Journal of Alloys and Compounds 971 (2024) 172681

Contents lists available at ScienceDirect

Journal of Alloys and Compounds

journal homepage: www.elsevier.com/locate/jalcom

Synergistic effect of bimodal structure and Cu-coated CNTs on the

properties of Cu/CNTs composites
Shakti Mishra a, Sudipta Mohapatra a, Bhaskar Santu Mudliyar b, Siddhartha Das a, Karabi Das a, *
a
Department of Metallurgical and Materials Engineering, Indian Institute of Technology Kharagpur, Kharagpur, West Bengal 721302, India
b
Central Research Facility, Indian Institute of Technology Kharagpur, Kharagpur, West Bengal 721302, India

A R T I C L E I N F O A B S T R A C T

Keywords: Multi-walled carbon nanotube-reinforced Cu matrix composites (Cu/MCNTs) present a challenge in achieving
Cu matrix composites high strength and good ductility. To address this issue, the bimodal structure composites were fabricated by
Carbon nanotubes vacuum mixing of coarse Cu powder (Cum) and fine Cu powder along with Cu-coated CNTs (CCNTs) followed by
Mechanical properties
spark plasma sintering (SPS). Two sets of bimodal composites with 1 wt% Cu-coated CNTs were fabricated by
Interface
Spark plasma sintering
varying the proportion of fine to coarse Cu powder (70:30 and 30:70). The unimodal composite was prepared by
mixing Cum and 1 wt% functionalized CNTs in a ball mill for 1 h followed by SPS. Using electron backscatter
diffraction (EBSD), the microstructure of the prepared composites was examined, and the influence of coarse
grain and fine grain, along with CCNTs, on the structure-property correlation in the composites was investigated.
The results show that the bimodal composites outperform the unimodal counterpart, and the bimodal composite
with fine to coarse Cu grain ratio of 70:30 has the highest compressive strength (323 MPa), compressive strain
(18.8%) electrical conductivity (85.4 ± 0.6% IACS), and thermal conductivity (295.4 ± 5 W/mK). The increase
in strength of the bimodal composite is attributed to the Cu-coated CNTs and the tuning between coarse and fine
grain. Load transfer, aided by the Cu coating on the CNT surface, is revealed to be the primary strengthening
mechanism for bimodal composites. This work sheds light on the mechanisms of strengthening and the structure-
property correlations in bimodal composites.

1. Introduction
Traditional metal like copper (Cu) has been extensively used in
various fields such as electronics, serving as electrical connectors, and in
power transportation as wires or cables due to its excellent thermal
conductivity (400 W/mK) and low electrical resistivity (1.72 × 10− 8
Ωm) [1,2]. Despite its good malleability, the engineering and industrial
use of Cu is limited due to its low mechanical strength [3,4]. Many re­
searchers have been focusing on developing Cu matrix nanocomposites
with ultrafine grains, containing nano-reinforcements. These contribute
to enhance the hardness and strength, a result of strengthening effects
[5].
Among various nano-reinforcements, Carbon nanotube (CNT) has
emerged as an ideal candidate due to their unique properties like
excellent mechanical strength, low coefficient of thermal expansion, and
high thermal and electrical conductivity [6–8]. In the past two decades,
CNTs reinforced Cu matrix (Cu/CNTs) composites have been fabricated
using several methods, including hot pressing [9], electroless deposition
[10], mechanical alloying [11], spray pyrolysis [12], and liquid metal­
lurgy [13]. However, CNTs have a strong "pinning" effect on the grain
boundaries, resulting in small grain sizes which enhances the strength,
but reduces the ductility and consequently impedes their structural
applications [14,15].
The trade-off between strength and ductility could be fixed by
tailoring the microstructure by introducing both coarse and fine grains
[16]. D. Mesguich et al. fabricated a bimodal microstructure of Cu
composites by mixing the two different-sized Cu powders and 1 vol%
CNT followed by the combination of spark plasma sintering and RT
wire-drawing. This process is favorable to enhance the strength while
limiting the increase in electrical resistivity [17]. L. Meng et al. [18]
provided a feasible way to fabricate the laminated Cu/CNTs composites
with a bimodal-grained structure where the CNTs were deposited on 30
and 100 µm thick Cu foil by electrophoretic deposition and then stacked
together followed by hot pressing and rolling. The composite fabricated
using 30 µm thick Cu foil exhibited the highest yield strength (225 MPa),
ultimate tensile strength (245 MPa) and ductility (32.7%). The

* Corresponding author.
E-mail address: karabi@metal.iitkgp.ac.in (K. Das).

