Texture and Browning Behavior of Allulose in Confectionery Cooking

D-Allulose (D-psicose, CAS 551-68-8) is a ketohexose epimerized at the C-3 position relative to D-fructose, and its behaviour in confectionery cooking is defined by a combination of reducing-sugar reactivity, high aqueous solubility, low molecular mass of 180.16 g/mol, and sweetness approximately 70% that of sucrose at 10% solids. The metabolizable energy value assigned under US FDA nutrition labelling is 0.4 kcal/g, and the ingredient is covered by GRAS notices GRN 000400 and GRN 000498, while FDA guidance excludes allulose from added sugars declarations on the basis of negligible postprandial glycaemic impact. In high-temperature confectionery processing, however, allulose does not act as a neutral bulk sweetener. Its reducing carbonyl participates directly in enediol-mediated dehydration and in nitrogen-carbonyl reactions with dairy proteins, gelatin, or free amino acids, so browning begins earlier and proceeds through different intermediate chemistry than sucrose-based formulas. The crystalline material shows a differential scanning calorimetry endotherm in the 96–112 °C region depending on crystal form, hydration state, and scan rate, with commercial thermal evaluations frequently reporting an onset near 109 °C for the anhydrous crystal. That thermal event lies below conventional hard-candy boiling endpoints of 145–155 °C, meaning allulose melts and dissolves before it can serve as a stable crystalline scaffold during final evaporative concentration. Furthermore, because allulose is a monosaccharide, it contributes roughly twice the osmotic mole count per unit mass as sucrose at equivalent solids, which depresses water activity but also increases hygroscopicity and modifies the glass-rubber transition of finished confectionery matrices. Formulators replacing sucrose with allulose must therefore re-engineer cook temperature, vacuum level, residual moisture, acid buffering, and package barrier properties rather than treat the substitution as a drop-in sweetener exchange. The industrial risk is not simply sweetness adjustment; it is a shift in the kinetic competition between sufficient drying, caramelization, Maillard colour formation, and low-molecular-weight plasticization that controls cold flow and stickiness on wrapped-piece storage lines.

How Does the Reducing Capacity of Allulose Alter Caramel Colour Development in Protein-Containing Systems?

In caramels, toffees, and other dairy-based confections, colour is generated principally through the Maillard reaction between reducing sugars and the free amino groups of milk proteins, with additional acid- or heat-driven sugar degradation contributing brown melanoidins and caramelization products. Allulose is a ketose and therefore enters the Maillard cascade through the formation of Heyns compounds following carbonyl-amine condensation, rather than through the Amadori products characteristic of aldoses such as glucose. The open-chain carbonyl concentration of allulose in hot aqueous syrup is governed by mutarotation and enolization equilibria, and commercial experience indicates that the rate of colour intensity increase is more rapid than sucrose-only controls when milk solids, evaporated skim milk, or whey protein concentrate with 80% protein are present. The pH of the caramel batch is a critical variable: Maillard browning accelerates in the weakly acidic to neutral range, commonly 6.0–7.2, and slows below pH 5.0 because protonation of the free amine suppresses the condensation step. In practice, allulose-containing caramel processed in a steam-jacketed batch cooker with scraped-surface agitation at 118–121 °C reaches a target CIE L* value of 48–52 measurably sooner than a sucrose-based batch at equivalent total solids. Published peer-reviewed kinetic data for this exact confectionery matrix are limited, but pilot-scale colourimeter logging on evaporated-milk caramel lines demonstrates that replacement of 25% of sucrose with allulose can shorten time-to-target colour by 10–20% when cook temperature and protein concentration are held constant. The measurement system is typically a 45/0 reflectance spectrophotometer or a tristimulus colorimeter reporting CIELAB coordinates under D65 illumination and a 10° observer, with colour difference expressed as ΔE*ab according to ISO 11664-4:2008. Because allulose also degrades through caramelization and acid-catalysed 5-hydroxymethylfurfural formation above 120 °C, the processing window between acceptable colour and burnt flavour narrows when the syrup contains reducing milk proteins and mineral buffers. Operators typically compensate by lowering the final cook temperature by 3–5 °C or increasing the vacuum pull to reduce residence time at high temperature, but both adjustments alter final moisture and yield stress. The colour trajectory is not linear with allulose dosage; at replacement levels above 50% of total sweetener solids, the browning rate can become sufficiently rapid that conventional batch kettles overshoot the target colour during discharge and transfer, especially if the downstream holding tank is unjacketed and retains heat for 10–20 min. For this reason, allulose caramels are best processed in continuous scraped-surface thin-film evaporators with short residence time, or in batch kettles with rapid vacuum cooling and discharge temperature no higher than 105 °C.

