In industrial hydrogen chloride synthesis, chlorine and hydrogen are combusted in a water-jacketed graphite chamber, where the gas-phase reaction between hydrogen and chlorine proceeds to near completion under a controlled excess of hydrogen. The graphite liner operates between a hot-face boundary layer that can exceed 1,200 °C and a water-cooled back face held below 80 °C, creating a radial thermal gradient that approaches 100 °C/mm in high-flux burner designs. Excess hydrogen is maintained in the product gas, typically within a target band of 2 vol% to 5 vol%, to ensure that residual free chlorine remains below 5 ppmv and to suppress oxidative degradation of downstream carbon steel and graphite. The selection of graphite for this service cannot be reduced to a single grade family; it is governed by the interaction between gas-phase composition, wall temperature, thermal stress, and the permeability of the graphite microstructure. Fine-grain isostatically pressed grades with bulk densities in the range of 1.78 g/cm³ to 1.86 g/cm³ and ash contents below 0.1 wt% are preferentially used where thermal shock resistance and low permeability are simultaneously required, but the choice of binder system, impregnation chemistry, and machining practice is equally significant. Published data from manufacturer datasheets and qualification programs for HCl synthesis service indicate that graphite grades with flexural strengths below 20 MPa can still survive low-stress burner liners, but higher-strength isostatic grades are required when nozzle ports, sight-glass bosses, or instrument penetrations create local stress concentrations. The following sections address the material selection logic under excess hydrogen control, with particular attention to the failure modes that are most frequently observed in production-scale equipment.

Why Does Excess Hydrogen Dictate Graphite Grade Selection in HCl Synthesis?

Excess hydrogen control is not merely a process setpoint; it directly shifts the electrochemical and thermochemical conditions at the graphite wall. Chlorine that is not consumed in the flame can adsorb on the carbon surface and form chlorinated carbon intermediates, which at wall temperatures above 200 °C can desorb as volatile chlorocarbons or react with water to generate HCl and oxygenated surface groups. In a stoichiometrically balanced H₂–Cl₂ flame, the equilibrium concentration of molecular chlorine is exceedingly low at the flame core but can rise sharply at the burner wall if mixing is incomplete. By maintaining excess hydrogen in the product gas, the molar ratio of hydrogen to chlorine at the boundary layer is shifted such that chlorine is preferentially consumed in the gas phase before reaching the graphite surface. The thermodynamic driving force for carbon oxidation by chlorine is therefore reduced, but not eliminated. Graphite grades with high catalytic ash content, particularly iron and vanadium, can accelerate the recombination of hydrogen and chlorine at the wall and create localized hot spots. For this reason, ash content is not a secondary property: iron concentrations above 50 ppm, depending on the specific mineral phase, have been associated with pitting and accelerated localized attack in chlorine-containing service when free chlorine excursions occur. The use of excess hydrogen also alters the gas-phase thermal conductivity because hydrogen has a thermal conductivity roughly seven to ten times higher than HCl at the relevant temperatures. This raises the convective heat transfer coefficient at the graphite hot face, increasing the heat flux into the water jacket and changing the temperature profile relative to a stoichiometric flame. Consequently, a graphite grade selected for a stoichiometric burner may fail under excess hydrogen operation if the thermal conductivity of the graphite is insufficient to conduct the increased heat flux without exceeding the maximum allowable wall temperature. In practice, isostatically pressed graphite with a thermal conductivity greater than 90 W/(m·K) at room temperature is generally preferred for high-flux HCl burners, although the final selection must be based on the actual heat flux and water-side fouling resistance.

The radial temperature distribution in an HCl burner graphite liner is rarely uniform because the flame is localized at the burner nozzle and the gas temperature decays along the chamber axis. A typical production-scale HCl synthesis unit with a capacity of 100 metric tons per day may use a cylindrical graphite combustion chamber with an internal diameter between 200 mm and 600 mm and a wall thickness between 25 mm and 50 mm. The flame zone occupies only the first 20% to 40% of the chamber length, producing an axial hot spot that can impose a thermal stress that exceeds the graphite flexural strength if the grade has insufficient thermal shock resistance. Thermal shock resistance of graphite is not captured by a single standard test; it is inferred from the ratio of flexural strength to the product of elastic modulus and coefficient of thermal expansion. Isostatic graphite with a flexural strength of 35 MPa, an elastic modulus of 10 GPa, and a coefficient of thermal expansion of 2.5 × 10⁻⁶/K may tolerate transient temperature differences on the order of 200 °C to 300 °C, whereas extruded grades with lower strength and higher anisotropy may crack under similar thermal gradients. Manufacturer qualification programs often use interrupted burner service cycles to introduce repeated thermal shocks, with the criterion that no new cracks exceeding 5 mm in length are detectable by dye penetrant testing after a defined number of cycles. The water-cooled back face must remain below the boiling point of water to avoid nucleate boiling and steam film formation, because film boiling can reduce the heat transfer coefficient by an order of magnitude and cause a rapid wall-temperature excursion. In practice, the maximum recommended continuous graphite hot-face temperature for HCl burners is often limited to 250 °C to 350 °C, depending on the oxygen and moisture content of the feed gases. The combination of high thermal conductivity, low elastic modulus, and fine grain size reduces the probability of thermal-stress cracking, but it also raises the cost and can complicate machining of large-diameter liners.

