High temperature testing of PEEK and PPS coatings on copper wires | npj Materials Degradation

npj Materials Degradation volume 9, Article number: 26 (2025) Cite this article

Metrics details

This study examined the high-temperature stability of polyether ether ketone (PEEK) and polyphenylene sulfide (PPS) in an oxygenated environment. Both polymers were extrusion-coated onto copper wires for electrical insulation in traction motors. Accelerated testing using thermogravimetry and calorimetry showed that copper catalyzed thermal oxidation of PEEK (at very high temperature), which was accelerated by a lower molar mass of the PEEK and an increased copper-polymer contact area. Both techniques indicated a complex thermal oxidation pattern for both polymers. Notably, the presence of copper seemed to reduce/retard the degradation of PPS. Overall, both polymers demonstrated high oxidation resistance at elevated temperature in an air environment, indicating long service life in electric motor, excluding factors like moisture, oil spray cooling and Joule heating.

The development and utilization of high-performance polymers,1,2 such as polyether ether ketone (PEEK) and polyphenylene sulfide (PPS), have attracted increasing attention3,4, due to their applicability in various demanding applications. PEEK and PPS are two of the most critical high-temperature thermoplastic engineering polymers, and they maintain their excellent mechanical and chemical resistance properties, even at high temperatures.5,6,7,8 PEEK, a member of the polyaryl ether ketone family, is extensively used in many applications, such as those where the polymer is exposed to relative motion,9,10 due to its intrinsically low friction coefficient and exceptional fatigue resistance.11 Today, PPS is utilized in areas as molding resins, fibers and matrices for thermoplastic composites.12,13 Research on the thermal properties of PEEK and PPS has been driven by the demand for electrically insulating materials that can withstand high temperatures, such as those found in electric motors.14,15,16,17,18 Although PEEK generally has a slightly higher thermal stability than PPS, it is more costly.19

Copper-containing oxidases play an essential role in accelerating many significant oxidative reactions in nature.20,21 These reactions range from electron transfer processes occurring outside of molecular frameworks (such as laccases)22 to the removal of hydrogen atoms (as exemplified by galactose oxidase)23. The catalytic properties of copper have been observed to speed up the decomposition of polymers by increasing the rate of oxidation,24 thereby influencing the formation of oxidized products.25,26,27 The impact of copper on the oxidation process of insulation polymers has garnered considerable interest, primarily due to its ability to hasten the aging and deterioration of these materials.28 However, the processes by which copper affects the properties of PEEK and PPS at high temperatures remain poorly understood.

This study provides new insights into the thermal stability/oxidation characteristics of two high-performance polymers, polyetheretherketone (PEEK) and polyphenylene sulfide (PPS), specifically regarding their interaction with copper, such as in wire coatings in electric motors; an application that is of high importance in an increasing fleet of electrified vehicles. The study shows the complex thermo-oxidation behavior of these polymers and the effects that copper has on it.

The degradation characteristics and kinetics were assessed by thermogravimetry and differential scanning calorimetry, whereas chemical changes were evaluated using infrared spectroscopy. By determining the degradation kinetics in both isothermal and non-isothermal conditions in systems of different wire/coating geometries the influence of interface geometry and wire/coating thickness on the coating stability was obtained. These insights are valuable when considering the design of future more effective (rapid) charging of electric vehicles, putting higher demands on the wire insulation coating.

In the first section, thermogravimetric results for LMPEEK (low molecular weight PEEK), HMPEEK (high molecular weight PEEK), PPS, and the coated copper wires are presented. The second section considers the kinetic analysis based on the results from the first section. The third section presents the calorimetric results, followed by the kinetic analysis of these data in the final section. For the nomenclature of the samples, refer to Table 3 in the Methods section. Briefly, (L) and (S) refer, respectively, to the large and small wire and coating sizes.

For all materials, no significant mass loss (<0.5%) was observed below 300 °C. Therefore, the region above 300 °C was examined further (Fig. 1). A higher heating rate generally leads to an increased maximum conversion rate.29 This is also evidenced here by the elevated maximum derivative thermogravimetry (DTG) values, and the higher temperature needed to achieve the same level of conversion at a high heating rate. The TG curves for the pure copper (sourced from the HMPEEK coated copper wire) showed that the mass increase up to 600 °C was less than 1.1% (Supplementary fig. 1). Most of the copper wire was also covered by the polymer in the coated system, which is why the effects on the copper were disregarded in subsequent analyses. As depicted in Fig. 1, the mass loss curves were, generally “disrupted” at several points, suggesting more than one stage/mechanism. This was also clearly observed with more than one peak/shoulder present in the DTG curves.

a LMPEEK, (b) LMPEEK(L)/W(S), (c) LMPEEK(S)/W(L), (d) HMPEEK, (e) HMPEEK/W(L), (f) PPS, (g) PPS(L)/W(S), (h) PPS(S)/W(L).

For both LMPEEK and HMPEEK, mass loss typically occurred at a lower temperature when coated onto a copper wire. An exception to this was the thicker LMPEEK coating on the thinner copper wire, where mass loss occurred in roughly the same temperature range as the pure polymer (Fig. 1a, b). The reason was probably the larger polymer mass, yielding a larger thermal lag. The earlier loss of polymer mass in the presence of the copper wire suggests the copper catalyzed thermal degradation. As observed in the LMPEEK system, the mass loss occurred earlier in the 2 mm wire system (LMPEEK(S)/W(L)), compared to the 0.8 mm wire system (LMPEEK(L)/W(S)). This suggests that a larger interfacial area between the polymer and the copper accelerated degradation.

