Oxidative acid treatment of carbon nanotubes

Surfaces and Interfaces 14 (2019) 1–8

Contents lists available at ScienceDirect

Surfaces and Interfaces journal homepage: www.elsevier.com/locate/surfin

Oxidative acid treatment of carbon nanotubes ⁎

T

Nurettin Sezer , Muammer Koç Division of Sustainable Development (DSD), Hamad Bin Khalifa University (HBKU), Qatar Foundation (QF), Education City (EC), Doha, Qatar

A R T I C LE I N FO

A B S T R A C T

Keywords: Carbon nanotubes Acid treatment Functionalization X-ray photoelectron spectroscopy Transmission electron microscopy Surface functional groups

This study investigated the oxidative acid treatment of multi-walled carbon nanotubes (MWCNT). Experiments were conducted by refluxing MWCNT with HNO3 and H2SO4/HNO3 (3:1 v/v) at two different molar concentrations. Treated MWCNT were characterized for their weight loss, surface chemistry, surface morphology, structural integrity, thermal stability and dispersion stability using a precision scale, X-ray Photoelectron Spectroscopy (XPS), Transmission Electron Microscopy (TEM), Thermogravimetric Analysis (TGA), and visual observation. The results showed that treatment with 15 M HNO3 introduced the highest atomic % of oxygencontaining chemical moieties to MWCNT surface with less sample weight loss. The TEM micrographs showed that during treatment with 15 M HNO3, MWCNT fragmented into smaller nanotubes of an average length of 900 nm. MWCNT experienced the most severe structural degradation after treatment with 15 M H2SO4/HNO3, where the nanotubes were completely dissolved during the refluxing process. Carboxyl and carbonyl groups were identified as the predominant functional groups present on treated MWCNT surfaces. TGA study revealed the preservation of graphitic MWCNT structure, increased water sorption capacity, and newly formed functional groups after the treatments. Dispersion of MWCNT was improved after all treatments. However, 10 M HNO3treated sample showed the weakest stability. According to the experimental results, MWCNT treatment with 15 M HNO3, yielded a formation of 5.2 at% surface oxygen with only 6.8% sample weight loss. The extent of surface oxidation in the case of 10 M H2SO4/HNO3 treatment was 3.2 at% and the weight loss was about 17%.

1. Introduction Carbon nanotubes (CNT) have attracted a growing attention by researchers of diverse fields due to their numerous unique properties such as structural, electrical, mechanical, thermal and gas storage [1–4]. However, the hydrophobic and chemically inert nature of the graphitic network presents a major challenge for the utilization of CNT in nanotechnology applications. Thus, functionalization of CNT becomes a prerequisite to improve their chemical reactivity [5,6]. By functionalization, chemical groups such as carboxyl (–COOH), carbonyl (–CO), and hydroxyl (-OH) are formed on the CNT surfaces, which exfoliate the CNT bundles, improve their wettability and enable their dispersion in polar media. In the presence of metal precursors, these surface groups help the diffusion and nucleation of metals on the CNT sidewalls for the synthesis of ideal CNT composites [7,8]. The interfacial adhesion of CNT with polymers is also improved by the functionalities on the surface which impart CNT a better compatibility with polymer matrix [9–11]. Through functionalization, hollow cavities on CNT surface are opened, which is essential in the application of gas storage [12]. The functionalization is widely conducted by oxidative acid treatment of CNT using various type of acids and acid mixtures such as HNO3, ⁎

H2SO4/HNO3, HCl/H2SO4/HNO3, H2O2, KMnO4, K2Cr2O7/H2SO4, and KMnO4/H2SO4 [8,13–18]. The strong interaction between the oxidizing acid molecules and CNT during the treatment results in creation of defect sites on the graphitic network. Generation of these oxidative defects involves replacement of one carbon atom from CNT lattice with one or more oxygen atoms to form functional groups such as carboxylic acid (–COOH) and hydroxyl (–OH) on the CNT surface [19]. These functional groups have rich chemistry [20]. Acid functionalization requires careful selection of acids and treatment conditions for the optimum result. Mild conditions may not introduce enough functionality to the CNT surface even after extended periods of treatment, while the CNT may suffer severe structural degradation in the very early stages of harsh acid treatment. The optimum treatment method should provide maximum functionality to the CNT surface with negligible structural degradation and weight loss [10]. In addition, the process should not be time-consuming, energy intensive, and costly. In the literature, several acids and acid mixtures have been investigated for CNT functionalization. For instance, Zhang et al. [21] used different acids and molar concentrations. The functional groups observed on SWCNT differed with the type of oxidant used during