Thermal Degradation Pathways at the Hot-Face Graphite Liner

At the hot face of an HCl burner, graphite is subjected to several simultaneous degradation pathways that are affected by excess hydrogen. Water vapor is always present in minor quantities because chlorine produced by membrane cells or diaphragm cells contains traces of oxygen, and oxygen reacts with hydrogen in the flame to form water. The reaction of carbon with water vapor, C + H₂O → CO + H₂, is thermodynamically favorable at temperatures above approximately 700 °C, but its rate at the wall is controlled by the wall temperature and the partial pressure of water. Excess hydrogen in the product gas suppresses the equilibrium conversion of carbon to CO by shifting the reverse reaction, but this protection diminishes if the wall temperature exceeds 400 °C or if oxygen breaks through the flame. Carbon dioxide can also gasify graphite, and its formation is favored by local oxygen ingress at joints or instrumentation ports. In chlorine-rich service, the most damaging pathway is the formation of volatile metal chlorides from ash constituents; iron, calcium, and aluminum in the graphite can react with HCl or chlorine to form chlorides that leave the surface and create pits. This is why high-purity graphite with ash content below 0.1 wt% is specified for HCl burner liners, and why certain manufacturers offer grades with ash contents below 0.02 wt% for critical flame-zone components. The binder phase in graphite is often more chemically reactive than the filler coke; impregnated pitch-carbon residues can preferentially oxidize or chlorinate, leading to grain loss. The selection of a graphite grade with a uniform microstructure and fine particle size below 15 µm reduces the probability of selective attack because the active phase is more homogeneously distributed. Published data for the exact oxidation rates of specific graphite grades in HCl burner atmospheres at 250 °C to 350 °C is limited, but qualification tests typically involve exposure to flowing HCl containing 0.5 vol% oxygen and 1 vol% water for periods of 500 h to 1,000 h, followed by measurement of mass loss and flexural strength retention. Under these conditions, acceptable graphite grades often exhibit mass loss of less than 0.5% and flexural strength retention greater than 85%, although such results are specific to the test apparatus and feed gas purity.

In an HCl burner, graphite porosity is not a trivial geometric property because gas permeability can allow HCl, chlorine, and moisture to migrate through the liner toward the water jacket or the atmosphere. Graphite is inherently porous with open porosities typically ranging from 8 vol% to 20 vol% depending on grade and manufacturing method. The permeability of graphite can range from 1 × 10⁻¹⁵ m² for fine-grain impregnated grades to 1 × 10⁻¹² m² for coarse extruded grades, and this range has a direct effect on the ability of the liner to contain the reaction gases. For HCl burner service, a graphite grade is usually impregnated with a thermosetting resin, most commonly phenolic, to reduce permeability. The resin fills the interconnected pore network but remains susceptible to thermal degradation at the hot face. Phenolic resin begins to decompose at temperatures above 180 °C to 200 °C, and prolonged exposure above 250 °C can lead to carbonization and shrinkage of the impregnant, which reopens permeability. In water-cooled burners, the hot-face temperature is often maintained below this threshold, so resin-impregnated graphite can provide adequate sealing for many years. However, if the water jacket is lost or the burner is operated at high turndown with a localized hot spot, the resin can volatilize and deposit carbonaceous residues on downstream equipment. An alternative is to specify a graphite grade with intrinsically low permeability, such as a fine-grain isostatically pressed grade with a median pore diameter below 1 µm, but even these grades may require surface sealing to achieve the helium leak rates required for safety-critical components. The selection of impregnant must also consider chemical compatibility: metal-bearing impregnants such as antimony or copper are generally avoided in HCl service because they can form volatile chlorides or catalyze local attack. Phenolic-impregnated graphite is compatible with HCl at moderate temperatures but should not be used in applications where the continuous wall temperature exceeds the resin degradation threshold.