Overall, the behavior of the PPS system was different from that of the PEEK systems. Generally, the thermal degradation of PPS did not increase noticeably in the presence of copper. In fact, the mass loss occurred later for the 0.8 mm coated wire (PPS(L)/W(S)) than for the pure polymer, at the lowest heating rate (Fig. 1g, h).

It should be noted that the degradation occurred above the melting point of the polymers. The melting points determined from the melting endotherms were 339, 344 and 346 °C for, LMPEEK, LMPEEK on the small wire, and LMPEEK on the large wire, respectively (Supplementary fig. 2). The melting points for HMPEEK, and HMPEEK on the large wire, were respectively 334 and 342 °C. The pure PPS was essentially amorphous, with the cold crystallized sample having a melting point of 280 °C, whereas the PPS on the small and large wires were semi-crystalline with melting points of 281 and 283 °C, respectively. Because of the complex melting endotherms of the polymers on the coated wires, it was not possible to determine any degree of crystallinity in these systems. However, it was possible to determine the degree of crystallinity for the pure LMPEEK and HMPEEK samples.30 Crystallinity was found to be 52% for LMPEEK and 45% for HMPEEK, indicating easier crystallization of the lower molar mass polymer.

The fit of the Ozawa-Flynn-Wall (OFW) and Kissinger-Akahira-Sunose (KAS) equations (Eqs. 5 and 6, refer to the experimental section) to the experimental TG data is shown in Fig. 2 and Supplementary fig. 3. These plots include three horizontal lines, representing different degrees of conversion (α) which, from bottom to top, indicate heating rates of 1, 2, and 4 °C min−1, respectively. At the highest heating rate (8 °C min−1), the degradation process was not fully completed at 600 °C for any of the samples, except for pure PPS. As a result, the kinetic analysis was strictly based on the TG results obtained at 1, 2 and 4 °C min−1. The activation energy for a specific degree of conversion α was deduced from the slope of the line connecting the data points corresponding to α at the different heating rates. The activation energy (Ea) and pre-exponential factor (A) derived from the kinetic analyses are shown in Fig. 3, Supplementary Figs. 4 and 5. The fact that the values of both A and Ea fluctuated along the degree of conversion implies that multiple mechanisms were involved in the mass loss process, as also mentioned when describing the TG and DTG curves. The pre-exponential factor in the rate equation indicates the frequency of molecular collisions necessary to initiate a reaction.31 This factor provides an additional viewpoint on reaction kinetics, supplementing the insights offered by the activation energy. The average values for the activation energy, pre-exponential factor, and the coefficient of determination (R2) from the kinetic analysis are detailed in Table 1. The analysis centered on a degree of conversion α from 0.1 to 0.9, to mitigate any effects of instability during the initial and final stages. Average activation energy values calculated using both the OFW and KAS methods showed impressive consistency in both value and trend (Fig. 3 and Supplementary Fig. 4), although the KAS results tended to come with larger error margins. As indicated in Table 1, kinetic analyses conducted with the OFW method revealed higher R2 values in all instances. Given this overall trend, the OFW results were primarily used to guide further discussions.

a LMPEEK and (b) LMPEEK(L)/W(S).

a LMPEEK, (b) LMPEEK(L)/W(S), (c) LMPEEK(S)/W(L), (d) HMPEEK, (e) HMPEEK/W(L), (f) PPS, (g) PPS(L)/W(S), (h) PPS(S)/W(L).

The average activation energies for the degradation of LMPEEK and HMPEEK were 126 kJ/mol and 153 kJ/mol, respectively (Table 1). The lower activation energy observed for LMPEEK could be attributed to its lower molecular weight, which requires fewer chain scissions to achive the same degree of degradation as the higher molecular weight HMPEEK. It is crucial to note that the determined activation energy, derived from thermogravimetry data, encompasses both the degradation process and the evaporation of the degradation products. The activation energy was essentially the same for LMPEEK and LMPEEK on the small wire (W(S): 0.8 mm). However, with a larger wire (W(L): 2 mm) that boasts a vast surface area between the polymer and copper, the activation energy decreased for both PEEK materials, this suggests that copper catalyses PEEK degradation. The average values of the pre-exponential factors were lowest when the PEEK was coated on the 2 mm wire. It should be observed that the size of the pre-exponential factor, gleaned from the data analysis, partly compensates for the substantial effect of changes in activation energy on k(T) in Eq. (1) (refer to the methods section). As the calculated activation energy decreases, the size of k(T) increases significantly (since the activation energy has an exponential effect) and hence a decrease in the pre-exponential factor is required to obtain “accurate” k(T) values.

The average activation energy, as well as the average pre-exponential factor, increased for PPS coated on the smaller 0.8 mm wire (PPS(L)/W(S)) compared to pure PPS. This once again indicates a different behavior for PPS as compared to PEEK. However, for the 2 mm wire system, both the activation energy and the pre-exponential factor were lower than they were for pure PPS.

