Storage trials on plain, filled and nut chocolate, and on beans and cocoa butter. Each measured one product under set conditions, which is exactly why they cannot be turned into a rule of months.
22 studies1997 to 2025Each checked against its published abstract
What these studies add up to
For plain chocolate, swings in temperature did the damage. Bars at a steady 20°C did not bloom in 90 days; bars cycled to 32°C did, and the speed of the swing changed how fast.
Appearance and flavour fail separately. Hot storage ruined how dark chocolate looked and felt while costing it the least aroma, and cooler, humid or frozen storage lost aroma without the visible damage.
In filled chocolate the filling's fat moves into the shell, and temperature governs the rate: measurable in days at 30°C, slight or absent at 18 to 20°C over weeks.
Where a filling holds water, yeasts spoil it. They were the commonest organisms in commercial pralines, and one grew at a water activity of 0.70, below the figure usually quoted as a limit.
Nuts add oxidation. Limiting oxygen in the pack extended the life of a hazelnut chocolate from about eight months to a year, the only controlled packaging comparison found.
Chocolate is not too dry for mould in every circumstance: dry-tolerant moulds were isolated from bars and grew where chocolate met moisture.
Moulds were isolated from dry chocolate bars, and grew where chocolate met a moist filling
Are xerophilic moulds present in chocolate, and can they grow there?
How
Isolation study using malt, yeast, 50% glucose agar as an enrichment medium, followed by a storage experiment in which chocolate containing Chrysosporium inops arthroconidia was coated on to agar of water activity 0.89 as a model filling. The number of bars examined is not given.
Finding
Three Chrysosporium species were isolated from commercial chocolate bars with a water activity of approximately 0.28. In the model filled chocolates, growth of C. inops occurred on the inside chocolate surface next to the filling, and on the outside surface after holding at 92% equilibrium relative humidity for 12 d. The white fungal growth was not unlike fat bloom to the naked eye but was clearly different under the electron microscope.
Does not establish: Isolation of spores from dry chocolate does not show that mould grows in plain chocolate under normal storage; growth was seen at the moist filling interface and at very high humidity.
Limits: A model system with an agar filling, one inoculated species, and an unstated number of commercial bars.
How does storage temperature affect texture, crystal structure, bloom and sensory quality of filled dark chocolate?
How
Filled dark chocolates stored at 18 and 30 °C for 8 weeks, with measurement of texture, polymorphic structure, triglyceride changes, bloom and sensory attributes. The abstract gives no sample size.
Finding
Storage at 18 °C for 8 weeks significantly retarded changes and the chocolates were free from bloom during the storage period. At 30 °C the rate of fat migration increased, texture declined, and the chocolates bloomed in the third week of storage.
Does not establish: It does not establish 18 °C as a general storage rule, and it does not cover nut or ganache fillings.
Limits: One filled product, whose filling included desiccated coconut, at two temperatures for 8 weeks.
Cocoa butter stored cool for two years oxidised slowly and stayed within limits
Hamid and Damit, 2004. Journal of the Science of Food and Agriculture. doi:10.1002/jsfa.1629
Question
How does the quality of cocoa butter change during long storage?
How
Pure prime pressed and deodorised cocoa butters from local grinders, about 5 kg of each packed in corrugated paper boxes lined with low-density polyethylene, stored at 15 +/- 4°C and 70+/- 4% RH for up to 24 months and sampled every 2 months.
Finding
Free fatty acid content, iodine value and peroxide value of both types increased with storage duration; saponification value and refractive index showed no significant changes. Free fatty acid content and peroxide value after 24 months were still below the limits specified in Malaysia.
Does not establish: It does not show how cocoa butter keeps at tropical ambient temperatures or in other packaging.
Limits: Two types of butter from Malaysian grinders at one cool storage condition. The abstract does not give the measured values or the limits.
How fast does oil migrate from a high-oil filling into chocolate, and how do chocolate formulation and temperature affect it?
