Spice authenticity explorer
Which spices are adulterated, with what, and how anyone knows. Each risk says whether it was found in goods on sale, is reported in handbooks and reviews, or is only possible. Each names the methods that can detect it, and each method says what it cannot show. Choose a spice, a kind of fraud, a method or a kind of evidence, and the three lists narrow together. The same records, spice by spice and with the home tests assessed, are in the authenticity register.
Showing all 48 risks, with 11 methods and 18 worked examples.
Risks
What is substituted or added, to which spice, and what the evidence for it amounts to. Each names the methods that can detect it.
Asafoetida · Bulking agent
Excess flour or starch
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Beyond legitimate compounding, compounded powder can carry excess starch and gum, down to products with very little resin at all.
Forms affected: ground. A buyer can sometimes tell.
Methods that can detect it: Smell and taste, Microscopy
Asafoetida · Species substitution
Cheaper Ferula gums and other resins
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Substitution with gums from cheaper Ferula species, and with other resins, has been recorded. India’s statutory standard names the resins it forbids in asafoetida of every kind: colophony, galbanum, ammoniacum and any other foreign resin. That is a prohibition, not a finding in goods on sale.
Forms affected: whole, ground. A buyer can rarely tell.
Methods that can detect it: Chromatography (HPLC, GC), Mass spectrometry
Black pepper · Bulking agent
Starch-based fillers — rice, buckwheat, wheat, barley
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
In the European Commission's 2021 survey, 70 of 421 pepper samples (17 per cent) were suspicious of adulteration, and 37 of them contained starchy fillers from rice, buckwheat or other cereals. The sample was risk-based rather than random, so the rate describes suspicion in targeted product rather than prevalence on a shelf.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: DNA analysis, Microscopy, Infrared spectroscopy (FTIR, NIR)
Black pepper · Bulking agent
Undeclared mustard seed
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Nine ground pepper samples in the same survey contained mustard, which is a regulated allergen in the European Union. An undeclared allergen turns an economic fraud into a health hazard for the people it affects.
See also: Mustard seed
Forms affected: ground. Needs a laboratory.
Methods that can detect it: DNA analysis
Black pepper · Species substitution
Papaya seeds
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
The classic pepper fraud: dried papaya seeds are a similar size and colour, and have been used to bulk both whole and ground pepper. It is recorded in the fraud history both the Commission survey and the literature review describe.
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Microscopy, DNA analysis, Chromatography (HPLC, GC)
Black pepper · Exhausted material
Low-piperine or spent pepper
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Fourteen samples in the survey fell below the 4 per cent piperine minimum, ten of them also carrying undeclared plant material. Low piperine is the chemical signature of dilution or of pepper from which oleoresin has been extracted.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Chromatography (HPLC, GC)
Chilli powder · Added colour
Sudan dyes and other illegal azo colours
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Sudan I in chilli powder used by food manufacturers triggered mass recalls in Britain and across Europe in 2005. The European Food Safety Authority’s review concluded that Sudan I to IV, Para Red and Rhodamine B are or may be genotoxic and carcinogenic; none is a permitted food colour. In 20 samples of hot chilli bought in markets in five Egyptian governorates and analysed in 2025, Sudan I was found above the limit of quantification in every one.
Forms affected: ground, flakes. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC), Mass spectrometry
Chilli powder · Bulking agent
Brick dust, starch and spent material
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Bulking of ground chilli with brick dust, starch and material from which the colour or pungency has already been extracted has been recorded; ground spice is the easy vehicle for all of it.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Microscopy, Elemental analysis (XRF, ICP), Chromatography (HPLC, GC)
Cinnamon (Ceylon) · Species substitution
Cassia sold as cinnamon
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
The commonest integrity issue for this spice, and in many jurisdictions a lawful one: "cinnamon" is defined broadly enough to cover the cassias, so an unnamed species is a disclosure problem rather than fraud. Where a product is sold as Ceylon or true cinnamon and contains cassia, it is misdescription. The consequence is not only price: the coumarin difference between the species is the basis of the European exposure assessment. Of 104 cinnamons bought from retailers in eleven EU countries, four ground samples labelled Ceylon were wholly or partly cassia, and a fifth, sold as Sri Lankan without naming the species, was cassia. Only one of 16 ground cinnamons from Spanish supermarkets closely resembled Ceylon cinnamon.
