Achieving comprehensive performance regulation in Cu matrix composites via CNT modified design Yuansen Chen a,b, Jinyang Cao a,b, Wenmin Zhao a,b,c,d,*, Yikun Li a,b,
Baixiong Liu a,b,c,**, Hui Guo a,b,c, Qiuwen Liu a,b,c a State Key Laboratory of Nonferrous Structural Materials, Jiangxi University of Science and Technology, Ganzhou, 341000, China b School of Materials Science and Engineering, Jiangxi University of Science and Technology, Ganzhou, 341000, China c Gannan Laboratory, Ganzhou, 341000, China d Yunnan Key Laboratory of New Materials Preparation and Processing, Kunming, 650093, China A R T I C L E I N F O Keywords: Cu matrix composites Carbon nanotube Carbonized polymer dot Mechanical properties Electrical conductivity A B S T R A C T To alleviate the trade-off between strength-conductivity in carbon nanotube (CNT)/Cu composites, by using a one-step hydrothermal method to synthesize CPD in situ and simultaneously grafting it onto the surface of CNT, a CPD@CNT composite reinforcement is successfully prepared. Subsequently, the corresponding composites are fabricated by powder metallurgy. The results show that the CPD@CNT prepared at 190 ◦C exhibits a significant strengthening effect on the Cu matrix composites. Specifically, it has excellent tensile strength (389 MPa), good electrical conductivity (91.64 % IACS), and outstanding thermal stability (softening temperature>510 ◦C). This is attributed not only to the abundant oxygen-containing functional groups on the CPD surface, through which the dispersibility of CNT is enhanced and the interface wettability between the reinforcement and the matrix is improved, but also to the differential thermal stability of these functional groups, partial degradation occurs at the interface, resulting in the formation of a Cu2O transition phase that ultimately achieves robust interface bonding. Furthermore, significant dislocation strengthening is achieved via dislocation pile-up and the Orowan mechanism, owing to CPD@CNT dispersed at grain boundaries and within grains. Concurrently, an efficient electron transport pathway is established by the robust interface bonding, whereby interface electron scattering is effectively suppressed. This study provide a novel strategy for the design of high-strength, high-electrical conductivity Cu matrix composites.

  1. Introduction Pure Cu exhibits excellent electrical and thermal conductivity, but it has relatively low mechanical strength and is prone to softening at high temperatures, which severely limits its application in today’s complex and diverse scenarios. Therefore, by adding high-performance rein­ forcement to prepare Cu matrix composites, while retaining the high electrical conductivity of the Cu matrix and enhancing its mechanical strength and high-temperature stability, has become a research hotspot in this field. Cu matrix composites are widely acknowledged as pivotal materials in critical sectors, including electronic packaging, power transmission, and aerospace, by virtue of this excellent electrical and thermal conductivity coupled with favorable manufacturability. They are considered among the cornerstone materials underpinning modern electronic information and energy technology systems [1–4]. Carbon nanotube (CNT), as typical representatives of one-dimensional nano­ carbon materials, are considered ideal reinforcements for Cu matrix composites owing to it ultra-high tensile strength, excellent elastic modulus, and outstanding electrical properties [5–7]. Theoretically, The properties of CNT and the Cu matrix are highly complementary, and a composite system that integrates high strength with high electrical conductivity is expected to be constructed [8]. However, during prac­ tical fabrication, CNT tend to severely agglomerate due to van der Waals forces. Additionally, the poor physical wettability between CNT and the Cu matrix results in insufficient interface bonding strength [9,10]. Furthermore, due to the significant differences in their thermal expan­ sion coefficients, interface stress concentration and structural defects are caused [11]. Not only are the load transmission and carrier transport
  • Corresponding author. State Key Laboratory of Nonferrous Structural Materials, Jiangxi University of Science and Technology, Ganzhou, 341000, China. ** Corresponding author. State Key Laboratory of Nonferrous Structural Materials, Jiangxi University of Science and Technology, Ganzhou, 341000, China. E-mail addresses: zhaowenmin@jxust.edu.cn (W. Zhao), liu_micro@126.com (B. Liu). Contents lists available at ScienceDirect Ceramics International journal homepage: www.elsevier.com/locate/ceramint https://doi.org/10.1016/j.ceramint.2026.04.440 Received 23 February 2026; Received in revised form 9 April 2026; Accepted 30 April 2026
    Ceramics International 52 (2026) 29145–29155 Available online 30 April 2026 0272-8842/© 2026 Published by Elsevier Ltd.

