Electrical cable filling compounds used in stranded power, control, and communication cables are formulated from low-polarity hydrocarbons, typically polyisobutylene or polybutene, blended with refined mineral oils, microcrystalline wax, and a thixotropic inorganic phase such as fumed silica or organoclay. Their primary function is to occupy interstitial voids between conductor strands, exclude moisture and oxygen, and suppress partial discharge initiation in the cable core. Dielectric stability after thermal cycling refers to the retention of volume resistivity, relative permittivity, dissipation factor, and AC breakdown strength after repeated temperature excursions between low-temperature contraction conditions and continuous operating conductor temperatures. In distribution cable designs rated from 1 kV to 36 kV, the filling compound is in direct contact with semiconductive conductor screens, crosslinked polyethylene insulation, and metallic shields, and its dielectric response therefore participates in the composite insulation system rather than operating as an isolated dielectric. Thermal cycling specifications for cable qualification commonly alternate between -25 °C and 90 °C, or between -40 °C and 105 °C depending on insulation class, with cycle durations of 4 h to 24 h. A compound that appears electrically benign at room temperature can exhibit progressive degradation of dielectric stability after 50 to 100 cycles because microstructural changes—oil exudation, wax recrystallization, polymer-filler network disruption, moisture condensation, and void coalescence—create continuous polar pathways or interfacial voids that increase dissipation factor and reduce breakdown strength. Measurement of dielectric stability is normally performed according to ASTM D257-14 for DC volume resistivity, ASTM D150-22a or IEC 62631-2-1 for relative permittivity and dissipation factor at power frequencies, and IEC 60243-1 for short-time electric strength. Guarded electrode cells with evaporated aluminum or conductive silver paint electrodes, Schering bridges or automated capacitance bridges, and temperature-controlled test chambers are required to resolve the small changes in tan δ that precede dielectric failure. Published data from compounding suppliers frequently report initial volume resistivity values of 1×10^12 Ω·cm to 5×10^14 Ω·cm at 23 °C, but after thermal cycling the lower bound can drop by one to two decades if oil separation, oxidative byproducts, or moisture ingress are not controlled. The technical problem is therefore defined not by the initial dielectric constant alone, but by the stability of the dielectric response after the filler network has been subjected to repeated phase transitions and thermomechanical stress.

What Limits Volume Resistivity After Repeated Thermal Excursions?

In low-polarity hydrocarbon filling compounds, volume resistivity is governed by ion mobility, electronic conduction through carbon or metallic contaminants, and field-enhanced impurity migration. Thermal cycling influences all three mechanisms. Ionic conduction becomes significant when residual moisture, metal soaps, or acid-functional oxidation products are present. The volume resistivity measured at 500 V DC after 60 s electrification in ASTM D257-14 can decline from 2×10^14 Ω·cm at 23 °C to 3×10^10 Ω·cm at 90 °C in a typical polyisobutylene-petrolatum system because the concentration of thermally dissociable ionic species increases and viscosity decreases by several orders of magnitude. Repeated cycling between -30 °C and 90 °C can produce irreversible changes in the distribution of polar oxidation products at the interface with copper conductors. Copper naphthenate formation, accelerated when free carboxylic acids are present, is a known contributor to elevated dissipation factor at low frequencies and at elevated temperature. Use of non-polar mineral oils with low aromatic content, typically <3 wt% aromatics by ASTM D2007 or equivalent clay-gel analysis, reduces ionic precursors. However, highly refined oils have lower solvency for the polymer and wax components, and phase separation can occur at low temperatures, creating oil-rich pockets with lower viscosity and different dielectric properties. Thermal cycling causes a pumping effect in stranded conductors: as copper expands and contracts with a linear thermal expansion coefficient of approximately 17×10^-6 K^-1, the compound is displaced from interstices and then reabsorbed at different rates depending on local viscosity and yield stress. If the compound lacks sufficient elastic recovery or yield stress above 50 Pa at 90 °C, gravity-assisted drainage and capillary flow can leave thin films or voids near the conductor. These void regions become sites for corona or partial discharge under voltage transients. The practical limit of volume resistivity after thermal cycling is therefore seldom governed by bulk polymer intrinsic properties; it is governed by interfacial degradation, ionic contamination, and void formation. To quantify this, cable manufacturers combine filling compound mechanical and thermal tests with dielectric measurements on compound-coated wire specimens before and after cycling. A change of volume resistivity greater than one decade, or an increase in dissipation factor greater than 0.005 at 50 Hz, is often treated as a rejection criterion in internal qualification specifications, although no single global standard defines a universal pass-fail value for all cable designs.

