Condensed Milk Shelf Life Science: Maillard Browning, Osmotic Preservation, and Can Stability

Executive Summary

Condensed milk — in both its sweetened (sucrose-added, ~45% sugar) and unsweetened (evaporated) forms — represents one of the most successful applications of food preservation engineering, achieving ambient-temperature stability measured in months to years through the synergistic application of water activity (a_w) control and thermal sterilization. Sweetened condensed milk (SCM) relies on osmotic preservation: the addition of sucrose at approximately 45% of total weight depresses a_w to 0.80-0.85, below the minimum growth threshold for all vegetative bacteria and most yeasts. Unsweetened condensed milk (evaporated milk) relies on commercial sterilization (retorting at 115-121°C for 15-20 minutes) to achieve commercial sterility, combined with hermetic can sealing to prevent recontamination. Despite their remarkable stability, both products undergo progressive quality deterioration through chemical pathways: non-enzymatic browning (Maillard reaction between reducing sugars and milk proteins), fat oxidation and separation, calcium mineral precipitation, and — in evaporated milk — age gelation and sedimentation. True microbial spoilage is rare and occurs only when preservation barriers fail: post-process contamination through compromised can seams, leakage, or — most critically — can swelling (hydrogen or CO₂ production from microbial metabolism or internal corrosion). This article provides a comprehensive examination of condensed milk stability from the dual perspectives of preservation engineering and degradation chemistry, suitable for food industry professionals, quality assurance specialists, and scientifically informed consumers.

Background

The invention of condensed milk in the mid-19th century — first by Nicolas Appert (thermal processing, 1810), then Gale Borden (vacuum evaporation with sugar addition, 1856) — represented a revolutionary advance in dairy preservation. Before refrigeration, fluid milk spoiled within hours in warm climates. Condensed milk enabled the distribution of dairy nutrition to armies, navies, tropical colonies, and urban populations without cold-chain infrastructure. Borden's key insight — that sugar could preserve milk as effectively as heat alone, without the "cooked" flavor of evaporated milk — created a product category that remains globally significant: worldwide production of condensed milk exceeds 4.5 million metric tons annually, with major markets in Southeast Asia, Latin America, the Middle East, and Africa.

The science underlying condensed milk stability integrates multiple disciplines: colligative chemistry (a_w depression by dissolved sucrose), thermal process engineering (F₀-value calculations for sterilization), non-enzymatic browning kinetics (Maillard reaction Arrhenius modeling), lipid chemistry (oxidation and physical separation), mineral chemistry (calcium phosphate precipitation equilibria), and packaging engineering (can integrity, internal lacquer coatings). Understanding these interacting systems is essential for shelf-life prediction, quality control, and spoilage investigation.

For related dairy preservation topics, see Sour Cream Shelf Life, Butter Shelf Life Science, and foundational principles in Water Activity and Food Stability.

Core Science: Two Preservation Mechanisms, Two Products

Sweetened Condensed Milk: Osmotic Preservation

The preservation of sweetened condensed milk (SCM) is achieved through a_w depression, not thermal sterilization. The manufacturing process involves: (1) standardization of milk to a specific fat-to-solids-not-fat ratio; (2) addition of sucrose (or other approved sweeteners) to achieve approximately 45% sugar in the finished product; (3) fore-warming (pre-heating) at 82-100°C for 10-30 minutes to destroy vegetative microorganisms, inactivate enzymes, and initiate controlled protein denaturation for viscosity development; (4) vacuum evaporation to concentrate total solids to approximately 70-74% (water content 26-30%); (5) cooling with seeding (addition of fine lactose crystals, typically ≤10 μm, to induce controlled crystallization and prevent large, sandy crystals during storage); and (6) aseptic packaging in cans or tubes.

The critical preservation parameter is the finished a_w. Sucrose at 45% w/w (approximately 62 g sucrose/100 g water) depresses a_w to 0.80-0.85 through Raoult's law colligative effects:

a_w = (moles water) / (moles water + moles solute × dissociation factor)

For sucrose (non-electrolyte, dissociation factor = 1) at 62 g/100 g water: moles water = 100/18 = 5.56; moles sucrose = 62/342 = 0.181. The ideal a_w = 5.56/(5.56 + 0.181) ≈ 0.968. The actual a_w (0.80-0.85) is substantially lower than this ideal prediction because: (a) the solution is highly non-ideal at these concentrations — sucrose-water interactions produce large negative deviations from Raoult's law; (b) milk salts and lactose contribute additional solute to the a_w depression; and (c) the lactose also depresses a_w independently.

