Formulation of a DGEBA-based floor topping mortar begins with control of the liquid epoxy resin equivalent weight and the amine hydrogen equivalent weight of the hardener. A standard bisphenol A diglycidyl ether resin with an epoxide equivalent weight of 188–192 g/eq is reacted with isophorone diamine at an amine hydrogen equivalent weight of 42.6 g/eq. For a 100 g resin charge, the stoichiometric addition is 22.4 g of isophorone diamine. Industrial mortar production commonly doses at 21–24 phr because the epoxide equivalent weight of delivered resin lots can shift by ±1.5% between batches. The cure reaction proceeds by nucleophilic attack of primary amine hydrogens on oxirane rings, generating secondary amines that subsequently react with additional oxirane groups to form a cross-linked network. Differential scanning calorimetry according to ASTM E1356 at a ramp rate of 10 K/min typically records an onset of cure near 55–75 °C, a peak exotherm between 120–140 °C, and total reaction enthalpy in the range 380–460 J/g for the neat resin-hardener mixture. In a 300 L planetary mixer fitted with a vacuum lid, the resin and hardener are blended at 40–60 rpm under 5–10 kPa absolute pressure to reduce air entrapment before graded aggregate is added. The use of a vacuum mixing stage reduces bubble-induced surface defects that would otherwise increase localized abrasion loss under ASTM D4060 wheel wear testing. Industrial floor topping systems based on this chemistry are specified where heavy-wheel traffic, impact, and aggressive cleaning demand a hard, chemically resistant surface with predictable wear behavior.

When Stoichiometric Imbalance Moves From 2 phr to 6 phr, the Wear Index Changes Nonlinearly

Under-cure from isophorone diamine addition below 21 phr and over-cure above 24 phr alter the network architecture differently. With an under-index of 0.85–0.95, residual oxirane groups remain after ambient cure, lowering the glass transition temperature and reducing surface hardness. With an over-index of 1.05–1.15, unreacted amine can plastify the network and increase hygroscopicity. Taber wear testing per ASTM D4060 using CS-17 wheels, a 1000 g load, and 1000 cycles on 6 mm thick castings demonstrates that a stoichiometric isophorone diamine/DGEBA binder typically yields a wear index of 55–75 mg/1000 cycles. At 2 phr under-index, the wear index may rise to 85–110 mg/1000 cycles, while at 4 phr under-index values above 130 mg are reported in industrial coating evaluations. The cliff-edge is sharper at over-index: increasing isophorone diamine from 24 phr to 26 phr can raise the wear index by 40–60% because free amine groups absorb atmospheric carbon dioxide and form surface carbamates that disrupt the wearing surface. Crosslink density calculated from rubber elasticity theory using the storage modulus in the rubbery plateau at 180 °C from dynamic mechanical analysis per ASTM E1640 declines from approximately 2.8–3.5 × 10⁻³ mol/cm³ to below 1.8 × 10⁻³ mol/cm³ when the hardener is shifted 4 phr off stoichiometry. Because floor mortar wear is a surface phenomenon, indentation hardness measured as Shore D per ASTM D2240 may remain above 80 even when bulk crosslink density has fallen, making hardness alone an unreliable proxy for abrasion resistance. The nonlinear response is compounded in filled systems because a soft binder matrix permits aggregate particles to rotate and pull out under shear from abrading wheels, producing craters rather than gradual polishing.