efficiency severely impaired by these issues, but their enhancement potential is also constrained. At the same time, intense interface electron scattering is triggered, ultimately leading to a significant deterioration in the performance of the composites [12–14]. Consequently, the prac­ tical application of CNT/Cu composites is severely restricted by these factors. Therefore, developing effective interface and structure control strategies to balance the competing performance bottlenecks of “strength - conductivity” has become an urgent research priority in this field. In recent years, many studies have focused on optimizing the struc­ ture and designing the functions of CNT through multi-dimensional synergistic strengthening strategies, and introducing secondary phases to construct composite reinforcement systems. Furthermore, interfacial synergistic effects and multi-mechanism coupling effects must be fully exploited to surmount the performance limitations inherent to single reinforcement phases, thus the comprehensive properties of composites to be markedly enhanced. For example, Feng et al. [15] fabricated Cu matrix composites reinforced with titanium carbide (TiC)-modified CNT and in-situ nano-TiC particles. By taking advantage of the respective merits of 1-D CNT and 0-D TiC, the simultaneous enhancement of strength and ductility are achieved in the Cu matrix composites. Luo et al. [16] coated the surface of CNT with Cu to promote the dispersion of CNT and enhance the interface bonding between CNT and Cu. This resulted in a more uniform distribution of strain in the composites, thereby the strength improved from 14% to 26.5%. Chen et al. [17] treated CNT with acid and uniformly deposited a Cu (or Ni) layer on the surface of CNT via the electroless deposition method. Compression tests demonstrated that the composite containing 2 vol% Ni-coated CNT exhibited a maximum compressive yield strength of 554 MPa, which is 55% and 30% higher than that of pure Cu and the composite reinforced with have significant advantages, the inherent properties of the CNT are sacrificed, and the operation process is also considered to be rather complex. Furthermore, not only are substantial economic costs incurred by the acids and metal salts required for these experiments, but signif­ icant environmental risks and challenges are also posed by waste solu­ tion treatment. As a novel zero-dimensional carbon material with a core-shell structure, carbonized polymer dot (CPD) are composed of layers of graphene-like microstructures in the core, which are stacked together. The shells of these CPD are modified with abundant oxygen-containing functional groups. As a result, it possess the ability to uniformly disperse in polar solvents and exhibit excellent electronic transport properties [18,19]. However, when using CPD as the rein­ forcement, the structure of CPD is inherently unstable and its single reinforcing effect is limited. Based on these properties, the CPD employed to improve the dispersion of CNT in the Cu matrix presents a feasible strategy. Consequently, in this work, a one-step hydrothermal method was adopted to synthesize CPD, and it was simultaneously fixed on the surface of CNT. This not only significantly shortened the time and eco­ nomic cost required for acid treatment of CNT, but also effectively improved the efficiency. Subsequently, high-strength and high- conductivity CPD@CNT/Cu composites were fabricated via powder metallurgy. The comprehensive performance of the CPD@CNT/Cu composites was significantly enhanced by taking full advantage of the synergistic performance of CPD and CNT. The hydrothermal reaction temperature was precisely regulated to achieve the optimal conditions for the anchoring and combination of CPD on CNT. This study provides an important technical approach and research insight for the prepara­ tion of Cu matrix composites with excellent dispersibility, strong inter­ face bonding and outstanding comprehensive properties, thereby offering a novel design concept and technical route for the development of high-performance Cu matrix composites with integrated structural and functional properties. 2. Experimental methods The raw materials used in the experiment included: Hydroxyl- functionalized multi-walled carbon nanotube (CNT, diameter: 10–20 nm, length: 0.5–2 μm, purity: 95%, Nanjing XFNANO Materials Tech. Co., Ltd., China), Citric Acid (C6H8O7, CA, 99.5%, Shanghai Macklin Biochemical Technology Co., Ltd., China), Ethylenediamine (C2H8N2, EDA, 98%, Shanghai Macklin Biochemical Technology Co., Ltd., China), Fig. 1. Schematic illustration of the preparation process of (a) CPD@CNT composite powder; (b) CPD@CNT/Cu composites. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29146

and electrolytic Cu powder (particle size: 325 mesh, purity: 99.9%, Shanghai Naiou Nano Technology Co., Ltd., China). 2.1. Synthesis of CPD@CNT and composites As illustrated in Fig. 1(a), a one-step hydrothermal method was adopted to synthesize CPD and graft it onto CNT, yielding the CPD@CNT hybrid. Specifically, CNT were first ultrasonically dispersed in deionized water, followed by the sequential addition of CA and EDA. The mixture was subjected to ultrasonic treatment to form a homogeneous suspen­ sion, which was then transferred into a supercritical reaction autoclave for hydrothermal reaction under continuous stirring. After dialysis and freeze-drying, the CPD@CNT powder was obtained. To investigate the effect of functionality degree on CNT, the synthesis reaction was con­ ducted at four different temperatures (150 ◦C, 170 ◦C, 190 ◦C, 210 ◦C). The resulting products were sequentially labeled as Samples A, B, C and D to facilitate their identification and differentiation in subsequent studies. As shown in Fig. 1(b), the CPD@CNT composite powder with a mass fraction of 0.3% was mixed with pure Cu powder in an ethanol medium via variable-speed ball milling (200 rpm, 8 h + 400 rpm, 5 h). Subse­ quently, the milled powder was reduced in a hydrogen atmosphere and consolidated by spark plasma sintering (SPS). The sintered bulk was subsequently subjected to multi-pass hot rolling with a total deformation of approximately 65%, and the CPD@CNT/Cu composites was ulti­ mately fabricated. 