At a_w 0.80-0.85, the microbiological ecology is severely constrained. No vegetative bacteria can grow below a_w 0.85 (except Staphylococcus aureus at 0.83 under ideal conditions, though the osmotic shock of direct inoculation into SCM typically precludes growth). Yeasts are generally inhibited below 0.88 (osmotolerant Zygosaccharomyces rouxii can grow to 0.62 but requires specific nutrient conditions rarely found in SCM). Molds — the most xerotolerant spoilage organisms — can grow at a_w as low as 0.61 for Xeromyces bisporus, but germination and growth at low a_w are extremely slow (weeks to months). In practice, below a_w 0.85, microbial spoilage of SCM is negligible under intact packaging.

Evaporated Milk: Thermal Sterilization

Unsweetened condensed milk (evaporated milk) contains no added sugar — it is simply milk concentrated to approximately 25-26% total solids (approximately 2.1:1 concentration ratio) by vacuum evaporation, then sterilized by retorting in hermetically sealed cans at 115-121°C for 15-20 minutes. This achieves a minimum F₀ value (sterilization equivalent at 121.1°C) of 5-8 minutes, sufficient to achieve commercial sterility — the destruction of all vegetative microorganisms and spores of public health significance (primarily Clostridium botulinum).

The absence of sugar preservation means evaporated milk has a high a_w (~0.97-0.99) permissive for microbial growth. Its preservation is therefore entirely dependent on: (1) the thermal sterilization step achieving commercial sterility; (2) hermetic can sealing preventing post-process recontamination; and (3) internal can lacquer integrity preventing metal corrosion that could compromise the seal or accelerate chemical degradation.

Core Science: Maillard Browning — The Dominant Quality Defect

Reaction Chemistry

Non-enzymatic browning via the Maillard reaction is the single most significant quality degradation pathway in condensed milk during ambient storage. The reaction proceeds between reducing sugars (lactose, at ~10-12% in SCM; ~9-10% in evaporated milk) and free amino groups on milk proteins (primarily lysine ε-amino groups on casein and whey proteins).

The reaction proceeds through three stages: (1) Early stage — reducing sugar carbonyl condenses with protein amino group to form a Schiff base, which rearranges to an Amadori compound (fructoselysine from lactose-lysine adduct). This stage is colorless and sensorially neutral. (2) Intermediate stage — Amadori compounds undergo dehydration, fission, and cyclization reactions producing furfurals, reductones, and α-dicarbonyl compounds (methylglyoxal, glyoxal, diacetyl). Some intermediate products are colored (yellow-brown) and possess characteristic caramelized, buttery, and malty aromas. (3) Advanced stage — polymerization of intermediate products forms high-molecular-weight brown nitrogenous polymers (melanoidins) producing dark brown color, bitter taste, and loss of nutritional quality through lysine blockage (unavailable for human metabolism).

Temperature Dependence

The Maillard reaction follows Arrhenius kinetics with an activation energy (E_a) of approximately 100-130 kJ/mol over the range 10-60°C — substantially higher than the E_a for most enzymatic reactions (~50-60 kJ/mol). This means Maillard browning is disproportionately accelerated by elevated storage temperatures:

Storage Temperature	Browning Rate Relative to 10°C	Time to Unacceptable Browning (SCM)
10°C	1× (baseline)	>24 months
20°C	4-6×	12-18 months
30°C	15-25×	6-9 months
40°C	50-100×	2-4 months
50°C	200-400×	2-4 weeks

Storage above 30°C — common in tropical warehouses and unrefrigerated shipping containers — can reduce the acceptable sensory shelf life of SCM by 50-70% compared to temperate storage. A product stored for 3 months at 35°C may exhibit browning and flavor deterioration equivalent to 12-18 months at 20°C. This is the primary reason tropical-market condensed milk often appears "older" at the same calendar age as temperate-market product.

