Articles
Articles
Articles
Fouling of the preheat train in a delayed coking unit processing significant ethylene tar fractions is governed by free-radical oligomerization of vinyl aromatic and diene moieties that survive the steam cracker quench system. The thermal degradation threshold for ethylene tar olefins is documented in petroleum refining literature as initiating at approximately 120°C to 150°C, with autocatalytic propagation accelerating rapidly above 200°C in the presence of dissolved oxygen and trace iron from storage tank corrosion products. This reactivity profile is quantified by bromine number determination per ASTM D1159-07 (reapproved 2017), with ethylene tar samples routinely exhibiting bromine numbers between 10 g Br2/100 g and 40 g Br2/100 g, compared to FCC decant oil values that typically fall below 5 g Br2/100 g. The consequence of this olefinic instability manifests as a progressive increase in pressure drop across the feed-bottoms exchanger network, where shell-and-tube exchangers with low-finned tubes of 19 mm outer diameter and 2.5 m to 6 m tube length experience deposit formation rates that reduce the overall heat transfer coefficient from design values of 250 W/m²·K to 350 W/m²·K to operational values below 150 W/m²·K within 60 to 90 days of continuous operation when undiluted ethylene tar constitutes more than 15 wt% of the total coker feed. Published industrial case data for specific fouling rates in ethylene tar service is limited; however, refinery operations literature consistently documents that coker feed originating from steam-cracker-derived streams requires either on-line spalling, scheduled off-line hydroblasting at 10,000 psi to 20,000 psi (69 MPa to 138 MPa) using rotating nozzle lance systems, or the installation of standby exchanger capacity to permit cleaning without unit shutdown. The mitigation of this fouling mechanism is further complicated by the fact that conventional antifoulant chemistries effective against asphaltene precipitation in vacuum residue service demonstrate reduced efficacy against the vinyl-polymerization fouling pathways characteristic of ethylene tar, with field data from coker operations indicating that phosphate ester dispersants and hindered phenol antioxidants provide only partial stabilization when injection rates are maintained at 50 ppm to 200 ppm based on feed volume.
At the coker furnace outlet, where the process requires coil outlet temperatures of 495°C to 510°C for needle coke feedstocks, the increased unsaturation of ethylene tar translates into measurable increases in coker furnace tube skin thermocouple readings that exceed baseline decant oil operation by 15°C to 30°C at equivalent absorbed duty. The furnace tubes, typically constructed from 9Cr-1Mo alloy steel per ASTM A335 Grade P9 with an outside diameter of 114.3 mm and a minimum wall thickness of 6.02 mm, are designed for a maximum tube skin temperature of approximately 650°C under decant oil service conditions. Ethylene tar blends exceeding 25 wt% have been reported in technical literature to approach skin temperatures of 630°C to 640°C after 30 to 45 days of continuous operation, reducing the thermal margin to the design limit to less than 20°C. The kinetic basis for this accelerated coking resides in the propensity of ethylene tar polyaromatic compounds to undergo thermal condensation reactions at temperatures above 450°C, generating mesophase precursors within the furnace coil itself rather than in the coke drum where anisotropic structure development is intended. The preheat system therefore operates under a dual constraint: sufficient residence time at high temperature to maintain feed fluidity and atomization, contrasted against excessive residence time that initiates premature mesophase nucleation in the furnace tube boundary layer where wall shear rates are insufficient to prevent deposition. This constraint defines an operational window in which the furnace pass velocity must be maintained between 2.1 m/s and 3.4 m/s for liquid-phase operation, corresponding to mass flux values of 1,800 kg/m²·s to 2,600 kg/m²·s, to simultaneously achieve adequate heat transfer and minimize coking deposition.
Mesophase development — the liquid-crystalline intermediate state that precedes anisotropic coke formation — is the governing kinetic step that determines whether ethylene tar-derived coke develops the needle-like microstructure required for high-performance graphite electrodes. The mesophase transformation for ethylene tar feedstocks initiates at approximately 410°C to 430°C under delayed coking drum conditions, with the rate of spherule nucleation following an Arrhenius relationship that produces observable anisotropic domains within 2 to 6 hours of drum residence at 440°C to 460°C. The critical operating constraint in needle coke production is the maintenance of drum vapor space temperatures between 440°C and 470°C, because mesophase viscosity — which decreases from approximately 10³ Pa·s at 420°C to 10¹ Pa·s at 460°C — must remain sufficiently low to permit coalescence of mesophase spherules into bulk anisotropic domains, yet sufficiently high to prevent turbulent disruption by volatile matter evolution. Ethylene tar, with its characteristic distillation profile showing 30 wt% to 50 wt% of the material volatilizing below 400°C per ASTM D7169-20 simulated distillation, generates a substantial vapor fraction within the drum that serves a dual function: the evolving light aromatics strip lower molecular weight isotropic liquid from the growing mesophase, improving orientation, while simultaneously creating gas bubbles that, if generated too rapidly, fracture the coalescing mesophase domains and produce isotropic inclusions in the final coke product. The drum pressure must therefore be controlled within the range of 0.3 MPa to 0.7 MPa absolute, with the lower end of this range preferred for needle coke production because reduced pressure enhances volatile removal and promotes unidirectional mesophase alignment parallel to the drum axis. Industrial practice documented in petroleum coking literature indicates that pressure excursions above 0.7 MPa for periods exceeding 2 hours during the coking cycle produce measurable increases in the coefficient of thermal expansion of the resulting coke, with CTE values shifting from the premium specification of ≤0.25 ppm/°C toward the regular needle coke range of 0.30 ppm/°C to 0.45 ppm/°C when measured by dilatometric methods per ASTM E228-17 on extruded graphite test specimens prepared from the calcined coke.