https://doi.org/10.1016/j.jallcom.2023.172681
Received 26 July 2023; Received in revised form 30 September 2023; Accepted 26 October 2023
Available online 30 October 2023
0925-8388/© 2023 Elsevier B.V. All rights reserved.
S. Mishra et al.
increment in yield strength and ductility is attributed to the introduced
CNTs for grain refinement, bearing the load, and restricted dislocation
motion [18]. In bimodal microstructure, the coarse grains provide more
space for newly generated dislocations during deformation, whereas the
dislocation density in fine grains is saturated very quickly with
increasing strain [19]. Thus, materials with high strength and ductility
can be designed with a bimodal grain size distribution in which fine
grains increase strength, while coarse grains maintain work hardening
and increase ductility [20,21]. As a result, the deformation behavior of
materials with bimodal grain size distribution differs from that of
coarse-grain or ultrafine-grain materials.
Taking into consideration the previously discussed points, the pre­
sent study employs a simple and easy technique for preparing unimodal
and bimodal Cu/MCNTs composites with 1 wt% of CNTs. The coarse Cu
powder was mixed with functionalized multi-walled carbon nanotubes
(FMCNT) using ball-milling for 1 h to fabricate unimodal Cu/MCNTs
composite powder. Similarly, bimodal Cu/MCNTs composite powder
was produced by mixing fine Cu powder along with Cu-coated CNT
(fabricated by wet chemical route (WCR)) and the coarse Cu powder,
using a vacuum mixer. All the composite powders were sintered using
spark plasma sintering (SPS) to obtain pellets. The primary aim of this
strategy is to modify the inherent interface structure without damaging
the CNT surface, and to achieve a bimodal structure by combining the
wet chemical route and vacuum mixing process. The second aim of this
study is to examine the relationship between structure and properties by
adjusting the amount of coarse (CG) and fine (FG) particles in Cu powder
while keeping the amount of CNTs fixed at 1 wt%. In addition, this study
reports the thermal and electrical conductivity values of unimodal and
bimodal Cu/MCNTs composites, which may provide new insights into
the synergistic effect of bimodal structures and Cu-coated CNT.
Furthermore, the strengthening mechanisms of unimodal and bimodal
Cu/MCNTs composites, including dislocation strengthening, strength­
ening due to load transfer, and grain strengthening, are also discussed.
2. Experimental
2.1. Preparation of composite powder and processing
The MCNTs with purity > 95%, outer diameter 5–20 nm, and length
1–10 µm were purchased from Plasma Chem GmbH Rudower Chaussee
29, Berlin. The functional groups C–
– O, O–H, and C–O are present on the
functionalized surface of the MCNTs. The particle size of the fine copper
powder, obtained by the wet chemical route is determined to be 188 ±
23 nm. Initially, the unimodal Cu/MCNTs composite powder was
fabricated using a high-energy planetary ball mill (Fritsch Vario-
Planetary ball mill, pulverisette-4). The purchased Cu powder (99.5%,
loba chem., − 40 + 100 mesh) was milled for 9 h with a rotation speed of
300 rpm. To prevent oxidation, the powder was milled in a stainless-
steel container with 10 mm-diameter steel balls in an Ar atmosphere
and a 10:1 ball to powder ratio. Toluene was used as a process control
agent during milling to prevent the increase in vial temperature caused
by particle and ball collisions. The milled Cu (Cum) was mixed with
functionalized CNTs (1 wt%) for 1 h in a ball mill to obtain a unimodal
(U-Cu/CNT) composite powder.
Further, the bimodal Cu/MCNTs composite powders were produced
by mixing coarse powder (Cum) and fine Cu powder (Cuw) along with
Cu-coated CNTs (CCNTs) in a vacuum mixer for one hour. The coarse Cu
powder was produced by ball milling as received powders for 9 h in a
ball mill, whereas fine Cu powder along with Cu-coated CNTs was
produced by wet chemical route.
In the present study, coarse (CG) and fine (FG) Cu powders are
denoted as Cum and Cuw, respectively. Two different compositions of
powders, one with XCu+ 1 wt% Cu-coated CNT (where, X = 70% CG +
30% FG) and the second with YCu+ 1 wt% Cu-coated CNT (where, Y =
30% CG + 70% FG) were used to fabricate bimodal B-70Cu/CCNT and
B-30Cu/CCNT, respectively. A schematic representation of the
2
Journal of Alloys and Compounds 971 (2024) 172681
fabrication process for pure Cu and Cu/MCNTs composite powder by
powder metallurgy (P/M) route is shown in Fig. 1. The nomenclatures
for representing the composites are listed in Table 1. The prepared
unimodal (U-Cu/CNT and U-Cu/CCNT) and bimodal (B-70Cu/CCNT
and B-30Cu/CCNT) composite powders were consolidated in a graphite
die (Ø 10 mm) to obtain pellets, followed via spark plasma sintering
(FUJI SPS 625, Fuji Electronic Industrial Co. Ltd., Japan). The SPS sin­
tering process was carried out under an Ar atmosphere at 900 ºC for 5
min holding time at 80 MPa applied pressure with 100 ºC/min heating
rate.
The Cu-coated CNTs along with fine Cu powder were produced by
wet chemical route by reducing the metal precursor CuSO4.5 H2O, in the
presence of CTAB and MCNTs under the reducing environment of NaOH
and NaBH4. This process was conducted while maintaining the pH (12)
and temperature at 5–8 ◦ C. The mixture was further stirred for 8 h to
ensure that all the precursor metal ions had been reduced to Cu. The
resulting precipitate was then cleaned, dried, and heat-treated under a
hydrogen atmosphere to yield Cu powder and Cu-coated CNTs. The
composition of the baths was chosen based on stoichiometric calcula­
tions to yield the required amount of Cu powder. The detailed synthesis
process has been described in our previous publication [22].
2.2. Structural characterization
The microstructural characterization of Cu composite powders and
sintered pellets was carried out using a scanning electron microscope
(Zeiss Merlin, Gemini 2), an emission scanning electron microscope (Zeiss
Merlin, Gemini 2) equipped with an EBSD system (TSL- EDAX), and an
analytical transmission electron microscope (TEM) (FEI TechnaiG2,
200 kV voltage). The preparation of TEM specimens involves cutting thin
slices (900 µm) followed by mechanical grounding down to 100 µm
thickness. In addition, a disc with a diameter of 3 mm was punctured
using a Gatan disc punch (Model number: 1104). Afterwards, ion milling
was performed using the Gatan precision ion polishing system (Model
691) to produce a transparent region. The samples were well polished
with progressive grades of emery paper (600–4000), followed by cloth
polishing with 0.25 µm diamond paste and colloidal solution (0.05 µm)
with the help of a vibratory polisher machine (VibroMetTM 2) for
observing the microstructure with SEM and EBSD. In this study, the
EBSD scans for all pellet samples were carried out over an area of 10 µm
× 10 µm using a step size of 0.25 µm. Further, post-processing of the
obtained raw data from the EBSD scan was done using orientation im­
aging microscopy (OIM) software (TSL-OIM version 7.2).
The Raman spectra and the normal phase identification of the
composites were obtained using a Raman spectrometer (T64000, Horiba
Jobin-Yvon) with Innova 70-C spectrum laser (coherent) and Bruker D8
Advanced X-ray diffractometer (1.54 Ȧ at a step size of 0.02◦ s− 1),
respectively, in the 2θ range of 20–100º. Using the Archimedes principle
and a Vickers hardness tester (Leica VMHT) with an applied load of
15 kgf and a dwell time of 10 s, the density and hardness of Cu and Cu/
MCNTs composites were determined. Additionally, an instrumented
nanoindentation tester (Hysitron Model: TI 950 Triboindenter) was used
to measure the elastic modulus and nanohardness values for sintered Cu
and Cu/MCNTs composite pellets with a Berkovich tip (specification: TI-
0039, tip diameter of 200 nm) at 8000 μN load. The thermal conductivity
and thermal diffusivity of the prepared Cu, unimodal and bimodal Cu/
MCNTs composite pellets (dimensions Ø 10 mm × 2.2 mm) were
calculated using a laser flash technique-assisted thermal conductivity
machine (Netzsch LFA 427) at room temperature. Further, the electrical
conductivity of prepared Cu and all Cu/MCNTs composite pellets were
obtained using a four-probe resistance measurement system (Keithley
2182 Nano Voltmeter). The compression tests were carried out using a
universal testing machine (Tinius Olsen-H50KS) with a strain rate of
1 × 10− 2 s− 1 at room temperature. The cylindrical samples (Ø
2.8 mm × 4.3 mm) were prepared for compression test as per ASTM E9
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 1. Schematic showing the synthesis of unimodal and bimodal Cu/MCNTs composites.

TEM micrograph of Cu-CNT composite powder, which is shown in

Table 1
Fig. 3b. The selected area diffraction (SAD) patterns of marked regions
The nomenclature of the composites.
(red box) in Fig. 3a correspond to FMCNT and the Cu in Cu-CNT com­
Sample Composition Nomenclature posite powder, as shown in Fig. 3d and e, respectively. The lattice fringe
Unimodal Coarse Cu powder (CG) + Functionalized CNT (1 wt U-Cu/CNT spacing and EDS spectrum of the selected region, as indicated in Fig. 3b
%) are shown in Fig. 3c and f, respectively. The interplanar spacing values
Fine Cu powder (FG) + Cu-coated CNT (1 wt%) U-Cu/CCNT
of 0.21 and 0.33 nm are assigned to the (111) plane of Cu and the (002)
Bimodal (Coarse + Fine) Cu powder + 1 wt% CCNTCoarse: B-70Cu/CCNT
fine = 70:30 plane of CNTs, respectively (Fig. 3c).
(Coarse + Fine) Cu powder + 1 wt% CCNTCoarse: B-30Cu/CCNT
fine = 30:70 3.2. Microstructural characterization of pellets

To characterize the microstructure of pellet samples, the SEM and

standards. After the deformation of the samples, the fracture behavior of
EBSD micrographs of sintered Cum and Cu/MCNTs composite pellets
the pellets was examined through the cracks present on the curved
(1 wt% CNT) are studied, as depicted in Fig. 4 and Fig. 5, respectively.
surface after the compression test.
The SEM micrographs of the pure Cum, U-Cu/CNT, B-70Cu/CCNT, and
B-30Cu/CCNT composite pellet are shown in Fig. 4a, b, c & d, respec­
3. Results
tively. As shown in Fig. 4b, the twins are present in large numbers in the
grains of unimodal Cu/MCNTs composite. In this composite, the critical
3.1. Microstructural characterization of powder
shear stress for twinning is exceeded by the shear stress (Pmax) induced
by ball milling during mechanical alloying of Cum and FCNTs, which
As depicted in Fig. 2(a-d), the SEM micrographs of Cu powder at
results in the formation of twins [24]. It is also observed that some pores
various milling times (a) 0 h, (b) 2 h, (c) 6 h, and (d) 9 h reveal
exist along the grain boundaries of the Cu matrix in U-Cu/CNT due to the
morphological changes. In the early stages of milling, the morphology of
presence of the cluster of shortened FMCNTs (Fig. 4b). The presence of
Cu powders turns from spherical to flaky particles. Due to continuous
twins can also be seen in bimodal B-70Cu/CCNT and bimodal
impact, fracturing, and welding, Cu particles acquire irregular shapes as
B-30Cu/CCNT composites, as shown in Fig. 5g and Fig. 4d. This is due to
the milling process continues. Functionalized MCNTs were mixed with
the presence of fine Cuw (synthesis by WCR) in composites [22]. Further,
Cu (9 h milled) powder and ball-milled for 1 h to produce a unimodal
the IPF and IQ map for Cum and Cu/MCNTs composite pellets are
composite. The short milling time was performed to avoid serious
illustrated in Fig. 5(a, c, g and j) and Fig. 5(b, d, h and k), respectively.
damage to the FMCNT surface in the Cu metal matrix [23]. Function­
The coarse grain (CG) is heterogeneously distributed in the bimodal
alized CNTs get fractured and shortened during 1 h of milling. In
B-70Cu/CCNT composite, as seen in Fig. 5h. In contrast, CG is uniformly
unimodal composite powder, FMCNTs are properly embedded into the
distributed within the fine grain (FG) in a bimodal B-30Cu/CCNT
deformed Cu matrix, as shown in Fig. 2e. In bimodal B-30Cu/CCNT
composite (Fig. 5k). The quantitative EBSD analysis of all the samples
composite powder, Cu-coated CNTs are uniformly distributed in the Cu
reveals that the fraction of low-angle grain boundaries (LAGBs) is
matrix, as revealed in Fig. 2f. The EBSD analysis was carried out for the
greater in bimodal Cu/MCNTs (Fig. 5i & l) composites than in unimodal
better understanding of the difference in microstructure between
Cu/MCNTs composites (Fig. 5f). The fraction of LAGBs in the
unimodal and bimodal composite.
B-70Cu/CCNT bimodal composite is 1.1 times that of the B-30Cu/CCNT
The coating of Cu on the surface of CNT can be observed from the
bimodal composite.

