Epichlorohydrin (ECH; CAS 106-89-8; 2-(chloromethyl)oxirane; molar mass 92.52 g mol⁻¹) is a bifunctional epoxide monomer used in liquid and solid epoxy resins, polyepichlorohydrin elastomers, epichlorohydrin-ethylene oxide copolymers, polyamide wet-strength resins, and cationic polyacrylamides. The product specification for ECH is not a single assay value but an impurity profile centered on chlorohydrin by-products that persist from the chlorohydrination route or arise during storage and handling. The term Epichlorohydrin Impurity Profile: Chlorohydrin By-products and Polymerization Risk refers to the coupled hazard in which residual dichloropropanols and monochloropropanediols increase hydrolyzable chloride in downstream epoxy resins while simultaneously functioning as moisture-adsorption centers that promote acid generation. The resulting acid species, together with dissolved iron from carbon steel handling, are initiating sites for oxirane ring-opening oligomerization and exothermic bulk polymerization. Merchant ECH is therefore refined by two-stage fractional distillation and stored under moisture-excluding nitrogen pads, with strict limits on water, color, density, and individual chlorohydrin isomers. The operational boundary between acceptable ECH and polymerizing ECH is set not by the epoxide concentration alone but by the concentration of ionic initiators, water, and temperature history.

Which Chlorohydrin Adducts Dominate the ECH Impurity Envelope?

In the allyl chloride route, chlorine-water addition to allyl chloride gives a mixture of regioisomeric dichloropropanols. The symmetrical 1,3-dichloro-2-propanol is favored at low chlorination temperatures and high chloride concentrations, while elevated chlorination temperatures increase the fraction of 2,3-dichloro-1-propanol. These two isomers are the principal chlorohydrin by-products carried into the saponification reactor. Saponification with sodium hydroxide or calcium hydroxide at 80–95 °C dehydrochlorinates the dichloropropanols to ECH. Incomplete conversion, emulsion layers in the brine phase, and entrainment leave residual dichloropropanols in crude ECH. The relative volatility between ECH and dichloropropanols is high, but separation in a vacuum-distilled final column is limited by reboiler fouling rather than thermodynamic selectivity. The boiling points of 1,3-DCP and 2,3-DCP are approximately 174 °C and 182 °C at 101.3 kPa, respectively, whereas ECH boils at 115–116 °C. A third chlorohydrin, 3-chloro-1,2-propanediol (3-MCPD), is not a direct saponification intermediate but is generated by hydration of ECH in the aqueous workup and by hydrolysis of residual dichloropropanols.

In continuous plants, the chlorohydrin saponification reactor is a steam-stripped distillation column rather than a batch stirred tank. The pH is maintained in the range 10–11 with lime slurry to avoid alkaline ECH polymerization. A local excess of sodium hydroxide above pH 12 initiates anionic oligomerization at the phase boundary; this produces a viscous polyepichlorohydrin rag layer that stabilizes the brine-organic emulsion and increases dichloropropanol carryover. Operators control this by staged alkali injection and by maintaining the saponification temperature in the bottom zone below 95 °C. Because ECH hydrolyzes to 3-MCPD in the aqueous phase, residence time in the stripper bottoms is minimized. Crude ECH is condensed and decanted; excess brine saturated with calcium chloride is recycled. Any calcium chloride carryover into the crude ECH can deposit in the reboiler and create local hot spots. Filtration of crude ECH through a coalescer before distillation reduces inorganic salt loading and reboiler scaling.

Quantitative impurity profiling of commercial ECH relies on gas chromatography with flame ionization detection on polar polyethylene glycol or cyanopropylphenyl capillary columns. The limiting variable in routine QC is injection-port artifact generation: hot split/splitless injectors above 200 °C can dehydrochlorinate dichloropropanols to ECH, producing false low chlorohydrin values and false high ECH assay. Cold on-column injection or a programmable temperature vaporizing inlet held at or below 180 °C is therefore specified. Water content is determined by coulometric Karl Fischer titration following ASTM E203; color is measured against platinum-cobalt standards following ASTM D1209; density is measured by ASTM D4052 at 20 °C. Industrial ECH specifications such as GB/T 13097 are supplemented by downstream limits for hydrolyzable chloride because epoxy resin producers correlate hydrolyzable chloride with resin color, reactor corrosion, and inconsistent advancement stoichiometry. The analytical lower limit for individual dichloropropanol isomers in ECH by modern GC-FID with cold injection is approximately 10 mg kg⁻¹; routine production control limits are typically higher because of sampling and transport stability. To avoid hydrolysis during sampling, glass bottles are dried, purged with nitrogen, and sealed with PTFE-lined caps; samples are not stored in polyethylene containers because chlorinated impurities could extract plastic additives and bias color readings.