Beyond temperature and pressure, the mesophase formation window for ethylene tar is constrained by the quantitative relationship between feed aromaticity and the solubility of mesophase in the isotropic liquid phase. The Hildebrand solubility parameter of ethylene tar, estimated from the aromatic carbon content measured by ASTM D5292-99 (reapproved 2019) and reported as approximately 21 MPa0.5 to 23 MPa0.5, determines the chemical potential driving force for mesophase precipitation. In feedstocks where ethylene tar constitutes more than 50 wt% of the coker feed without dilution by a lower-solubility-parameter component such as FCC decant oil (which exhibits a solubility parameter closer to 19 MPa0.5 to 20 MPa0.5), the mesophase precipitation onset shifts to lower temperatures by approximately 10°C to 20°C, reducing the thermal margin for controlled mesophase growth before solidification arrests structural development. This shift is critical because the difference between the mesophase onset temperature and the temperature at which the mesophase transforms fully to green coke is the effective processing window; for ethylene tar-rich feeds, this window narrows from the typical 30°C to 50°C range observed with decant oil to a range of 20°C to 35°C, requiring tighter furnace outlet temperature control of ±3°C versus the ±5°C tolerance generally accepted for conventional needle coke operations. Published data for specific ethylene tar-rich industrial coking runs is limited; however, laboratory-scale coking studies using autoclave reactors and thermogravimetric analysis provide the kinetic parameters that define these boundaries, with the mesophase content of semicoke samples determined by polarized light microscopy per the methodology described in ASTM D4616-95 (reapproved 2023) for pitch characterization and adapted for green coke.
Characterization of the mesophase transformation kinetics in ethylene tar systems requires simultaneous measurement of volatile evolution rate, mesophase content, and anisotropic domain size distribution. Thermogravimetric analysis coupled with differential scanning calorimetry, conducted at heating rates of 5°C/min to 20°C/min under inert atmosphere with nitrogen purge rates of 50 mL/min, reveals that ethylene tar loses 50 wt% to 65 wt% of its mass between 200°C and 500°C, with the maximum rate of volatile evolution occurring between 380°C and 430°C — a temperature range that overlaps directly with the mesophase nucleation regime. This overlap creates a process conflict unique to ethylene tar among needle coke feedstocks: the volatile fraction that must be removed to concentrate the mesogenic components is released at precisely the temperatures where mesophase spherules are forming and must remain undisturbed. The practical resolution of this conflict in industrial delayed coking involves staging the temperature ramp within the drum: the feed is initially distributed at the drum inlet with a temperature of 480°C to 490°C, but the inherent heat transfer dynamics of the coke drum establish a vertical thermal gradient of 2°C/cm to 5°C/cm, such that the mesophase formation zone occupies a vertical band 1.5 m to 3 m in depth that migrates upward as the drum fills over the 24-hour to 48-hour coking cycle. The feed rate for needle coke production is consequently set at 1.5 m³/h to 4 m³/h per drum for drums of 6.1 m to 8.5 m diameter and 24 m to 30 m tangent-to-tangent height, yielding superficial vapor velocities of 0.05 m/s to 0.15 m/s at the drum top that are sufficiently low to permit mesophase coalescence without channel formation. The maximum anisotropic domain size reported in technical literature for ethylene tar-based needle coke produced under these conditions ranges from 200 μm to 600 μm, compared to domain sizes exceeding 1,000 μm reported for high-quality decant oil-derived needle coke, reflecting the inherent limitation imposed by the ethylene tar volatile evolution profile.