Syneresis and oil migration are the most common physical failure modes that reduce dielectric stability after thermal cycling. Microcrystalline wax and polyisobutylene form a gel network that immobilizes the liquid oil fraction; the degree of immobilization depends on wax crystallite size, polymer molecular weight distribution, and the cooling rate from processing temperatures. When the compound is cooled from a processing temperature of 110 °C to 130 °C to ambient, microcrystalline wax crystallizes into a three-dimensional network with crystalline lamellae typically between 2 µm and 10 µm in lateral dimension, as observed by polarized light microscopy. Slow cooling produces larger crystallites and a more open network, while rapid cooling produces finer and more uniform crystallites. Thermal cycling through the wax melting range, which for microcrystalline grades is approximately 60 °C to 85 °C, repeatedly destroys and reforms this network. Each cycle can produce slight syneresis, visible as oil droplets on the compound surface or as a thin oil film at the copper interface. The exuded oil has a relative permittivity near 2.2 to 2.4, while the bulk compound may have a relative permittivity of 2.3 to 2.8, so the immediate dielectric effect of pure oil separation can be small. However, the oil film acts as a trap for moisture and ionic debris, and it can penetrate into the semiconductive screen or insulation interface, altering the local electric field distribution. A standardized oil separation test based on ASTM D1742 or equivalent press filtration at 80 °C for 24 h exposes the compound to a static pressure differential, but it does not simulate the dynamic capillary forces generated by thermal expansion of stranded copper. Modified thermal cycling oil-separation tests are therefore used, in which a wire specimen is cycled between -20 °C and 100 °C inside a sealed glass tube, and any separated oil is collected and weighed after 20 cycles. A mass loss greater than 0.5 wt% is often associated with a measurable increase in tan δ of the bulk compound after cycling, particularly if the oil contains polar oxidation products. The combination of oil separation with moisture ingress is especially damaging; water has a high relative permittivity near 80 and high dielectric loss, and even 100 ppm of dispersed water in a non-polar medium can raise dissipation factor from 0.001 to 0.01 at low frequencies. Filling compounds intended for water-blocking applications are evaluated for water absorption by methods adapted from ASTM D570, and a mass gain above 0.1 wt% after 24 h immersion at 23 °C generally indicates excessive polar content or filler moisture. The interaction of syneresis, moisture ingress, and ionic contamination explains why single-property tests cannot adequately predict dielectric stability after prolonged cycling.