Protein-Lactose Interaction in High-Solids Systems

The Maillard reaction in SCM is further accelerated by the product's high solids concentration. At 70-74% total solids, molecular crowding increases the effective concentration of reactants (lactose and proteins are forced into closer proximity) and reduces water activity (which favors condensation reactions). This concentration effect partially offsets the a_w-mediated slowing of the Maillard reaction that would otherwise occur at a_w 0.80-0.85 (the Maillard rate maximum in most food systems occurs at a_w 0.6-0.8; it decreases at both lower and higher a_w). In practice, SCM exhibits significant Maillard activity throughout its shelf life despite the low water activity.

Core Science: Physical Instabilities — Fat Separation and Sedimentation

Fat Separation (Creaming)

Both SCM and evaporated milk are oil-in-water emulsions, with milkfat globules (1-8 μm) dispersed in the aqueous serum phase. The density difference between liquid milkfat (~0.92 g/cm³ at processing temperature) and the concentrated serum phase (~1.15-1.25 g/cm³, depending on sugar and solids concentration) drives fat globule flotation (creaming). The creaming rate follows Stokes' law:

v = [2r²(ρ_s - ρ_f)g] / (9η)

Where v is the creaming velocity, r is the fat globule radius, ρ_s and ρ_f are the densities of serum and fat respectively, g is gravitational acceleration, and η is serum viscosity.

In SCM, the extremely high serum viscosity (η ~1-10 Pa·s, compared to ~1 mPa·s for water) dramatically slows creaming — but does not eliminate it over storage periods of months to years. Partial fat coalescence can also contribute: fat globules adhere to form larger effective radii (r term in Stokes' law is squared), accelerating creaming. Homogenization prior to evaporation reduces fat globule size to <2 μm, greatly slowing separation, but imperfect homogenization or post-process globule aggregation can produce visible fat layers.

Calcium Sediment

A white, granular sediment at the bottom of evaporated milk cans is a common physical instability, not a microbial defect. The sediment consists primarily of calcium phosphate (Ca₃(PO₄)₂), calcium citrate (Ca₃(C₆H₅O₇)₂·4H₂O), and magnesium phosphate (Mg₃(PO₄)₂) — minerals naturally present in milk that have limited solubility at the high concentrations achieved through evaporation and that precipitate during the heat sterilization process and subsequent storage.

The precipitation is thermodynamically favored by the concentration effect (evaporated milk contains approximately 2× the mineral concentration of fluid milk) and kinetically accelerated by the heat treatment, which reduces calcium phosphate solubility (retrograde solubility — calcium phosphate is less soluble at elevated temperatures). The sediment is harmless but represents a commercial quality defect — consumers find it visually objectionable and it can indicate inadequate homogenization or excessive storage time.

Core Science: Packaging Integrity — The Microbial Barrier

Can Integrity and Spoilage

Commercial sterility in evaporated milk and osmotic preservation in SCM are contingent on packaging integrity. The can (typically tinplate with internal lacquer coating, or aluminum for tubes) serves as a hermetic barrier preventing recontamination by environmental microorganisms. Any breach in this barrier — seam defects, pinhole corrosion, dent-related micro-fractures — creates a route for microbial ingress and spoilage.

Can swelling (胀听) is the most diagnostically significant packaging defect. Three mechanisms can produce internal positive pressure:

Hydrogen swell: Internal corrosion of uncoated or damaged tinplate produces H₂ gas from the reaction of tin/iron with acidic milk components. This is a purely chemical reaction — the product remains sterile but the can integrity is compromised, and the product should be discarded (metal dissolution products may have entered the food).

Microbial gas production: Contaminant microorganisms — typically spore-forming bacteria that survived inadequate sterilization (under-processing) or entered through a compromised seam — metabolize milk components producing CO₂ and/or H₂. Common organisms: Bacillus spp. (facultative anaerobes), Clostridium spp. (obligate anaerobes that produce H₂ and CO₂ through butyric acid fermentation). Any can swelling with microbial origin represents a food safety hazard — the organisms that produce the gas may include pathogens (Clostridium botulinum, though its spores require ≥F₀ 2.5-3.0 for 12-log reduction, and commercially sterile processes typically achieve F₀ >5.0).

