Multiphase Topology of Impact Polypropylene Copolymers and Its Effects on Electrical and Mechanical Performance for Recyclable HVDC Cable Insulation

1. Introduction

The development of high-voltage direct-current (HVDC) transmission has imposed increasingly stringent requirements on the electrical reliability, thermal stability, processability, and environmental compatibility of cable insulation materials [1,2]. Crosslinked polyethylene (XLPE) remains the principal insulation material for extruded power cables because of its excellent insulating properties and well-established industrial applications [3,4]. However, the crosslinked network in XLPE makes the material difficult to recycle by simple remelting, while the crosslinking and degassing steps increase manufacturing complexity [3,5]. Polypropylene (PP)-based insulation materials, which are thermoplastic and recyclable, have therefore emerged as important candidates for sustainable HVDC cable insulation because of their high melting temperature, good thermal stability, and low dielectric loss [2,6].

Despite its favorable electrical properties and processing advantages, neat isotactic polypropylene (iPP) has high rigidity and limited low-temperature toughness, which restrict its further use during cable manufacture, installation, and operation [4,5,7]. Impact copolymer polypropylene (ICP) improves impact resistance and flexibility by incorporating an ethylene–propylene rubber (EPR)-rich phase into a semicrystalline PP matrix [8,9]. However, introducing a multiphase structure also changes the spatial organization of the crystalline, amorphous, and rubber-rich regions and may affect interfacial polarization, charge injection, carrier transport, and space-charge accumulation [9,10,11,12]. Maintaining stable high-field insulation performance while preserving the mechanical flexibility of ICP is therefore a central challenge in designing PP-based HVDC cable insulation materials [8,13,14,15,16].

Considerable effort has been devoted to jointly optimizing mechanical flexibility and high-field insulation reliability in PP-based insulation materials through compositional control and structural design. These strategies include elastomer-phase regulation, blend design, crystalline-phase regulation, heat treatment, and interfacial-structure optimization [12,17,18,19,20,21,22]. Previous studies have shown that the electrical and mechanical responses of PP-based insulation materials depend not only on their overall composition, but also on chain-sequence regularity, the distribution of component crystallizability, crystalline structure, and the state of multiphase interfaces [10,16,23,24,25]. For reactor-grade ICP multiphase systems, however, systematic understanding remains limited regarding how changes in molecular sequence affect crystalline-framework formation, connectivity among the multiphase regions, and interfacial charge behavior [8,9,19]. In particular, the coupled evolution of the matrix-chain structure, crystalline organization, and phase-interface topology in ICP, together with its influence on high-field charge transport and macroscopic insulation performance, requires further clarification [8,9,19,26].

On this basis, three representative ICP materials were selected to establish a multiscale framework for structural analysis. Homopolymer-matrix impact copolymer polypropylene (HICP) represents a conventional homopolymer PP matrix. Random copolymer-matrix impact copolymer polypropylene (RICP) was prepared by introducing a small amount of ethylene during matrix polymerization to form a PP matrix with random-copolymer characteristics, thereby enabling examination of the effects of matrix-sequence changes on crystalline structure and multiphase topology. Commercial high-phase-continuity impact copolymer polypropylene (CICP) served as an industrial reference with a highly flexible multiphase structure, allowing the combined structural and property responses of a material containing a high proportion of low-crystallizability components and a high-phase-continuity organization to be examined. CICP differs from the laboratory-prepared HICP and RICP in composition, material origin, and industrial production system. It was therefore used primarily as a commercial high-phase-continuity reference rather than as a strictly controlled single-variable comparison.

Molecular composition and component-distribution analyses, crystalline-structure characterization, multiphase morphology observations, dynamic mechanical and rheological analyses, dielectric spectroscopy, high-field conduction measurements, thermally stimulated depolarization current (TSDC) measurements, pulsed electro-acoustic (PEA) space-charge measurements, and direct-current breakdown tests were combined to investigate the evolution of ICP from molecular structure and crystalline organization to multiphase topology and charge behavior. By establishing relationships among molecular sequence, crystalline framework, multiphase topology, and mechanical and electrical properties, this study aimed to elucidate how reactor-grade ICP multiphase structures regulate the performance of PP-based HVDC insulation and to provide a theoretical basis for the structural design of recyclable PP-based cable insulation materials.

2. Materials and Methods

2.1. Materials

The polypropylene materials, solvents, and auxiliary media used in this study are listed in Table 1. HICP and RICP were prepared at the Petrochemical Research Institute of PetroChina using an in-house-developed Ziegler–Natta catalyst system. CICP was the commercial Hifax CA10A grade supplied by LyondellBasell and was used without further modification.

2.2. Equipment

The principal instruments used for specimen preparation, structural characterization, and property testing are listed in Table 2.

2.3. Specimen Preparation

2.3.1. Polymerization of HICP and RICP

HICP and RICP were prepared in a 10 L two-stage stirred gas-phase polymerization unit. Before polymerization, the reactor was evacuated and purged three times with high-purity nitrogen. After the temperature reached 70 °C, an in-house-developed MgCl₂-supported TiCl₄ Ziegler–Natta catalyst from the Petrochemical Research Institute of PetroChina was added. Triethylaluminum was used as the cocatalyst, and dicyclopentyldimethoxysilane was used as the external electron donor. The Al/Ti and Al/Si molar ratios were 120 and 10, respectively.

The first-stage reaction was conducted at 70 °C and a total pressure of 2.30 MPa for 60 min. The hydrogen/propylene molar ratio was 4.0×10⁻³ for both materials during this stage. For HICP, only propylene was fed continuously to form a homopolymer PP matrix. For RICP, ethylene was continuously introduced into the propylene feed, and the feed was adjusted so that the random-copolymer matrix obtained in the first stage contained approximately 2.0 wt% ethylene.

After the first-stage reaction, the temperature was reduced to 65 °C and the total pressure was adjusted to 1.80 MPa for the second-stage gas-phase copolymerization of ethylene and propylene. For both materials, ethylene and propylene were fed continuously at a mass ratio of 60:40. The molar ratio of hydrogen to the combined amount of ethylene and propylene was maintained at 1.0×10⁻³, and the reaction was continued for 45 min. The second-stage monomer conversion was controlled so that the rubber-rich component accounted for approximately 20.0% of the total product mass in each of HICP and RICP. After completion, the reactor was rapidly depressurized and purged with high-purity nitrogen for 30 min to remove residual monomers. The resulting polymer powders were then melt-pelletized for subsequent use.

Commercial Hifax CA10A pellets supplied by LyondellBasell were used as CICP and were dried before subsequent specimen preparation and testing.

2.3.2. Preparation of Compression-Molded Specimens

Pellets of the three ICP materials were dried in a vacuum oven at 80 °C for 8 h. The dried pellets were distributed uniformly in a stainless-steel mold and preheated without pressure at 200 °C for 5 min to ensure complete melting. The material was then compression-molded at 15 MPa for 5 min, with one pressure-release step for venting. While the pressure was maintained, the molded material was cooled to room temperature with circulating cooling water at an average rate of 15 °C min⁻¹.

Specimens of different thicknesses were prepared according to the test requirements. Films approximately 0.20 mm thick were used for high-field conduction, broadband dielectric, TSDC, PEA, DC breakdown, and haze measurements. Sheets 1.0 mm thick were used for X-ray scattering, dynamic rheological, and dynamic mechanical measurements. Plates 4.0 mm thick were used for tensile, flexural, and impact tests. All specimens were conditioned at 23 °C and 50% relative humidity for 48 h before testing.