Microcrystalline Wax Content and Crystallization Shrinkage Boundaries

Microcrystalline wax content in polyisobutylene-based filling compounds is typically adjusted between 5 wt% and 15 wt% to provide drop point and heat resistance. The drop melting point measured by ASTM D127 rises from approximately 75 °C to 95 °C as wax content increases from 5 wt% to 12 wt%, but the same addition raises the crystallization shrinkage coefficient and increases brittleness at temperatures below -20 °C. Crystallization shrinkage is a critical boundary because cable compounds are forced into stranded interstices at elevated temperatures and then constrained by conductor geometry during cooling. If the compound undergoes excessive volumetric contraction, localized delamination from the conductor surface occurs, creating air-filled microvoids with breakdown strength below 3 kV/mm compared with the bulk breakdown strength of a well-processed hydrocarbon gel of 20 kV/mm to 40 kV/mm measured by IEC 60243-1. Differential scanning calorimetry at a heating rate of 10 K/min measures the wax melting endotherm and estimates the crystallinity that drives this shrinkage. A compound with a broad endotherm between 40 °C and 85 °C is generally less prone to void formation than one with a narrow melting peak because the phase transition is distributed over the thermal cycling range and does not create a single sharp contraction event. Selection of microcrystalline wax with high oil content and low n-paraffin content reduces large crystal growth, while addition of 2 wt% to 4 wt% of a low-density polyethylene or an ethylene-vinyl acetate copolymer can further refine crystal size through nucleation. However, copolymer addition changes the non-polar character and may raise dissipation factor at elevated temperature if the comonomer contains polar vinyl acetate units. Wax crystallite size refinement is measured by polarized light microscopy on a hot stage; the mean crystallite size after cooling at 5 K/min from 100 °C should remain below 10 µm to avoid visible oil separation and excessive light scattering. The same hot-stage test shows that repeated thermal cycles gradually increase the thickness of wax lamellae by Ostwald ripening, which reduces the network surface area available to immobilize oil. This coarsening is one reason why dielectric stability determined after 10 cycles may not predict performance after 100 cycles. Internal qualification programs therefore include extended cycling of 50 to 200 cycles before measuring tan δ and volume resistivity, with intermediate sampling at 10, 25, and 50 cycles to detect nonlinear degradation trends. Published data for specific formulations with exact crystallite dimensions and cycle counts are limited because such data are typically retained as proprietary compound supplier information.

The influence of fumed silica on dielectric stability is dominated by its surface chemistry, aggregate structure, and concentration. Hydrophilic fumed silica with a BET surface area of 200 m²/g to 380 m²/g creates a strong thixotropic network in hydrocarbon media through hydrogen bonding between surface silanol groups. This network prevents oil drainage and provides the yield stress required for vertical or overhead cable sections. Rheological measurements with a controlled-stress rheometer at 25 °C typically show a yield stress of 50 Pa to 500 Pa at silica loadings of 2 wt% to 5 wt%, depending on shear history and polymer molecular weight. The same silanol groups, however, are polar adsorption sites for water and low-molecular-weight polar contaminants. At filler loadings above approximately 3.5 wt%, the dissipation factor of the compound at 50 Hz and 23 °C can increase from below 0.001 to above 0.005, and the increase is more pronounced after humidity exposure at 85 % RH because adsorbed water contributes to Maxwell-Wagner-Sillars interfacial polarization. Surface treatment with dimethyldichlorosilane or hexamethyldisilazane converts many silanol groups to non-polar methyl groups, reducing moisture sensitivity and lowering dielectric loss, but it also reduces the yield stress at a given loading by diminishing hydrogen bonding. A formulation gradient is therefore required to balance dielectric stability and anti-sag behavior. The processing path is equally important: fumed silica must be dispersed under high shear to break down agglomerates; production-scale dispersion is typically performed in a twin-screw extruder with a length-to-diameter ratio of at least 40:1, using screw elements that generate extensional mixing, followed by vacuum devolatilization at residual pressure below 80 mbar to remove moisture and low-molecular-weight volatiles. Inadequate dispersion leaves agglomerates larger than 20 µm, which act as dielectric heterogeneities and can initiate partial discharge in high-voltage cable joints. Inadequate devolatilization leaves residual moisture above 200 ppm, which increases tan δ and reduces volume resistivity after thermal cycling. Batch-to-batch variability in silica moisture content and silanol density is a documented production issue; incoming silica should be tested for moisture by Karl Fischer titration and for surface area by ASTM D1993. The dielectric stability of the finished compound after thermal cycling is therefore not an inherent property of the raw materials alone but a consequence of the dispersion and devolatilization history.

