DSM Somos ProtoGen™ 18420 is an epoxy-based photopolymer for laser stereolithography where thermal endurance, low moisture uptake, and stable dimensions after postcure are specified. The liquid resin is processed on standard 355 nm SLA platforms and is documented for build layer thicknesses in the 0.100 mm to 0.125 mm range, although the installed machine profile and recoating settings control the useful lower limit. The product is a cationic epoxide formulation rather than a radical acrylic; photoacid generation during laser scanning initiates ring-opening polymerisation, and the reaction continues after scanning by dark cure. Because green-state conversion is incomplete, the supplier specifies a combined ultraviolet postcure and heated oven cure. The notation UV postcure at HOC +3 refers to a 3-hour heated oven hold after UV exposure. Without this thermal segment, the published heat deflection response and dimensional stability are not reached.

Green-part cleaning has a greater influence on this resin than on many low-viscosity non-epoxy grades. Blind holes, internal lattice cells, and stepped bores retain liquid resin; residual photoacid can continue to catalyse local dark cure if the solvent flush is incomplete. Production lines typically use a two-stage wash in a supplier-approved or proprietary solvent such as tripropylene glycol monomethyl ether, followed by air displacement and drying at ambient temperature. The parts are then placed in a UV chamber with 365 nm UV-A sources; total exposure is best controlled by radiometric dose rather than cycle time because lamp age and reflector fouling degrade irradiance. The subsequent HOC +3 step is a forced-air oven soak. Thick sections should be ramped rather than dropped directly onto the soak temperature, because exothermic epoxide advancement can overshoot the setpoint and produce surface discoloration or interlayer cracking.

Reported mechanical and thermal property envelope for initial process planning

Representative values from supplier technical literature are summarised in Table 1. They are not batch-certified and must not replace lot-specific certification for mould inserts, functional test parts, or production fixtures. Tensile measurements follow ASTM D638-14; flexural measurements follow ASTM D790-17; heat deflection temperature follows ASTM D648-18; impact resistance follows ASTM D256-23; hardness follows ASTM D2240-15; density follows ASTM D792-20; and viscosity follows ASTM D2196-20. The property envelope places the material as a stiff, low-ductility epoxy resin. Tensile strength is commonly reported between 45 MPa and 60 MPa, tensile modulus near 3,000 MPa, and elongation at break below 5%. The low elongation means snap-fit features, living hinges, and high-strain clips are not appropriate without geometric modification or metal inserts.

The largest source of variability in these values is postcure completeness. HDT at 0.46 MPa is particularly sensitive to the thermal segment; parts removed from the oven before the core reaches soak temperature can test below the published envelope. Conditioning also matters. Water absorbed during cleaning or storage can reduce the dry HDT, so dimensional inspection should follow ASTM D618 conditioning and the supplier’s recommended drying protocol. The lower HDT at 1.82 MPa should be used for load-bearing fixtures or inserts exposed to melt pressure.

In high-humidity environments, cleaned green parts should be pre-dried before the HOC +3 segment when relative humidity exceeds 60%. Moisture in the green part can vaporise during thermal postcure, creating microvoids at interlayer boundaries. This is most visible in thick sections and in parts with large flat surfaces where layer interfaces are oriented perpendicular to the primary shrinkage gradient.

Where injection mould inserts are printed for low-pressure prototype moulds, the material can be used to evaluate gate placement and wall-thickness distribution, but tool life is limited when melt temperatures exceed the 1.82 MPa HDT. The epoxy surface can soften at the interface and develop compression set under repeated cycles. For short-shot or low-pressure moulding of polyolefins with melt temperatures below 100 °C, insert life may be acceptable for a limited number of cycles. Published data for this specific configuration is limited. The user should instrument the insert near the gate and record cavity pressure; if the local temperature approaches the lower HDT threshold, the cycle time should be extended or the insert redesigned with metal cooling channels.

Wind-tunnel test parts and aerodynamic fixtures exploit the sandable and machinable surface after full cure. Dimensional stability in the test envelope depends on keeping stagnation temperatures below the 0.46 MPa HDT only if stresses are low; for loaded load paths, the 1.82 MPa value governs. When polished and sealed, the low moisture uptake reduces change in weight between dry and humid environments. However, surface coatings should be tested for solvent compatibility because solvent can interact with residual epoxide groups.

What does the HOC +3 thermal segment change at the network level?

The UV postcure is necessary but not sufficient. During laser scanning, the cationic photoacid is generated and epoxide ring-opening begins, but vitrification quickly limits molecular mobility. The result is a green part that is dimensionally stable enough for handling but contains residual oxirane groups and constrained reactive sites. The heated oven cure at HOC +3 supplies thermal mobility above the current glass transition, allowing further chain extension and crosslink formation. The network then shifts toward higher crosslink density, increasing the heat deflection temperature and reducing solvent sensitivity. Omission of the thermal segment leaves a partially converted network with lower elevated-temperature resistance and a tendency toward slow dark polymerisation during storage. Slow dark cure can produce dimensional drift in assemblies because the part continues to advance after mating features have been machined or pinned.