How
Two-layer chocolate and peanut butter paste model confectionery followed by magnetic resonance imaging. Five chocolate formulations varying in particle size, milk fat content and emulsifier level were assessed at 20 and 30 °C over 17 d.
Finding
A Fickian diffusion model fitted the data at 30 °C. No significant mass transfer took place in the 20 °C samples over the experimental time frame.
Does not establish: It does not show that oil migration stops at 20 °C over a product's full shelf life.
Limits: A layered model system observed for 17 d, so the absence of transfer at 20 °C applies to that time frame only.
Do packaging oxygen permeability, modified atmosphere and oxygen absorbers change how well dark chocolate with hazelnuts keeps?
How
Dark chocolate with hazelnuts packed in two laminates (PET//LDPE, and PET coated with SiOx//LDPE) under vacuum, nitrogen or with an oxygen absorber, stored in the dark at 20 °C for 12 months. Commercial controls were packed in aluminium foil in air. Peroxide value, hexanal, colour, fatty acids, volatiles and sensory attributes were followed. Sample numbers are not given.
Finding
Peroxide value rose from 0.80 in fresh product to 6.51 meq of oxygen per kg of chocolate fat in commercially packed samples after 12 months, and hexanal from 0.53 to 7.56 mg/kg. Acceptable quality was retained for about 8 months in commercial packages, 8 to 9 months in PET//LDPE, 11 months in the SiOx-coated laminate, and at least 12 months with an oxygen absorber irrespective of material.
Does not establish: These are measured results for a nut-containing dark chocolate, not shelf-life figures for plain chocolate, and the study does not test paper, compostable films or light exposure.
Limits: One product from one producer, stored in the dark at one temperature. Shelf life was judged by the study's own sensory and chemical criteria.
What drives colour change and deterioration of white chocolate in storage: lipid oxidation, non-enzymatic browning, or both?
How
Three white chocolate samples (with a synthetic antioxidant, with casein peptides as a natural antioxidant, and with none) stored for 10 months at 20 and 28 °C, with measurement of acidity, thiobarbituric acid reactive substances, peroxide value, water activity and colour.
Finding
Acidity, thiobarbituric acid reactive substances and peroxide values increased with time. Samples at 20 °C often showed lower values than those at 28 °C, although the differences were not always significant. Water activity increased from 0.4 to 0.53-0.57 over the 10 months. Adding antioxidants did not significantly influence most parameters.
Does not establish: It does not measure the effect of light or of packaging material on white chocolate, and it does not set a shelf life.
Limits: Three laboratory formulations and two temperatures. The abstract does not describe the packaging or light exposure.
Can cocoa beans be stored safely from insects without fumigation, by sealing them airtight?
How
A field trial of 6.7 tonnes of beans in a sealed flexible enclosure in Makassar, Indonesia, and trials by Ghana's cocoa board comparing three sealed stacks with one unfumigated and one fumigated stack. Some authors are from the maker of the enclosure.
Finding
Oxygen fell to 0.3% after 5.5 days in the Indonesian trial and no insects survived. In Ghana all insects in the sealed stack were dead by the sixth week, and the beans kept their grade through nine weeks.
Does not establish: How sealed storage compares with fumigation over a full season, or what it does to flavour.
Limits: Short trials, and written in part by the product's manufacturer.
Hot storage wrecked how dark chocolate looked and felt, and cost it the least aroma
Nightingale and colleagues, 2012. Journal of Agricultural and Food Chemistry. doi:10.1021/jf204718z
Question
How do storage conditions that lead to fat bloom or sugar bloom change the flavour volatiles and sensory quality of dark chocolate?
How
Laboratory storage trial. Dark chocolate was held under different conditions leading to either fat or sugar bloom and analysed at 0, 4 and 8 weeks by volatile analysis and descriptive sensory evaluation. The abstract does not give the number of samples or the exact temperatures.