Oxidation induction time (OIT) measurement, to our knowledge, have not previously been reported for PEEK and PPS. The technique was used here to reveal if further information could be obtained to explain the complex degradation pattern obtained with thermogravimetry, and also to obtain further information on the degradation kinetics. To pinpoint a temperature range where the OIT would yield meaningful results, non-isothermal (temperature-sweep) measurements, referred to as Tox measurements were done. These differential scanning calorimetry (DSC) curves unveiled a complex degradation/oxidation pattern for PEEK and PPS (Supplementary fig. 6). Notably, an observable exothermic process initiated above 400 °C. It is also remarkable that the coated wire systems seemed to register a marginally lower temperature where significant increases in exothermic heat flow occurred (commenced around 500 °C). Based on the Tox curves, the OIT was determined to be recorded at 420, 425 and 430 °C. Figure 4 displays the 420 °C curves for the pure polymers, alongside FTIR spectra determined on pristine samples and samples post-OIT test. The PEEK samples subjected to oxygen demonstrated a broad but limited exothermic (oxidation) signal from the start of oxygen exposure to a point where a second more substantial exothermic (oxidation) process took off. This larger process began after approximately 120 and 135 min for LMPEEK and HMPEEK, respectively. Therefore, just as the TG data suggests, the higher molecular weight grade appeared more stable. In nitrogen (thermal induction time), the isothermal curves were entirely featureless. We conducted OIT experiments using pure copper at the three temperatures and found no thermal oxidation (Supplementary fig. 7). Thus, the OIT results of the coated copper wires reflected exclusively the thermal oxidation of the polymer part.

Left column: OIT results of pure LMPEEK, HMPEEK and PPS after an 8 °C min−1 heating in nitrogen to 420 °C and a subsequent switch to oxygen (blue curves) and, as a reference, using nitrogen throughout the 2 h isothermal test (red curves). Right columns: FTIR results of pure LMPEEK, HMPEEK and PPS before (green curves) and after an 8 °C min−1 heating in nitrogen to 420 °C and a subsequent switch to oxygen (blue curves) are displayed in the right column. As a reference, FTIR was also obtained on samples exposed to the same heating cycle, but with nitrogen throughout the 2 h isothermal test (red curves). Exothermal heatflow is downwards.

In the case of PPS, the isothermal segment of the OIT curve remained relatively flat until a brief, minor exothermic heat flow occurred (after 120 min). This was then followed by a major exothermic event (post 140 min). These particular features were not present in nitrogen, thereby suggesting that the exothermic processes were correlated with oxidation.

FTIR spectroscopy was performed on the outer surface of the coatings (Fig. 4). The main FTIR peaks for all three polymers appeared below 1700 cm−1, which aligns with previously reported data for LMPEEK, HMPEEK32 and PPS33. For PEEK, the peaks located around 1250 and 1730 cm−1 correspond to ether and ketone functional groups, respectively.34 Several sharp peaks within the region of 1225 to 950 cm−1 can be attributed to C–H in plane bend vibrations. The absorption from 900 to 670 cm−1 relates to C–H out-of-plane bend vibrations. Peaks around 1500 and 1600 cm−1 suggest the presence of aromatic rings in the polymer structure.35 In the range of 2800 to 3000 cm−1, aliphatic C–H stretch vibrations give rise to two less pronounced peaks.36

For PPS, the peaks at 1576, 1465 and 1388 cm−1 are associated with aromatic ring stretching in S-C6H4-S. Peaks at 1092 and 1077 cm−1 correspond to C–S bond stretching in the same structure. Peaks at 1010 and 802 cm−1 are assigned to C–H bending modes, with the 745 cm−1 peak indicating ring bending37. The peak at 802 cm−1 also signifies para-substitution in the aromatic ring (S-C6H4-S)38. Additionally, the absorption band at 3063 cm−1 is related to the C–H stretching vibration in the aromatic ring, as observed in PEEK.

Following the “OIT” experiment in nitrogen, most peaks in PEEK were still discernible, albeit with reduced intensity. Additionally, a broad signal from roughly 1800 cm−1 and below overlapped with the narrower peaks, signifying a carbon-rich pyrolysis product. The observations for PPS, closely mirrored these but without the overlapping broad absorbance. It should be noted that the FTIR spectra were always obtained at room temperature after cooling the specimens from the high-temperature tests.

Following the OIT experiment (in oxygen), the original peaks in the region below 1700 cm−1 became smaller than those in the nitrogen case. Indeed, in the case of LMPEEK, we observed an almost flat absorbance across a wide wavenumber range, indicating heavily degraded material without the typical aromatic structure. HMPEEK, however, had less flat absorbance in this region, indicating a material with a lesser degree of degradation than LMPEEK. This discrepancy is likely due to the greater number of end-groups in LMPEEK. The enhanced reactivity of end groups in polymers39, can be attributed to their lessened steric hindrance and improved accessibility to reactive species like oxygen.40

The FTIR spectrum of PPS in oxygen displayed different information compared to PEEK (Fig. 4). A broad, low-intensity absorbance in the ~3500–3200 cm−1 region indicated the presence of hydroxyl groups and the stronger peaks at 2800 and 2900 cm−1 suggested the presence of aliphatic C-H groups. Two strong peaks at 1730 and 1590 cm−1 demonstrated that carbonyls were formed during the OIT experiment. Furthermore, the sulfur found in PPS can undergo thermal oxidation. For instance, the pronounced peaks at 1158 and 1048 cm−1 in the FTIR spectrum of oxidized PPS have been noted in prior studies.41 These peaks can be attributed to the symmetric stretching of newly formed sulfone (–SO2–)42 and sulfoxide (–SO–) groups. The formation of these groups during the OIT test signaled significant chemical changes in the PPS structure driven by the thermal oxidation process.