See also: Cassia
Forms affected: ground, whole. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Microscopy, Infrared spectroscopy (FTIR, NIR), Chromatography (HPLC, GC), DNA analysis
Cinnamon (Ceylon) · Wrong plant part
Root, seed and other parts of the plant in ground bark
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
In the same EU retail study, six samples carried 5 to 40 times more camphor than the normal range. Camphor is the main volatile of the cinnamon root, so the authors read this as bark partly replaced by root, or by another camphor-rich Cinnamomum species. Fatty acid methyl esters in eight samples suggested seed in the powder. Substitution of Ceylon by cassia was not the commonest problem found: 66.3 per cent of the 104 samples failed an ISO quality criterion or an EU legal limit, were suspected of fraud, or held enough coumarin to concern children.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC)
Cloves · Wrong plant part
Clove stems in ground cloves
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Stems are a legitimate product in their own right and a recognised extender of ground cloves, where they lower the oil content while keeping the smell. Microscopy shows the stem tissue. The United States regulator acts on cloves at an average of 5 per cent or more stems by weight and lists the defect as aesthetic and as economic adulteration; India’s statutory standard caps what it calls tendrils and mother cloves together at 2 per cent in whole cloves. Neither is a finding in goods on sale.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: Microscopy, Chromatography (HPLC, GC)
Cloves · Exhausted material
Exhausted cloves
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Cloves from which the essential oil has been distilled keep their shape and colour, and have been returned to trade mixed with sound cloves. Low volatile oil and eugenol give them away.
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: Smell and taste, Chromatography (HPLC, GC)
Cumin · Bulking agent
Peanut and almond
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
In 2014 and 2015 ground cumin carrying undeclared peanut and almond protein was recalled across North America and Europe, and allergic reactions were reported. The contamination was traced back to the supply chain at origin; ground nut shells or flour used to bulk the spice is the explanation generally given. A Canadian regulator’s survey of 299 retail packs of cumin and paprika in late 2015 found undeclared peanut in seven, at 1.6 to 12 parts per million, levels it thought could come from cross-contamination. It left almond out of the survey because its laboratory method for almond cross-reacts with mahaleb.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: DNA analysis, Mass spectrometry
Cumin · Bulking agent
Other seeds, husk and plant material
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Bulking with cheaper seeds, husk and other plant material is the routine form of cumin fraud; about one cumin sample in seven was flagged as suspicious in the European Union’s 2021 coordinated survey.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Microscopy, Infrared spectroscopy (FTIR, NIR), DNA analysis
Cumin · Species substitution
Caraway sold as cumin
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
In the European Union’s 2021 coordinated survey, seven of 250 cumin samples contained a substantial amount of caraway while labelled as cumin, confirmed both by DNA and by carvone, the aroma compound typical of caraway. The survey also recorded why the confusion is easy to make innocently: in some member states the everyday word for cumin names caraway, so a label without the botanical name is ambiguous.
See also: Caraway
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: DNA analysis, Chromatography (HPLC, GC), Smell and taste
Cumin · Bulking agent
Undeclared mustard
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Mustard is a declared food allergen in the EU, and the 2021 coordinated survey found mustard DNA in nine cumin samples and mustard seed above the extraneous-matter limit in three. Whether added as filler or picked up through shared equipment, undeclared mustard is a hazard to anyone allergic to it.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: DNA analysis
Fennel seed · Exhausted material
Spent seed from oil distillation
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Seed left over after its essential oil has been distilled off reaches the market as fennel seed. It looks broadly like fennel and tastes of very little.
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: Smell and taste, Visual inspection, Chromatography (HPLC, GC)
Ginger · Exhausted material
Exhausted ginger
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Ginger left over after oleoresin extraction — its pungency and aroma removed — has been dried, ground and sold as spice. It looks like ground ginger and tastes of very little.
Forms affected: ground. A buyer can sometimes tell.
Methods that can detect it: Smell and taste, Chromatography (HPLC, GC)
Ginger · Bulking agent
Starch and other fillers
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Starch bulking of ground ginger has been recorded.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Microscopy
Mace · Species substitution
Bombay mace (Myristica malabarica)
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
The aril of a wild Indian relative of nutmeg, almost without aroma, has been used to extend and substitute for true mace. Its distinctive pigments make it detectable by chemical tests and chromatography. India’s statutory standard for mace shuts it out by name, together with a second aril it calls wild mace; a prohibition in a statute is not a finding that either is on sale.
Forms affected: ground, whole. A buyer can rarely tell.
Methods that can detect it: Chromatography (HPLC, GC), Microscopy, Smell and taste
Nutmeg · Species substitution
Myristica argentea sold as nutmeg and mace
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Of 22 nutmeg seeds and mace samples bought in Thai spice markets, 12 — six of ten mace samples and six of twelve seeds — were an oval-shaped type that DNA barcoding identified as Myristica argentea rather than M. fragrans. The two differed in their chemical profiles, including a lower myristicin content in the oval mace tested.
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Chromatography (HPLC, GC), DNA analysis
Oregano · Bulking agent
Olive leaves
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
The European Union’s 2021 coordinated survey flagged close to half of the oregano samples it tested as suspicious — by far the worst of the six spices surveyed — and olive leaf was the commonest adulterant found. A UK study using infrared spectroscopy and mass spectrometry found the same adulterant pattern.
Forms affected: leaf, ground. A buyer can rarely tell.
