Literature · 16 studies, 1990–2026
What has actually been measured about how magnesium salts are absorbed
Supplement marketing ranks magnesium forms confidently. The published literature is thinner and less tidy than the marketing: 14 of these 16 studies were run in people, 2 in animals, sample sizes run from eleven to several hundred, and several use different read-outs — urinary excretion, intracellular magnesium, stable isotopes — which are not interchangeable. This page reports what each study did and found, and links the original. It does not rank the salts and does not tell anyone what to take.
Studies listed16
Run in people14
Salts covered12
How to read this
Why the studies disagree
- The read-out differs. Urinary excretion, serum magnesium, intracellular magnesium and stable-isotope tracing measure different things. A salt can stand out on one read-out and look unremarkable on another.
- The dose differs, and absorption depends on dose. A 2020 profiling study treats absorption as dose-dependent, which is enough on its own to explain why single-dose head-to-head results conflict.
- The population differs. One study here is in patients after ileal resection — people with impaired absorption by definition. Another is in magnesium-depleted rats, and depletion itself raises absorption.
- “Amino acid chelate” is not one substance. It is a class, so a trial arm using it is not reproducible from the name alone. That is also why this site cannot compute a compound mass for labels that say only “chelate”.
- Solubility is not absorption. A 1990 paper found magnesium oxide virtually insoluble in a beaker across the whole range of stomach acidity. That is a measurement about a beaker; the same paper then tested people separately.
Published measurements · peer-reviewed sources
All 16 studies
Ordered newest first. Each entry states the design, what the study measured, and the limits that bound how far the result travels.
These entries report what each study measured and on whom. They are not advice, not a ranking, and not a claim about any product on this site. Every reference links to its PubMed record so the original can be read in full.
Randomized, open-label study of the short-term pharmacokinetics of oral magnesium oxide in healthy volunteers.
- Design
- Randomized, open-label pharmacokinetic study in 24 healthy male volunteers aged 18-45 years, equally assigned to a single oral dose of either 750 mg or 2000 mg magnesium oxide.
- What it measured
- Serum magnesium concentrations at baseline and multiple time points up to 12 hours post-dose (colorimetric assay); pharmacokinetic parameters Cmax, Tmax, and AUC0-12 via non-compartmental analysis; safety via vital signs, laboratory parameters, ECG, and adverse events.
- Limits
- Male volunteers only, no female data; only two dose levels tested (750 mg, 2000 mg) so dose-proportionality across a wider range is not established; terminal elimination parameters (t1/2, AUC0-infinity, CL/F, MRT) could not be estimated within the 12-hour sampling window; single-dose (not chronic) administration; sample size n=24 (12 per dose group).
Comparing the Bioavailability of Two Seawater-Derived Magnesium Preparations.
- Design
- Double-blind, randomized, three-arm crossover study in 20 healthy young men and women: placebo (maltodextrin), seawater-derived magnesium citrate (Aquamin-Mg, minimum 10% elemental magnesium), and seawater-derived magnesium hydroxide (Aquamin-MgTg, minimum 33% elemental magnesium). Total magnesium doses were standardized to the RDA for elemental magnesium.
- What it measured
- Incremental 18-hour urinary magnesium excretion as the relative-bioavailability marker; serum magnesium concentration at 1 and 2 hours post-ingestion (colorimetric assay, run in triplicate).
- Limits
- Small sample (n=20); healthy young adults only, not a magnesium-deficient or older population; only two seawater-derived preparations plus placebo were tested, no head-to-head comparison against standard magnesium citrate/oxide/glycinate salts; short observation window (18 h urine, 2 h serum).
Comparative Clinical Study on Magnesium Absorption and Side Effects After Oral Intake of Microencapsulated Magnesium (MAGSHAPE™ Microcapsules) Versus Other Magnesium Sources.
- Design
- Double-blind, randomized, cross-over clinical study in 40 healthy women and men. Participants followed a low-magnesium diet for 7 days and fasted 8 hours, then received a single oral dose of each of four products in cross-over fashion: microencapsulated magnesium (Mg-MS / MAGSHAPE), magnesium oxide (MgO), magnesium citrate (Mg-C), and magnesium bisglycinate (Mg-BG). Blood was drawn by digital puncture before dosing (0 h) and at 1, 4, and 6 hours after each product.
- What it measured
- Plasma magnesium concentration at 0, 1, 4, and 6 hours after oral intake of each of the four products; gastrointestinal side effects (intestinal motility, sensations of gastric heaviness) associated with each product.
- Limits
- Single-dose exposure per product, not chronic supplementation; sample size n=40; a specific proprietary microencapsulation technology was tested, which may not generalize to other encapsulation methods; plasma magnesium was the only absorption marker used (no urinary or tissue measurement).
The bioaccessibility and tolerability of marine-derived sources of magnesium and calcium.