When Excess Hydrogen Operation Drops Below 2 vol% at Burner Turndown

Burner turndown is a critical operational boundary for graphite selection because the ability to maintain excess hydrogen becomes more difficult at low firing rates. At turndown below 30% of design capacity, the hydrogen and chlorine flows are reduced, and the flame length shortens, shifting the hot zone closer to the burner nozzle and reducing the mixing intensity in the chamber. If the excess hydrogen concentration falls below 2 vol% due to ratio control drift or analyzer lag, free chlorine can appear at the graphite wall and attack the liner. The corrosion rate of graphite in dry chlorine is generally low below 100 °C, but in the presence of water vapor and HCl, free chlorine can generate hydrochloric acid and hypochlorous species that increase the electrochemical potential at the surface. Graphite grades with higher electrical conductivity and lower ash content are less susceptible to electrochemical corrosion, but the most effective design response is to specify a graphite with a fine pore structure that limits the diffusion of chlorine into the bulk. Under these conditions, the graphite surface may experience transient temperatures above the normal cooling limit because the shorter flame reduces the wetted area for heat transfer to the water jacket. A graphite with a higher thermal conductivity will spread the local heat flux and reduce the peak temperature, so a grade with a conductivity above 100 W/(m·K) is often recommended for burners with turndown ratios exceeding 3:1. In addition, the graphite nozzle tip, which is exposed to the flame base, must be fabricated from the same or a higher-grade graphite and may require a replaceable insert to accommodate erosion and thermal cracking. The use of excess hydrogen as a chlorine scavenger is effective only when the mixing is adequate; at turndown, the hydrogen injection velocity may be insufficient to prevent chlorine from contacting the graphite before the reaction is complete. Consequently, the graphite selection cannot compensate for poor burner hydraulics, but a high-density, low-permeability isostatic grade can reduce the damage caused by short-duration chlorine excursions.

The chemical purity of the graphite in an HCl burner is directly linked to product quality because ash constituents can be volatilized or leached into the gas stream and transported to downstream absorption or compression equipment. In semiconductor-grade HCl applications, metal contamination limits are extremely tight, and graphite with an ash content above 0.05 wt% may be unacceptable. The ash in graphite typically contains silicon, aluminum, iron, calcium, and sometimes boron or vanadium, depending on the coke source and graphitization temperature. During HCl synthesis, iron and calcium can react with HCl to form volatile or entrained metal chlorides, which may deposit on downstream piping or contaminate the HCl product. For this reason, qualification procedures often include an acid digestion or inductively coupled plasma analysis of the graphite according to ASTM C561 or equivalent methods, with acceptance criteria defined in terms of individual metal concentrations rather than total ash alone. A graphite grade with a total ash content of 0.1 wt% may still contain iron at 30 ppm to 50 ppm, and if the application is sensitive, the iron concentration should be specified below 10 ppm. The graphitization temperature, which typically exceeds 2,500 °C, influences the degree of ash volatilization and the crystallinity of the graphitic structure. High graphitization temperatures reduce residual hydrogen and oxygen in the lattice and improve chemical stability, but they also increase cost. In practice, high-purity isostatic graphite grades with ash levels below 0.02 wt% are available, but their flexural strength may be lower than that of conventional fine-grain grades because certain ash constituents can act as grain-boundary strengtheners. The selection must therefore balance purity requirements against the mechanical stresses imposed by the burner geometry.

Machining Tolerance, Assembly Stress, and Chloride-Induced Spalling

The manner in which graphite is machined and assembled into an HCl burner has a first-order effect on service life. Graphite is brittle and notch-sensitive, so the design of flanges, thread roots, and sealing faces must avoid sharp corners with radii below 1 mm. In production-scale HCl burners, graphite liners are often fabricated from multiple cylindrical sections joined by cemented or flanged connections, and the sealing faces must be flat within 0.05 mm over a 300 mm diameter to prevent gas bypass. The tightening of steel flanges against graphite flanges can induce tensile stresses that exceed the graphite tensile strength if the bolt torque is not controlled. Graphite tensile strength is typically one-half to one-third of its flexural strength, so a grade with a flexural strength of 35 MPa may have a tensile strength of only 12 MPa to 18 MPa. Assembly procedures should specify a maximum bolt preload and the use of compressible gaskets or spring washers to accommodate differential thermal expansion between the graphite liner and the steel shell. Differential thermal expansion is not negligible: graphite has a coefficient of thermal expansion of about 2.0 × 10⁻⁶/K to 4.0 × 10⁻⁶/K, while carbon steel expands at about 12 × 10⁻⁶/K. If the graphite liner is rigidly constrained, heating from ambient to operating temperature can generate compressive or tensile stresses that exceed the material capacity. For this reason, the liner is often free to expand axially, and the hot end is allowed to float. Chloride-induced spalling is a specific failure mode in which chloride salts or hydrated metal chlorides form at the cold face or within pores and generate crystallization pressure. This is more commonly observed in burner outlet sections where the gas temperature falls below the dew point of HCl, which depends on water content but is typically in the range of 80 °C to 110 °C for wet HCl. The use of a graphite grade with low permeability and fine pore size reduces the ingress of moisture and the associated salt formation, but the cold end must also be protected by thermal insulation or trace heating if the feed gases contain water.