When Partial Discharge Exposure Superimposes on Thermal Cycling in High-Voltage Joints

High-voltage cable joints and terminations often contain filling compounds in direct contact with geometric stress control elements, where the electric field distribution is non-uniform. In these configurations, thermal cycling is not the only stress; the compound may also experience partial discharge activity if microvoids develop during cooling cycles. The sequence of events leading to dielectric instability begins with void formation at the conductor or insulation interface due to shrinkage and syneresis, followed by localized field enhancement above the breakdown strength of the gas-filled void, which is typically below 3 kV/mm at atmospheric pressure. Partial discharge pulses then bombard the inner surfaces of the void with electrons and ions, producing reactive species that oxidize the hydrocarbon compound and deposit conductive carbonaceous residues. The apparent charge measured according to IEC 60270 may initially remain below 10 pC, but after repeated thermal cycles the discharge magnitude can increase to 50 pC or more as void dimensions grow and surfaces become more conductive. The oxidation products, including carboxylic acids, ketones, and conjugated unsaturation, increase dissipation factor and reduce volume resistivity. A cyclic test combining thermal cycling with partial discharge monitoring is therefore more severe than thermal cycling alone. The test cell consists of a needle-plane electrode embedded in the compound within a temperature-controlled chamber, with a sinusoidal voltage of 5 kV rms to 15 kV rms at 50 Hz or 60 Hz. Discharge inception voltage, extinction voltage, and pulse repetition rate are recorded at each temperature plateau. A compound with a high concentration of aromatic mineral oil may exhibit better gas absorption and slower carbonization than a purely paraffinic oil, but aromatic oils are restricted by health and environmental regulations under REACH and may be unsuitable for water-contact applications. The dielectric dissipation factor after combined partial discharge and thermal cycling is often worse than after thermal cycling alone by a factor of 2 to 10, depending on the discharge energy and the antioxidant package. Since published data for this specific configuration are limited, qualification tests typically rely on comparative evaluation of a candidate formulation against an approved reference compound under identical electrical and thermal stress.

Oxidative degradation during thermal cycling is a kinetic process that couples oxygen diffusion, antioxidant depletion, and free-radical chain branching in the hydrocarbon matrix. The rate of oxygen uptake increases exponentially with temperature, and the cyclic residence time at the upper temperature plateau controls the cumulative oxidation. Differential scanning calorimetry using ASTM D3895 measures oxidative induction time at a specified temperature, often 190 °C or 200 °C, but this is an accelerated isothermal test that may not fully represent cycling below 105 °C. In service, hydrocarbons degrade through a slower radical chain mechanism that forms hydroperoxides, which decompose to carbonyl compounds and alcohols. Fourier-transform infrared spectroscopy monitoring of the carbonyl absorption near 1720 cm^-1 can quantify oxidation after cycling, and an increase in carbonyl index greater than 0.05 absorbance units relative to the 1460 cm^-1 methylene reference is associated with increased dissipation factor and reduced volume resistivity. The antioxidant system for cable filling compounds usually combines a hindered phenol primary antioxidant and a phosphite secondary antioxidant, with total concentrations between 0.2 wt% and 0.6 wt%. Hindered phenols are effective radical scavengers but can form quinoid chromophores when oxidized, which are themselves polar and can modestly increase dielectric loss at high frequencies. Phosphites decompose hydroperoxides but are sensitive to hydrolysis, and their activity can be lost during storage at high humidity. The pre-drying requirement at relative humidity above 60 % is therefore relevant not only for the filler but also for the antioxidant package. Acid scavengers such as zinc stearate or epoxy-functionalized soybean oil are sometimes added at 0.1 wt% to 0.3 wt% to neutralize acidic processing residues and corrosion products from copper conductors. However, metal stearates can increase ionic conductivity at elevated temperature, and their concentration should be kept below the threshold at which the volume resistivity at 90 °C falls below 1×10^10 Ω·cm. The balance between oxidation resistance and dielectric loss is formulation-specific, and it is evaluated by measuring the dissipation factor at 50 Hz and 0.1 Hz before and after forced air aging at 100 °C for 168 h according to IEC 60216 or equivalent thermal endurance methods.

