
Alternative Flours: Chickpea, Buckwheat, Quinoa
Only one of these three flours is clearly more nourishing than wheat flour: chickpea, with 22.4 g of protein and 10.8 g of fibre per 100 g against 13.6 g and 3.5 g for a medium-strength white wheat flour (Ciqual, ANSES). Buckwheat has less protein than white wheat flour. Quinoa has roughly the same. None of the three replaces wheat gram for gram in a risen dough, because none of them forms gluten. And a “gluten-free” label is not the same thing as “naturally gluten-free” from a bulk bin, which is the one point in this article that is about safety rather than taste.
That is the short version. The rest is about dosing, about why a dough tears, and about the two or three situations where these flours need genuine care.
What these flours actually contain
The figures below come from the ANSES Ciqual 2020 table, per 100 g of raw flour. I have put three wheat flours alongside them, because comparing an alternative flour to “wheat flour” without saying which one is meaningless: between a fine pastry flour and a wholemeal one, the fibre gap is fourfold.
| Per 100 g | Energy | Protein | Carbohydrate | Fibre | Iron | Magnesium | Folate |
|---|---|---|---|---|---|---|---|
| Chickpea flour | 359 kcal | 22.4 g | 47 g | 10.8 g | 4.86 mg | 166 mg | 437 µg |
| Buckwheat flour | 348 kcal | 11.5 g | 68.4 g | 4.2 g | 2.5 mg | 157 mg | 66.8 µg |
| Quinoa (raw grain) | 358 kcal | 13.2 g | 58.1 g | 7 g | 4.57 mg | 197 mg | 184 µg |
| White pastry flour (T45) | 356 kcal | 9.06 g | 75.9 g | 2.5 g | 1.18 mg | 25 mg | 18 µg |
| Plain white flour (T65) | 346 kcal | 13.6 g | 67.6 g | 3.5 g | 1.31 mg | 30.4 mg | 18.3 µg |
| Wholemeal flour (T150) | 342 kcal | 11.4 g | 64.9 g | 10.2 g | 2.7 mg | 85.1 mg | 52.4 µg |
Three things stand out. First, chickpea flour is playing a different sport: two and a half times the protein of pastry flour, four times the fibre, four times the iron, and 437 µg of folate per 100 g when the adult reference intake is 200 µg a day according to the NHS. It is a pulse flour, not a cereal flour, and it shows.
Second, buckwheat. Its nutritional reputation runs ahead of its composition. At 11.5 g of protein it has less than plain white wheat flour, and its 4.2 g of fibre sit well below the 10.2 g of a wholemeal. Where it earns its place is magnesium: 157 mg against 30.4 mg, a fivefold difference.
Third, quinoa, which is neither the disaster nor the miracle it gets described as. Its 13.2 g of protein matches a good white wheat flour. Its real interest lies elsewhere, in iron, magnesium and folate, plus a fat content of 6.07 g that explains both its flavour and its short shelf life.
One thing to keep in mind throughout: these values are for bare flour. The moment you put it in a recipe with water, sugar or oil, everything dilutes. A flour with 22 g of protein, in a batter hydrated at 200 per cent, gives a pancake with 5.6 g. That is not a flaw, it is just how a recipe works.
Protein deserves a number, not an adjective
“Complete protein”, “high-quality protein”: these phrases circulate widely and mean very little without a score. The current measure is DIAAS, which combines amino acid profile with actual measured digestibility.
A study published in Plant Foods for Human Nutrition in 2025 measured these values for the flours it used to formulate muffins: 34 per cent for wheat flour, 50 per cent for quinoa flour, 77 per cent for white bean flour (Europe PMC). Quinoa therefore does considerably better than wheat while staying a long way from the score of 100 that would mark a complete protein in the strict sense. The word “complete” in popular writing comes from the fact that quinoa contains lysine, the amino acid cereals lack. That is true and useful, but it is a different claim.
Staying with quinoa, an analysis of three varieties published in Food Chemistry: X in 2024 reports in vitro protein digestibility above 76.9 per cent for all three, with oxalate levels from 396.9 to 715.2 mg per 100 g (Europe PMC). Oxalates are why anyone prone to calcium oxalate kidney stones should raise it with their doctor before making quinoa a daily staple. For everyone else, the amount eaten in a bake stays modest.
If the goal is the protein density of a meal, the simplest route is not to hunt for the perfect flour but to pair a pulse with a grain in the same dish, a principle covered in our article on combining grains and legumes.
