon dioxide as an anti-caking agent at 2% of the food's weight 3. That 2% figure is a food rule; capsules generally need far less, because a little silica goes a long way on flow.

How much silicon dioxide you swallow, in real numbers

Two conversions turn a frightening-sounding ingredient into a number. First, silicon dioxide is only about 47% silicon by weight, because silicon (28.09) makes up less than half of the molecule's mass (60.08). Second, colloidal silica is one of the least absorbable forms of silicon ever measured: in a 2009 study that tracked how much silicon from different sources turned up in urine, about 1% of colloidal silica was absorbed 4.

From one capsule to your bloodstream

1
The capsule
A 500 mg capsule fill with 1% silicon dioxide holds 5 mg of silicon dioxide, the top of the usual glidant range.
2
The silicon in it
Silicon is about 47% of silicon dioxide by mass, so 5 mg of silicon dioxide carries about 2.3 mg of silicon.
3
What the gut absorbs
About 1% of colloidal silica is absorbed 4, so roughly 0.02 mg of silicon reaches the blood from one capsule.
4
Out again
The kidneys clear absorbed silicon into urine; urinary silicon is how researchers measure absorption in the first place 4.
Worked example using a 500 mg capsule at 1% silicon dioxide. Absorption figure from Sripanyakorn and colleagues, 2009.
0.02 mgSilicon absorbed from a typical capsule5 mg silicon dioxide, 47% silicon, about 1% absorbed

Food supplies far more. In the Framingham Offspring cohort, men took in 30 to 33 mg of silicon a day and women 24 to 25 mg, with beer, bananas and string beans the biggest sources 5. Food silicon is also much easier to absorb than the silica in a capsule: on average 41% of it turned up in urine 5. At that rate, the lower end of that diet puts about 10 mg of silicon into the blood each day, several hundred times what one capsule adds.

Source of siliconWhat it isShare absorbed
Alcohol-free beerDissolved silicon from barley and hops64%
MonomethylsilanetriolA liquid silicon supplement64%
Green beansA high-silicon vegetable44%
Orthosilicic acidThe dissolved form of silicon in water43%
Choline-stabilised orthosilicic acidA popular hair and nail supplement17%
BananasSilicon bound in a less soluble form4%
Magnesium trisilicateAn old-style antacid4%
Colloidal silicaThe silicon dioxide in capsules1%

The absorption figures in the table come from Sripanyakorn and colleagues, who measured urinary silicon after each source in healthy volunteers 4. A common error in articles about this additive is to set milligrams of silicon dioxide beside milligrams of dietary silicon as if they were the same thing, but 10 mg of silicon dioxide is 4.7 mg of silicon before absorption is even counted. Six capsules a day at 5 mg each comes to 30 mg of silicon dioxide, 14 mg of silicon and about 0.14 mg absorbed, still a small fraction of the roughly 10 mg a day absorbed from food.

Where the bad reputation comes from

Silicosis is permanent scarring of the lungs caused by breathing respirable crystalline silica, dust fine enough to reach the air sacs. It affects people who mine, quarry, sandblast, and cut or polish engineered stone benchtops. The International Agency for Research on Cancer (IARC) classifies inhaled crystalline silica from occupational sources as a Group 1 carcinogen, the category for agents that cause cancer in humans, and classifies amorphous silica as Group 3, not classifiable. OSHA limits a worker's exposure to respirable crystalline silica to 50 micrograms per cubic metre of air, averaged over an 8-hour shift.

A capsule differs from that dust on both counts. The silica is amorphous, and it goes to the gut, where it passes through or dissolves, rather than settling in lung tissue for decades. Factories that handle bulk silica powder still use dust controls, because no fine powder is good to breathe all day, but that is a workplace matter with no bearing on swallowing a capsule. The chemical-sounding name and the link to sand do the rest of the reputational damage.

What regulators concluded, and what they left open

The FDA lists silicon dioxide as a permitted food additive within the 2% cap 3. International expert committees (JECFA and the former EU Scientific Committee on Food) gave it an acceptable daily intake (ADI) of "not specified", the label for substances so low in toxicity that no numerical limit is needed. EFSA re-examined the additive in 2018, found it poorly absorbed and saw no toxicity at the levels people eat, with the highest intakes at least ten times below the doses that showed no effect in studies 1.