Corresponding author. E-mail address: [email protected] (N. Sezer).

https://doi.org/10.1016/j.surfin.2018.11.001 Received 11 July 2018; Received in revised form 30 October 2018; Accepted 2 November 2018 Available online 03 November 2018 2468-0230/ © 2018 Elsevier B.V. All rights reserved.


N. Sezer, M. Koç

Fig. 1. (a) FESEM and (b) TEM micrographs of high purity MWCNT in as-purchased condition. Outside diameter: < 20 nm and length: 1–12 µm. A negligible amount of amorphous carbon adsorbed on the MWCNT sidewalls.

In the literature, there are numerous published researches on acid functionalization of CNTs especially with HNO3 and H2SO4/HNO3 under different experimental conditions. However, there is no available study to investigate the functionalization of MWCNT by HNO3 and H2SO4/HNO3 under identical treatment conditions. To insightfully understand and fairly assess the functionalization performance of HNO3 and H2SO4/HNO3 on MWCNT, it is essential to conduct an experimental study using these two acids under the identical treatment conditions. In this study, functionalization performance of HNO3 and H2SO4/HNO3 on MWCNT is systematically investigated under the same treatment conditions using two acid concentrations. The surface morphology, structural integrity, weight loss, surface chemistry, thermal stability, and dispersion stability of the functionalized MWCNT were characterized by Transmission Electron Microscopy (TEM), precision scale, X-ray Photoelectron Spectroscopy (XPS), Thermogravimetric Analysis (TGA), as well as the visual observation of aqueous MWCNT dispersions, respectively. In brief, this paper aims to provide a critical insight into understanding the effect of acid type (HNO3 and H2SO4/HNO3) and concentration (diluted [10 M] and concentrated [15 M]) on the surface functionalization of MWCNT.

treatment. KMnO4 was identified as a powerful agent for generating new defect sites at short treatment time, whereas diluted HNO3 (2.6 M) was not effective to create defects even after prolonged treatment time. Osorio et al. [22] systematically studied the effect of acid mixtures on functionalization of CNT. Among the studied acids; H2SO4/HNO3/HCl, H2SO4/HNO3, and HNO3, the best dispersion stability were obtained by CNT sample which was treated with H2SO4/HNO3/HCl. The results from Raman spectroscopy indicated that the graphitic structure of CNT was maintained. The addition of carboxyl and hydroxyl groups on CNT surface at all the treatment conditions was confirmed by FT-IR. Saleh [23] studied the effect of varying the acid refluxing temperature (60, 80, 100, 120 and 140 ⁰C) on the functionalization performance of MWCNT using H2SO4/HNO3 and HNO3. The ideal reaction temperatures for an efficient oxidation of MWCNT with HNO3 and H2SO4/HNO3 were determined by titration method as 120 ⁰C and 100 °C, respectively. It was concluded that both the acid type and treatment temperature has strong effects on the extent of the surface oxygen content of functionalized-MWCNT. The time has also a pronounced effect on the functionalization of CNT [23]. For instance, Chiang et al. [7] demonstrated the performance of acid functionalization of CNT at different time periods. 4 M aqueous mixture of H2SO4/HNO3 (3/1, v/v) was refluxed at the boiling temperature for several time periods of 0.5, 1, 2, 4 and 7 days. According to their results, the graphitic structure of the CNT was preserved and the concentration of oxygen-containing moieties increased with increasing treatment time. Dispersion of CNT in solutions is hindered by the formation of CNT bundles due to the strong Van-der Waals interaction among the asgrown CNT and high surface energy caused by the strong curvature of the thin nanotubes [10,24]. The functional groups, introduced on CNT surface by acid functionalization, impart repulsive forces to exfoliate these bundles [11,25]. Zhu et al. [26] achieved a considerable improvement in the dispersion of SWCNT by H2SO4/HNO3 treatment, which finally yielded a well-integrated SWCNT in the epoxy composites. Tchoul et al. [27] functionalized SWCNT under sonication using 8 M HNO3 to improve their dispersion in solutions. The mild treatment exfoliated SWCNT bundles, which resulted in an improved dispersibility in polar solvents. Naseh et al. [28] observed a good dispersion after chemically modifying the MWCNT with HNO3 by 4 h refluxing. It was attributed to the -OH groups formed after acid treatment that make hydrogen bonding with the water molecules. Solubility was further studied by a prolonged period of treatment time [29]. High solubility in the range of 20–40 mg/ml was obtained after long exposure of 24–48 h of MWCNT in concentrated HNO3 (> 60%), but 60–90% MWCNT weight was lost.

