⚠ Evidence warning: Clinical trials or other evidence do not support the claim that silicon at typical Western dietary intakes is a major reason historical low-calcium populations kept healthy bones, or that dietary silicon is a proven part of osteoporosis prevention. This page keeps it as a report of what people, websites or other media say.

Silicon for Bone Density

Barley, a grain rich in silicon
Barley, a grain rich in silicon

The single most influential observational study of silicon and bone health is Jugdaohsingh, Tucker, Qiao and colleagues' analysis of the Framingham Offspring cohort — 2,847 community-dwelling adults whose dietary silicon intake was estimated from food-frequency questionnaires and whose hip bone mineral density (BMD) was measured by dual-energy X-ray absorptiometry (DXA). Adults in the highest quintile of dietary silicon intake (over 40 mg/day) had hip BMD up to 10% higher than adults in the lowest quintile (under 14 mg/day) — a magnitude comparable to the effect of weight-bearing exercise or vitamin D supplementation. The link was seen in men and premenopausal women but not in postmenopausal women, and a later study in Aberdeen tied that difference to estrogen status. This page walks through the Framingham finding in detail, the silicon-collagen-vitamin-D-K2 axis in bone matrix, the unusually high silicon bioavailability from beer, and what the one supplement trial and the animal studies have and have not shown.


Table of Contents

  1. The Bone Matrix Two-Phase Architecture
  2. The Framingham Offspring Silicon-BMD Study
  3. Estrogen Modulation of the Silicon Effect
  4. Bioavailability: Beer, Mineral Water, Whole Grains
  5. The Silicon-Aluminum Interaction
  6. The Silicon-Vitamin-D-Vitamin-K2 Axis
  7. Osteoblast Mechanism: Stimulation of Bone Formation
  8. Clinical Trials in Osteopenic and Osteoporotic Patients
  9. Bone-Density Nutrients: Amounts in Studies and Reference Values
  10. Cautions
  11. What Is Not Yet Known About Silicon and Bone
  12. Key Research Papers
  13. Connections
  14. Featured Videos

The Bone Matrix Two-Phase Architecture

Bone is a composite material with two distinct phases: a flexible organic matrix (roughly 30% of dry bone weight) and a hard mineral phase (roughly 70%). The organic matrix is mostly type I collagen with smaller contributions from osteocalcin, osteonectin, osteopontin, bone sialoprotein, and proteoglycans. The mineral phase is hydroxyapatite, Ca10(PO4)6(OH)2, deposited as plate-like crystals along the collagen fibrils.

The two phases serve different mechanical functions. The collagen matrix provides tensile strength and flexibility — it is the rope that resists pulling apart. The hydroxyapatite provides compressive strength and rigidity — it is the cement that resists crushing. A bone with normal mineral density but weakened collagen matrix can be just as fracture-prone as a bone with weakened mineralization. This is why BMD measured by DXA does not fully predict fracture risk: it measures the mineral phase but ignores the collagen quality. The clinical mismatch between BMD and fracture in atypical femur fractures associated with long-term bisphosphonate use reflects exactly this disconnect — bisphosphonates preserve mineral density at the cost of collagen turnover quality.

Silicon supports both phases. It is required for the collagen biosynthesis pathway (through prolyl hydroxylase activation, see Connective Tissue) and it appears to participate in hydroxyapatite nucleation at the mineralization front, where new mineral is deposited on the collagen template. The Carlisle electron-microprobe studies of the 1970s first identified silicon enrichment specifically at sites of active osteoid mineralization in developing bone, with silicon concentration dropping as mineralization completed. Silicon thus functions as both a collagen-synthesis cofactor and a mineralization-localization signal — a unique dual role among trace minerals.

Back to Table of Contents

The Framingham Offspring Silicon-BMD Study

The Framingham Offspring Cohort is the second-generation arm of the original Framingham Heart Study, started in 1971 to study cardiovascular risk in the children of the original Framingham participants. By the late 1990s, the cohort was middle-aged and had been extensively phenotyped for cardiovascular outcomes; the silicon and BMD analysis was a secondary investigation using existing food-frequency questionnaire data and DXA measurements taken in 1996-2001.