2.2. Materials characterization The phase composition of the composite powders was determined using X-ray diffraction (XRD, Shimadzu XRD-6100). The powder morphology and composite microstructure were examined by scanning electron microscopy (SEM, MLA650F). The morphology of the rein­ forcement and the interfacial structure of the composite were charac­ terized by transmission electron microscopy (TEM, Tecnai G2 F20). Raman spectroscopy (Horiba LabRAM HR Evolution) was utilized to investigate the carbon structure of the CPD@CNT hybrid material. Electron backscatter diffraction (EBSD, attached to the MLA650F SEM) was used to evaluate the average grain size of the composite. The Vickers hardness of the CPD@CNT/Cu composites was measured at room tem­ perature using an HV-1000A tester under a 100 g load with a 10 s dwell time, and the electrical conductivity was determined at room tempera­ ture using an eddy current conductivity meter (Sigma 2008A). Finally, the mechanical properties of the composites were evaluated with a universal testing machine (UTM 4204) at a tensile rate of 0.1 mm/min. 3. Results and discussion 3.1. Microstructural characterization The microstructure of the pristine CNT and the as-prepared CPD@CNT composite powders is displayed in Fig. 2. The macroscopic state of pristine CNT and Samples A–D, after they are ultrasonically dispersed in ethanol and subsequently allowed to stand for 24 h under UV irradiation, is illustrated in Fig. 2(a). The highest fluorescence in­ tensity under UV light is exhibited by Sample C, indicating that the Fig. 2. (a) Photos of the prepared CPD@CNT in UV light (upper) and daylight (bottom), respectively; (b) TEM and HRTEM images (insets) of original CNT; (c) sample A; (d) sample B; (e) sample C; (f) sample D. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29147

formation of CPD is rendered most effective under this temperature condition. After standing, the pristine CNT are completely agglomerated and settled, whereas the CPD@CNT composite powders remain uni­ formly dispersed, which demonstrates excellent dispersion stability. A TEM image of the pristine CNT is shown in Fig. 2(b), the corresponding high-resolution TEM (HRTEM) image in the lower-right corner reveals that pristine carbon nanotubes have a hollow tubular structure, with a diameter of approximately 30 nm and a length of 0.5–2 μm, and their walls are characterized by smoothness without obvious defects. How­ ever, due to the strong van der Waals forces and high specific surface area between CNT, they are prone to agglomeration [20]. TEM images of the CPD@CNT composite powders for Samples A–D are presented in Fig. 2(c–f), and the dispersion state of the CNT and the generation of CPD on their surfaces are reflected by these images. With the increase in hydrothermal reaction temperature, the agglomeration degree of CNT first decreases and then intensifies, while the amount of CPD generated on their surfaces also shows a trend of initial increase followed by subsequent decrease. This indicates that a significant influence is exer­ ted by the reaction temperature exerts on the surface structure of CNT and the nucleation behavior of CPD. After CPD are successfully gener­ ated on the surface of CNT, the dispersion stability of CNT in solvent is improved by the abundant oxygen-containing functional groups on CPD. The upper-right and lower-right insets in Fig. 2(e) are HRTEM images, lattice fringes with interplanar spacings of 0.35 nm and 0.21 nm are displayed by these images, which correspond to the (002) plane of the CNT and the (100) plane of the CPD deposited on their surface, respectively. This further confirms that CPD are successfully formed on the surfaces of CNT. Distinct diffraction peaks at 2θ = 26.14◦are indicated by XRD diffraction analysis (Fig. 3(a)) for all CPD@CNT composite powders, and these peaks are assigned to the (002) crystal plane of CNT. It is note­ worthy that variations in the diffraction peak profiles of Sample A to D are shown with the change of hydrothermal reaction temperature: relatively broadened peaks are presented by Sample A and D, whereas the sharpest and most intense diffraction characteristics are demon­ strated by Sample C. This phenomenon indicates that an enhanced de­ gree of graphitization and a more perfect crystal structure are achieved for Sample C, from which it can be inferred that this specific processing parameter is most favorable for the improvement of graphitization in CPD@CNT composite reinforcement. Fourier Transform Infrared (FTIR) spectroscopy is utilized to sys­ tematically analyze the surface functional groups of the CNT. As shown in Fig. 3(b), the presence of O-H (~3420 cm−1), C-H (~2930 cm−1), C=O (~1700 cm−1), C=C (~1580 cm−1), N-H (~1380 cm−1), and C-N Fig. 3. Microstructural characterization of CPD@CNT: (a)XRD; (b) Infrared spectrum; (c) Raman spectrum. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29148

(~1210 cm−1) is indicated by the FTIR spectra. It is noteworthy that the C-N bond is clearly visible only in Sample A and B, whereas in Sample C and D, it is significantly weakened or even disappears. The reason for this phenomenon may be that at relatively low hydrothermal reaction temperatures, the synthesis of CPD is primarily dominated by poly­ merization and dehydration reactions between the precursors CA and EDA. During this process, a “polymer framework” mainly composed of polyamide-like structures is formed, accompanied by initial carboniza­ tion. As a result, a large number of amide bonds (-CO-NH-) are retained by the CPD, which are manifested as strong C-N and C=O bonds in the FTIR spectra [21]. On the other hand, explains why the C=O peak in the FTIR spectra of Sample A and B has a higher intensity, while it is weakened in Sample C and D. When the hydrothermal reaction tem­ perature is further increased, the degree of carbonization is intensified, which leads to the pyrolysis of unstable C-N bonds. Some heteroatoms are escaped in the form of volatile small molecules, while structural rearrangement is undergone by the remaining carbon atoms, and larger and more ordered sp2 structures are formed accordingly. Charge trans­ port scattering is thus greatly reduced, and the electrical conductivity of the material is further enhanced. It is confirmed by the FTIR results that CPD are successfully grafted onto the surfaces of CNT. The polar func­ tional groups contained in them not only enhance the stability and hy­ drophilicity of the composite powder in polar solvents, so