None of them replaces wheat gram for gram
Gluten is not an ingredient, it is a network. Two wheat proteins, gliadin and glutenin, hydrate and get worked, and the resulting structure holds fermentation gases and gives dough its elasticity. None of the three flours here has it. A dough made entirely of buckwheat or chickpea will not rise, will not stretch and tears when you shape it.
That does not disqualify them, it defines what they are for. They handle beautifully anything that needs no elasticity: thin batters poured and cooked flat, mixtures bound by something other than gluten, shortcrust pastry where the absence of a network is the whole point, and sauce thickening.
For anything that needs structure, you either keep to partial substitution of wheat or add a binder. A rheology study published in Food Science and Technology International in 2021 measured what happens as wheat flour is progressively replaced by chickpea flour from 0 to 40 per cent: farinograph water absorption falls, the mixing tolerance index rises, and resistance to extension drops (Europe PMC). In plain terms, the dough gets softer and tears more easily the further you push it.
In practice, these rates hold up:
- Cakes, muffins and soft bakes: up to 30 per cent chickpea or quinoa flour with no other change. Beyond that the crumb turns dense and the flavour takes over.
- Shortcrust or sweet pastry: up to 50 per cent, because the missing network is an advantage here. The pastry will be more crumbly, so roll it between two sheets of paper.
- Risen bread: 15 to 20 per cent at most. The remaining wheat is doing all the structural work.
- Pancakes and galettes: 100 per cent is fine, since nothing has to rise.
- Sauce thickening: the same proportions as wheat flour, roughly 10 g per 250 ml, slaked cold.
The nutritional gain from partial substitution is measurable. On a 200 g flour base, moving from all white pastry flour to 70 per cent wheat and 30 per cent chickpea raises protein from 18.1 to 26.1 g and fibre from 5 to 10 g, at essentially identical energy (712 against 714 kcal). The share of energy coming from protein goes from 10.2 to 14.6 per cent. That detail matters: Annex 13 of the Swiss ordinance on food information sets 12 per cent of energy as the threshold for a “source of protein” claim and 20 per cent for “high in protein”. The all-white-flour base misses the first threshold; the 30 per cent chickpea base clears it.
Hydration is the setting that matters
This is where most attempts go wrong. Pulse and pseudocereal flours do not take up water the way wheat does, and more importantly they do not release it the same way either.
Chickpea flour drinks fast and forms lumps with alarming ease. The fix is simple: mix it with the dry ingredients first, then add liquid in three goes while whisking, never the other way round. A rest of 15 to 30 minutes does the rest, letting the starch finish hydrating.
Buckwheat behaves much like wheat flour on hydration but thickens cold during the rest. A galette batter that looks perfect under the whisk will be too thick an hour later. Allow 10 to 15 per cent more liquid, or let it down again before cooking.
Quinoa absorbs less and gives a batter that looks runny when it is actually correct. The urge to add flour to fix the texture is exactly what produces dry muffins.
For bakes with no wheat at all, you need a binder. Psyllium husk at 5 to 8 g per 250 g of flour swells and creates a viscosity that crudely mimics gluten’s role. Ground flaxseed or chia, mixed with three times their volume of water, does something similar. Egg works too and is not always necessary: substitution techniques are covered in our egg-free baking guide. And if the aim is to build a flour blend that behaves predictably, our guide to gluten-free flours for baking deals with that specifically.
Gluten-free: what the law says and what the analyses find
This is the most important section here for anyone with coeliac disease. Coeliac disease is an autoimmune condition in which the immune system attacks the lining of the small intestine in the presence of gluten, and the only treatment is strict, permanent avoidance (NHS). It is not a dietary preference.
In Switzerland, article 41 of the ordinance on food information permits a “gluten-free” statement when the content does not exceed 20 mg/kg in the product as sold to the consumer, and “very low gluten” up to 100 mg/kg for products based on specially processed cereals. Chickpeas, buckwheat and quinoa contain no gluten by nature. The plant is never the problem. The neighbouring field, the combine, the silo, the mill, the shop aisle and the shared scoop are.
The measurements are unambiguous. A market study published in Frontiers in Nutrition in 2025 analysed 163 flours by R5 ELISA: detectable contamination was found in 16.67 per cent of flours carrying a gluten-free claim, in 50 per cent of packaged but unlabelled naturally gluten-free flours, and in 84.91 per cent of naturally gluten-free flours sold loose (Europe PMC). Oats and buckwheat were the worst affected, and the use of a shared scoop at the point of sale raised measured levels significantly.