Thermal expansion: A can that swells when heated and contracts when cooled exhibits thermal expansion only — this is not spoilage but a normal physical phenomenon.

Inspection Protocol

Industrial and commercial quality control for canned condensed milk follows a standardized inspection sequence: (1) visual inspection for dents, rust, leaking, or swelling; (2) tap test — a normal can produces a dull, solid sound; a swollen can produces a high-pitched, hollow "drum" sound; (3) opening inspection — normal product should have uniform color, smooth consistency, and clean dairy aroma; (4) sensory evaluation after opening — any off-odor, gas release upon opening, visible mold, or abnormal color/texture → reject and investigate lot.

Research Evidence

Study	Key Finding	n	Methodology	Industrial Implication
Martins et al. (2001)	Maillard browning Ea 100-130 kJ/mol in dairy systems; rate at 30°C: 15-25× rate at 10°C	Model dairy systems	Kinetic modeling	Tropical storage reduces SCM shelf life 50-70%
van Boekel (2006)	Maillard in concentrated milk: lysine blockage 10-30% after 12 months at 25°C; nutritional quality loss accompanies browning	Storage study, HPLC	Furosine measurement	Nutritional degradation parallels color/aroma changes
Lewis (2016)	Calcium phosphate precipitation in evaporated milk: concentration factor 2.1× → solubility product exceeded; sedimentation rate temperature-dependent	Physicochemical analysis	Solubility modeling, sediment analysis	Calcium sediment is physical instability, not microbial defect
Fox & McSweeney (2015)	Lipid oxidation in high-solids dairy: aw 0.80-0.85 → oxidation rate maximum in SCM; concurrent Maillard products may mask rancid aromas	Lipid chemistry review	PV, TBARS	Oxidation in SCM may be sensorially masked by Maillard aromas
Fleet (1990)	Yeast spoilage of SCM: osmotolerant Z. rouxii capable of growth at aw 0.62, but sucrose concentration in SCM (62% w/v in water) is fungistatic	Dairy yeast survey	aw-controlled growth curves	SCM microbiological stability: robust under intact packaging
Walstra et al. (2006)	Homogenization effect on evaporated milk: globule size <2 μm → creaming rate reduced 10-30×	Dairy technology textbook	Stokes' law, particle sizing	Adequate homogenization essential for physical stability
Jay et al. (2005)	Can swells: 3 types (hydrogen, microbial CO₂, thermal); microbial swells → Clostridium or Bacillus → food safety hazard	Food microbiology textbook	Differential diagnosis	Never consume product from a swollen can
Codex Alimentarius (2018)	SCM standard: min 8% milkfat, min 28% total milk solids, max 27% water	International standard	Specification	Regulatory identity standard for SCM
Nieuwenhuijse et al. (1998)	Age gelation in evaporated milk: calcium-mediated protein cross-linking during storage; gelation onset 6-18 months	Storage study, rheology	Viscosity monitoring	Age gelation limits evaporated milk shelf life independent of sterility
Fox & McSweeney (1998)	Lactose crystallization in SCM: seeding with ≤10 μm crystals controls size to ≤25 μm (below oral detection threshold)	Dairy chemistry textbook	Microscopy, sensory	Seed crystal technology critical for SCM smoothness
Rahman (2009)	aw threshold for microbial growth: vegetative bacteria >0.85; SCM aw 0.80-0.85 → no vegetative growth possible	Review	aw meter, challenge studies	SCM microbiological stability foundation validated
Hendrickx et al. (1995)	Retort thermal process design: F₀ >5.0 for evaporated milk; C. botulinum 12-log reduction at F₀ 3.0 → safety margin	Process engineering textbook	Heat penetration, F₀ calculation	Evaporated milk: thermal sterilization robustness confirmed

FAQ: Condensed Milk Shelf Life and Spoilage

Q1: Does sweetened condensed milk ever truly spoil, or does it just degrade in quality?

