Ingredient guides
No. 2
The Verateca TablePeriodic Table of Supplements

Magnesium

Share

Email article

Published

Magnesium is an essential mineral: the human body cannot make it and gets it from food and drinks. An adult body holds about 25 g, mostly in bone. Supplements add to it, in forms such as magnesium citrate, glycinate and oxide.

Magnesium in the body

Inside cells, magnesium is the second most common positively charged mineral, after potassium. It takes part in more than 300 enzyme systems. It is needed for normal muscle and nerve function, energy metabolism, protein synthesis, and normal bones and teeth.

Terms and sources
positively charged mineral
a cation. de Baaij et al., Physiological Reviews, 2015 (PMID 25540137).
enzyme systems
Enzymes are proteins that run chemical reactions in cells. The 300 figure is from NIH ODS. A 2015 review puts the number of enzyme reactions at over 600.
energy metabolism
How cells turn food into usable energy.

About 50–60% of the body's magnesium sits in bone. Most of the rest is in muscle and other soft tissue. Less than 1% is in the blood. A blood test sees only this small part.

Terms and sources
in the blood
In serum. The usual range is 0.75–0.95 mmol/L.

Where magnesium comes from

Good sources in food include seeds, nuts, beans, whole grains and green leafy vegetables. In plants, magnesium sits at the center of chlorophyll, the green pigment. One ounce (28 g) of roasted pumpkin seeds has 156 mg; half a cup of boiled spinach has 78 mg. Drinking water adds from about 1 mg to over 120 mg per liter, depending on the source. The body absorbs about 30–40% of the magnesium in food.

Terms and sources
chlorophyll
Plants use chlorophyll to turn carbon dioxide into glucose. Royal Society of Chemistry, Education: Magnesium.
156 mg
NIH Office of Dietary Supplements, table of food sources. Almonds: 80 mg per ounce. Black beans, cooked: 60 mg per half cup.

In supplements, magnesium comes as part of a compound, such as magnesium citrate or magnesium oxide. Magnesium compounds are made from sources that include seawater, natural brines and magnesite, a mined mineral.

Terms and sources
magnesite
The mineral is named after Magnesia, a district of Thessaly in Greece. US Geological Survey, Mineral Commodity Summaries 2025; WebElements. The USGS data cover all industrial uses, not only supplements.

Daily amounts and the upper limit

United States

US upper limit for adults: 350 mg of magnesium a day (National Academies; a reference value, not a legal limit). The US limit counts magnesium from supplements and medicines, not from food or water. It is set to avoid loose stools.

Terms and sources
upper limit
Tolerable Upper Intake Level (UL), National Academies, 1997.
loose stools
The National Academies (1997) based it on mild loose stools in some adults. These appeared at 360 mg a day from supplements.

US recommended amounts for adults are 400–420 mg a day for men and 310–320 mg for women. These amounts count food, drinks and supplements together. In the US, 48% of people get less magnesium from food and drinks than their estimated need.

Terms and sources
recommended amounts
RDA, National Academies. Men 19–30: 400 mg, 31+: 420 mg. Women 19–30: 310 mg, 31+: 320 mg.
estimated need
Estimated Average Requirement (EAR). National survey NHANES 2013–2016, all ages.

The Supplement Facts panel states the amount of magnesium itself, not the weight of the whole compound. In 1,481 US labels of magnesium-only products, the middle half list 144–400 mg per serving.

Terms and sources
magnesium itself
elemental magnesium. In our label count, some labels list the compound weight instead.
1,481 US labels
NIH Dietary Supplement Label Database, including products no longer sold. Labels whose only ingredient is named magnesium, in mg. A serving is not always a full day's amount.

In the same download, 29,894 labels list magnesium. Of them, 95% list something besides magnesium, and 71% list calcium.

Terms and sources
29,894 labels
Labels with an ingredient whose name includes magnesium, excluding magnesium stearate and silicates. We counted these.

Upper limits in other countries: see the table of limits by country on this page.

Forms of magnesium and how they differ

The partner compound decides how much magnesium a gram of the compound holds. 500 mg of magnesium oxide holds about 300 mg of magnesium. The same 500 mg of magnesium L-threonate holds about 40 mg. In small studies, the body absorbed more magnesium from citrate, chloride, lactate and aspartate than from oxide and sulfate. We do not rank forms.

Terms and sources
about 40 mg
8% by our calculation from the formula. The Lopresti 2025 trial states that 2 g of Magtein holds 145 mg, about 7%.
small studies
Summary by the NIH Office of Dietary Supplements.

NIH notes: "Magnesium may interact with certain medications, such as oral bisphosphonates, tetracyclines, and quinolone antibiotics."