Figure 5 shows the OIT outcomes for pure LMPEEK, HMPEEK, PPS, and coated copper wires at temperatures of 420, 425, and 430 °C. The observed curves were mixed, with multiple possible mechanisms contributing to the degradation. Consequently, determining a distinct oxygen induction time was not straightforward, particularly for the systems with fairly “flat” curves. Nevertheless, to facilitate the estimation of degradation kinetics at the three temperatures and to determine activation energy, the OIT of the PEEK materials was defined as the onset of the larger exothermal process (as indicated by the arrow in Fig. 4). It is worth noting that the curves in Figs. 4 and 5 for the pure materials were not identical, due to differing heating rates up to isothermal conditions, which influenced the absolute values of the OIT. The OIT was comparable for pure LMPEEK and LMPEEK on the small wire (LMPEEK(L)/W(S)). However, in the system with a larger polymer-copper contact (2 mm wire, LMPEEK(S)/W(L)) the OIT decreased relative to that of the pure polymer, similar to what was observed also in the HMPEEK system. For instance, the OIT values (after the switch to oxygen) at 430 °C were nearly 50 min for both pure LMPEEK and LMPEEK coated on the small wire (0.8 mm), but approximately 40 min for the large wire system (Fig. 5a–c). These findings, in conjunction with the TG results, indicate that the copper wire acts as a catalyst, speeding up the thermal oxidation of PEEK polymers at the investigated OIT temperatures.

a LMPEEK, (b) LMPEEK(L)/W(S), (c) LMPEEK(S)/W(L), (d) HMPEEK, (e) HMPEEK/W(L), (f) PPS, (g) PPS(L)/W(S), (h) PPS(S)/W(L). Orange, green and purple colors refer to 420, 425 and 430 °C isothermal conditions, respectively. Exothermal heatflow is downwards.

While the OIT of pure PPS could be determined as the time at which major oxidation commenced, this methodology was not applicable to the PPS on the wires. The OIT curves in these systems were steady yet somewhat “unstable”, signifying a series of minimal exothermic heat flow events. As indicated by the thermogravimetry data (Fig. 1), the PPS seemed to exhibit increased thermal (oxidation) stability in the presence of copper, at least when coated on the small wire. Meanwhile, the FTIR spectrum of PPS on the copper wire, following the OIT test, indicated a higher degree of molecular change compared to the same test conducted in nitrogen, However, these changes were less significant than those seen in the pure polymer (Fig. 6a, refer to the size of the absorbance in the region between ~1500 and 1750 cm−1). The distinction in the behavior between PPS and PEEK, when in contact with copper is likely attributable to the presence of sulfur in PPS, which can react with copper to produce copper sulfide.43 Copper sulfide, if in the form of e.g. a layer, may, subsequently decrease the rate of copper-catalyzed PPS oxidation. In the case of PEEK (LMPEEK), the large change in the FTIR spectra after the OIT test made it difficult to assess whether the coating was oxidized to a larger or smaller extent than the material oxidized in the absence of copper (Fig. 6b).

a pure PPS (PPS curve) and PPS coated on the large wire after the OIT tests in oxygen (O2 curves) and in nitrogen (N2 curves); (b) pure LMPEEK (LMPEEK curve) and LMPEEK coated on the small wire after the OIT tests in oxygen (O2 curves) and in nitrogen (N2 curves).

To estimate the activation energy for oxidation/degradation from OIT, “Arrhenius” curves were generated. These were based on the OIT versus temperature data (Fig. 7), and the corresponding parameters and coefficient of determination are presented in Table 2. Although not statistically significant, there was a general trend. The activation energy decreased in the presence of copper in the PEEK systems, and with increasing contact area between LMPEEK and the copper wire. This strengthened the observation of a degradation-catalyzing effect on PEEK by copper. The same general trend was also observed for the activation energy determined from TG data; nevertheless, the absolute magnitude of these diverged between the two methods. This is unsurprising, given that TG measures the loss of mass due to the evaporation of degradation products. In contrast, OIT measures the calorimetric response during the degradation, which may include reactions that do not immediately lead to the evaporation of species. As seen with the TG data, the pre-exponential factor decreased in the presence of copper. Again, this is, at least in part, a compensation effect for the significant influence of changes in the activation energy on the k(T) values in Eq. 1. The OIT activation energy for the pure PPS, derived from the onset of the exothermal signal was substantially larger than for the PEEK systems. However, the Arrhenius fit was also inferior (Fig. 7). The OIT for the 430 °C curve (Fig. 5f) was challenging to assess, but choosing an OIT closer to the start of the large exothermal signal (at approximately 65 min), improved the Arrhenius fit and considerably decreased the activation energy (Fig. 7). Notably, the Arrhenius parameters for the PPS coated wires were not possible to determine because of their unique OIT behavior (Fig. 5). This, combined with the poorer Arrhenius fits of the PEEK coated wires (lower R2, Table 2), suggests that the presence of copper modified the degradation/oxidation mechanisms observed in the pure polymers. This is also suggested by the intricate behavior of the activation energy and pre-exponential parameter with conversion rate in the TG data (Fig. 3, Supplementary Figs. 4 and 5).

The red and gray curves in the PPS case correspond to the use of an OIT at 430 °C of 40 and 65 min, respectively.

A comprehensive investigation revealing the different degradation products (paths) of PEEK, oxidized at different oxygen partial pressures has been reported by Courvoisier et al.44 By the use of FTIR, the kinetics of the degradation (formation of degradation products) could be assessed at significantly lower temperatures (250 to 320 °C) than was possible here using OIT from DSC. Notably, an increase in oxygen pressure reduced the time for the formation of the reaction products. By the use of the induction time to the first formation of reaction products we here combined our measurements (OIT) and those by Courvoisier et al.44, to be able to estimate the kinetics over a significantly larger temperature region than was used here. As shown in Fig. 8 the data could be represented by a single Arrhenius relationship (R2 = 0.978) with an activation energy of 140 kJ/mol (and a Log (A) of 6.82 s−1). Notably, this activation energy is in the middle of those for LMPEEK and HMPEEK, deduced form the thermogravimetry data (Table 1). One should have in mind that this quite good correlation was obtained despite the differences in the types of measurements and measurement conditions and the state of the polymer. The OIT was measured in the molten state and the FTIR in the semicrystalline state. The FTIR data in the plot corresponded to ageing at an oxygen pressure of 0.21 bar, which was the pressure that resembled the OIT test the best (DSC, 1 bar O2). It is clear that the activation energies determined in the narrow temperature range with OIT, were overestimated for pure PEEK (Table 2). A similar analysis, combining DSC and FTIR, was not possible to do for the copper-coated systems, and also not for PPS, due to the absence of reported such data for these systems.

from OIT (lower set, pure LMPEEK and HMPEEK) and FTIR (upper set) as a function of the inverse of temperature44.

