What Is Silica Good For? Trials, Forms and the UK Rules
Silica, meaning the dietary silicon in food, water and supplements, has no authorised health claim in Great Britain: the register lists 23 silicon wordings (collagen and connective tissue, bone, hair and nails, joints, skin, aluminium in the brain, cardiovascular health, stomach, immune cells) and every one is Non-authorised, and there is no nutrient reference value and no "source of silicon" claim either.1 The trial record is short, all of one form at 3 to 12 mg of silicon a day: forearm skin roughness fell over 20 weeks in 50 women, hair fibres held their strength over 9 months in 48 women, a bone-collagen marker rose over 12 months in 136 women while spine density did not change, and 166 people with knee osteoarthritis showed no overall difference.2, 3, 4, 5 This is the silica hub within the vitamins and minerals guide.
What is silica, and is it the same as silicon or silicone?
Silicon is an element that "occurs chiefly as the oxide and silicates". The UK expert group's definition: "The term 'silica' is used to refer to naturally occurring materials composed principally of silicon dioxide (SiO2), whereas 'silicone' (organosiloxane) refers to man-made siloxane polymers".6 So silica is silicon dioxide, and silicone, the material in bakeware, implants and the replicas that measure wrinkle depth in collagen skin trials, is a manufactured polymer no supplement contains.
The silica in this article is dietary silicon: the silicon in food, water and supplements, which is swallowed, dissolved and passed out in urine. It is not crystalline silica dust, which causes silicosis when breathed in at work and which the HSE and the food regulators assess separately, and it is not the silica gel in desiccant sachets, which is not poisonous if swallowed but whose packet can choke a small child (call NHS 111, or 999 if someone is choking).6, 7, 8 Homeopathic "silicea" and tissue-salt products are highly diluted preparations sold as remedies; they are not a dietary source of silicon and this article is not about them.
Is silicon a nutrient you need?
Not officially. The European committee: "The essentiality of silicon for man has not been established and a functional role for silicon in humans has not yet been identified"; "A Recommended Intake for silicon has not been established due to insufficient data and lack of functional criteria".9 The UK expert group: "Silicon deficiency has not been observed in humans"; "Methods of assessing silicon status have not been established".6 The case for essentiality is animal: chicks on a low-silicon diet "showed significantly retarded growth and development within 2 to 3 weeks".10 One review suggests "near 25mg/d would be a reasonable suggestion for an adequate intake", but no official body has adopted a figure, so there is no reference intake, no "% NRV" on any label, and no recognised deficiency state or test.11 Thinning hair or brittle nails have causes a GP can test for (iron, thyroid, vitamin D, zinc), not a silica level.
What may a label say about silicon in Great Britain?
"Only authorised claims in the Great Britain NHC register may be used in Great Britain." Its 23 silicon rows, covering silicon, orthosilicic acid, "silica / silicious earth" and choline-stabilised orthosilicic acid, are all Non-authorised.1
| Claimed effect | Register IDs | Reason recorded |
|---|---|---|
| Contribution to normal formation of collagen and connective tissue | 287, 288, 333, 334, 335, 1652, 1718, 1719 | "has not been substantiated" |
| Maintenance of normal bone | 287, 335, 1652, 1718 | "has not been substantiated" |
| Contribution to normal formation of hair and nails | 334, 1652, 1719 | "has not been substantiated" |
| Maintenance of normal joints | 1652 | "has not been substantiated" |
| Maintenance of normal appearance and elasticity of the skin | 288, 333 | "has not been linked to a function of the body, as required by Article 13" |
| Protection against aluminium accumulation in the brain | 290 | "has not been substantiated" |
| Cardiovascular health; a protective coat on the stomach; stimulating immune cells | 289, 345, 358 | "not sufficiently defined to be able to be assessed"; "not a beneficial physiological effect"; Non-authorised |
The rejected consumer phrases include "helps improve hair thickness" and "helps maintain strong bones".1 Lawful hair, skin, nail and bone wording belongs to other nutrients, where a serving supplies 15% of the reference value: zinc contributes to the maintenance of normal skin, hair and nails; biotin and selenium to the maintenance of normal hair; copper to maintenance of normal connective tissues; vitamin D to the maintenance of normal bones; vitamin C to normal collagen formation for the normal function of skin, bones and cartilage.1
What have the trials actually measured?
