DSM Somos ProtoGen™ 18420 is an epoxy-based liquid photopolymer supplied for 355 nm solid-state laser stereolithography. The material is identified by the product model number 18420 and is formulated for a two-stage cure sequence: laser-induced crosslinking during the build, followed by separate ultraviolet and thermal postcure steps. In the liquid state, the resin is a low-viscosity cationic epoxy; after full cure, it becomes a rigid, high-modulus thermoset. The final mechanical response is therefore not determined by the liquid formulation alone, but by the entire process chain, including vat temperature, layer thickness, laser energy dose, cleaning solvent, drying time, UV chamber irradiance, and oven ramp rate. Fully UV- and thermally postcured specimens conditioned at 23 ± 2 °C and 50 ± 5 % RH per ASTM D618 exhibit the supplier-published typical property ranges in Table 1. These ranges are not minimum batch specifications and must be confirmed against the lot-specific certificate of analysis.

How Does ProtoGen 18420 Differ from Impact-Modified and Acrylate SLA Resins?

ProtoGen 18420 belongs to the class of cationic epoxy SLA resins. Unlike acrylate and methacrylate resins that rely primarily on radical propagation, epoxy rings continue to react during dark conditions after the laser has moved to the next layer. This dark cure produces more complete conversion in thick sections but also creates a longer effective reaction time. Compared with impact-modified SLA epoxies that may exhibit notched Izod values above 40 J/m and elongation at break above 10 %, ProtoGen 18420 has a lower energy-absorption capacity before fracture. The trade-off is a tensile modulus above 2,400 MPa and an HDT at 0.46 MPa above 95 °C after full postcure. Acrylate resins often have lower viscosity and faster working curves; however, their thermal deformation resistance at equivalent filler content is generally lower, and moisture uptake can produce measurable dimensional change under high humidity. The data do not support selection of ProtoGen 18420 for snap-fit or impact-sensitive parts; its mechanical response is closer to a rigid thermoset than to an ABS-like SLA resin.

Thermal Deflection and Flexural Modulus Under ASTM D648 and ISO 75 Conditions

Heat deflection temperature is measured on a flat specimen under a defined surface stress, not as a maximum continuous-use temperature. Under ASTM D648-18 and ISO 75-2:2013, specimens of rectangular cross-section are immersed in a heat-transfer medium and heated at 2 °C/min while loaded to either 0.46 MPa or 1.82 MPa. For fully postcured ProtoGen 18420, the lower-stress HDT value falls in the 95–105 °C range, while the higher-stress value falls in the 82–90 °C range. The drop between the two stresses is a direct consequence of the crosslinked epoxy network; the material retains stiffness through the glass transition onset but deflects more rapidly as network segmental mobility increases. Because the HDT test measures deflection of a small bar, it should not be used alone for load-bearing flanges or press-fit inserts. Designers should combine HDT with flexural modulus and creep data generated on the actual section thickness.

In flexural testing per ASTM D790-17 or ISO 178, fully postcured samples typically fail in a brittle mode at low strain. Flexural strength values are sensitive to surface condition, build orientation, and postcure depth. Specimens built parallel to the build plane often show tensile-dominated failure and lower strengths than those tested perpendicular to the layer plane; this anisotropy is a known limitation of laminated stereolithography structures and must be accounted for in finite-element models.

Wind tunnel model production and investment casting pattern fabrication subject ProtoGen 18420 to opposing process requirements. In aerodynamic testing, the epoxy network provides dimensionally stable thin-wall shells and elevated heat deflection under moderate stagnation temperatures, but the low notched impact resistance limits assembly and disassembly cycles. Foundry burnout trials require slow ramps through the ceramic shell thermal expansion range; published data for this specific configuration is limited, so maximum pattern wall thickness, shell permeability, and burnout schedule should be established with sacrificial test patterns. Because the cured epoxy has low ash residue only when burnout is complete, incomplete burnout can leave carbonaceous defects in castings. Ventilation and personal protective equipment are required during burnout because decomposition products of epoxy networks include carbon monoxide and low-molecular-weight organics. The liquid resin should not be allowed to contaminate water systems; disposal must follow the safety data sheet and local regulations.

When Thermal Postcure Is Omitted, What Property Cliff-Edges Appear?

Green-state ProtoGen 18420 parts contain unreacted epoxy groups, active cationic species, and photoinitiator residues. If thermal postcure is omitted, the network remains partially crosslinked and the 0.46 MPa HDT may remain below 60 °C, with flexural modulus 15–25 % below the fully postcured value. The resin’s cationic chemistry continues to propagate in the dark, so the degree of conversion depends on time, temperature, and section thickness. A delay of 24–48 h between green part cleaning and oven cure can produce nonuniform dark cure: thick sections retain exothermic heat and crosslink further than thin walls, creating internal stress gradients. When the part is subsequently heated, differential shrinkage between skin and core manifests as bowing, twist, or delamination. This behavior is a known process conflict in production environments where parts are built over a weekend and postcured on Monday. Operators should either postcure immediately following the standard solvent-drying step or store green parts at controlled low temperature to slow dark cure, and should verify geometric fixtures during postcure to prevent creep collapse of unsupported walls.