Thermal conductivity and heat transfer also influence dielectric stability because local temperature gradients control the distribution of liquid phases and the rate of void formation. Hydrocarbon filling compounds have thermal conductivities in the range of 0.15 W/(m·K) to 0.30 W/(m·K), significantly lower than copper at approximately 400 W/(m·K) and lower than XLPE insulation near 0.35 W/(m·K). During cable load cycling, the conductor reaches its maximum temperature while the outer insulation remains cooler, so the filling compound directly adjacent to the conductor experiences the highest temperature and the lowest viscosity. This gradient drives radial migration of low-molecular-weight oil fractions outward toward the insulation screen, where they can accumulate at the interface. The accumulation is often observed in cable dissection after thermal cycling as a thin, greasy film between the conductor screen and the insulation screen. The film itself may have volume resistivity above 1×10^12 Ω·cm, but it creates a discontinuity that can weaken mechanical coupling between semicon and insulation. The relevant test is not solely dielectric; mechanical compatibility tests such as tensile elongation at break and peel strength between semicon and insulation after aging provide supporting data. For medium-voltage cables, qualification testing according to IEC 60502-1 requires that non-metallic components be compatible with each other and that no component adversely affect the dielectric performance of the insulation. The standard does not prescribe a dedicated dielectric test for the filling compound alone, so cable manufacturers rely on composite tests on full cable cores. A full-size cable sample is subjected to load cycling at a conductor temperature of 90 °C or 105 °C for 8 h on load and 16 h off load, with dielectric loss or partial discharge measurements repeated after 10, 50, and 100 cycles. The interpretation of results requires separating dielectric losses in the insulation from those caused by the filling compound or interfacial contamination, which is often accomplished by comparative measurements on dry and filled cores or by high-resolution dielectric spectroscopy in the frequency range 10^-3 Hz to 10^6 Hz.

Dielectric spectroscopy across a wide frequency range provides more diagnostic information than single-frequency tan δ. The complex relative permittivity of a filling compound is governed by electronic polarization, atomic polarization, and dipolar or interfacial polarization. In a non-polar hydrocarbon matrix, the relative permittivity is expected to remain near 2.2 to 2.8 across the frequency range from 10^-3 Hz to 10^6 Hz, with a slight increase at low frequencies if ionic conduction contributes to loss. The dissipation factor tan δ is related to the loss index by the ratio ε″/ε′. In a well-processed compound, tan δ at 50 Hz and 23 °C is typically below 0.002, while at 90 °C it may rise to 0.05 because ionic mobility increases. After thermal cycling, the low-frequency tan δ may increase disproportionately, indicating the formation of interfacial polarization sites or the accumulation of ionic species at boundaries. The measurement requires a frequency response analyzer and a temperature-controlled parallel-plate or guard-ring cell with electrode separation of 0.5 mm to 2.0 mm. The applied voltage is kept below 1 V rms to avoid nonlinear effects. The frequency sweep is repeated at 23 °C, 60 °C, and 90 °C before and after cycling, and the activation energy of low-frequency conduction is calculated from the Arrhenius plot of conductivity versus reciprocal temperature. An activation energy below 0.3 eV suggests impurity conduction or moisture pathways, while values between 0.5 eV and 0.9 eV are more typical of intrinsic ionic conduction in hydrocarbon gels. Changes in activation energy after cycling can distinguish between bulk oxidation and interfacial contamination. Because the experimental setup is sensitive to electrode polarization, interpretation of low-frequency data requires subtraction of electrode effects using guarded configurations described in ASTM D257-14 and IEC 62631-3-1.