A Canadian survey published in 2026 in the Journal of Food Protection tested 657 samples: 4.6 per cent exceeded the Codex level of 20 mg/kg, but only 0.4 per cent of those carrying a gluten-free claim, against 18.9 per cent of those carrying only a precautionary statement (Europe PMC). Nineteen of the thirty non-compliant samples listed oats.
The practical conclusion fits in one line: for someone with coeliac disease, a certified gluten-free flour in a sealed bag and a buckwheat flour bought loose are not the same product, whatever the plant of origin. For someone cutting back on gluten for digestive comfort, the stakes are far lower. Cross-contact in a shared kitchen is covered more broadly in our article on cooking without gluten, nuts or soy.
The two risks nobody mentions
Buckwheat allergy
It is little known in Europe and it is serious. A review published in 2025 in the Asian Pacific Journal of Allergy and Immunology notes that it can present as urticaria, rhinoconjunctivitis, asthma or anaphylaxis, that the allergen Fag e 2 is recognised by IgE in up to 90 per cent of symptomatic patients, and that no approved immunotherapy exists, leaving strict avoidance as the only management (Europe PMC). A review in Current Allergy and Asthma Reports in 2024 also describes food-dependent, exercise-induced anaphylaxis after ingestion (Europe PMC).
Two things follow. Occupational exposure through inhaling flour is a documented route of sensitisation, which matters for anyone handling large volumes. And because buckwheat turns up more and more in products marketed as gluten-free, someone with the allergy has to read those ingredient lists as carefully as any other. Swelling of the lips or throat, difficulty breathing or collapse after eating are emergency symptoms and need immediate help (NHS).
Raw flour
A pancake batter resting an hour on the worktop, a scraped bowl, a taste of raw cookie dough: it is the ordinary gestures that cause the trouble. Flour is a raw agricultural product, not a sterilised one.
A study of 70 flours bought from Swiss retailers, published in the Journal of Food Protection in 2019, found stx genes in 12.9 per cent of samples, with Shiga toxin-producing Escherichia coli successfully isolated from eight of the nine positives, including serogroups O103 and O146 associated with human cases (Europe PMC). The authors state plainly that consuming raw or undercooked flour is a risk.
Nothing dramatic follows from this, just three habits. Rest batters in the fridge rather than at room temperature. Do not taste raw batter, and do not let a child do it. Cook things through: a pancake or galette is done after a few minutes a side, which is not the issue, but a thick batter barely set in the middle is.
Three weighed recipes, with what they actually deliver
The compositions below are calculated ingredient by ingredient from Ciqual values, on the raw preparation. They do not account for evaporation during cooking, which concentrates things slightly.
Chickpea pancakes, socca style
150 g chickpea flour, 300 ml water, 15 ml rapeseed oil, 3 g salt, 1 teaspoon cumin.
Mix the flour, salt and spices dry. Add the water in three goes, whisking. Stir in the oil. Rest for 30 minutes in the fridge. Cook in a properly hot pan, 2 to 3 minutes a side.
For 6 pancakes: 112 kcal, 5.6 g protein, 2.7 g fibre, 1.2 mg iron and 109 µg folate each. Across the whole preparation, protein supplies 20 per cent of energy, which matches the “high in protein” threshold in Annex 13. It is comfortably the best of the three recipes on that measure, and it comes entirely from the pulse.
Buckwheat galettes
250 g buckwheat flour, 550 ml water, 5 g salt.
Add the water gradually while whisking until the batter is fluid. Rest an hour in the fridge, then let it down again if it has thickened. Cook on a very hot, barely oiled pan.
For 8 galettes: 109 kcal, 3.6 g protein, 1.3 g fibre and 49 mg magnesium each. Protein accounts for 13.2 per cent of energy. A plain galette is light and nutritionally neutral: the filling makes the meal, which is good news rather than a shortcoming.
Quinoa and chocolate muffins
100 g quinoa flour, 100 g rice flour, 50 g cornflour, 40 g sugar, 200 ml oat drink, 35 ml rapeseed oil, 60 g baking chocolate, 1 sachet baking powder, 1 teaspoon cider vinegar, a pinch of salt.
Mix the dry ingredients. Combine the plant drink, vinegar and oil, then wait two minutes. Bring together without working the batter. Fold in the chopped chocolate. Bake for 20 minutes at 180 °C.
