What Drives Epichlorohydrin Cost: Glycerol, Propylene and the Route Switch

Across epoxy resin, epichlorohydrin rubber, synthetic glycerin and wet-strength resin supply chains, the delivered cost of epichlorohydrin is not set by a single raw material but by the interplay among propylene valuation, chlorine and hydrogen chloride balance, caustic soda demand and crude glycerin availability. The propylene-based allyl chloride route historically supplied the marginal tonne because it represents the larger share of installed capacity. The glycerol-to-ECH route introduced a competing C3 platform from biodiesel and oleochemical processing, allowing producers to decouple ECH from steam cracker and refinery propylene balances when glycerin economics are favorable. The resulting route switch is not instantaneous and is constrained by chlor-alkali integration, salt effluent permits, purification loads and materials of construction. The following sections examine what drives epichlorohydrin cost across the propylene chain, the glycerol route, energy and capital differences, and the operational thresholds that trigger substitution.

Why Does the Propylene-to-Allyl Chloride Chain Remain the Marginal Cost Setter?

The propylene route converts propylene to allyl chloride through high-temperature substitutive chlorination. The reactor is typically operated at approximately 500 °C and 0.1–0.2 MPa with a significant propylene excess to suppress addition reactions and improve allyl chloride selectivity. Allyl chloride is then converted with hypochlorous acid to a mixture of 1,3- and 1,2-dichloropropanol, and the dichloropropanol mixture is dehydrochlorinated with lime or sodium hydroxide to epichlorohydrin. The sequence can be represented as:

• C₃H₆ + Cl₂ → C₃H₅Cl + HCl
• C₃H₅Cl + HOCl → C₃H₆Cl₂O
• 2 C₃H₆Cl₂O + Ca(OH)₂ → 2 C₃H₅ClO + CaCl₂ + 2 H₂O

Within this route, propylene and chlorine typically form the largest variable-cost block. Propylene price discovery follows steam cracker output, refinery fluid catalytic cracking utilization and propane dehydrogenation economics, while chlorine cost is tied to membrane-cell electricity consumption and caustic soda selling prices. An ECH unit without captive chlorine and caustic must purchase both streams, exposing production cost to simultaneous petrochemical and electrochemical shocks. Integrated mass balances commonly show chlorine intensity in the range of 1.4–1.6 t per tonne ECH when hypochlorous acid generation, chlorohydrin losses and byproduct neutralization are included. Lime-based saponification adds calcium chloride effluent treatment and steam demand; caustic-based saponification consumes approximately 0.7–1.0 t of sodium hydroxide per tonne ECH and generates sodium chloride brine.

The marginal cost contribution of the allyl chloride route also depends on byproduct hydrogen chloride valorization. The chlorination step co-produces HCl that can be routed to an oxychlorination reactor producing 1,2-dichloroethane in a vinyl chloride complex. If the ECH unit is not integrated with ethylene dichloride or vinyl chloride monomer production, HCl becomes a low-value or negative-value stream. On commercial allyl chloride reactors, tube-skin temperature control and metallurgy are critical; operating data from manufacturing lines show that metal-catalyzed byproduct formation increases sharply when reactor outlet temperatures exceed design limits, reducing allyl chloride selectivity and increasing coking in the effluent heat recovery system. These site-specific effects mean that the propylene route remains the marginal cost setter during periods of strong propylene and chlorine demand, but its cost floor rises when HCl disposal or caustic integration is unavailable.

Glycerol-to-epichlorohydrin units alter the variable cost structure because the C3 skeleton is supplied by glycerin rather than propylene, but the benefit depends on crude glycerin quality, hydrogen chloride balance and salt disposal. The two-stage reaction sequence converts glycerol with HCl gas in the presence of a carboxylic acid catalyst at 105–130 °C and near-atmospheric pressure to a dichloropropanol mixture, which is then dehydrochlorinated with alkali to ECH and chloride salt. Stoichiometrically, pure glycerol yields close to 1.0 t ECH per tonne glycerol because the molecular masses are nearly equal; with 80 wt% crude glycerin from biodiesel, the feed requirement rises to approximately 1.3–1.5 t per tonne ECH after water removal, MONG rejection and process losses. Vendor data on continuous GTE trains indicate that feedstock pre-treatment and evaporator fouling are the main operability constraints when switching from refined glycerol to crude grades. The overall chemistry may be summarized as:

• C₃H₈O₃ + 2 HCl → C₃H₆Cl₂O + 2 H₂O
• C₃H₆Cl₂O + NaOH → C₃H₅ClO + NaCl + H₂O

Unlike the propylene route, the glycerol pathway does not require an allyl chloride reactor or hypochlorous acid generation. However, it adds saline evaporation, catalyst recovery, thin-film evaporation for heavy organic rejection, and feed drying. Hydrogen chloride is consumed in the hydrochlorination step, and caustic soda or lime is consumed in the dehydrochlorination step. If sodium chloride brine is returned to a chlor-alkali cell room, organic contamination must be reduced by activated carbon, chemical oxidation or thermal treatment. The economics therefore shift from propylene and chlorine purchases to glycerin, HCl gas, caustic, steam and brine treatment. Route selection therefore becomes an electrochemical and environmental balance as much as a feedstock-price decision.

Glycerol Hydrochlorination Capital Intensity and Salt Effluent Constraints

Capital cost comparisons between propylene-based and glycerol-based trains are complicated by integration boundaries. A glycerol unit substitutes storage and purification for high-temperature corrosion-resistant allyl chloride reactors, but adds vacuum distillation capacity for water removal from the glycerol feed, a hydrochlorination loop in hot wet HCl service, and salt management units. Materials of construction in the hydrochlorination section typically include glass-lined steel, graphite blocks, tantalum heat exchangers and high-nickel alloys such as C276. The selection depends on chloride stress corrosion cracking risk in distillation overheads and the presence of carboxylic acid catalyst at elevated temperature.