that uniform dispersion of the powder in the matrix is promoted, but also facilitate the improvement of interface wettability between the CNT and the Cu ma­ trix, such that effective pathways for interface electron transport are provided [22]. The defect density and structural ordering of carbon materials are effectively reflected by the intensity ratio of ID/IG in Raman spectros­ copy [23]. As shown in Fig. 3(c), the ID/IG ratios for Sample A to D are 0.297, 0.322, 0.386, and 0.616, respectively, and a clear monotonic increasing trend is exhibited by these ratios. From XRD analysis, the sharpness of the characteristic peaks is interpreted as a reflection of the overall crystalline integrity of the material, with a typical detection depth in the micrometer range. In contrast, the surface and local states of the material are primarily probed by Raman spectroscopy, with a typical analysis depth ranging from several to tens of nanometers. With the increase in hydrothermal temperature, the amount of CPD generated on the surfaces of CNT is continuously increased. Concurrently, a signifi­ cant rise in the intensity of the D-band in Raman spectra is collectively led to by abundant surface functional groups and new structural defects formed between the CPD and CNT, which is manifested as a continuous increase in the ID/IG ratio. The discrepancy between the XRD and Raman results is precisely demonstrated to indicate that CPD are successfully nucleate and grown on the CNT surfaces without significantly compro­ mise to the overall skeletal integrity of the CNT. A sudden increase in the values observed for Sample D indicates that excessively high tempera­ tures also cause certain damage to the CNT structure itself. The distribution state of the reinforcement phase in the CPD@CNT/ Cu composite powder is systematically characterized by SEM. As shown in Fig. 4(a–d), the CPD@CNT reinforcements are uniformly distributed throughout the composite powder as a whole. However, a small number of clusters formed by the entanglement of CPD@CNT are still observed in Sample A (see Fig. 4(a)). With the increase of hydrothermal temper­ ature, the agglomeration phenomenon is significantly alleviated in Sample B (see Fig. 4(b)). For Sample C (see Fig. 4(c)), the CPD@CNT reinforcements are evenly distributed without any obvious agglomera­ tion. This indicates that under these condition, the nucleation effect of CPD on the surface of CNT is the most optimal. Thus, the surface func­ tional groups of CPD can more effectively promote the dispersion of CNT. Meanwhile, bridge-like and penetrating structures are observed, which demonstrates that close physical contact and geometric Fig. 4. Microstructure of CPD@CNT/Cu composite powders: (a) Sample A; (b) Sample B; (c) Sample C; (d) Sample D. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29149

compatibility are established between the reinforcements and the Cu powder particles, rather than simple physical adsorption. Nevertheless, when the temperature is further increased to the condition corre­ sponding to Sample D, local agglomeration reoccurs in the composite powder. It is inferred that excessively high hydrothermal temperature causes structural damage to the CNT themselves, the nucleation sites of CPD is reduced, and the surface modification and dispersion-promoting effects of CPD are weakened. This observation is consistent with the conclusions drawn from the TEM morphology and Raman spectrum analysis of the aforementioned CPD@CNT composite powder, which collectively verifies the significant influence of hydrothermal tempera­ ture on the dispersion behavior of the reinforcement phase. TEM and HRTEM are employed to analyze the interfacial micro­ structure of Sample C in Fig. 5. It is shown in Fig. 5(a) that CPD@CNT are not only present at grain boundaries but also distributed within the Cu matrix. A clean interface without obvious gaps or pores is revealed by magnified observation of a CPD@CNT at a grain boundary (Fig. 5(b)). The interface bonding between CPD@CNT and the Cu matrix is further subjected to high-resolution analysis (see Fig. 5(c)). In Fig. 5(d), a distinct transition zone of Cu2O is observed between the CPD@CNT and the Cu matrix. This zone is attributed to the decomposition of oxygen- containing functional groups on the CPD during the SPS sintering pro­ cess, by which active oxygen atoms are released. These active oxygen atoms are then reacted with Cu atoms from the Cu matrix, leading to the formation of a Cu2O transition zone at the interface. This observation is consistent with related studies in which oxygen-containing functional groups are utilized to improve the interfacial bonding of CNT/Cu com­ posites [22,24,25]. The FFT and IFFT images of the region within the white box are shown in Fig. 5(e1, e2). The observed interplanar spacings of 0.209 nm and 0.213 nm are assigned to the Cu (111) plane and the Cu2O (020) plane, respectively. A lattice mismatch of 10.57% is calcu­ lated. The clear lattice fringes observed indicate the formation of a semi-coherent interface. Lattice strain accommodation at the interface is achieved by the generation of dislocations, which is beneficial for the formation of a strong and tough interface bond. [26]. The Cu2O nano­ particles act as a bridge between CNT and the Cu matrix, and the Fig. 5. TEM images of CPD@CNT/Cu composite of sample C; (a-b) TEM images of CPD@CNT/Cu; (c) enlarged TEM image of the white box in b; (d) enlarged TEM image of the white box in c; (e1-e2) FFT and IFFT of the white box in d; (f) enlarged TEM image of the white box in c; (g1-g2) FFT and IFFT of the white box in f. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29150

effective enhancement of interfacial wettability as well as the reducing of interfacial energy are both considered to be associated with this. In Fig. 5(f), direct and tight bonding between the CPD@CNT and the Cu2O interface is observed. Based on the corresponding FFT and IFFT images, lattice spacings of ~0.206 nm and ~0.2135 nm are identified, corre­ sponding to the CPD (1 12) and Cu2O (00 2) crystal planes, respectively. This interfacial relationship collectively contributes to a continuous, Fig. 6. Formation of the transition zone. Fig. 7. (a-c) Different dislocation types of sample C; (d) IFFT image; (e) Inverse pole Fig (IPF) of the EBSD image; (f) Grain size statistics chart. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29151