For 10 muffins: 177 kcal, 2.6 g protein, 7.6 g sugars and 1.7 g fibre each. Protein supplies just 5.8 per cent of energy. A quinoa muffin is still a cake, and swapping wheat flour for quinoa flour does not change that. It tastes good; it is not a protein food.
Storage, buying and flavour
Quinoa flour goes rancid faster than the others, and the reason is quantifiable: 6.07 g of fat per 100 g against 0.82 g for white pastry flour. It is those fats that oxidise. Airtight jar, away from light, in the fridge if the bag has to last more than two months. Chickpea flour, at 6.69 g of fat, deserves the same treatment. Buckwheat, at 2.19 g, tolerates the cupboard.
On flavour, each has real character. Chickpea is slightly sweet and frankly beany when raw, far mellower once cooked. Buckwheat is rustic with a bitterness that comes from its tannins; toasted buckwheat, or kasha, adds roasted notes. Quinoa is earthy and a little bitter, that bitterness coming from saponins in the seed coat, removed by washing or abrasion before milling. Dry-toasting quinoa flour for five minutes softens the note noticeably.
In Switzerland all three are available in supermarkets and health food shops. Chickpea is the easiest to find, often sold as besan or gram flour in Indian grocers, where the price per kilo is markedly lower. Local supply chains exist for chickpeas too, a subject covered in our article on local chickpea varieties for the Swiss climate, alongside our piece on ancient Swiss grains.
Grind size is worth a second look, because it is rarely declared and changes the result more than the variety does. Finely milled chickpea flour from an Indian grocer gives smoother pancakes than a coarsely ground organic product, which leaves a faintly gritty texture. With buckwheat, what also matters is whether the hull went into the mill: dark buckwheat flour containing hull is stronger and more bitter, pale flour from hulled groats is milder and easier to get right first time. If the packet says nothing, the colour tells you.
There is a practical question about quantities too. These flours rarely get used up in kilo bags, and a half-empty bag of quinoa flour sitting in the cupboard for a year will turn rancid before it runs out. Buying small and using deliberately works better: chickpea flour for savoury pancakes and as a protein lift in cakes, buckwheat for galettes and rustic bases, quinoa flour in blends at no more than a third. If you only want to keep one of the three, make it chickpea, which delivers the most both nutritionally and in the pan. Our recipe for homemade fresh pasta with chickpea flour shows what it does in a dough.
What the environmental arguments are worth
Chickpea is a legume: it fixes atmospheric nitrogen and reduces the nitrogen fertiliser needed by the following crop. That is a solid agronomic argument, independent of any nutritional claim.
Buckwheat is described everywhere as a hardy, sustainable crop. Agronomically that is true, and yet its presence in Switzerland is residual. According to an article in Agrarforschung Schweiz published in 2019, the Swiss buckwheat area fell below one hectare in 1972 and the crop no longer appears separately in agricultural statistics; Agroscope and HAFL trials run from 2015 to 2018 on fourteen varieties gave yields ranging from under 10 to over 20 dt/ha, averaging 14.7 dt/ha (Agrarforschung Schweiz). For comparison, Swiss bread wheat routinely exceeds 55 dt/ha. Buckwheat is interesting in a rotation and for pollinators, not as a yield crop.
As for quinoa, the controversy about Andean communities needs revisiting. The familiar accusation is that Western demand priced quinoa out of reach for Peruvians. Work published in World Development in 2018 tested that claim against eleven years of nationally representative household surveys: quinoa accounts for less than 1 per cent of the budget of the Peruvian households that consume it, roughly 30 per cent of households; and rising international prices are associated with a small but statistically significant increase in household welfare in districts and provinces where quinoa consumption or production is more concentrated (Bellemare, Fajardo-Gonzalez and Gitter, 2018). The authors conclude that international quinoa trade has not harmed household welfare in Peru.
The simple story was more morally satisfying. It does not survive the data. That does not make South American quinoa equivalent to European-grown quinoa in transport terms, but it is no longer the same argument.
Five recurring mistakes
Replacing all the wheat in a risen recipe. This is the number one cause of failure. With no gluten and no substitute binder, a bread or brioche dough cannot hold gas.
Pouring water onto chickpea flour all at once. The lumps that form will not whisk out. Mixing dry first, then adding liquid gradually, solves it.
Adding flour to thicken a runny batter. With quinoa and rice, the batter always looks too loose and turns dry in the oven if you correct it. Trust the resting time, not the immediate appearance.
Buying loose flour for someone with coeliac disease. The 84.91 per cent detectable contamination in loose naturally gluten-free flours is not an abstract statistic.
