Lithium: The Mineral, Its Natural Sources and How It Acts

The medicine John Cade reported in 1949 was not a molecule built in a laboratory. It was a simple salt of lithium, the third element of the periodic table — a soft, silvery metal that never occurs free in nature but is dissolved, in tiny amounts, in seawater, in rock, in soil, in many spring and river waters, and in the plants that grow from them. Lithium was a mineral long before it was a psychiatric medicine, and for most of the nineteenth century it was better known as a supposed remedy for gout and a selling point for “lithia” waters than as anything to do with the mind.

This page follows lithium as a natural substance: how it was found in a Swedish mineral in 1817 and named after the Greek word for stone, where it comes from today, how it reaches food and drinking water, the gout and healing-spring era that Cade himself wrote about, his own idea that lithium might be a trace element, the later argument that it is an essential nutrient, the drinking-water studies on suicide and dementia and their limits, the narrow gap between an effective and a toxic amount, and what researchers think lithium actually does inside cells. It reports history and research findings only; it gives no dosing and no supplement guidance.

Table of Contents

  1. Element Number Three
  2. Petalite, Spodumene and the Sea
  3. Lithium in Food and Water
  4. Gout, Lithia Tablets and Healing Springs
  5. Cade’s Trace-Element Idea
  6. Is Lithium an Essential Nutrient?
  7. Lithium in Tap Water and Suicide Rates
  8. Lithium in Tap Water and Dementia
  9. A Narrow Margin
  10. How Lithium Acts: Inositol and GSK-3
  11. Key Research Papers
  12. Connections

1. Element Number Three

Lithium sits third in the periodic table, after hydrogen and helium. Each lithium atom has three protons, and it is the lightest of all metals and the least dense of all solid elements — a block of pure lithium would float on water, though it reacts with the water as it does. Like sodium and potassium, its neighbours in the same column of the table, it gives up one electron very readily. That is why it is never found as a bare metal in nature: it is always bound up in minerals or dissolved as the positively charged lithium ion, Li+.

The element was discovered in 1817 by the Swedish chemist Johan August Arfwedson, who was analysing a mineral called petalite from the island of Utö, in the Stockholm archipelago. His analysis left a fraction he could not account for with the known alkali metals, and he concluded it held a new one. Because it had been found in a mineral, rather than in plant ash as potassium and sodium had been, the new element was named from the Greek lithos, “stone”. The metal itself was first isolated in 1821 by the English chemist William Thomas Brande, who passed an electric current through a lithium compound.

It is worth keeping in mind what “lithium” means in medicine. Patients are never given the metal. They are given a salt — lithium carbonate, as described on this site’s Lithium Carbonate page, or historically lithium citrate — which dissolves in the body to release the lithium ion. Everything that follows, from spring water to brain chemistry, is about that small charged ion.

Back to Table of Contents

2. Petalite, Spodumene and the Sea

Petalite, the mineral in which lithium was first found, is one of several lithium-bearing minerals found in granite pegmatites — coarse-grained rocks that crystallise late from cooling magma and gather unusual elements. The main ores mined for lithium are spodumene and petalite, together with lepidolite, a lilac-coloured lithium mica. All three are silicates: lithium locked into a framework of silicon, oxygen and aluminium.

Lithium is also present in the oceans. Seawater holds roughly 0.14 to 0.25 parts per million — a tiny concentration, but spread through so much water that the sea contains an enormous total amount. Salty lakes and underground brines, where water has evaporated and concentrated what it carried, hold far richer amounts.

Rain and groundwater slowly dissolve lithium out of rocks and soils, which is how it enters rivers, springs, wells and eventually the food chain. The amount varies enormously from one place to another, depending on the local geology. That geographical variation is the starting point for nearly every story on this page: the reputation of certain healing springs, the uneven lithium content of crops, and the modern drinking-water studies all depend on the simple fact that some regions’ water carries much more lithium than others.

Back to Table of Contents

3. Lithium in Food and Water

Everyone takes in some lithium every day, without noticing, from food and water. In a 2002 review in the Journal of the American College of Nutrition, G. N. Schrauzer described grains and vegetables as the primary food sources of lithium, with drinking water able to make a significant contribution in some areas. He also noted that traces of lithium in human organs and tissues had already been recorded in the late nineteenth century.

Because plants take up lithium from the soil and water around them, how much ends up in a given food depends on where it was grown. Schrauzer’s review stressed that daily intakes vary widely from region to region for exactly this reason. Animal foods carry some lithium as well, reflecting what the animals themselves ate and drank.

The amounts in food and water are tiny compared with the amounts used in psychiatric treatment. That difference in scale matters when reading the drinking-water research below: the studies look at whether very small, lifelong exposures might be linked with patterns of health across whole populations, not at medicinal amounts. More on hydration and the minerals carried by water appears on this site’s Hydration page, and the wider family of trace elements is described on the Trace Minerals page.