Abrasion resistance in mortars is not governed solely by binder network density. Aggregate packing controls the area fraction of softer polymer exposed to the abrading medium. Floor topping formulations with quartz sand graded from 0.1 mm to 2.0 mm and silica flour with a median particle size of 10–20 µm approach maximum packing density when fine particles occupy interstices between coarse particles. The Fuller curve exponent n between 0.37 and 0.45 is commonly used to proportion aggregates. Formulations that remain above a filler-to-binder mass ratio of 4:1 but below the critical pigment volume concentration of approximately 55–65 vol% provide a balance of compressive strength, substrate wetting, and low wear loss. Below 4:1, the surface layer contains more binder and Taber wear loss increases; above the critical pigment volume concentration, insufficient binder wets the aggregate surfaces and loose particle generation dominates wear because interaggregate contacts lack a load-transferring polymer film. Viscosity measurement per ISO 3219 using a cone-plate rheometer at 25 °C and a shear rate of 1 s⁻¹ shows mixed mortar viscosity rising from 8–15 Pa·s at 50 wt% quartz loading to 40–80 Pa·s at 65 wt% loading, and above 70 wt% the material can exceed 150 Pa·s, which is beyond the pumping limits of continuous mortar mixers and rotor-stator feed pumps. Published data for this specific configuration is limited at aggregate loadings above 72 wt% because flow becomes paste-like and air release requires vacuum degassing below 2 kPa, with extended mixing that can consume working time before aggregate incorporation is complete.

What Limits Pot Life and Aggregate Wetting in Floor Topping Mortars?

Pot life is measured by the change in apparent viscosity and flow. In a 100 g static cup at 25 °C, a stoichiometric DGEBA/isophorone diamine mixture with no filler has an initial mixed viscosity of 800–1,500 mPa·s when measured by ASTM D2196 using a Brookfield LV rotational viscometer, spindle 4 at 20 rpm. The viscosity reaches 4,000–8,000 mPa·s after 30–40 min. In a 20 kg batch mixed in a planetary mixer at 20 °C jacket temperature, the exotherm can raise the core temperature to 45–60 °C within 15 min, shortening the application window to 10–20 min. Aggregate addition after gel onset leads to poor wetting and stress concentrations at the particle boundaries. A wetting threshold can be determined by measuring the spread of the mortar on a glass plate under vibration: a slump diameter of 160–200 mm at 5 min is often used as a production release criterion, but this test is qualitative. A more reproducible method is the spiral flow channel at 25 °C, where a 300 g charge should travel 600–900 mm at 10 min; below 400 mm, blade coating and troweling become difficult and entrapped air increases. Production-scale planetary mixers with 7.5 kW drive power per 100 kg batch and hydraulically driven agitator speeds from 10 rpm to 60 rpm are used to disperse fine fillers. These mixers cannot reverse gelation, and a loss of cooling water during high-shear dispersion can cause localized gelation along the vessel wall. Coating thicknesses of 3–6 mm require workability retention longer than 30 min, which is why isophorone diamine is sometimes blended with benzyl alcohol or aliphatic reactive diluents. Such diluents reduce crosslink density and hardness, and their emission potential is controlled under GB/T 23986 and related low-VOC standards. Extended pot life can also be obtained by pre-cooling the resin and aggregate to 15–18 °C, but lower temperatures increase viscosity by approximately 10–20% per 5 °C decrease and can compromise wetting of fine aggregate fractions.

Binder-Aggregate Interphase Failure and Abrasion Debris Formation

Wear in filled epoxy mortars proceeds by plowing of the resin matrix, fracture of weakly bonded aggregates, and pull-out of particles at the interphase. In falling sand abrasion per ASTM C501, the volume loss of an unfilled isophorone diamine/DGEBA coating can be three to five times higher than a mortar filled with 55–65 wt% fused alumina of particle size 0.25–0.50 mm. The use of a silane coupling agent such as 3-glycidoxypropyltrimethoxysilane at 0.3–0.8 wt% of filler is reported to reduce the Taber wear index by 20–40% because the covalent bond between silanol groups and glass or alumina surfaces resists particle pull-out. Surface roughness after abrasion, measured by contact profilometry per ISO 21920-2, shows that aggregate loss produces peaks and pits with Ra values above 5–15 µm in poorly coupled systems, whereas well-bonded filled systems retain Ra below 2–4 µm after 1000 Taber cycles. Abrasion debris from quartz-filled mortars contains fractured quartz particles below 10 µm, which can be quantified by laser diffraction per ISO 13320. The absence of free crystalline silica in the dust of aluminum oxide and silicon carbide filled systems is often a regulatory advantage under REACH and OSHA 29 CFR 1926.1153, which specifies an action limit of 25 µg/m³ for respirable crystalline silica. The choice of aggregate hardness is not the only variable: quartz has a Mohs hardness of 7, fused alumina 9, and silicon carbide 9–9.5. Silicon carbide filled mortars exhibit the highest abrasion resistance, but mix viscosity rises due to angular particle shape and high surface area, and trowel finishing becomes more difficult. Abrasion debris morphology also changes from rounded polymer-rich particles in neat resin systems to sharp mineral fragments in highly filled mortars, which influences the wear mechanism of opposing wheels and cleaning pads.