Terms and sources
certain medications
NIH Office of Dietary Supplements, Magnesium fact sheet for health professionals, section on interactions with medications.
Magnesium glycinate, citrate, oxide and other forms
FormMagnesium in the compoundWhat is known
Oxide60%Less was absorbed than from citrate in small studies.
Hydroxide42%Not in the NIH summary of absorption by form.
Carbonate29%Not in the NIH summary of absorption by form.
Citrate16%More was absorbed than from oxide in small studies.
Malate16%Not in the NIH summary of absorption by form.
Glycinate (bisglycinate)14%Labels use both names for magnesium bound to the amino acid glycine. Not in the NIH summary of absorption by form.
Chloride12%More was absorbed than from oxide in small studies.
Taurate9%Not in the NIH summary of absorption by form.
L-threonate8%Brand: Magtein. In one trial, 2 g a day gave 145 mg of magnesium, about 7%.

Share of magnesium by mass of the anhydrous compound; chloride as the hexahydrate. We calculated these from the chemical formula.

Published studies of magnesium

6 studies in healthy adults: 3 found a difference, 2 found none, 1 had mixed results.

Our one-line summary comes first. The original is on PubMed.

Show the 6 studies
  • Healthy adults took magnesium citrate, amino-acid form or oxide for 60 days; blood magnesium was highest with citrate.

    Trial · 46 people · 60 days · funding not stated

    Walker et al., Magnesium Research, 2003

    Terms and sources
    magnesium citrate
    300 mg of magnesium itself a day in each group. Another group took a dummy pill (placebo).
    amino-acid form
    magnesium bound to amino acids (amino-acid chelate)
    oxide
    magnesium oxide. Blood magnesium with oxide did not differ from the dummy pill.
    blood magnesium
    serum magnesium. Magnesium in 24-hour urine was also higher with citrate and the amino-acid form than with oxide.
    funding not stated
    The abstract does not name who paid for the study or supplied the product. We could not open the full text.

    Study · Abstract only · about 1 min to read

    Our estimate at about 200 words a minute, from the PubMed abstract.

    Study summary

    Researchers compared three forms of magnesium with a dummy pill in 46 healthy UK adults. Three groups took 300 mg of magnesium a day: as citrate, bound to amino acids, or as oxide. A fourth took the dummy pill. After 60 days, more magnesium appeared in 24-hour urine with citrate and the amino-acid form than with oxide. Blood magnesium was highest with citrate. These differences were statistically significant. Oxide did not differ from the dummy pill. Health and symptoms were not measured. Funding: not named in the abstract; we could not open the full text.

  • Healthy adults drank one magnesium chloride dose; free blood magnesium rose more than with a dummy drink, total did not.

    Trial · 17 people · 1 dose, 24 hours · maker-funded

    Zhan et al., Nutrients, 2020

    Terms and sources
    magnesium chloride dose
    ReMag, 300 mg of magnesium itself, in water with lemon juice, with a low-magnesium breakfast
    free blood magnesium
    ionized magnesium in whole blood, over 24 hours
    a dummy drink
    placebo: water with lemon juice only. Each person had both drinks, one after the other (crossover), at least 7 days apart.
    total
    Total magnesium in blood serum and magnesium in urine did not differ from the dummy drink.
    maker-funded
    New Capstone, maker of ReMag, paid for the study through an educational grant.

    Study · Free full text · about 15 min to read

    Our estimate at about 200 words a minute, from the PMC full text.

    Study summary

    A pilot study tested ways to measure magnesium after one dose. 17 healthy adults drank 300 mg of magnesium as magnesium chloride (ReMag) in water with lemon juice. On another day, at least 7 days apart, they drank the same drink without magnesium. Over 24 hours, free magnesium in whole blood rose more after the magnesium drink. The difference was statistically significant. Total blood magnesium and urine magnesium did not differ. Complete data for free magnesium came from 14 people. Funding: New Capstone, maker of ReMag, paid through an educational grant.

  • Adults who slept poorly took magnesium glycinate for 4 weeks; sleep-problem scores fell slightly more than with a dummy pill.

    Trial · 155 people · 4 weeks

    Schuster et al., Nature and Science of Sleep, 2025

    Terms and sources
    Adults who slept poorly
    healthy adults aged 18–65 who rated their own sleep as poor; no diagnosis
    magnesium glycinate
    magnesium bisglycinate, 250 mg of magnesium itself a day
    sleep-problem scores
    A 0–28 sleep questionnaire: −3.9 vs −2.3 points at week 4. Effect size 0.2, which the authors call small; statistically significant. Self-rated stress scores did not differ.
    a dummy pill
    placebo

    Study · Free full text · about 20 min to read

    Our estimate at about 200 words a minute, from the PMC full text.