- Design
- Two-part study. (1) In vitro digestion (INFOGEST method) comparing solubility of Aquamin-Mg (seawater-derived, 12% magnesium), a commercially available magnesium bisglycinate, and a specific branded magnesium bisglycinate (PrizMAG) during simulated gastric and intestinal digestion phases, with and without co-administered food. (2) A randomized, double-blind, placebo-controlled 12-week study in a small cohort of healthy older adults given a combined Aquamin-Mg/Aquamin-F supplement.
- What it measured
- In vitro mineral solubility during simulated gastric and intestinal digestion phases; adverse-event reporting and tolerability over the 12-week human study arm.
- Limits
- The solubility comparison was performed in vitro (simulated digestion model), not as in vivo human absorption; the human arm assessed tolerability/safety only, not comparative bioavailability; the human cohort was described only as small and limited to older adults, with no numeric sample size given in the abstract.
Effects of short-term magnesium supplementation on ionized, total magnesium and other relevant electrolytes levels.
- Design
- Short-term, 10-day supplementation study in 61 healthy young female adults, comparing three magnesium supplement forms: magnesium oxide (powder/granulate), magnesium citrate, and magnesium carbonate (effervescent tablets).
- What it measured
- Ionized magnesium concentration and its percentage of total magnesium; serum total magnesium concentration; other serum electrolyte levels — each assessed at baseline and after the 10-day intervention.
- Limits
- Healthy young female adults only, no male or older/deficient populations studied; short 10-day intervention with no longer-term follow-up; three specific product formulations were tested, so results may not generalize to other products of the same salts; the abstract does not report the per-arm breakdown of the 61 participants.
Dose-Dependent Absorption Profile of Different Magnesium Compounds
- Design
- Absorption of several magnesium compounds profiled across doses rather than at a single dose.
- What it measured
- Absorption is treated as dose-dependent, so a comparison made at one dose need not hold at another — which is why single-dose head-to-head results disagree.
- Limits
- Compound coverage and sample size vary by arm; read the paper before carrying any single figure across.
Timeline (Bioavailability) of Magnesium Compounds in Hours: Which Magnesium Compound Works Best?
- Design
- Single-dose administration of 400 mg/70 kg magnesium as one of five compounds (magnesium sulfate, magnesium oxide, magnesium acetyl taurate, magnesium citrate, magnesium malate) to Sprague Dawley rats, with a control group, followed by time-dependent tissue and serum sampling.
- What it measured
- Time-dependent magnesium absorption profile (area under the curve) in serum; magnesium tissue penetration and concentration, including brain tissue; behavioral indicators (e.g., anxiety-related behavior) recorded in the animals.
- Limits
- Animal model (Sprague Dawley rats), not humans — extrapolation of pharmacokinetic and behavioral findings to people is not established; single dose only, no chronic-dosing data; per-compound group sample size is not stated in the abstract; behavioral endpoints were measured only in rodents. The tested acetyl taurate is distinct from the unmodified taurate in the chemistry table.
Scottsdale Magnesium Study: Absorption, Cellular Uptake, and Clinical Effectiveness of a Timed-Release Magnesium Supplement in a Standard Adult Clinical Population.
- Design
- Placebo-controlled study at two clinics in 91 adults; 53 received MagSRT (500 mg dimagnesium malate plus vitamins B6, B12, and folate, dosed as 2 tablets containing 250 mg elemental magnesium) and the remainder received placebo. A subset of 24 MagSRT participants continued the trial to 90 days.
- What it measured
- Serum magnesium at 4 and 8 hours post-dose; red blood cell (RBC) magnesium at baseline, 30 days, and 90 days (subset); magnesium status questionnaire score at baseline, 30 days, and 90 days.
- Limits
- Comparator was placebo only, not a head-to-head comparison against other magnesium salts; the tested product combined dimagnesium malate with B-vitamins and folate, so effects cannot be isolated to the malate salt alone; 90-day follow-up was limited to a subset (n=24) of the original 53 treated participants; outcome included a self-reported symptom questionnaire alongside biochemical markers.
Comparison of magnesium status using X-ray dispersion analysis following magnesium oxide and magnesium citrate treatment of healthy subjects
- Design
- Oral magnesium citrate versus magnesium oxide in healthy subjects; intracellular magnesium measured by X-ray dispersion analysis, alongside platelet function.
- What it measured
- The comparison used intracellular magnesium rather than serum magnesium, which changes what the read-out means.
- Limits
- Healthy subjects, small sample; intracellular magnesium by X-ray dispersion is not a routine clinical measure.
Study of magnesium bioavailability from ten organic and inorganic Mg salts in Mg-depleted rats using a stable isotope approach
- Design
- Eighty male Wistar rats fed a magnesium-depleted diet for three weeks, then randomised into ten groups, each receiving a different magnesium salt labelled with a stable isotope.