The oven segment should be counted from part-core temperature, not from chamber air temperature. A forced-air oven with a timer and a separate thermocouple inserted in a sacrificial block of similar section thickness is a practical control. When processing large tools or mould inserts, the soak may need to be extended beyond the 3-hour nominal hold to allow the core to approach setpoint. Rapid temperature ramps are not recommended for sections above 10 mm; a stepped ramp of 20–30 °C per stage is used on some production lines to avoid exothermic overshoot. Published data for this specific configuration is limited, so each new chamber and part geometry should be qualified with a thermal audit.

A common failure observed on production lines is the formation of a tightly crosslinked outer skin during UV exposure before the core receives sufficient thermal cure. The skin can restrict shrinkage and lock in stress; later machining then relieves the stress unevenly, causing part bow. Another failure mode is local darkening when UV chambers leak shorter-wavelength emission; the resulting chromophore formation is not reversible. These process failures are controlled by mapping chamber irradiance, using a pyrometer or embedded thermocouple during oven validation, and maintaining lot-to-lot records of photoacid concentration.

When replacing a radical acrylic SL resin with this material on an existing machine, the operator should not carry over laser working-curve parameters. Epoxy systems can have a different penetration depth and critical exposure; the working curve \(C_d = D_p \ln(E/E_c)\) is less predictive because dark polymerisation contributes to gelation after the scan. Small features may build with lower exposure than acrylic systems, but overcure can close holes and fuse narrow gaps. A build platform-level calibration with test pins and thin walls is required before committing to production geometry.

In a typical wind-tunnel model or underhood test part application, the part can be assembled after full UV and thermal postcure with minimal immediate dimensional movement. However, any subsequent coating or paint bake should not exceed the lower HDT threshold unless the part is supported, because the epoxy network can soften under load at temperatures above the 1.82 MPa HDT. Unsupported bending loads at elevated temperature should be evaluated using ASTM D648-18 data rather than dry ambient flexural values.

When this material is compared with other stereolithography resins

The main differentiation is the epoxy cationic network. General-purpose SL resins based on acrylate or hybrid chemistry often have higher elongation and lower viscosity but less thermal resistance. DSM Somos ProtoGen™ 18420 is selected when the application is dominated by elevated-temperature stiffness and dimensional stability rather than impact ductility. Mineral-filled ceramic grades can offer higher modulus but tend to settle more readily in the vat, require more frequent recirculation, and may demand more aggressive recoating. The unfilled nature of this epoxy material keeps viscosity in the 250–400 cP range, which simplifies vat refilling and reduces separation of dense fillers during long builds. However, the low notched Izod values mean it is not a direct replacement for ABS-like resins such as DSM Somos NeXt in snap-fit or impact-loaded housings.

Compared with high-temperature unfilled resins that require only UV postcure, the presence of the HOC +3 thermal segment introduces additional equipment and scheduling constraints. If a production cell does not have a forced-air oven with sufficient airflow and part-core temperature monitoring, the material cannot be processed to its published thermal envelope. The resin also has a finite dark-cure window; cleaned green parts should not be held for extended periods before postcure because partial conversion continues at room temperature and can alter dimensions before the intended thermal cycle begins.

System qualification for this resin is more involved than for low-viscosity unfilled acrylics. On a 355 nm galvanometric platform, laser power is mapped across the entire build surface because the depth-of-cure equation is influenced by local intensity distribution. A build-platform calibration using a test bar array with thin ribs and pins should be performed whenever a new vat lot is introduced. The test parts are measured immediately after cleaning and again after full UV and HOC +3 postcure; the difference quantifies the batch-to-batch dark-cure contribution. If the postcure shrinkage shifts significantly, the build parameters should be reviewed before releasing the lot into production.

Vat maintenance requires the resin to be filtered through a fine paint or resin sieve after a failed build to remove gel particles and cured debris. Unlike mineral-filled grades, settling is minimal, but the epoxy can stratify slightly if the vat is held above 30 °C for extended periods. The recoating blade should be checked for wear; a damaged blade can produce layer thickness variation and increase the exposure-to-cure mismatch across the platform. Operators record ambient humidity and vat temperature in the build log because viscosity changes affect the wetting film and can shift the working curve.

The compliance status of the liquid resin and cured article must be verified against the current safety data sheet and the applicable jurisdiction. Supplier documentation typically addresses chemical control under REACH Regulation (EC) No 1907/2006 and the RoHS Directive 2011/65/EU, but those documents do not automatically qualify a finished printed assembly for end-use electrical equipment. Flammability classification, smoke density, and gas release for aircraft interiors are separate certifications and are not implied by the base resin data.

The operational boundary for this product is defined by the need for combined UV and thermal postcure, low impact ductility, and sensitivity to amine-contaminated solvents. Basic amine-based cleaning agents or hardeners should be avoided because they can neutralise the cationic photoacid and prevent full epoxide conversion. Alcohol rinses should be short because prolonged immersion can swell the green epoxy network and promote microcracks at layer interfaces. Vat life is not indefinite; viscosity rise, moisture uptake, and batch-to-batch photoacid content should be monitored with a rotational viscometer and recorded in the build log. If a layer is skipped or the recoater speed is increased beyond the resin profile, the resulting thin film may not wet the previous cured layer sufficiently, producing delamination that is not corrected by postcure.