Sulfur heteroatom chemistry represents the most quantitatively restrictive limitation confronting ethylene tar utilization in needle coke delayed coking. Total sulfur in ethylene tar, determined by wavelength-dispersive X-ray fluorescence per ASTM D4294-21, spans 0.5 wt% to 3.5 wt% depending on the sulfur content of the naphtha or gas oil charged to the upstream steam cracker and the severity of the cracking operation. The needle coke specification frameworks maintained by graphite electrode manufacturers — documented in procurement specifications referenced against ISO 12980:2019 (carbonaceous materials used in the production of aluminium, green coke and calcined coke for electrodes, analysis using X-ray fluorescence spectrometry) — dictate a maximum sulfur content in calcined needle coke of 0.6 wt% for regular grade material and 0.5 wt% for premium grade material destined for ultra-high-power electric arc furnace electrodes. The sulfur distribution between the liquid and vapor phases during delayed coking is not uniform: approximately 30% to 50% of the sulfur present in the ethylene tar feed reports to the green coke, while the remainder partitions into the coker vapor and liquid products, resulting in a sulfur concentration factor of 1.5 to 2.5 between feed sulfur and green coke sulfur when expressed on a weight percentage basis. For an ethylene tar containing 1.0 wt% sulfur used as 25 wt% of a blend with a decant oil containing 0.4 wt% sulfur, the blended feed sulfur of approximately 0.55 wt% would yield green coke sulfur in the range of 0.6 wt% to 0.9 wt% after concentration — a result that straddles the acceptability boundary for regular needle coke and definitively excludes premium grade production. Published industrial data on sulfur partitioning in delayed coking of ethylene tar blends is limited, but the general thermodynamic framework of sulfur speciation in carbochemical systems is documented in the petroleum refining literature, identifying thiophenic sulfur compounds as the dominant sulfur species retained in the coke matrix due to their thermal stability and incorporation into the developing aromatic lamellae.
Beyond the specification compliance calculus, sulfur exerts a direct structural and electrochemical degradation effect on needle coke quality through multiple independent mechanisms. First, sulfur atoms incorporated at the periphery of aromatic lamellae during coking create localized lattice discontinuities that increase the interlayer spacing of the graphitic structure, with a reported increase of 0.002 nm to 0.005 nm in d002 spacing for each 0.1 wt% increment in coke sulfur above 0.3 wt%, as measured by X-ray diffraction per ASTM D5187-10 (reapproved 2021) and calculated using the Scherrer equation. Second, the presence of sulfur in the coke increases the coefficient of thermal expansion through disruption of the oriented lamellar arrangement, with technical literature reporting an approximate 0.02 ppm/°C to 0.05 ppm/°C increase in longitudinal CTE for each 0.1 wt% increase in sulfur content above 0.4 wt%. Third, during graphitization at temperatures of 2,800°C to 3,000°C in the Acheson or longitudinal graphitization furnace, sulfur is released as sulfur dioxide and carbon disulfide through a process known as "puffing," which generates internal porosity and can cause dimensional instability or cracking in the graphite electrode if the sulfur content exceeds 0.6 wt% to 0.8 wt%. The puffing phenomenon is quantified by measuring the irreversible expansion of electrode test specimens during graphitization heat treatment, with the puffing index defined as the percentage increase in specimen length after heating at a rate of 5°C/min to 2,800°C under argon atmosphere. Ethylene tar-derived needle coke samples with sulfur contents above 0.5 wt% have been reported in petroleum coking literature to exhibit puffing indices exceeding 0.5%, compared to values below 0.2% for low-sulfur decant oil-derived needle coke.
Metals contamination, ash content, and the filtration requirements for quinoline insolubles (QI) removal define the third cluster of limitations governing ethylene tar acceptance in needle coke delayed coking operations. Ethylene tar produced from naphtha crackers typically carries low concentrations of vanadium and nickel — the two metals of greatest concern in carbon product specifications — with values reported in petroleum refining literature below 1 ppm for each element when measured by inductively coupled plasma atomic emission spectroscopy per ASTM D5185-18. However, the iron content of ethylene tar can reach 20 ppm to 50 ppm when storage and transfer equipment is constructed from carbon steel, because the acidic components in the tar (phenolic oxygenates and carboxylic acid derivatives formed during steam cracking) slowly corrode ferrous surfaces and suspend the resulting iron naphthenates in the feedstock. Iron contamination is particularly detrimental to needle coke quality because iron acts as a catalytic poison during the solid-state graphitization step, accelerating the formation of disordered graphitic material at the expense of oriented lamellar domains. The ash content of green and calcined needle coke, determined by complete combustion per ASTM D4422-13, is limited to 0.1 wt% maximum in premium needle coke specifications, corresponding to a maximum allowable metals loading of approximately 100 ppm total ash-forming elements in the coker feed when accounting for the concentration factor during coking. For ethylene tar containing 50 ppm iron, the contribution to green coke ash is approximately 0.03 wt% per 25 wt% of ethylene tar in the feed blend — a contribution that consumes 30% of the total ash budget and leaves insufficient margin for ash contributions from the co-feed components. The mitigation strategy for ethylene tar-derived iron contamination involves either the use of stainless steel storage tanks constructed from ASTM A240 Type 316L alloy, the installation of magnetic filtration systems with 1 μm to 5 μm capture ratings upstream of the coker feed drum, or chemical iron chelation using citric acid-based additives at injection rates of 10 ppm to 50 ppm.