Methods that can detect it: Infrared spectroscopy (FTIR, NIR), Microscopy, Chromatography (HPLC, GC), Mass spectrometry, DNA analysis
Oregano · Bulking agent
Myrtle, sumac, cistus and hazelnut leaves
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Other dried leaves — myrtle, sumac, cistus and hazelnut among them — have been identified in commercial oregano.
Forms affected: leaf, ground. A buyer can rarely tell.
Methods that can detect it: Infrared spectroscopy (FTIR, NIR), Mass spectrometry, DNA analysis
Paprika · Added colour
Sudan dyes and other illegal azo colours
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Sudan I found in chilli powder used by food manufacturers triggered mass recalls across Britain and Europe in 2005. The European Food Safety Authority’s review concluded that Sudan I to IV, Para Red and Rhodamine B are or may be genotoxic and carcinogenic. None is a permitted food colour. Of 20 paprika samples bought in Egyptian markets and analysed in 2025, the authors report Sudan IV above the limit of quantification in 12 and Sudan I in 3.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC), Mass spectrometry
Paprika · Bulking agent
Ground stems, seeds and other plant material
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Ground paprika is bulked with cheaper plant material — the pepper’s own stems and seeds among it — which dilutes colour and flavour while passing as paprika by eye. The European Union’s 2021 coordinated survey found maize, carrot, tomato, sunflower seed and onion or garlic above the 1% extraneous-matter limit, and ten of its 27 suspicious paprika and chilli samples carried non-authorised dyes.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Microscopy, Infrared spectroscopy (FTIR, NIR), UV-visible spectrophotometry
Pomegranate molasses · Weight gain
Added glucose syrup
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
A study bought seven brands sold as pomegranate molasses in the shops of Bursa and tested each for glucose syrup: four contained it. The authors suspected it had been added to raise the dry matter. The registrations for the Hatay and Oğuzeli products both state that no sweetener and no additive is used.
One study, seven brands, one city, published in 2010, using a qualitative test that shows presence and not amount. A later study of five brands bought in Malatya measured sugars and did not test for added syrup, so it neither confirms nor contradicts the finding.
Forms affected: liquid. A buyer can rarely tell.
Methods that can detect it: Qualitative chemical test
Saffron · Species substitution
Safflower florets
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Safflower florets are the classic saffron substitute — sold outright as saffron in markets and mixed into genuine threads. They colour food yellow-orange and contribute none of saffron’s aroma or bitterness. In ground saffron the official colour-strength method cannot detect safflower, marigold or turmeric added below about 20 per cent; a mass-spectrometry screening method published in 2025 detected safflower down to 5 per cent and turmeric down to 3.
See also: Safflower
Forms affected: threads, powder. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Microscopy, Chromatography (HPLC, GC), Mass spectrometry, DNA analysis
Saffron · Wrong plant part
Yellow style mixed into red cuts
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Padding red saffron with the flower’s own yellow styles adds weight from the same plant, so it passes any test for species and fails only on colouring strength.
Forms affected: threads, powder. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, UV-visible spectrophotometry
Saffron · Wrong plant part
Dyed fibres, corn silk and other plant parts
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Dyed styles, corn silk, calendula petals and other plant fibres have all been recorded made up to look like saffron threads, alongside mixing with older or already-extracted saffron.
Forms affected: threads, powder. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Microscopy, DNA analysis
Saffron · Added colour
Added synthetic colour
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Synthetic dyes are added to lift the apparent colour of weak or padded saffron. The ISO test methods include checks for artificial colourants, because a colour reading alone can be fooled by them.
Forms affected: threads, powder. A buyer can rarely tell.
Methods that can detect it: Chromatography (HPLC, GC)
Saffron · Weight gain
Moisture, honey, glycerol, oil or sand
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Saffron is sold by the gram at a price that rewards anything adding weight: syrup, honey, glycerine and oil have all been recorded, and one analysed sample appeared to have been weighted with sand.
Forms affected: threads, powder. A buyer can rarely tell.
Methods that can detect it: Visual inspection, Elemental analysis (XRF, ICP), Chromatography (HPLC, GC)
Saffron · False origin
Saffron from elsewhere sold as Spanish
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Spain has long exported far more saffron than it grows — an average of about 36 tonnes a year against under 3 tonnes of harvest over 1997–2013 — because saffron bought from Iran and elsewhere is repacked there. That trade is lawful when labelled honestly; the fraud is the label. A metabolomic study of 44 commercial samples classed more than half of those sold as "Spanish saffron" as neither grown nor processed in Spain. Only the protected designation, Azafrán de La Mancha, guarantees Spanish origin.
Forms affected: threads, powder. Needs a laboratory.
Methods that can detect it: Mass spectrometry
Saffron · Added colour
Sudan I and other non-authorised dyes
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
In the European Union’s 2021 coordinated survey, non-authorised dyes were found in eleven saffron samples that contained no other adulterant, some with several dyes, and Sudan I in one sample made mostly of another plant. Sudan dyes are industrial colours; the European Food Safety Authority’s review concluded they are or may be genotoxic and carcinogenic, and none is a permitted food colour.