2.3.3. Preparation of SEM and POM Specimens

Extruded strands for scanning electron microscopy (SEM) were prepared with a melt-flow-rate tester at 230 °C under a load of 2.16 kg. The strand diameter was approximately 2 mm. The strands were cooled in liquid nitrogen for 10 min and then rapidly cryofractured. The fracture surfaces of HICP and RICP were first etched in xylene at 80 °C for 8 h and then etched at 60 °C for a further 16 h. The CICP fracture surface was etched in xylene at room temperature for 1 h. After etching, the specimens were rinsed sequentially with xylene and absolute ethanol, vacuum-dried at 40 °C for 12 h, sputter-coated with gold, and observed by SEM.

For polarized optical microscopy (POM), films approximately 30 μm thick were placed between a glass slide and a coverslip, and their crystalline morphology was recorded at room temperature.

2.4. Structural Characterization and Property Testing

2.4.1. Molecular Composition, Molecular Weight, and Component Distribution

¹³C NMR: An 80 mg sample was placed in a high-temperature NMR tube and completely dissolved in 3.0 mL of deuterated 1,2,4-trichlorobenzene at 120 °C. ¹³C NMR spectra were acquired using inverse-gated decoupling with a pulse angle of 90°, a relaxation delay of 10 s, and 4096 scans. The total ethylene content was calculated from the integrated areas of the characteristic carbon-resonance regions, and relative changes in the different sequence structures were compared.

GPC-IR: Each sample was prepared in 1,2,4-trichlorobenzene containing 0.0125 wt% of the antioxidant BHT at a concentration of 2.0 mg mL⁻¹ and dissolved at 150 °C for 90 min. Measurements were performed at 150 °C with a mobile-phase flow rate of 1.0 mL min⁻¹. Elution signals were recorded using an IR4 infrared detector. The number-average molecular weight (M_n), weight-average molecular weight (M_w), and molecular-weight distribution (PDI) were obtained by universal calibration.

CFC/TREF: Each sample was dissolved in 1,2,4-trichlorobenzene at 140 °C for 60 min and then cooled to 30 °C at 1 °C min⁻¹ to complete crystallization. After being held at 30 °C for 30 min, the system was subjected to programmed-temperature elution from 30 to 140 °C. An IR4 detector simultaneously recorded the concentration and molecular-weight distribution of the fractions eluting at each temperature. Fractions eluting at or below 35 °C were defined as the room-temperature soluble fraction (SF).

2.4.2. Thermal Properties and Crystalline Structure

Differential scanning calorimetry (DSC) was performed under high-purity nitrogen. A 5–8 mg sample was heated from room temperature to 200 °C at 10 °C min⁻¹ and held for 5 min to erase its thermal history. It was then cooled to 40 °C at 10 °C min⁻¹ to record crystallization and reheated to 200 °C at the same rate to record the second melting process. The apparent crystallinity was calculated using Equation (1), with 207 J g⁻¹ as the equilibrium enthalpy of fusion of 100% crystalline iPP.

Wide-angle X-ray scattering (WAXS) and small-angle X-ray scattering (SAXS) measurements used Cu Kα radiation with a wavelength of λ = 0.154 nm. The WAXS scanning range was 2θ = 10°–30°, with a step size of 0.02° and a scanning rate of 2° min⁻¹. Two-dimensional WAXS patterns were collected with an area detector using an exposure time of 300 s. The average crystallite size perpendicular to a specific crystallographic plane was calculated using the Scherrer equation. For SAXS, the scattering-vector range was 0.10–2.50 nm⁻¹. After background subtraction and Lorentz correction of the original curves, the long period (L_p) was calculated from the scattering-peak position. The mass crystallinity (X_c) was first converted to a mass fraction as w_c = X_c/100 and then to the volume crystallinity (φ_c) using crystalline- and amorphous-region densities of 0.936 and 0.850 g cm⁻³, respectively. The effective crystalline-layer contribution (d_c) and amorphous-layer thickness (d_a) in the long-period model were then calculated.

$$X_c={\displaystyle \frac{{\Delta H}_m}{{{\Delta H}_m}^0}}\times 100\%$$

(1)

$$D_{hkl}={\displaystyle \frac{K\lambda }{\beta cos\theta }}$$

(2)

$$L_p={\displaystyle \frac{2\pi }{q_{max}}}$$

(3)

$${\phi }_c={\displaystyle \frac{\frac{w_c}{{\rho }_c}}{\frac{w_c}{{\rho }_c}+\frac{1-w_c}{{\rho }_a}}}$$

(4)

$$d_c=L_p{\phi }_c,d_a=L_p-d_c$$

(5)

In these equations, ΔH_m is the melting enthalpy measured during the second heating scan, and ΔH_m⁰ is the enthalpy of fusion of fully crystalline iPP. X_c is the mass crystallinity, and w_c = X_c/100 is its fractional form. K was set to 0.89, β is the full width at half maximum of the diffraction peak after correction for instrumental broadening, and θ is the Bragg angle. q_max is the peak position in the Lorentz-corrected curve. ρ_c and ρ_a are the densities of the crystalline and amorphous regions of PP, respectively.

2.4.3. Morphological, Optical, Dynamic Mechanical, and Rheological Characterization

The etched fracture surfaces of HICP and RICP were observed using a JEOL JSM-7500F field-emission scanning electron microscope at an accelerating voltage of 1 kV. CICP was observed using an FEI Nova NanoSEM 450 at an accelerating voltage of 5 kV. At least 5 different regions of each sample were imaged. The crystalline morphology of the films described in Section 2.3.3 was recorded by POM at room temperature. Haze was measured using a BYK-Gardner haze meter. For each material, 5 films with thicknesses close to 0.20 mm were selected, and 3 positions on each film were measured and averaged.

Dynamic mechanical analysis (DMA) was performed in tensile mode at a frequency of 1 Hz and a heating rate of 3 °C min⁻¹ over a temperature range of −80 to 150 °C. The storage modulus (E′) and loss factor (tan δ) were recorded. Dynamic rheological measurements used 25 mm parallel plates with a gap of 1.0 mm at 200 °C, a strain amplitude of 1%, and an angular-frequency range of 0.1–600 rad s⁻¹. The complex viscosity (η*), storage modulus (G′), and loss modulus (G″) were recorded.

2.4.4. Electrical Properties and Charge Behavior

DC breakdown tests were performed in accordance with IEC 60243-1. Specimens approximately 0.20 mm thick were immersed in silicone oil and tested using brass sphere–sphere electrodes with a diameter of 25 mm. The voltage was increased continuously at 1 kV s⁻¹ until breakdown. At least 10 valid measurements were obtained for each material and analyzed using a two-parameter Weibull distribution. The characteristic breakdown field α corresponds to a cumulative failure probability of 63.2%, while the shape parameter β describes the data dispersion.

$$P(E)=1-exp[-{\left({\displaystyle \frac{E}{\alpha }}\right)}^{\beta }]$$

(6)

Here, P(E) is the cumulative breakdown-failure probability, E is the breakdown field, α is the characteristic breakdown field, and β is the Weibull shape parameter.