Filtration for QI control is mandatory in needle coke feedstock preparation regardless of whether ethylene tar is present, but the nature of the QI in ethylene tar differs fundamentally from that in FCC decant oil. In decant oil, the QI consists of refractory catalytic fines (aluminosilicate zeolite fragments and matrix material) that require mechanical separation. In ethylene tar, the QI is predominantly composed of pyrolytic carbon — carbonaceous particulates formed by homogeneous nucleation in the steam cracker radiant coils and quench system — plus entrained coke fines from the quench oil system. The particle size distribution of pyrolytic carbon in ethylene tar, measured by laser diffraction particle size analysis, typically exhibits a bimodal distribution with one population centered at 0.5 μm to 2 μm and a second population at 5 μm to 20 μm. The total QI content of unfiltered ethylene tar ranges from 0.2 wt% to 1.0 wt% per ASTM D2318-20 (quinoline-insoluble content of tar and pitch), but this value must be reduced below 0.5 wt% and preferably below 0.2 wt% to prevent QI particles from acting as nucleation sites that disrupt mesophase coalescence and produce isotropic domains within the needle coke structure. The filtration configuration in industrial needle coke feedstock preparation typically employs a two-stage system: primary filtration through backwashable sintered metal mesh filters with 5 μm to 10 μm nominal retention ratings, followed by secondary depth filtration through diatomaceous earth or cellulose fiber filter aids in pressure leaf filters operating at 0.3 MPa to 0.5 MPa differential pressure and temperatures of 150°C to 200°C to maintain feed viscosity below 100 cSt during filtration. The filtration of ethylene tar-rich blends is complicated by the presence of the previously described olefinic oligomerization products, which form gelatinous deposits on filter media and reduce the effective filtration cycle time from the 8-hour to 24-hour range typical for decant oil service to 2 hours to 6 hours for ethylene tar-containing feeds.
Blending of ethylene tar with FCC decant oil or thermal tar represents the most common industrial strategy for exploiting ethylene tar's high aromaticity while mitigating its sulfur, olefin, and volatility liabilities. The quantitative relationship between blend ratio and needle coke quality is not linear; rather, it exhibits threshold behavior that defines a practical maximum blending ratio of 30 wt% to 40 wt% for ethylene tar in decant oil-based feeds, beyond which the aggregate properties of the blend fail one or more critical feedstock criteria. The blended feed aromaticity, expressed as BMCI, follows a mass-weighted average mixing rule that is documented in petroleum refining literature: a decant oil with BMCI of 110 blended with ethylene tar of BMCI 125 at 30 wt% ethylene tar yields a blended BMCI of approximately 115, which remains above the industry-accepted threshold of 110 to 120 for needle coke feedstock suitability. However, when the blending ratio reaches 50 wt%, the blended BMCI exceeds 120, a value that is superficially favorable but is accompanied by a corresponding increase in the blended bromine number from approximately 3 g Br2/100 g (decant oil alone) to 8 g Br2/100 g to 12 g Br2/100 g — a threshold at which the fouling rate in the preheat exchangers begins to accelerate non-linearly, with pressure drop doubling within 30 days of continuous operation according to refinery fouling case studies published in heat exchanger design literature. The maximum practical blending ratio is therefore determined not by aromaticity or BMCI considerations but by the olefinic reactivity limit, which imposes a ceiling of approximately 35 wt% to 40 wt% for ethylene tar from naphtha crackers and 25 wt% to 30 wt% for ethylene tar from gas oil crackers, the latter carrying higher sulfur and nitrogen loads that compound the reactivity issue. Published industrial data for specific blending optimization cases is limited; the thresholds cited above are derived from the combination of bromine number specifications, fouling rate models documented in heat transfer engineering literature, and the sulfur concentration factors established through coker mass balance calculations.
Within the blend compatibility framework, the asphaltene and QI interaction between ethylene tar and the co-feed components deserves particular attention. Ethylene tar contains essentially no pentane-insoluble asphaltenes when measured per ASTM D3279-19, with typical values below 0.5 wt%, because the steam cracking process converts the heavy alkyl aromatic structures that would otherwise form asphaltenes into polyaromatic hydrocarbons and light olefins. Decant oil, by contrast, can contain 0.5 wt% to 5 wt% asphaltenes depending on the FCC unit conversion level and the effectiveness of the slurry settler system. The stable dispersion of decant oil asphaltenes depends on the aromatic solvency of the surrounding liquid phase, and the addition of ethylene tar — with its high aromatic carbon content measured at 60% to 75% by ASTM D5292-99 (reapproved 2019) — initially improves asphaltene peptization and reduces the risk of asphaltene deposition in the preheat train. However, this solvency benefit is offset by the lower volumetric average boiling point of ethylene tar, which shifts the blend distillation profile such that the 50 wt% recovery temperature decreases by 15°C to 30°C relative to decant oil alone per ASTM D7169-20 simulated distillation. This shift reduces the concentration of high-boiling solvent molecules that remain in the liquid phase during the mesophase formation regime, potentially destabilizing the mesophase in the late stages of the coking cycle when the residual liquid fraction becomes enriched in asphaltenic and pre-asphaltenic components. The operational manifestation of this instability is an increase in the isotropic coke content of the drum product, detectable by polarized light microscopy as domains of less than 5 μm in maximum dimension that fail to coalesce into the oriented needle-like structure. For blend ratios where ethylene tar exceeds 30 wt%, the proportional increase in the isotropic fraction becomes statistically significant, with laboratory coking studies reporting anisotropy indices (the ratio of anisotropic to isotropic domain area) that fall below 0.8 compared to values of 0.9 to 0.95 for decant oil-only operation.