Continuous hard candy manufacture with allulose requires a downward revision of the vacuum boiling endpoint and a longer cooling-tunnel dwell because the cooked glass enters the rubbery condition at ambient temperature more readily than sucrose-based glass. A typical sugar-free or reduced-sugar hard candy line uses a thin-film evaporator or a continuous vacuum cooker operating at 145–155 °C with vacuum of 0.15–0.25 bar absolute and residual moisture after cooking of 1.5–2.5% by Karl Fischer titration according to ASTM E203-16. In allulose-rich formulas, the monosaccharide's lower molecular mass and high hygroscopicity depress the glass transition temperature, so the same residual moisture produces a softer piece that may block in the cooling tunnel or flow on the wrapping conveyor. Glass transition is measured by differential scanning calorimetry according to ASTM E1356-08(2014), and production troubleshooting on hard candy lines indicates that replacement of 50% of sucrose solids with allulose can reduce the characteristic onset of the glass transition by several kelvin at equal moisture, although exact values depend on residual water, corn syrup solids, acid inversion, and any added polyols. The practical consequence is observed at the cooling and wrapping stages: allulose hard candy discharged from a rotary die former at 85–95 °C may remain deformable for 20–30% longer than conventional sucrose syrup when the tunnel relative humidity exceeds 30%. Some manufacturing lines retrofit the cooling tunnel with dehumidified air at 20–25% RH and extend the residence time by 15–25% to achieve a surface hardness sufficient for flow-wrap packaging. Failure to condition the environment produces cold flow within 48 h at 25 °C and 60% RH, where individual pieces stick to cold-seal film and create jammed transfer belts. The operational boundary is sharp: allulose content above 30% of total sweetener solids generally demands a package water vapour transmission rate below 0.5 g/(m²·day) at 38 °C and 90% RH as determined by ASTM F1249-20. Without such a barrier, the glass transition falls below ambient storage temperature because absorbed moisture plasticizes the amorphous matrix, and hardness measured by a texture analyser with a 3-point bend fixture declines below acceptable fracturability within days. The same moisture ingress raises surface tack and accelerates Maillard or caramelization residual colour development even in a finished piece because allulose remains chemically available rather than becoming inert once the glass solidifies.

When Fondant and Fudge Systems Are Rebalanced Around Allulose Crystallization

Fondant and fudge systems rely on controlled sucrose crystallization to produce a soft, fine-grained texture with shear-thinning flow and low water activity after cooling. Allulose disrupts this texture paradigm because it is a highly soluble monosaccharide that does not crystallize readily during the agitation and cooling steps that develop the desired fondant grain. Traditional fondant is cooked to 114–118 °C, cooled to 45–50 °C, and then beaten or crystallized on a swept-surface cooling crystallizer where a seed slurry or mechanical shear generates fine crystals with mean diameters commonly targeted below 20 µm. When allulose replaces 25–50% of the sucrose solids, the supersaturation state of the cooling mass shifts because allulose remains largely dissolved in the intercrystalline syrup phase. The syrup phase becomes more fluid and hygroscopic, while crystal growth of the remaining sucrose may be slowed or constricted because the increased solute viscosity of the concentrated allulose phase reduces molecular diffusion to the crystal interface. The resulting fondant is softer, with lower yield stress and a greater tendency to dry out slowly or remain tacky at room temperature. Production equipment observations on swept-surface cooling units with rotor speeds below 20 rpm show that allulose batches require a lower final moisture content, often 8–10%, to avoid a greasy surface sheen and to maintain sufficient rigidity for depositing into centres or enrobing. Particle size distribution is measured by laser diffraction using ISO 13320:2020, and texture is commonly assessed with a cone penetrometer or a texture analyser using a 45° cone at 5 mm penetration. Fudge processed with allulose may exhibit a coarser perceived texture despite a smaller mean crystal size because the residual syrup phase is soft and wet, making the product chewy rather than short. The browning behaviour of allulose in fudge is also significant: because fudge is cooked at 112–116 °C with milk solids and often salt, the Maillard reaction is active, and allulose-containing batches can darken during the cooling and beating stages to a greater degree than sucrose controls, particularly if the cooling mass remains above 70 °C for extended agitation. Published data for allulose fondant crystal morphology under industrial swept-surface crystallization are limited, so process developers generally rely on pilot-scale validation with the specific rotor speed, wall temperature, and seed addition 0.5–1.0% by mass before committing to production scale.