Forms affected: threads, powder. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC), Mass spectrometry
Smoked paprika · Misdescribed product
Unsmoked paprika with added smoke flavouring
Possible, with no case recorded
The sources describe how this product is exposed. They report no case of it in this product.
Unsmoked paprika treated with smoke flavouring and sold as smoked is the risk specific to this product, and one reason the La Vera designation certifies process as well as place.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Smell and taste, Chromatography (HPLC, GC)
Smoked paprika · Added colour
Sudan dyes and other illegal azo colours
Possible, with no case recordedHealth hazard
The sources describe how this product is exposed. They report no case of it in this product.
The illegal dye cases of 2005 involved chilli and paprika products generally, and the European Food Safety Authority found Sudan I to IV, Para Red and Rhodamine B are or may be genotoxic and carcinogenic. Smoked paprika is exposed on the same terms as unsmoked.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC), Mass spectrometry
Star anise · Species substitution
Japanese star anise (Illicium anisatum)
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Japanese star anise has been found in consignments of culinary star anise, and infusions made from contaminated product have caused poisonings, including in infants. The United States Food and Drug Administration issued an advisory on star anise teas in 2003, and the European Medicines Agency's assessment records the adulteration and the methods used to detect it.
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Microscopy, Chromatography (HPLC, GC), Mass spectrometry, DNA analysis
Sumac · Weight gain
Undeclared or excessive salt
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Salt is added freely to ground sumac, sometimes to a substantial fraction of the packet and not always prominently declared.
Forms affected: ground. A buyer can usually tell.
Methods that can detect it: Smell and taste, Visual inspection
Sumac · Added colour
Added red colour and beetroot
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Red colourants and beetroot powder are used to deepen the colour of faded sumac.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: UV-visible spectrophotometry, Chromatography (HPLC, GC)
Sumac · Misdescribed product
Added citric acid
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Citric acid is added to restore sourness to weak or old material, masking the loss of the fruit’s own acidity.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Chromatography (HPLC, GC)
Turmeric · Added colour
Lead chromate
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Lead chromate, a yellow industrial pigment, was documented being dusted onto turmeric rhizomes during polishing in Bangladesh to brighten them, and the practice was linked to elevated blood lead in the population studied. The evidence is specific to the supply chains investigated and should not be generalised to all turmeric. India’s statutory standard names the pigment: whole turmeric and turmeric powder must each be free from lead chromate and pass a test for it.
Forms affected: whole, ground. Needs a laboratory.
Methods that can detect it: Elemental analysis (XRF, ICP)
Turmeric · Added colour
Non-permitted dyes such as metanil yellow
Found in traded goodsHealth hazard
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Synthetic yellow dyes not permitted in spices have been added to ground turmeric to improve colour — metanil yellow is the one most often reported. In the European Union’s 2021 coordinated survey three of 316 samples carried non-authorised dyes: one Sudan I and two tartrazine, a colour permitted in some foods and not in spices. A 2026 review of 48 studies covering 2,235 commercial turmeric samples counted 20.0 per cent as adulterated; metanil yellow was the dye most often tested for, and 150 of 947 Indian samples tested for it were positive, while Sudan dye was reported in 28.2 per cent of 238 samples from Pakistan.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC), UV-visible spectrophotometry
Turmeric · Bulking agent
Starch, chalk and other fillers
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Ground turmeric is bulked with starches and mineral fillers, which dilute colour and are then often compensated for with added colour. India’s statutory standard requires turmeric powder to be free of foreign starch and caps total starch at 60 per cent by weight.
Forms affected: ground. A buyer can rarely tell.
Methods that can detect it: Microscopy, Infrared spectroscopy (FTIR, NIR)
Turmeric · Bulking agent
Undeclared chilli and cereal material
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Of 316 turmeric samples in the European Union’s 2021 coordinated survey, 24 contained DNA of undeclared plant material above the 2% extraneous-matter limit of the ISO specification — mostly chilli, and starchy crops such as maize, rice, oats and wheat.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: DNA analysis, Microscopy
Turmeric · Exhausted material
Low-curcuminoid or exhausted material
Found in traded goods
A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
Six samples in the same survey fell below the 2% curcuminoid minimum of the ISO specification, and one contained curcumin alone, without the demethoxycurcumin and bisdemethoxycurcumin that turmeric always carries, and no detectable turmeric DNA — consistent with a colour made to look like turmeric rather than the spice.
Forms affected: ground. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC), DNA analysis
Vanilla · Misdescribed product
Synthetic or lignin-derived vanillin in "natural" vanilla
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Vanillin made from guaiacol or from lignin costs a small fraction of bean vanillin and is chemically identical, so passing it off as natural extract is the central fraud in the trade. It is detected by stable-isotope ratios, which differ with the plant or petrochemical origin of the carbon and hydrogen, and by the absence of the minor compounds a real bean extract carries.
