Is pyrrhotite magnetic?
Usually. Its sheet says magnetic, varying inversely with iron. Bennett and Graham call monoclinic Fe7S8 ferrimagnetic and hexagonal FeS to Fe9S10 normally not.
Usually, not always. The Handbook sheet for pyrrhotite, Fe1−xS (x = 0 to 0.17), hardness 3.5 to 4.5, says magnetic, varying in intensity inversely with iron content. USGS pulled 7 samples at a best 0.01 to 0.05 amp. Bennett and Graham (1980) call monoclinic Fe7S8 ferrimagnetic and the hexagonal phases, FeS to Fe9S10, normally antiferromagnetic.
In short
- Three sources, one direction. The Handbook of Mineralogy pyrrhotite sheet says magnetic, varying in intensity inversely with iron content. USGS Open-File Report 99-529, Table 1 pulled 7 samples, total range below 0.01 to 0.70 amp, best 0.01 to 0.05, and the report names pyrrhotite ferromagnetic. Bennett and Graham 1980 (American Mineralogist 65, 800–807) say the monoclinic phase is ferrimagnetic and the hexagonal phases normally antiferromagnetic, non-magnetic.
- The Fe9S10 line. Bennett and Graham give hexagonal for FeS to Fe9S10, monoclinic for Fe7S8, rhombohedral for Fe9S11. Worked: FeS 50.00 atomic % Fe, 63.53 wt%; Fe9S10 47.37%, 61.05 wt%; Fe7S8, the IMA formula, 46.67%, 60.38 wt%; Fe9S11 45.00%, 58.77 wt%. FeS to Fe9S10 is the composition range they call normally non-magnetic.
- Worked on the sheet. The Homestake mine and Kongsberg analyses convert to Fe0.868S, 46.46 and 46.48 atomic % Fe, beside Fe7S8’s 46.67. The Setregruben analysis converts to Fe0.917S, 47.85%, on the iron-rich side of the Fe9S10 line. The sheet does not say which structure each analysed piece had, so composition places a piece; it does not prove its structure.
- Documented failure: the line moves. Bennett and Graham report that hexagonal pyrrhotites may be made magnetic at will by different thermal cycles, and that fine magnetite lamellae are present in many pyrrhotites. Haines et al. 2020 (Journal of Physics: Condensed Matter 32, 385401) say it is generally held that monoclinic pyrrhotite will have a magnetic moment and hexagonal will not, but cite an exception. And USGS calls the whole span Fe5S6 to Fe16S17 ferromagnetic.
- The decision. A pull is consistent with pyrrhotite; no pull does not rule it out, since the iron-rich hexagonal kind is normally non-magnetic. Confirm with the dark grayish black streak, hardness 3.5 to 4.5 and bronze colour, as on pyrrhotite. Troilite, FeS, has no vacancies and no net moment in Bennett and Graham’s account.
