Studies of uses for the husk, pulp and shell of cacao. Nearly all are laboratory work or small feeding trials, and each summary says so, because the commonest error in this subject is to describe an experiment as an industry.
16 studies1984 to 2026Each checked against its published abstract
What these studies add up to
Pectin can be extracted from pod husk, at 7.5 to 19.5% of its dry weight in three laboratory studies. It is a weakly gelling kind, and no commercial production was found.
Husk and shell as feed run into theobromine. Rats were harmed by husk until a fungus had removed it, and broilers did worse above 1% shell meal.
Husk biochar carried polycyclic aromatic hydrocarbons, in amounts that varied 25-fold with how it was made.
Pulp is mostly sugar and varies widely by hybrid and ripeness. It has been brewed into beer at pilot scale and used to sweeten a laboratory chocolate.
Shell is the cautious case: Salmonella is hard to kill in it, and the one study of shell infusion tested its own samples only.
Two processes reached pilot scale: brewing with pulp and extracting compounds from shell. Nothing studied was an operating business.
Broilers did worse once cocoa shell meal passed 1% of the diet
How toxic are cocoa shell meal and its theobromine to young broilers?
How
Research feeding trial: two experiments with broilers fed 1, 2, 4 and 6% dietary cocoa shell meal, which contained 1.3% theobromine by analysis, or pure theobromine at matching levels. Bird numbers are not in the abstract.
Finding
Performance tended to be depressed with dietary cocoa shell meal levels above 1%. Pure theobromine was a little more toxic than the same amount supplied by the meal.
Does not establish: It does not set a safe level for other poultry classes or other species.
Limits: A 1984 trial with young chicks; no growth figures in the abstract.
How much treated cocoa pod husk can go into broiler finisher diets without harming performance?
How
Research feeding trial: 35-day trial with 3-week-old finishing broilers on diets containing 100, 200 or 300 g per kg of husk fermented with Pleurotus ostreatus or supplemented with fibre-degrading enzymes, plus a control; seven treatments replicated three times. Bird numbers are not in the abstract.
Finding
The authors concluded that fermented dried husk up to 200 g per kg and enzyme-supplemented husk up to 100 g per kg could be included without deleterious effects.
Does not establish: It does not show that husk feeding is practised commercially or that these inclusion rates respect legal theobromine limits.
Limits: Small trial; the husk had been fermented or enzyme-treated, so the result does not apply to raw husk. Theobromine is not mentioned in the abstract.
What is cacao pod husk made of, and what are its water-soluble pectins like?
How
Laboratory scale. Milled and dried pod husks extracted with hot water at 50 and 100 °C for 90 min at 1:25 w/v, with chemical, spectroscopic and rheological analysis of the fractions. Source and number of husk batches not given in the abstract.
Finding
The two fractions yielded 7.5 and 12.6% pectins. Both were low-methoxyl homogalacturonans (degree of esterification 37.0 and 42.3%) with high acetyl contents (29.0 and 19.2%). A 5% solution of the 50 °C fraction showed weak gel behaviour.
Does not establish: It does not show that husk pectin is produced or sold commercially; the authors say further investigations are required.
Limits: Bench extraction from one source of husk. The best fraction showed only weak gel behaviour.
Which conditions maximise pectin yield when cacao pod husk is extracted with citric acid?
How
Laboratory scale. Fractional factorial screening followed by response surface methodology with a central composite design, varying extraction conditions.
Finding
Extraction with aqueous citric acid at pH 3.0 for 95 min at 95 °C gave an experimental yield of 10.1 ± 0.3 g/100 g dry cacao pod husks, against a predicted yield of approximately 9.0 g/100 g. The pectin contained 65.1 ± 0.8 g uronic acid/100 g fraction and formed gels at acidic pH and high sucrose content.
Does not establish: It gives no costs, no food-grade specification and no evidence of production beyond the laboratory.
Limits: Optimisation on the bench with one husk source. Degree of esterification was 40.3%, so this is a low-methoxyl pectin.
How heat resistant is Salmonella in dry cocoa bean shell and hazelnut shell?
How
Laboratory scale. Heat resistance of outbreak and non-outbreak Salmonella strains measured in crushed cocoa bean and hazelnut shells at low moisture contents (≤4% w/w) and at about 7% w/w.
Finding
For the two most heat-resistant strains, S. Oranienburg and S. Enteritidis PT30, decimal reduction times at the higher test temperature were about 2.5 min in crushed cocoa bean shells and 7–11 min in crushed hazelnut shells, at about 4% moisture. Adding moisture to about 7% markedly reduced them, to 2–4.5 min at the lower test temperature.