Sweetened condensed milk (SCM) with intact packaging and a_w 0.80-0.85 will not undergo microbial spoilage — the water activity is below the growth threshold for all vegetative bacteria and most osmotolerant yeasts. Quality deterioration through Maillard browning (color darkening from cream to tan to brown; flavor shifting from sweet dairy to caramelized and eventually bitter), fat oxidation, and lactose crystallization will progressively degrade sensory quality, but the product remains microbiologically safe. Evaporated milk, by contrast, relies on thermal sterilization and hermetic sealing — if either barrier fails, rapid microbial spoilage can occur because its a_w (~0.97-0.99) is fully permissive for bacterial growth. The key distinction: SCM quality degrades but safety persists; evaporated milk safety is seal-dependent.

Q2: Is browning in condensed milk a sign it's unsafe?

No. Maillard browning (non-enzymatic browning) is a chemical reaction between reducing sugars (lactose) and milk proteins that produces brown polymers (melanoidins). It is accelerated by elevated storage temperatures (rate 15-25× faster at 30°C vs. 10°C) and prolonged time. Browned SCM is sensorially degraded (bitter, caramelized, "burnt" flavors) and nutritionally compromised (lysine blockage reduces available protein quality), but it is not microbiologically hazardous. If the can is intact, unswollen, and there is no off-odor upon opening, heavily browned SCM is safe to consume — though its flavor may be objectionable. If browning is accompanied by can swelling, off-odor, or gas release upon opening, discard — these indicate microbial involvement, not Maillard chemistry alone.

Q3: What causes the white sediment at the bottom of evaporated milk cans?

The white, granular sediment is composed of calcium phosphate (Ca₃(PO₄)₂), calcium citrate, and magnesium phosphate — naturally occurring milk minerals that precipitate from the concentrated solution during heat sterilization and subsequent storage. The mechanism: (a) milk concentration during evaporation increases mineral concentrations ~2×; (b) the retort sterilization (115-121°C) reduces calcium phosphate solubility (retrograde solubility — less soluble at high temperature); (c) mineral crystals nucleate on can surfaces and sediment over time. The sediment is a physical instability, not a microbial or chemical spoilage issue. It is safe to consume, though its presence indicates either extended storage or inadequate homogenization/stabilization in the manufacturing process. Vigorous shaking before opening can resuspend the sediment if it is fine (non-cemented).

Q4: Can I use condensed milk after the "Best Before" date?

Unopened SCM with intact packaging: yes, typically for 6-12 months beyond the printed date, provided the can is undamaged and has been stored at moderate temperatures (<30°C). The "Best Before" date is a quality indicator — color and flavor may have degraded, but the product is unlikely to be unsafe. Unopened evaporated milk: more cautiously. While commercial sterility should persist indefinitely in an intact can, internal can corrosion can develop over time (particularly in acidic products), potentially introducing metal ions and compromising seal integrity. Evaporated milk >6 months past its date should be carefully inspected before use (can integrity, odor upon opening, no gas release). Once opened, both SCM and evaporated milk must be treated as perishable — refrigerate and use within 3-5 days (SCM) or 2-3 days (evaporated milk).

Q5: Why does my condensed milk sometimes crystallize and become grainy?

Lactose crystallization is the cause. Lactose has relatively low solubility (~18 g/100g water at 25°C) compared to sucrose (~200 g/100g water). In SCM, the total water content is only 26-30%, and the lactose concentration — approximately 10-12% of total weight or 35-45 g/100g water — is far above the saturation limit. The solution is thermodynamically supersaturated with respect to lactose. Manufacturers control crystallization by seeding: they add finely ground lactose crystals (≤10 μm) to the cooled, concentrated product, inducing controlled, uniform crystallization into crystals ≤25 μm — below the oral detection threshold for grittiness (~15-20 μm). If seeding is inadequate, or if the product experiences temperature fluctuations that dissolve small crystals and regrow large ones (Ostwald ripening, analogous to ice crystal growth in ice cream), crystals can exceed 25 μm and produce a sandy, gritty texture. The product is safe but sensorially defective. Large crystal formation is irreversible.

Q6: What happens if I boil an unopened can of condensed milk (to make dulce de leche)?