3
S. Mishra et al.
Fig. 2. SEM micrographs of (a-d) Cu powder ball-milled for zero h, 2 h, 6 h,
and 9 h; (e) unimodal Cu/MCNTs, and (f) bimodal B-30Cu/CCNT composites.
The grain orientation spread (GOS) maps of all the samples before
the compression test are presented in Fig. 6(a-d). The grains showing
misorientation between 0 and 1ᵒ are considered as recrystallized grains
and those showing misorientation above 1ᵒ are considered as deformed
grains [25]. The GOS values of recrystallized grains (sum of the fractions
˂ 1ᵒ) in Cu pellet, unimodal Cu/MCNTs composite (U-Cu/CNT), bimodal
Cu/MCNTs composite (B-70Cu/CCNT), and bimodal Cu/MCNTs com­
posite (B-30Cu/CCNT) sample are found to be 0.13, 0.18, 0.51, and
0.24, respectively. The bimodal Cu/MCNTs composites exhibit higher
GOS values as compared to the unimodal Cu/MCNTs composite. There is
a significant difference in GOS values between B-70Cu/CCNT (0.51) and
B-30Cu/CCNT (0.24) samples. It is attributed to the incorporation of
more amount of coarse Cu in B-70Cu/CCNT as compared to the
B-30Cu/CCNT sample. The highest GOS value in bimodal Cu/MCNTs
composite (B-70Cu/CCNT) among all the samples indicates that more
amount of strain can be accommodated during deformation when the
sample is subjected to a compression test.
4
Journal of Alloys and Compounds 971 (2024) 172681
3.3. Structural characterization of composite powders and pellets
The XRD patterns of Cu powder (0, 2, 6, and 9 h of milling), and
sintered Cu, unimodal and bimodal Cu/MCNTs composite pellets are
shown in Fig. 7a and b, respectively. The XRD analysis (Fig. 7a) suggests
that no impurity is incorporated into pure metallic Cu powder and the
absence of Cu oxidation during the milling process. The XRD pattern for
three different Cu/MCNTs composite pellets with the same composition
(1 wt%) is plotted, and the resulting peaks are matched with the ICDD
reference data for Cu (00–004–0836), as shown in Fig. 7b. The MCNTs
peak is not found in this XRD pattern, as only small amount of MCNTs is
present in the Cu matrix as reinforcement.
The structural defects of CNT in unimodal U-Cu/CNT and bimodal B-
Cu/CCNT composites are evaluated using Raman spectroscopy of sin­
tered pellet, as shown in Fig. 7c. Appearances of only two peaks, at
1337 cm− 1 and 1580 cm− 1 (D and G band) [26], in Raman spectra of the
respective composites confirms that the structural integrity of CNT
tubular is maintained in the final composites. The ID/IG ratio increases
from 1.07 in FMCNT to 1.34 in the U-Cu/CNT composite pellet. This
indicates that the defect density is 1.25 times higher in the U-Cu/CNT
composite than that of FCNT. Additionally, it is also observed that the
ID/IG ratio in the U-Cu/CNT composite is higher than that of the bimodal
composites. These defects are presumably introduced during milling for
1 h. On the other hand, the intensity ratio (ID/IG) of bimodal
B-70Cu/CCNT and B-30Cu/CCNT composite pellets is found to be 1.10
and 1.12, respectively, which are almost similar values. These results
show the nominal structural damage of MCNTs, leading to better
properties in final bimodal Cu/MCNTs composites. The results show that
the Cu coating can be an appropriate shield for MCNTs to avoid large
structural destruction. The broadening and shifting of the G band in the
spectrum are mainly attributed to the existence of residual strain, which
modifies the inter-atomic spacing between carbon atoms in the MCNTs
[27].
The oxidation state of the constituent elements present on the MCNT
surface is analyzed using X-ray Photoelectron Spectroscopy (XPS). The
survey scan of the unimodal U-Cu/CNT composite indicates the exis­
tence of Cu and C (Fig. 7d). The peaks corresponding to the Cu 2p3/2 and
Cu 2p1/2 appear at 933.5 and 953.7 eV, respectively (Fig. 7e). Also,
Fig. 7f shows the three deconvoluted peaks of the C at 284.3, 286 and
287.2 eV, which arise from carbon-carbon interactions, including C-C
sp2 bond, carbon-oxygen interactions including C-O bond, and carbonyl
bond C– – O, respectively [28,29]. These C– – O and C-O bonds indicate
the existence of oxygen functional groups on the MCNTs surface [22].
3.4. Properties of the Cu/MCNTs composite pellets
3.4.1. Density and hardness measurement
The relative density and hardness of sintered pure Cum and all the
composites are depicted in Fig. 8(a) & (b), respectively. The relative
density of Cum, unimodal U-Cu/CNT, unimodal U-Cu/CCNT, bimodal B-
70Cu/CCNT, and bimodal B-30Cu/CCNT are 98.2%, 92.2%, 93.2%,
95.1% and 96%, respectively. It can be noticed that the relative density
of the bimodal Cu/MCNTs composite pellets is better than that of the
unimodal U-Cu/CNT composite pellet. This is mainly attributed to the
homogenous dispersion of Cu-coated CNT in bimodal composites. In U-
Cu/CNT, few FMCNTs are agglomerated, and few are uniformly
embedded into the Cu matrix during milling (Fig. 2e). The hardness of
the pure Cum, unimodal U-Cu/CNT, unimodal U-Cu/CCNT, bimodal B-
70Cu/CCNT, and bimodal B-30Cu/CCNT composites pellets is obtained
as 43 ± 2.3, 69.3 ± 2.1, 124 ± 3.2, 87.7 ± 1.2 and 113 ± 4 HV,
respectively (Fig. 8b). The high hardness of functionalized CNT as
reinforcement results in a higher hardness value of the unimodal U-Cu/
CNT composite than that of the Cu pellet. The presence of pores is
responsible for deteriorating the hardness of the unimodal U-Cu/CNT
composite, which is 1.6 times lower than that of the bimodal B-30Cu/
CCNT composite. Additionally, the hardness value of the bimodal B-
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 3. TEM micrographs of (a) Cu-CNT composites fabricated via WCR method, (b) showing the uniform Cu coating on MCNT, (c) lattice fringes and d spacing value
of the Cu coated MCNT represented by a circular box in (b), (d) thickness of CNT and Cu-coated CNT, (e) SAD pattern of Cu particles which exist in Cu-CNT
composites powder as marked in square box in (a), and (f) energy dispersive spectrum of Cu coated MCNT marked in (b).

Fig. 4. SEM micrographs of (a) Cu pellet, (b) unimodal U-Cu/CNT composite, (c) bimodal B-70Cu/CCNT composite, (d) bimodal B-30Cu/CCNT composite, and (e &
f) showing the magnified images of the square box in Figures (c) and (d), respectively.

5
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 5. Inverse pole figure (IPF), Image quality (IQ), and corresponding misorientation angle distribution maps of (a-c) Cu pellet, (d-f) unimodal Cu/MCNTs
composite (U-Cu/CNT), (g-i) bimodal B-70Cu/CCNT composite and (j-l) bimodal B-30Cu/CCNT composite, respectively.