- What it measured
- Stable isotopes allow absorption and retention to be traced directly rather than inferred from urine, across ten salts at once — the widest head-to-head comparison in the literature.
- Limits
- Rats, not people, and magnesium-depleted rats at that: depletion itself raises absorption. The breadth of the comparison is its value; the species is its limit.
Mg citrate found more bioavailable than other Mg preparations in a randomised, double-blind study
- Design
- Randomised, double-blind, placebo-controlled parallel study in 46 healthy individuals, 300 mg of elemental magnesium per day, comparing an amino-acid chelate, citrate and oxide.
- What it measured
- Relative bioavailability of the three preparations was compared at an equal elemental dose.
- Limits
- 46 participants, healthy, 300 mg per day; “amino-acid chelate” is a class rather than one defined compound, so the chelate arm is not a single substance.
Bioavailability of US commercial magnesium preparations
- Design
- Four commercially available US preparations given to healthy volunteers; bioavailability measured as the increment in urinary magnesium excretion.
- What it measured
- The study set out to test the claim that organic magnesium salts are more easily absorbed than inorganic ones, using urinary excretion as the read-out.
- Limits
- Healthy volunteers, small sample, urinary excretion as a proxy for absorption rather than a direct measurement of retention.
Bioavailability of magnesium diglycinate vs magnesium oxide in patients with ileal resection
- Design
- Comparison of magnesium diglycinate (a chelate) with magnesium oxide in patients who had undergone ileal resection.
- What it measured
- The work followed in vitro and in situ findings suggesting diglycinate is absorbed by a different route than inorganic salts, and tested that in patients with impaired absorption.
- Limits
- Patients after ileal resection are a specific clinical population with poor absorption to begin with; results do not transfer to people with an intact gut.
Blood and urinary magnesium kinetics after oral magnesium supplements.
- Design
- Randomized, crossover study in 12 healthy normomagnesemic subjects: a baseline period followed by three separate one-week supplementation trials at 16 mmol/dose — magnesium chloride solution, slow-release magnesium chloride tablets, and magnesium gluconate tablets — with a standardized diet during each trial.
- What it measured
- Urinary magnesium excretion in 0-4, 4-8, 8-12, and 12-24 hour collections; intraleukocyte, total serum, and ultrafiltrable serum magnesium at 0, 1, 2, 3, 4, 8, 12, and 24 hours; area-under-curve for each compartment relative to baseline for each supplement form.
- Limits
- Very small sample (n=12); only two chemical forms tested (chloride in two formulations, plus gluconate); short trial duration per arm (24 hours, with weekly crossover intervals); normomagnesemic (not magnesium-deficient) subjects only.
Magnesium bioavailability from magnesium citrate and magnesium oxide
- Design
- In vitro solubility of 25 mmol of each salt across hydrochloric acid concentrations mimicking achlorhydric to peak acid secretion, plus in vivo gastrointestinal absorbability.
- What it measured
- Magnesium oxide was virtually insoluble across the acid range tested, while citrate dissolved; absorbability was then compared in vivo.
- Limits
- Solubility in a beaker is not absorption in a person; the in vivo arm is small. Published 1990.
Effect of magnesium citrate and magnesium oxide on the crystallization of calcium salts in urine
- Design
- Seven normal subjects and four patients with recurrent calcium oxalate stones; 486 mg of magnesium per day for two weeks, taken on an empty stomach and again with food.
- What it measured
- Taken on an empty stomach, urinary magnesium rose by only 77–79 mg per day out of 486 mg given — and the increase changed when the dose was taken with food.
- Limits
- Eleven subjects; the endpoint is urinary chemistry in stone formers, not absorption in general. Published 1990.
Separate question · fixed chemistry
How much magnesium each salt carries, which is not the same question
Absorption is contested. Mass is not. How much of a salt's weight is magnesium follows from its molecular formula and does not depend on any study — and it varies more than tenfold across the salts on US labels.
| Salt | Magnesium by mass | Compound needed per 100 mg elemental |
|---|---|---|
| oxide | 60.3% | 166 mg |
| hydroxide | 41.7% | 240 mg |
| carbonate | 28.8% | 347 mg |
| chloride | 25.5% | 392 mg |
| sulfate | 20.2% | 495 mg |
| citrate | 16.2% | 619 mg |
| malate | 15.5% | 644 mg |
| glycinate | 14.1% | 709 mg |
| glycerophosphate | 12.5% | 800 mg |
| lactate | 12.0% | 833 mg |
| taurate | 8.8% | 1,130 mg |
| aspartate | 8.4% | 1,188 mg |
| L-threonate | 8.3% | 1,212 mg |
| orotate | 7.3% | 1,376 mg |
| gluconate | 5.9% | 1,706 mg |
See the label mass-budget audit for what this implies about what will physically fit in a capsule.
Food vs supplement: food grams and compound milligrams for 100 mg elemental magnesium →