Boiling a sealed can of SCM — the traditional method for making dulce de leche — accelerates the Maillard reaction dramatically, producing the characteristic dark brown color and caramelized flavor. The can must remain completely submerged in water throughout the heating process (typically 2-3 hours at a gentle boil). Critical safety warnings: (a) never let the water level drop below the top of the can — an exposed can top can overheat and explode violently from internal steam pressure; (b) never heat an unopened can in a microwave; (c) always allow the can to cool completely before opening — opening a hot, pressurized can releases superheated product that can cause severe burns. The risk of can explosion is real and has caused documented injuries. A safer alternative: open the SCM, transfer to a heatproof bowl over a water bath (bain-marie), cover with foil, and bake or steam. The chemistry is identical but the physical hazard is eliminated.

Q7: How can I tell if an unopened can of condensed milk is spoiled?

Inspection sequence: (a) External — check for rust, severe dents (especially on seams), leakage, or can swelling (bulging ends that do not depress or that spring back when pressed). Any of these → discard without opening. (b) Opening — a hiss of gas release upon piercing is normal for evaporated milk (internal vacuum from retort cooling) but abnormal for SCM under vacuum-sealed cans. A positive pressure release (spurting) → discard (microbial or chemical gas production). (c) Visual — the product should be uniform in color (cream to light tan for SCM; white to cream for evaporated milk) and consistency (smooth, viscous, pourable). Dark brown color, separated layers, curdled appearance, or visible mold → discard. (d) Olfactory — clean, sweet, milky aroma. Sour, putrid, yeasty, rancid, or "off" odors → discard.

Q8: Does sweetened condensed milk have more shelf life than evaporated milk?

Yes — and the fundamental reason is the preservation mechanism. SCM relies on osmotic preservation (a_w 0.80-0.85): even if the can integrity is compromised post-manufacture, the low a_w prevents microbial growth. The product may dry out, absorb odors, or be contaminated with dust — but it will not support bacterial proliferation. Evaporated milk relies on thermal sterilization + hermetic sealing: if the can integrity is compromised, the high a_w (~0.97-0.99) allows rapid microbial spoilage. Additionally, SCM's high sugar content provides some antioxidant effect (sucrose can scavenge free radicals), modestly slowing lipid oxidation. In practice: unopened SCM stored at 20-25°C maintains acceptable quality for 18-24 months from production; evaporated milk for 12-18 months. Once opened, both require refrigeration; SCM typically lasts 3-5 days (sugar provides some protection); evaporated milk 2-3 days (no preservation barrier remaining).

Q9: Why does evaporated milk sometimes gel (become solid) in the can?

Age gelation is a progressive increase in viscosity during extended storage of evaporated milk, sometimes culminating in a semi-solid, yogurt-like gel. The mechanism involves: (a) slow protein-protein interactions — primarily calcium-mediated cross-linking of casein micelles — that gradually build a three-dimensional protein network; (b) Maillard reaction products that can participate in protein cross-linking; (c) plasmin (native milk proteinase) — incompletely inactivated by the sterilization process in some products — slowly hydrolyzes casein, producing peptides that participate in gel network formation. Age gelation is a physical-chemical quality defect, not microbial spoilage. The product is safe (commercially sterile) but sensorially unacceptable. Gelation onset typically occurs at 6-18 months of storage, accelerated by elevated temperatures. Products with higher total solids, higher calcium content, and lower pre-heat treatment during manufacture are more susceptible. There is no reliable way to reverse gelation; gelled product should be discarded on quality grounds.

Q10: Can condensed milk develop botulism?

The risk of Clostridium botulinum toxin production in commercially produced condensed milk is effectively zero, but the mechanism differs between product types. In SCM, a_w 0.80-0.85 is below the minimum for C. botulinum growth and toxin production (minimum a_w ~0.94 for proteolytic types; ~0.97 for non-proteolytic types). Even if spores survived the forewarming heat treatment (which they may, as forewarming temperatures of 82-100°C for 10-30 minutes are below the thermal death time for spores), they cannot germinate, grow, or produce toxin at the product a_w. In evaporated milk, the retort sterilization process achieves F₀ >5.0, which exceeds the F₀ 2.5-3.0 required for 12-log reduction of C. botulinum spores — providing a substantial safety margin. The only botulism risk from condensed milk would arise from home-prepared condensed milk or from a gross manufacturing failure (severely under-processed evaporated milk in a can with an undetected seam leak allowing post-process contamination). Both scenarios are extremely rare in modern commercial production under HACCP control.