2. Methodology 2.1. Materials MWCNT of high purity (> 99%) synthesized by the CatalyticChemical Vapor Deposition (CCVD) method was purchased from Cheaptubes Inc. (Grafton, USA) and used as received. As-purchased MWCNT was already purified using a Dielectric Barrier Discharge process. Morphology of the as-purchased MWCNT was analyzed using FESEM (JEOL FESEM 7500) and TEM (Talos F200X). Fig. 1 shows the micrographs of the as-received MWCNT, which was of outside diameter and length of < 20 nm and 1–12 µm, respectively. The MWCNTs were in the form of big bundles due to the strong Van der Waals interactions between the nanotubes (Fig. 1(a)). The TEM micrograph shows that the as-purchased MWCNT has high purity with negligible amorphous carbon adsorbed on the MWCNT sidewalls (Fig. 1(b)). In the experiments, ultrapure HNO3 69% (w/w) (Romil, Cambridge, UK) and H2SO4 95–98% (w/w) (Honeywell Fluka, Leicestershire, UK) were employed. 2.2. CNT functionalization Acid treatment experiments were conducted as follows. 250 ml acid solutions were prepared at required concentrations and added into a 2


N. Sezer, M. Koç

Table 1 Experimental conditions for each treatment processes. Experiment

Acid type and concentration

Treatment conditions

MWCNT weight (mg)

Acid solution volume (ml)

Treatment Treatment Treatment Treatment

15 M 10 M 15 M 10 M

5 h refluxing at 120 ⁰C

250

250

1 2 3 4

HNO3 HNO3 H2SO4/HNO3 (3:1 v/v) H2SO4/HNO3 (3:1 v/v)

cycles using centrifugation (Eppendorf 5810 R). After pH of the MWCNT solutions reached above 5, they were dried in an electric furnace overnight at 80 °C and then ground into powder form using mortar and pestle. For MWCNT functionalization, four treatment approaches were investigated as summarized in Table 1. 2.3. Characterization The functionalized MWCNT samples were characterized for their weight loss, dispersion stability in aqueous media, surface chemistry, surface morphology, structural integrity, and thermal stability. 2.3.1. Weight loss During acid treatment, it is vital to know the effect of the applied method on the amount of sample weight loss. Acid treatment is a labor and energy intensive process. The substantial weight losses during treatment would result in an even higher energy consumption rate per unit weight of functionalized CNTs produced. Therefore, the MWCNT samples, before and after each treatment processes, were weighed in a precision scale to calculate percent weight losses.

Fig. 2. Sample weight loss as a function of treatment condition. H2SO4/HNO3 is a stronger acid relative to HNO3, therefore, it leads to a substantial weight loss of MWCNTs during the treatment process. However, minimal weight losses were observed with MWCNTs treated by HNO3.