Processing Window Constraints in Continuous Compound Injection

The processing window for filling compound application is often narrower than the window for compounding itself. On a three-layer cable line with a screw-fed filling station, the compound is transferred from a heated reservoir through a gear pump and die into the interstitial area of the stranded conductor, typically just before the semiconductive conductor screen is applied. The reservoir temperature is maintained between 105 °C and 125 °C to reduce viscosity while avoiding thermal degradation of the polymer and antioxidant. If the temperature falls below 100 °C, microcrystalline wax begins to crystallize, apparent viscosity rises sharply, and the gear pump may cavitate or produce pressure spikes. If the temperature exceeds 135 °C, the mineral oil may begin to form volatile oxidation products and the antioxidant is consumed prematurely, causing dielectric loss to increase before any service aging has occurred. The allowable processing window is therefore approximately ±5 °C around the optimum injection temperature for some formulations, which is a critical threshold risk. Viscosity at the injection shear rate of 10 s^-1 to 100 s^-1 should be maintained between 20 Pa·s and 50 Pa·s; below this range, the compound drains from vertical strands before the screen is applied, and above this range, the compound does not fully penetrate the inner layers of the conductor. Production-scale experience with gear pumps of 5 cm³/rev to 20 cm³/rev displacement indicates that small changes in filler content—for example, an increase in fumed silica from 2.0 wt% to 3.0 wt%—can raise pressure at the die from 20 bar to 40 bar and require a higher reservoir temperature to maintain flow. The higher temperature may exceed the degradation threshold of the oil, so the formulation must be adjusted rather than simply raising the setpoint. This conflict between rheological processing and thermal stability is one of the main reasons that filling compounds are qualified with both rheological and dielectric tests before full-scale cable trials. Batch-to-batch variance in filler moisture or wax crystallinity can shift this processing window enough to produce intermittent void formation, which may not be detectable by routine dielectric tests until after thermal cycling.

Low-temperature extremes in thermal cycling introduce a different set of dielectric stability concerns. At temperatures below -20 °C, the hydrocarbon oil fraction may solidify or become highly viscous, causing the compound to shrink and embrittle. The glass transition temperature of polyisobutylene is near -65 °C, but mineral oils and microcrystalline wax can solidify or partially crystallize at much higher temperatures. The resulting differential contraction can produce cracks that propagate from the conductor interface into the bulk compound. These cracks may close during the heating cycle, but their surfaces may be contaminated with moisture or oxidation products and may not fully heal. The dielectric recovery after crack closure is often incomplete, and the volume resistivity measured at -20 °C after cycling can be lower than the initial value despite the low temperature. Low-temperature flexibility tests for filling compounds, such as those based on cold bending, evaluate whether the compound cracks under deformation at specified temperatures. These tests are usually mechanical rather than dielectric, but they are important because a mechanically cracked compound cannot function as a moisture barrier or partial discharge suppressant. A compound designed for arctic service may be formulated with a higher proportion of low-viscosity polybutene and a lower wax content to maintain flexibility at -40 °C, but this reduces the drop point and increases the tendency for oil separation at high temperature. The formulation conflict between low-temperature flexibility and high-temperature drip resistance is resolved by using higher molecular weight polyisobutylene, which improves elastic network formation without contributing to wax crystallization, but this raises processing viscosity. Selection of the polymer therefore has a direct effect on dielectric stability after thermal cycling because it determines whether the compound remains a cohesive gel with no voids or cracks across the entire temperature range.