Forms affected: extract, powder, paste. Needs a laboratory.
Methods that can detect it: Stable isotope ratio analysis, Chromatography (HPLC, GC), Mass spectrometry
Vanilla · Misdescribed product
Coumarin in cheap vanilla-flavoured extracts
Reported in the literatureHealth hazard
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Tonka bean, which is rich in coumarin, has a vanilla-like smell and has been used in inexpensive "vanilla" extracts. Coumarin is prohibited as a food additive in the United States and limited in the European Union on account of its liver toxicity at high intake.
See also: Tonka bean
Forms affected: extract. Needs a laboratory.
Methods that can detect it: Chromatography (HPLC, GC)
Voatsiperifery · Species substitution
Other Malagasy wild peppers
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Because of its popularity and price, voatsiperifery has been adulterated with related Malagasy wild peppers — Piper malgassicum and Piper tsarasotrae are named — which are gathered in the same forests and look much the same.
Forms affected: whole. A buyer can rarely tell.
Methods that can detect it: Chromatography (HPLC, GC), Mass spectrometry
Voatsiperifery · Species substitution
Cultivated pepper
Reported in the literature
A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
Cultivated peppers of commerce have been recorded mixed into lots sold as voatsiperifery.
See also: Black pepper
Forms affected: whole, ground. A buyer can sometimes tell.
Methods that can detect it: Visual inspection, Smell and taste, Chromatography (HPLC, GC)
Methods, and where each stops
The methods named by the risks shown above: what each measures, what it can tell apart, and what it cannot.
Visual inspection
What it shows, and where it stops:Looking at the product, with a lens if needed. Useful for whole spices and almost useless for ground ones: saffron threads, cinnamon quills, peppercorns and star anise can be examined for shape, colour and foreign pieces; a powder cannot, and the fillers that matter most in powders are chosen because they look the same.
Smell and taste
What it shows, and where it stops:Smell and taste. They reliably reveal staleness, exhausted material and gross substitution — ground cloves with no numbing warmth, cinnamon with the sharp bite of cassia — but not the adulterants that matter most: a dye, a heavy metal or a well-chosen filler has no flavour of its own at the levels used.
Microscopy
What it shows, and where it stops:Identifies plant tissue by its cells: olive or myrtle leaf in oregano, starch grains in a spice that should have none, stems and husk in a powder. Cheap and fast, and still one of the most reliable screens for bulking with other plant material. It needs a trained analyst and reference material, and it cannot see a dissolved dye or a mineral pigment.
Qualitative chemical test
What it shows, and where it stops:A reagent added to a sample that gives a colour or a precipitate when a class of substance is present. One study of bottled pomegranate sour tested for glucose syrup this way and found it in four of seven brands. Such a test shows that something is there and not how much, it answers only for the substance its reagent reacts with, and its result is no better than the judgement of the analyst reading it.
UV-visible spectrophotometry
What it shows, and where it stops:Measures a compound by how much light a solution of it absorbs at a defined wavelength. It is the basis of saffron grading, where crocin, picrocrocin and safranal are read at three wavelengths, and of paprika colour measurement. It shows dilution well; it cannot always say what the diluent was, and a synthetic dye absorbing in the same region can inflate a colour reading, which is why the saffron standard adds a separate test for artificial colours.
Saffron grading by ISO 3632
What it measures:The ISO 3632 test reads a water extract of saffron at three wavelengths, for crocin (colouring strength), picrocrocin (bitterness) and safranal (aroma), and sorts the result into commercial categories. It measures how much of the compounds that make saffron’s colour and flavour a sample carries, which is what a saffron grade is.
What it cannot prove:A grade is not proof of purity. The readings are absorbances, so a synthetic dye absorbing in the same region can inflate the colouring figure; and saffron bulked with look-alike plant material can still fall within a commercial category — in one survey, removing the adulterants moved samples from Category III to Category II. In a study comparing methods, the ISO safranal reading did not correlate with safranal measured by chromatography.
Infrared spectroscopy (FTIR, NIR)
What it shows, and where it stops:A fingerprint of a sample’s overall chemical composition, read in seconds and compared against authentic references. Suited to screening many samples for anything unusual — it was the main screen of the largest European survey, with suspicious samples confirmed by other methods. It flags that something is wrong more readily than it identifies what, and it is only as good as its reference library.
Infrared screening with a chemometric model
What it measures:Infrared spectrometers — mid-infrared (FTIR) and near-infrared (NIR), benchtop or handheld — record a spectrum of the whole sample in seconds without separating it, and a statistical model trained on authentic and adulterated samples classifies new ones. For black pepper, benchtop FTIR-ATR and FT-NIR models told authentic pepper from adulterant samples with 100% prediction accuracy, in a study covering other peppers and 27 adulterants.