2.3.2. Aqueous dispersion stability CNT functionalized through different treatment routes were dispersed in deionized water at a solid concentration of 0.1 wt% using an ultrasonication probe (Vibra-Cell from Sonics&Materials) for 15 min at 20 kHz and – 450 W with 3 s pulse 1 s wait cycles. The colloidal suspensions were preserved at ambient conditions and visually observed for their stability for a duration of three months. 2.3.3. Surface morphology The micrographs of MWCNT before and after the treatments were taken by using TEM (Talos F200X) at 200 kV. The samples were prepared for TEM observation by drop casting of the dilute dispersions of MWCNT onto the carbon coated copper micro-grids followed by solvent evaporation in air at room temperature. 2.3.4. X-ray photoelectron spectroscopy X-Ray Photoelectron Spectroscopy (XPS) was employed to confirm the formation of different surface chemical functional groups and defect sites on MWCNT surface. The XPS spectra of all samples were acquired using a spectrophotometer (Thermo Fisher Scientific ESCALAB 250Xi) with a microfocused monochromator AlKα X-ray source (1486.68 eV). The survey scan spectra were collected at a range of 0–1350 eV binding energy with a step size of 1 eV, in order to identify the chemical composition of the functionalized nanotube surface. Moreover, high-resolution XPS spectra with a step size of 0.1 eV was employed over C1s and O1s regions. For quantification, background subtraction (Shirley type) followed by curve fittings (mixed Gaussian-Lorentzian type) were studied in Thermo Avantage™. The peak fitting practice was repeated until an acceptable fit was acquired. The atomic ratios were calculated in Thermo Avantage™ from the photoelectron peak area, using sensitivity factors available in the software library.

Fig. 3. Photographs of pristine and treated MWCNT dispersions; (a) after sonication, (b) after 3 months. Treatments of 15 M HNO3 and 10 M H2SO4/HNO3 maintained a good dispersion of MWCNT over three months. However, pristine MWCNT and MWCNT treated with 10 M HNO3 showed a weak dispersion behavior.

round bottom flask. Then, 250 mg of as-received MWCNT were weighed and added into the acid solution. The round bottom flask was immersed into an oil bath and the oil bath was heated using a hot plate which was placed underneath it. The temperature of the oil bath was set to 120 °C and 5 h refluxing time was selected for each treatment. The time "zero" of the experiments was considered the moment of appearance of the initial reflux drops. The acid refluxing process was carried out under vigorous magnetic stirring. After completion of the refluxing process, the MWCNT solution was undertaken many dilutions and decantation

2.3.5. Thermogravimetric analysis Oxidation resistance of the MWCNT was determined under continuous oxygen flow with a heating rate of 10 °C/min using 3


N. Sezer, M. Koç

Fig. 4. TEM micrographs at high (0.1) and low (0.2) magnifications. (a) Pristine MWCNT, (b) 10 M H2SO4/HNO3-treated MWCNT, (c) 10 M HNO3-treated MWCNT, and (d) 15 M HNO3-treated MWCNT. The MWCNT surfaces were almost completely free from amorphous carbon. The initial nanotube structure was maintained after treatment. Following acid treatment, however, the nanotube bundles were noticeable exfoliated and curled. Nanotubes were apparently fragmented into smaller tubes after treatment with 15 M HNO3.

Fig. 5. XPS survey spectrum of pristine and treated MWCNT. The survey scan spectra were collected at a range of 0–1350 eV binding energy with a step size of 1 eV, in order to identify the chemical composition of the functionalized nanotube surface. Noticeable peaks are observed due to carbon and oxygen. Fig. 6. High-resolution XPS peak deconvolutions of pristine and treatedMWCNT over C1s region, inset figure is a closer view of the curves in the energy range of 285–294 eV. The highest peak at 284.3 eV is due to the graphitic network, defects on MWCNT surface were detected at 285.2 eV. The peaks due to carbons attached to different oxygen-containing moieties are shown at 287.1, 288.9 and 290.4 eV. The π–π* transition loss peak was identified at 291.5 eV.

thermogravimetric analyzer (Discovery TGA from TA Instruments) which measures the changes in sample weight as a function of temperature. 3. Results and discussion 3.1. Weight loss

stronger acid than HNO3 [30]. Nevertheless, HNO3 treatment resulted in only 5% and 6.8% weight loss at diluted (10 M) and concentrated (15 M) treatment conditions, respectively.