Operation of delayed coking drums with ethylene tar-containing feedstocks requires a deliberate trade-off between the low-pressure conditions that favor anisotropic mesophase development and the high-pressure conditions that suppress premature volatile evolution and maintain acceptable drum vapor velocities. The drum pressure control system, typically consisting of a back-pressure control valve on the combined coker vapor line to the main fractionator, maintains the drum pressure setpoint within ±0.02 MPa of the target value. For needle coke production with ethylene tar blends, the pressure setpoint is positioned at 0.35 MPa to 0.55 MPa absolute, representing a compromise that accommodates the higher volatile content of ethylene tar relative to decant oil. At pressures below 0.30 MPa, the accelerated release of light aromatic vapors from the mesophase formation zone generates a three-phase flow regime (liquid mesophase, isotropic liquid, and vapor) that disrupts mesophase domain growth and increases the coefficient of thermal expansion of the resulting coke. At pressures above 0.60 MPa, suppressed volatile release traps isotropic liquid within the mesophase matrix, and the resulting coke exhibits reduced anisotropic domain size and increased microporosity. The drum pressure constraint is further complicated by the interaction with the coker furnace operation: higher drum pressures require higher furnace outlet temperatures to maintain the same drum vapor space temperature, and the furnace outlet temperature is itself constrained by the maximum tube skin temperature limit of 650°C for 9Cr-1Mo steel per ASTM A335 Grade P9. For every 0.1 MPa increase in drum pressure, the furnace outlet temperature must be increased by approximately 3°C to 5°C to maintain target drum temperatures, reducing the thermal margin between operating conditions and the furnace tube coking threshold. This coupling between drum pressure and furnace severity defines the upper pressure limit for ethylene tar service at approximately 0.55 MPa to 0.60 MPa absolute for feeds containing 25 wt% or more ethylene tar.
The furnace coil coking rate in ethylene tar service is dominated by the same reactive species that cause preheat exchanger fouling, but the mechanism transitions from olefin oligomerization to thermal condensation and mesophase deposition at the elevated temperatures of the coker furnace. The furnace design for needle coke operations typically employs multiple parallel passes — commonly 4 to 6 passes for units processing 2,000 m³/day to 8,000 m³/day of feed — with each pass containing 20 to 30 horizontal tubes in a double-fired cabin configuration. Tube inner diameters are maintained at 100 mm to 125 mm to provide the high volumetric flow capacity required for the high recycle rates that characterize needle coke coker operations. The addition of ethylene tar to the feed increases the fouling rate in the upper radiant section of the furnace, where the bulk fluid temperature exceeds 450°C and the homogeneous thermal condensation of polyaromatic compounds initiates. Refinery operational data reported in petroleum processing literature indicates that a 10 wt% ethylene tar blend increases the rate of furnace tube pressure drop by 20% to 40% relative to decant oil-only operation, a consequence of the higher aromatic condensation reactivity of ethylene tar's low-molecular-weight polyaromatic species. The practical response in industrial operations involves either increasing the steam injection rate to the furnace passes from the baseline of 0.5 wt% to 1.0 wt% based on feed (which increases mass flux and reduces wall residence time) or accepting a shortened furnace run length between decoking operations. Furnace decoking for units processing ethylene tar blends is typically scheduled at 4-month to 6-month intervals, compared to 6-month to 9-month intervals for decant oil-only operation, with the decoking procedure employing steam-air spalling at temperatures of 650°C to 700°C followed by high-velocity steam purge at 2.5 m/s to 3.5 m/s superficial velocity to remove detached coke particles.