What it cannot prove:A model recognises only what it was trained on, and its accuracy describes the samples used to build and test it — in the pepper study, samples spiked with known adulterants at 5–95%. It flags that a sample differs from the references more readily than it identifies what is wrong, so a flagged sample needs a targeted method to confirm.
Chromatography (HPLC, GC)
What it shows, and where it stops:Separates a sample into its individual compounds — liquid chromatography for dyes and pigments, gas chromatography for volatile aroma compounds — so an analyst can find a compound that should not be there, such as a Sudan dye in chilli, or the absence or imbalance of one that should, such as coumarin revealing cassia in a product sold as Ceylon cinnamon. It answers the question it is set up to ask; an adulterant nobody looked for passes.
Targeted LC-MS/MS for illegal dyes
What it measures:Liquid chromatography separates an extract and tandem mass spectrometry identifies each compound by its mass and fragments, so the method confirms and quantifies named illegal dyes such as the Sudan dyes at trace levels. One method validated to EU requirements covers fourteen dyes in chilli powder, sauces and flavourings, with detection limits of 0.03–0.44 µg/kg.
What it cannot prove:A targeted method finds only what it is built to look for. A clean result means none of the listed dyes was present above its limit, not that the product contains no added colour.
Volatile profiling with a chemometric model
What it measures:Samples the volatile compounds above a spice and separates them by gas chromatography — here followed by ion mobility spectrometry — to give a profile of its aroma chemistry. Used untargeted, the whole profile is compared with authentic material: for oregano adulterated with olive leaf, one model classified every validation sample correctly and a second estimated the percentage of olive leaf.
What it cannot prove:The model was built and validated against two kinds of olive leaf. A different adulterant lies outside what it was validated for, and its percentages are estimates from calibration against those two.
Mass spectrometry
What it shows, and where it stops:Identifies compounds by their mass, usually after chromatography (LC-MS or GC-MS). It confirms what a screen has flagged and finds very small amounts. In untargeted use it compares a sample’s whole chemical profile with authentic references, which is how most saffron sold as "Spanish" in one study was shown not to have been grown in Spain — a question of origin that no single marker compound answers. Used in targeted form it finds only the compounds it was set up to look for, so a clean result covers those and no others.
Targeted LC-MS/MS for illegal dyes
What it measures:Liquid chromatography separates an extract and tandem mass spectrometry identifies each compound by its mass and fragments, so the method confirms and quantifies named illegal dyes such as the Sudan dyes at trace levels. One method validated to EU requirements covers fourteen dyes in chilli powder, sauces and flavourings, with detection limits of 0.03–0.44 µg/kg.
What it cannot prove:A targeted method finds only what it is built to look for. A clean result means none of the listed dyes was present above its limit, not that the product contains no added colour.
Ambient (DART) tandem mass spectrometry screening
What it measures:Direct Analysis in Real Time ionises a sample in open air in front of a triple-quadrupole mass spectrometer, which watches for fragment transitions specific to known adulterants, without a chromatographic separation. A method of this kind detected turmeric in saffron down to 3 per cent and safflower down to 5 per cent, at about twenty analyses in ten minutes.
What it cannot prove:It is targeted: it looks for the transitions of the adulterants it was built for, here safflower and turmeric, and says nothing about others such as marigold, dyes or exhausted saffron. The authors present it as a rapid screening tool for testing laboratories.
DNA analysis
What it shows, and where it stops:Identifies which plant species are present, by barcoding or targeted PCR. It is the method for species substitution — cassia sold as cinnamon, one herb for another, other plant material in a powder. It cannot detect dyes, minerals or exhausted material, and it weakens on heavily processed or heat-treated product, where DNA is degraded.
DNA barcoding and metabarcoding
What it measures:Reads short, standard regions of plant DNA — such as ITS2, matK, rbcL and trnH-psbA — that are conserved within a species and differ between species, to identify which plants are present. Metabarcoding sequences many at once, so one test can list the species in a mixed product without being told what to look for; in a proof-of-concept study it found undeclared mint in oregano and species substitutions, and three markers together were more sensitive than one.
What it cannot prove:It identifies species, not where or how they were grown or how much aroma is left, and it cannot see dyes, minerals or exhausted material. A species it does not find is not proven absent: in the same proof-of-concept study, DNA degraded by heavy processing, gaps in the reference database and natural variation within species limited detection in some cases.
Stable isotope ratio analysis
What it shows, and where it stops:Distinguishes chemically identical molecules by the ratios of stable isotopes their origin leaves in them. It is how vanillin from vanilla beans is told apart from vanillin made from lignin or petrochemicals, which no other test can separate because the molecule is the same. A ratio places a sample only against reference values measured on authentic material, and those reference sets are still being built.
Stable isotope ratio analysis of vanillin
What it measures:Vanillin is the same molecule whether it comes from cured vanilla beans, lignin or petrochemicals, so no test of its structure can tell the sources apart. The ratios of carbon-13 to carbon-12 and of deuterium to hydrogen can, because each source leaves its own isotopic signature; they are measured on the vanillin itself, and are also used to trace a vanilla’s geographical origin.