Counting on a bake for protein. A quinoa muffin delivers 2.6 g of protein. The serious sources are elsewhere, as our plant protein guide sets out, and as our piece on high-protein labels shows, packaging often promises more than the product contains.
Adapting to your situation
Coeliac disease: certified flours, sealed bags, dedicated utensils and toaster. Buckwheat and oats deserve particular care given the contamination data.
Known buckwheat allergy: strict avoidance, including in commercial gluten-free products, and systematic reading of ingredient lists.
Vegetarian looking to raise nutrient density: chickpea flour is the only one of the three that genuinely changes anything. Its 4.86 mg of iron per 100 g counts, but this is non-haem iron, whose absorption depends heavily on the rest of the meal; our article on plant-based iron absorption sets out the levers that actually work.
Prone to oxalate kidney stones: quinoa contains 396.9 to 715.2 mg of oxalate per 100 g depending on variety. Occasional use raises no particular question; daily use in quantity is worth a medical opinion.
Children: no issue with these flours once cooked. The one thing to watch is raw batter, which children are statistically more exposed to.
What this article does not say
Ciqual values are averages from measurements on French samples. Chickpea flour from another origin, quinoa of another variety, or a coarser grind will give different results, sometimes by 10 to 20 per cent. The orders of magnitude hold; the third decimal does not.
The substitution rates suggested here come from a mix of published rheology data and ordinary kitchen practice. They do not replace a trial in your oven with your flour.
No Swiss prices are given, because they vary too much between supermarkets, health food shops and specialist grocers for any range to be honest.
Finally, none of the above is medical advice. Coeliac disease has to be diagnosed by a doctor before any dietary change, because starting a gluten-free diet too early distorts the tests.
Frequently asked questions
Do these flours help with weight loss? No. They provide between 348 and 359 kcal per 100 g, which is exactly what wheat flour provides. The difference is in composition, not energy.
Can you make 100 per cent buckwheat bread? Yes, but it will be dense and closed-textured, closer to a thick flatbread. A binder such as psyllium improves it considerably, without ever producing something resembling a wheat sourdough.
Can chickpea flour be eaten raw? No, for two reasons: raw pulses contain antinutritional factors that cooking inactivates, and the bacterial contamination issue applies to all flours.
Is buckwheat a cereal? No, it is a polygonaceous plant, botanically closer to rhubarb than to wheat. The French name “black wheat” describes its use, not its family.
Why is my quinoa batter bitter? Probably saponin residue, or flour that has started to go rancid. Five minutes of dry toasting helps with the first; nothing rescues the second.
Should these flours be sifted? For chickpea, yes, it prevents lumps. For buckwheat and quinoa it is optional.
Are commercial gluten-free products nutritionally worse? Not systematically. An analysis of 913 products on the Greek market published in 2024 concluded that gluten-free products and their gluten-containing counterparts had broadly comparable profiles, while noting that a quarter of the gluten-free products had an unfavourable profile (Europe PMC). Read the label rather than the category.
How much fibre do these flours provide relative to requirements? The NHS recommends 30 g of fibre a day for adults, with average intake around 20 g (NHS). A hundred grams of chickpea flour provides 10.8 g, which is substantial, but you rarely eat 100 g of flour on its own.
Sources
- Ciqual 2020 table, ANSES: composition of chickpea, buckwheat, wheat and rice flours and raw quinoa.
- Swiss FDHA ordinance on food information, SR 817.022.16: article 41 and Annex 13.
- NHS: coeliac disease, food allergy, iron, B vitamins and folate, fibre.
- Frontiers in Nutrition, 2025: gluten contamination in 163 flours.
- Journal of Food Protection, 2026: Canadian survey of 657 samples.
- Journal of Food Protection, 2019: STEC in 70 Swiss retail flours.
- Asian Pacific Journal of Allergy and Immunology, 2025 and Current Allergy and Asthma Reports, 2024: buckwheat allergy.
- Plant Foods for Human Nutrition, 2025: DIAAS values for flours.
- Food Chemistry: X, 2024: quinoa oxalates and digestibility.
- Food Science and Technology International, 2021: rheology of wheat-chickpea doughs.
- International Journal of Food Sciences and Nutrition, 2024: nutritional profiles of gluten-free products.
- Bellemare, Fajardo-Gonzalez and Gitter, World Development, 2018: welfare impacts of quinoa prices in Peru.
- Agrarforschung Schweiz 10(5), 2019: buckwheat in Switzerland, Agroscope and HAFL trials.