Five small controlled studies, four of one form, choline-stabilised orthosilicic acid (ch-OSA); doses are milligrams of silicon.
| Study | Form, dose, length | People | Measured | Found |
|---|---|---|---|---|
| Barel 20052 | ch-OSA, 10 mg a day, 20 weeks | 50 women with photodamaged skin | Forearm roughness by instrument; self-rated brittleness | Roughness rose on placebo (Rt +8%, Rm +11%) and fell on silicon (Rt −16%, Rm −19%), two of three parameters significant; brittleness fell against baseline only |
| Wickett 20073 | ch-OSA, 10 mg a day, 9 months | 48 women with fine hair | Hair fibre strength and thickness | Break load fell 10.8% on placebo and 2.20% on silicon; cross-section rose against baseline; growth and shedding not measured |
| Spector 20084 | ch-OSA, 3, 6 or 12 mg a day with calcium and vitamin D3, 12 months | 184 women with osteopenia, 136 completed | Bone markers; spine and hip density | PINP up at 6 and 12 mg "without a clear dose response effect"; "Lumbar spine BMD did not change significantly"; a post-hoc femoral-neck finding; "no ch-OSA related adverse events" |
| Geusens 20175 | ch-OSA, 12 weeks | 211 adults with knee osteoarthritis, 166 analysed | WOMAC pain, stiffness, function | "no differences were observed between the two treatment groups" overall; men on silicon improved in a pre-specified gender analysis (46 men in all) |
| Li 201012 | Water at 86 mg/L silica, 1 litre a day, 12 weeks | 17 postmenopausal women | Urinary silicon; bone markers | Urinary silicon rose from 0.016 to 0.037 mg/mg creatinine; bone markers unchanged |
The four supplement trials share authors, and none has been repeated independently: a fact about the literature, not about the form.2, 3, 4, 5 None measured hair growth, shedding, acne, facial skin or fracture.
Does silica help bones?
The observations are larger than the trial. In 2,847 adults of the Framingham Offspring cohort, "Dietary silicon correlated positively and significantly with BMD at all hip sites in men and premenopausal women, but not in postmenopausal women", with differences "up to 10%" between intakes above 40 mg a day and below 14 mg; the study was cross-sectional, and the authors noted that moderate alcohol's apparent effect on BMD "might be attributed to Si intake", beer being a main source.13 In 3,198 Scottish women aged 50 to 62, femoral-neck density was 2% lower in the bottom silicon quartile (16 mg a day) than the top (31.5 mg), "for oestrogen-replete women only".14 In 400 Korean men aged 19 to 25, "daily total silicon intake had no correlation with calcaneus bone density".15 In the laboratory, orthosilicic acid at 10 and 20 µM increased collagen type 1 synthesis 1.45- to 1.8-fold in human osteoblast-like cells; the committee holds that "a functional role for silicon in humans has not yet been identified", and the register refused the bone wordings.16, 9, 1 No silicon trial has been run in osteoporosis, a GP-managed condition; can you rebuild bone density and what is osteopenia cover the nutrients with authorised bone wordings.
Does silica help skin, hair and nails?
The two trials in the table are the whole record: forearm roughness in 50 women over 20 weeks, with brittleness scored against baseline only, and hair-fibre strength in 48 women over 9 months, with growth and shedding not measured.2, 3 No trial of any silicon form has measured hair loss, acne, pores or facial skin, and the register refused the hair, nail, wrinkle and elasticity wordings.1 Shedding has causes a GP can test for; collagen for hair and nails carries the NHS list and the silica FAQs; is biotin good for hair growth covers the biotin wording.
Silica or collagen: do you need both?
They are different things. Collagen is a protein, taken in grams, with its own trial record summarised in what collagen does to the body. Silicon is a trace element, taken in milligrams, with the trials above. No trial has compared them or tested them together, so nothing can be said about one adding to the other; taking both is a label question, each at its own dose, with no interaction documented. Products sold as "collagen builders" pair silica with vitamin C, zinc and copper, whose wordings are their own; is there a vegan collagen explains what those packs contain.
How does the body handle silicon, and does the form matter?