Green coke yield from ethylene tar-containing delayed coking operations is systematically lower than that obtained from decant oil feedstocks at equivalent operating conditions, reflecting the lower Conradson carbon residue (CCR) of ethylene tar relative to conventional needle coke feedstocks. The micro carbon residue of ethylene tar, determined per ASTM D4530-15, spans 8 wt% to 20 wt% depending on the severity of the upstream steam cracker and the extent of quench oil recycle; by contrast, FCC decant oil used for needle coke production typically exhibits micro carbon residue values of 10 wt% to 25 wt%, and thermally cracked residue can exceed 30 wt%. The green coke yield from a delayed coking unit is a function of the feed micro carbon residue, the coker recycle ratio, and the drum operating conditions, with a practical yield range for needle coke operations of 20 wt% to 35 wt% of fresh feed. When ethylene tar at 30 wt% blend ratio is introduced into a decant oil-based feed, the blended micro carbon residue decreases by 1 wt% to 3 wt%, and the green coke yield correspondingly decreases by 2 wt% to 5 wt% — a reduction that directly impacts the economic viability of the operation because the drum cycle time must be extended to produce the same coke mass per cycle. The extended cycle time is not merely an economic penalty; it also extends the residence time of the mesophase at elevated temperature, which increases the risk of over-coking and the development of a fine-grained, high-density coke that behaves like anode-grade rather than needle-grade material. The trade-off between yield and quality is managed in industrial practice by adjusting the recycle ratio — the ratio of coker heavy gas oil recycled from the main fractionator to the fresh feed — within the range of 1.05 to 1.40 for needle coke operations. Higher recycle ratios improve needle coke quality by concentrating the heavy aromatic fraction of the feed, but they also increase the furnace duty and coking rate. For ethylene tar blends, the recycle ratio is typically maintained at the lower end of this range (1.05 to 1.15) because the ethylene tar itself supplies the light aromatic solvency that would otherwise require higher recycle rates to achieve.
Hydrodesulfurization (HDS) of ethylene tar prior to delayed coking represents a technically feasible but operationally challenging pre-treatment strategy for overcoming the sulfur limitation. The catalytic hydrotreating of ethylene tar requires elevated hydrogen partial pressures of 8 MPa to 18 MPa and reactor temperatures of 320°C to 400°C, with liquid hourly space velocities of 0.5 h⁻¹ to 1.5 h⁻¹ over nickel-molybdenum or cobalt-molybdenum catalysts supported on alumina (catalyst specifications per ASTM D3907-13 for activity testing of fluid catalytic cracking catalysts, adapted for hydrotreating catalyst evaluation). The HDS performance on ethylene tar is limited by three factors: (1) the high olefin content consumes hydrogen through saturation reactions at approximately 50 Nm³/m³ to 150 Nm³/m³ of hydrogen per unit volume of feed, increasing the hydrogen demand above that required for sulfur removal alone; (2) the aromatic saturation reactions are thermodynamically constrained at the temperatures exceeding 350°C that are necessary for deep sulfur removal, limiting the achievable sulfur reduction to approximately 70% to 85% for ethylene tar feeds with initial sulfur contents of 1.0 wt% to 3.0 wt%; and (3) the polymerization of olefinic species in the catalyst bed — the same chemistry that fouls the coker preheat exchangers — accelerates catalyst deactivation, reducing catalyst cycle life from the 12-month to 24-month range typical for straight-run gas oil hydrotreating to 3 months to 6 months for ethylene tar service. Published data on long-term ethylene tar hydrotreating is limited, but the fundamental catalyst deactivation mechanisms of olefin polymerization and coke deposition on the catalyst surface are well documented in hydroprocessing literature, with the relevant kinetic parameters (activation energy for coking: 90 kJ/mol to 130 kJ/mol; hydrogen consumption for olefin saturation: 1.0 Nm³/m³ per wt% olefin) available from published kinetic studies of hydrotreating of aromatic feeds.
At the interface between feedstock chemistry and graphite electrode manufacturing, the mechanical and thermal property specifications of the final electrode product impose a backward-propagating constraint on ethylene tar flexibility. The graphite electrode specifications for ultra-high-power electric arc furnace service — defined in procurement documents aligned with ISO 14420:2020 (determination of coefficient of thermal expansion of carbonaceous materials) and ISO 10143:2019 (electrical resistivity of calcined coke granules) — require a longitudinal coefficient of thermal expansion of ≤0.25 ppm/°C measured over the temperature range of 25°C to 520°C, a transverse electrical resistivity of ≤6.5 μΩ·m at ambient temperature, and a flexural strength of ≥10 MPa per ASTM D7972-21 (three-point flexural strength of carbon and graphite). The needle coke that serves as the aggregate for these electrodes must exhibit a real density of ≥2.13 g/cm³ after calcination at 1,400°C, measured by helium pycnometry per ASTM D2638-10 (reapproved 2021), and a sulfur content of ≤0.5 wt% to ≤0.6 wt% per ASTM D4239-18 (sulfur in the analysis sample of coal and coke using high-temperature tube furnace combustion, adapted for petroleum coke). Ethylene tar that has not been hydrodesulfurized and retains its inherent sulfur content of 0.5 wt% to 3.5 wt% produces calcined coke that fails the sulfur specification at blend ratios above 15 wt% to 20 wt%, depending on the co-feed sulfur content. Furthermore, the nitrogen content of ethylene tar — typically 0.1 wt% to 0.4 wt% per ASTM D5291-21 — introduces heterocyclic nitrogen species into the coke microstructure that affect the electrochemical performance of the graphite electrode, with nitrogen content above 0.5 wt% in the calcined coke associated with increased electrode consumption rates in electric arc furnace service. The combination of these property constraints creates a practical limit for ethylene tar incorporation in needle coke feedstocks that is significantly lower than the limit imposed by aromaticity or mesophase considerations alone, reinforcing the conclusion that ethylene tar serves best as a moderate (10 wt% to 30 wt%) blending component rather than a primary feedstock.