Commercial ECH quality is bounded by residual chlorohydrin content rather than total chlorine

ECH itself contains 38.3 wt% chlorine, so total chlorine measurements do not distinguish monomer from chlorohydrin impurities. The operationally meaningful parameter is hydrolyzable or saponifiable chlorine derived from residual dichloropropanols and chloropropanediols. Merchant-grade ECH certificates of analysis commonly list ECH assay at or above 99.9 wt% on a dry basis, water at or below 0.005 wt%, and individual dichloropropanol isomer contents at or below 0.05 wt%. These limits are not independent: a batch with 0.05 wt% 1,3-DCP and 0.05 wt% 2,3-DCP introduces roughly 1,000 mg kg⁻¹ combined chlorohydrin, which can consume sodium hydroxide during liquid epoxy resin advancement and raise the hydrolyzable chloride of the finished resin above the 500 mg kg⁻¹ limit used in many industrial resins. In a molten bisphenol A epoxy resin reactor, ECH is charged in molar excess and the ring-closing dehydrochlorination is carried out with aqueous sodium hydroxide. Residual dichloropropanols compete with the intermediate chlorohydrin ether for alkali, form additional epichlorohydrin-derived impurities, and can leave bound non-hydrolyzable chloride in the resin backbone. Production-scale batch records show that when dichloropropanol isomer contents are not held constant, the epoxy equivalent weight and saponifiable chloride of the final resin shift even if the ECH assay remains unchanged.

Normal boiling points in the following impurity table are at 101.3 kPa; operational impacts are observed qualitatively in production refinement, epoxy resin advancement, and storage stability.

When Trace Water and Iron Residues Accelerate Ring-Opening Oligomerization

The oxirane ring in ECH is strained and susceptible to both cationic and anionic ring-opening polymerization. Cationic initiation is the dominant storage risk because residual hydrogen chloride from chlorohydrin decomposition and ferric chloride from carbon steel corrosion generate terminal oxonium sites. Propagation occurs by nucleophilic attack of another ECH oxygen on the activated ring, forming polyepichlorohydrin chains. The bulk polymerization is exothermic and can be self-accelerating once the temperature reaches the threshold for acid formation from residual chlorohydrins. Water is not an initiator but is a chain-transfer agent and hydrolysis reactant. Under neutral aqueous conditions, ECH hydrolyzes to 3-MCPD at a rate that is low at ambient temperature but increases with temperature and chloride ion activity. In the presence of free hydrochloric acid, hydrolysis and oligomerization occur simultaneously, and the mixture can reach a visible gel point without an external heat source. This is the reason storage specifications keep water below 0.005 wt% and require nitrogen blanketing or desiccant vent dryers on tanks.

The kinetic distinction between hydrolysis and polymerization is important. Hydrolysis of ECH to 3-MCPD consumes water and forms a vicinal diol that can act as a chain-transfer agent in cationic polymerization. The diol does not necessarily initiate polymerization but lowers the average molecular weight of any polyepichlorohydrin formed and increases the amount of solvent-soluble oligomers. Free acid generated by hydrolysis of residual dichloropropanols then catalyzes further ring opening. This autocatalytic sequence explains why wet ECH with high dichloropropanol content can remain within specification for assay and moisture at the start of storage but later drift into a polymerizing regime. For this reason, ECH held in storage for more than 30 days is resampled for water, dichloropropanols, and acid number before use in epoxy resin batches. The acid number is usually kept below 0.01 mg KOH g⁻¹ by use of an acid scavenger or by immediate consumption.