Silicon crosses the gut as dissolved orthosilicic acid and leaves in urine within a day: in 8 volunteers, median uptake was 50.3% (range 21.9 to 74.7%), "readily absorbed from the gastrointestinal tract of man and then readily excreted in urine".17 The one head-to-head study gave 32 fasted adults aged 19 to 40, in sub-studies of 3 to 14, two sources a week apart at the makers' maximum doses, measuring six-hour urinary silicon. Recovery was 64% for monomethylsilanetriol (6.9 mg of silicon, 14 people), 64% for alcohol-free beer, 44% for green beans, 43% for an orthosilicic acid solution, 17% for a ch-OSA concentrate (20 mg, 8 people, 1 ml diluted in 199 ml of water and drunk at once), 4% for bananas and for a magnesium trisilicate antacid (200 mg, 8 people) and 1% for a colloidal silica (780 mg, 3 people). The authors' rule: "Monomeric silicates were readily absorbed, while particulate silicates were decreasingly well absorbed with increasing polymerisation."18 Doses differed more than a hundredfold and the window was six hours, so the percentages are not dose-matched. Over weeks, 10 mg a day of ch-OSA raised serum silicon in 50 women and urinary silicon in 48, and 10.5 mg a day of monomethylsilanetriol for 4 weeks raised both in 22 women with "no reported adverse effects".2, 3, 19 Silicon from solid foods is absorbed once it dissolves: "a mean of 41% of the ingested silicon was excreted in urine".20 Bamboo extract has not been measured in people. Retained supplement law lists choline-stabilised orthosilicic acid, silicon dioxide, silicic acid and organic silicon (monomethylsilanetriol) as silicon sources; bamboo and horsetail are botanicals, not listed sources.21 How much a product delivers depends on how readily its silicon dissolves into orthosilicic acid, measured for some sources and not others; absorption is a measurement, not a health outcome. Types of silica explained sets out each form and its study.
How much silicon do people get, and from what?
In 209 Scottish women over 60, mean intake was 18.6 mg a day, cereals providing about 30%.22 The European committee's British estimate is 20 to 50 mg a day, "55% from water, coffee and beer, 14% from grain products and 8% from vegetables"; Framingham men took in 30 to 33 mg a day and women 24 to 25, "beer and bananas in men and bananas and string beans in women" the main sources.9, 20 The UK database of 207 foods found "The highest concentrations of Si were found in cereals and cereal products, especially less refined cereals and oat-based products", "substantial amounts" in Kenyan, French and runner beans, spinach, dried fruit, bananas and red lentils, "undetectable amounts in tomatoes, oranges and onions", and beer the richest drink.23 Content and absorption differ: bananas gave 4% in the head-to-head study, green beans 44%.18 Horsetail is a herb sold as a silicon source; it is a traditional herbal medicine in its own right, with its own cautions, and is not covered here. Foods high in silica has the UK portions.
Is it safe to take silica every day?
The European committee: "In humans, apart from occasional reports of renal stones, mainly associated with long-term use of silicate-containing antacids, there is little evidence of adverse effects of orally ingested silicon. The available data are inadequate to derive a tolerable upper intake level." Typical intakes of 20 to 50 mg a day "are unlikely to cause adverse effects".9 The UK expert group, working from animal data, set a safe upper level for supplemental silicon of 700 mg a day for a 60 kg adult "for daily consumption over a lifetime", and noted that "UK food supplements contain up to 500 mg silicon".6 The European panel's missing ceiling reflects thin data, not proof that any amount is fine. In the trials, 3 to 12 mg a day for 12 months produced "no ch-OSA related adverse events" and 10.5 mg a day for 4 weeks none.4, 19 Three groups ask first: people with kidney disease, because "Patients with renal failure generally have a higher concentration of silicon in plasma", so a GP decides;9 pregnant and breastfeeding women, because no supplement trial exists, so ask a GP or midwife; and anyone on regular medicines, because herbal silicon products carry their own notes, so a pharmacist checks. There is no reference intake to meet, so no amount of silica, bamboo or otherwise, is "needed"; the figure that matters on a label is milligrams of silicon, which not every label states. Is it safe to take silica every day has the detail. Ring 111 out of hours, and 999 for swelling of the face or throat or difficulty breathing.
Frequently asked questions
Is silicon dioxide the same as silica?
Yes. On an ingredients list it is usually E551, an anti-caking agent that EFSA found "appears to be poorly absorbed" with "no indication for toxicity of E 551 at the reported uses and use levels".24 It is an additive, not a silicon dose.
When is the best time to take silica, and how long does it take to work?