| Property | Ethylene Tar (Naphtha Cracker) | Ethylene Tar (Gas Oil Cracker) | FCC Decant Oil | Thermally Cracked Residue |
|---|---|---|---|---|
| Specific gravity at 15.6°C (ASTM D4052-22) | 1.03–1.08 g/cm³ | 1.05–1.10 g/cm³ | 1.00–1.06 g/cm³ | 1.05–1.15 g/cm³ |
| Kinematic viscosity at 50°C (ASTM D445-21) | 50–300 cSt | 100–500 cSt | 100–800 cSt | 1,000–10,000 cSt at 100°C |
| Conradson carbon residue (ASTM D189-06, reapproved 2019) | 8–15 wt% | 10–20 wt% | 10–25 wt% | 25–45 wt% |
| Total sulfur (ASTM D4294-21) | 0.3–1.5 wt% | 1.0–3.5 wt% | 0.3–1.5 wt% | 0.5–3.0 wt% |
| Aromatic carbon (ASTM D5292-99, reapproved 2019) | 60–70% | 65–75% | 50–65% | 45–60% |
| BMCI (calculated) | 100–130 | 110–140 | 90–120 | 70–100 |
| Quinoline insolubles (ASTM D2318-20) | 0.2–1.0 wt% | 0.3–1.5 wt% | 0.5–5.0 wt% | 0.1–0.5 wt% |
| Bromine number (ASTM D1159-07, reapproved 2017) | 10–30 g Br₂/100 g | 15–40 g Br₂/100 g | 1–5 g Br₂/100 g | 2–8 g Br₂/100 g |
| Nickel + vanadium (ASTM D5185-18) | <2 ppm | <5 ppm | 1–20 ppm | 10–200 ppm |
| 50 wt% recovery temperature (ASTM D7169-20) | 300–380°C | 320–400°C | 400–480°C | 480–550°C |
The molecular architecture of ethylene tar — a distribution of two-ring to five-ring polyaromatic hydrocarbons with pendant alkyl groups and residual olefinic unsaturation — generates characteristic behavior during the delayed coking cycle that differs fundamentally from the behavior of petroleum-derived aromatic feedstocks such as FCC decant oil. In the coker drum, where feed is introduced at 480°C to 490°C and the liquid phase undergoes thermal condensation to form progressively larger polyaromatic species, ethylene tar's relatively low mean molecular weight (typically 180 Da to 350 Da per mass spectrometric characterization reported in petroleum chemistry literature) requires a longer condensation pathway to reach the mesophase size threshold than does decant oil with a mean molecular weight of 300 Da to 500 Da. This extended condensation pathway consumes additional residence time and releases additional volatile matter during the mesophase growth phase, both of which act contrary to the requirements for optimal anisotropic structure development. The volatile matter released during the latter stages of the coking cycle contains substantial quantities of naphthalene, alkylnaphthalenes, and acenaphthylene — compounds with boiling points between 218°C and 280°C — that are recovered in the coker main fractionator as heavy coker naphtha and light coker gas oil. The presence of these aromatic compounds in the coker liquid products has a secondary consequence: they increase the aromaticity of the coker gas oil product, which, when recycled to the drum per the recycle ratio setting, has already been thermally cracked and therefore exhibits different mesophase behavior than fresh feed. The recycle stream from ethylene tar-derived coker gas oil is characterized by a higher aromaticity (70% to 80% aromatic carbon) and lower olefin content than the fresh ethylene tar feed, because the most reactive olefinic species have been consumed during the first pass through the drum.
Storage stability of ethylene tar prior to introduction into the delayed coking unit constitutes an additional operational limitation that is frequently underestimated in process design. Ethylene tar stored in atmospheric-pressure, cone-roof tanks at temperatures of 60°C to 120°C undergoes gradual oxidative polymerization and condensation reactions that increase its viscosity and promote the formation of a surface skin and a bottom sludge layer. The viscosity of ethylene tar stored at 80°C for 30 days has been reported in petroleum storage literature to increase by 20% to 50% relative to freshly produced material, with the magnitude of the increase correlated with the initial bromine number and the oxygen ingress rate of the tank vent system. The bottom sludge — comprising polymerized olefins, iron corrosion products, and entrained quench oil solids — can accumulate to depths of 0.5 m to 2 m in tanks of 15 m diameter over a 6-month storage period, necessitating periodic tank cleaning that involves either manual entry and mechanical removal or the use of heated crude oil wash cycles at 120°C to 150°C. The storage-related degradation of ethylene tar is mitigated in industrial practice through the use of nitrogen-blanketed, floating-roof tanks with heating coils sized to maintain 70°C to 90°C bulk temperature and suction heaters rated for 100 kW/m² to 200 kW/m² heat flux at the pump-out nozzle. Additionally, the transfer lines between the storage tanks and the coker feed drum are insulated and heat-traced with steam or hot oil at 120°C to 150°C, with line sizes selected to maintain a minimum transfer velocity of 0.5 m/s to prevent deposition of polymerized material on the pipe walls.