Comparison of the weight losses after acid treatment of each sample is shown in Fig. 2. MWCNT treated with H2SO4/HNO3 experienced significant weight losses of 17% and 100%, at 10 M and 15 M acid concentrations, respectively, which was due to that H2SO4/HNO3 is a 4


N. Sezer, M. Koç

Table 2 Quantitative results (atomic %) of the fits in XPS C1s and O1s regions. Sample

Pristine MWCNT MWCNT-HNO3, 10 M MWCNT-H2SO4/HNO3,10 M MWCNT-HNO3, 15 M

Binding energy (eV) Carbon 284.3 285.2 C (sp2) C (sp3)

287.1 C]O

288.9 -COOH

290.4 Carbonates

291.5 π–π*

Oxygen 531.2 C]O, O–C]O

533.2 C–O–C, –COOH, C–OH

535.5 H2O

62.8 61.2 62.8 58.6

4.2 4.6 4.0 5.1

2.6 3.3 3.2 3.4

3.6 2.2 2.4 2.4

3.3 4.0 3.4 3.0

0.34 1.21 1.91 1.92

0.31 0.91 1.14 2.93

0.05 0.08 0.15 0.30

22.8 22.5 21.0 22.4

Fig. 7. High-resolution XPS peak deconvolutions of pristine and treated-MWCNT over O1s region. The peaks are identified for oxygen bonded to carbon, including double bonds; C] O, OeC]O at 531.2 eV, and single bonds; CeO, CeOH, CeOeC at 533.2 eV. Moreover, chemisorbed oxygen was detected at 535.5 eV.

the samples but after one week, sedimentation started with the pristine sample and at the end of three months, 10 M HNO3-treated sample also started to sediment. However, 10 M H2SO4/HNO3- and 15 M HNO3treated samples exhibited excellent stabilities with no visual sedimentation throughout the observation period. The suspension samples photographed after preparation and after three months are shown in Fig. 3. 3.3. Surface morphology From TEM micrographs it is possible to assess the diameter, number of walls, structure, and relative concentration of defect sites (defect site density) and the presence of adsorbed amorphous carbon in the CNT sidewalls before and after treatment. The comparison of the representative TEM micrographs of the pristine MWCNT and functionalized MWCNT is presented in Fig. 4. For the pristine MWCNTs, the TEM image (Fig. 4 (a.1)) shows that there was a negligible amorphous carbon content on the MWCNT surface and it does not increase following any of the applied acid treatment (Fig. 4 (b.1), (c.1), and (d.1)). The as-received MWCNTs were purified by the manufacturer and most of the amorphous carbon and catalytic metal nanoparticles was possibly eliminated during the purification process. Few defects present on the MWCNTs sidewalls (Fig. 4 (a.1)) could be generated during this purification process. Presence of some defects in the sidewall of the acid treated MWCNT can also be observed in Fig. 4 (b.1), (c.1), and (d.1). The amorphous carbon content and defective site density of the MWCNT surface before and after acid treatment in all treatment conditions appeared to be similar. This was indicative of the preservation of the initial structure of the MWCNT, in full agreement with XPS results. After oxidation with 10 M H2SO4/

Fig. 8. Change in surface atomic % of C1s and O1s, and O/C ratio as a function of acid type and concentration. After treatment with 15 M HNO3, the oxygen content (at%) of the pristine MWCNT increased from 0.72% to a maximum of 5.15% and the O/C ratio increased from 0.71% to 5.43%. Similarly, the oxygen content (at%) was increased to 2.18% and 3.13% on MWCNT surface treated with 10 M HNO3 and 10 M H2SO4/HNO3, respectively.

3.2. Aqueous dispersion stability Dispersion stability of treated MWCNT was visually investigated. Firstly, the MWCNT were dispersed in deionized water through sonication. Then, all the treated MWCNT samples were observed for their stability. In the initial stage, no sedimentation was observed with any of 5


N. Sezer, M. Koç

Fig. 9. Variations of the surface atomic ratio of sp2bonded carbon and sp3-bonded carbon as a function of the applied treatment condition. The sp2-bonded carbon and sp3-bonded carbon content of MWCNT did not vary much with any of the applied treatment methods. Therefore, no significant change in the ratio between sp2 and sp3-orbitals of the MWCNT was observed after treatment, indicating the initial structure of MWCNT was preserved.