6
S. Mishra et al.
Fig. 6. Grain orientation spread (GOS) distribution graphs of (a) Cu pellet, (b) U-Cu/CNT, (c) B-70Cu/CCNT, (d) B-30Cu/CCNT samples before the compression test.
30Cu/CCNT composite is 1.3 times higher than that of the bimodal B-
70Cu/CCNT composite. Additionally, the hardness value of the unim­
odal U-Cu/CCNT composite is the highest among all the composites,
despite adding the same wt% of CNT in the Cu matrix. This is primarily
attributed to the higher concentration of fine-grained Cu compared to
bimodal composites.
3.4.2. Nanoindentation test
The elastic modulus and nanohardness of Cum, unimodal U-Cu/CNT,
bimodal B-70Cu/CCNT, and bimodal B-30Cu/CCNT composites pellets
are estimated from the nanoindentation test, as shown in Fig. 8(c & d).
The elastic modulus of unimodal U-Cu/CNT is 94.3 ± 7.6, which is also
1.5 times higher than that of Cum pellet, but lower than unimodal U-Cu/
CCNT and bimodal composites. The elastic modulus of unimodal U-Cu/
CCNT is 117.3 ± 1.1, which is also 1.8 times higher than that of Cum
pellet. In addition, the elastic modulus of B-70Cu/CCNT and B-30Cu/
CCNT composites is 111.9 ± 6.9 and 126.7 ± 3.6 GPa, respectively,
which are 1.8 and 2 times higher than that of pure Cu pellet (63.7
± 5.4 GPa).
A similar trend is observed in the hardness of the composites,
calculated from Vickers and nanoindentation tests, as presented in Fig. 8
(b & d). The nanohardness of U-Cu/CNT, U-Cu/CCNT, B-70Cu/CCNT,
and B-30Cu/CCNT composites are 1.1 ± 0.19, 1.9 ± 0.04, 1.3 ± 0.12
and 2.1 ± 0.06 GPa, respectively, which are 1.3, 2.3, 1.5 and 2.5 times
higher than that of pure Cu pellet (0.84 ± 0.13 GPa). In addition, the
nanohardness of U-Cu/CNT is 1.2 and 1.9 times lower than that of B-
70Cu/CCNT and B-30Cu/CCNT composites, respectively. It is also
noticed that the nanohardness of the bimodal B-30Cu/CCNT composite
is 1.1 times higher than that of the unimodal U-Cu/CCNT. It may happen
due to the homogenous dispersion of Cu-coated CNT and coarse Cu in
7
Journal of Alloys and Compounds 971 (2024) 172681
the fine Cu region of bimodal composites. Based on the Hall-Petch
relationship, the decrease in the average grain size of the unimodal
and bimodal Cu/MCNTs composite pellets compared to pure Cu pellets
yields improvement in the hardness of composites. The presence of fine
Cu particles (fabricated by WCR) also reduces the average grain size in
bimodal composites.
3.4.3. Compression test
The mechanical properties including the yield strength (YS), uniaxial
compressive strength (UCS), and compressive strain of unimodal and
bimodal Cu/MCNTs composite pellets with 1 wt% CNT, are shown in
Fig. 9b, which are obtained from stress-strain curves, as shown in
Fig. 9a. The yield strength of Cum, U-Cu/CNT, U-Cu/CCNT, bimodal B-
70Cu/CCNT and B-30Cu/CCNT is 44 ± 11, 87 ± 13, 143 ± 15, 118
± 15, 134 ± 18 MPa, respectively. The compressive strength of Cum, U-
Cu/CNT, U-Cu/CCNT, bimodal B-70Cu/CCNT and B-30Cu/CCNT is 155
± 16, 243 ± 11, 282 ± 17, 270 ± 16, 323 ± 12 MPa, respectively. The
U-Cu/CNT composite has a yield strength of 87 MPa and a compressive
strength of 243 MPa, which is 2.0 and 1.6 times higher than that of pure
Cu pellets. The yield and compressive strength of the U-Cu/CNT com­
posite is higher than that of Cu pellets due to the presence of function­
alized CNTs in the Cu matrix but lower than that of U-Cu/CCNT and
bimodal Cu/MCNTs composites. The yield strength of the U-Cu/CNT
composite is 1.4 and 1.5 times lower than that of the bimodal B-70Cu/
CCNT and B-30Cu/CCNT composites, respectively. Additionally, the
compressive strength of the U-Cu/CNT composite is 1.1 and 1.3 times
lower compared to that of the bimodal B-70Cu/CCNT and B-30Cu/CCNT
composites, respectively. Further, it can be observed that the yield
strength and compressive strength of the bimodal B-70Cu/CCNT com­
posites increase from 44 to118 MPa and 154–270 MPa, respectively,
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 7. (a) XRD pattern of Cu powder as a function of milling time, (b & c) XRD pattern and Raman spectra for FMCNT, unimodal and bimodal Cu/MCNTs composite
pellets, respectively, (d) XPS spectra of unimodal (U-Cu/CNT) composite pellet, (e & f) Survey scan showing the binding energy for core-level spectra for Cu and C
with their peak deconvolutions.

which is 2.7 and 1.8 times higher than that of pure Cu pellet. Addi­
tionally, the yield strength and compressive strength of B-30Cu/CCNT
composite are found to be 134 and 323 MPa, respectively, which is 3.0
and 2.1 times higher than that of pure Cu pellet. The yield strength of the
bimodal B-30Cu/CCNT composite is almost similar to the U-Cu/CCNT,
but the compressive strength of the bimodal B-30Cu/CCNT composite is
1.1 times higher than that of unimodal U-Cu/CCNT. This is mainly
because FG around the MCNT and Cu coating on MCNTs surface
strengthen the bimodal Cu/MCNTs composites.
The bimodal B-70Cu/CCNT exhibits superb deformability with the
highest compressive strain of 21.3% among all the composites. The
compressive strain of bimodal B-30Cu/CCNT and B-30Cu/CCNT
composites is 2.7 and 2.6 times higher than that of unimodal U-Cu/CNT
and unimodal U-Cu/CCNT, respectively. Although the compressive
strain of the B-30Cu/CCNT composite is 1.1 times lower, its strength is
1.2 times higher than that of the B-70Cu/CCNT composite. The existence
of bimodal-grain provides significantly higher compressive strength
along with better compressive strain as compared to those of unimodal
composite. The presence of coarse grains accommodates more disloca­
tions with a good dislocation storage capacity, leading to a high
deformability. In addition, bimodal B-70Cu/CCNT and bimodal B-30Cu/
CCNT composites possess high GOS values which indicate these can
accommodate more strain (Fig. 6c & d).
Fig. 10(a-d) depicts the grain orientation spread (GOS) maps of each

