Peanut butter production is four operations: sort the kernels, roast them, grind them to a paste, then pack. Only one of those steps cannot be undone, and it is the grinding, because a mill mixes every kernel you feed it and spreads whatever a single bad one was carrying through the entire batch. Ugandan groundnuts are grown mainly in the east and the north and eaten countrywide as roasted seed, sauce, flour and butter.
What Peanut Butter Production Involves on a Small Scale
Commercial peanut butter is ground dry roasted groundnut kernels, usually with salt, sometimes with a sweetener, and sometimes with a stabiliser that stops the oil rising. A Ugandan producer working at village or small business scale needs a way to shell, a way to sort, a roaster, a mill and a filling and capping arrangement. Nothing on that list is exotic. What separates a product a shop will stock from one it will not is the sorting and the packaging, neither of which is the expensive part.
Indian food processing training material for small enterprises describes two routes through the grinding. The older single stage method grinds roasted kernels once to a coarse paste. The two stage method grinds to a medium texture, then reduces again to a fine smooth one, and it warns that putting the paste through repeatedly heats it to excessive temperatures. That is the whole technical argument between a village mill and a proper butter grinder: the second machine gets to smooth without cooking the paste on the way.
What a Sack of Unshelled Groundnuts Yields in Kernels
Start from the pod, because every cost and every price you quote later rests on this conversion and a surprising number of small producers have never measured it.
Then the composition of the kernel, which explains most of what the product does later. Groundnut kernels run about 45 to 50 percent oil and 26 to 34 percent protein, and a pooled set of analyses puts the oil at a mean of about 48 percent of dry matter across a range of roughly 44 to 53. Nearly half the weight of your raw material is oil, which is why the paste separates, why it goes rancid rather than mouldy, and why it carries the energy that makes it worth eating. Oleic, linoleic and palmitic acids together make up more than 90 percent of that oil, with oleic acid around half of it.
Sorting Groundnut Seed Before Any Other Step
This is the step that decides whether the product is safe, and it happens before the roaster is lit. Groundnuts are among the crops most prone to contamination by Aspergillus flavus and Aspergillus parasiticus, the moulds that produce aflatoxins, and the International Agency for Research on Cancer classifies naturally occurring mixtures of aflatoxins as a Group 1 human carcinogen, meaning known to cause cancer in humans. The ministry's own extension material states that aflatoxins can cause stunting in young children and cancer, and that they are heat stable so remain poisonous after cooking or boiling.
A review of Ugandan aflatoxin research reports that visual sorting, winnowing, washing, crushing and dehulling together contribute a 40 to 80 percent reduction in aflatoxin levels in grains, and that sorting is particularly recommended for peanuts. That is the largest single improvement available to a small producer and it costs labour rather than money.
That last point is where sorting quietly fails in practice. Sorting only works if the rejected fraction leaves the food chain. A pile of mouldy kernels set aside and then ground on a slow day has achieved nothing except concentrating the problem into one jar.
Why Grinding Is the One Step Production Cannot Undo
Aflatoxin in a sack of groundnuts is not spread evenly. It sits in a small number of kernels at high concentrations, which is why a sample can test clean while the sack is not, and it is also why grinding is dangerous. Grinding takes a distribution with a few very hot points and turns it into a uniform average across the whole batch. Nothing downstream reverses that.
Ugandan measurement bears this out about as directly as evidence can. A study screening 120 peanut samples from Kampala, Mubende, Gulu, Pader, Mbarara, Masindi and Kaberamaido found milled groundnuts carrying a mean total aflatoxin of about 1,278 micrograms per kilogramme against about 85 for whole groundnut seeds from the same survey, roughly fifteen times as much. The study implicates the milling of fungus damaged peanuts by traders, which hides the visible evidence of spoilage from the buyer. Overall, 72 percent of samples were aflatoxin positive, and 52 percent exceeded the threshold the study measured against.
Read as a production instruction, that finding is simple. Grinding is not a processing step you can do carelessly and fix later. It is the point at which a sorting decision becomes permanent.
Aflatoxin Limits for Groundnuts in Uganda, and Whose Limit Each One Is
Several different aflatoxin numbers circulate for groundnuts, and arguments about them are usually arguments about whose rule is being quoted. There are at least four, they sit at four different levels, and which one applies depends entirely on where the jar is going.
| Who sets it | What it covers | Total aflatoxins |
|---|---|---|
| Uganda | All foods | 10 per kg |
| East African Community | Cereals, pulses | 10 per kg |
| European Union | Groundnut, cereal | 4 per kg |
| United States | Food, dairy feed | 20 per kg |
All four figures are micrograms per kilogramme, the unit often written as parts per billion. Alongside its total, the East African Community harmonised limit sets aflatoxin B1 on its own at 5 micrograms per kilogramme, B1 being the most carcinogenic of the four and usually the most abundant. Uganda's own national figure is a total across B1, B2, G1 and G2 applied to all foods, recorded in a peer reviewed continental review of African mycotoxin regulation which cites the African Union's aflatoxin control situation analysis for Uganda.