Finding
Chocolates stored at high temperature were the most visually and texturally compromised, but their volatile concentrations were affected the least. Samples stored at ambient, frozen and high relative humidity conditions had significant volatile loss. High-temperature storage caused a polymorphic shift to form VI and increased hardness. The condition with the least overall impact was storage at constant temperature and 75% relative humidity.
Does not establish: It does not establish a recommended storage temperature or a shelf life in months, and it says nothing about milk or white chocolate.
Limits: One dark chocolate over 8 weeks. The abstract does not state the temperatures used, and the higher volatile retention after hot storage was measured instrumentally while the sensory panel detected no difference in chocolate flavour.
How much does dark chocolate swell when fat or moisture migrates into it, and do the two behave alike?
How
Method development and laboratory measurement using a confocal chromatic displacement sensor to scan the height profile of dark chocolate samples before and after exposure to fat or moisture migration. Sample numbers are not given.
Finding
Moisture migration resulted in rapid swelling once a water activity of 0.8 was reached in the chocolate. Migrating fat induced a more pronounced swelling than the same amount of absorbed moisture, which the authors take as evidence of different mechanisms.
Does not establish: It does not relate swelling to cracking or consumer-visible defects in commercial pralines.
Limits: Laboratory confectionery systems measured with a new method; the proposed mechanisms are described as probable or possible.
What does storage under typical northern Australian climatic conditions do to a military ration chocolate?
How
Storage trial of one military ration chocolate under simulated tropical conditions, with sensory rating, instrumental colour, microscopy and differential scanning calorimetry. The abstract gives no sample size and does not state the temperatures.
Finding
Sensory quality decreased during storage and after seven days the chocolate was no longer of acceptable appearance. The decline was strongly correlated with the degree of blooming. Thermal analysis showed a shift from predominantly polymorph form V in fresh samples to a greater proportion of form VI in bloomed samples.
Does not establish: It does not show how quickly ordinary retail chocolate blooms in hot weather, and it does not show that the bloomed chocolate was unsafe.
Limits: A single product designed as a ration item, under one simulated climate. The storage temperatures are not given in the abstract.
Which micro-organisms are present in chocolate praline fillings, and how tolerant are they of low water activity, low pH, ethanol and preservatives?
How
Isolation on media containing 40-50% glucose from chocolate pralines (water activity 0.70-0.898) of nine manufacturers and from sugar syrups used by one of them. In total 677 isolates were identified by conventional and sequencing methods, and fifteen isolates were screened for growth under stress conditions.
Finding
Yeasts were the most frequently isolated group and were found in numbers up to 10(7) CFU/g; bacteria and moulds were found in low numbers. Z. rouxii grew at water activities down to 0.70, ethanol concentrations up to 6.0%, pH down to 2.0 and sorbic acid at the highest concentration tested. None of the bacteria were able to grow at the conditions tested.
Does not establish: It does not give a water activity below which every filling is safe from spoilage, and it concerns spoilage organisms, not pathogens.
Limits: Growth limits come from laboratory screening of fifteen isolates in media, not in pralines. The abstract does not say which country the manufacturers were in.
Are the factory environment and common filling ingredients sources of xerophilic fungi in chocolate confectionery?
How
Air sampling in Belgian chocolate confectionery factories on DG18 and MY50G media, and direct plating of four basic filling ingredients, with identification to species level. The number of factories is not given in the abstract.
Finding
The general fungal spore load was in the range of 50-250 colony forming units per cubic metre of air and the strictly xerophilic load was below 50. Penicillium species were the most prevalent in the factory environment. Among ingredients, nuts were the most likely potential source of direct contamination, with Eurotium the most prevalent genus.
Does not establish: It does not show how often fillings actually spoil, or that removing alcohol or preservatives leads to spoilage in a given recipe.
Limits: A survey of presence, not of spoilage. Factory numbers and sampling periods are not stated.