There was no indication for toxicity of E 551 at the reported uses and use levels.EFSA Panel on Food Additives and Nutrient Sources added to Food, 2018 1

EFSA did not stop there. In 2018 it declined to confirm the "not specified" ADI, because the official specifications for E551 did not describe particle size or rule out nanoparticles 1. The 2024 follow-up closed most of that gap. It concluded that E551 raises no safety concern for any population group at reported uses, including food for infants under 16 weeks old 2. Instead of an ADI, the panel used a margin of exposure, which measures how far the highest estimated intake sits below the dose that caused no harm in studies; it required a margin of at least 36 and every intake estimate cleared it 2. The panel also asked for tighter limits on lead, mercury and arsenic in E551 and a new maximum for aluminium 2.

BodyYearWhat it concluded
US FDACurrent rulePermitted anti-caking agent in food up to 2% of weight 3
Institute of Medicine2001No upper limit set for silicon; no harm from food silicon shown
EFSA2018Poorly absorbed, no toxicity at use levels; ADI unconfirmed over particle-size gaps 1
EFSA2024No safety concern for any age group; stricter heavy-metal and aluminium limits; nano uncertainty remains 2

The nano question

Some particles in food-grade silicon dioxide fall in the nano range, under 100 nanometres across, far smaller than a red blood cell. The worry is that particles this small might cross the gut wall and reach organs. EFSA's 2024 opinion said that a lack of studies using proper dispersion methods leaves uncertainty about nano-sized aggregates 2, so this is the one part of the safety file that regulators still call open.

Two lines of evidence frame how big that uncertainty is. In a simulated digestion model, nano-sized silica from foods containing E551 was present at the mouth stage and largely disappeared in the stomach stage, where acid and salts made the particles clump into much larger agglomerates. In the intestinal stage that followed, the nano-sized particles reappeared, in higher amounts than in the mouth, and the authors concluded that the gut lining is most likely exposed to nano-sized silica 7. In a 2014 preclinical feeding study, 84 days of 1,000 mg per kg of body weight per day of a nanostructured reference silica led to more liver fibrosis (scarring), and 2,500 mg per kg per day of food-grade silica left silica deposits in the spleen; the authors called for studies at lower doses closer to human intake 6.

0.07 mg/kgSilicon dioxide per kg of body weight from one 5 mg capsule in a 70 kg adultThe preclinical liver finding came at 1,000 mg/kg a day, more than 10,000 times higher

Set against a capsule, the preclinical doses are enormous. One 5 mg capsule gives a 70 kg adult about 0.07 mg per kg, and six capsules give about 0.43 mg per kg. The fair summary is low concern with some uncertainty left: the preclinical signal came at doses far beyond supplement use, the gut lining probably does meet some nano-sized particles, and EFSA still wants better studies on the aggregates 2.

Does silicon dioxide build up, or hurt your kidneys?

In people with working kidneys, absorbed silicon does not build up: the kidneys clear it into urine, and after colloidal silica, blood levels peaked about 2 hours after the dose 4. That is why researchers can read absorption from a urine collection. The Institute of Medicine set no upper limit for silicon in 2001, judging the evidence too thin, and noted that silicon from food had not been shown to cause harm.

The only kidney harm on record involves a different product at a different scale. In 1973, doctors described a man who had taken about 2 g of magnesium trisilicate antacid after every meal for many years and formed silica kidney stones 8. That is around 6 g of a silicate a day for years, more than a thousand times the 5 mg of silicon dioxide in a capsule.

Advanced chronic kidney disease (CKD) and dialysis change the picture for many substances, silicon included, because clearance falls as kidney function falls. For anyone on dialysis or with an eGFR under 30, the kidney team should see the full list of supplements, inactive ingredients too. Excipient silica is a minor item on that list; the active ingredients usually matter more. Our guide to kidney function explains what eGFR and creatinine measure.

CalculatoreGFR calculatorTurn a creatinine result into an eGFR to see which kidney stage it falls in.Work it out

Should you avoid silicon dioxide?

The evidence gives most people no health reason to avoid silicon dioxide in supplements. Preferring fewer additives is a reasonable personal choice, and some brands market themselves on it. The usual swaps are rice hull concentrate or rice flour as the flow agent, and oil-filled softgels, which need no flow agent at all.