High-field conduction was measured in accordance with IEC 62631-3-1 using a three-electrode configuration and a Keithley 6517B electrometer. Measurements were performed at 30, 70, and 90 °C. The electric field was increased stepwise from 5 to 50 kV mm⁻¹, and the quasi-steady-state current was recorded after holding each field for 10 min. The current density j was calculated to obtain the j–E curves. Before broadband dielectric spectroscopy, gold electrodes 20 mm in diameter were evaporated onto both sides of each film. The relative permittivity (ε′) and dielectric loss (tan δ) were recorded from 10⁻¹ to 10⁶ Hz at 30, 70, and 90 °C.

For TSDC measurements, each specimen was heated to 140 °C and polarized under a DC electric field of 40 kV mm⁻¹ for 30 min. While the field was maintained, the specimen was rapidly cooled to −80 °C, short-circuited for 5 min, and then heated to 150 °C at 3 °C min⁻¹ while the depolarization current was recorded. The characteristic trap depth (E_t) was calculated using the half-peak-width method, and the volumetric trap density (N_t) was calculated by integrating the depolarization current.

$$E_t={\displaystyle \frac{2.47k{T_p}^2}{\Delta T}}$$

(7)

$$N_t={\displaystyle \frac{60\int I\left(T\right)dT}{{\beta }_{h}eSd}}$$

(8)

Here, k is the Boltzmann constant, T_p is the absolute temperature corresponding to the depolarization-current peak, ΔT is the peak width at half maximum, I(T) is the depolarization current, β_h is the heating rate, e is the elementary charge, S is the effective electrode area, and d is the specimen thickness. The factor of 60 in Equation (8) converts the heating rate from K min⁻¹ to K s⁻¹.

PEA measurements were performed at room temperature. A DC electric field of 40 kV mm⁻¹ was applied for 2400 s, after which the applied voltage was removed and the specimen was short-circuited for 600 s. The through-thickness space-charge-density distribution and its evolution with time were recorded continuously.

2.4.5. Mechanical Properties

Tensile properties were measured in accordance with GB/T 1040.2-2022 using Type 1A dumbbell specimens with nominal dimensions of 170 mm × 10 mm × 4 mm. A test speed of 1 mm min⁻¹ was used to determine the tensile modulus, and a speed of 50 mm min⁻¹ was used to determine the tensile strength and nominal strain at break. For each material, 5 replicate specimens were tested.

Flexural properties were measured in accordance with GB/T 9341-2008 using specimens measuring 80 mm × 10 mm × 4 mm, a span of 64 mm, and a test speed of 2 mm min⁻¹. The flexural modulus and flexural stress at the specified deflection were obtained. Charpy notched impact properties were measured in accordance with GB/T 1043.1-2008 using specimens measuring 80 mm × 10 mm × 4 mm with a notch depth of 2 mm. Tests were performed at 23, 0, and −20 °C using 10 replicate specimens under each condition.

3. Results and Discussion

3.1. Molecular Composition, Sequence Structure, and Component Distribution

Figure 1 presents the normalized ¹³C NMR spectra of HICP, RICP, and CICP. Based on the integrated areas of the characteristic carbon resonance regions associated with ethylene, the total ethylene contents of HICP, RICP, and CICP were 11.5 wt%, 13.6 wt%, and 23.3 wt%, respectively. The total ethylene contents of HICP and RICP differed by only 2.1 percentage points, providing a relatively similar compositional basis for comparing the effects of random copolymerization in the matrix on structure and properties. The higher ethylene content of RICP was consistent with the design in which ethylene was introduced during the first-stage propylene polymerization, indicating that part of the ethylene was incorporated into the matrix chains rather than being concentrated entirely in the rubber-rich component formed during the second stage. The total ethylene content of CICP was substantially higher than those of the other two materials, reflecting its larger proportion of low-crystallizability chain segments and rubber-rich components. CICP was therefore used as a commercial high-phase-continuity ICP reference to examine the structural response when composition and topology vary concurrently in an industrial material.

Figure 2 and Table 3 show that all three materials had the broad molecular-weight distributions characteristic of reactor-grade impact copolymer polypropylene. The number-average molecular weights of HICP, RICP, and CICP were 34.1, 38.4, and 37.5 kg mol⁻¹, respectively, and their weight-average molecular weights were 334.4, 411.8, and 314.6 kg mol⁻¹, corresponding to PDI values of 9.80, 10.73, and 8.39. The molecular-weight-distribution curves of HICP and RICP were generally similar. Although the M_w of RICP was slightly higher than that of HICP, the two values were of the same order of magnitude. Because the two laboratory-prepared samples were produced under the same principal polymerization conditions, this difference may be related to changes in the monomer-insertion environment around the active centers and in the balance between chain growth and chain transfer after a small amount of ethylene was introduced during the first stage of RICP synthesis; the difference may therefore be regarded as a concomitant change associated with random copolymerization. The subsequent structure-property analysis therefore focuses on the matrix sequence, crystallizability distribution, crystalline framework, and multiphase topology while also considering the modest difference in molecular weight. Although the M_w of CICP was lower than that of HICP, its complex viscosity and storage modulus at 200 °C were higher than those of HICP, as shown later by the melt-rheology results. This comparison indicates that the spatial organization of the multiphase components exerted a stronger influence on network constraints than the modest differences in molecular weight.

To further distinguish the crystallizability associated with different sequence structures, the three materials were fractionated by TREF and CFC. Table 4 summarizes the total ethylene content, the content of the soluble fraction eluted at 35 °C or below (SF), and the main matrix elution-peak temperature. HICP had an SF content of 18.92 wt% and a main elution peak at 125.0 °C, indicating a large proportion of highly regular iPP matrix chains with a relatively concentrated crystallizability distribution. The SF content of RICP increased to 27.21 wt%, while its main elution peak decreased to 112.0 °C. Relative to HICP, the total ethylene content of RICP increased by only 2.1 percentage points, whereas the SF content increased by 8.29 percentage points. This result indicates that the ethylene introduced during the first stage not only increased the total comonomer content but also substantially altered the sequence regularity of part of the matrix chains, shifting more chain segments into the low- and intermediate-crystallizability ranges. The SF content of CICP reached 73.49 wt%, while a high-temperature eluting fraction remained near 123.0 °C, indicating that CICP consisted predominantly of low-crystallizability or rubber-rich chain segments but also contained a small amount of highly regular PP chains.

Figure 3. TREF elution profiles of HICP, RICP, and CICP.

Figure 3. TREF elution profiles of HICP, RICP, and CICP.

The two-dimensional CFC contour maps further reveal the coupled distributions of elution temperature and molecular weight (Figure 4). The main signal of HICP was concentrated in the high-elution-temperature region, corresponding to the highly regular PP matrix, while a relatively distinct distribution appeared in the low-temperature soluble-fraction region. The center of the RICP distribution shifted toward lower temperatures, and a broader band of intermediate components appeared at approximately 80–100 °C, indicating that chain segments with different crystallizabilities extended continuously from the highly crystalline matrix toward the low-crystallizability components. In conjunction with the first-stage random-copolymerization route used for RICP, these intermediate components could provide a molecular basis for a more gradual change in segmental composition between the crystalline PP-rich and rubber-rich regions. CFC reveals the continuous distribution of chain-segment crystallizability; when considered together with the SEM and rheological results presented later, it can further clarify the roles of these intermediate components in interphase transitions and physical connectivity.