UV and Thermal Postcure Protocol Bounds

Postcure uniformity is determined by UV irradiance, spectral output, temperature ramp rate, oven airflow, and part packing density. Supplier process sheets typically specify UV exposure in the 320–400 nm band. A representative cycle for sections near 3 mm is 30 min per side in a UV chamber followed by 2 h at 80 °C in forced air. Thicker sections require longer thermal soak but not necessarily higher UV dose; excessive UV exposure can cause surface overcure and embrittlement while the core remains undercured. Table 2 compares indicative mechanical behavior across the green, UV-only, and UV-plus-thermal states. The values are presented to illustrate the process-dependent property gradient rather than as independent specification limits.

Green parts are typically cleaned in tripropylene glycol monomethyl ether or isopropanol within a controlled wash station. Solvent retention at layer interfaces can cause pore formation during thermal postcure; therefore, a forced-air drying step of 30–60 min at 25–35 °C is used before UV exposure. Excessive solvent soak, particularly longer than 10 min in bulk solvent, can absorb into the low-crosslink-density green network and lower local glass transition; this produces chalky surfaces after cure. Ultrasonic cleaning is not recommended for thin walls because cavitation energy can initiate microcracking at layer interfaces. Acetone and chlorinated solvents are incompatible with the uncured resin and can cause stress cracking of green supports.

If a Production Line Switches from Low-Viscosity Acrylates to ProtoGen 18420, Recoater Calibration Becomes Critical

On 355 nm solid-state laser machines with nominal beam diameters of 0.15–0.25 mm and laser powers in the 100–200 mW range, the working curve of ProtoGen 18420 generally requires a higher energy dose per unit volume than low-viscosity acrylate resins. To maintain adequate depth of cure, vat temperature is typically controlled between 28 °C and 32 °C, which lowers the liquid viscosity into the 220–280 mPa·s range. Layer thickness is normally set at 0.05 mm or 0.10 mm; 0.15 mm is possible only after laser power and resin age are validated because cured layer thickness and green modulus change with resin batch and beam profile. The recoater blade gap should be re-established when switching resin families because the higher viscosity and different wetting behavior alter the resin film formed over the previous layer. A blade gap that is too wide causes wavy surfaces and fresh resin starvation on down-facing regions; a gap that is too narrow can drag or smear partially cured regions and introduce layer shifting. Field observations on production SLA platforms show that level detection sensors may require offset adjustment because the meniscus shape and surface tension differ from acrylate formulations. Dedicated vats and recoater blades are recommended.

Machining and support removal are performed after UV-only or after thermal postcure depending on part geometry. When supports are removed from fully thermally postcured ProtoGen 18420, the low notched impact resistance causes a high incidence of fracture at contact points if the support tips are not pre-sanded or if cutters are dull. Heating parts to 35–40 °C before support removal reduces notch sensitivity and lowers the stress required to separate supports. CNC machining of the cured epoxy produces fine particulate; local exhaust and wet sanding are required to prevent airborne epoxy dust. Tool wear is high compared with ABS due to the glassy network; carbide burrs and diamond-coated abrasives are preferred. The material can be bonded using epoxy structural adhesives after surface abrasion and solvent wipe with isopropanol, but joint strength is limited by the low peel resistance of the base resin.

Build orientation changes both mechanical performance and postcure response. Specimens built flat on the build platform typically show higher flexural strength but more curl at free edges; vertical builds show better z-axis dimensional fidelity but lower interlayer tensile strength. Published data for this specific configuration is limited across all orientations, so pre-production builds should include tensile and flexural bars in the same orientation as the part. The anisotropic response is less than glass-filled nylon but remains measurable: interlayer tensile strength is commonly 10–20 % lower than in-plane tensile strength. This must be included when designing pressure-retaining parts or bolted joints.

Quality control specimens should be built in the same orientation, on the same substrate location, and with the same layer thickness as production parts. Tensile bars per ASTM D638-14 Type I and flexural bars per ASTM D790-17 should be postcured alongside parts and conditioned for 24 h at 23 ± 2 °C and 50 ± 5 % RH before testing. Dimensional inspection should include a thermal conditioning step because the epoxy network continues to relax during early thermal cycling. The resin is incompatible with amine-based post-treatment coatings and should not be combined with reactive diluents intended for acrylate resins; such additions can disrupt cationic propagation and produce tacky surfaces or brittle networks. The liquid material should be stored in sealed light-protective containers away from heat and moisture. If the resin has been warmed above 35 °C for more than a few days, viscosity and cure speed should be checked against a fresh lot before production use.