Can cocoa butter stearin or sorbitan monostearate delay fat bloom in dark chocolate, and under what storage conditions does bloom appear?
How
Three dark chocolate formulations (standard, with cocoa butter stearin, with sorbitan monostearate) stored as bars for 90 days at 20 °C and under oscillating temperature between 20 and 32 °C. Bloom was followed by whiteness index, melting behaviour and polymorphism. The number of bars is not given.
Finding
All samples stored at 20 °C did not develop fat bloom and kept the form V crystal structure. Under oscillating temperature, cocoa butter stearin (6.0%, w/w) and sorbitan monostearate (0.15%, w/w) delayed surface fat bloom by at least 45 and 15 days respectively compared with the standard chocolate.
Does not establish: It does not show that chocolate kept at 20 °C never blooms over longer periods, and it does not test milk or filled chocolate.
Limits: Laboratory formulations over 90 days. One constant temperature and one cycling regime were tested.
Do fermentation and moisture content change how badly stored cocoa beans are attacked by the tropical warehouse moth?
How
Fermented and unfermented beans at 7 or 9% moisture, stored for 6 months in ventilated jars with moth larvae introduced at the start.
Finding
Moth numbers and insect-damaged beans were lower on fermented than on unfermented beans, and lower at 7% moisture than at 9%. Weight loss was higher at the higher moisture.
Does not establish: A safe moisture content. It compares two levels.
Limits: Jars of one kilogram, not sacks in a warehouse. The index dates the paper 2019; the work is older.
What condition are cocoa beans in as they pass from farmer to trader to exporter in South Sulawesi?
How
Surveys in the dry season of 1997 and the wet season of 1998 in three regencies and the port city, with 9 samples from farmers, 21 from traders and 15 from exporters each season.
Finding
Moisture was above the Indonesian standard's limit of 7.5% at all three levels, and higher in the wet season. Insects were found on beans from traders and exporters, the red flour beetle predominating. Some trader and exporter samples exceeded the 4% limit for mouldy beans.
Does not establish: What condition beans leave South Sulawesi in today.
Limits: A small survey in one province in the late 1990s.
How does water activity affect the heat resistance of xerophilic moulds in a confectionery filling without preservatives?
How
Thermal death time curves for one strain of Aspergillus pseudoglaucus and two strains of Aspergillus fischeri in a low-acid (pH 6) confectionery model of evaporated milk and fructose adjusted to water activities of 0.70, 0.75 and 0.80. Experiments were performed in triplicate.
Finding
A. fischeri was the more heat tolerant: at 90°C its decimal reduction times at a water activity of 0.75 were 11.5 to 34.8 min. A. pseudoglaucus became more heat resistant as water activity fell. The authors note that confectionery of low water activity is typically considered resistant to moulds, but that several xerophilic fungi cause considerable losses through spoilage.
Does not establish: It does not give a water activity at which a ganache is safe from mould, or a heat treatment suitable for any real filling.
Limits: A model filling, three strains, and heat survival only. Growth during storage was not measured.
Can accelerated testing predict the oxidative stability of raw and roasted hazelnuts in normal storage?
How
Accelerated shelf-life testing at 55, 65 and 75 °C at a water activity of 0.43, with peroxide value, para-anisidine value and total oxidation value fitted to an Arrhenius model, validated against samples kept for 8 months at 20-30 °C. Sample numbers are not given.
Finding
The maximum activation energy and Q10 (1.871) were obtained for peroxide value in raw hazelnuts and the minimum (Q10 1.552) in roasted hazelnuts, which the authors read as a negative effect of roasting on oxidative stability. Predicted and actual storage values correlated with R2 = 0.91-0.98.
Does not establish: It does not give a time to rancidity for hazelnut pralines or gianduja, where the chocolate matrix and packaging change oxygen exposure.