The CFC signal of CICP retained a small amount of highly crystallizable component at approximately 123 °C and showed a larger, continuously distributed intermediate band at lower temperatures. Thus, rubber-rich and low-crystallizability chain segments predominated, while a small amount of highly regular PP chains remained capable of forming crystalline supports. The molecular compositions of the three materials therefore exhibited a progressive change from the relative separation of the highly regular matrix and low-temperature components in HICP, through the increased fraction of intermediate-crystallizability chain segments in RICP, to the predominance of low-crystallizability components together with a small fraction of highly crystallizable chains in CICP. These differences in component distribution provided the molecular basis for the subsequent evolution of the crystalline framework, lamellar stacking, and multiphase interfacial organization.

3.2. Evolution of the Crystalline Framework and Lamellar Organization

Differences in molecular-sequence regularity and component crystallizability directly govern nucleation, crystal growth, and lamellar stacking during cooling. Figure 5 shows the DSC second-heating and cooling curves of the three ICPs, and the corresponding parameters are listed in Table 5. The melting-peak temperature, crystallization-peak temperature, and crystallinity of HICP were 166.4 °C, 116.4 °C, and 51.1%, respectively. Its relatively narrow melting and crystallization peaks indicate that the highly regular PP matrix rapidly formed a comparatively well-developed crystalline framework at elevated temperatures.

The melting- and crystallization-peak temperatures of RICP decreased to 147.2 °C and 107.0 °C, respectively, its crystallinity decreased to 32.1%, and its melting peak broadened markedly. These changes agree with the shift of the main CFC elution peak toward lower temperatures and the increase in intermediate-crystallizability components, indicating that the ethylene introduced during the first stage interrupted continuous isotactic propylene sequences, reduced the capacities for chain folding and lamellar thickening, and broadened the distribution of crystal perfection. The crystallinity of CICP was only 13.7%, yet its melting peak remained at 164.6 °C. The low melting enthalpy indicates a very small crystalline fraction, whereas the high melting-peak temperature indicates that the residual crystalline component still contained relatively long, highly regular propylene sequences. The thermal characteristics of CICP are therefore more appropriately described as sparse but thermally stable PP crystalline supports, rather than as crystals that are all substantially thinner or less perfect.

The one-dimensional WAXS results are shown in Figure 6a. All three materials exhibited the characteristic α-iPP (110), (040), and (130) reflections at 2θ ≈ 14.1°, 16.8°, and 18.5°, respectively, indicating that their crystalline frameworks were dominated by the α form. RICP also exhibited a β(300)-related diffraction signal at 2θ ≈ 16.1°, indicating that random copolymerization in the matrix not only reduced crystal perfection but also altered the polymorphic selection of some chain segments. The β-crystal signal reflects the greater diversity of the crystalline structure of RICP; its influence on charge behavior must be evaluated together with the crystallite dimensions, the state of the crystalline-amorphous interfaces, and the multiphase organization.

The crystallite coherence lengths D₁₁₀ calculated from the full width at half maximum of the (110) peak were 13.87, 11.75, and 12.44 nm for HICP, RICP, and CICP, respectively. HICP had the narrowest (110) peak and the largest D₁₁₀, consistent with its high crystallinity and comparatively well-developed crystalline framework. The peak width increased from 0.571° for HICP to 0.674° for RICP, with a corresponding decrease in D₁₁₀, indicating that random copolymerization markedly restricted the lateral coherence of the crystals. CICP had the lowest integrated diffraction intensity, but its D₁₁₀ remained 12.44 nm, which was higher than that of RICP. Thus, the principal characteristic of CICP was the substantial reduction in the amount of crystalline material, whereas the coherence length of its residual crystals was not the smallest among the three samples. This result is consistent with its low crystallinity and relatively high melting-peak temperature.

The two-dimensional WAXS patterns are shown in Figure 7. All three samples exhibited continuous, complete Debye-Scherrer rings without pronounced arc-like concentration in the azimuthal direction, indicating that the crystallites remained randomly oriented under the compression-molding conditions used in this study. The diffraction rings of RICP were comparatively diffuse, whereas the crystalline diffraction rings of CICP were superimposed on a stronger amorphous-scattering background. No pronounced difference in macroscopic orientation was observed among the three materials; the subsequent changes in properties were therefore primarily associated with crystalline content, polymorphic composition, lamellar stacking, and multiphase interfacial organization.

The Lorentz-corrected SAXS profiles were used to further characterize the long-period structure of the lamellar stacks. The long periods L_p obtained from the peak positions were 16.10, 13.47, and 14.60 nm for HICP, RICP, and CICP, respectively. The volume crystallinity φ_c was converted from the DSC mass crystallinity, and the effective crystalline-layer contribution d_c and amorphous-layer thickness d_a were then calculated using the linear two-phase long-period model; the results are listed in Table 6. For HICP, d_c was 7.84 nm and d_a was 8.26 nm. HICP also had the smallest SAXS peak full width at half maximum, indicating a relatively high crystalline-layer volume fraction and a comparatively narrow long-period distribution.

For RICP, d_c decreased to 4.05 nm, 48.3% lower than that of HICP, while d_a increased to 9.42 nm and the SAXS peak full width at half maximum increased to 0.310 nm⁻¹. These results indicate that random copolymerization in the matrix reduced the effective crystalline contribution along the lamellar stacking direction and broadened the long-period distribution, consistent with the broader DSC melting peak and smaller WAXS crystallite size. The crystalline framework changed from a thicker, more concentrated lamellar structure to more dispersed crystalline layers separated by wider amorphous regions, providing the structural conditions for the subsequent formation of roughened and more gradual interphase transition regions.

The effective crystalline-layer contribution of CICP was 1.84 nm, and its amorphous-layer thickness reached 12.76 nm. This value was calculated from the long period and volume crystallinity of the whole sample and represents the mean effective contribution of the crystalline layers within the multiphase long-period model. Because the crystallinity of CICP was only 13.7%, this parameter primarily reflects the sparse distribution of the crystalline framework within the overall structure. Together with D₁₁₀ = 12.44 nm and a melting-peak temperature of 164.6 °C, these results show that the residual PP crystals retained appreciable dimensions and thermal stability but were dispersed at a low volume fraction within the large population of low-crystallizability components.

The POM results reflect these differences in crystalline organization at a larger length scale (Figure 8). The HICP field of view showed strong birefringent textures and comparatively well-developed spherulitic structures, indicating that its highly regular matrix had a strong capacity for crystal growth during cooling. The birefringent units of RICP were markedly finer, with smaller bright regions and overlapping boundaries, consistent with its lower crystallization temperature, smaller crystallite size, and broader long-period distribution. The CICP field of view was dark overall and showed only dispersed, weak birefringent textures, indicating that the small crystalline fraction could not form large, complete spherulites and instead existed mainly as dispersed crystalline regions within the large population of low-crystallizability components.

Taken together, the DSC, WAXS, SAXS, and POM results show that HICP formed an abundant, comparatively well-developed crystalline framework with a relatively concentrated size distribution; random copolymerization in the RICP matrix lowered the crystallization temperature, reduced the crystallite size and effective crystalline-layer contribution, and broadened the lamellar distribution; and CICP contained low-volume-fraction, dispersed PP crystalline supports embedded in a large population of low-crystallizability chain segments. This sequence of crystalline organizations subsequently altered the connectivity and interfacial states of the PP-rich and rubber-rich regions on the micrometer scale.