Substrate moisture and surface preparation determine whether the hardened mortar remains bonded under abrasive loads. Concrete slabs with moisture vapor emission rates above 3 lb/1000 ft²/24 h measured by ASTM F1869 or relative humidity above 75–80% measured by ASTM F2170 often require a moisture-control primer before mortar placement. Isophorone diamine-cured epoxy mortars applied to saturated surface-dry concrete can suffer interfacial cure inhibition, reducing pull-off adhesion measured by ASTM D4541 using 50 mm diameter aluminum dollies to below 1.0 MPa, while properly prepared and dried substrates yield 1.5–2.5 MPa or higher with cohesive failure in the concrete. Surface profiling per ASTM D4417 using replica tape or a depth micrometer should achieve CSP 3–5 for floor toppings, corresponding to a profile depth of 0.3–0.5 mm. Acid etching is not recommended for epoxy mortars because residual acids inhibit amine cure. Concrete pH per ASTM F710 should be below 10 before application. The placement of a primer layer at 0.2–0.4 L/m², followed by broadcast of 0.3–0.6 mm fused alumina at 2–3 kg/m² before mortar application, creates a mechanical lock that improves interlayer adhesion and prevents delamination under heavy-wheel traffic. Without this surface preparation sequence, wear testing of the topping may show localized peeling rather than uniform mass loss, and the Taber wear index value becomes dominated by adhesive failure rather than cohesive wear.

Because Water Contact Before Gelation Depresses Crosslink Density

When isophorone diamine-cured DGEBA mortar is exposed to water during the first 6–12 h of cure, primary and secondary amine groups react with carbon dioxide and water to form carbamate and bicarbonate species. This carbamation reaction reduces the number of amine protons available for oxirane ring opening. Infrared spectroscopy per ASTM E1252 may show a carbonyl absorption at 1,630–1,680 cm⁻¹ in the surface layer, while the underlying bulk material remains uncarbamated. The result is a gradient in crosslink density across the 0.5–1.0 mm surface, which lowers abrasion resistance and causes chalking. In accelerated weathering per ASTM G154 cycle 1, water condensation can produce surface whitening within 24–72 h if the mortar was not post-cured. To avoid this defect, floor topping installations in high-humidity environments often require post-curing for 24–48 h at 30–40 °C using forced-air heaters or low-pressure steam, followed by 7 days at 23 °C before heavy trafficking. The post-cure protocol raises the surface glass transition temperature above 120 °C. Without it, surface glass transition temperature may remain below 70 °C and Taber wear loss can double relative to the same formulation cured under dry conditions. Published data for exterior installations under combined freeze-thaw and wheel traffic is limited, but the standard conditioning protocols in ASTM D4060 and ASTM C501 require reporting temperature and relative humidity during specimen preparation because of this sensitivity.