Back to Table of Contents

4. Gout, Lithia Tablets and Healing Springs

Lithium’s first medical career had nothing to do with mood. In the mid-nineteenth century chemists found that lithium urate — the lithium salt of uric acid — dissolves in water far more readily than the other urates. Gout is caused by needle-like crystals of uric acid salts deposited in the joints (see this site’s History of Gout page), so the idea followed naturally that lithium might dissolve those deposits. The London physician Alfred Baring Garrod introduced lithium salts for gout, and John Cade cited Garrod’s 1859 work in the historical review that opened his own 1949 paper.

The idea spread well beyond the clinic. As Cade recounted in that review, “lithia tablets” were sold and taken for a wide range of ailments, and medical journals of the early twentieth century — he cited reports in The Practitioner in 1907 and 1909 — began recording toxic effects among people who took them. Mineral waters were advertised for their lithium content, and spas whose springs carried lithium were credited with healing powers. The pharmacy historians Strobusch and Jefferson later surveyed this era under the title “the checkered history of lithium in medicine”.

The gout theory itself did not hold up, and lithium faded from rheumatology. But one thread from this era caught Cade’s attention. He noted a textbook remark that the waters of certain wells had been considered helpful in mental illness, and wrote — as quoted by the historian Johan Schioldann — that it was “very likely that their supposed efficacy was a real efficacy and directly proportional to the lithium content of the waters”. That speculation about well water, written in 1949, anticipated by more than half a century the modern studies of lithium in tap water described below.

The link back to gout also explains how Cade came to lithium at all. He had been testing uric acid in his animal experiments, and because uric acid itself dissolves so poorly he used its most soluble salt, lithium urate — the same property that had made lithium attractive to the gout doctors. That chain of reasoning is told on the Lithium Salts and Mania: Cade’s 1949 Discovery page.

Back to Table of Contents

5. Cade’s Trace-Element Idea

Cade did not think of lithium only as a drug. Having seen manic patients calm when given lithium salts and relapse when the salts were stopped, he speculated in his 1949 paper that mania might involve a deficiency of lithium ions in the body. In the words quoted by Schioldann, he wrote that lithium “may well be an essential trace element”, pointing out that it was “widely distributed” and had been “detected in sea-water and in many spring and river waters, in the ash of many plants, and in the animal ash”.

The idea fitted his background. Cade had spent three and a half years as a prisoner of war in Changi, where, his biographers write, he had watched deficiency diseases such as beriberi and pellagra — caused by a lack of vitamin B1 and niacin — affect the men around him. The thought that a missing trace substance could disturb the mind was not a strange one for a doctor with that experience. His life and his wartime years are covered on the John Cade: Life and Career page.

Cade’s deficiency theory was not confirmed. People with mania do not, as far as later research has shown, simply lack lithium, and the doses used in treatment are far above anything found in food or water. Yet his instinct that lithium belongs to the world of natural trace elements, rather than only to the pharmacy, has kept returning in the nutrition and drinking-water research of the following decades.

Back to Table of Contents

6. Is Lithium an Essential Nutrient?

An essential nutrient is one the body needs but cannot make, so that a lack of it causes harm. Iron, zinc and iodine are accepted examples. Whether lithium belongs on that list is unsettled, and no official nutrition body has set a requirement for it.

The strongest argument for essentiality came from Schrauzer’s 2002 review. He summarised animal experiments in which rats and goats kept on low-lithium diets showed higher mortality and abnormalities of reproduction and behaviour, and population studies in which lower lithium in drinking water was associated with higher rates of suicide, homicide and drug-related arrests. On that basis he argued that the evidence was sufficient to regard lithium as essential, and he proposed a provisional daily intake figure for adults. That figure is one author’s proposal, published in a review article; it is not a recommendation of any official body, and it is reported here only as part of the scientific record.

Schrauzer also acknowledged the main gap: no specific human lithium-deficiency disease has been characterised. Unlike scurvy for vitamin C or goitre for iodine, there is no recognised illness that appears when lithium is missing and disappears when it is restored. Until that link is shown, lithium’s status as a nutrient remains an open question rather than an established fact. The site’s Inositol page describes a comparable case — inositol is a “vitamin-like” compound that, as section 10 explains, is also central to one theory of how lithium works.

Back to Table of Contents

7. Lithium in Tap Water and Suicide Rates

If lithium in medicinal doses reduces suicide among people with mood disorders — a finding described on the From Cade to Schou page — researchers wondered whether the much smaller amounts in drinking water might show a faint echo across whole populations. Because the lithium content of tap water varies with local geology, different towns form a kind of natural experiment.