3.3. Multiphase Morphology and Network Constraints

The spatial distribution of the crystalline framework and low-crystallizability components ultimately produced distinct multiphase morphologies on the micrometer scale. Figure 9, Figure 10 and Figure 11 show the cryofractured surfaces of HICP, RICP, and CICP, respectively, after selective xylene etching. HICP and RICP were subjected to the same stepwise etching conditions. Because prolonged high-temperature etching caused pronounced swelling and deformation of the residual CICP structure, CICP was etched at room temperature for 1 h to preserve the overall fracture-surface morphology. In view of these differences in specimen preparation, the SEM results were used to compare phase-domain connectivity, interfacial morphology, and phase continuity; pore size and porosity were not quantitatively compared among the samples.

The HICP fracture surface contained relatively isolated, round-edged phase domains within a comparatively continuous, dense matrix. The boundaries of some domains were clearly defined, producing a typical discrete sea-island morphology. In conjunction with the high crystallinity, large effective crystalline-layer contribution, and concentrated high-temperature eluting fraction of HICP, this morphology can be understood as discrete rubber-rich domains enclosed by a highly regular PP matrix. The relatively abrupt phase-domain boundaries make spatial heterogeneity in dielectric response and electrical conduction more likely, providing structural sites for the interfacial polarization and local charge accumulation discussed subsequently.

Discrete phase domains remained identifiable on the RICP fracture surface, but the surface was markedly rougher. Plate-like, fibrillar, and folded structures were widely distributed at the edges of the phase domains and throughout the matrix, and the comparatively smooth boundaries observed in HICP were replaced by a more irregular transitional morphology. This feature is consistent with the increase in the intermediate-crystallizability components at 80–100 °C in the CFC results, the reduced effective crystalline-layer contribution determined by SAXS, and the broader long-period distribution. The low-crystallizability chain segments and refined crystalline regions together increased the spatial extent of interpenetration between segments of different phases. Consequently, the PP-rich and rubber-rich regions no longer exhibited only simple closed interfaces, but showed a stronger tendency toward physical connectivity and interfacial bridging.

The mildly etched fracture surface of CICP displayed labyrinth-like interconnected pores and a continuous residual framework throughout the field of view, rather than the isolated circular domains that predominated in HICP. This two-dimensional morphology is consistent with the 73.49 wt% low-temperature soluble fraction, low crystallinity, and large proportion of amorphous chain segments of CICP, indicating high spatial continuity of the low-crystallizability/rubber-rich component, with a small number of PP crystalline regions dispersed in the continuous phase. The two-dimensional morphology is consistent with a continuous or co-continuous multiphase organization, and the fractionation results together with the rheological results discussed below further support the increased phase continuity.

The haze data provide a macroscopic optical response to the changes in characteristic scale of the multiphase morphology (Table 7). The three films had similar thicknesses, and the haze values of HICP, RICP, and CICP were 97.3%, 78.3%, and 64.7%, respectively. The large-scale crystalline organization and discrete phase domains of HICP produced pronounced refractive-index fluctuations, resulting in the highest haze. In RICP, crystal refinement and the broader interfacial transition regions reduced the concentration of large-scale scattering centers. CICP had the lowest haze, indicating a further reduction in the large-scale structural heterogeneity responsible for strong visible-light scattering. The change in haze was consistent with the crystal refinement and increased phase continuity observed by POM and SEM and provides auxiliary evidence for homogenization of the multiphase morphology.

Differences in the multiphase morphology and crystalline framework further altered segmental relaxation and dynamic load-bearing capacity. Figure 12 and Table 8 present the DMA results. The storage moduli of HICP at −50, 25, and 100 °C were 4559, 1760, and 372 MPa, respectively, and were the highest among the three materials at all three temperatures, reflecting the strong constraint imposed on deformation by its high crystallinity and continuous PP crystalline framework. At the same temperatures, the storage moduli of RICP decreased to 3043, 983, and 186 MPa, indicating that crystal refinement and the increased amorphous fraction substantially reduced the material stiffness. The storage moduli of CICP at −50, 25, and 100 °C were only 2574, 74.9, and 5.4 MPa, respectively, indicating that the large proportion of rubber-rich components dominated its mechanical response above room temperature, while the sparse PP crystalline regions provided only limited physical support.

HICP exhibited relaxation responses of the rubber-rich phase and the amorphous regions of the PP matrix near −31.2 and 18.3 °C, respectively. The two peaks were relatively distinct, consistent with its discrete multiphase morphology. The corresponding peaks of RICP shifted to −38.0 and 17.9 °C, respectively, and increased in intensity, indicating that the increased population of low-crystallizability chain segments raised the fraction of amorphous segments participating in relaxation and altered the local motional environment of the rubber-rich regions. CICP exhibited a dominant, broad, intense relaxation peak near −8.4 °C and retained a shoulder at approximately −36.2 °C, indicating substantial overlap among the relaxation processes of different chain segments. This broadening and overlap indicate smaller phase-domain dimensions and a more continuous distribution of segmental motional environments; their structural significance must be interpreted together with the SEM and rheological results.

In the high-temperature region, the α* relaxation peaks of HICP and RICP occurred at 111.4 and 96.0 °C, respectively. The lower peak temperature of RICP was consistent with its refined lamellae, lower crystallinity, and weaker constraints from the crystalline regions. CICP exhibited a diffuse high-temperature response at approximately 129.9 °C. This peak reflects the combined contributions of residual crystalline supports, network relaxation, and the geometric stability of the specimen; the peak position itself does not represent the overall degree of crystal perfection.

Dynamic melt rheology further characterized the long-range relaxation of the molecular chains and multiphase components after crystal melting (Figure 13). At 200 °C, all three materials exhibited shear-thinning behavior, while the complex viscosity and storage modulus consistently followed the order CICP > RICP > HICP. The molecular-weight distributions of HICP and RICP were generally similar, and the slightly higher M_w of RICP may have contributed in part to its higher melt viscosity. When considered together with the CFC, crystalline-structure, and SEM results, the stronger rheological response of RICP was also associated with the increased population of low-crystallizability chain segments, a broader spatial range of interphase chain interpenetration, and enhanced interfacial connectivity. Because CICP had a lower weight-average molecular weight than HICP but exhibited the highest complex viscosity and storage modulus, its enhanced melt elasticity cannot be explained by average molecular weight. Instead, it was more likely associated with long-range entanglement of the low-crystallizability/rubber-rich components in the melt and interactions among the components.

The low-frequency region corresponds to segmental relaxation over longer time scales. The higher η* and G′ values of CICP and RICP in this region indicate stronger constraints on long-range molecular-chain relaxation. This behavior is consistent with the continuous distributions of low- and intermediate-crystallizability components revealed by CFC and the increased interfacial connectivity observed by SEM. Rheology characterizes the physical network and entanglement state in the melt; together with molecular fractionation, crystalline organization, and etched morphology, these results can further confirm the evolution from discrete phase domains to interfacial bridging and a high-phase-continuity organization across the three materials.

Differences in the molecular compositions of the three materials were successively reflected in their crystalline frameworks, micrometer-scale phase morphologies, and segmental dynamics. HICP comprised discrete rubber-rich domains enclosed by a highly regular, highly crystalline PP matrix and therefore exhibited relatively well-defined interfaces and high stiffness. Random copolymerization in the RICP matrix increased the population of intermediate-crystallizability chain segments, refined the crystalline framework, and broadened the interphase transition regions, producing a stronger tendency toward interfacial bridging. CICP formed a high-phase-continuity organization comprising a continuous low-crystallizability/rubber-rich phase and dispersed PP crystalline regions. These differences provided the structural basis for the subsequent changes in charge trapping, detrapping, and high-field transport.