Each swap has a trade-off. Rice-based flow agents are less effective, so capsule fill can vary more and some powders clump sooner. Rice hulls are naturally rich in silica, so a rice hull concentrate does not remove silica; it changes the source. "Clean label" describes a short ingredient list, and it says nothing about safety.

  • People on dialysis or with advanced CKD: worth raising every supplement, inactive ingredients included, with the kidney team.
  • People with past or current work exposure to silica dust: the lung risk comes from the dust, and a clinician should know the work history.
  • Infants: EFSA's 2024 review specifically assessed E551 in foods for infants under 16 weeks and found no safety concern at current exposure levels 2.

To check a product, read the "Other ingredients" line under the Supplement Facts panel; US labels list those ingredients from most to least by weight, so silicon dioxide usually sits near the end. Our guide to supplements covers the rest of the label.

Silica as the main ingredient is a different question

The safety reviews of silicon dioxide cover silica as an excipient, a few milligrams added to help manufacturing. Some products sell silicon itself as the active ingredient: orthosilicic acid and horsetail extract for hair, nails and bone, and gram-dose porous silica marketed for blood sugar and weight. Those are different doses with different evidence, and the excipient safety data cannot vouch for them.

Silica as an excipientSilicon or silica as the active ingredient
Typical amountMilligrams per capsuleMilligrams of dissolved silicon up to grams of porous silica
Why it is therePowder flow and anti-cakingMarketed for hair, nails, bone, blood sugar or weight
How much is absorbedAbout 1% for colloidal silica 417 to 43% for orthosilicic acid forms 4
Evidence behind itSafety reviews by the FDA and EFSA 2 3Observational links to bone density and small, early human studies

Anyone taking glucose-lowering medicine who is considering a product sold for blood sugar should raise it with the clinician who manages that medicine first, because added effects on glucose can push levels too low.

Can a blood test show whether silicon dioxide is affecting you?

No routine blood test measures silicon, and normal kidney or liver numbers cannot vouch for a single inactive ingredient. A blood test is useful for two other supplement questions: whether the active ingredient is registering (vitamin D, B12 or ferritin, for example), and whether a supplement is skewing a result. Biotin is the best-known example; high doses can throw off some thyroid and heart tests, as our guide to biotin and blood tests explains, and the wider list is in supplements that affect blood tests.

Free, no accountRead your blood testTaking supplements and holding a recent blood test? Read every line of it, free.Read my results

For silicon dioxide itself, the useful number is about 0.02 mg of absorbed silicon per capsule, against roughly 10 mg a day from ordinary food.

Free, no accountDrop in your lab report and see every marker explained in plain words, placed on the body, against your lab's range. It's read on your device: nothing is uploaded and nothing is kept.
The distinctions
  1. 01
    Form and route decide the risk

    The silica that causes silicosis and lung cancer is crystalline dust breathed into the lungs at work. The silica in supplements is amorphous powder that is swallowed, and only about 1% of it is absorbed, so a capsule adds about 0.02 mg of silicon to the 24 to 33 mg a day that food provides.

Citations
  1. EFSA ANS Panel (2018). Re-evaluation of silicon dioxide (E 551) as a food additive. EFSA Journal 16(1):5088
  2. EFSA FAF Panel (2024). Re-evaluation of silicon dioxide (E 551) in foods for infants below 16 weeks and follow-up for all population groups. EFSA Journal 22(10):e8880
  3. US FDA. 21 CFR 172.480 Silicon dioxide
  4. Sripanyakorn S et al. (2009). The comparative absorption of silicon from different foods and food supplements. Br J Nutr 102(6):825-834
  5. Jugdaohsingh R et al. (2002). Dietary silicon intake and absorption. Am J Clin Nutr 75(5):887-893
  6. van der Zande M et al. (2014). Sub-chronic toxicity study in rats orally exposed to nanostructured silica. Part Fibre Toxicol 11:8
  7. Peters R et al. (2012). Presence of nano-sized silica during in vitro digestion of foods containing silica as a food additive. ACS Nano 6(3):2441-2451
  8. Joekes AM, Rose GA, Sutor J (1973). Multiple renal silica calculi. BMJ 1(5846):146-147

Educational context only, not medical advice or a diagnosis. Always discuss your results with a clinician.