| Composition | Where it comes from | x | Atomic % Fe | Fe by mass | Structure (source) | Magnetism (source) |
|---|---|---|---|---|---|---|
| FeS | Troilite (IMA, sheet); B&G’s iron-rich end | 0 | 50.00 | 63.53% (sheet ideal 63.53) | Hexagonal (troilite sheet; B&G) | B&G: no vacancies, no net moment. Sheet: none printed |
| Fe16S17 | USGS OFR 99-529, iron-rich end of its pyrrhotite span | 0.059 | 48.48 | 62.11% | Not stated | USGS: ferromagnetic |
| Fe0.917S | Setregruben analysis (sheet) | 0.083 | 47.85 | 61.57% measured | Not stated | Sheet: magnetic, varying inversely with iron content |
| Fe9S10 | B&G: iron-poor limit of the hexagonal phases | 0.100 | 47.37 | 61.05% | Hexagonal (B&G) | B&G: normally antiferromagnetic, non-magnetic |
| Fe7S8 | IMA formula; B&G monoclinic | 0.125 | 46.67 | 60.38% | Monoclinic (B&G, 4C) | B&G: ferrimagnetic |
| Fe0.868S | Homestake and Kongsberg analyses (sheet) | 0.132 | 46.46–46.48 | 60.18%, 59.83% measured | Not stated | Sheet: magnetic, varying inversely with iron content |
| Fe5S6 | USGS, iron-poor end of its span | 0.167 | 45.45 | 59.21% | Not stated | USGS: ferromagnetic |
| Fe0.83S | Handbook limit, x = 0.17 | 0.17 | 45.36 | 59.11% | Monoclinic, pseudohexagonal (sheet) | Sheet: magnetic, varying inversely with iron content |
| Fe9S11 | B&G rhombohedral; their smythite formula | 0.182 | 45.00 | 58.77% | Rhombohedral (B&G) | B&G: ferrimagnetic |
Is pyrrhotite magnetic? What the Handbook sheet says
The pyrrhotite sheet says magnetic, varying in intensity inversely with iron content. The Handbook of Mineralogy pyrrhotite sheet gives Fe1−xS with x = 0 to 0.17, monoclinic, pseudohexagonal, point group 2/m; crystals typically tabular or platy on {0001}, to 40 cm, also as rosettes and commonly massive, granular; twinning on {10-12}; parting on {0001}, distinct; hardness 3.5 to 4.5, density 4.58 to 4.65; bronze-yellow to pinchbeck-brown, tarnishing quickly; streak dark grayish black. The September 2026 IMA list writes the formula Fe7S8, status G, 1835.
What the sheet does not tie together. Its polymorphism line records numerous polytypes, with 4-M, 6-M, 5-H, 7-H and 11-H known, and a high-temperature hexagonal polymorph which may remain stable at ordinary temperatures. The sheet does not say which of these is magnetic. That link comes from the papers in the next section, not from the sheet.
Where magnetism sits among its neighbours. On the Handbook sheets, magnetite is strongly magnetic and cubanite strongly magnetic; pyrite prints paramagnetic; troilite, pentlandite and chalcopyrite print nothing. Which sheets record magnetism across the site is on which minerals are magnetic.
Monoclinic or hexagonal: which pyrrhotite is magnetic?
Bennett and Graham: the hexagonal phases are normally antiferromagnetic, non-magnetic, and the monoclinic and rhombohedral phases ferrimagnetic. Bennett and Graham 1980 (American Mineralogist 65, 800–807) write that one has tended to speak of hexagonal phases in the composition range FeS to Fe9S10, monoclinic for the Fe7S8 phase and rhombohedral for Fe9S11, though the situation is more complicated than this. They put the magnetic properties to a great extent down to how the vacant iron sites are ordered. In the monoclinic 4C structure every second iron layer is complete and the alternate layer has one site in four vacant, so that of the seven iron atoms in a formula unit, one has its spin unpaired.
A second, later source. Haines et al. 2020 (Journal of Physics: Condensed Matter 32, 385401) say crystals are expected to be ferrimagnetic if the vacancies are confined to alternate iron layers and antiferromagnetic if they occur equally in all layers; as a consequence, it is generally held that monoclinic pyrrhotite will have a magnetic moment whereas hexagonal pyrrhotite will not, with a pointer to Horng and Roberts (2018) as an exception. They give a Néel temperature of about 590 K, nearly constant across the solid solution.
Troilite. Bennett and Graham: if there are no vacancies, as in troilite, all spins are paired and there is no net moment. The troilite sheet gives FeS, hexagonal, hardness 3.5 to 4.5, density 4.67 to 4.79, and prints no magnetic property; USGS Table 1 has no troilite row.