In conclusion the degradation characteristics of PEEK and PPS, examined using thermogravimetric analysis, oxidation induction time analysis, and kinetic modeling indicated a complex pattern implying the occurrence of several processes. The effects of copper were different for PEEK and PPS and depended often on the wire geometry. A larger contact area increased in general the degradation rate of PEEK. A thinner coating, which was the case when the large wire was used, possibly also speeded up the degradation rate (more copper per volume of polymer). This was considered by calculating an Li/Ac value for each system (wire/polymer interfacial circumferential length per coating cross-sectional area, refer to Table 3 in the Methods section). The values in the small wire systems were 0.04 (PPS) and 0.07 (LMPEEK), and the corresponding values in the large wire systems were substantially larger; 0.40 (HMPEEK, PPS) and 0.62 (LMPEEK). For PPS the copper did not have the same catalysing effect on the oxidation, probably due to the formation of copper-sulfur species. The trends in degradation were similar across the TG and OIT data, although the size of the activation energies differed; as expected due to the different nature of the methods.

To have a rough indication of the stability of the PEEK and PPS in an electric motor service condition, the induction time can be extrapolated from the Arrhenius curve in Fig. 8 to a typical service temperature of 120 °C (with peaks up to 180 °C). This yields induction times that go safely beyond the expected lifetime of a car or a truck. Note that we then neglect the effects of crossing the glass transition, which, would however increase the induction period. In this estimation we are also neglecting any effects of the copper (and Joule heating) and the presence of moisture and oil (spray cooling). As we show here, the choice of material type for the coating, as well as its inherent properties (molar mass) will have an influence on the lifetime of the electric motor. Further work is needed to establish the effects of especially the presence of oil, but also the effects of moisture and Joule heating on the performance of the coating.

Copper wires, extrusion coated with LMPEEK (lower molecular weight poly(ether ether ketone) (PEEK), KetaSpire® KT880NT), HMPEEK (higher molecular weight PEEK, KetaSpire® KT820NT) and poly(phenylene sulfide) (PPS, Ryton® QA200P) were supplied by Syensqo, Italy. The melt mass flow rates (MFR) of LMPEEK and HMPEEK are 36 g 10 min−1 and 3 g 10 min−1 (400 °C 2.16 kg, ASTM D1238), respectively, and that of PPS is 100 g 10 min−1 (316 °C, 5.0 kg. ASTM D1238). Other than the use of a lubricant (calcium stearate, less than 0.5%) in the two PEEK grades, the polymers were pure. To determine the effect of different contact areas between the circular copper wire and the polymer coating, two different copper wire diameters were used for both the LMPEEK and the PPS systems. To investigate the influence of PEEK’s molar mass, the thicker copper wire was also coated with HMPEEK. Since the thermal properties of the coated wire system depend on the size of the wire-polymer contact area and the amount of polymer coated on the wire, the wire-polymer interfacial circumferential length (Li) was divided by the cross-sectional area of the coating (Ac) to calculate (Li Ac−1), which, along with the wire and coating geometry, is presented in Table 3. Unextruded polymer pellets and films were also used. LMPEEK and HMPEEK pellets, and PPS film were supplied by Syensqo, Italy. The coated LMPEEK, HMPEEK and PPS wires were extruded on a prototyping line at Syensqo using a Sterlingh extruder (barrel diameter 38 mm, L/D 30, with conventional screw type having mixing elements). Typical processing conditions are reported in the following table. The extruded PPS films are commercially available products supplied by Syensqo’s Ajedium division and the PPS pellets are commercially available products supplied by Syensqo Specialty Polymers. The temperature profile, pressure and line speed for each wire/coating production are summarized in Table 4.

A TGA/DSC 3+ STARe System (Mettler Toledo, Switzerland) was used to conduct thermogravimetry. The samples were heated from 30 to 600 °C under a 50 ml/min O2 atmosphere. Four different heating rates of 1, 2, 4, and 8 °C/min were used to determine the kinetics of degradation for the polymers and the polymers coated on copper wires. To eliminate systematic error, a blank test was performed before the actual sample tests at each heating rate. Thermogravimetric results were obtained by subtracting the blank test data from the sample test results. Specimens consisting of sections of the coated wire were cut out perpendicular to the wire direction and placed in 100 µl alumina crucibles before starting the test. The weight of the specimens was approximately 150 mg for those with a 2 mm thick wire and 30 mg for those with a 0.8 mm wire. The weight used for the pure polymers was 5.0 ± 0.1 mg.