When selecting between isostatically pressed, molded, and extruded graphite grades for HCl burner liners, the property differences are sufficiently large that a single design cannot be qualified for all materials without testing. The table below consolidates typical property ranges reported by graphite manufacturers for fine-grain isostatically pressed and medium-grain extruded grades. These ranges are not acceptance criteria but are representative of commercial materials that have been used in corrosion-resistant graphite equipment. The selection for a specific burner must be based on the actual vendor datasheet and qualified by the end user under the intended operating conditions, because property values vary with raw coke, pitch, graphitization temperature, and impregnation.

The data in the table illustrate why extruded grades are generally not preferred for high-integrity HCl burner liners: their higher open porosity and ash content increase the risk of gas permeation and localized attack, while their lower flexural strength reduces tolerance to assembly stress. However, extruded grades may be acceptable for large-diameter outer shells or downstream sections where the gas temperature is below 150 °C and mechanical loads are modest. The final material selection should be linked to a finite element analysis of the thermal stress field, using temperature-dependent properties. Graphite thermal conductivity typically decreases with increasing temperature; a material that conducts 110 W/(m·K) at room temperature may conduct only 60 W/(m·K) to 80 W/(m·K) at 500 °C. The elastic modulus may also change with temperature, but the change is less pronounced. Published data for the specific combination of HCl, excess hydrogen, and water vapor at the graphite surface is limited, and most qualifications rely on comparative testing rather than absolute predictions.

What Limits the Use of Resin-Impregnated Graphite in Chlorine-Bearing Service?

Phenolic resin impregnation is the most common method for reducing the permeability of graphite in corrosion-resistant equipment, but its use in HCl burners is bounded by thermal and chemical limits. The resin is typically cured after impregnation at temperatures between 150 °C and 180 °C, and the resulting composite has a permeability that is one to three orders of magnitude lower than the base graphite. In an HCl burner, the hot face of the impregnated graphite may reach temperatures that exceed the resin degradation threshold if the water jacket is not effective. At temperatures above 200 °C, the resin begins to decompose, releasing water, phenolics, and low-molecular-weight organic compounds into the gas stream. In semiconductor-grade or pharmaceutical-grade HCl production, even trace organic contamination is unacceptable, and resin-impregnated graphite may be disqualified for this reason. The chemical compatibility of phenolic resin with wet HCl is generally acceptable at temperatures below 80 °C, but at higher temperatures the resin can hydrolyze or soften. In addition, when free chlorine is present during turndown excursions, the resin can be chlorinated, producing chlorinated phenols and other compounds that may accelerate resin degradation. For these reasons, some HCl burner designs avoid organic impregnants altogether and rely on high-density, fine-grain graphite with an intrinsically low open porosity. However, such grades still have some interconnected porosity, and leaks may occur at flanges or joints unless additional measures are taken. In critical burner sections, a two-layer approach is sometimes used: a hot-face liner of high-purity, unimpregnated isostatic graphite and a cold-face sleeve of phenolic-impregnated graphite separated by a cooling gap. This arrangement keeps the organic impregnant below its degradation temperature while the hot face is protected from chlorine attack by the excess hydrogen. Published data for long-term service life of this two-layer arrangement in excess hydrogen HCl burners is limited, and end users typically qualify the design through pilot-scale testing.