Carbon black and other conductive fillers are generally avoided in cable filling compounds because they suppress volume resistivity and promote partial discharge. In semiconductive layers, carbon black is intentionally used to provide a surface resistance between 10^3 Ω and 10^5 Ω, but in filling compounds even trace carbon from processing equipment can reduce volume resistivity below acceptable limits. Cleaning of compounding equipment between batches is therefore critical. Production lines that alternate between conductive semiconductive compounds and non-conductive filling compounds must use dedicated screws, barrels, and pumps, or perform purging with a high-viscosity hydrocarbon purging compound. Cross-contamination of 0.1 wt% carbon black can reduce volume resistivity by two decades and increase tan δ at 50 Hz above 0.01. The dielectric stability after thermal cycling of a contaminated compound is poor because the carbon particles provide conductive bridges that are stable even when the hydrocarbon matrix shrinks or exudes oil. The purity of mineral oil and wax is also evaluated by measuring the dissipation factor of the liquid component at 100 °C according to ASTM D924-22. A liquid tan δ below 0.001 at 100 °C and 60 Hz indicates low polar contamination, while values above 0.01 suggest the presence of ionic residues, metal soaps, or excessive aromatic content. Solidification of wax around such contaminants can produce localized high-field regions because the solid wax phase has a slightly different permittivity than the oil phase. The resulting Maxwell-Wagner-Sillars polarization is frequency-dependent and can dominate tan δ in the 10 Hz to 100 Hz range. Dielectric stability after thermal cycling therefore begins with raw material purity and equipment segregation, not merely with final compound formulation.

Compatibility with crosslinked polyethylene and ethylene-propylene rubber insulation is a non-negotiable constraint. Hydrocarbon oils and low-molecular-weight fractions can migrate into XLPE insulation during thermal cycling and plasticize or swell the polymer. XLPE has a solubility parameter near 16.5 MPa^0.5 to 17.5 MPa^0.5, while mineral oil and polyisobutylene have lower solubility parameters; the difference is usually small enough for limited interaction, but high concentrations of low-molecular-weight oils can migrate under sustained thermal gradients. The migrated oil reduces the insulation glass transition temperature and can lower voltage endurance. The standard compatibility test for cable components is often based on weight change, tensile properties, and dielectric loss of insulation plaques after aging in contact with the filling compound at 100 °C for 168 h. A weight gain of XLPE greater than 3 wt% or a reduction in elongation at break greater than 30 % is generally considered unacceptable in internal cable manufacturer specifications. The dielectric stability of the filling compound itself can also be affected by migration of peroxide decomposition products from the insulation during the crosslinking process. Cumyl alcohol, acetophenone, and other polar byproducts of dicumyl peroxide decomposition can diffuse into the filling compound during cable manufacture and increase its dissipation factor. This contamination is often removed or reduced by conductor heating or by post-curing, but residual levels may remain in thick insulation walls. Thermal cycling accelerates redistribution of these byproducts, and their presence in the filling compound can be detected by an increase in tan δ at 23 °C and by characteristic IR absorptions near 1700 cm^-1 and 3400 cm^-1. The compound formulation can be made more resistant to this contamination by selecting mineral oils with low aromatic content and by avoiding fillers that adsorb polar species and then release them at high temperature.

Quality control of dielectric stability after thermal cycling requires a combination of batch-level raw material tests, in-process rheological tests, and finished-compound electrical tests. Mineral oil is tested for kinematic viscosity at 40 °C by ASTM D445, acid number by ASTM D974, and dissipation factor by ASTM D924-22. Microcrystalline wax is tested for drop melting point by ASTM D127 and oil content by ASTM D721. Fumed silica is tested for moisture content by Karl Fischer titration and for surface area by ASTM D1993. The finished compound is tested for cone penetration by ASTM D937, viscosity at 110 °C by rotational viscometer, volume resistivity by ASTM D257-14, and dissipation factor by ASTM D150-22a. For thermal cycling qualification, a reduced batch set is passed through the full cycling profile, and the electrical properties are compared with the initial values. A batch that exhibits a volume resistivity drop greater than one decade or a tan δ increase greater than 0.005 at 50 Hz and 23 °C is rejected. These limits are typically company-specific because no single global standard defines a universal pass-fail criterion for all cable filling compounds. Traceability of these measurements to national metrology institutes is maintained through calibrated resistance standards and capacitance standards, and the test cells are revalidated annually. This level of control is necessary because the filling compound, although a small mass fraction of the cable, contacts every conductor strand and forms part of the interfacial dielectric system that must remain stable across the entire thermal cycling envelope.