8
S. Mishra et al.
Fig. 8. (a & b) Relative density and hardness, and (c & d) elastic modulus and hardness obtained from nanoindentation test, for Cum and Cu composite pellets.
specimen after compression testing. The GOS values of deformed grains
(sum of fractions ˃ 1ᵒ) are 0.82, 0.63, 0.84 and 0.81 for pure Cu, U-Cu/
CNT, B-70Cu/CCNT and B-30Cu/CCNT samples, respectively. The GOS
values of the bimodal Cu/MCNTs composites are greater than those of
the unimodal Cu/MCNTs composite. It indicates a higher amount of
strain is accommodated when the samples are subjected to compression
test which results in higher ductility in bimodal Cu/MCNTs composites
(Fig. 9b). In addition, the higher GOS value (0.84) in B-70Cu/CCNT
sample indicates its higher compressive strain compared to other
samples.
3.5. Conductivity of composite pellets
The thermal diffusivity and conductivity of pure Cu pellets, and
unimodal and bimodal Cu/MCNTs composite pellets are measured at
room temperature, which is presented in Fig. 11a. The thermal diffu­
sivity and thermal conductivity for the pure Cu pellets are 96.3 mm2/s
and 359.2 ± 8 W/mK, respectively. The thermal diffusivity for the
unimodal U-Cu/CNT, U-Cu/CCNT, bimodal B-70Cu/CCNT and B-30Cu/
CCNT composites are 39.8, 65, 80.1 and 79.8 mm2/s, respectively.
Further, the thermal conductivity of unimodal, U-Cu/CNT, U-Cu/CCNT,
bimodal B-70Cu/CCNT and B-30Cu/CCNT composite pellets with 1 wt%
CNT is 186.4 ± 1, 287.4 ± 1, 326.5 ± 6 and 295.4 ± 5 W/mK, respec­
tively. The thermal conductivity of all the composites is lower than the
pure Cu pellet. The primary reason for the lower conductivity of the
composite samples is the randomly oriented MCNTs in the Cu matrix.
The MCNTs exhibit low thermal conductivity along the radial direction
and high thermal conductivity along the tube’s axial direction [30].
Hence, the randomly oriented MCNTs disrupt the unidirectional heat
transfer mechanism in the Cu matrix, which reduces the conductivity of
composites. However, a drastic difference in thermal conductivity is
observed between unimodal U-Cu/CNT and bimodal composites despite
9
Journal of Alloys and Compounds 971 (2024) 172681
having the same amount of reinforcement (1 wt%). A significant
reduction in the thermal conductivity of the unimodal U-Cu/CNT
composite can be directly attributed to the presence of pores that arise
from the cluster of functionalized MCNTs, as seen in the SEM image and
IPF map of the unimodal composite (Figs. 4b & 5 d). It can be noticed
that the thermal conductivity of unimodal U-Cu/CCNT, bimodal
B-70Cu/CCNT and B-30Cu/CCNT composites, is higher than that of
unimodal U-Cu/CNT composite due to the incorporation of Cu-coated
CNTs. Additionally, the thermal conductivity of bimodal
B-70Cu/CCNT composite is the highest among all the composites. In the
case of B-30Cu/CCNT, the proportion of fine to coarse Cu grains (70:30)
is higher as compared to B-70Cu/CCNT (30:70), which increases the
phonons scattering by increasing the dislocation density [31], as seen in
Fig. 12(e & f). This increment in the dislocation density in the
B-30Cu/CCNT (3.8×1014/m2) is 0.7 times more than that of the
B-70Cu/CCNT (2.8×1014/mm2) sample. Thereby, the thermal conduc­
tivity of the B-30Cu/CCNT composite is 9.5% lower than that of the
B-70Cu/CCNT composite.
The electrical conductivity for Cu and composite pellets is also
depicted in same Fig. 11a. The electrical conductivity for the unimodal
U-Cu/CNT, and bimodal B-70Cu/CCNT and B-30Cu/CCNT composites
are obtained to be 94.2 ± 0.2, 63.9 ± 0.1, 90.7 ± 0.1 and 85.4 ± 0.6%
IACS, respectively. A significant reduction in the electrical conductivity
(63.9% IACS) of unimodal U-Cu/CNT is observed due to the presence of
porosity and the 26% reduction in the grain size of Cu. Additionally, the
Cu and CNTs have a huge difference in the coefficient of thermal
expansion, which causes a high density of interface misfit dislocations
that lowers the electrical conductivity. It is also observed that the
electrical conductivity of the B-30Cu/CCNT composite is lower than that
of the bimodal B-70Cu/CCNT composite. The proportion of FG is higher
in the B-30Cu/CCNT composite, which increases the grain boundary
volume. This results in the increase of obstacles to the conductive path,
S. Mishra et al.
Fig. 9. (a) Compressive stress and compressive strain curves of Cu, and unimodal and bimodal Cu/MCNTs composites, (b) corresponding obtained results from the
compressive stress-strain curves (c) preparation of the samples for compression testing.
which lowers the electrical conductivity of the composite.
4. Discussion
4.1. Strengthening mechanism
In the present study, the contribution of CNT in unimodal (CG) and
bimodal (CG + FG) to the strengthening can be summarized based on the
underlying strengthening mechanisms, including grain boundary
strengthening [36], Orowan strengthening [36], load-transfer
strengthening [37], and dislocation strengthening [6]. The compres­
sive strength (σC) for Cu/MCNTs composites can be expressed in the
mathematical term as Eq. (1):
σc = σm + ΔσGR +ΔσLT + ΔσD +ΔσOR (1)
where ΔσGR, ΔσLT, ΔσD, and ΔσOR, are the grain boundary, load transfer,
dislocation, and Orowan strengthening, respectively. In the present
study, the compressive strength (σm) of the coarse Cu matrix is 155 MPa.
4.1.1. Grain boundary strengthening (Δσ GR,)
The strengthening effect of grain boundary in all the composites
could be ascribed to the synergistic effects of grain size and distribution
of Cu-coated CNTs. Even though the contribution in strengthening by
grain boundary is the lowest among other strengthening mechanisms,
the distribution of CG and FG plays an important role in enhancing the
strength and compressive strain in the present study. The strengthening
effect of grain boundary can be evaluated from the Hall-Petch equation
as (Eq. (2)) [35]:
( )
Δσ GR = K dC − 1/2 − dCum − 1/2 (2)
10
Journal of Alloys and Compounds 971 (2024) 172681
where dCum and dc are the average grain sizes for Cum pellet and pre­
pared Cu/MCNTs composites (unimodal and bimodal). K is the Hall-
Petch coefficient for Cum (0.07 MPa.m1/2) [35]. The EBSD phase maps
of Cum and Cu/MCNTs composites at different scales are shown in
Fig. 12, and their corresponding grain size partition histogram (Fig. 12e)
is obtained by counting the grains in the images. It can be observed from
the histogram map that a higher fraction of FG of Cu is present in the
B-30Cu/CCNT composite as compared to other composites, which is
quite obvious.
Also, the average grain size of pure Cum and Cu/MCNTs composites
is estimated from three EBSD images and used to calculate the contri­
bution of grain boundary to overall strengthening. The average grain
size for the Cum, unimodal U-Cu/CNT, unimodal U-Cu/CCNT, bimodal
B-70Cu/CCNT and B-30Cu/CCNT composites are obtained to be 4.7
± 3.2, 3.5 ± 2.1, 1.1 ± 3, 2.8 ± 2.3 and 1.6 ± 1.2 µm, respectively. A
significant reduction (66%) in the grain size of B-30Cu/CCNT composite
is observed due to the presence of a higher fraction of fine Cu grains
(0.73), as can be seen in Fig. 12e. The presence of fine grains offers more
grain boundaries, which eventually enhances the strength based on the
Hall-Petch relationship.
4.1.2. Load transfer strengthening (ΔσLT)
The load transfer strengthening mechanism plays a crucial role in
accurately predicting the degree of strength exhibited by the interface
bonding between the matrix and reinforcement. The load transfer
strengthening can be estimated using the modified shear-lag model
established by Kelly and Tyson [38]:
(
lc
)
( )
ΔσLT = σf V f 1− + σm 1 − V f (l> lc ) (3)
2l
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 10. Grain orientation spread (GOS) distribution graphs of (a) Cu pellet, (b) unimodal Cu/MCNTs composite (U-Cu/CNT), (c) bimodal Cu/MCNTs composite (B-
70Cu/CCNT) and (d) bimodal Cu/MCNTs composite (B-30Cu/CCNT) samples after the compression test.

Fig. 11. (a) Thermal conductivity & thermal diffusivity, and electrical conductivity in % IACS (International Annealed Copper Standard) for unimodal and bimodal
Cu/MCNTs composites. (b) comparison of thermal and electrical conductivity of bimodal Cu/MCNTs composites in this study and CNTs/Cu composites as reported in
the literature [6,32–35].

(
σ CNT
) diameter (12 nm), and the average length (5.5 µm) of CNTs [39]. The
lc = (dCNT ) (4) σCNT and τy are the compressive strength of CNTs (416 MPa) [40] and
2τ y
the interfacial shear stress (half of the yield strength of the matrix),
Where lc, dCNT, and l are the critical length (0.08 µm), the average respectively.

11
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 12. (a-d) Grain size distribution, (e) partitioning of grain size distribution, (f) grain size and dislocation density for Cu pellet and Cu/MCNTs composites.

Theoretically, the contribution of load transfer strengthening of all
the composites (unimodal U-Cu/CNT, bimodal B-70Cu/CCNT and B-
30Cu/CCNT) is 61.7 MPa. This is because the same volume fraction of
CNT is being used in all of the composites. Practically, it is not feasible
because the different fabrication methods are followed. In the case of
unimodal U-Cu/CCNT, the contribution of load transfer strengthening
value of unimodal U-Cu/CCNT is 65.3 MPa, which is different from the
value obtained from unimodal U-Cu/CNT and bimodal composites The
disparity in theoretical values can be attributed to the difference in yield
strength values between the fine Cu (141 MPa) [39] and the coarse Cu
matrix (155 MPa). However, the load transfer strengthening depends on
the interfacial bonding between the Cu matrix and CNT. The

Fig. 13. (a) The experimental and theoretical compressive strength of Cu/MCNTs composites, showing the strength contribution by various strengthening mech­
anisms, also (b) the strengthening efficiency of unimodal and bimodal Cu/MCNTs composites.