3.4. Topology-Related Trap Distribution and High-Field Carrier Transport

Different topological structures alter the spatial connectivity between crystalline and amorphous regions and between PP-rich and rubber-rich regions, thereby affecting the polarization response, charge-carrier injection, trapping–detrapping processes, and high-field migration behavior. To establish the relationship between structural evolution and charge transport, broadband dielectric spectroscopy, high-field conduction, and thermally stimulated depolarization current results are analyzed together in this section.

Figure 14 shows the broadband dielectric responses of the three materials at 30, 70, and 90 °C. The relative permittivities of all three ICPs remained within the low range of 2.01–2.15 over 10⁻¹–10⁶ Hz, reflecting the weakly polar nature of polyolefin systems. At 30 °C, the relative permittivities of HICP, CICP, and RICP were approximately 2.14, 2.09, and 2.07, respectively. After the temperature increased to 70 and 90 °C, the values decreased for all three materials and followed the order HICP > RICP > CICP. The higher crystalline content and more concentrated density fluctuations in HICP increased the polarizable response per unit volume. In RICP and CICP, the increased proportion of low-crystallizability chain segments and the more dispersed crystalline framework reduced the overall polarizability. The relative permittivity of CICP remained low at elevated temperatures, indicating that its high proportion of low-crystallinity phases did not translate into a pronounced increase in the overall polarization response within the tested frequency range.

The dielectric loss generally remained on the order of 10⁻⁵–10⁻³ at 30 and 70 °C, and the fluctuations over some frequency ranges approached the instrumental resolution limit for low-loss measurements. At 90 °C, HICP showed a pronounced low-frequency upturn, with the loss approaching 5 × 10⁻³ near 0.1 Hz, whereas RICP and CICP remained primarily on the order of 10⁻³. The temperature sensitivity of the low-frequency loss was mainly associated with the long-range migration of thermally activated charge carriers and interfacial conduction polarization. The weaker high-temperature, low-frequency losses of RICP and CICP are consistent with their lower quasi-steady-state current densities discussed below, indicating stronger restriction of effective charge transport in these two materials at elevated temperatures.

Figure 15 presents the current-density–electric-field (j–E) curves of the three materials at 30, 70, and 90 °C. On double-logarithmic coordinates, all curves exhibited two regions: an approximately linear response at low fields and a nonlinear increase at high fields. This behavior indicates that electrode injection, trap occupancy, and charge-carrier migration changed concurrently with increasing electric field. Crystalline, amorphous, and rubber-rich regions coexist within the ICPs; therefore, the transition in each curve represents a characteristic field arising from the combined effects of electrode injection, trap occupancy, and migration across multiple microregions.

Across all tested temperatures and electric-field ranges, the current density consistently followed the order HICP > RICP > CICP. At 30 °C, the characteristic transition fields of HICP, RICP, and CICP were 14.8, 24.3, and 34.2 kV mm⁻¹, respectively. At 70 °C, they decreased to 11.6, 19.5, and 25.4 kV mm⁻¹, respectively, and at 90 °C they further decreased to 9.7, 15.6, and 19.1 kV mm⁻¹. Increasing temperature intensified segmental thermal motion and increased the probability of charge-carrier detrapping, shifting the transition field of all three materials toward lower fields, while the topology-related order remained unchanged.

At 30 °C and 50 kV mm⁻¹, the current densities of HICP, RICP, and CICP were approximately 1.5 × 10⁻⁷, 5.5 × 10⁻⁸, and 9.5 × 10⁻⁹ A m⁻², respectively. Relative to HICP, the current density was lower by approximately 2.7-fold for RICP and 15.8-fold for CICP. In HICP, the relatively abrupt interfaces of the discrete phases and the concentrated crystalline-amorphous boundaries provide structural locations for local-field enhancement and charge accumulation. In RICP, the intermediate-crystallizability chain segments and roughened interfaces subject charge carriers to more trapping and scattering events as they cross different microregions. In CICP, the low-crystallinity/rubber-rich phase is continuously distributed, whereas a small population of PP crystalline regions is present as dispersed rigid domains. The constrained crystalline-amorphous transition segments around the crystalline regions, together with the differences in composition and packing that develop from PP-rich to ethylene-rich regions, provide local sites with different binding strengths for charge carriers. Meanwhile, the reduction in abrupt isolated interfaces also weakens local-field concentration and interfacial injection. When considered together with the subsequent TSDC and PEA results, regulation of electrode-interface injection and delayed bulk trapping–detrapping in CICP jointly suppress effective charge-carrier transport, resulting in the lowest high-field current density.

Table 9. High-field-conduction parameters of HICP, RICP, and CICP at different temperatures.

Sample	Transition field at 30 °C (kV mm−1)	Transition field at 70 °C (kV mm−1)	Transition field at 90 °C (kV mm−1)	j at 30 °C and 50 kV mm−1 (A m−2)
HICP	14.8	11.6	9.7	1.5 × 10−7
RICP	24.3	19.5	15.6	5.5 × 10−8
CICP	34.2	25.4	19.1	9.5 × 10−9

Table 9. High-field-conduction parameters of HICP, RICP, and CICP at different temperatures.

Sample	Transition field at 30 °C (kV mm−1)	Transition field at 70 °C (kV mm−1)	Transition field at 90 °C (kV mm−1)	j at 30 °C and 50 kV mm−1 (A m−2)
HICP	14.8	11.6	9.7	1.5 × 10−7
RICP	24.3	19.5	15.6	5.5 × 10−8
CICP	34.2	25.4	19.1	9.5 × 10−9

TSDC further reveals the charge-trapping and detrapping characteristics associated with the transport differences described above (Figure 16). The trap density obtained from TSDC represents the effective trap population that participates in charge capture and thermal release under the specified polarization conditions. HICP exhibited a narrow, intense main peak near 92.5 °C, with a characteristic trap depth of 0.92 eV and a volumetric trap density of 2.84 × 10¹⁶ m⁻³. The main peak of RICP shifted to 105.0 °C and its peak current decreased; the trap depth increased to 1.05 eV and the trap density was 1.35 × 10¹⁶ m⁻³. CICP exhibited a broad, gradual high-temperature peak at approximately 136.6 °C while retaining a weaker release peak in the low-temperature region; its characteristic trap depth and volumetric trap density reached 1.25 eV and 2.69 × 10¹⁶ m⁻³, respectively. From HICP to RICP and then to CICP, the main peak shifted continuously toward higher temperatures and progressively broadened, indicating an increase in the average detrapping barrier that charge carriers must overcome and a transition from a relatively concentrated to a broader trap-energy distribution.

The trap parameters correspond to the structural changes described above. The highly regular matrix and discrete-phase interfaces of HICP correspond to a relatively concentrated and shallow TSDC response. In RICP, crystal refinement and the increased number of intermediate-crystallizability chain segments diversify the segmental constraints in crystalline-amorphous and interphase transition regions. In CICP, a small population of PP crystalline regions is dispersed within the continuous low-crystallinity phase. Chain segments surrounding the crystalline regions that can neither crystallize completely nor relax freely, together with the compositional and packing differences formed as PP-rich chain segments transition toward ethylene-rich chain segments, may constitute charge-trapping sites with different binding strengths. The different amounts of thermal energy required for charge detrapping from these sites are consistent with the marked shift and broadening of the CICP main peak toward higher temperatures.

Table 10. Characteristic TSDC parameters of HICP, RICP, and CICP.