gradient-strengthened interface structure extending from the CNT, through the Cu2O transition layer, to the Cu matrix. As shown in Fig. 6, the formation process of Cu2O is directly visu­ alized. During the sintering process, oxygen atoms are released from the oxygen-containing functional groups on the CPD surface, and a tighter transition layer is constructed between the CNT and Cu. The weak physical contact relying on van der Waals forces between CNT and Cu is upgraded to strong chemical bonding achieved through C-O-Cu link­ ages. In terms of mechanical properties, this transition layer acts as uniformly distributed nano-rivets, enabling efficient stress transfer from the Cu matrix to the CNT, which provides a fundamental guarantee for the remarkable strength enhancement of the material. In terms of electrical properties, due to the low content and dense structure of Cu2O, it mainly serves as a bridging connection, and its influence on the overall electron transport is minimized. Typical dislocation configurations observed in the CPD@CNT/Cu composites are displayed in Fig. 7. As is shown in Fig. 7(a), significant dislocation pile-ups are observed near grain boundaries. It is indicated that the transmission of dislocations into adjacent grains is effectively impeded by grain boundaries, leading to the generation of localized stress concentration. When mobile dislocations encounter CPD@CNT reinforcements, two distinct interaction mechanisms are observed. The first is illustrated in Fig. 7(b), where a large number of dislocations gliding on the same slip system are strongly pinned upon encountering a Fig. 8. (a) Bar chart of relative conductivity and hardness; (b) Engineering stress-strain curve; (c) True stress-strain curve; (d) Hardening rate curve; (e) Comparison of tensile and electrical conductivity properties with previously reported Cu composites [5,9,28–36]. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29152

CPD@CNT, and a zone of high-density dislocation accumulation is formed around it. This phenomenon is primarily attributed to the mismatch in the coefficients of thermal expansion between the CNT and the Cu matrix [27]. In the second mechanism, when a mobile dislocation meets an impenetrable CPD@CNT, the dislocation line is not simply piled up but can be bowed around and bypassed by the dislocation line, as shown in Fig. 7(c). It is revealed from the cross-sectional view that the dislocation line is circumvented by the reinforcement and its continuous motion is maintained, while a residual dislocation loop encircling the reinforcement is simultaneously left behind. These dislocation loops and the surrounding high-density dislocation network are clearly resolved at the atomic scale by HRTEM (Fig. 7(d)). The key microscopic mechanism enabling the composites to achieve high strength and high work-hardening capability is constituted by these multi-scale disloca­ tion interactions, which range from grain boundaries to grain interiors and from pinning to bypassing mechanisms. It is indicated by the IPF analysis of the EBSD image in Fig. 7(e) that a typical equiaxed grain structure with random grain orientation distribution and no significant texture formation is exhibited by the CPD@CNT/Cu composite. The average grain size of approximately 1.83 μm is obtained from the cor­ responding grain size statistical distribution (see Fig. 7(f)). This grain size is found to be significantly smaller than that of conventional pure Cu or single-reinforcement composites, and the notable grain refinement effect of the CPD@CNT composite reinforcement on the Cu matrix is directly confirmed. This refinement is primarily attributed to the pinning effect of the reinforcements on grain boundary migration during sintering and hot rolling, by which grain growth is effectively inhibited. The refined equiaxed grain structure is shown to contribute significantly to strengthening via the Hall-Petch mechanism, while the coordination of multiple slip systems is also facilitated, thereby enhancing the uni­ form deformation capability of the material. This microstructural characteristic, combined with the aforementioned dislocation strengthening mechanisms, collectively forms the microstructural foundation for the composite to achieve high strength and good ductility. 3.2. Electrical conductivity and mechanical properties Bar charts of the electrical conductivity and hardness values of the composites are presented in Fig. 8(a). The overall variations are rela­ tively minor, and high electrical conductivity (90.18%–91.64% IACS) is maintained by all samples. The preservation of conductivity can be attributed to the intrinsic sp2-carbon character and conductivity of the CPD themselves, coupled with the positive role played by them in improving the CNT-Cu interface and reducing electron scattering. The engineering stress-strain curves of the CPD@CNT/Cu compos­ ites are shown in Fig. 8(b). A marked enhancement in tensile properties is observed for Sample C, with an ultimate tensile strength (UTS) of 389.82 MPa and a yield strength (YS) of 331.23 MPa. This is attributed to the formation of a Cu2O transition zone between CPD@CNT and the Cu matrix. Efficient load transfer is enabled by this type of interface via strong chemical bonding, while electron scattering is simultaneously minimized due to its clean and well-defined nature. The enhancement in strength demonstrates that the introduction of CPD enables the achievement of significant strengthening while successfully circum­ venting the dilemma of a sharp decline in electrical conductivity induced by traditional reinforcement phases. However, a significant enhancement in strength is accompanied by a moderate decrease in plasticity. The fracture elongation of Sample C (13.32%) is found to be lower than that of Sample B (21.73%), where optimal ductility is exhibited. This observation is attributed to the fact that excessively high strength renders dislocation motion extremely difficult, and the critical stress for microcrack nucleation is reduced accordingly [15]. The true stress-strain and work hardening rate curves of the com­ posites are shown in Fig. 8(c) and (d), respectively. The work hardening rate curves clearly reveal the intrinsic differences in the deformation behavior between the CPD@CNT/Cu composites. A very high