Dissolved iron is controlled by specifying 316L stainless steel or lined carbon steel, and by filtering through 0.2 µm elements during unloading. Zinc-galvanized fittings are not used because zinc chloride is a Lewis acid initiator. Equipment experience in ECH distillation shows that tube-sheet hot spots above 120 °C accelerate polyepichlorohydrin deposition and reduce run length. Vacuum distillation is therefore preferred to keep bottoms temperatures below 120 °C, and reboilers are designed with low skin temperatures rather than high-pressure saturated steam.

Differential scanning calorimetry of ECH samples drawn from storage tanks and road tankers is used as a screening test for active polymerization initiators. A sample is sealed in a high-pressure gold-plated crucible and ramped from 30 °C to 250 °C at a scan rate of 5 °C min⁻¹; an exotherm onset below 100 °C is considered a rejection criterion for long-term storage. Accelerating rate calorimetry or adiabatic pressure Dewar testing supplies time-to-maximum-rate and pressure-rate data for emergency relief design. Because polyepichlorohydrin can form as a dispersed gel before visible viscosity increases, inline viscometers do not always provide early warning. Storage tanks are therefore fitted with low-point and near-surface sampling points, and samples are checked for filterable gel, pH shift, and copper strip discoloration. A low-point sample with elevated iron and water relative to the top sample indicates stratified contamination that should be removed before the tank is circulated.

Pressure Relief Sizing and Inhibitor Compatibility Limits

Runaway ECH polymerization is treated as a two-phase relief scenario in emergency vent sizing. The design basis assumes a credible Lewis acid contamination event, with the initiating temperature at or below the normal storage maximum. DIERS-based homogeneous two-phase vent sizing is required because the polymerizing mixture can foam and expel liquid droplets, reducing the effective mass capacity of a vapor-only relief device. A rupture disk upstream of a pressure relief valve is standard when gel formation could block the valve inlet. Storage vessels with nitrogen pads typically use conservation vents set at 3.5 kPa gauge, but the emergency relief device is sized for the external fire case and the polymerization case rather than normal breathing. Pressure relief capacity is based on adiabatic calorimetry of contaminated ECH, not on hydrocarbon vapor-pressure correlations. No single generic relief size applies across terminals; published data for specific ECH adiabatic rates is often site-specific because initiator identity and concentration vary.

Epichlorohydrin is not usually stabilized by phenolic free-radical inhibitors because the principal degradation route is ionic. Acid-scavenging stabilizers and dry storage are used instead. Any proposed stabilizer must not contain primary or secondary amines, because amines initiate anionic ECH polymerization and generate chlorohydrin adducts. Organotin mercaptides used in vinyl chloride polymerization are incompatible. Peroxide inhibitors are ineffective against acid-initiated ring opening but may be used to control allyl chloride-derived peroxide species in recovered light ends. The operational boundary for storage is therefore set by three simultaneous conditions: water below 0.005 wt%, total iron below 0.5 mg kg⁻¹, and storage temperature below 30 °C. Higher temperatures can be tolerated only for short-duration transit, provided the ECH is consumed promptly and the vessel is free of rust, water pockets, and caustic residues.

At a production-scale liquid epoxy resin facility, ECH is unloaded from tank trucks through a positive-displacement pump and a 0.2 µm filter into a nitrogen-blanketed day tank. The transfer line is purged with nitrogen after unloading to prevent moist air ingress. The day tank is equipped with a desiccant dryer on the conservation vent to maintain water below 0.005 wt% during the campaign. Batch charging of ECH to the resin reactor is performed by mass flow meter with temperature compensation; the reactor is purged with nitrogen and the ECH is added before solid bisphenol A to avoid hot bisphenol A initiating oligomerization. The reactor jacket is held below 60 °C during the initial ECH dissolution phase, and caustic is added stepwise to control the exotherm and prevent local pH excursions above 12. After reaction, excess ECH is stripped under vacuum at a bulk temperature below 120 °C; the recovered ECH is returned to the day tank only if water and dichloropropanol contents are within the same limits as fresh feed. This closed-loop use reduces polymer risk but requires continuous impurity monitoring because recycled ECH can accumulate light ends and acid species.