Sample	Main-peak temperature, Tp (°C)	Characteristic trap depth, Et (eV)	Trap volume density, Nt (m−3)	Peak-shape characteristic
HICP	92.5	0.92	2.84 × 1016	Narrow and concentrated
RICP	105.0	1.05	1.35 × 1016	Shifted to higher temperature with lower peak amplitude
CICP	136.6	1.25	2.69 × 1016	Broad high-temperature peak

Table 10. Characteristic TSDC parameters of HICP, RICP, and CICP.

Sample	Main-peak temperature, Tp (°C)	Characteristic trap depth, Et (eV)	Trap volume density, Nt (m−3)	Peak-shape characteristic
HICP	92.5	0.92	2.84 × 1016	Narrow and concentrated
RICP	105.0	1.05	1.35 × 1016	Shifted to higher temperature with lower peak amplitude
CICP	136.6	1.25	2.69 × 1016	Broad high-temperature peak

Under the approximation of thermally activated detrapping, the characteristic residence time of a charge carrier in a trap can be expressed as:

$${\tau }_t={\tau }_{0}exp\left({\displaystyle \frac{E_t}{k_{B}T}}\right)$$

(9)

Here, τ₀ is the attempt time, k_B is the Boltzmann constant, and T is the absolute temperature. At 303 K and for comparable attempt frequencies, increasing the trap depth from 0.92 to 1.05 eV increases the characteristic residence-time scale by approximately 1.45 × 10²; increasing it to 1.25 eV increases the scale by approximately 3.06 × 10⁵ relative to HICP. This estimate reflects the exponential amplification of thermally activated detrapping by the trap barrier, while the actual residence process is also jointly affected by the trap density, energy-level distribution, and local segmental motion.

The high-field current can be expressed in terms of the free-charge-carrier concentration and effective mobility as:

$$J=q{n}_f{\mu }_{eff}E$$

(10)

Here, q is the charge carried by a charge carrier, n_f is the free-charge-carrier concentration, and μ_eff is the effective mobility incorporating trapping–detrapping effects. Deeper traps prolong the residence time of trapped charge carriers and reduce the effective time available for long-range drift, thereby decreasing n_f and μ_eff. For RICP and CICP, early-stage charge trapping near the interface may also modify the local electric field at the electrode-dielectric interface and affect subsequent injection. The j–E and TSDC results jointly indicate that both bulk migration limitation and interfacial injection regulation contribute to current suppression, with their relative contributions varying with electric field and temperature.

3.5. Space-Charge Evolution and DC Breakdown Response

The PEA results directly show how the trap and transport differences described above affect the distribution of charge in the bulk (Figure 17). After polarization for 2400 s under a DC electric field of 40 kV mm⁻¹, HICP developed a pronounced charge-density gradient through the thickness, with a maximum local space-charge density of approximately 65.5 C m⁻³; appreciable residual charge remained observable after short-circuiting. Charge penetration into the bulk was markedly reduced in RICP, and its maximum space-charge density decreased to approximately 28.4 C m⁻³. CICP maintained low bulk charge levels during both polarization and short-circuiting, with a maximum below 12.5 C m⁻³. Relative to HICP, the maximum space-charge density was reduced by approximately 56.6% in RICP and by at least 80.9% in CICP.

TSDC and PEA respectively characterize the thermally releasable effective trap population after high-temperature polarization and the net charge distribution through the thickness at room temperature; these two sets of results must be considered together with the high-field current. In HICP, the shallow traps provide weak long-term confinement of charge carriers, allowing charge to penetrate readily into the bulk. The interfacial transition regions and deeper traps in RICP reduce the charge-migration depth. CICP simultaneously exhibits the lowest high-field current, the deepest TSDC response, and the lowest bulk space charge, indicating strong suppression of both electrode injection and effective bulk transport.

The effect of space charge on the local electric field can be described by the one-dimensional Poisson relation:

$${\displaystyle \frac{d{E}_{sc}\left(x\right)}{dx}}={\displaystyle \frac{{\rho }_{sc}\left(x\right)}{\epsilon }}$$

(11)

$$E_{loc}\left(x\right)=E_0+E_{sc}\left(x\right)$$

(12)

Here, ρ_sc(x) is the space-charge density, ε is the permittivity, E₀ is the applied average field, and E_sc(x) is the additional field generated by space charge. Greater space-charge density and a steeper spatial gradient produce a stronger additional local field and a larger deviation between the actual peak field within the material and the applied average field. Therefore, the lower bulk space charge in RICP and CICP provides a direct charge-transport basis for mitigating local-field distortion and delaying high-field failure.

Figure 18 shows the Weibull distributions of DC breakdown strength for the three materials. The characteristic breakdown strengths of HICP, RICP, and CICP were 226.2, 310.8, and 384.4 kV mm⁻¹, respectively. Relative to HICP, the values increased by 37.4% for RICP and 69.9% for CICP. Notably, the total ethylene contents of HICP and RICP differed by only 2.1 percentage points. The pronounced increase in breakdown strength of RICP despite its reduced modulus strongly indicates that crystal refinement, interfacial transition, and trap reconstruction induced by matrix randomization improve high-field stability.

The Weibull shape parameters of the three materials were 14.8, 5.6, and 10.5, respectively, and did not vary monotonically with characteristic breakdown strength. The higher β value of HICP indicates a more concentrated distribution of breakdown data, although its characteristic breakdown strength was the lowest. The lower β value of RICP reflects the greater dispersion of microscopic failure paths introduced by the randomized matrix and multilevel transition structure. CICP retained an intermediate β value while achieving the highest breakdown strength, indicating favorable breakdown stability.

A combined comparison of the TSDC, PEA, and breakdown data shows that, as the trap depth increased from 0.92 eV to 1.05 and 1.25 eV, the maximum space-charge density decreased from 65.5 C m⁻³ to approximately 28.4 and below 12.5 C m⁻³, while the characteristic breakdown strength increased from 226.2 kV mm⁻¹ to 310.8 and 384.4 kV mm⁻¹. Charge transport is controlled by both electrode-interface injection and migration within the material bulk. Deeper traps prolong the localized residence time of charge carriers and reduce the effective time available for long-range drift, whereas the reduction in abrupt isolated interfaces helps mitigate local-field concentration and reduce net injection. Consequently, the bulk space charge and its additional electric field decrease, delaying breakdown. Within the sample set investigated here, trap deepening, reduced space charge, and enhanced breakdown form a consistent semiquantitative association. Differences in composition and material origin between CICP and the laboratory-prepared materials also contribute to this overall response.

Table 11. Relationships among trap, space-charge, and DC-breakdown parameters of HICP, RICP, and CICP.

Sample	Et (eV)	Nt (m−3)	Maximum space-charge density (C m−3)	Characteristic breakdown strength (kV mm−1)	Weibull β
HICP	0.92	2.84 × 1016	65.5	226.2	14.8
RICP	1.05	1.35 × 1016	28.4	310.8	5.6
CICP	1.25	2.69 × 1016	12.5	384.4	10.5

Table 11. Relationships among trap, space-charge, and DC-breakdown parameters of HICP, RICP, and CICP.