Published data for this specific configuration is limited because licensed glycerol-to-ECH processes are proprietary, but engineering studies available from process licensors show that the glycerol route generally reduces raw material cost exposure when propylene is priced above glycerin but increases site complexity if brine electrolysis is not available. Steam consumption is a decisive factor because water is both a reaction product and a feed impurity. Multi-effect evaporation of dilute salt streams can consume 6–12 t of low-pressure steam per tonne ECH in standalone units, depending on the number of effects and the final brine concentration target. In contrast, propylene-based saponification also requires steam stripping but frequently operates with fewer evaporation stages because calcium chloride can be disposed or sold at higher concentration. The actual energy gap between the two routes is therefore site-specific and depends on the availability of low-pressure steam from integrated cogeneration.

The table indicates that the route switch is not a single-variable choice. A producer with surplus chlorine and ethylene dichloride capacity will likely retain propylene-based ECH even if glycerin is temporarily cheap, because the co-product HCl and chlorine logistics are already sunk. A biodiesel-integrated producer with excess crude glycerin and access to low-cost HCl gas may run a glycerol unit despite higher steam load, provided that brine disposal or chlor-alkali recycle can absorb the sodium chloride effluent.

Crude Glycerol Impurities Limit Feedstock Flexibility in Biodiesel-Integrated Plants

Biodiesel-derived crude glycerin is not a uniform feedstock. Depending on transesterification catalyst and upstream oil quality, crude glycerin can contain glycerol in the range of 60–85 wt%, water at 5–15 wt%, methanol at 0.5–3 wt%, and ash or matter organic non-glycerol at 5–10 wt%. Methanol must be recovered to control flash point and to avoid side reactions in the hydrochlorination reactor. Ash and organic salts increase catalyst consumption and foul reboilers; continuous GTE units typically require thin-film evaporators or wiped-film evaporation whenever MONG exceeds approximately 5 wt%. Production-scale experience from biodiesel-integrated sites shows that reboiler cleaning frequency and catalyst makeup are the main operational differences between refined glycerol and crude glycerin campaigns. The pre-treatment train may include acidulation, vacuum drying, distillation and activated carbon adsorption. These operations reduce the basic raw material cost advantage of crude glycerin when the feed is heavily contaminated, but they permit the GTE unit to accept lower-value glycerin streams that would otherwise require disposal or further refining.

Glycerol route reliability is therefore tied to biodiesel plant operation, not only to glycerin market price. If biodiesel plants reduce run rates or increase variety in feedstock oils, crude glycerin composition can change within one batch campaign. The ECH unit must then adjust acidulation, distillation and catalyst addition rates to hold dichloropropanol selectivity. This linkage creates a different operational risk profile from the propylene-based route, where feedstock quality is controlled by refinery and steam cracker specifications but supply is tied to petrochemical cycles. The route switch decision thus involves feed impurity variability, not just price alone.

When the Route Switch Alters Chlor-Alkali and Epoxy Resin Feedstock Exposure

Process economics are often summarized through the propylene-to-glycerin price ratio, but the actual switching threshold is a matrix of co-product credits, electricity prices and environmental permits. A propylene-based ECH unit consumes chlorine and produces HCl, which can be returned to a vinyl chloride oxychlorination reactor. A glycerol-based unit consumes HCl and produces brine, which can be returned to the membrane cell only if organic impurities are below cell-room limits. At sites with surplus caustic demand and limited ethylene integration, the glycerol route can reduce dependence on purchased chlorine. At sites with high electricity prices and no brine electrolysis, the glycerol route may be constrained by the cost of purchased hydrogen chloride and the disposal cost of sodium chloride.

Independent of route, epoxy resin producers require consistent oxirane reactivity and low chlorinated impurity content. Commercial epichlorohydrin specifications commonly include purity of 99.9 wt% or higher, water content by ASTM E1064 not exceeding 100 ppm, and APHA color by ASTM D1209 not exceeding 15. Route-specific impurities differ: propylene-derived material may contain traces of allyl chloride and chloropropenes, while glycerol-derived material may carry traces of acrolein, glycidol or chloropropanediol. These differences can influence epoxy resin catalyst consumption and should be managed through distillation column controls and finish purification. The ECH molecule itself has a normal boiling point of 115.6 °C, a density of 1.18 g/cm³ at 20 °C and a closed-cup flash point of 31 °C; storage vessels therefore require temperature control, inert gas padding and secondary containment regardless of production route.

Published market cost models often simplify the route decision to a propylene-to-glycerin ratio, but plant-level experience indicates that the effective threshold shifts with the chlor-alkali operating point. A chlor-alkali complex running at maximum caustic production has surplus chlorine, favoring the propylene route because elemental chlorine can be consumed internally. A biodiesel-integrated site with limited steam and no HCl consumer can run the glycerol route only if crude glycerin is pre-treated and saponification brine is accepted by existing waste treatment or chlor-alkali units. Process control in both routes is dominated by reactor selectivity and distillation purity. Allyl chloride reactors require continuous feed ratio control to limit polychlorinated byproducts and coking. Glycerol hydrochlorination requires water removal to shift equilibrium toward dichloropropanol while avoiding excessive catalyst decomposition and acrolein formation. Because ECH is classified as a flammable liquid and acute toxic substance under the Globally Harmonized System, shipping and storage add non-feedstock costs that become more visible when a route switch increases import dependency or lengthens supply lines. These logistics costs do not alter the fundamental chemistry, but they influence delivered cost and inventory requirements at epoxy resin plants and downstream consumers.