initial work hardening rate is observed for Sample C, which, however, is decreased drastically with increasing strain. It is indicated that the reinforcement (CPD@CNT) within this sample was extremely effective at hindering dislocation motion in the early stages of deformation. However, swift exhaustion of its strain hardening capacity is induced by the rapid accumulation of dislocations, thereby limiting its plastic deformability. In sharp contrast, although a slightly lower initial work hardening rate is exhibited by Sample B, its curve displays the most gradual decline. Excellent and stable work hardening capability is maintained throughout the entire deformation process. This sustained strengthening effect enables effective resistance to localized necking, thus achieving the maximum uniform plastic deformation. Good electrical conductivity is exhibited by Sample C while consid­ erable tensile strength is possessed. Although its elongation at break is decreased compared with that of Sample B, which exhibits optimal plasticity, this moderate sacrifice of ductility is considered acceptable in the design of multifunctional structural materials, and its absolute value is found to be significantly higher than that of many reported advanced Cu matrix composites [5,9,28–36]., as comparatively illustrated in Fig. 8 (e). The hardness evolution curve of Sample C after annealing for 1 h at temperatures ranging from 300 to 800 ◦C is displayed in Fig. 9. It is observed that as the annealing temperature increases, the hardness is first decreased slowly, then dropped sharply, after which the rate of decrease is moderated again. This corresponds to distinct microstruc­ tural evolution stages: the recovery stage, in which internal stresses are relieved without alteration of the deformed microstructure. The recrystallization stage, in which new, strain-free grains are nucleated and grown. And the grain growth stage, in which grains are grown to reduce the overall system energy [37]. According to the Chinese Na­ tional Standard (GB/T 33370–2016), the softening temperature of a specimen is defined as the annealing temperature at which its micro­ hardness is decreased to 80% of the original value. The material’s sta­ bility and tolerance under high-temperature conditions are reflected by the softening temperature, and it is served as a key indicator for eval­ uating the high-temperature resistance of Cu. As is indicated in Fig. 8, the softening temperature of sample C is greater than 510 ◦C, with a corresponding hardness of 90.448 HV at that point. This softening temperature is found to be significantly higher than the typical recrys­ tallization temperature range of pure Cu (200–300 ◦C), indicating that the high-temperature softening resistance of the composite is substan­ tially enhanced by the introduction of CPD@CNT. This improvement is primarily attributed to the synergistic pinning effect of the Fig. 9. High-temperature softening resistance curve. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29153

multi-dimensional reinforcements. Grain boundary migration is effec­ tively hindered by CPD@CNT distributed at grain boundaries, by which nucleation and growth during recrystallization are suppressed. Mean­ while, dislocations are subjected to stable obstacles by those dispersed within the grains, delaying dislocation recovery and rearrangement at elevated temperatures. The temperature at which significant softening occurs is collectively elevated by approximately 210 ◦C due to this multi-scale pinning effect at both interfaces and defects, fully demon­ strating that the core advantage of nano-carbon reinforcements is real­ ized in enhancing the thermal stability of Cu matrix composites. 4. High-strength, high-conductivity mechanism To achieve high strength and high electrical conductivity in Cu matrix composites, the aim is to simultaneously meet the load-bearing requirements of these materials as structural components and the effi­ cient energy transmission demands of them as electrical conductors. The multi-scale strengthening and conductive scattering transport pathways constructed by CPD@CNT within the Cu matrix are illustrated in Fig. 10. Its strengthening mechanism is manifested as dual dislocation obstruction at grain boundaries and within grains. As is shown in Fig. 10 (a), rigid non-deformable obstacles are acted as by CPD@CNT located at grain boundaries, leading to the accumulation of a large number of dislocations at their frontiers. The slip of dislocations across grain boundaries to adjacent grains is thereby effectively hindered, which significantly enhances the grain boundary strengthening effect of the composites [38]. According to the dislocation pile-up theory, a strong back stress is generated by the accumulation of dislocations in front of grain boundaries, thereby remarkably increasing the yield strength of the material. The contribution of grain boundary strengthening is thus constituted. Within grains, when a dislocation line moves on the slip plane and encounters CPD@CNT, the dislocation line is bent and grad­ ually bypasses the reinforcement phase, as direct cutting through the latter is not permitted. Subsequently, driven by the sustained applied stress, the bent dislocation line continues to bypass the reinforcement phase. Owing to the considerable length of CPD@CNT, the dislocation line proceeds to move forward after bypassing the reinforcement phase, a behavior analogous to the Orowan bypass mechanism. This process not only directly elevates the yield strength but also enables the continuous multiplication of residual dislocation loops, which in turn endows the material with excellent work-hardening capacity from another aspect. At the same time, CPD@CNT, through its unique “candied haw­ thorn” structure, not only greatly reduces electron scattering but also constructs efficient charge transport channels, thereby ensuring high conductivity. The core mechanism is found to lie in the fact that excel­ lent electrical conductivity is possessed by the sp2 carbon core of CPD itself, with which an efficient hybrid charge transport channel is formed together with CNT. A pathway with a lower energy barrier for electron transport from Cu to CNT and then back to Cu is provided by the formed transition layer, by which electron scattering at the interface is greatly mitigated and interfacial contact resistance is thereby significantly reduced (see Fig. 10(b)). 