The study was led by Ravin Jugdaohsingh of the Gastrointestinal Laboratory at the Rayne Institute, St Thomas' Hospital, London, with co-authors including Katherine Tucker, Douglas Kiel and Jonathan Powell. The study (published in Journal of Bone and Mineral Research, 2004) included 1,251 men and 1,596 women with usable dietary silicon estimates and hip and lumbar spine DXA. The major findings:

An earlier analysis of the original Framingham and Framingham Offspring cohorts (Jugdaohsingh 2002) estimated average silicon intake at 30 and 33 mg/day in men and 24 and 25 mg/day in women, with intake falling with age. The main food sources were beer and bananas in men, and bananas and string beans in women. On average 41% of the silicon eaten in foods was excreted in urine, the study's measure of absorption, and the authors concluded that solid foods are a major source of available silicon.

The study was observational and so cannot prove causation, but the large BMD differences between the highest and lowest of the five intake groups (quintiles), the consistency across both men and premenopausal women, and the biological plausibility of the silicon-collagen mechanism made the Framingham finding the strongest single piece of evidence in the silicon-bone literature. A later observational study in Aberdeen, UK (Macdonald 2012) found a similar link in women with adequate estrogen, and a Korean study in ovariectomized rats (Kim 2009) found that added silicon raised femur and tibia BMD only in calcium-deficient animals.

Back to Table of Contents

Estrogen Modulation of the Silicon Effect

The finding that the silicon-BMD association was absent in postmenopausal women prompted a focused follow-up study by Macdonald, Hardcastle, Jugdaohsingh and colleagues using the Aberdeen Prospective Osteoporosis Screening Study. In 3,198 women aged 50-62, average femoral-neck BMD was 2% lower in the quarter of women with the lowest silicon intake (about 16 mg/day) than in the quarter with the highest (about 31.5 mg/day). The link held only in women with adequate estrogen: late premenopausal women (r=+0.21) and women on HRT (r=+0.09), while estrogen-depleted women showed no such association. The interaction between estrogen status and silicon intake stayed significant after adjustment for other factors. Higher silicon intake also went with lower urine markers of bone breakdown and a higher blood marker of bone formation. The authors concluded that estrogen status appears important for how silicon acts in bone, and that further work is needed to explain the mechanism.

The proposed mechanism is that estrogen is required for orthosilicic acid to exert its full osteoblast-stimulating effect. In cell culture, the combination of estradiol plus orthosilicic acid produces synergistic increases in osteoblast collagen synthesis greater than either agent alone. The molecular pathway is not resolved. In the Reffitt 2003 cell study, silicon raised collagen type 1 production without changing the level of collagen type 1 messenger RNA (mRNA), so silicon appears to act after the collagen gene is read, through the collagen-finishing enzyme prolyl hydroxylase, rather than by switching the gene on.

One reading of these results is that silicon alone may do less for bone in postmenopausal women with low estrogen. Sources of estrogen or estrogen-like activity discussed in this context are hormone replacement therapy (including estradiol patches) and phytoestrogen-rich foods such as flaxseed, a source of lignans. The silicon-estrogen interaction is also relevant to interpretation of trial data: trials in osteopenic or osteoporotic women (e.g. Spector 2008) typically include calcium and vitamin D as background and may underestimate the silicon effect if endogenous estrogen is uniformly low across both arms.

Men do not face the equivalent issue because testosterone aromatizes to estradiol throughout adult life, maintaining a baseline estrogenic signal in male bone.