12
S. Mishra et al.
strengthening efficiency R for all composites is estimated using Eq. (5).
σyc − σym
R= (5)
vCNT σym
The strengthening efficiency (R values) of all the composites is
calculated using the yield strength of the composite (σyc) and Cu pellet
(σym), obtained from the stress-strain curve, as presented in Fig. 13b.
The R-values of the unimodal U-Cu/CNT, U-Cu/CCNT, and bimodal B-
70Cu/CCNT and B-30Cu/CCNT composites are 20.4, 44.0, 35 and 42.2,
respectively. In actuality, the contribution of load transfer strengthening
in all the composites is determined using the following Eq. [41]:
R compressive strength of bimodal B-30Cu/CCNT composites are almost
ΔσLT = × vf × σ ym (6)
2
The strengthening efficiency R can be written as S/2, where S is the
aspect ratio of CNT reinforcement [42]. The actual contribution of load
transfer strengthening for bimodal B-70Cu/CCNT and B-30Cu/CCNT
composites are 37 and 47.2 MPa, respectively. In the case of unimodal
U-Cu/CNT, the contribution of load transfer strengthening is lower
(21.5 MPa) than the theoretical value (61.7 MPa). It indicates that the
load transfer from FMCNT to Cu is poor. The load transfer strengthening
in bimodal composites is higher than that in unimodal U-Cu/CNT due to
the better interfacial bonding between Cu and MCNTs. However, the
actual contribution of load transfer strengthening of the bimodal
B-30Cu/CCNT and unimodal U-Cu/CCNT composites is 51.1 MPa and
52.4 MPa, respectively, which is nearly similar.
4.1.3. Dislocation strengthening(Δσ D)
The contribution of dislocation strengthening (ΔσD) in unimodal and
bimodal CNTs/Cu composites is determined using the formula as pro­
posed by Arsenault et al. [43]:
(√̅̅̅̅̅ √̅̅̅̅̅̅̅̅̅̅ )
Δσ D = αGb ρc − ρm (7)
The geometric constant (α) and shear modulus (G) of Cu are 1.25 and
42.1 GPa, respectively. The ρc and ρm are the dislocation density of
composites and Cu matrix, respectively. Further, the dislocation density
(ρ) for composites is estimated using Eq. (8).
√̅̅̅
2 3ε
ρ= (8)
db
where d, b, and ε are the crystallite size, Burgers vector, and lattice
strain, respectively. The d and ε are estimated from the XRD pattern of
composite pellets using the Williamson-Hall equation [44] (Eq. (9)).
Kλ
βhkl .cosθ = + 4ε.sinθ (9)
d (Fig. 5e and f) in U-Cu/CNT composite can act as barriers to dislocation
Where λ, K, d, ε and βhkl are X-ray wavelength (0.154 nm), constant,
crystallite size, lattice strain and full-width half maxima (FWHM),
respectively. The contribution of ΔσD in bimodal B-30Cu/CCNT is
51.7 MPa which is 1.7 times higher than that of the unimodal composite
(30.3 MPa) due to the presence of a higher proportion of FG (Fig. 13).
4.1.4. Orowan strengthening(ΔσOR)
Apart from that, the improvement in the strength of unimodal and
bimodal composites resulted from dislocation restraining by CNTs or
dislocation bypassing during plastic deformation. It can be expressed
through the Orowan strengthening mechanism (Orowan-Ashby equa­
tion) [45,46] as follows:
( )
0.4MGb
ln dCNT
b difference between the CG and FG region is low (Fig. 12d), Thereby, the
ΔσOR = √̅̅̅̅̅̅̅̅̅̅̅ . (10)
π( 1 − γ ) (λ − dCNT )
√̅̅̅̅̅̅̅̅̅̅̅̅̅
1 3π
λ = dCNT (11)
2 2V CNT
13
Journal of Alloys and Compounds 971 (2024) 172681
The VCNT and dCNT are CNT’s volume fraction and diameter (12 nm),
λ is interparticle spacing for CNTs, and M is the Taylor factor for Cu
(3.06). The contribution of Orowan strengthening in all the composites
is found to be the same (44.5 MPa) due to the similar volume fraction of
CNTs.
Based on the overall discussion, the ΔσLT accounts for the main
strengthening contribution in bimodal Cu/MCNTs (B-70Cu/CCNT and
B-30Cu/CCNT) composites. Moreover, Fig. 13a displays the theoreti­
cally calculated strength contribution from various mechanisms and
experimentally determined compressive strength values. It may be noted
that the experimentally measured and theoretically calculated values of
similar. Further, the actual contribution of different strengthening
mechanisms in experimentally obtained compressive strength for the
composites is described in Table 2.
4.2. Interface microstructure analysis
The TEM and Inverse FFT (IFFT) images for unimodal U-Cu/CNT and
bimodal B-30Cu/CCNT composites are shown in Fig. 14, to investigate
the interface region between the Cu and MCNT (FCNT or CCNTs). The
accumulation of dislocations at the grain boundary junction can be
observed in the micrograph of the unimodal composite (Fig. 14a). The
presence of FCNT along the grain boundary (Fig. 14c) implies difficulty
in dislocation motion and pile-up, which is also an indirect indication of
dislocation formation. To further verify this, the inverse FFT (Fig. 14e)
of the marked region in (c) is examined. The measured interplanar
spacing value is 0.33 nm which corresponds to the CNTs (0002), as
shown in Fig. 14f. Also, existence of dislocations at FCNT can be seen in
inverse FFT image (Fig. 14e). Further, a combination of fine and coarse
Cu grains (marked by yellow) is observed in the TEM micrograph of
bimodal B-30Cu/CCNT composites (Fig. 14b). The high magnification
image of the composite shows a Cu-coated CNT (Fig. 14d). The IFFT of
this micrograph also confirms the presence of Cu and CNT, as seen in
Fig. 14 (h and k). The measured interplanar spacing values are 0.21 and
0.34 nm which correspond to the Cu (111) and CNTs (0002), as shown
in Fig. 14(i and j).
The SEM images of cracks present on the curved surface of unimodal
and bimodal deformed composite samples are investigated to under­
stand the effect of grain size of Cu (CG and FG) and MCNTs on the
fracture behavior during compressive loading. A large number of dim­
ples are visible on the fracture surface of the Cu pellet, showing that the
failure mode is completely ductile, as depicted in Fig. 15a. Further, the
presence of a significant proportion of HAGBs and twin boundaries
motion, thereby forcing dislocations to accumulate and become entan­
gled at these grain boundaries. During the deformation, the piled-up
dislocations will further develop into micro-crack, as shown in
Fig. 15b. The bimodal B-70Cu/CCNT composite exhibits an inhomoge­
neous distribution of FG and CG, as seen in Fig. 5g and h (EBSD phase
maps). During deformation, high-stress concentrations occur at the
boundaries between the CG and FG due to the significant deformation
mismatch. The dislocations within the CG region facilitate slipping and
tend to move towards forming slip bands (Fig. 15c). The lower strength
of the CG region compared to the FG region, and significant accumu­
lation of deformation in the slip bands would reduce the strength of
composite and increase the ductility and toughness [47].
Further, it is noticed that the CG regions are uniformly distributed in
the FG region of the bimodal B-30Cu/CCNT composite. The grain size
stress concentration at the boundaries between the CG and FG in the B-
30Cu/CCNT composite is low. Furthermore, the strain localization in the
FG region is decreased due to the uniformly distributed CG in the FG
region [48]. As a result, the B-30Cu/CCNT composite has achieved a
higher strength than the B-70Cu/CCNT composite.
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Table 2
The contribution of different strengthening factors in experimentally determined compressive strength of unimodal and bimodal composites.
Experimentally determined compressive strength Contribution of different strengthening factors in experimentally determined compressive strength

Samples Δσc (MPa) Δσm(MPa) ΔσGR(MPa) ΔσD(MPa) ΔσOR(MPa) ΔσLT(MPa)

U-Cu/CNT 243 155 5.1 30.3 31.6 21.5
U-Cu/CCNT 282 141 26.3 34.4 27.9 52.3
B-70Cu/CCNT 270 155 9.5 29.2 39.3 39.5
B-30Cu/CCNT 323 155 26.8 51.7 42.2 51.1

Fig. 14. TEM image of sintered (a) unimodal U-Cu/CNT, (b) bimodal B-Cu/CCNT, (c) CNT present in unimodal composite and (d) Cu-coated CNT present in bimodal
composite, (e & f) IIFT image and corresponding d spacing value of CNTs (0002), (h & k) IIFT image of marked area as 1 and 2 in Cu-coated CNT image (g) and
corresponding d spacing value of (i & j) Cu (111) and CNT (0002), respectively. Figure (l) SAD pattern of sintered unimodal U-Cu/CNT composite and (m) The
schematic of bimodal B-Cu/CCNT composites.

5. Conclusions significantly enhance the strengthening efficiency and promote

In the present work, bimodal Cu composites consisting of coarse
milled Cu and fine Cu (synthesized by WCR), in two different ratios
(70:30 and 30:70), with 1 wt% of Cu-coated CNT, were successfully
synthesized. Similarly, a unimodal Cu composite consisting of milled Cu
and functionalized CNT (1 wt%) was also synthesized using the same
powder metallurgy method. The following outcomes from this study are
drawn:
• Compared to unimodal U-Cu/CNT and U-Cu/CCNT composite, the
bimodal B-30Cu/CCNT composite shows enhanced compressive
strength, as well as improved thermal and electrical properties.
• The B-30Cu/CCNT composite exhibits the most favorable combina­
tion of properties, including a hardness of 2.1 ± 0.06 GPa, a strength
of 323 MPa, a compressive strain of 18.8% with balance electrical
(85.4 ± 0.6% IACS), and thermal conductivity (295.4 ± 6 W/mK).
• The strengthening efficiency of B-70Cu/CCNT and B-30Cu/CCNT
composites is determined to be 35 and 42.2, respectively, which are
1.7 and 2.1 times higher than that of the U-Cu/CNT composite (20).
In addition, the strengthening efficiency of B-30Cu/CCNT is com­
parable to the U-Cu/CCNT composite. The Cu-coated MCNTs
strong interfacial bonding between Cu and MCNT in bimodal Cu/
MCNTs and U-Cu/CCNT composites.
• The Orowan and dislocation strengthening (ΔσOR and ΔσD, respec­
tively) mechanisms primarily contribute to strengthening in unim­
odal U-Cu/CNT composites. Conversely, load transfer is the
dominant strengthening mechanism in bimodal Cu/MCNTs and U-
Cu/CCNT composites.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
CRediT authorship contribution statement
Shakti Mishra: Investigation, Methodology, Writing – original draft,
Writing – review & editing. Sudipta Mohapatra: Formal analysis,
Writing – review & editing. Bhaskar Santu Mudliyar: Data acquisition.
Siddhartha Das: Conceptualization, Supervision, Writing – review &
editing. Karabi Das: Conceptualization, Supervision, Writing – review
& editing.