Limits: Hazelnuts alone, not hazelnuts in chocolate or as a paste. Packaging and oxygen access are not described in the abstract.
How quickly do dried cocoa beans gain moisture in storage at different relative humidities?
How
Laboratory measurement of moisture adsorption kinetics of whole dried cocoa beans over a relative humidity range of 64–93% at 25 °C, fitted with a Fickian diffusion and adsorption isotherm model. The number of beans or lots is not given.
Finding
The model predicted moisture gain by individual beans accurately (R2 > 0.994). The authors propose an operational storage map to predict average bean moisture during storage at constant relative humidity (64% ≤ RH ≤ 93%) and temperatures from 5 to 35 °C.
Does not establish: The abstract does not state a safe moisture content or a relative humidity limit, and it does not measure mould growth or free fatty acids.
Limits: Simulated storage of individual beans in the laboratory; bagged or bulk cocoa in a warehouse or container was not measured.
How do the tropical warehouse moth and the red flour beetle develop on fermented stored cocoa beans?
How
Laboratory rearing of both insects on cocoa beans in Ghana.
Finding
The moth, Ephestia cautella, took four days as an egg, 31 as a larva and seven as a pupa. The beetle, Tribolium castaneum, took nine, 40 and five days on broken beans.
Does not establish: How much damage either does in a warehouse.
Limits: Laboratory conditions; the abstract does not give the temperature or humidity.
How do storage temperature and time change the texture, colour and sensory quality of dark and milk filled chocolate bars?
How
Storage experiment at 6, 12, 20 and 30 °C with assessment at 2, 6, 10, 18 and 26 weeks. The abstract does not give the number of bars or describe the filling.
Finding
Temperatures of 6 and 12 °C were the most suitable; those samples were evaluated as not significantly deteriorated by the storage period. Samples at 20 °C started to deteriorate after 10 weeks, mainly in meltdown rate and off-flavours, and this was more evident in dark chocolate bars. Holding bars at 24 °C for 24 h immediately after production improved resistance to fat bloom.
Does not establish: It is a result for these filled bars, not a universal shelf life for chocolate at 20 °C, and it does not address condensation when chilled chocolate is brought back to room temperature.
Limits: The abstract gives neither the filling composition nor the source of the bars.
Can white chocolate be made heat resistant by curing it at controlled humidity and temperature?
How
Laboratory study of white chocolate (30% whole milk powder, 44% sucrose and 26% cocoa butter) cured as solid samples at 83% RH at 29°C for 1 week, or at 50% RH at 18°C or 29°C for 12 weeks, with Karl Fischer moisture, penetration force and electron microscopy. Sample numbers are not given.
Finding
The aim was a white chocolate able to be picked up at 33 and 55°C, and the authors report that heat resistance was achieved. Moisture content increased from an average of 0.84% to 4.6% during curing, and penetration force increased, indicating development of internal structure. Curing at 50% RH for 12 weeks also gave a heat-resistant chocolate.
Does not establish: It does not show that the cured chocolate tastes or melts in the mouth like the original, or that the approach is used commercially.
Limits: One laboratory recipe. The abstract reports no sensory testing of the cured chocolate.
Does the rate at which temperature cycles, and not only its range, affect fat bloom in dark chocolate?
How
Laboratory experiment. Dark chocolate was cycled between 20 and 32 °C at fast, middle, slow and slow-long rates for 10 cycles, with measurement of fat bloom, polymorphism, texture and microstructure. The abstract gives no sample size.
Finding
The fast rate, especially for cooling, significantly retarded fat bloom formation by limiting fat migration. Polymorphic transition showed less correlation with early fat bloom. The slower rates gave a harder texture and could promote fat migration and accelerate bloom.
Does not establish: It does not show that rapid temperature swings are harmless to chocolate, and it does not give a time to visible bloom for any real transport route.
Limits: One dark chocolate, 10 cycles, one temperature range chosen to represent tropical conditions.
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.