Does not establish: It does not report how often cocoa shell carries Salmonella.
Limits: Inoculated laboratory material, not naturally contaminated shell.
Can sun-dried cocoa pod husk replace maize in the diet of growing rabbits?
How
Research feeding trial: twenty seven mixed-breed growing rabbits aged 10-12 weeks (1.2-1.3 kg) in a 6-week trial, with husk replacing maize at 0, 50 and 100 %; 3 replicates of 3 animals per diet.
Finding
Digestibility of dry matter and metabolisable energy decreased as husk inclusion increased. The authors report a cost advantage from including 200 g of sun-dried husk per kg of diet in place of maize, and conclude that husk can totally replace maize as a cheap energy source.
Does not establish: It does not address theobromine. Rabbits are among the species for which EU law sets the lowest theobromine limit in feed.
Limits: Nine animals per diet for six weeks. The abstract's statement on growth differences is garbled in the published text, so the growth result cannot be read from it.
Pilot scale: a bioreactor with a 100-liter capacity and a usable volume of 60 liters, at 22°C for 96 hours, with 30% cocoa pulp added to the wort at 12°P; two yeasts compared, and sensory acceptance tests.
Finding
Beer was produced with the adjunct, and the beer made with the yeast SC52 showed better fermentation performance and better acceptance in sensory analysis than the commercial ale yeast.
Does not establish: It does not show a commercial product or that the pulp supply chain exists to support one.
Limits: One pilot batch size at a Brazilian university; no figures in the abstract.
Pod husk compost improved soil and seedlings and reduced black pod on plantlets
Doungous and colleagues, 2018. Environmental science and pollution research international. doi:10.1007/s11356-018-2591-0
Question
Can compost made from cocoa pod husk improve soil fertility and suppress black pod disease?
How
Laboratory and plantlet scale: in vitro antagonism tests of compost water extracts against Phytophthora megakarya, and cocoa plantlets grown on soil amended with compost at several rates. No field trial is described.
Finding
Compost water extracts inhibited mycelial growth, with inhibition reaching 100%. Plantlets on amended soil had significantly lower disease scores. Compost raised soil pH and most essential elements and lowered aluminium. A dose of 20% (v/v) significantly increased stem length and leaf number.
Does not establish: It does not show a yield benefit, and it does not address whether returning husks to the farm carries Phytophthora with them.
Limits: Plantlets and laboratory tests, not bearing trees. The abstract does not say whether the husks used came from diseased pods.
Can green extraction of valuable compounds from cocoa shell be scaled up from the laboratory?
How
Laboratory comparison of ultrasound, hydrodynamic cavitation and conventional extraction of cocoa shells, then scale-up to a pilot flow reactor using a water, ethanol and hexane mixture.
Finding
At pilot scale the process afforded 20.5 w/w % hydrophilic and 15.8 w/w % lipophilic fractions. Extracts contained theobromine at 32.7 ± 0.12 mg per g of shells and caffeine at 1.76 ± 0.08 mg per g. The abstract notes that removed husk fibre is rerouted into animal feedstock components and crop mulching.
Does not establish: It does not show an operating industrial process; the authors say the work should facilitate industrial design.
Limits: A pilot run by one group; the solvent mixture includes hexane, and no costs are given.
Is cocoa pod husk safe and nutritious in feed once a fungus has removed its theobromine?
How
Laboratory animal study in Sprague-Dawley rats: feeds in which 30% or 50% of the maize was replaced by husk treated with Talaromyces verruculosus, or by husk treated with inactivated fungus, with control groups with and without theobromine. Group sizes are not in the abstract.
Finding
Rats ingesting theobromine in the inactivated-fungus husk diet or by gavage showed marked deleterious effects on body weight, thymus weight and tissue histology. Treated husk increased body weight. Substitution at 30% caused no mortality or organ-specific toxicity, unlike substitution at 50%.
Does not establish: It does not show that detheobrominated husk feed exists as a product or works in farm animals.
Limits: A rat study, not a trial in livestock. The treatment is a research procedure.
Can pectin, xylooligosaccharides and bioethanol be obtained from cocoa pod husk in one process?
How
Laboratory scale. Citric acid-assisted hydrothermal pretreatment of cocoa pod husk, enzymatic hydrolysis of the solid fraction, and fermentation of the hydrolysates by Candida tropicalis and Saccharomyces cerevisiae.