A DSC 820 (Mettler-Toledo, Switzerland) was utilized to determine the calorimetric properties of the samples. The non-isothermal oxidation pattern and oxidative onset temperature (Tox) were determined using a heating rate of 8 °C/min from 25 to 600 °C in an oxygen atmosphere with a flow rate of 50 ml/min. The OIT was achieved by initially heating the specimen in nitrogen (flow rate of 50 ml/min) to the target temperature starting from 25 °C employing a 50 °C per minute heating rate. Upon reaching the test temperature, the gas was switched to oxygen at the same flow rate. Isothermal tests were additionally performed in nitrogen to distinguish the effects of oxygen from other thermal events. The specimen sizes consistent with those in the TG test were employed here as well. The crystallinity (Xc) was determined from the DSC melting endotherm of samples heated at a rate of 10 °C per minute in nitrogen. Xc was calculated as the ratio of the melting enthalpy to the enthalpy of 100% crystalline polymer (130 J/g for PEEK30)45.

The reaction kinetics can often be expressed based on the Arrhenius law (Eq. (1)):

where A represents the pre-exponential factor, R denotes the universal gas constant, which is equal to 8.314 J mol−1, Ea stands for the activation energy (kJ mol−1), and k(T) is the rate constant term in the Arrhenius equation, which changes based on the temperature (T)46,47.

The reaction conversion α can be estimated as shown in Eq. (2):

where m0, m f, and mT are the initial mass, final mass, and mass at temperature T of the reactant.

The reaction rate can be expressed by the following equation:

where t represents the time of reaction, dα/dt represents the reaction rate, and f(α) is the reaction model function.

When the material is heated using a constant heating rate dT/dt = β (K min−1), the following Eq. (4), reaction model can be derived by combining Eq. (1) and Eq. (3) in an integral form g(α):

The kinetic parameters of the thermal degradation behavior were calculated using the Netzsch Kinetics Neo software (ver. 2.5.0.1, Germany). This software adheres to the recommendations of the International Confederation for Thermal Analysis and Calorimetry (ICTAC)48. We employed different calculation methods to estimate the kinetic parameters from isothermal and kinetic results.

The model-free methods of OFW and KAS were used to uncover the kinetics of thermal degradation reactions49.

The OFW method can be expressed as50

and the KAS method can be expressed as51

The kinetics of degradation were determined using the calorimetry data. The activation energy for the thermal degradation was calculated based on the OIT results, utilizing the isothermal Arrhenius equation.52 The corresponding pre-exponential factor A was estimated according to the first-order reaction model when the reaction conversion α reached a value of 5%, to negate any instability influence at the start of the test.

A PerkinElmer Spectrum 2000 (PerkinElmer Inc., U.S.A.), equipped with a MKII Golden GateTM (diamond crystal) attenuated total reflection (ATR) device (Diamond Crystal, Kent, England), was utilized for generating ATR Fourier-transform infrared spectroscopy (FTIR) spectra. The scanning procedure incorporated a resolution of 4.0 cm–1 and the range was spanning from 4000 to 600 cm–1. Each spectrum involved the employment of 32 scans.

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Ta, X., Chen, X. & Wang, Z. L. Design and synthesis of triboelectric polymers for high performance triboelectric nanogenerators. Energy Environ. Sci. 16, 3654–3678 (2023).

Article Google Scholar

Yuan, P., Sun, Y., Xu, X., Luo, Y. & Hong, M. Towards high-performance sustainable polymers via isomerization-driven irreversible ring-opening polymerization of five-membered thionolactones. Nat. Chem. 14, 294–303 (2022).

Article CAS PubMed Google Scholar

Ren, Y. et al. Exploring the self-nucleation effect: Transforming crystalline morphology for improved mechanical performance of carbon fiber reinforced polyphenylene sulfide composites. Compos. B: Eng. 272, 111231 (2024).

Siraj, N., Azhar, S., Hashmi, R. & Verma, S. State-of-the-art review on the high-performance poly (etherether ketone) composites for mechanical, tribological andbioactive characteristics. Polym. Adv. Technol. 33, 3049–3077 (2022).

Article CAS Google Scholar

Li, L. et al. Evaluation of the mechanical and tribological behavior of polyether ether ketone fiber-reinforced resin-based friction materials fabricated by wet granulation. Polymers 15, 4732 (2023).

Article CAS PubMed PubMed Central Google Scholar

Ren, T., Zhu, G., Zhang, C. & Wang, S. Preparation of CF/PEEK composites with high mechanical performance using PEEK derivatives as the sizing agent. Macromol. Rapid Commun. 44, e2200738 (2022).

Zhao, P. et al. PPS ultrafine fiber enhanced aramid fiber filter with superior thermal stability and excellent chemical resistance for efficient PM2.5 removal. Reactive Funct. Polym. 188, 105605 (2023).

Yang, G., Zhang, M., Su, K. & Li, Z. OPPS fibers with high temperature resistance and excellent antioxidant properties by an oxidation method. ACS Appl. Mater. Interfaces 14, 50225–50234 (2022).

Article CAS Google Scholar

Zol, S. M. et al. Description of poly(aryl-ether-ketone) materials (PAEKs), polyetheretherketone (PEEK) and polyetherketoneketone (PEKK) for application as a dental material a materials science review. Polymers 15, 17 (2023).

Article Google Scholar

Lovinger, A. J., Hudson, S. D. & Davis, D. D. High-temperature crystallization and morphology of poly (aryl ether ether ketone). Macromolecules 25, 1752–1758 (1992).

Article CAS Google Scholar

Feng, C. et al. Preparation and properties of the poly(ether ether ketone) (PEEK)/nano-zinc oxide (ZnO)–short carbon fiber (SCF) artificial joint composites. ACS Appl. Polym. Mater. 4, 8869–8877 (2022).

Article CAS Google Scholar

Zuo, P., Tcharkhtchi, A., Shirinbayan, M., Fitoussi, J. & Bakir, F. Overall investigation of poly (phenylene sulfide) from synthesis and process to applications—a review. Macromol. Mater. Eng. 304, 1800686 (2019).