Starch and pectin gummy depositing lines with allulose show reduced initial gel strength and increased brown discoloration in the drying room because the monosaccharide's high solubility and reducing activity persist at water activity levels between 0.60 and 0.75. Gelatin gummies are commonly deposited into starch trays at 60–70% total solids and 65–75 °C, then aged at 18–25 °C and 35–45% RH for 18–48 h to reach the desired moisture and gel firmness. Allulose changes the moisture release rate because it retains water more strongly than sucrose under declining RH, slowing the drying step and leaving the piece surface sticky at the demoulding point. In pectin confections, where the pH is typically 3.0–3.4 to permit gelation with slow-set high-methoxyl pectin, allulose is exposed to acidic conditions during extended mogul storage, and acid-catalysed dehydration to hydroxymethylfurfural can produce yellow-brown surface coloration over a multi-day drying window. The colour shift is measurable as a decrease in L* and an increase in b* on the piece surface using the same CIELAB geometry described in ISO 11664-4:2008. Gel strength is evaluated with a texture analyser using a cylindrical probe of 12.7 mm diameter and a compression depth of 4 mm at 1 mm/s; allulose-containing gelatin formulas typically exhibit lower initial bloom values before drying and require an additional 6–12 h in the starch room to reach equivalent firmness. Depositing viscosity also shifts because allulose lowers the apparent viscosity of the hot gum syrup at equal total solids, and this can increase tailing and uneven deposit weights unless the depositor nozzle temperature is reduced or the solids are increased. On a starch mogul line with depositor nozzle orifices in the range of 2.0–3.5 mm, tailing is managed by raising the syrup solids by 2–3% or lowering the deposit temperature from 75 °C to 70 °C, but these adjustments cannot eliminate the browning risk caused by the reducing carbonyl group. The operational limitation is therefore twofold: the drying curve must be extended to overcome allulose-mediated moisture retention, and the acidulated allulose phase must be protected by minimizing holding time above 70 °C and by adding buffering salts where flavour permits.

Viscosity Depression, Burn-On Fouling, and Scraped-Surface Heat Transfer

The lower molecular mass of allulose relative to sucrose changes the rheology of concentrated confectionery syrups in ways that directly affect heat transfer and fouling behaviour in continuous scraped-surface thin-film cookers. At the same total solids and temperature, allulose syrups tend to show lower apparent viscosity than sucrose syrups because monosaccharides generate less molecular entanglement and a different water-binding structure than disaccharides. On a steam-jacketed scraped-surface evaporator with wall temperature 145–165 °C, rotor speed 100–200 rpm, and residence time 30–90 s, lower viscosity permits faster film drainage under gravity and can increase the frequency of transient dry patches on the upper wall. These dry patches become sites of accelerated caramelization and Maillard reaction when proteins or dairy solids are present, leading to burn-on fouling that reduces heat transfer and generates dark specks in the finished confectionery. The fouling layer acts as an insulating film; production records on caramel and hard candy lines indicate that allulose batches tend to require more frequent clean-in-place cycles, commonly with 2% sodium hydroxide solution at 80 °C for 45–60 min, to restore heat transfer coefficients to baseline. Rotational viscometry using a Brookfield RVDV-II+ Pro with small sample adapter at 60 °C and 10 rpm provides a comparative log of syrup viscosity during development, but the relevant parameter in the scraped-surface film is not bulk viscosity alone; it is the wall shear stress and film thickness under centrifugal and gravitational drainage. Allulose's reducing behaviour aggravates fouling because the carbonyl group reacts with amino compounds at the heated wall even when the bulk syrup does not reach severe discoloration. The heat transfer coefficient may decline by 10–20% over a production day at high allulose dosage, requiring operators to increase steam pressure or reduce throughput to hold the same boiling endpoint. In contrast, sucrose-only syrups with similar total solids tend to foul more slowly because sucrose is non-reducing until it undergoes thermal inversion; acid-induced inversion can equalize this behaviour, but neutral syrups show a marked difference. The practical control strategy is to lower the temperature differential between the wall and the boiling film, use shorter residence time, and avoid high-protein recipes on thin-film equipment running at allulose replacement above 50% of total sweetener solids. The same viscosity depression is less detrimental in batch kettles with strong top-driven agitation, but it still influences the boil-off rate and requires recalibration of the refractometer endpoint because refractive index readings of allulose syrups do not correlate identically with sucrose-based calibration curves at high solids.