No trial has compared times of day; the outcome trials gave the dose daily for 20 weeks to 12 months, so follow the label.2, 3, 4 The trials measured instruments and markers at those endpoints, not visible change, so "how long" has no tested answer.
Does silica make you look younger?
The register refused the wrinkle and elasticity wordings because the effect "has not been linked to a function of the body".1 The one skin trial measured forearm roughness by instrument, not facial appearance, and no trial has studied sagging or crepey skin.2
Can I put silica on my face, and will it clog pores?
Silica in make-up and skincare is a texture ingredient under cosmetics law, a different use from dietary silicon. No trial here measured acne or pores; persistent spots are a pharmacist or GP question.
Does silica increase collagen?
In human osteoblast-like cells, orthosilicic acid raised collagen type 1 synthesis 1.45- to 1.8-fold, and in 136 women a collagen-formation marker rose at 6 and 12 mg a day; the register judged the collagen wording "has not been substantiated".16, 4, 1 Collagen supplements supply the protein itself.
Sources
- Great Britain nutrition and health claims (NHC) register, silicon entries, 19 May 2026. gov.uk.
- Barel A, et al. Choline-stabilized orthosilicic acid and skin, nails and hair. Arch Dermatol Res, 2005. PubMed 16205932.
- Wickett RR, et al. Choline-stabilized orthosilicic acid and hair tensile strength. Arch Dermatol Res, 2007. PubMed 17960402.
- Spector TD, et al. Choline-stabilized orthosilicic acid in osteopenic females. BMC Musculoskelet Disord, 2008. PubMed 18547426.
- Geusens P, et al. Choline-stabilized orthosilicic acid in knee osteoarthritis. BMC Musculoskelet Disord, 2017. PubMed 28056936.
- Expert Group on Vitamins and Minerals. Safe Upper Levels for Vitamins and Minerals, 2003. UK Government Web Archive.
- Health and Safety Executive. Silicosis. hse.gov.uk.
- Lavon O, Bentur Y. Silica gel: non-toxic ingestion. Isr Med Assoc J, 2015. PubMed 26665312.
- Scientific Committee on Food. Tolerable upper intake level of silicon, 2004. efsa.europa.eu.
- Carlisle EM. Silicon: an essential element for the chick. Science, 1972. PubMed 5086395.
- Nielsen FH. Update on the possible nutritional importance of silicon. J Trace Elem Med Biol, 2014. PubMed 25081495.
- Li Z, et al. Silicon from artesian aquifer water and bone health in postmenopausal women. Nutr J, 2010. PubMed 20946656.
- Jugdaohsingh R, et al. Dietary silicon intake and bone mineral density, Framingham Offspring cohort. J Bone Miner Res, 2004. PubMed 14969400.
- Macdonald HM, et al. Dietary silicon interacts with oestrogen to influence bone health. Bone, 2012. PubMed 22173054.
- Choi MK, Kim MH. Dietary silicon intake of Korean young adult males and bone status. Biol Trace Elem Res, 2017. PubMed 27525412.
- Reffitt DM, et al. Orthosilicic acid stimulates collagen type 1 synthesis in human osteoblast-like cells. Bone, 2003. PubMed 12633784.
- Reffitt DM, et al. Silicic acid: gastrointestinal uptake and urinary excretion in man. J Inorg Biochem, 1999. PubMed 10612067.
- Sripanyakorn S, et al. The comparative absorption of silicon from different foods and food supplements. Br J Nutr, 2009. PubMed 19356271.
- Jugdaohsingh R, et al. Monomethylsilanetriol is safe and increases the body pool of silicon. Nutr Metab, 2013. PubMed 23622499.
- Jugdaohsingh R, et al. Dietary silicon intake and absorption. Am J Clin Nutr, 2002. PubMed 11976163.
- Directive 2002/46/EC on food supplements, Annex II (retained law). legislation.gov.uk.
- McNaughton SA, et al. Dietary silicon intake in post-menopausal women. Br J Nutr, 2005. PubMed 16277786.
- Powell JJ, et al. A provisional database for the silicon content of foods in the United Kingdom. Br J Nutr, 2005. PubMed 16277785.
- EFSA ANS Panel. Re-evaluation of silicon dioxide (E 551) as a food additive. EFSA Journal, 2018. PubMed 32625658.
This article is general information, not medical advice. Kidney disease, pregnancy or a regular medicine mean asking a GP, midwife or pharmacist first.