One consequence of that table matters for anyone reading Ugandan research. The published Ugandan peanut surveys mostly report compliance against the 20 microgram figure, which is a United States action level and twice Uganda's own national limit. So when a study says half its samples were above the limit, the share failing Ugandan law is larger than the share it reported, and the share that would fail a European buyer is larger again. The gap is not academic: of the shipments of African agricultural products rejected by the European Union for mycotoxin contamination over a fifteen year period, groundnut and groundnut based products were 88 percent of those affected.
Drying and Storage of Groundnut Kernels Before Processing
The controlling variable for groundnuts is not the same as the one used for maize, and carrying the grain figure across is a mistake worth naming. Maize is graded on a moisture percentage, and the ceilings published for it sit at 13.0 and 13.5 percent depending on the edition of the standard, which the maize storage guide covers. Groundnuts are a different crop under a different specification.
What the aflatoxin literature gives for groundnuts is a water threshold rather than a grade figure. Aspergillus flavus and Aspergillus parasiticus are drought tolerant moulds that can still grow and produce aflatoxin at a relative moisture of around 85 percent, and keeping storage below that level, or below 10 degrees Celsius, prevents aflatoxin production during storage. Post harvest work names two controlling requirements and only two: the moisture of the seed and the level of insect infestation.
Practically, that means dry the pods hard and fast, on a tarpaulin or a raised rack rather than on bare soil, keep them off the ground once dry, and keep insects out. Contamination can start in the field, continue after the crop is lifted but before it is gathered, then carry on through transport and storage, so there is no single point at which the risk is closed. Drought stress and high temperature both raise aflatoxin production in the field, with temperature the stronger of the two, which means a bad season arrives at your store as a worse sack. On the insect side, the containers and hygiene that keep beetles out of grain work here too, and are covered on the pages about hermetic grain storage bags and preventing weevils in stored grain.
Roasting Groundnuts After Harvest and What Heat Does Not Fix
Roasting does two things. It develops the flavour that makes the product recognisable, through browning reactions in the kernel, and it drives the water out. Small enterprise processing guidance gives roasting at around 160 degrees Celsius for around 45 minutes, varying with the moisture the kernels went in at, and takes the water content down to about 1 percent. The kernels then cool in air before grinding, and blanching, the removal of the papery skin, usually happens between the two.
What roasting does not do is make contaminated groundnuts safe. Aflatoxins are heat stable, which the ministry states in plain terms, and they remain poisonous after cooking or boiling. A roaster is not a decontamination step and no temperature available to a food processor turns a bad batch into a good one. This is the reason the sorting table above sits before the roasting section rather than after it.
Two things follow for the operator. The roast is where you can still smell a problem, because a mouldy note survives into the roasted kernel and is easier to detect than in the raw nut, so a sniff at the cooling tray is a free final check. And the roast is your control over consistency: the same time and the same temperature on kernels sorted to the same size give a repeatable product, which is what a shop buyer is actually asking for when they ask about quality.
Milling to a Paste and Why the Oil Separates in Storage
Grinding shears the kernel apart and releases the oil that was held inside the cells, so the paste that comes out is solid particles suspended in liquid oil. Suspensions of that kind separate, and a layer of oil on top of a jar of natural peanut butter is physics rather than spoilage.
Commercial producers stop it by adding a stabiliser at the grinding stage, along with salt and sometimes a sweetener, which holds the oil in place and gives the smooth spreadable texture consumers expect from a factory product. A producer who adds nothing has a natural product that will separate and needs stirring, and both are legitimate as long as the label matches the jar. If you do add anything, every addition has to appear on the label, which is where a small producer most often falls foul of a formal buyer.
Texture is the other decision. A single grind gives a coarse paste and the two stage route gives a fine one, with the warning already noted that pushing paste through a mill repeatedly overheats it. Overheated paste tastes of the machine rather than the nut. Crunchy products are made by holding back a portion of coarsely ground kernel and folding it in, rather than by grinding badly.
Packaging, Labelling and Shelf Life for the Market
Peanut butter fails by oxidation, not by mould, because the water was driven out in the roast and there is very little left for a microbe to use. What there is plenty of is oil, and oil goes rancid on contact with air, light and warmth. So the packaging job is to exclude air.
Small enterprise guidance names vacuum packing as the main route to reducing oxidation in a commercial line. At a smaller scale, the practical equivalents are filling jars full so the headspace is small, capping immediately while the paste is still warm enough to flow and settle, sealing tightly, and storing finished stock out of sunlight and away from heat. A jar half full of air on a shop shelf in the sun is a short shelf life whatever went into it.