14
S. Mishra et al. Journal of Alloys and Compounds 971 (2024) 172681

Fig. 15. SEM fractographs of edge-cracked (a) pure Cu, (b) unimodal U-Cu/CNT, (c) bimodal B-70Cu/CCNT and (d) B-30Cu/CCNT composites, and the repre­
sentation of the fracture model for (e) unimodal U-Cu/CNT and (f) bimodal Cu/MCNTs composites.

Declaration of Competing Interest References

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.
Data availability
The raw data used in this work can be obtained from the corre­
sponding author on request.
Acknowledgment
Shakti Mishra would like to thank MHRD, Govt. of India for a Ph.D.
research fellowship. The authors would also like to thank IIT Kharagpur
for supporting this research work.
[1] T. Yang, W. Chen, H. Zhang, L. Ma, Y.-Q. Fu, In-situ generated graphene from
wheat flour for enhancing mechanical and electrical properties of copper matrix
composites, Mater. Sci. Eng.: A 835 (2022), 142662, https://doi.org/10.1016/j.
msea.2022.142662.
[2] E. Khaleghi, M. Torikachvili, M.A. Meyers, E.A. Olevsky, Magnetic enhancement of
thermal conductivity in copper–carbon nanotube composites produced by
electroless plating, freeze drying, and spark plasma sintering, Mater. Lett. 79
(2012) 256–258, https://doi.org/10.1016/j.matlet.2012.03.117.
[3] R. Jiang, X. Zhou, Z. Liu, Electroless Ni-plated graphene for tensile strength
enhancement of copper, Mater. Sci. Eng.: A 679 (2017) 323–328, https://doi.org/
10.1016/j.msea.2016.10.029.
[4] J.L. Song, W.G. Chen, L.L. Dong, J.J. Wang, N. Deng, An electroless plating and
planetary ball milling process for mechanical properties enhancement of bulk
CNTs/Cu composites, J. Alloy. Compd. 720 (2017) 54–62, https://doi.org/
10.1016/j.jallcom.2017.05.208.
[5] S. Mohammed, D.L. Chen, Z.Y. Liu, D.R. Ni, Q.Z. Wang, B.L. Xiao, Z.Y. Ma,
Deformation behavior and strengthening mechanisms in a CNT-reinforced
bimodal-grained aluminum matrix nanocomposite, Mater. Sci. Eng.: A 817 (2021),
141370, https://doi.org/10.1016/j.msea.2021.141370.