5. Conclusion In this work, CPD@CNT/Cu composites were prepared through a one-step hydrothermal method and SPS process. CNT are evenly dispersed in the Cu matrix with the help of CPD. CPD@CNT achieves dislocation strengthening through the accumulation of dislocations at grain boundaries and within grains, as well as the Orowan mechanism. Moreover, the abundant oxygen-containing functional groups on the CPD surface improve interfacial wettability. At the same time, the unique structure of CPD@CNT constructs efficient charge transport channels, significantly reducing electron scattering, thereby enabling the composite material to possess both excellent mechanical properties and electrical conductivity. CRediT authorship contribution statement Yuansen Chen: Writing – original draft, Validation, Investigation. Jinyang Cao: Methodology, Formal analysis, Data curation. Wenmin Zhao: Writing – review & editing, Supervision, Resources, Project administration, Investigation. Yikun Li: Methodology, Investigation, Formal analysis. Baixiong Liu: Project administration, Methodology, Investigation. Hui Guo: Supervision, Investigation. Qiuwen Liu: Vali­ dation, Investigation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgment This work was supported by the National Science Foundation of China (Grant No. 52304382), Yunnan Key Laboratory of New Materials Preparation and Processing (2024KF08), the Program of Jiangxi Pro­ vincial Laboratory of Advanced Metal Materials and Critical Minerals Development (2025GNJC02), Advanced Materials-National Science and Technology Major project (2025ZD0619700), and the Pioneer Research Program of Jiangxi Provincial Center for Interdisciplinary Studies in Future-oriented Materials. Data availability Data will be made available on request. Fig. 10. Strengthening mechanism diagrams: (a) Dislocation strengthening mechanism; (d) Conductive strengthening mechanism. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29154

References [1] X.Y. Wang, Z.D. Yang, L.T. Bian, W.T. Liu, G.Q. Zhang, J.C.T. Chen, P. Liu, Copper- based composite sintering materials and reliability analysis for power electronics packaging, J. Sci. Adv. Mater. Devices 10 (3) (2025) 100963, https://doi.org/ 10.1016/J.JSAMD.2025.100963, 100963. [2] M.M. Kang, X.C. Wang, Y.J. Wang, W. Meng, Y. Li, Z. Liu, Multispectral stealth textiles based on electromagnetic cage structure: design and performance regulation of graphene/copper composite system, J. Alloys Compd. (2025) 1036181852, https://doi.org/10.1016/J.JALLCOM.2025.181852, 181852. [3] H. F Li, Z.Z. Yong, D.D. Liu, K.J. Wu, L. Guo, X.N. Wang, H.H. Jin, Q.W. Li, Interface engineering for high-strength and high-ampacity of carbon nanotube/ copper composite wires, Carbon (2024) 219118845, https://doi.org/10.1016/J. CARBON.2024.118845. [4] O.O. Ekerenama, A. Ikeubab, C.N. Njoku, D.I. Njokud, W. Emorie, I.K. Nwokolon, I. N. Etimf, B.O. Okonkwog, I.I. Udohh, E.F. Danieli, P.C. Uzomaj, B.O. Awonusin, S. K. Kolawolek, I.P. Etiml, O.S. Olanrelem, Advancements in corrosion studies and protective measures for copper and copper-based alloys in varied environmental conditions, Results Eng. (2025) 105257, https://doi.org/10.1016/j. rineng.2025.105257. [5] X.F. Zhao, X.N. Ren, Z.P. Chen, C.C. Ge, Enhancement of mechanical and electrical properties of copper matrix composites by different types of carbon nanotubes, Mater. Charact. 218 (P2) (2024) 114575, https://doi.org/10.1016/J. MATCHAR.2024.114575, 114575. [6] Y.Y. Yu, C. Yang, Q. Gong, et al., Improving the electromagnetic interference shielding performance of CNTs/copper composites by interface modification with a carbon buffer layer, Mater. Lett. (2025) 391138489, https://doi.org/10.1016/J. MATLET.2025.138489, 138489. [7] Y.Y. Yu, C. Yang, Q. Gong, Y.T. Niu, H.L. Fu, Z.M. Luo, Y.Y. Zhang, Synergistic influence of carbon nanotube-graphene oxide hybrid and nanosized interfacial TiC on the mechanical performance of Cu matrix composites, J. Mater. Res. Technol. (2023) 252866–252879, https://doi.org/10.1016/J.JMRT.2023.06.140. [8] Y.L. Jia, K. Zhou, W.T. Sun, M. Ding, Y. Wang, X.Q. Kong, D.Z. Jia, M.H. Wu, Y. Fu, Enhancement mechanisms of mechanical, electrical and thermal properties of carbon nanotube-copper composites: a review, J. Mater. Res. Technol. (2024) 321395–321415, https://doi.org/10.1016/J.JMRT.2024.07.181. [9] S.L. Fu, X.H. Chen, P. Liu, Preparation of CNTs/Cu composites with good electrical conductivity and excellent mechanical properties, Mater. Sci. Eng., A (2020) 771138656, https://doi.org/10.1016/j.msea.2019.138656, 138656. [10] G.J. Yang, R.R. Wang, D. Fang, T. Hu, C.X. Bao, J.H. Yi, Nano-silver modified carbon nanotubes to reinforce the copper matrix composites and their mechanical properties, Adv. Powder Technol. 33 (8) (2022), https://doi.org/10.1016/J. APT.2022.103672. [11] F. Long, X.X. Guo, K.X. Song, et al., Synergistic strengthening effect of carbon nanotubes (CNTs) and titanium diboride (TiB 2) microparticles on mechanical properties of copper matrix composites, J. Mater. Res. Technol. 9 (4) (2020) 7989–8000, https://doi.org/10.1016/j.jmrt.2020.05.036. [12] F. Long, X.H. Guo, K.X. Song, S.G. Jia, V. Yakubov, S.L. Li, Y.B. Yang, S.H. Liang, Effect of CNTs content on the microstructures and properties of CNTs/Cu composite by microwave sintering, J. Alloys Compd. (2018) 771498–771504, https://doi.org/10.1016/j.jallcom.2018.08.315. [13] M.K. Z, G. Mahdi, B. Marek, W. Lidia, ´C. Nordin, B.P. D, K.K.K. K, Breaking the electrical barrier between copper and carbon nanotubes, Nanoscale 9 (24) (2017) 8458–8469, https://doi.org/10.1039/c7nr02142a. [14] F. Daneshvar, H.X. Chen, K. Noh, H.J. Sue, Critical challenges and advances in the carbon nanotube-metal interface for next-generation electronics, Nanoscale Adv. 3 (4) (2021) 942–962, https://doi.org/10.1039/D0NA00822B. [15] J.Q. Feng, J.M. Tao, X.F. Chen, Y.H. Liu, C.J. Li, J.H. Yi, Achieving strength–ductility balance in Cu matrix composite reinforced with double nanophase of CNT and intragranular in-situ TiC, J. Mater. Sci. Technol. (2025) 225247–225260, https://doi.org/10.1016/J.JMST.2024.11.035. [16] S.W. Luo, B. Chen, M. Song, Z.G. Zhang, J.H. Yi, S.F. Zhou, B.S. Guo, Z.T. Yu, W. Li, Improving the strength-ductility synergy of carbon nanotubes reinforced Cu matrix composites through interfacial regulation, Composites Part A 175 (2023), https:// doi.org/10.1016/J.COMPOSITESA.2023.107787. [17] X.F. Chen, J.M. Tao, J.H. Yi, Y.H. Liu, R. Bao, C.J. Li, S.L. Tan, X. You, Enhancing the Strength of Carbon Nanotubes Reinforced Copper Matrix Composites by Optimizing the Interface Structure and Dispersion Uniformity, Diamond & Related Materials, 2018, pp. 8874–8884, https://doi.org/10.1016/j.diamond.2018.06.026. [18] W.M. Zhao, R. Bao, J.H. Yi, Y. Zhang, Y.F. Pu, J.T. Zhang, Influence of carbonized polymer dot (CPD) structure on mechanical and electrical properties of copper matrix composite, Mater. Charact. 