The Oita study

In 2009 Ohgami and colleagues measured lithium in the tap water of 18 municipalities in Oita prefecture, Japan, and compared it with each municipality’s standardised mortality ratio for suicide over 2002–2006. They reported a negative association: municipalities with more lithium in their water tended to have lower suicide rates. The study was short, a letter-length report, and covered one prefecture.

The 2020 meta-analysis

By 2020 enough studies had accumulated for a systematic review. Memon and colleagues, writing in the British Journal of Psychiatry, pooled 15 ecological studies from several countries and found an inverse association between naturally occurring lithium in drinking water and total suicide rates. When results were split by sex, the association was statistically significant for women; for men it fell just short of significance.

What these studies cannot show

These are ecological studies: they compare average water lithium with average suicide rates in whole areas, not the lithium intake and outcome of individual people. That makes them open to the “ecological fallacy” — a pattern between areas need not hold for the people living in them — and to confounding by anything else that differs between regions, such as income, rural life, health services or other minerals in the water. Memon and colleagues stated these limits themselves and called for randomised community trials to test whether the association reflects cause and effect. The findings are associations, not proof.

Back to Table of Contents

8. Lithium in Tap Water and Dementia

A second line of drinking-water research asks about dementia. Laboratory work on lithium’s effects on brain cells, including the enzyme GSK-3 discussed in section 10, led researchers to ask whether long-term exposure to natural lithium in water might be linked with lower rates of dementia.

The largest study of this question used Denmark’s national registers. Kessing and colleagues, publishing in JAMA Psychiatry in 2017, compared 73,731 people diagnosed with dementia with 733,653 people without it, and estimated each person’s long-term exposure to lithium from the water supplies where they had lived. People whose water carried more than 15.0 micrograms of lithium per litre had a lower incidence of dementia than those exposed to 2.0 to 5.0 micrograms per litre, the study’s reference group (incidence rate ratio 0.83).

The result was not a straight line, however. People in a middle band of exposure, 5.1 to 10.0 micrograms per litre, had a higher incidence of dementia than that reference group (incidence rate ratio 1.22). The authors described the relationship as non-linear and noted that confounding by other factors could not be excluded. In other words, the study fits the idea that higher natural lithium is associated with less dementia, but it does not show a simple “more is better” pattern, and as an observational study it cannot establish that lithium itself is the cause.

Back to Table of Contents

9. A Narrow Margin

The single most important fact about lithium as a medicine is how small the gap is between the amount that has an effect and the amount that poisons. Pharmacologists describe this as a narrow therapeutic index. The history of lithium in the 1940s and 1950s shows that gap more starkly than any textbook.

The salt-substitute poisonings of 1949

In March 1949, months before Cade’s paper appeared, the Journal of the American Medical Association published reports of lithium poisoning in heart patients. People on low-sodium diets had been using lithium chloride, sold as a table-salt substitute, to season their food; some were seriously poisoned and some died. Corcoran, Taylor and Page described “lithium poisoning from the use of salt substitutes”, and Hanlon and Romaine reported their observations on the toxicity of lithium chloride used in place of sodium chloride. Lithium and sodium are handled by the kidney in closely linked ways, and people who were eating little sodium — often with heart or kidney disease — were especially exposed.

Cade’s first patient

Cade met the same danger in his own work. According to the reviews of his modern biographies, his first patient, the man he called W.B., died of lithium poisoning in 1950, and Cade then set lithium aside. The episode is described on the 1949 discovery page.

Learning to measure

What made lithium usable was measurement. In Melbourne in the early 1950s, E. M. Trautner and his colleagues studied how ingested lithium is excreted and retained, and how it affects the body’s balance of other ions, publishing a detailed account in 1955. Mogens Schou’s 1957 review of the biology and pharmacology of the lithium ion gathered what was then known. Blood-level measurement became the basis of lithium treatment, and Schou, writing in 1997 after forty years of experience, described regular monitoring of serum lithium and of kidney function as part of its long-term use. Effects of long-term treatment on the kidney and thyroid are summarised on the legacy page and on this site’s Nephrogenic Diabetes Insipidus and Thyroid Panel pages.

The narrow margin is also why the natural-source story and the medical story have to be kept apart. The traces in food and tap water are far below medicinal amounts, while the amounts used in psychiatry sit close to the toxic range and are managed with blood tests. Lithium sold outside medicine, including as a supplement, is outside the scope of this page.

Back to Table of Contents

10. How Lithium Acts: Inositol and GSK-3

Seventy-five years after Cade’s paper, nobody can yet say exactly why lithium steadies mood. Lithium is a small ion that touches many processes in the cell, and researchers have proposed several mechanisms. Two have dominated the field.