15
S. Mishra et al.
[6] S. Fu, X. Chen, P. Liu, Preparation of CNTs/Cu composites with good electrical
conductivity and excellent mechanical properties, Mater. Sci. Eng.: A 771 (2020),
138656, https://doi.org/10.1016/j.msea.2019.138656.
[7] K. Ma, Z.Y. Liu, B.S. Liu, B.L. Xiao, Z.Y. Ma, Improving ductility of bimodal carbon
nanotube/2009Al composites by optimizing coarse grain microstructure via hot
extrusion, Compos. Part A: Appl. Sci. Manuf. 140 (2021), 106198, https://doi.org/
10.1016/j.compositesa.2020.106198.
[8] C. Schilde, M. Schlömann, A. Overbeck, S. Linke, A. Kwade, Thermal, mechanical
and electrical properties of highly loaded CNT-epoxy composites – A model for the
electric conductivity, Compos. Sci. Technol. 117 (2015) 183–190, https://doi.org/
10.1016/j.compscitech.2015.06.013.
[9] A.K. Shukla, N. Nayan, S.V.S.N. Murty, S.C. Sharma, P. Chandran, S.R. Bakshi, K.
M. George, Processing of copper-carbon nanotube composites by vacuum hot
pressing technique, Mater. Sci. Eng. A 560 (2013) 365–371, https://doi.org/
10.1016/j.msea.2012.09.080.
[10] J.M. Tao, X.F. Chen, P. Hong, J.H. Yi, Microstructure and electrical conductivity of
laminated Cu/CNT/Cu composites prepared by electrodeposition, J. Alloy. Compd.
717 (2017) 232–239, https://doi.org/10.1016/j.jallcom.2017.05.074.
[11] N. Nayan, A.K. Shukla, P. Chandran, S.R. Bakshi, S.V.S.N. Murty, B. Pant, P.
V. Venkitakrishnan, Processing and characterization of spark plasma sintered
copper/carbon nanotube composites, Mater. Sci. Eng.: A 682 (2017) 229–237,
https://doi.org/10.1016/j.msea.2016.10.114.
[12] L. Zhang, R. Bao, J. Yi, S. Guo, J. Tao, C. Li, D. Fang, Y. Liu, F. Li, Improving
comprehensive performance of copper matrix composite by spray pyrolysis
fabricated CNT/W reinforcement, J. Alloy. Compd. 833 (2020), 154940, https://
doi.org/10.1016/j.jallcom.2020.154940.
[13] S.C. Tjong, Recent progress in the development and properties of novel metal
matrix nanocomposites reinforced with carbon nanotubes and graphene
nanosheets, Mater. Sci. Eng.: R: Rep. 74 (2013) 281–350, https://doi.org/10.1016/
j.mser.2013.08.001.
[14] Z.Y. Liu, B.L. Xiao, W.G. Wang, Z.Y. Ma, Modelling of carbon nanotube dispersion
and strengthening mechanisms in Al matrix composites prepared by high energy
ball milling-powder metallurgy method, Compos. Part A: Appl. Sci. Manuf. 94
(2017) 189–198, https://doi.org/10.1016/j.compositesa.2016.11.029.
[15] S. Dong, J. Zhou, D. Hui, Y. Wang, S. Zhang, Size dependent strengthening
mechanisms in carbon nanotube reinforced metal matrix composites, Compos. Part
A: Appl. Sci. Manuf. 68 (2015) 356–364, https://doi.org/10.1016/j.
compositesa.2014.10.018.
[16] E. Ma, Eight routes to improve the tensile ductility of bulk nanostructured metals
and alloys, JOM 58 (2006) 49–53, https://doi.org/10.1007/s11837-006-0215-5.
[17] D. Mesguich, C. Arnaud, F. Lecouturier, N. Ferreira, G. Chevallier, C. Estournès,
A. Weibel, C. Josse, C. Laurent, High strength-high conductivity carbon nanotube-
copper wires with bimodal grain size distribution by spark plasma sintering and
wire-drawing, Scr. Mater. 137 (2017) 78–82, https://doi.org/10.1016/j.
scriptamat.2017.05.008.
[18] L. Meng, X. Wang, X. Hu, H. Shi, K. Wu, Role of structural parameters on strength-
ductility combination of laminated carbon nanotubes/copper composites, Compos.
Part A: Appl. Sci. Manuf. 116 (2019) 138–146, https://doi.org/10.1016/j.
compositesa.2018.10.021.
[19] H. Zhang, H. Wang, J. Wang, J. Rong, M. Zha, C. Wang, P. Ma, Q. Jiang, The
synergy effect of fine and coarse grains on enhanced ductility of bimodal-
structured Mg alloys, J. Alloy. Compd. 780 (2019) 312–317, https://doi.org/
10.1016/j.jallcom.2018.11.229.
[20] M. Zha, Y. Li, R.H. Mathiesen, R. Bjørge, H.J. Roven, Microstructure evolution and
mechanical behavior of a binary Al–7Mg alloy processed by equal-channel angular
pressing, Acta Mater. 84 (2015) 42–54, https://doi.org/10.1016/j.
actamat.2014.10.025.
[21] Q. Shi, C. Wang, K. Deng, Y. Fan, K. Nie, Wei-Liang, Work hardening and softening
behaviors of Mg-Zn-Gd-Ca alloy regulated by bimodal microstructure, J. Alloy.
Compd. 938 (2023), 168606, https://doi.org/10.1016/j.jallcom.2022.168606.
[22] S. Mishra, S. Jena, S. Das, K. Das, Multiwalled carbon nanotubes coated with twin-
strengthened copper, Mater. Chem. Phys. 280 (2022), 125799, https://doi.org/
10.1016/j.matchemphys.2022.125799.
[23] Z.Y. Liu, S.J. Xu, B.L. Xiao, P. Xue, W.G. Wang, Z.Y. Ma, Effect of ball-milling time
on mechanical properties of carbon nanotubes reinforced aluminum matrix
composites, Compos. Part A: Appl. Sci. Manuf. 43 (2012) 2161–2168, https://doi.
org/10.1016/j.compositesa.2012.07.026.
[24] J.Y. Huang, Y.K. Wu, H.Q. Ye, Deformation structures in ball milled copper, Acta
Mater. 44 (1996) 1211–1221, https://doi.org/10.1016/1359-6454(95)00231-6.
[25] S. Mohapatra, G. Poojari, L. Marandi, S. Das, K. Das, A systematic study on
microstructure evolution, mechanical stability and the micro-mechanical response
of tensile deformed medium manganese steel through interrupted tensile test,
Mater. Charact. 195 (2023), 112562, https://doi.org/10.1016/j.
matchar.2022.112562.
[26] G.J.H. Melvin, Q.-Q. Ni, Y. Suzuki, T. Natsuki, Microwave-absorbing properties of
silver nanoparticle/carbon nanotube hybrid nanocomposites, J. Mater. Sci. 49
(2014) 5199–5207, https://doi.org/10.1007/s10853-014-8229-9.
[27] A. Jorio, R. Saito, Raman spectroscopy for carbon nanotube applications, J. Appl.
Phys. 129 (2021), 021102, https://doi.org/10.1063/5.0030809.
16
Journal of Alloys and Compounds 971 (2024) 172681
[28] T.I.T. Okpalugo, P. Papakonstantinou, H. Murphy, J. McLaughlin, N.M.D. Brown,
High resolution XPS characterization of chemical functionalised MWCNTs and
SWCNTs, Carbon 43 (2005) 153–161, https://doi.org/10.1016/j.
carbon.2004.08.033.
[29] B.-J. Kim, J.-P. Kim, J.-S. Park, Effects of Al interlayer coating and thermal
treatment on electron emission characteristics of carbon nanotubes deposited by
electrophoretic method, Nanoscale Res. Lett. 9 (2014) 236, https://doi.org/
10.1186/1556-276X-9-236.
[30] K. Chu, Q. Wu, C. Jia, X. Liang, J. Nie, W. Tian, G. Gai, H. Guo, Fabrication and
effective thermal conductivity of multi-walled carbon nanotubes reinforced Cu
matrix composites for heat sink applications, Compos. Sci. Technol. 70 (2010)
298–304, https://doi.org/10.1016/j.compscitech.2009.10.021.
[31] J. Zou, D. Kotchetkov, A.A. Balandin, D.I. Florescu, F.H. Pollak, Thermal
conductivity of GaN films: Effects of impurities and dislocations, J. Appl. Phys. 92
(2002) 2534–2539, https://doi.org/10.1063/1.1497704.
[32] P.G. Koppad, H.R.A. Ram, C.S. Ramesh, K.T. Kashyap, R.G. Koppad, On thermal
and electrical properties of multiwalled carbon nanotubes/copper matrix
nanocomposites, J. Alloy. Compd. 580 (2013) 527–532, https://doi.org/10.1016/
j.jallcom.2013.06.123.
[33] H. Wang, Z.-H. Zhang, Z.-Y. Hu, Q. Song, S.-P. Yin, Interface structure and
properties of CNTs/Cu composites fabricated by electroless deposition and spark
plasma sintering, Mater. Res. Express 5 (2018), 015602, https://doi.org/10.1088/
2053-1591/aaa167.
[34] C. Guo, Z. Zhan, L. Quan, Study of the preparation and properties of 0.5 vol% Ni-
CNTs/Cu nanocomposites with magnetic alignment, J. Alloy. Compd. 781 (2019)
261–269, https://doi.org/10.1016/j.jallcom.2018.12.028.
[35] X. Wei, J. Tao, Y. Hu, Y. Liu, R. Bao, F. Li, D. Fang, C. Li, J. 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), 141248, https://doi.org/10.1016/j.msea.2021.141248.
[36] W.J. Kim, T.J. Lee, S.H. Han, Multi-layer graphene/copper composites: Preparation
using high-ratio differential speed rolling, microstructure and mechanical
properties, Carbon 69 (2014) 55–65, https://doi.org/10.1016/j.
carbon.2013.11.058.
[37] X. Zhang, C. Shi, E. Liu, F. He, L. Ma, Q. Li, J. Li, N. Zhao, C. He, In-situ space-
confined synthesis of well-dispersed three-dimensional graphene/carbon nanotube
hybrid reinforced copper nanocomposites with balanced strength and ductility,
Compos. Part A: Appl. Sci. Manuf. 103 (2017) 178–187, https://doi.org/10.1016/j.
compositesa.2017.09.010.
[38] B. Chen, S. Li, H. Imai, L. Jia, J. Umeda, M. Takahashi, K. Kondoh, Load transfer
strengthening in carbon nanotubes reinforced metal matrix composites via in-situ
tensile tests, Compos. Sci. Technol. 113 (2015) 1–8, https://doi.org/10.1016/j.
compscitech.2015.03.009.
[39] S. Mishra, S. Jena, S. Mohapatra, S. Das, K. Das, Mechanical behavior and texture
evolution of integrated MCNTs/Cu composites with balanced electrical/thermal
conductivity, Powder Technol. (2023), 118751, https://doi.org/10.1016/j.
powtec.2023.118751.
[40] M. Zu, W. Lu, Q.-W. Li, Y. Zhu, G. Wang, T.-W. Chou, Characterization of carbon
nanotube fiber compressive properties using tensile recoil measurement, ACS Nano
6 (2012) 4288–4297, https://doi.org/10.1021/nn300857d.
[41] S. Xiang, X. Wang, M. Gupta, K. Wu, X. Hu, M. Zheng, Graphene nanoplatelets
induced heterogeneous bimodal structural magnesium matrix composites with
enhanced mechanical properties, Sci. Rep. 6 (2016) 38824, https://doi.org/
10.1038/srep38824.
[42] S.I. Cha, K.T. Kim, S.N. Arshad, C.B. Mo, S.H. Hong, Extraordinary strengthening
effect of carbon nanotubes in metal-matrix nanocomposites processed by
molecular-level mixing, Adv. Mater. 17 (2005) 1377–1381, https://doi.org/
10.1002/adma.200401933.
[43] R.J. Arsenault, N. Shi, Dislocation generation due to differences between the
coefficients of thermal expansion, Mater. Sci. Eng. 81 (1986) 175–187, https://doi.
org/10.1016/0025-5416(86)90261-2.
[44] Á. Révész, T. Ungár, A. Borbély, J. Lendvai, Dislocations and grain size in ball-
milled iron powder, Nanostruct. Mater. 7 (1996) 779–788, https://doi.org/
10.1016/S0965-9773(96)00048-7.
[45] M. Yang, L. Weng, H. Zhu, T. Fan, D. Zhang, Simultaneously enhancing the
strength, ductility and conductivity of copper matrix composites with graphene
nanoribbons, Carbon 118 (2017) 250–260, https://doi.org/10.1016/j.
carbon.2017.03.055.
[46] Z. Zhang, D. Chen, Consideration of Orowan strengthening effect in particulate-
reinforced metal matrix nanocomposites: A model for predicting their yield
strength, Scr. Mater. 54 (2006) 1321–1326, https://doi.org/10.1016/j.
scriptamat.2005.12.017.
[47] K. Ma, X.N. Li, J.F. Zhang, Z.Y. Liu, B.L. Xiao, Z.Y. Ma, Tension-compression fatigue
behaviors of uniform and bimodal carbon nanotube/Al-Cu-Mg composites, Mater.
Charact. 192 (2022), 112193, https://doi.org/10.1016/j.matchar.2022.112193.
[48] K. Ma, X.N. Li, K. Liu, X.G. Chen, Z.Y. Liu, B.L. Xiao, Z.Y. Ma, Improving the high-
cycle fatigue strength of heterogeneous carbon nanotube/Al-Cu-Mg composites
through grain size design in ductile-zones, Compos. Part B: Eng. 222 (2021),
109094, https://doi.org/10.1016/j.compositesb.2021.109094.