Where is the line? Fe9S10, worked
Fe9S10 is the boundary Bennett and Graham give: FeS to Fe9S10 is the hexagonal range they call normally non-magnetic. Worked on standard atomic weights (Fe 55.845, S 32.06): FeS is x = 0, 50.00 atomic % iron, 63.53 wt%; Fe9S10, x = 0.100, 47.37%, 61.05 wt%; Fe7S8, x = 0.125, 46.67%, 60.38 wt%; Fe9S11, x = 0.182, 45.00%, 58.77 wt%. The Handbook’s limit, x = 0.17, is 45.36%. Fe9S11 lies beyond it; Bennett and Graham also give Fe9S11 as smythite, which the IMA list writes (Fe,Ni)3+xS4.
Worked on the sheet’s analyses. Homestake mine, South Dakota (60.18 Fe, 39.82 S) and Kongsberg, Norway (59.83 Fe, 39.55 S) both convert to Fe0.868S, 46.46 and 46.48 atomic % iron, beside Fe7S8; the sheet prints Fe0.87S1.00 for both. Setregruben, Norway (61.57 Fe, 38.53 S) converts to Fe0.917S, 47.85%, on the iron-rich side of Fe9S10; the sheet prints Fe0.92S1.00. The sheet does not say which structure or polytype each analysed piece was, so this places each by composition only.
Documented failure: the line moves. Bennett and Graham report that hexagonal pyrrhotites may be made magnetic at will by using different thermal cycles, and that fine magnetite lamellae are present in many pyrrhotites. A heated piece, or one carrying magnetite, can pull from the wrong side of the line. The composition sets an expectation, not a result.
What USGS measured, and where the words disagree
USGS pulled all its pyrrhotite early and calls the whole span ferromagnetic. USGS Open-File Report 99-529 (Rosenblum and Brownfield) ran 7 pyrrhotite samples: total range below 0.01 to 0.70 amp, best 0.01 to 0.05. Only magnetite and pentlandite, at a best 0.01, start as low in its 377-row table. The report defines ferromagnetic as strongly attracted to a magnet, like a piece of iron, and contrasts pyrite, FeS2, which it calls diamagnetic, with pyrrhotite, Fe5S6 to Fe16S17, which it calls ferromagnetic, putting the difference down to the atomic structures.
Documented failure: two vocabularies. USGS’s span runs from x = 0.059 to 0.167, so it includes compositions, such as Fe9S10 and Fe16S17, that Bennett and Graham call hexagonal and normally antiferromagnetic. USGS uses ferromagnetic as a practical separator term; the papers use ferrimagnetic and antiferromagnetic as physical terms. The two do not contradict each other on a sample USGS actually ran, but USGS does not say which structures its 7 samples had.
Not repeated here. The pentlandite row and the three-row line are on is pentlandite magnetic; the nickel rows on is millerite magnetic; hematite, ilmenite and chromite on is hematite magnetic, is ilmenite magnetic and is chromite magnetic.
How to read a magnet on a pyrrhotite specimen
A pull is consistent with pyrrhotite; no pull does not rule it out. No source fetched for this page tests a hand magnet, so this is a reading of the sources, not a test result. The monoclinic kind is ferrimagnetic and the iron-rich hexagonal kind normally not, per Bennett and Graham, so a bronze sulfide that ignores a magnet can still be pyrrhotite. Confirm with the streak, dark grayish black; hardness 3.5 to 4.5; and colour, bronze-yellow to pinchbeck-brown. The methods are on how to use a streak plate and how to test mineral hardness properly.
When the pull may be something else. The pyrrhotite sheet lists pentlandite and magnetite among its associates, and both pull at a best 0.01 amp on the USGS separator. A 20.00 g solid piece reads 15.63 to 15.70 g in water as pyrrhotite and 15.65 to 16.00 g as pentlandite, so weight will not split them; the streak will. The species page is pyrrhotite; the large crystals the sheet lists include Dalnegorsk, Primorskiy Kray.
On the label. Write pyrrhotite when streak, hardness and colour agree, whatever the magnet does. The magnetism questions are indexed at the collecting guides. Nothing on this site is offered for sale; the wanted list is the only commercial route.