Control of the Exotherm During Mass-Curing Is the Central Processing Conflict

The processing window for a 500 kg batch of isophorone diamine-cured DGEBA floor topping mortar is governed by heat transfer from the mixing vessel. The reaction kinetics accelerate rapidly with temperature, and a jacket temperature tolerance of ±2 °C around 20 °C is often specified. At a jacket temperature of 24 °C, the pot life can fall by 50% relative to operation at 18 °C. At jacket temperatures below 16 °C, initial mixed viscosity increases and aggregate wetting becomes incomplete, leading to weak boundary layers. Production-scale planetary mixers for such mortars are typically fitted with an 18.5 kW main motor, a 7.5 kW high-speed disperser, and a 2.2 kW vacuum pump, with vessel internal pressure controlled between 5–10 kPa absolute during final mixing. The disperser tip speed under load should be maintained between 18–25 m/s to deagglomerate fine filler without trapping air. Specific energy input per batch is commonly recorded in the range 0.3–0.6 kWh/kg. If the cooling jacket is not held at 18–20 °C, the core temperature can exceed 70 °C within 20 min for a fully charged 500 kg mixer, causing gelation in the vessel or discharge line and requiring mechanical removal of partially cured material. Twin-screw extruders with defined L/D ratios are not normally used for ambient cure mortars because thermoset gelation in the barrel can cause screw seizure. Instead, low-speed planetary mixing followed by vacuum discharge into transfer pumps is preferred. The temperature-rise profile measured by a thermocouple immersed in the batch should be logged at 1 min intervals and compared with the differential scanning calorimetry kinetic curve obtained by ASTM E1356. A deviation of more than 5 °C above the predicted isothermal mass temperature indicates poor jacket heat removal or a resin lot with higher reactivity, requiring rework or disposal. Because the abrasion resistance of the cured mortar is reduced by both premature gelation and incomplete mixing, the exotherm is a direct quality gate for surface wear performance.

In continuous production, batch-to-batch variance in epoxide equivalent weight, filler moisture, and aggregate particle size distribution shifts the relationship between mixing time and final wear resistance. Resin lots with an epoxide equivalent weight at the lower end of 188–192 g/eq generate a higher crosslink density when the same hardener dosage is used, but can also generate more exotherm. Filler moisture above 0.3 wt% measured by ASTM D280 or ISO 787-2 can introduce defects and lower pull-off adhesion. Aggregate median particle size drifting from 0.35 mm to 0.45 mm may reduce paste demand slightly, but coarse fractions above 2.0 mm can segregate during transport and produce uneven wear across a floor. Production records from mortar plants show that a daily check of filler moisture and resin epoxide equivalent weight is necessary to keep the formulated system within 21–24 phr hardener loading. A single-point deviation outside this range is not always visible at trowel application, but becomes measurable as a 20–40% increase in Taber wear index after cure. These observations are consistent with laboratory data showing that the relationship between formulation variables and wear loss is nonlinear near the critical pigment volume concentration and near stoichiometry.

The table below compiles representative values for quartz-filled isophorone diamine-cured DGEBA floor topping mortars tested under controlled laboratory conditions. Values are derived from typical industrial datasheet ranges and interlaboratory studies; absolute wear loss varies with aggregate type, post-cure schedule, wheel refacing procedure, and specimen preparation. The data illustrate the nonlinearity of wear behavior as filler loading approaches the critical pigment volume concentration.

Compliance assessment for industrial floor topping mortars uses a combination of mechanical, thermal, and surface methods. The following matrix summarizes the core test methods, equipment types, and typical acceptance criteria applied to isophorone diamine-cured DGEBA systems. The values are not absolute design guarantees; project specifications may require additional chemical exposure testing or in-situ adhesion verification after installation.

Chemical contact conditions can alter the ranking of abrasion-resistant mortar systems. A mortar that shows a low Taber wear index under dry room-temperature testing may lose more mass after immersion in dilute organic acids if the filler-matrix interphase is not fully cured. Testing according to ISO 175 with 10% acetic acid or 10% lactic acid solutions at 23 °C for 7 days is used to assess chemical attack on the polymer network and aggregate surface. Isophorone diamine-cured DGEBA systems generally resist alkali and aliphatic hydrocarbons, but prolonged contact with concentrated organic acids, some ketones, or chlorinated solvents can soften the matrix and increase wear loss under subsequent ASTM D4060 testing. In such cases, the combined chemical and abrasive service may require an additional 3–5 mm sacrificial wear layer or the substitution of quartz with chemically resistant fused alumina. The durability of the mortar is therefore not a single abrasion number but a matrix of wear resistance, adhesion, cure completeness, and compatibility with service fluids. Standardized wear tests performed without reporting the substrate moisture history, post-cure temperature, and aggregate type do not adequately differentiate floor topping failures that originate from interfacial delamination rather than surface abrasion.