What it cannot prove:A ratio places a sample only against the reference values of authentic material it is compared with, which is why reference profiles of authentic pods are still being built. The authors of one such study note that increasingly sophisticated fraud requires the methods to keep being refined, and they added minor aroma compounds as further markers.
Isotope ratio and elemental profiling for origin
What it measures:Combines stable isotope ratios of the whole spice — carbon, nitrogen and sulphur — with the concentrations of many trace elements, and classifies samples by a statistical model trained on samples of known origin or production method. In one market study, markers such as δ13C, δ34S, rubidium, caesium, vanadium, iron and aluminium classified paprika from Hungary, Serbia and Spain with about 90 per cent accuracy and cinnamon from Sri Lanka, Madagascar and Indonesia with 89 per cent; organic paprika showed higher δ15N, δ34S and zinc.
What it cannot prove:The model can only assign a sample to an origin it was trained on, and its accuracy describes the samples used to build it. The authors call their classification preliminary and say larger, more balanced sets of samples are needed before the markers can be generalised; a sample from an origin outside the reference set cannot be placed at all.
Elemental analysis (XRF, ICP)
What it shows, and where it stops:Measures elements such as lead and chromium, by X-ray fluorescence or plasma methods, and so detects mineral pigments like the lead chromate found in turmeric, which no organic method would see. It measures elements, not compounds: lead chromate is inferred from lead and chromium found together, not identified as such.
X-ray fluorescence screening for lead
What it measures:Measures elements directly in a sample, without dissolving it, and handheld analysers work in a market. For lead in turmeric it was calibrated against ICP-MS, the laboratory reference: results on powder agreed closely (R² 0.99, mean error 0.82%), and on whole dried roots less closely (R² 0.98, mean error 18.36%), because a root is uneven where a powder is uniform. Lead chromate shows as chromium found alongside the lead.
What it cannot prove:It measures elements, not compounds: lead chromate is inferred from lead and chromium occurring together, not identified as such. On roots it is semi-quantitative, within about 24%, and tends to read slightly high at higher concentrations. A screen is not a confirmation — the tiered approach proposed for turmeric uses portable XRF to screen and mass spectrometry to confirm.
Isotope ratio and elemental profiling for origin
What it measures:Combines stable isotope ratios of the whole spice — carbon, nitrogen and sulphur — with the concentrations of many trace elements, and classifies samples by a statistical model trained on samples of known origin or production method. In one market study, markers such as δ13C, δ34S, rubidium, caesium, vanadium, iron and aluminium classified paprika from Hungary, Serbia and Spain with about 90 per cent accuracy and cinnamon from Sri Lanka, Madagascar and Indonesia with 89 per cent; organic paprika showed higher δ15N, δ34S and zinc.
What it cannot prove:The model can only assign a sample to an origin it was trained on, and its accuracy describes the samples used to build it. The authors call their classification preliminary and say larger, more balanced sets of samples are needed before the markers can be generalised; a sample from an origin outside the reference set cannot be placed at all.
Worked examples
What one study reported when it used one of these methods on one spice. An example is evidence about the method. It is not always about a risk listed above, and some used samples the researchers prepared themselves.
Saffron · Saffron grading by ISO 3632
The official approach, UV-visible spectroscopy with liquid chromatography of crocin, is reported to miss safflower, marigold or turmeric when they are added in small proportions.
A method of this kind: UV-visible spectrophotometry
Saffron · Saffron grading by ISO 3632
In a survey reported alongside a microscopy method for spotting look-alikes, only 40% of saffron samples met ISO Category I, and the rest averaged 36.25% adulteration.
A method of this kind: UV-visible spectrophotometry
Cinnamon (Ceylon) · DNA barcoding and metabarcoding
Of six barcode regions compared, trnH-psbA told Cinnamomum verum apart from C. aromaticum, C. burmannii and C. loureiroi, and a high-resolution melting assay built on it distinguished the four species in commercial products; DNA of the mould Aspergillus flavus spiked into a mixture was also detected.
A method of this kind: DNA analysis
Saffron · DNA barcoding and metabarcoding
Of three barcodes compared, psbA-trnH gave different product sizes for saffron, safflower and calendula, and was judged the most useful for detecting those two as adulterants.
A method of this kind: DNA analysis
Turmeric · DNA barcoding and metabarcoding
Of three loci tested (rbcL, ITS and matK), ITS best told the Curcuma species apart.
A method of this kind: DNA analysis
Oregano · DNA barcoding and metabarcoding
In a proof-of-concept study, nanopore metabarcoding detected mint that had not been declared in oregano.