The control of excess hydrogen is achieved through continuous analysis of the product gas, usually by a thermal conductivity analyzer or an optical cell that measures free chlorine. The response time of the analyzer and the actuation speed of the hydrogen control valve determine how quickly the burner can recover from a chlorine excursion. Graphite selection interacts with this control system because a high-conductivity, low-porosity graphite liner can tolerate a longer excursion before the onset of irreversible attack. For example, if the analyzer response time is 10 s to 20 s and the valve stroke time is 5 s to 10 s, the graphite may be exposed to free chlorine for up to 30 s. In this short period, the damage is likely limited to surface roughening rather than structural failure, provided the wall temperature is below 250 °C. If the wall temperature exceeds 350 °C, the same excursion can cause measurable mass loss. This is why the maximum allowable continuous hot-face temperature is often set at 300 °C for graphite grades with ash contents below 0.1 wt%, with a recommendation to alarm at 320 °C and trip at 350 °C. The alarm and trip limits should be derived from the specific graphite grade because oxidation kinetics are sensitive to catalytic impurities. A graphite with a higher surface area due to fine porosity may also have a higher oxidation rate, even at the same bulk density, because the reaction rate scales with accessible surface area. Mercury intrusion porosimetry data, which report the pore size distribution, are therefore more useful than bulk density alone for predicting oxidation behavior. A narrow pore size distribution with a median pore diameter below 1 µm is generally associated with better resistance to gas penetration and lower effective diffusivity.

In production-scale HCl burner fabrication, batch-to-batch variations in graphite billets can affect the machining yield and final assembly leak rate. Isostatically pressed graphite billets are formed by compressing powder under high isostatic pressure, typically 100 MPa to 200 MPa, followed by carbonization and graphitization. The resulting billet has a more uniform grain structure than extruded or molded grades, but the graphitization cycle can introduce residual stresses if the heating rate is not controlled. Manufacturer certificates for high-purity grades typically include bulk density, flexural strength, ash content, and thermal diffusivity for each billet, but the values can vary within a single billet by up to 5%. For critical flame-zone components, the end user may require that test coupons be cut from the same billet and machined in the same orientation as the finished part, because graphite properties can be anisotropic even in near-isotropic isostatic grades. The orientation of the part relative to the billet pressing axis can affect flexural strength by 10% to 15% and thermal conductivity by similar amounts. Machining operations such as turning, milling, and drilling create surface damage that can reduce flexural strength by up to 20% if improper tools or speeds are used. Diamond tooling, low feed rates, and adequate dust extraction are standard practice. After machining, the graphite surfaces should be cleaned with dry, oil-free compressed air or solvent wiped with a non-chlorinated solvent to remove machining fines, because loose graphite particles can enter the burner and contaminate the product.

Flange connections in graphite burners require particular attention to creep and relaxation. Graphite does not creep significantly at ambient temperatures, but the gaskets and steel bolts do. The use of PTFE or flexible graphite gaskets is common, but PTFE is limited to temperatures below 200 °C and can deform under load. Flexible graphite gaskets, which are manufactured from expanded graphite, have excellent chemical resistance and can operate at temperatures up to 400 °C in non-oxidizing atmospheres. However, flexible graphite is soft and requires controlled compression to avoid extrusion into the gas stream. The bolt preload for graphite flanges should be calculated to provide a gasket seating stress without exceeding the graphite compressive strength. As a rule, the maximum allowable compressive stress on graphite flanges is typically limited to 50% of the published compressive strength. For a graphite with a compressive strength of 80 MPa, the design compressive stress should not exceed 40 MPa. If the graphite liner is heated, the steel bolts expand more than the graphite flange, which can reduce the gasket seating stress and cause leakage. This is why spring washers or disc springs are recommended and why the bolt torque must be rechecked after the first thermal cycle.

Production-scale experience with HCl burners indicates that the most frequent material-related failures are not uniform corrosion but localized cracking from thermal stress and joint leakage due to inadequate compression. Graphite liners that survive 10,000 h of continuous operation often show no measurable change in bulk density but may exhibit increased helium leak rates at the hot end, indicating microcracking or resin degradation. The use of excess hydrogen does not eliminate all chlorine-related degradation; it reduces the frequency and severity of chlorine excursions. In systems where the excess hydrogen analyzer is out of service, the burner should be operated with a conservative hydrogen bias, but the graphite wall temperature must be monitored because high hydrogen concentrations can increase heat flux. Graphite grades containing metal-bearing impregnants or high-ash filler cokes are incompatible with excess hydrogen HCl service when used downstream of the flame zone, because metal chlorides can form and deposit in the absorption system. The final acceptance of a graphite grade for a specific HCl burner should include a pilot-scale or side-stream test that exposes the material to the actual product gas composition, including the worst-case chlorine excursion and the maximum turndown condition. Without such a test, published data for this specific configuration is limited and the qualification remains provisional.