On the label, a buyer looks for the product name, the ingredients in order, the net weight and a date. Getting a formal retailer to stock the product will also involve national standards registration, and requirements there change, so confirm the current position with the district commercial office rather than working from anything you read online. General packaging decisions for produce are covered on how to package produce for sale, and the agricultural product packaging guide goes wider.
What Ugandan Surveys Found in Market Peanut Butter and Paste
It is worth knowing what the existing market looks like, because it sets the bar a careful producer can clear. The picture is not reassuring and it is also the opportunity.
One row of that farm level table does not add up: a fourth village is printed with 75 percent positive and 15 percent negative, which sums to 90, so it is left out above rather than reconciled. The three villages quoted are the rows whose positive and negative shares total 100.
Two more things are worth knowing and neither is available to buy yet. One Ugandan groundnut variety in the Serenut series has been cited as more resistant to fungal growth and aflatoxin production, while the international breeding literature says progress on aflatoxin resistance has been limited, so treat variety as a small help rather than a solution. And Ugandan laboratory work has isolated non aflatoxin producing moulds from household foods that cut aflatoxin B1 by more than 95 percent when grown alongside the toxin producing strain, with the researchers stating that field evaluation is the next step. That is promising and it is not a product a processor can order.
Questions About Peanut Butter Production in Uganda
Does roasting get rid of aflatoxin? No. Aflatoxins are heat stable and remain poisonous after cooking or boiling, which the ministry's own extension material states directly. Roasting develops flavour and drops the water content to around 1 percent, both of which matter for the product, and neither of which cleans up contaminated kernels. Sorting before the roast is the control.
How do I know whether my groundnuts are safe to grind? Honestly, you cannot know by looking, and that is the uncomfortable answer. Contamination sits in a few kernels at high levels, so a clean looking sample proves little. What you can do is stack the odds: sort out everything discoloured, shrivelled, cracked or insect damaged, float off the buoyant ones, smell the roasted kernels before grinding, and reject any batch that came to you damp or smelling musty. Testing needs a laboratory, and your district agricultural office can say which testing services are currently reachable from your area.
What should I do with the nuts I sort out? Keep them out of food, including animal feed for milking or laying stock. The Ugandan review warns in as many words against the practice of setting mouldy peanuts aside and then feeding them to children or grinding them into paste. Destroy or dispose of them away from the store, and do not let them sit in the processing room where they can end up in a batch on a busy day.
Why does my peanut butter separate, and have I done something wrong? Nothing wrong. Kernels are close to half oil, grinding releases it, and a suspension of solids in oil separates on standing. A commercial product holds together because a stabiliser was added at the mill. If you want the natural product, say so on the label and tell buyers to stir it. If you want the factory texture, the addition has to be declared in the ingredients.
How long will my peanut butter keep? The limit is rancidity in the oil rather than microbial spoilage, because the roast leaves very little water behind. Air, light and heat all shorten it, so a full jar, a tight seal and a cool dark shelf are worth more than anything you can add. No shelf life figure is given here because it depends on your fill, your seal and your storage conditions, and the way to establish it is to keep sealed samples of your own batches and taste them at intervals until you find the point where they turn.
Can I use the same mill that grinds maize flour? It is common and it is a real risk. A shared mill carries residue from whatever went through before, so a machine that has just ground a mouldy cereal batch will pass some of it into your paste, and an oily paste is harder to clean out than flour. If the mill is shared, run it clean first, go first in the day where you can, and be aware that the hygiene of the plate is now part of your product.
What does the equipment cost to set up? No shilling figure is published here, because the only prices available are individual dealer quotations that move with the exchange rate, and a roaster and a butter grinder vary by more than an order of magnitude across capacities. The cost structure is more useful than a total: your biggest recurring line is the kernel, and the conversion above tells you that a 100 kg sack of pods gives around 70 kg of kernel before quality rejection, so the raw material cost per kilogramme of butter is well above the cost per kilogramme of pods. Track current groundnut prices against your own measured yield, then get three written quotations for a roaster, a two stage grinder and a filler on one written specification.
Finding Processing Equipment and Packaging Suppliers
Write your specification before you talk to anyone: throughput in kilogrammes per hour, whether the grinder is single or two stage, whether the roaster is a rotating drum with its own heater, and what jars and caps you intend to fill. Ask three suppliers to quote against that same list, because a quotation for an unnamed machine cannot be compared with anything. Spare parts matter more than the headline price, since bearing failure is the ordinary way these machines stop, so ask what is stocked locally before you buy. Your district commercial office can point you at equipment dealers and at the current standards registration route, and the value addition guides cover the neighbouring processing businesses, including sunflower oil processing and soybean processing, which share most of the same equipment problems.