Back to Table of Contents

Bioavailability: Beer, Mineral Water, Whole Grains

Dietary silicon comes from three principal sources: cereal grains, beverages (beer, mineral water, coffee), and root and leafy vegetables. The bioavailability differs dramatically by source. Sripanyakorn and colleagues at the Rayne Institute, St Thomas' Hospital, London, measured the urinary recovery of silicon from each source as a proxy for absorption:

A 2009 comparison by the same London group (Sripanyakorn 2009) measured absorption, based on urinary silicon, from eight high-silicon sources: 64% from alcohol-free beer and from monomethyl silanetriol (a supplement form), 44% from green beans, 43% from an orthosilicic acid solution, 17% from choline-stabilized orthosilicic acid (ch-OSA, a supplement form), 4% from bananas and from a magnesium trisilicate antacid, and 1% from colloidal silica. Silicon in its single-molecule (monomeric) form was readily absorbed, while particle forms were absorbed less the more polymerized they were.

The upshot: a diet that includes oatmeal, high-silicon mineral water and some beer delivers substantially more bioavailable silicon than a typical low-grain Western pattern. The Framingham silicon quintile-1 vs quintile-5 contrast (under 14 vs over 40 mg/day) roughly separates diets low in grains and beer from diets rich in whole grains that include beer, and the Framingham authors noted that some of the bone benefit seen with moderate drinking of alcoholic beverages might come from their silicon.

For non-drinkers and people who leave out beer for caloric or health reasons, high-silicon mineral water (1 liter per day) plus daily oatmeal delivers roughly equivalent silicon without alcohol.

Back to Table of Contents

The Silicon-Aluminum Interaction

One of silicon's most useful properties is its competitive binding with aluminum in the gastrointestinal tract and renal tubule. Orthosilicic acid forms hydroxyaluminosilicate complexes with Al3+ ions, reducing aluminum absorption across the intestine and increasing urinary aluminum excretion when aluminum is already in circulation. This effect was first demonstrated in the 1970s by Birchall and colleagues at ICI Mond Research, and subsequently developed by Exley and Birchall into a hypothesis-driven research program on silicon as a natural aluminum chelator.

The relevance for bone is that aluminum is itself bone-toxic. Aluminum interferes with hydroxyapatite crystal formation, accumulates in the mineralization front, and produces a distinctive osteomalacia pattern (aluminum bone disease) most clinically familiar from dialysis patients in the 1970s and 80s before phosphate binders were switched away from aluminum hydroxide. Even at sub-clinical exposure levels, aluminum may contribute to age-related bone fragility, particularly in individuals with chronic kidney disease who cannot effectively excrete aluminum.

Silicon's protective effect on bone may therefore have two components: direct osteoblast stimulation and collagen support (the primary effect), plus indirect protection from aluminum-mediated bone matrix interference (a secondary effect). The Davenward 2013 Alzheimer pilot demonstrated that the silicon-aluminum binding works in vivo at the doses obtainable from drinking high-silicon mineral water — an important proof-of-concept for the dietary approach.

Aluminum exposure in modern populations comes primarily from antiperspirants, aluminum cookware in acidic dishes (tomato sauce, coffee), aluminum-containing antacids (such as aluminum-magnesium antacids), aluminum food additives (some baking powders, processed cheese), and tap-water aluminum sulfate flocculants. Less exposure to these sources lowers aluminum intake, and dietary silicon, by the binding described above, may reduce how much of the remainder is absorbed.

Back to Table of Contents

The Silicon-Vitamin-D-Vitamin-K2 Axis

No single nutrient builds bone in isolation. The bone-building network includes calcium (the mineral substrate), vitamin D (calcium absorption), vitamin K2 (osteocalcin carboxylation and calcium routing to bone), magnesium (cofactor for vitamin D activation and bone matrix), boron (osteoblast function), and silicon (collagen and mineralization). Nutrition approaches to osteoporosis prevention commonly address all six.

The presence of silicon at typical Western dietary levels is a major reason why historical populations with much lower calcium intake than modern Western recommendations (1000-1200 mg/day) nevertheless maintained healthy bone — their grain-and-mineral-water-rich diets delivered abundant silicon and the full bone-building cofactor network.