Related Research

• Sour Cream Shelf Life Science — Another fermented dairy product with distinct preservation chemistry
• Hydrolytic Rancidity in Butter — Dairy fat degradation at the lipid-water interface
• Water Activity and Food Stability — The foundational role of a_w in food preservation
• Microbial vs Chemical Spoilage Explained — Framework for distinguishing spoilage types
• Ingredients & Additives: Their Role in Food Stability — How formulation determines shelf life

References

1. Martins, S. I. F. S., Jongen, W. M. F., & van Boekel, M. A. J. S. (2001). A review of Maillard reaction in food and implications to kinetic modelling. Trends in Food Science & Technology, 11(9-10), 364-373. https://doi.org/10.1016/S0924-2244(01)00022-X

2. van Boekel, M. A. J. S. (2006). Formation of flavour compounds in the Maillard reaction. Biotechnology Advances, 24(2), 230-233. https://doi.org/10.1016/j.biotechadv.2005.11.004

3. Lewis, M. J. (2016). Physical Properties of Foods and Food Processing Systems. Woodhead Publishing. https://doi.org/10.1016/C2014-0-02694-0

4. Fox, P. F., & McSweeney, P. L. H. (2015). Advanced Dairy Chemistry: Volume 2 — Lipids (3rd ed.). Springer. https://doi.org/10.1007/978-1-4614-4714-6

5. Fleet, G. H. (1990). Yeasts in dairy products. Journal of Applied Bacteriology, 68(3), 199-211. https://doi.org/10.1111/j.1365-2672.1990.tb02566.x

6. Walstra, P., Wouters, J. T. M., & Geurts, T. J. (2006). Dairy Science and Technology (2nd ed.). CRC Press. https://doi.org/10.1201/9781420028010

7. Jay, J. M., Loessner, M. J., & Golden, D. A. (2005). Modern Food Microbiology (7th ed.). Springer. https://doi.org/10.1007/978-0-387-23413-7

8. Codex Alimentarius Commission. (2018). Standard for Sweetened Condensed Milks (CXS 282-1971, Rev. 2018). FAO/WHO. https://www.fao.org/fao-who-codexalimentarius

9. Nieuwenhuijse, J. A., van Vliet, T., & Walstra, P. (1998). Age-thinning and age-thickening of concentrated milk products. Netherlands Milk and Dairy Journal, 42(3), 271-290.

10. Fox, P. F., & McSweeney, P. L. H. (1998). Dairy Chemistry and Biochemistry. Springer. https://doi.org/10.1007/978-1-4615-2055-1

11. Rahman, M. S. (2009). Food stability determination by water activity. Critical Reviews in Food Science and Nutrition, 49(7), 535-552. https://doi.org/10.1080/10408390802544458

12. Hendrickx, M., Maesmans, G., De Cordt, S., Noronha, J., Van Loey, A., & Tobback, P. (1995). Evaluation of the integrated time-temperature effect in thermal processing of foods. Critical Reviews in Food Science and Nutrition, 35(3), 231-262. https://doi.org/10.1080/10408399509527701

13. Lund, B. M., Baird-Parker, T. C., & Gould, G. W. (Eds.). (2000). The Microbiological Safety and Quality of Food. Aspen Publishers. https://doi.org/10.1007/978-1-4615-4447-0

14. Frankel, E. N. (2014). Lipid Oxidation (2nd ed.). Woodhead Publishing. https://doi.org/10.1016/C2011-0-07245-4

15. International Commission on Microbiological Specifications for Foods. (2005). Microorganisms in Foods 6: Microbial Ecology of Food Commodities (2nd ed.). Springer. https://doi.org/10.1007/0-387-28801-4

About the Author

Martin Wang — Food Scientist | Industrial Processing Expert

Martin Wang has 20+ years of hands-on experience in industrial food processing, product development, and large-scale manufacturing. He has led multiple commercial food projects from factory to market and specializes in shelf-life control, water activity management, and process optimization. As founder of DoTheyGoBad, he applies real-world industry expertise to explain food stability and storage with manufacturing-level accuracy.

View author profile · Back to all articles