181 (2021), https://doi.org/10.1016/J. MATCHAR.2021.111463. [19] W.M. Zhao, R. Bao, J.H. Yi, Y. Zhang, Y.F. Pu, J.T. Zhang, Achieving a better mechanical enhancing effect of carbonized polymer dots than carbon nanotubes and graphene in copper matrix, Compos. Commun. (2021) 100906, https://doi. org/10.1016/J.COCO.2021.100906. [20] H. Deng, J.H. Yi, C. Xia, Y. Yi, Mechanical properties and microstructure characterization of well-dispersed carbon nanotubes reinforced copper matrix composites, J. Alloys Compd. (2017) 727260–727268, https://doi.org/10.1016/j. jallcom.2017.08.131. [21] C.L. Xia, J.D. Zhong, X. Han, S.J. Zhu, Y.F. Li, H. Liu, B. Yang, The formation mechanism of carbonized polymer dots: crosslinking-induced nucleation and carbonization, Angew. Chem. 63 (44) (2024) e202410519, https://doi.org/ 10.1002/ANIE.202410519. [22] S.R. M, S. Atsuko, Y. Takeo, K. Ken, H. Kenji, Improving carbon nanotube/copper film composite electrical performances by tailoring oxygen interface through gaseous ozone treatment of carbon nanotube films, Synth. Met. 288 (2022), https://doi.org/10.1016/J.SYNTHMET.2022.117103. [23] H. Geng, B. Chen, J. Wan, J. Shen, K. Kondoh, J.S. Li, Matrix effect on strengthening behavior of carbon nanotubes in aluminum matrix composites, Mater. Charact. 195 (2023), https://doi.org/10.1016/J.MATCHAR.2022.112484. [24] X.F. Chen, J.M. Tao, J.H. Yi, Y.C. Liu, R. Bao, C.J. Li, S.L. Tan, X. You, Enhancing the Strength of Carbon Nanotubes Reinforced Copper Matrix Composites by Optimizing the Interface Structure and Dispersion Uniformity, Diamond & Related Materials, 2018, pp. 8874–8884, https://doi.org/10.1016/j.diamond.2018.06.026. [25] M. Park, B.H. Kim, S. Kim, D.S. Han, G. Kim, K.R. Lee, Improved binding between copper and carbon nanotubes in a composite using oxygen-containing functional groups, Carbon 49 (3) (2010) 811–818, https://doi.org/10.1016/j. carbon.2010.10.019. [26] J.J. Liu, J.R. Huang, Y.B. Zhang, X. Sun, X. Yang, S.H. Du, T.J. Li, T.M. Wang, A novel strategy of dual in-situ synthesis for graphene/Cu composites with high strength and high electrical conductivity, Carbon (2025) 244120645, https://doi. org/10.1016/J.CARBON.2025.120645, 120645. [27] M. Shakti, M. Sudipta, M.B. Santu, D. Siddhartha, D. Karabi, Synergistic effect of bimodal structure and Cu-coated CNTs on the properties of Cu/CNTs composites, J. Alloys Compd. 971 (2024), https://doi.org/10.1016/J.JALLCOM.2023.172681. [28] Y.Z. Li, R. Long, J.M. Tao, X.F. Chen, Y.C. Liu, C.J. Li, Z.Y. Xu, J.H. Yi, Regulation of the interfacial structure in CNTs/Cu composites by chemical vapor deposited WC to enhance mechanical properties and conductivity, Ceram. Int. 51 (17) (2025) 23725–23740, https://doi.org/10.1016/J.CERAMINT.2025.03.060. [29] H.Q. Wu, L.P. Deng, X.C. Pan, B.S. Wang, H.L. Xiang, Study on the microstructure, properties and strengthening mechanisms of electrophoretic deposited carbon- nanotubes/Cu composites, Mater. Charact. (2025) 221114786, https://doi.org/ 10.1016/J.MATCHAR.2025.114786, 114786. [30] L.Q. Zhang, R. Bao, J.H. Yi, S.D. Guo, J.M. Tao, C.J. Li, D. Fang, Y.C. Liu, F.X. Li, Improving comprehensive performance of copper matrix composite by spray pyrolysis fabricated CNT/W reinforcement, J. Alloys Compd. 833 (2020), https:// doi.org/10.1016/j.jallcom.2020.154940. [31] B. Ya, F.Q. Meng, Y. Xu, B.W. Zhou, L.G. Meng, X.G. Zhang, Enhancing the mechanical properties of copper matrix composites through nickel-grafted carbon nanotubes prepared by molecular-level mixing and self-reduction methods, J. Alloys Compd. (2025) 1020179546, https://doi.org/10.1016/J. JALLCOM.2025.179546, 179546. [32] X. Wei, J.M. Tao, Y. Hu, Y.C. Liu, R. Bao, F.X. Li, D. Fang, C.J. Li, J.H. Yi, Enhancement of mechanical properties and conductivity in carbon nanotubes (CNTs)/Cu matrix composite by surface and intratube decoration of CNTs, Mater. Sci. Eng., A 816 (2021), https://doi.org/10.1016/J.MSEA.2021.141248. [33] P. Yang, X. You, J.H. Yi, D. Fang, R. Bao, T. Shen, Y.C. Liu, J.M. Tao, C.J. Li, S. L. Tan, S.D. Guo, Simultaneous achievement of high strength, excellent ductility, and good electrical conductivity in carbon nanotube/copper composites, J. Alloys Compd. (2018) 752431–752439, https://doi.org/10.1016/j.jallcom.2018.03.341. [34] S.Y. Mu, J.M. Tao, X.F. Chen, Y.C. Liu, R. Bao, C.J. Li, F.X. Li, J.H. Yi, The configuration design of cross-linked CNT networks to realize heterostructure in Cu matrix composite towards prominent mechanical-electrical property synergy, J. Mater. Res. Technol. (2024) 307447–307461, https://doi.org/10.1016/J. JMRT.2024.05.131. [35] Y.Q. Zhang, H. Zhang, J.M. Tao, Y.C. Liu, R. Bao, C.J. Li, F.X. Li, X.F. Chen, D. Ye, J. H. Yi, Synergistic optimization of properties in carbon nanotubes reinforced Cu matrix composites prepared by co-deposition, Ceram. Int. 50 (11PA) (2024) 18337–18346, https://doi.org/10.1016/J.CERAMINT.2024.02.318. [36] Y. Pan, S.Q. Xiao, X. Lu, C. Zhou, Y. Li, Z.W. Liu, B.W. Liu, W. Xu, C.C. Jia, X.H. Qu, Fabrication, mechanical properties and electrical conductivity of Al2O3 reinforced Cu/CNTs composites, J. Alloys Compd. (2018) 7821015–7821023, https://doi. org/10.1016/j.jallcom.2018.12.222. [37] J.C. Dong, H.T. Liu, W.Q. Li, Y.X. Hua, S.Y. Zhang, K.X. Song, Effect of microalloying with different contents of Al and Sc on the microstructure and properties of pure copper, J. Mater. Res. Technol. (2025) 391480–391493, https:// doi.org/10.1016/J.JMRT.2025.09.126. [38] S.Q. Zhou, W. Zhang, M.B. Liu, W.J. Ren, Q.H. Zhou, J.R. Wei, P.F. Wu, Strength- toughness combination in nickel matrix composites reinforced by hybrid graphene nanoplatelets-titanium diboride, Carbon (2023) 2011137–2011148, https://doi. org/10.1016/J.CARBON.2022.10.033. Y. Chen et al.
Ceramics International 52 (2026) 29145–29155 29155