Back to Table of Contents

Osteoblast Mechanism: Stimulation of Bone Formation

The Reffitt 2003 in vitro study established the direct osteoblast-stimulating effect of orthosilicic acid, tested at 0-50 micromolar (blood levels are normally 5-20 micromolar). Human osteoblast-like cells (the MG-63 line, plus two other human osteoblast-like cell types) exposed to orthosilicic acid showed:

The molecular mechanism is incompletely characterized but appears to act directly on osteoblasts rather than through calcium or phosphate handling. For collagen, the effect worked through prolyl hydroxylase, after the collagen gene is read; the genes for alkaline phosphatase and osteocalcin did show increased expression. The collagen effect was seen in all three cell types tested.

The evidence on bone breakdown (resorption) is mixed. In vitro studies of osteoclast precursors (RAW264.7 line) exposed to orthosilicic acid are described as showing no increase in osteoclast differentiation or bone resorption activity, and some studies suggest mild osteoclast inhibition. In calcium-deficient ovariectomized rats, added silicon raised femur and tibia BMD by reducing bone resorption, with no change in alkaline phosphatase or osteocalcin (Kim 2009). In the Spector 2008 trial, the bone-resorption marker collagen type I C-terminal telopeptide (CTX-I) showed a trend to rise with dose. Whether silicon's net effect in people is mainly bone-building (anabolic, like teriparatide) or mainly a brake on resorption (like bisphosphonates) is therefore not settled.

If silicon's main effect is bone-building, it would complement the more familiar antiresorptive bone agents. The idea that a person on a bisphosphonate could benefit from added silicon, to support the remaining osteoblast bone-formation pathway and the matrix-collagen quality, is a proposal, not a trial result.

Back to Table of Contents

Clinical Trials in Osteopenic and Osteoporotic Patients

The Spector 2008 trial (BMC Musculoskeletal Disorders) randomized 184 osteopenic women (spine T-score below -1.5), of whom 136 completed the study, to placebo or ch-OSA at 3, 6 or 12 mg silicon per day, on top of background calcium 1000 mg/day and vitamin D3 800 international units (IU, 20 micrograms)/day for 12 months. All four arms received the same calcium and vitamin D background.

The findings:

The trial was relatively short (12 months) and the bone-density changes were modest. However, the PINP biomarker increase was the predicted effect of an osteoblast-stimulating, collagen-supporting agent, and the femoral-neck signal, from a post-hoc subgroup only, was consistent with that mechanism. Longer, larger trials would be needed to fully establish silicon as an osteoporosis treatment; the trial's authors concluded that ch-OSA with calcium and vitamin D3 had a potential beneficial effect on bone collagen compared with calcium and vitamin D3 alone.

Eisinger and Clairet (1993) reported a retrospective study of 53 osteoporotic women followed for 14-22 months. The 8 women treated with silicon had a significant increase in femoral BMD, while etidronate (13 women) produced a slight, statistically non-significant increase in vertebral BMD.

Back to Table of Contents

Bone-Density Nutrients: Amounts in Studies and Reference Values

The amounts below are those used in the studies on this page, set in official reference values, or commonly described for bone health. They are reported as information.

Back to Table of Contents

Cautions

This content is provided for informational purposes only and does not constitute medical advice. The cautions above matter most for people with osteoporosis or chronic kidney disease, and for people taking osteoporosis medicines or warfarin.

Back to Table of Contents

What Is Not Yet Known About Silicon and Bone

A 2024 umbrella review of controlled animal studies (Pritchard and Nielsen) found that the silicon amounts tested ranged from 0.003 to 863 mg per kg of body weight per day. Silicon consistently helped bone and mineral metabolism only at around 139 mg per kg of body weight per day, a level the authors called likely unfeasible to attain in humans; for comparison, the review cites a current human intake recommendation of 25 mg per day in total. Missing data in the published animal studies made an effective threshold hard to pin down, and the authors asked for future studies to test it directly.

In people, the evidence described on this page is observational (the Framingham and Aberdeen studies), one small retrospective study (Eisinger 1993), and one 12-month trial that measured bone markers and BMD rather than fractures (Spector 2008).