Finding
At 120° C, 10 min and 2% citric acid, pectin recovery was 19.5% with 41.9% uronic acids. Enzymatic hydrolysis of the solid fraction gave glucan conversion of 60%. Fermentation reached 30.9 g and 45.2 g of bioethanol per kg of husk.
Does not establish: It is not a techno-economic assessment and does not show that husk ethanol is viable; the ethanol yield per kilogram is small.
Limits: Bench results under optimised conditions; no energy balance or costs are reported in the abstract.
Can cocoa bean shell replace part of the concentrate in dairy goat diets?
How
Research feeding trial: twenty-two Camosciata delle Alpi goats in two groups of 11; in the test group 200 g of the 1.20 kg per day concentrate was replaced by the same amount of pelleted cocoa bean shell.
Finding
Total dry matter intake was reduced by the shell. Milk yield and milk fat, protein and casein were unaffected, and milk urea fell. The authors add that use may be hindered by theobromine and that nutritionists must avoid exceeding the legal theobromine limits.
Does not establish: It does not show shell is safe at higher inclusion or for other species.
Limits: 11 goats per group; the abstract does not state the trial's length or the theobromine content of the diet.
How much does the composition of cocoa pulp differ between hybrids, origins and stages of ripeness?
How
Laboratory analysis of pulp from several cocoa hybrids from Espírito Santo and Bahia states, Brazil, at different degrees of ripeness. The number of hybrids is not in the abstract.
Finding
Pulp was on average 76 % sugar, with a wide variation (20 %) between hybrids and a prevalence of reducing sugars. Ripe pods had more fructose and glucose, while unripe pods held mainly sucrose. Enzyme activities also varied between hybrids.
Does not establish: It does not give a single composition that holds for cocoa pulp in general.
Limits: The abstract does not say on what basis the 76 % is expressed, and it gives no figures for individual hybrids.
What passes from cocoa shell into an infusion, and is the drink safe?
How
Laboratory scale. Shell from toasted Colombian Criollo cocoa beans, ground to three particle sizes, infused and analysed for bioactive compounds, volatiles, sensory attributes, biogenic amines, ochratoxin A and microbial load. One source of shell.
Finding
Smaller particle sizes gave greater bitterness, acidity and astringency, linked to more epicatechin, melanoidins and proanthocyanidins. Analysis of biogenic amines, ochratoxin A and microbial load showed the infusion is not a risk to public health, in the authors' words.
Does not establish: A clean result on one batch does not show that shell infusions in general are free of ochratoxin A, which other studies find concentrated in shell.
Limits: One batch of shell from one origin. Heavy metals and theobromine intake are not mentioned in the abstract.
Can parts of the cocoa pod that are usually discarded replace refined sugar in chocolate, and with what environmental effect?
How
Laboratory formulation in which pod endocarp is turned into a gel using cocoa pulp juice concentrate in place of beet sugar, with a cradle-to-factory life cycle assessment of hypothetical large-scale production.
Finding
The chocolate formulation contains up to 20 wt% gel, with comparable sweet taste, higher fibre and reduced saturated fatty acid content. The life cycle assessment shows that large-scale production could reduce land use and global warming potential compared with average European dark chocolate.
Does not establish: It does not show the product is manufactured, that it is permitted to be sold as chocolate, or that farmers earn more from it.
Limits: The environmental result is modelled for a scale that does not exist. The abstract gives no figures for the reductions.
How do pyrolysis temperature and atmosphere affect the PAH content of cocoa pod husk biochar, and is the biochar acutely toxic?
How
Laboratory scale. Biochar produced across 300-900 °C under CO2 or N2, analysed for polycyclic aromatic hydrocarbons, and tested on zebrafish embryos from 0 to 4 dpf.
Finding
Biochar made under CO2 at 700 °C had the greatest PAH burden (2478 ± 228 ng per g), while treatment under N2 at 900 °C left 104 ± 22 ng per g. High-ring PAHs dominated. Zebrafish embryos showed no detectable acute effects under short-term exposure.
Does not establish: It does not show that husk biochar is safe in long-term agricultural use or that anyone produces it commercially.
Limits: Laboratory biochar and a short embryo assay. No soil or crop was tested.
We would like to set two Google Analytics cookies to count page views, and to count them in our own cookie-free measurement as well. They do not identify you and nothing is shared with advertisers. Decline and the site works exactly the same — see cookies for what each one does.