A method of this kind: DNA analysis
Vanilla · Stable isotope ratio analysis of vanillin
Measured on authentic pods by compound-specific isotope analysis, vanillin of Vanilla planifolia gave δ13C of −20.5‰ to −19.1‰ and V. tahitensis about −16.5‰, a significant difference between the species; δ2H ran from −99‰ to −63‰ and varied with origin. The authors place vanillin from lignin or petroleum at −36.2‰ to −24.9‰ and from glucose near −12.5‰, which is why a blend of the two can imitate the natural δ13C and why δ2H is measured as well.
A method of this kind: Stable isotope ratio analysis
Turmeric · X-ray fluorescence screening for lead
Of 503 turmeric samples bought in five eastern Indian states in 2021–2023, 30% exceeded India’s permissible lead limit of 10 µg/g; in Bihar the geometric mean was 48 µg/g and the highest sample 6,416 µg/g. Every sample above 10 µg/g also held chromium at levels suggesting lead chromate.
A method of this kind: Elemental analysis (XRF, ICP)
Chilli powder · Targeted LC-MS/MS for illegal dyes
Applied to 2,350 samples in Egypt over four years, the method found illegal dyes in 18.62% of chilli powders; Sudan IV and Red B were the commonest, each in 15.86% of them.
A method of this kind: Chromatography (HPLC, GC), Mass spectrometry
Cinnamon (Ceylon) · Infrared screening with a chemometric model
In 46 cinnamons from the Slovenian market, FTIR spectra told Ceylon from cassia cinnamon.
A method of this kind: Infrared spectroscopy (FTIR, NIR)
Paprika · Infrared screening with a chemometric model
In 45 paprikas from the Slovenian market, FTIR showed spectral features the authors read as consistent with removal of the oleoresin or addition of an azo dye — a suspicion that, in their words, needs further verification, not a finding of fraud.
A method of this kind: Infrared spectroscopy (FTIR, NIR)
Black pepper · Infrared screening with a chemometric model
Two handheld near-infrared instruments did less well than the benchtop ones: the microNIR 1700ES classified 91.30% of samples correctly and the SCiO 86.96%.
A method of this kind: Infrared spectroscopy (FTIR, NIR)
Coriander seed · Infrared screening with a chemometric model
On 200 authentic Indian coriander seed samples and 90 that the researchers had themselves cut with starch, salt or sawdust, the best models from a benchtop near-infrared instrument sorted every sample correctly, the best from a portable one classed 98.5% of the authentic and all of the adulterated samples correctly, and a handheld consumer device caught the adulterated samples while passing somewhat fewer of the genuine ones, though still over 90%. These were prepared samples, not coriander found adulterated in trade.
A method of this kind: Infrared spectroscopy (FTIR, NIR)
Cinnamon (Ceylon) · Volatile profiling with a chemometric model
Volatile and other chemical markers separated Ceylon from cassia: eugenol, β-phellandrene, δ-3-carene and cryptone were found only in Ceylon cinnamon, and coumarin and a set of sesquiterpenes characterised cassia. Of 16 ground cinnamons from supermarkets in Alicante, Spain, one resembled Ceylon, one was a mixture and the rest were cassia.
A method of this kind: Chromatography (HPLC, GC)
Oregano · Volatile profiling with a chemometric model
Applied to 15 commercial oregano samples, the method classified two as adulterated, with an estimated 31% and 43% olive leaf.
A method of this kind: Chromatography (HPLC, GC)
Paprika · Isotope ratio and elemental profiling for origin
Forty-five paprika samples from the Slovenian market were classified by country of origin with a preliminary accuracy of 90 per cent.
A method of this kind: Stable isotope ratio analysis, Elemental analysis (XRF, ICP)
Cinnamon (Ceylon) · Isotope ratio and elemental profiling for origin
Among 46 cinnamons, sulphur isotopes and barium, with other elements, separated production practices with 95 per cent accuracy.
A method of this kind: Stable isotope ratio analysis, Elemental analysis (XRF, ICP)
Saffron · Ambient (DART) tandem mass spectrometry screening
Turmeric was detected in saffron at levels as low as 3 per cent and safflower at 5 per cent, both far below the roughly 20 per cent that the official spectrophotometric approach is reported to miss.
A method of this kind: Mass spectrometry
How to read a risk
Every risk shown has a source. What its sources establish varies, and each card says which.
- Reported in the literature
- A handbook, a review or a trade body records that it happens. No specific finding in goods on sale is cited.
- Found in traded goods
- A survey, a recall, an official advisory or a published study found it in goods that were on sale or in trade.
- Possible, with no case recorded
- The sources describe how this product is exposed. They report no case of it in this product.
A finding belongs to the samples that were tested. Surveys of this kind choose products thought likely to be adulterated, so a rate describes those samples and is not how often a product on a shelf is affected. A risk recorded for a spice says nothing about any one packet of it.
A method is not a verdict. A result answers the question the method was set up to ask: a clean result from a test for named dyes covers those dyes, and a DNA result names species and says nothing about dyes, minerals or origin. Where no risk appears for a spice, none has been read. It does not mean the spice is never adulterated.