The inositol-depletion hypothesis

Many hormones and nerve signals act on cells by triggering a chain of messenger molecules built from inositol, a sugar-like compound (described on this site’s Inositol page). After each round of signalling, the cell recycles inositol using an enzyme called inositol monophosphatase. Lithium blocks that enzyme. In 1989 M. J. Berridge, C. P. Downes and M. R. Hanley, writing in the journal Cell, put forward what they called “a unifying hypothesis”: by blocking inositol recycling, lithium would gradually starve the most active signalling pathways of inositol, damping down overactive cells while leaving quieter ones largely alone. The idea offered an explanation of how one simple ion could calm an overexcited system without sedating everything.

GSK-3

In 1996 the developmental biologists P. S. Klein and D. A. Melton, working on frog embryos, showed that lithium inhibits an enzyme called glycogen synthase kinase-3 (GSK-3), a master switch involved in cell signalling, growth and survival. Lithium inhibited GSK-3β at concentrations in the low millimolar range (an inhibition constant of about 2 mM), and its effects on the developing embryo matched those of blocking GSK-3. In the same study, complete inhibition of inositol monophosphatase did not reproduce lithium’s developmental effects, which suggested that in that system, at least, GSK-3 rather than inositol depletion was the key target. GSK-3 has since become a major focus of research into mood disorders and into the dementia questions raised in section 8.

An open question

The two hypotheses are not mutually exclusive, and lithium affects other targets too. Evidence from frog embryos and cell cultures does not automatically explain what happens in the human brain over years of treatment. The honest summary is that lithium’s clinical effects are well documented in trials, while its mechanism remains an active area of research — a fitting legacy for a medicine that Cade himself reached by a chain of reasoning that later scientists judged partly mistaken. For the clinical picture, see this site’s Bipolar Disorder page.

Back to Table of Contents

Key Research Papers

  1. Cade JF. Lithium salts in the treatment of psychotic excitement. Med J Aust. 1949;2(10):349-52. PubMed PMID: 18142718
  2. Schrauzer GN. Lithium: occurrence, dietary intakes, nutritional essentiality. J Am Coll Nutr. 2002;21(1):14-21. PubMed PMID: 11838882
  3. Strobusch AD, Jefferson JW. The checkered history of lithium in medicine. Pharm Hist. 1980;22(2):72-6. PubMed PMID: 11610717
  4. Ohgami H, Terao T, Shiotsuki I, Ishii N, Iwata N. Lithium levels in drinking water and risk of suicide. Br J Psychiatry. 2009;194(5):464-5. PubMed PMID: 19407280
  5. Memon A, Rogers I, Fitzsimmons SMDD, Carter B, Strawbridge R, Hidalgo-Mazzei D, Young AH. Association between naturally occurring lithium in drinking water and suicide rates: systematic review and meta-analysis of ecological studies. Br J Psychiatry. 2020;217(6):667-678. PubMed PMID: 32716281
  6. Kessing LV, Gerds TA, Knudsen NN, Jørgensen LF, Kristiansen SM, Voutchkova D, Ernstsen V, Schullehner J, Hansen B, Andersen PK, Ersbøll AK. Association of Lithium in Drinking Water With the Incidence of Dementia. JAMA Psychiatry. 2017;74(10):1005-1010. PubMed PMID: 28832877
  7. Corcoran AC, Taylor RD, Page IH. Lithium poisoning from the use of salt substitutes. J Am Med Assoc. 1949;139(11):685-8. PubMed PMID: 18110875
  8. Hanlon LW, Romaine M 3rd. Lithium chloride as a substitute for sodium chloride in the diet; observations on its toxicity. J Am Med Assoc. 1949;139(11):688-92. PubMed PMID: 18128981
  9. Trautner EM, Morris R, Noack CH, Gershon S. The excretion and retention of ingested lithium and its effect on the ionic balance of man. Med J Aust. 1955;42(8):280-91. PubMed PMID: 13264856
  10. Schou M. Biology and pharmacology of the lithium ion. Pharmacol Rev. 1957;9(1):17-58. PubMed PMID: 13431415
  11. Schou M. Forty years of lithium treatment. Arch Gen Psychiatry. 1997;54(1):9-13. PubMed PMID: 9006394
  12. Berridge MJ, Downes CP, Hanley MR. Neural and developmental actions of lithium: a unifying hypothesis. Cell. 1989;59(3):411-9. PubMed PMID: 2553271
  13. Klein PS, Melton DA. A molecular mechanism for the effect of lithium on development. Proc Natl Acad Sci U S A. 1996;93(16):8455-9. PubMed PMID: 8710892
  14. Malhi GS. Cade’s lithium: an extraordinary experiment with a not-so-ordinary element. Med J Aust. 2014;201(1):24-5. PubMed PMID: 24999881

PubMed Topic Searches

  1. Lithium in drinking water and suicide
  2. Lithium in drinking water and dementia
  3. Lithium as a trace element
  4. Lithium mechanism: GSK-3 and inositol

Further Reading

Back to Table of Contents

Connections