Article Google Scholar

Vinicius Osti de Moraes, D., Ferreira de Melo Morgado, G., Magnabosco, R. & Rezende, M. C. Study of the influence of PPS, PAEK, and PEEK thermoplastic matrices on the static mechanical behavior of carbon fiber-reinforced composites. Polym. Compos. 45, 1726–1736 (2023).

Article Google Scholar

Youcef Kemari, G. B., Zarel Valdez-Nava, F. F. & Diaham, S. in High Temperature Polymer Dielectrics: Fundamentals and Applications in Power Equipment (eds Jun-Wei Zha & Zhi-Min Dang) Ch. 4 (WILEY, 2023).

Giants, T. W. Crystallinity and dielectric properties of PEEK, poly(ether ether ketone). IEEE Trans. Dielectr. Electr. Insulation 1, 991–999 (1994).

Article CAS Google Scholar

Baferani, M. A., Li, C., Shahsavarian, T., Ronzello, J. A. & Cao, Y. High temperature insulation materials for DC cable insulation — Part I: space charge and conduction. IEEE Trans. Dielectr. Electr. Insul. 28 (2021).

Sen Niu, L. Z., Qiannan, C. & Zhang, Y. in Polymer Composites for Electrical Engineering (eds Xingyi Huang & Toshikatsu Tanaka) Ch. 3 (WILEY, 2021).

Bulut, M. S., Ordu, M., Der, O. & Basar, G. Sustainable thermoplastic material selection for hybrid vehicle battery packs in the automotive industry: a comparative multi-criteria decision-making approach. Polymers 16, 2768 (2024).

Article CAS PubMed PubMed Central Google Scholar

Deng, S., Lin, Z., Cao, L., Xian, J. & Liu, C. PPS/recycled PEEK/ carbon nanotube composites: structure, properties and compatibility. J. Appl. Polym. Sci. 132, 42497 (2015).

Article Google Scholar

Khan, M. A., Khan, S., Sengupta, S., Naath, M. B. & Naskar, S. Copper-containing oxidases play a crucial role in accelerating many important oxidative reactions in nature. Coordination Chem. Rev. 504 (2024).

Tucci, F. J. & Rosenzweig, A. C. Direct methane oxidation by copper- and iron-dependent methane monooxygenases. Chem. Rev. 124, 1288–1320 (2024).

Article CAS PubMed PubMed Central Google Scholar

Bento, I. et al. Mechanisms underlying dioxygen reduction in laccases. Structural and modelling studies focusing on proton transfer. BMC Struct. Biol. 10, 14 (2010).

Article Google Scholar

Toftgaard Pedersen, A. et al. Process requirements of galactose oxidase catalyzed oxidation of alcohols. Org. Process Res. Dev. 19, 1580–1589 (2015).

Article CAS Google Scholar

Osawa, Z. Role of metals and metal-deactivators in polymer degradation. Polym. Degrad. Stab. 20, 203–236 (1988).

Article CAS Google Scholar

Oh, S. & Stache, E. E. Recent advances in oxidative degradation of plastics. Chem. Soc. Rev. 53, 7309–7327 (2024).

Article CAS PubMed Google Scholar

Liang, C. et al. The critical role of minerals in persulfate-based advanced oxidation process: catalytic properties, mechanism, and prospects. Chem. Eng. J. 482, 148969 (2024).

Okpara, E. C., Wojuola, O. B., Quadri, T. W. & Banks, C. E. An overview of advanced oxidation processes using copper-based catalytic degradation of organic pollutants in water. Appl. Mater. Today 36, 102053 (2024).

Li, J. et al. Copper-catalyzed oxidation caused by copper-rich impurities in cross-linked polyethylene cable insulation. J. Mater. Sci.: Mater. Electron. 27, 806–810 (2016).

CAS Google Scholar

Mani, T., Murugan, P., Abedi, J. & Mahinpey, N. Pyrolysis of wheat straw in a thermogravimetric analyzer: effect of particle size and heating rate on devolatilization and estimation of global kinetics. Chem. Eng. Res. Des. 88, 952–958 (2010).

Article CAS Google Scholar

Blundell, D. J. & Osborn, B. N. The morphology of poly(aryl-ether-etherketone). Polymer 24, 953–958 (1983).

Article CAS Google Scholar

Wen, Y. et al. Synergistic effect of the co-pyrolysis of cardboard and polyethylene: a kinetic and thermodynamic study. Energy 229, 120693 (2021).

Article CAS Google Scholar

Rangasamy, V. S., Thayumanasundaram, S., Seo, J. W. & Locquet, J.-P. Vibrational spectroscopic study of pure and silica-doped sulfonated poly (ether ether ketone) membranes. Spectrochim. Acta A: Mol. Biomol. Spectrosc. 138, 693–699 (2015).

Article CAS PubMed Google Scholar

Mahmud, M. B., Anstey, A., Shaayegan, V., Lee, P. C. & Park, C. B. Enhancing the mechanical performance of PA6 based composites by altering their crystallization and rheological behavior via in-situ generated PPS nanofibrils. Compos. B: Eng. 195, 108067 (2020).

Article CAS Google Scholar

Coates, J. Interpretation of Infrared Spectra, A Practical Approach. Encyclopedia Anal. Chem. 10815–10837 (2000).

Ahmad, A. et al. Stability of amorphous PEEK in organic solvents. J. Chem. Soc. Pak. 40, 810 (2018).