Sample	Et (eV)	Nt (m−3)	Maximum space-charge density (C m−3)	Characteristic breakdown strength (kV mm−1)	Weibull β
HICP	0.92	2.84 × 1016	65.5	226.2	14.8
RICP	1.05	1.35 × 1016	28.4	310.8	5.6
CICP	1.25	2.69 × 1016	12.5	384.4	10.5

3.6. Mechanical–Electrical Synergy and Multiscale Mechanism

Regulation of the multiphase topology ultimately produces a coordinated response in flexibility and high-field insulation stability. Table 12 summarizes the macroscopic mechanical properties of the three materials. The tensile modulus and flexural modulus of HICP were 1000 and 962 MPa, respectively, reflecting the high stiffness imparted by its highly regular PP matrix. For RICP, these two moduli decreased to 612 and 590 MPa, representing reductions of 38.8% and 38.7%, respectively, relative to HICP, while the nominal strain at break increased from 500% to 610%. These changes are consistent with the lower crystallinity of RICP, the reduced contribution of effective crystalline lamellae, the increase in interphase transition segments, and the decrease in DMA storage modulus.

The tensile modulus and flexural modulus of CICP further decreased to 89.4 and 72.8 MPa, respectively. CICP did not fracture before the 630% instrument limit was reached, demonstrating the greatest flexibility. Its impact strength at room temperature was 35 kJ m⁻², lower than the 68 kJ m⁻² of RICP, possibly because of differences in the failure modes. At 0 and −20 °C, however, the impact strength reached 56 and 97 kJ m⁻², respectively, far exceeding those of HICP and RICP. These results indicate that the continuous low-crystallinity/rubber-rich phase dominating CICP can maintain large-scale deformation and crack blunting at low temperatures, while the dispersed PP crystalline regions provide the necessary physical support.

Increasing the rubber-rich-component content reduces crystallinity and increases the amorphous-phase fraction, changes that are generally accompanied by greater segmental mobility and free volume and therefore favor thermally activated charge-carrier migration. The total ethylene content and 35 °C-soluble fraction of CICP reached 23.3 wt% and 73.49 wt%, respectively, providing this potential driving force for transport. However, its low-crystallinity/rubber-rich phase is distributed with high continuity, with a small population of PP crystalline regions dispersed within it. This arrangement reduces abrupt isolated interfaces and helps mitigate local-field concentration, while the constrained segments around the crystalline regions and the compositional transition regions provide trapping sites with different binding strengths. The high-field current, TSDC, and PEA results indicate that, under the present test conditions, interfacial injection regulation and delayed bulk trapping–detrapping dominate and offset the transport-promoting effect of the high amorphous content.

The comparison between HICP and RICP, which have similar ethylene contents and were prepared by the same laboratory polymerization route, reveals a relatively clear pathway for structural regulation: randomization of the first-stage matrix reduces chain regularity, causing crystal refinement and an increase in intermediate-crystallizability components and expanding the range of chain interpenetration between phases. These changes subsequently increase the trap barrier, reduce space charge, enhance breakdown strength, and markedly decrease the macroscopic modulus.

Figure 19 summarizes the multiscale structure-property relationships of the three ICPs. HICP has a highly regular crystalline matrix and discrete rubber-rich phases separated by relatively abrupt interfaces, corresponding to a shallow and concentrated trap distribution and a high bulk space-charge density; it therefore retains a high modulus but exhibits the lowest breakdown strength. Through matrix randomization, RICP develops a refined crystalline framework and expanded interfacial transition regions, increasing the trap depth from 0.92 to 1.05 eV and reducing the space-charge density while simultaneously lowering the modulus and increasing the breakdown strength. CICP consists of a continuous low-crystallinity/rubber-rich phase and dispersed PP crystalline regions. Abrupt isolated interfaces are reduced, while the regions surrounding the crystallites and the compositional transition regions form charge-trapping sites with different binding strengths. This structure is consistent with its broad high-temperature TSDC peak, low high-field current, and low bulk space charge and enables the material to achieve the highest breakdown strength together with an extremely low modulus and excellent low-temperature toughness.

The mechanical and electrical properties of ICPs are jointly controlled by the molecular sequence, crystalline framework, continuity of the rubber-rich phase, and spatial organization of the interfaces. The evolution from discrete phase domains toward progressive interfacial bridging and a highly continuous multiphase structure provides an effective structural pathway for reducing material stiffness while maintaining high-field stability.

4. Conclusions

This study investigated HICP, RICP, and the commercial high-phase-continuity material CICP and analyzed the influence of ICP structural evolution on DC insulation behavior through the progressive relationships among molecular composition, crystalline framework, multiphase topology, charge transport, and mechanical-electrical properties. The principal conclusions are as follows.

(1) Polymerization design altered the molecular sequences and crystalline frameworks. The total ethylene contents of HICP, RICP, and CICP were 11.5, 13.6, and 23.3 wt%, respectively, and their 35 °C-soluble fractions were 18.92, 27.21, and 73.49 wt%, respectively. Matrix randomization in RICP shifted the main elution peak from 125.0 to 112.0 °C, increased the intermediate-crystallizability components, reduced the crystallinity from 51.1% for HICP to 32.1%, and decreased the effective crystalline-layer contribution from 7.84 to 4.05 nm. CICP was dominated by low-crystallizability components and had a crystallinity of 13.7%, with a small population of PP crystalline regions of relatively high thermal stability dispersed within it.

(2) Reconstruction of the crystalline framework altered the spatial organization of the PP-rich and rubber-rich phases. HICP consisted of a highly regular PP matrix surrounding discrete rubber-phase domains with relatively distinct interfaces. Crystal refinement and the increase in intermediate components in RICP expanded the interphase transition range and produced pronounced interfacial-bridging features. CICP exhibited a highly continuous multiphase structure consisting of a continuous low-crystallinity/rubber-rich phase and dispersed PP crystalline regions. The SEM, DMA, and melt-rheology results jointly showed that phase-domain connectivity and long-range constraints in the melt progressively increased from HICP to RICP and then to CICP, while the material stiffness decreased markedly.

(3) Topological evolution reshaped the trap distribution and high-field charge transport. The characteristic trap depth increased from 0.92 eV to 1.05 and 1.25 eV, the maximum space-charge density decreased from 65.5 C m⁻³ to approximately 28.4 and below 12.5 C m⁻³, and the breakdown strength increased from 226.2 kV mm⁻¹ to 310.8 and 384.4 kV mm⁻¹. Deeper traps prolonged the residence time of trapped charge carriers, while the reduction in abrupt isolated interfaces helped mitigate local-field concentration; together, these effects reduced the bulk charge and its additional electric field. Within the sample set investigated here, these results establish a semiquantitative association linking deeper traps, reduced space charge, and enhanced breakdown.

(4) Regulation of the multiphase topology improved the synergy between flexibility and high-field stability. The tensile modulus of RICP decreased from 1000 MPa for HICP to 612 MPa, its strain at break increased from 500% to 610%, and its breakdown strength increased by 37.4%. These results demonstrate that matrix randomization and interfacial bridging can simultaneously reduce stiffness and improve insulation stability at similar ethylene contents. The tensile modulus of CICP was only 89.4 MPa, and CICP did not fracture before the 630% instrument limit was reached; its impact strength at −20 °C was 97 kJ m⁻², and its breakdown strength reached 384.4 kV mm⁻¹. Its highly continuous multiphase structure mitigated local-field concentration at abrupt isolated interfaces, while the constrained segments surrounding the dispersed crystalline regions and the compositional transition environments enhanced charge trapping, offsetting the transport-promoting effect of the high amorphous content.

This study establishes the structure-property relationship linking molecular sequence, crystalline framework, multiphase topology, traps and space charge, and mechanical-electrical properties, thereby providing a basis for the topological design of PP-based HVDC insulation materials.