| Property | Test Method | Premium Needle Coke Specification | Regular Needle Coke Specification | Ethylene Tar Implication |
|---|---|---|---|---|
| Sulfur (calcined coke) | ASTM D4239-18 / ISO 12980:2019 | ≤0.5 wt% | ≤0.6 wt% | Feed sulfur ≤0.3 wt% required at 30 wt% ET blend |
| Ash (calcined coke) | ASTM D4422-13 | ≤0.1 wt% | ≤0.2 wt% | Iron contamination from storage must be <25 ppm |
| Real density (calcined coke at 1,400°C) | ASTM D2638-10 (reapproved 2021) | ≥2.13 g/cm³ | ≥2.10 g/cm³ | Volatile-rich ET reduces density if drum temp <440°C |
| Coefficient of thermal expansion | ISO 14420:2020 / ASTM E228-17 | ≤0.25 ppm/°C | 0.25–0.45 ppm/°C | Pressure excursions >0.6 MPa increase CTE |
| Electrical resistivity | ISO 10143:2019 | ≤6.5 μΩ·m | ≤8.0 μΩ·m | Sulfur >0.5 wt% increases resistivity |
| Feedstock QI | ASTM D2318-20 | ≤0.2 wt% | ≤0.5 wt% | Pyrolytic carbon QI requires depth filtration |
| Feedstock aromaticity (BMCI) | Calculated | ≥120 | ≥110 | ET alone meets BMCI; blend dilution must be monitored |
| Feedstock metals (Ni+V) | ASTM D5185-18 | ≤5 ppm | ≤10 ppm | ET typically meets; Fe from storage is main risk |
Operating at the confluence of furnace severity, drum pressure, and feed quality constraints, the delayed coking unit processing ethylene tar blends requires a rigorous process control strategy that integrates on-line analytical data with predictive coking models. The feed quality analyzer suite in needle coke service typically includes an on-line sulfur analyzer operating on the principle of pulsed X-ray fluorescence with a detection limit of 10 ppm sulfur, an on-line viscosity analyzer operating at the feed pump discharge temperature with a repeatability of ±2% of reading, and a density analyzer based on vibrating element technology with an accuracy of ±0.0005 g/cm³. These continuous measurements are supplemented by laboratory determinations of micro carbon residue per ASTM D4530-15 on 4-hour composite samples, quinoline insolubles per ASTM D2318-20 on 8-hour composites, and simulated distillation per ASTM D7169-20 on daily composites. The control system responds to feed quality excursions by adjusting the furnace outlet temperature within the range of 495°C to 510°C, the drum pressure setpoint within 0.35 MPa to 0.55 MPa, and the recycle ratio within 1.05 to 1.40, with the adjustment hierarchy determined by which constraint is closest to violation. When the feed sulfur content rises above the threshold that would produce calcined coke exceeding 0.6 wt% sulfur, the primary corrective action is a reduction in ethylene tar blend ratio — an adjustment that can be implemented within 2 to 4 hours through feed blending valve repositioning — rather than an increase in furnace temperature or a reduction in drum pressure, both of which compromise coke morphology. This feed-blending-based correction strategy is documented in refinery operations literature as the most effective, most rapidly implementable, and least disruptive response to ethylene tar quality excursions.
The coker gas oil and distillate products derived from ethylene tar-containing feeds exhibit distinctive compositional fingerprints that affect downstream processing and product disposition. The coker light gas oil, with a distillation range of 200°C to 350°C, contains elevated concentrations of naphthalene, methylnaphthalenes, and indene — aromatic species that require hydrotreating for sulfur and nitrogen removal before blending into diesel fuel pools or require dedicated disposition to aromatic extraction units. The coker heavy gas oil, boiling between 350°C and 520°C, exhibits an aromatic carbon content of 50% to 65% and a BMCI of 70 to 90 when derived from ethylene tar-containing feeds, values that make this stream unsuitable for fluid catalytic cracking feedstocks without severe hydrotreating to reduce aromaticity and increase hydrogen content. The main fractionator overhead product — coker naphtha with a boiling range of 30°C to 200°C — is characterized by high olefin and diolefin content (bromine number 50 to 100 g Br₂/100 g) that requires immediate stabilization and hydrotreating to prevent gum formation during storage. These product quality effects impose incremental hydrotreating capacity requirements that must be accounted for in the overall refinery economics of ethylene tar utilization. Published refinery configuration studies for ethylene tar processing indicate that the incremental hydrogen consumption for hydrotreating the coker liquid products from a 30 wt% ethylene tar feed blend amounts to 0.5 Nm³/m³ to 1.5 Nm³/m³ of fresh feed above the baseline decant oil case, primarily due to the higher olefin and nitrogen content of the coker distillates.