CAS Google Scholar

Díez-Pascual, A. M. & Díez-Vicente, A. L. Poly (3-hydroxybutyrate)/ZnO bionanocomposites with improved mechanical, barrier and antibacterial properties. Int. J. Mol. Sci. 15, 10950–10973 (2014).

Article PubMed PubMed Central Google Scholar

Durani, S. M. et al. IR laser ablative desulfurization of poly (1, 4-phenylene sulfide). J. Anal. Appl. pyrolysis 73, 145–149 (2005).

Article CAS Google Scholar

Fan, Z.-Z. et al. Fabrication of ultrafine PPS fibers with high strength and tenacity via melt electrospinning. Polymers 11, 530 (2019).

Article PubMed PubMed Central Google Scholar

Ebnesajjad, S. Fluoroplastics, vol. 2: Melt Processible Fluoropolymers-the Definitive User’s Guide and Data Book (William Andrew, 2015).

Ravve, A. Principles of Polymer Chemistry (Springer Science & Business Media, 2013).

Zhang, M. et al. Preparation and properties of polyphenylene sulfide/oxidized-polyphenylene sulfide composite membranes. React. Funct. Polym. 160, 104842 (2021).

Article CAS Google Scholar

Yuan, S. et al. New promising polymer for organic solvent nanofiltration: oxidized poly (arylene sulfide sulfone). J. Membr. Sci. 549, 438–445 (2018).

Article CAS Google Scholar

Webster, H., Wightman, J. & Johnson, T. Analysis of the molecular structure at the PPS/copper interphase and its role in adhesion. J. Adhes. 53, 229–244 (1995).

Article CAS Google Scholar

Courvoisier, E., Bicaba, Y. & Colin, X. Multi-scale and multi-technique analysis of the thermal degradation of poly (ether ether ketone). Polym. Degrad. Stab. 151, 65–79 (2018).

Article CAS Google Scholar

Kong, Y. & Hay, J. N. The measurement of the crystallinity of polymers by DSC. Polymer 43, 3873–3878 (2002).

Article CAS Google Scholar

Guan, H. et al. Energetic and environmental optimizations and byproduct valorization of pyrolysis of textile dyeing sludge with FeCl3. J. Clean. Prod. 434, 139940 (2024).

Article CAS Google Scholar

Arrhenius, S. On the influence of carbonic acid in the air upon the temperature of the ground. J.Philos. Mag. J. Sci. 5, 237–276 (1896).

Article Google Scholar

Koga, N. et al. ICTAC Kinetics Committee recommendations for analysis of thermal decomposition kinetics. Thermochim. Acta 719, 179384 (2023).

Article CAS Google Scholar

Xie, C. et al. Co-combustion thermal conversion characteristics of textile dyeing sludge and pomelo peel using TGA and artificial neural networks. Appl. energy 212, 786–795 (2018).

Article CAS Google Scholar

Xie, C. et al. Quantifying thermal decomposition regimes of textile dyeing sludge, pomelo peel, and their blends. Renew. Energy 122, 55–64 (2018).

Article CAS Google Scholar

Liu, J. et al. Co-) combustion of additives, water hyacinth and sewage sludge: thermogravimetric, kinetic, gas and thermodynamic modeling analyses. Waste Manag. 81, 211–219 (2018).

Arcenegui-Troya, J., Sánchez-Jiménez, P. E., Perejón, A. & Pérez-Maqueda, L. A. Determination of the activation energy under isothermal conditions: Revisited. J. Therm. Anal. Calorim. 148, 1679–1686 (2023).

Article CAS Google Scholar

Download references

Swedish Energy Agency is acknowledged for financial support (project number: 52718-1)

Open access funding provided by Royal Institute of Technology.

Department of Fibre and Polymer, School of Sciences in Chemistry Biotechnology and Health, KTH Royal Institute of Technology, Teknikringen 58, 114 28, Stockholm, Sweden

Sirui Liu, Fritjof Nilsson & Mikael S. Hedenqvist

Lubrizol Limited, The Knowle, Nether Lane, Hazelwood, Derby, DE56 4AN, UK

Greg Hunt

Materials Engineering Centre, Volvo Car Corporation, 405 31, Göteborg, Sweden

Kai Kallio

Syensqo, Viale Lombardia 20, 20021, Bollate, (MI), Italy

Stefano Montani

FSCN research centre, Mid Sweden University, 851 70, Sundsvall, Sweden

Fritjof Nilsson

RISE, Isafjordsgatan 28 A, 164 40, Kista, Sweden

Love Pallon

Materials Technology for Electrification-EMEME, Scania CV AB, 151 87, Södertälje, Sweden

Negin Yaghini

Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore, 117585, Singapore

Yuming Wen

Department of Materials Science and Engineering, School of Materials Science and Engineering KTH Royal Institute of Technology, Brinellvägen 23, 114 28, Stockholm, Sweden

Yuming Wen

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

S.R.L., Y.M.W., and M.S.H. initiated this project. Y.M.W., and M.S.H. supervised the study. S.R.L. did the experiement, performed the measurements and data analysis. Y.M.W. did the kinetic and dynamic activation energy calculation. S.R.L wrote the manuscript with the help of all other authors. G.H., K.K., S.M., F.N.,L.P.,N.Y. revised the manuscript.

Correspondence to Yuming Wen or Mikael S. Hedenqvist.

The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

Liu, S., Hunt, G., Kallio, K. et al. High temperature testing of PEEK and PPS coatings on copper wires. npj Mater Degrad 9, 26 (2025). https://doi.org/10.1038/s41529-025-00574-x

Download citation

Received: 16 October 2024

Accepted: 08 March 2025

Published: 17 March 2025

DOI: https://doi.org/10.1038/s41529-025-00574-x

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative