Physica E 146 (2023) 115566

Contents lists available at ScienceDirect

Physica E: Low-dimensional Systems and Nanostructures

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

Annealing effect on the thermoelectric properties of multiwall

carbon nanotubes
E.M. Elsehly a, b, **, E.M.M. Ibrahim c, Medhat A. El-Hadek d, A. El-Khouly a, b, V. Khovaylo b,
Z.M. Elqahtani e, N.G. Chechenin f, A.M. Adam c, g, *
a
Faculty of Science, Physics Department, Damanhour University, 22516, Damanhour, Egypt
b
National University of Science and Technology MISIS, Moscow 119049, Russian Federation
c
Physics Department, Faculty of Science, Sohag University, 82524 Sohag, Egypt
d
Department of Production and Mechanical Design, Faculty of Engineering, Port-said University, Egypt
e
Physics Department, Faculty of Science, Princess Nourah Bint Abdulrahman University, Riyadh, Saudi Arabia
f
Skobeltsyn Institute of Nuclear Physics, Lomonosov Moscow State University, 119991, Moscow, Russia
g
Faculty of Engineering, King Salman International University, South Sinai, Egypt

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

Keywords: In this article, it is demonstrated that transport properties and thus thermoelectric performance of multiwall
Carbon nanotubes carbon nanotubes (MWCNTs) can be enhanced significantly by annealing. Two different samples of MWCNTs
Thermoelectric properties with small and large diameters were investigated. The as-prepared MWCNT samples were treated at different
Annealing
conditions of annealing. The pristine and annealed MWCNTs were investigated using x-ray diffraction, Raman
Nanotube diameter
spectroscopy and scanning electron microscope in order to investigate the microstructure, defect content and
morphology before and after the annealing process. The electrical conductivity, Seebeck coefficient and thermal
conductivity were measured over the temperature range of (300–450 K) and were found to correlate with the
tube diameter and the annealing conditions. The electrical conductivity and Seebeck coefficient measurements
show that the conduction mechanism in MWCNTs is thermal activation as the samples have degenerate
semiconductor-like behavior. Upon annealing, the Seebeck coefficient increases, followed by a reduction in the
thermal conductivity. The thermoelectric power factor was enhanced by annealing to realize the highest value of
1.06 μW m− 1 K− 2 at 475 K for the sample with the larger diameter. The thermoelectric figure of merit ZT was
calculated to determine the performance of the materials. ZT of the sample of the larger tube’s diameter was
enhanced by about 247% at 450 K.

1. Introduction [8–10]. Several strategies have been introduced to improve the prop­
erties of bulk CNT materials, such as chemical or metallic doping [11],
Development of everyday applications needs new materials with solvent densification [12], and annealing [13,14]. The main factors
enhanced physical properties such as good electrical conductivity, high contributing to the transport properties of bulk CNTs are the intrinsic
thermoelectric power and lower production cost. Carbon nanotubes properties of the CNT and the contact resistant between adjacent CNTs
(CNTs) are considered as one of the most promising candidates for assembled in the bulk form [15]. Despite using the most precise syn­
achieving these tasks. They exhibit outstanding physical properties on thesis conditions, the CNTs are always produced with many types of
the individual nanotube level, including high thermal [1,2], electrical defects. The in-situ born defects such as the Stone–Wales defects or even
conductivities [3] and tensile strength [4]. However, most real world simple vacancies decreases the mean free path of charge carriers and
applications require three-dimensional bulk CNT materials, such as ar­ phonons due to scattering processes. Besides, CNTs always contain some
rays [5], sheets [6], and yarns [7]. The physical properties, of these kinds of contamination, such as residual catalyst, other forms of carbon,
carbon materials, are significantly low compared to individual CNTs etc., which have significant effects on their transport properties [16].

* Corresponding author. Physics Department, Faculty of Science, Sohag University, 82524 Sohag, Egypt.
** Corresponding author. Faculty of Science, Physics department, Damanhour University, 22516, Damanhour, Egypt.
E-mail addresses: elsehlyfigo@yahoo.com (E.M. Elsehly), alaa.mohamed@science.sohag.edu.eg (A.M. Adam).

https://doi.org/10.1016/j.physe.2022.115566
Received 1 February 2022; Received in revised form 3 October 2022; Accepted 8 November 2022
Available online 11 November 2022
1386-9477/© 2022 Elsevier B.V. All rights reserved.
E.M. Elsehly et al. Physica E: Low-dimensional Systems and Nanostructures 146 (2023) 115566

High-temperature thermal annealing can improve the structure of in­ measurement zone. The thermal diffusivity was measured in both lon­
dividual CNTs by improving crystallinity, healing defects, and gitudinal and transverse directions by laser flash measurement tech­
straightening the walls of MWCNTs [17]. In turn, annealing is expected nique using Netzsch laser flash apparatus (LFA447). The density (ρ) of
to improve the thermal and electrical transport properties of CNT bulk each MWNT tablet was calculated from its mass (m), area (A), and
materials [13, 17]. Annealing provides a reagent-free approach to thickness (t) (Table 1). The thermal conductivity κ was calculated from
remove catalyst particles, which is necessary for applications where measurements of the thermal diffusivity, D, the specific heat, Cp, and the
toxicity is a concern [18]. The CNT structural is highly dependent on the density ρ, using the relationship: κ = D•Cp•ρ. The heat capacities (Cp)
annealing temperature and the tube quality [13]. Moderated annealing were measured using a differential scanning calorimeter (DSC 200 F3
temperatures improve the crystallinity of highly disordered tubes. Maia, NETZSCH) under nitrogen flow (50 ml/min).
However, defect healing may require higher temperature ranges [14].
Thermoelectric (TE) properties of CNTs were measured using a 3. Results and discussions
micro-fabricated device [19,20]. The experiments show positive See­
beck coefficient (S) values due to oxygen adsorption in the range 3.1. SEM analysis
100–102 μVK-1 for individual MWCNTs. The S values were roughly
linear temperature dependent. The observed larger values of S for SEM images of the pristine and annealed MWCNTs are shown in
randomly oriented semiconducting nanotubes compared to individual Fig. 1. It can be observed that pristine TM tubes are with smaller
CNT were attributed to inter-tube barriers relative to the temperature diameter and highly entangled showing a spaghetti-like shape, Fig. 1a.
gradient [21]. Usability of CNTs in TE generation also depends on the Upon annealing, the morphology of the tubes of the TM samples has
electrical conductivity σ and thermal conductivity k beside the Seebeck been remarkably enhanced, Fig. 1b. For pristine TMD samples, the SEM
coefficient. The performance of a thermoelectric material is represented image, Fig. 1c, shows random distribution and knotted tubes, while after
by the so-called dimensionless figure of merit ZT: the heat treatment, Fig. 1d, the structure was enhanced, and the tubes
showed relatively better alignment. Annealing in vacuum is expected to
S2 σ
ZT = T (1) improve the structure, purity and crystallinity of the nanotubes by
k
removal of the structural defects as has been demonstrated in the next
where T is the temperature (in Kelvin). A ZT value of about ~1.2 × sections by XRD and Raman spectroscopy analyses. Heat treatment of
10− 5and ~2 × 10− 5at 300 K for nitrogen- and boron-doped MWCNTs, can significantly improve their mechanical, electrical and thermal
respectively were reported [22]. properties due to reduced defects and increased crystallinity [24,25].
In this work, it has been revealed that annealing in vacuum improves The diameter of CNTs, was determined based on a survey of 250 MWNTs
the thermoelectric figure of merit of MWCNTs. It has been illustrated randomly chosen among resolvable CNTs of each sample. The average
that the enhancement in ZT is due to an increase in the Seebeck coef­ external tube diameter increased upon annealing from 15 nm to 20 nm
ficient and electrical conductivity, simultaneously, with a decrease in for TM and from 30 to 35 nm for TMD, Table 1. This increase could be
the thermal conductivity upon annealing. attributed to amorphous carbon present before annealing being incor­
porated into the CNTs upon annealing.
2. Experimental methods
3.2. XRD characterization
Powder of commercial MWCNTs, Taunit-M (TM) and Taunit-MD
(TMD), are used as initial materials. TM tubes are with length of 5 XRD patterns of pristine and annealed MWNTs are depicted in Fig. 2.
μm, average external diameter of 15–25 nm, and specific surface area The most significant graphite-like Bragg peaks (002) and (100) are
(SSA) of 234 m2/g. TMD tubes are 8 μm in length, with diameter range clearly shown. It is worthwhile to note, the diffraction peak (002)
from 30 to 50 nm, and SSA of 199 m2/g. Similarly, to previous work characterizes the interlayer spacing between the graphene planes while
[23], MWCNTs were purified using mixture of hydrochloric acid and the Bragg peak (100) characterizes the two-dimensional in-plane sym­
H2O procured from Shenzhen Nanotech, China. Purified MWCNT sam­ metry along the graphene layers. Considering the stacking of MWNTs
ples were pressed in the form of pellets at pressure of 30 MPa with may be turbostratic, the two-dimensional reflection peak (100) could be
diameter of 10 mm and height 2 mm. Annealing of the MWNTs under referred as (10), as explained in Ref. [26]. The position and width of the
study was performed at 1 × 10− 6 mbar using a quartz tube furnace (002) peak denotes the structural ordering of the carbon materials [27].
equipped with a turbo molecular pump. Heating was started at 5 ◦ C/min The XRD patterns of the pristine and annealed TM samples exhibit (002)
of reaching the high vacuum in the quartz chamber. The samples were peak at angles 2θ = 38.78 and 38.98◦ , respectively [28,29], taking into
annealed at 1200 ◦ C for 8 h and cooled naturally in vacuum. The pristine account that usually the Cu–K alpha ray is used instead of Cr in the XRD
and annealed Samples were investigated by a desktop X-ray diffrac­
tometer (Difray 401, Scientific Instruments, Russia) using Cr-Kα radia­ Table 1
tion (λ = 2.2909 Å) at room temperature in order to investigate the The inter-planar d-spacing, average diameter, the density, Raman parameters,
microstructural changes that associated with the annealing process. The crystalline size, and (C–C) (sp2) % of TM and TMD samples before and after
MWCNTs morphology was studied using a field emission scanning annealing.
electron microscope (FE-SEM) (Carl Zeiss Ultra 55 SEM). A confocal MWCNT d (Å) Average Density IG/ I2D/ La (C–C)
micro-Raman spectroscopy analysis was performed using a NTEGRA samples diameter (ρ) ID ID (nm) (sp2)
Spectra system in backscattering mode with a laser of 632.8 nm (1.96 (nm) (gm/ %
cm3)
eV) excitation wavelength. Chemical analysis of the samples before and
after treatment was conducted using a PHI 5000 C XPS system and TM 3.452 15 ± 0.5 0.312 0.8 0.7 4.9 ± 96.61
1.2
energy-dispersive X-ray spectroscopy. The purity of the samples was also
A-TM 3.426 18 ± 0.5 0.241 1.25 1.15 7.2 ± 98.12
indirectly evaluated from a Thermogravimetric analysis (TGA). 1.5
The standard four-probe and differential methods using laboratory- TMD 3.355 30 ± 0.5 0.424 0.85 0.65 5.4 ± 97.32
made system (Cryotel, Russia) were employed to measure the elec­ 1.3
trical conductivity and Seebeck coefficient, respectively. The measure­ A-TMD 3.331 35 ± 0.5 0.337 1.33 1.24 7.9 ± 99.45
1.6
ments were carried out in He2 atmosphere to avoid contamination and
ensure homogeneous distribution of the temperature inside the EDS analysis was performed to characterize the studied materials revealing the
constituting elements. Results are illustrated in Table 2.

2
E.M. Elsehly et al. Physica E: Low-dimensional Systems and Nanostructures 146 (2023) 115566

Fig. 1. SEM images of TM-MWNT before annealing (a), and after annealing (b), and TMD- MWNT before annealing (c), and after annealing (d). The SEM images have
been taken from the surface.

Fig. 2. X-ray diffraction pattern for pristine, and annealed TM (a) and TMD samples (b).

measurements. The TMD samples exhibit (002) peak at angles of 39.75◦

Table 2
and 40.11◦ for pristine and annealed samples, respectively. Experi­
Elemental composition of MWCNTs samples before and after treatment deter­
mentally, a d-002 spacing of 3.45 Å was determined for pristine and
mined by EDX [30,31].
3.42 Å for annealed TM samples. For TMD sample that has larger
Wt (%) C O Mg Al Co
diameter, we observed a systematic decrease in d-002 spacing from 3.35
Å to 3.33 Å after annealing. This variation in the XRD peak position can TM 84.83 13.73 0.38 0.15 0.78
be attributed to the diameter difference between TM and TMD samples. A-TM 98.89 0.94 0.08 0.05 0.04
TMD 86.35 12.78 0.28 0.38 0.21
It is understood that the decrease in the inter-shell spacing with higher
A-TMD 99.1 0.87 0.00 0.03 0.00
diameter nanotubes results primarily from interlayer interactions and
defect mediated structural stabilization of nanotube layer. Note that
lower diameter tubes possess high curvature and thus high strain that sp3-hybridized carbon atoms in the nanotubes) at 1340 cm− 1, G band
results in a higher interplanar spacing. As expected, interlayer d-spacing (graphite band of sp2-hybridized carbon atoms) at 1580 cm− 1, and G′
decreases with increasing MWCNT diameter [29].. band (second overtone of the defect induced D band and is related to the
three-dimensional order) at 2700 cm− 1. These peaks are typical features
3.3. 4- Raman spectra of the MWCNTs and used for determining their graphitic nature and
structural ordering [32]. The relative intensities (the peak height) of the
Fig. 3 presents Raman spectra of pristine and annealed MWCNTs. G-band to D-band (IG/ID) well-known indicator for the structure quality
Raman investigations were carried out using 632.8 nm laser source and crystallinity of a nanotube sample. The Raman spectra of TM and
within Raman shift in the range of 1000–3200 cm− 1. The spectra show TMD samples show a reduction in the D-band intensity by annealing
three characteristic peaks corresponding to the D band (disorder band of with a simultaneous increase in the intensity of the G-band. Therefore,

3
E.M. Elsehly et al. Physica E: Low-dimensional Systems and Nanostructures 146 (2023) 115566

Fig. 3. Raman spectra of MWCNTs after different annealing conditions.

the IG/ID ratio increases from 0.8 to 1.25 and from 0.85 to 1.33 for TM
and TMD samples, respectively, indicating an improvement in the
crystallinity with annealing in accordance with the SEM images in
Fig. 1a and b. The increase in the IG’/ID or I2D/ID ratio from 0.7 to 1.15
and from 0.65 to 1.24 for TM and TMD samples, respectively (see
Table 1); confirming the crystallinity enhancement and the considerable
reduction in the structural graphitic disorder in the MWCNTs by
annealing [33–37]. The increase of the MWCNT quality is expected to
improve the thermoelectric properties of the MWCNT. Raman spectra
reveal a narrowing of the G-band after annealing. Worth mentioning
that we can refer to G′ band as 2D for standard analysis.
The width of G-band can be used to extract the crystallite size within
the CNTs, using the empirical formula proposed by Mallet-Ladeira et al.
[38, 39]:
Γ G = (68 ± 4) − (5.2 ± 0.5)La (2)

where ΓG is the half width at half maximum (HWHM) of the G-peak and
La is the average in-plane crystallite size within the CNT. From the data
in Fig. 3, we have La values for the pristine and annealed samples
respectively, Table 1. Annealed samples showed large La compared with
the pristine CNTs. These values show a clear and strong indication of
Fig. 4. TGA data for pristine and vacuum annealed TMD samples.
densifying the stacking of graphene layers through annealing.
The MWCNTs purity could be obtained from the TGA experiment
based on the difference between the burn temperatures of CNTs and annealing (see Table 1). It is worthwhile to note, the σ-values match well
amorphous carbon. The TGA curves of TMD samples before and after with the values reported by Kunadian et al. [22] for randomly oriented
annealing are shown in Fig. 4. It is clear that after saturation, the total MWCNTs pressed in tablet form.
weight losses of the non-annealed TMD sample and of the vacuum Conduction mechanisms in semiconductors are described by various
annealed one were approximately 86.8 and 95.3%, respectively. It is models. The σ-T plots presented in Fig. 5 (b) are found to be well rep­
shown from Fig. 4 that the residual components decreased after vacuum- resented by the Arrhenius model according to the following equation:
annealing. Moreover, the oxidation process of the pristine TMD sample − Eg
started earlier than the annealed one [40]. The TGA data can be σ (T) = σ o (3)
2KB T
explained in terms of a decrease in the amorphous carbon and an
enhancement in the graphitization for the vacuum-annealed CNT where, σo is constant, Eg is the band gap energy KB is the Boltzmann
samples. constant. This indicates that the intrinsic band-to-band transition
Fig. 5a illustrates the electrical conductivity (σ) vs. temperature (T) mechanism dominates through thermal activation process. The linear
of the pristine and annealed MWCNTs for all samples, the electrical behavior of Ln(σ) vs. 1000/T plots, which are presented in Fig. 5 (b),
conductivity increases as the temperature increases indicating a typical confirm the well coincidence of the temperature dependency of σ with
semiconductor behavior. The σvalues of the pristine and annealed bulk Eq. (3). Additionally, appearance of the Ln(σ) vs. 1000/T plot in a one
MWCNTs samples are lower than the values reported for an individual linear portion over the whole temperature range of measurement sug­
CNT [22]. This can be attributed to the charge carriers scattering at the gests that the band-to-band transition conduction dominates as a unique
interfaces between the randomly oriented tubes and hence the decrease mechanism in the samples with the increase in temperature. The Ln (σ)
of their mean free bath. On the other hand, annealing improves signif­ vs. 1000/T plots allow to determine the Eg values. The calculations
icantly the electrical conductivity of both the TM and TMD bulk samples revealed that the annealing process leads to remarkable reduction on the
due to the crystallinity enhancement. The data are in consistence with Eg value from 9.5 ± 0.01 to 8.7 ± 0.007 meV for TMD CNTs and from 8.5
results of the Raman spectroscopy analyses where the IG/ID and IG’/ID ± 0.009 to 8.1 ± 0.005 meV for TMD CNTs. The decrease in the Eg values
ratios increases indicating to a decrease in the defect content with with annealing matches well with the associated increase in σ values

4
E.M. Elsehly et al. Physica E: Low-dimensional Systems and Nanostructures 146 (2023) 115566

Fig. 5. Temperature dependence of the (a) electrical conductivity, (b) Ln(σ) vs. 1000/T, (c) Seebeck coefficient, (d) power factor, (e) thermal conductivity and (f)
figure of merit of MWCNTs.

which can be attributed to the decrease in the defects and the crystal­ conducted in air, the samples were probably oxygen doped with hole-
linity improvement. Worth noting that the detected increase in the like carrier charges resulting in the observed positive Seebeck coeffi­
electrical conductivity can also be attributed to the enhancement of the " cient. Previously, Collins has reported that as-produced SWCNTs can
σ o " (intercept) instead of the energy gap (slope) [41], as shown in Fig. 5 show p-type semiconducting behavior where they become hole-doped
(b). due to oxygen adsorption [42]. Similar results have been reported for
Fig. 5 (c) shows the Seebeck coefficient (S) of the samples under as-produced xylene-derived MWCNTs and B-doped MWCNTs because of
study as a function of temperature (T). It can be observed that the S contribution of the hole-like carriers [43].
values for pristine and annealed samples are on the positive axis where S The room temperature Seebeck coefficient (S300) of the TMD and TM
has positive values over the whole range of temperature indicating that increases from 9.5 to 11.5 μVK− 1 and from 6 to 7 μVK− 1, respectively
the holes are the dominant charge carriers. Since the experiments were with annealing. The pristine nanotubes have lower S values compared to

5
E.M. Elsehly et al. Physica E: Low-dimensional Systems and Nanostructures 146 (2023) 115566

the annealed nanotubes possibly due to the presence of defects created Seebeck coefficient of the MWCNTs under investigation while decreases
on the outer walls of the nanotube because of adsorbed oxygen. significantly the thermal conductivity. The thermoelectric performance
It is worthwhile to note, evaluation of the thermoelectric efficiency of MWCNTs enhances by annealing due to the enhancement of the
of a material can be estimated from the power factor (PF=S2σ) as a electrical conductivity and Seebeck coefficient and the simultaneous
simplified expression [44]. Fig. 5 (d) illustrates the temperature decrease in the thermal conductivity. The figure of merit ZT increases by
dependence of the PF of the samples. Generally, the plots indicate that about 147% at 300 K to 160% at 450 K for TM sample and about 229% at
PF increases with increasing the temperature in the same behavior of σ 300 K to 247% at 450 K for the TMD sample. The difference in the ZT
and S. The annealing process significantly enhances the PF value of the enhancement of the two types of MWCNT at the same temperature of
MWCNTs under study where the room temperature PF300 increases from measurement by annealing is attributed to the difference in the tube’s
0.13 to 0.2 μWm− 1K− 2 and from 0.63 to 1.07 μWm− 1K− 2 for TM and diameter and aspect ratio. Although the MWCNTs have very low ther­
TMD, respectively. The improved PF is attributed to the improvement of moelectric performance due to their ultra-high thermal conductivity,
the electrical conductivity and Seebeck coefficient upon annealing. their performance can be enhanced significantly by annealing.
Fig. 5 (e) shows the thermal conductivity (k) of pristine and annealed
MWCNTs. The thermal conductivity of the MWCNTs was about 2times Declaration of competing interest
lower than those measured by Kunadian et al. and 4–6 times lower than
those reported by Huang et al. [45]. The room temperature thermal The authors declare that they have no known competing financial
conductivity (k300) values for the samples are much lower than those of interests or personal relationships that could have appeared to influence
the individual MWCNTs (3000–6000 W/m K) [46–48]. The low k values the work reported in this paper.
of our randomly oriented carbon nanotubes are attributed to phonon
scattering at the resistive thermal junctions between the CNTs [49]. Data availability
Adsorption of oxygen during the annealing process also results in
structural changes that increases phonon scattering and thus decreases Data will be made available on request.
the k values. This could explain the decrease in the thermal conductivity
of the TMD samples from 0.23 to 0.17 W/mK. Acknowledgements
The figure of merit was calculated using Eq. (1) to obtain the effi­
ciency of the samples for electricity generation. Fig. 5 (f) shows the This study was supported financially by the Academy of Scientific
temperature dependence of ZT plots which shows an increase in the ZT Research and Technology (ASRT), Egypt, Grant No. 6735.
values upon increasing the temperature. This behavior can be attributed Princess Nourah bint Abdulrahman University Researchers Sup­
to the increase in both the electrical conductivity and Seebeck coeffi­ porting Project number (PNURSP2022R124), Princess Nourah bint
cient. The maximum ZT value was realized for the A-TMD sample to be Abdulrahman University, Riyadh, Saudi Arabia.
3.7 × 10− 3 at 450 K. Annealing shows varied effects on the efficiency of
the samples. The ZT increased about 147% at 300 K to 160% at 450 K for References
TM sample and about 229% at 300 K to 247% at 450 K for the TMD
sample which can be attributed to the enhancement in the electrical [1] S. Berber, Y.K. Kwon, D. Tománek, Unusually high thermal conductivity of carbon
conductivity and Seebeck coefficient and the decrease in the thermal nanotubes, Phys. Rev. Lett. 84 (2000) 4613–4616.
[2] E. Pop, D. Mann, Q. Wang, K. Goodson, H. Dai, Thermal conductance of an
conductivity shown above. Compared to results of other attempts re­ individual single-wall carbon nanotube above room temperature, Nano Lett. 6
ported for enhancing the ZT values of the MWCNTs, the observed (2006) 96–100.
enhancement by annealing shows comparable results, for example, with [3] T.W. Ebbesen, H.J. Lezec, H. Hiura, J.W. Bennett, H.F. Ghaemi, T. Thio, Electrical
conductivity of individual carbon nanotubes, Nature 382 (1996) 54–56.
the results reported in previous work [50–53], where doping with boron
[4] M.-F. Yu, O. Lourie, M.J. Dyer, K. Moloni, T.F. Kelly, R.S. Ruoff, Strength and
increases the ZT of MWCNTs 200%. Although the obtained PF values in breaking mechanism of multiwalled carbon nanotubes under tensile load, Science
this study are smaller than previously obtained values for other CNT vol. 287 (2000) 637–640.
[5] M.B. Jakubinek, M.A. White, G. Li, C. Jayasinghe, W. Cho, M.J. Schulz, V. Shanov,
samples [54], our systematic study proves that heat treatment can be
Thermal and electrical conductivity of tall, vertically aligned carbon nanotube
used as an effective technique to improve the thermoelectric perfor­ arrays, Carbon 48 (2010) 3947–3952.
mance (PF or ZT) of randomly oriented CNT. Other reports obtained [6] J.H. Pöhls, M.B. Johnson, M.A. White, R. Malik, B. Ruff, C. Jayasinge, M.J. Schulz,
relatively high PF using high density and vertically aligned CNT array V. Shanov, Physical properties of carbon nanotube sheets drawn from nanotube
arrays, Carbon 50 (2012) 4175–4183.
[55]. On the other hand, the most synthesized CNTs are randomly ori­ [7] M.B. Jakubinek, M.B. Johnson, M.A. White, C. Jayasinghe, G. Li, W. Cho, M.
ented because obtaining well-aligned CNTs is very difficult in lab as is J. Schulz, V. Shanov, Thermal and electrical conductivity of array-spun multi-
requires expensive techniques and high cost. In the application point of walled carbon nanotube yarns, Carbon 50 (2012) 244–248.
[8] A.E. Aliev, M.H. Lima, E.M. Silverman, R.H. Baughman, Thermal conductivity of
view, our study tends to save energy and cost by controlling the struc­ multi-walled carbon nanotube sheets: radiation losses and quenching of phonon
ture of random CNTs rather than consuming the efforts in obtaining modes, Nanotechnology 21 (2010) 1–11, 035709.
well-oriented tubes. Also, it is expected that randomly oriented CNT [9] M.F.L. De Volder, S.H. Tawfick, R.H. Baughman, A.J. Hart, Carbon nanotubes:
present and future commercial applications, Science 339 (2013) 535–539.
materials have lower thermal conductivity compared with that of [10] A.M. Marconnet, M.A. Panzer, K.E. Goodson, Thermal conduction phenomena in
well-aligned CNT samples. As known, lower thermal conductivity means carbon nanotubes and related nanostructured materials, Rev. Mod. Phys. 85 (2013)
better thermoelectric performance. However, more study is needed to 1295–1326.
[11] N. Behabtu, C.C. Young, D.E. Tsentalovich, O. Kleinerman, X. Wang, A.W.K. Ma, E.
enhance the thermoelectric performance of randomly oriented CNT
A. Bengio, R.F. terWaarbeek, J.J. de Jong, R.E. Hoogerwerf, S.B. Fairchild, J.
materials. B. Ferguson, B. Maruyama, J. Kono, Y. Talmon, Y. Cohen, M.J. Otto,
M. PasqualiStrong, Light, multifunctional fibers of carbon nanotubes with
ultrahigh conductivity, Science 339 (2013) 182–186.
4. Conclusions [12] X. Zhang, Q. Li, T.G. Holesinger, P.N. Arendt, J. Huang, P.D. Kirven, T.G. Clapp, R.
F. DePaula, X. Liao, Y. Zhao, L. Zheng, D.E. Peterson, Y. Zhu, Ultrastrong stiff, and
Two sets of MWCNTs samples with different diameters were inves­ lightweight carbon-nanotube fibers, Adv. Mater. 19 (2007) 4198–4201.
[13] Agathe Figarol, Jeremie Pourchez, Delphine Boudard, Valerie Forest,
tigated before and after vacuum annealing. Raman spectroscopy showed
Sarah Berhanu, Jean-Marc Tulliani, Jean-Pierre Lecompte, Michele Cottier,
an increase in the IG/ID ratio after annealing at 1200 ◦ C, indicating an Didier Bernache-Assollant, Philippe Grosseau, J. Nanoparticle Res. 17 (2015) 1–14.
improvement of the crystallinity of the CNTs tablets. The conduction [14] J.Y. Huang, F. Ding, K. Jiao, B. Yakobson, Self-templated growth of carbon-
mechanism in pristine and annealed MWCNTs is thermally activated nanotube walls at high temperatures, Small 3 (2007) 1735–1739.
[15] J.F. Niven, M.B. Johnson, S.M. Juckes, M.A. White, N.T. Alvarez, V. Shanov,
where the charge carriers contribution to conduction increases by Influence of annealing on thermal and electrical properties of carbon nanotube
temperature. Annealing increases the electrical conductivity and yarns, Carbon 99 (2016) 485–490.

6
E.M. Elsehly et al. Physica E: Low-dimensional Systems and Nanostructures 146 (2023) 115566

[16] E.M.M. Ibrahim, Vyacheslav O. Khavrus, Albrecht Leonhardt, Silke Hampel, [36] S.L.H. Rebelo, et al., Progress in the Raman spectra analysis of covalently
Steffen Oswald, Mark H. Rümmeli, Bernd Büchner,Synthesis, characterization, and functionalized multiwalled carbon nanotube:unraveling disorder in graphitic
electrical properties of nitrogen-doped single-walled carbon nanotubes with materials, Phys. Chem. Chem. Phys. 18 (2016) 12784–12796.
different nitrogen content, Diam. Relat. Mater. 19 (2010) 1199–1206. [37] Y. Wei, L. Jiang Liu, Z. Fan Chen, Vacuum-breakdown-induced needle-shaped ends
[17] Vyacheslav O. Khavrus, E.M.M. Ibrahim, Albrecht Leonhardt, Silke Hampel, of multiwalled carbon nanotube yarnsand their field emission applications, Nano
Steffen Oswald, Christine Täschner, Bernd Büchner, Conditions of simultaneous Lett. 7 (12) (2007) 3792–3797.
growth and separation of single- and multi-walled CNx nanotubes, J. Phys. Chem. C [38] E.M. Elsehly, N.G. Chechenin, A.V. Makunin, A.A. Shemukhin, H.A. Motaweh, Eur.
114 (2010) 843–848. Phys. J. D 71 (2017) 79.
[18] A.M. Adam, P. Petkov, Impacts of thickness reduction and heat treatment on the [39] P. Mallet-Ladeira, P. Puech, C. Toulouse, M. Cazayous, N. Ratel-Ramond,
optical properties of thin chalcogenide films, Ceram. Int. 43 (2017) 11015–11022. P. Weisbecker, G.L. Vignoles, M. Monthioux, Carbon (80) (2014), 629–39.
[19] P. Kim, L. Shi, A. Majumdar, P.L. McEuen, Thermal transport measurements of [40] R. Andrews, D. Jacques, D. Qian, E.C. Dickey, Carbon 39 (2001) 1681–1687.
individual multiwalled nanotubes, Phys. Rev. Lett. 87 (2001), 215502, 1–4. [41] Ryo Shikata, et al., Nanotechnology 33 (2022), 235707.
[20] N. Kang, L. Lu, W.J. Kong, J.S. Hu, W. Yi, Y.P. Wang, et al., Observation of a [42] P.G. Collins, Extreme oxygen sensitivity of electronic properties of carbon
logarithmic temperature dependence of thermoelectric power in multiwall carbon nanotubes, Science 287 (5459) (2000) 1801–1804.
nanotubes, Phys. Rev. B 67 (2003), 033404, 1–4. [43] A.M. Adam, A.K. Diab, M. Ataalla, Maged F. Alotaibi, Abdulaziz N. Alharbi, E.
[21] J. Hone, M.C. Llaguno, N.M. Nemes, A.T. Johnson, J.E. Fischer, D.A. Walters, et al., M. Elsehly, Optimized thermoelectric performance in thin (Bi2Se3)1− x(Bi2Te3)x
Electrical and thermal transport properties of magnetically aligned single wall alloyed films, J. Alloys Compd. 898 (2022), 162888.
carbon nanotube films, Appl. Phys. Lett. 77 (2000) 666–668. [44] A. El-Khouly, A.M. Adam, Y. Altowairqi, I. Serhiienko, E. Chernyshova, A. Ivanova,
[22] Illayathambi Kunadian, Rodney Andrews, M. Pinar Mengüç, Dali Qian, Carbon 47 V.L. Kurichenko, A. Sedegov, D. Karpenkov, A. Novitskii, A. Voronin,
(2009) 589–601. Parkhomenko Yu, V. Khovaylo, Transport and thermoelectric properties of Nb-
[23] E.M. Elsehly, N.G. Chechenin, A.V. Makunin, D.A. Pankratov, H.A. Motaweh, doped FeV0.64Hf0.16Ti0.2Sb half-Heusler alloys synthesized by two ball milling
J. Water Proc. Eng. 25 (2018) 81. regimes, J. Alloys Compd. 890 (2022), 161838.
[24] A.M. Adam, E. Lilov, E.M.M. Ibrahim, P. Petkov, L.V. Panina, M.A. Darwish, [45] Q. Huang, L. Gao, Y. Liu, J. Sun, Sintering and thermal properties of multiwalled
J. Mater. Process. Technol. 264 (2019) 76–83. carbon nanotube–BaTiO3 composites, J. Math. Chem. 15 (2005) 1995–2001.
[25] M.G. Nematov, M.M. Salem, M. Adam, A.M. Ahmed, L.V. Panina, A.T. Morchenko, [46] L. Shi, D. Li, C. Yu, W. Jang, D. Kim, Z. Yao, et al., Measuring thermal and
Effect of stress on magnetic properties of annealed glass-coated Co71Fe5B11Si10Cr3 thermoelectric properties of one-dimensional nanostructures using a
amorphous microwires, IEEE Trans. Magn. 53 (11) (2017), 7921694. microfabricated device, J. Heat Tran. 125 (2003) 881–888.
[26] C.R. Houska, B.E. Warren, "X-Ray study of the graphitization of carbon black, [47] M. Osman, D. Srivastava, Temperature dependence of the thermal conductivity of
J. Appl. Phys. 25 (1954) 1503–1509, https://doi.org/10.1063/1.1702373. single-wall carbon nanotubes, Nanotechnology 12 (2001) 21–24.
[27] A. Cao, C. Xu, J. Liang, D. Wu, Bingqiug Wei, Chem. Phys. Lett. 344 (2001) 13. [48] J. Che, T. Cagin, W.A. Goddard, Thermal conductivity of carbon nanotubes,
[28] Dilip K. Singh, P.K. Iyer, P.K. Giri, Diameter dependence of interwall separation Nanotechnology 11 (2000) 65–69.
and strain in multiwalled carbon nanotubes probed by X-ray diffraction and Raman [49] P. Kim, L. Shi, A. Majumdar, P.L. McEuen, Thermal transport measurements of
scattering studies, Diam. Relat. Mater. 19 (10) (2010) 1281–1288. individual multiwalled nanotubes, Phys. Rev. Lett. 87 (2001), 215502, 1–4.
[29] P. Nie, C. Min, H.J. Song, et al., Preparation and tribological properties of [50] G. Yamamoto, K. Shirasu, Y. Nozaka, Y. Sato, T. Takagi, T. Hashida, Structure-
polyimide/carboxyl-functionalized multi-walled carbon nanotube nanocomposite property relationships in thermally-annealed multi-walled carbon nanotubes,
films under seawater lubrication, Tribol. Lett. 58 (2015) 7. Carbon 66 (2014) 219–226.
[30] E.M. Elsehly, N.G. Chechenin, A.V. Makunin, H.A. Motaweh, E.G. Leksina, Water [51] W. Huang, Y. Wang, G. Luo, F. Wei, 99.9% purity multi-walled carbon nanotubes
Sci. Technol. 75 (2017) 1564. by vacuum high-temperature annealing, Carbon 41 (2003) 2585–2590.
[31] E.M. Elsehly, N.G. Chechenin, A.V. Makunin, H.A. Motaweh, Mater. Res. Express 3 [52] D. Mattia, M.P. Rossi, B.M. Kim, G. Korneva, H.H. Bau, Y. Gogotsi, Effect of
(2016), 105013. Graphitization on the wettability and electrical conductivity of CVD-carbon
[32] M. Endo, T. Hayashi, H. Muramatsu, Y.A. Kim, H. Terrones, M. Terrones, M. nanotubes and films, J. Phys. Chem. B 110 (2006) 9850–9855.
S. Dresselhaus, Nano Lett. 4 (2004) 1451–1454. [53] A.M. Adam, E.M.M. Ibrahim, Panbude Anshu, K. Jayabal,
[33] E.M. Elsehly, A.P. Evseev, E.A. Vorobyeva, et al., The effect of argon irradiation on Pandiyarasan Veluswamy, A.K. Diab, Thermoelectric power properties of Ge doped
the filtration properties of multiwalled carbon nanotubes, Tech. Phys. Lett. 48 PbTe alloys, J. Alloys Compd. 872 (2021), 159630.
(2022) 58–61, https://doi.org/10.1134/S1063785022030026. [54] M.T.Z. Myint, T. Nishikawa, K. Omoto, et al., Controlling electronic states of few-
[34] C.J. An, Y.H. Kang, H. Song, Y. Jeong, S.Y. Cho, High-performance flexible walled carbon nanotube yarn via joule-annealing and p-type doping towards large
thermoelectric generator by control of ElectronicStructure of directly spun carbon thermoelectric power factor, Sci. Rep. 10 (2020) 7307, https://doi.org/10.1038/
nanotube webs with various molecular dopants, J. Mater. Chem. A. 5 (2017) s41598-020-64435-0.
15631–15639. [55] M. Hada, T. Hasegawa, H. Inoue, M. Takagi, K. Omoto, D. Chujo, et al., One-minute
[35] A.R. Hopkins, A.C. Labatete-Goeppinger, H. Kim, H.A. Katzman, Space Joule annealing enhances the thermoelectric properties of carbon nanotube yarns
survivability of carbon nanotube yarn material inLow earth orbit, Carbon 107 via the formation of graphene at the interface, ACS Appl. Energy Mater. 2 (2019)
(2016) 77–86. 7700–7708, https://doi.org/10.1021/acsaem.9b01736.