The Packaging Barrier and Shelf-Life Failure Modes of Allulose Boiled Sweets

Allulose boiled sweets fail most often through moisture uptake that drives the amorphous glass into the rubbery state, producing cold flow, surface stickiness, and package adhesion before any microbial or flavour deterioration is observed. The moisture sorption isotherm of allulose is shifted upward relative to sucrose in the low and intermediate water activity range, so the same packaging film and sealing conditions do not provide equal shelf-life protection. A wrapped hard candy stored at 25 °C and 60% RH will absorb moisture until its surface layer reaches local equilibrium; because allulose is hygroscopic and low in molecular mass, absorbed water sharply reduces the glass transition and initiates flow under the piece's own weight. The failure mode is quantified by dimensional change, stick force measured with a tensile probe after 1 h of ambient conditioning, and glass transition onset by ASTM E1356-08(2014). Packaging specifications for allulose-containing boiled sweets generally require a water vapour transmission rate below 0.5 g/(m²·day) at 38 °C and 90% RH when measured by ASTM F1249-20, and in high-humidity distribution markets the limit may be tightened to 0.2 g/(m²·day). Film structures that meet this requirement include metallized biaxially oriented polypropylene of 18–20 µm thickness with a cold-seal coating, or multilayer cast polypropylene with ethylene vinyl alcohol barrier layers. Single-layer orientated polypropylene is generally insufficient at allulose replacement above 25%, as field complaints of sticking within 2–4 weeks are common in tropical distribution channels. The second shelf-life failure mode is progressive browning inside the package, driven by residual moisture, trace amino compounds, and the continued availability of allulose as a reducing sugar. Even in a sealed package, the piece can darken over weeks at 30–35 °C, especially if the formula contains fruit acids, milk solids, or protein-based flavour systems. Therefore, the packaging barrier must address both the physical plasticization pathway and the chemical browning pathway; a moisture-impermeable film alone cannot prevent colour drift if the piece is filled or laminated with a protein-containing layer that retains reactive amines. Production teams commonly specify finished-piece water activity below 0.45 for allulose hard candy to slow both stickiness and browning, but this is not always achievable with traditional boiling endpoints and requires vacuum distillation or extended cooking.

Colour and Texture Quantification Under Standardised Confectionery Test Protocols

The development of allulose-containing confectionery requires a coordinated measurement matrix because colour, texture, and moisture are mutually coupled through the glass transition and the Maillard reaction. A single analytical technique cannot resolve the processing conflicts; instead, the same sample should be evaluated for glass transition, water activity, moisture content, apparent viscosity, surface colour, and mechanical hardness under defined geometry and extension rate. The table below lists the principal methods used on confectionery pilot lines and the parameters that are most sensitive to allulose substitution. The methods are not exhaustive, but they provide a compliance-linked framework for comparing allulose formulas against sucrose controls without relying on subjective sensory language.

When a formulation crosses from sucrose to allulose, the most diagnostically informative measurements are not the absolute hardness or colour values but the rate of change of glass transition with moisture and the time-dependent decrease in L* at constant temperature and protein load. A confectionery batch that meets its immediate hardness target on the cooling tunnel may still fail after 72 h under retail blister pack conditions because the glass transition falls below the ambient dry-bulb temperature following moisture absorption. Similarly, a caramel that leaves the cooker within a target L* range may continue to brown in an insulated holding tank if allulose and milk proteins remain in contact above 100 °C, because the Maillard reaction has not been quenched by sufficient cooling or pH adjustment. Analytical protocols should therefore include kinetic sampling: measure the product at 0, 24, 48, and 72 h after production under controlled RH and temperature, and record the glass transition, surface tack, and colour difference. The use of standardized methods does not remove the need for pilot-scale confirmation on actual manufacturing equipment, because no benchtop measurement fully reproduces the film drainage, shear history, and thermal history of a continuous scraped-surface cooker or a starch mogul drying room. For allulose-based confectionery, the most actionable engineering specification is the combination of a tightly specified water activity, a measured glass transition at the finished moisture content, and a package barrier that keeps the piece below the critical moisture threshold for its intended shelf-life. Formulations that ignore this coupling between texture and browning are prone to field failure even when the cooked syrup meets the target refractive solids and the fresh piece passes an immediate snap test.