    Study summary

    Researchers compared magnesium bisglycinate with a dummy pill. In total, 155 adults aged 18–65 rated their own sleep as poor. They took 250 mg of magnesium or the dummy pill daily for 4 weeks. On a 0–28 sleep questionnaire, scores fell by 3.9 points with magnesium and by 2.3 with the dummy pill. The difference was statistically significant; the authors call the effect small. Stress, mood and tiredness scores did not differ. Sleep was self-rated only. Funding: Leibniz University Hannover, from its own budget; Biogena made the capsules.

  • Poor sleepers over 50 took magnesium citrate for 7 weeks; sleep scores did not differ from a dummy pill.

    Trial · 100 people · 7 weeks · funding not stated

    Nielsen et al., Magnesium Research, 2010

    Terms and sources
    Poor sleepers over 50
    aged 51–85, sleep questionnaire score above 5; no diagnosis required
    magnesium citrate
    320 mg of magnesium itself a day
    sleep scores
    Pittsburgh Sleep Quality Index; lower scores mean fewer sleep complaints. Scores fell in both groups, and the groups did not differ.
    a dummy pill
    placebo (sodium citrate)
    funding not stated
    We could not find who paid for this study or supplied the product.

    Study · Abstract only · about 2 min to read

    Our estimate at about 200 words a minute, from the PubMed abstract.

    Study summary

    Researchers compared magnesium citrate with a dummy pill made of sodium citrate. 100 adults aged 51–85 with poor sleep scores took part. They took 320 mg of magnesium or the dummy pill daily for 7 weeks; 96 completed. Sleep scores fell in both groups, and the groups did not differ. Overall, blood magnesium did not differ between groups. In 37 people with low blood magnesium at the start, it rose with magnesium only. Sleep was self-rated only. Funding: not named in the abstract; PubMed tags the work as not funded by the US government.

  • Adults unhappy with their sleep took magnesium threonate for 6 weeks; ring-tracked sleep did not differ from a dummy pill.

    Trial · 100 people · 6 weeks · maker-funded

    Lopresti et al., Frontiers in Nutrition, 2025

    Terms and sources
    Adults unhappy with their sleep
    healthy adults aged 18–45; no sleep diagnosis
    magnesium threonate
    Magtein, 2 g a day = 145 mg of magnesium itself
    ring-tracked sleep
    Sleep from an Oura ring did not differ. On questionnaires, daytime sleep-related impairment scores fell more; sleep disturbance did not differ. The trial's main measure, a thinking-test score, rose more with magnesium. The paper reports one-sided tests.
    a dummy pill
    placebo
    maker-funded
    Threotech, which supplied Magtein, paid for the study and was involved in its design.

    Study · Free full text · about 50 min to read

    Our estimate at about 200 words a minute, from the PMC full text.

    Study summary

    Researchers compared magnesium L-threonate (Magtein) with a dummy pill. 100 healthy adults aged 18–45 were unhappy with their sleep. They took 2 g a day or the dummy pill for 6 weeks. 2 g of Magtein holds 145 mg of magnesium. Sleep measured by an Oura ring did not differ between groups. Daytime sleep-related impairment scores fell more with magnesium; sleep disturbance scores did not differ. The main measure, a thinking-test score, rose more with magnesium. The authors used a less strict, one-sided test. Funding: Threotech, which supplied Magtein, paid for the study and was involved in its design.

  • Older women in exercise classes took magnesium oxide for 12 weeks; physical test scores rose more than with no pill.

    Trial · 139 people · 12 weeks · funding not stated

    Veronese et al., American Journal of Clinical Nutrition, 2014

    Terms and sources
    Older women
    healthy women, average age 71.5
    magnesium oxide
    300 mg of magnesium itself a day
    physical test scores
    Short Physical Performance Battery: +0.41 points more than with no pill. Chair-stand time fell 1.31 s more and 4-m walking speed rose 0.14 m/s more. Grip and leg strength did not differ.
    no pill
    The other group took nothing; there was no dummy pill. 124 of 139 women were analyzed.
    funding not stated
    The abstract does not name who paid for the study or supplied the product. We could not open the full text.

    Study · Abstract only · about 2 min to read

    Our estimate at about 200 words a minute, from the PubMed abstract.

    Study summary

    Researchers compared magnesium oxide with no pill. 139 healthy women, average age 71.5, attended a weekly exercise program for 12 weeks. One group took 300 mg of magnesium a day; the other took nothing. 124 women were analyzed. The physical performance score rose 0.41 points more with magnesium, a statistically significant difference. Chair-stand time fell more and 4-m walking speed rose more with magnesium. Grip and leg strength did not differ. The researchers who tested the women did not know their group. Funding: not named in the abstract.

Sources

Information, not a prescription. Do not start, stop or combine supplements or medicines on your own.