Back to Table of Contents

Key Research Papers

  1. Jugdaohsingh R, Tucker KL, Qiao N et al. (2004). Dietary silicon intake is positively associated with bone mineral density in men and premenopausal women of the Framingham Offspring cohort. Journal of Bone and Mineral Research. — PubMed PMID: 14969400
  2. Macdonald HM, Hardcastle AC, Jugdaohsingh R, Fraser WD, Reid DM, Powell JJ (2012). Dietary silicon interacts with oestrogen to influence bone health: evidence from the Aberdeen Prospective Osteoporosis Screening Study. Bone. — PubMed PMID: 22173054
  3. Spector TD, Calomme MR, Anderson SH et al. (2008). Choline-stabilized orthosilicic acid supplementation as an adjunct to calcium/vitamin D3 stimulates markers of bone formation in osteopenic females. BMC Musculoskeletal Disorders. — PubMed PMID: 18547426
  4. Reffitt DM, Ogston N, Jugdaohsingh R et al. (2003). Orthosilicic acid stimulates collagen type 1 synthesis and osteoblastic differentiation in human osteoblast-like cells in vitro. Bone. — PubMed PMID: 12633784
  5. Carlisle EM (1970). Silicon: a possible factor in bone calcification. Science. — PubMed PMID: 5410261
  6. Carlisle EM (1981). Silicon: a requirement in bone formation independent of vitamin D1. Calcified Tissue International. — PubMed PMID: 6257332
  7. Sripanyakorn S, Jugdaohsingh R, Elliott H et al. (2004). The silicon content of beer and its bioavailability in healthy volunteers. British Journal of Nutrition. — PubMed PMID: 15005826
  8. Davenward S, Bentham P, Wright J et al. (2013). Silicon-rich mineral water as a non-invasive test of the 'aluminum hypothesis' in Alzheimer's disease. Journal of Alzheimer's Disease. — PubMed PMID: 22976072
  9. Eisinger J, Clairet D (1993). Effects of silicon, fluoride, etidronate and magnesium on bone mineral density: a retrospective study. Magnesium Research. — PubMed PMID: 8292498
  10. Price CT, Koval KJ, Langford JR (2013). Silicon: a review of its potential role in the prevention and treatment of postmenopausal osteoporosis. International Journal of Endocrinology. — PubMed PMID: 23762049
  11. Jugdaohsingh R (2007). Silicon and bone health. Journal of Nutrition, Health & Aging. — PubMed PMID: 17435952
  12. Powell JJ, McNaughton SA, Jugdaohsingh R et al. (2005). A provisional database for the silicon content of foods in the United Kingdom. British Journal of Nutrition. — PubMed PMID: 16277785
  13. Kim MH, Bae YJ, Choi MK, Chung YS (2009). Silicon supplementation improves the bone mineral density of calcium-deficient ovariectomized rats by reducing bone resorption. Biological Trace Element Research. — PubMed PMID: 19034393
  14. Sripanyakorn S, Jugdaohsingh R, Dissayabutr W et al. (2009). The comparative absorption of silicon from different foods and food supplements. British Journal of Nutrition. — PubMed PMID: 19356271
  15. Jugdaohsingh R, Anderson SH, Tucker KL et al. (2002). Dietary silicon intake and absorption. American Journal of Clinical Nutrition. — PubMed PMID: 11976163
  16. Pritchard A, Nielsen BD (2024). Silicon Supplementation for Bone Health: An Umbrella Review Attempting to Translate from Animals to Humans. Nutrients. — PubMed PMID: 38337624
  17. Marshall D, Johnell O, Wedel H (1996). Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. — PubMed PMID: 8634613
  18. Institute of Medicine (2011). Dietary Reference Intakes for Calcium and Vitamin D. Summary. Washington, DC: The National Academies Press. — doi:10.17226/13050

PubMed Topic Searches

  1. PubMed: silicon + bone mineral density
  2. PubMed: orthosilicic acid + osteoblast
  3. PubMed: silicon + postmenopausal osteoporosis
  4. PubMed: silicon + aluminum chelation
  5. PubMed: Framingham osteoporosis nutrition

Back to Table of Contents

Connections

Back to Table of Contents