Fig. 10. Sample weight loss as a function of temperature, inset graph is a closer look to the TGA plot for initial weight losses in temperature range 150–450 ⁰C. The weight loss between 150 and 400 ⁰C in treated MWCNTs is due to the decomposition of surface-attached moieties. The disordered carbon was degraded during a temperature rise from 400 to 450 ⁰C. The temperature corresponding to the highest weight loss rate is due to the oxidation of the graphitic structure.

of 0.1 eV. Deconvolution of high-resolution XPS spectra over the C1s region for reference and functionalized samples was conducted, as shown in Fig. 6. The deconvolution of the C1s photoemission peak showed the highest peak at 284.3 eV due to the graphitic network [31], whereas the defects on MWCNT surface were detected at a peak value of 285.2 eV [32]. The peaks corresponding to the carbons attached to different oxygen-containing moieties are shown at 287.1, 288.9 and 290.4 eV [2]. Lastly, the π-π* transition loss peak was identified at 291.5 eV [7]. In Table 2, 15 M HNO3-treated MWCNT had the highest atomic % of the carbon from C]O and COOH groups, and the lowest from π–π* transitions, which indicates that the treatment with 15 M HNO3 gives the highest functionality to MWCNT surface among the tested treatment conditions. The deconvolution of XPS O1s photoemission peaks confirms the presence of oxygen bonded to carbon, including double bonds; C]O, OeC]O at 531.2 eV, and single bonds; CeO, CeOH, CeOeC at

HNO3 (Fig. 4 (b.2)), 10 M HNO3 (Fig. 4 (c.2)), and 15 M HNO3 (Fig. 4 (d.2)), some bundles noticeably exfoliated and curled. Especially in the case of 15 M HNO3 treated MWCNT, severe fragmentation of the MWCNT takes place as clearly seen in (Fig. 4 (d.2)). The average length of the nanotubes was estimated to be reduced to 900 nm. These indicate the shortening effect of a strong oxidizing agent on the nanotube length. However, in the case of treatment with 10 M H2SO4/HNO3, the MWCNT still maintained their original length. The above observations are in good agreement with the XPS results presented in the following section. 3.4. X-ray photoelectron spectroscopy Fig. 5 shows the XPS survey scan for pristine and treated MWCNT. The detected elements are labeled as shown. Since the XPS survey spectra showed noticeable peaks due to carbon and oxygen only, the high-resolution XPS spectra of C1s and O1s were studied over the binding energy ranges of 280–295 eV and 525–545 eV with a step size 6


N. Sezer, M. Koç

Fig. 11. Derivative weight loss curves as a function of temperature for pristine and acid-treated MWCNTs (a) in the temperature range of 100–900 °C, (b) in the temperature range of the highest sample weight loss rates.

533.2 eV on treated MWCNT surfaces (Fig. 7) in good agreement with the previous research [32]. Finally, chemisorbed oxygen was detected at 535.5 eV [7]. Results showed that the major oxygen contribution was from C]O and CeO bindings. The atomic % of oxygen from C]O, and CeO and chemisorbed water was calculated and presented in Table 2. Change in surface atomic percent of C1s and O1s, and O/C ratio as a function of acid type and concentration are illustrated in Fig. 8. After treatments, the concentration of carbon atoms decreased with the formation of new oxygen bearing moieties on the surface of MWCNT. As such, oxygen concentration and the oxygen/carbon (O/C) ratio on the surface of MWCNT increased. In the literature, pristine MWCNT typically contains around 1 at% surface oxygen [33]. Pristine MWCNT, in this study, contained only 0.72 at% oxygen on the surface which was then increased to a maximum of 5.15 at.% after treatment with 15 M HNO3. Likewise, the O/C ratio increased from 0.71% to 5.43% during treatment with 15 M HNO3. Dilute acid treatments also yielded higher oxygen % on MWCNT surface. The oxygen content was 2.18 at% and 3.13 at% on MWCNT surface treated with 10 M HNO3 and 10 M H2SO4/ HNO3, respectively. The contribution of chemisorbed water was also found in the pristine sample which increased after treatment with 15 M HNO3. The presence of an increased amount of oxygen-containing functionalities on MWCNT surface supports the colloidal stability of the 10 M H2SO4/HNO3 and 15 M HNO3 treated MWCNT in polar media as presented in Section 3.2. sp2-carbon atoms present in the graphitic structure of the MWCNT sidewalls [34], while sp3-carbon atoms exist at defect sites in the MWCNT sidewalls or in the amorphous carbon [35]. The ratio of sp2- and sp3-bonded carbon atom can provide a metric for the structural integrity of the overall CNT. The variations of the surface atomic ratios of sp2-bonded carbon and sp3-bonded carbon corresponding to the applied treatment method are plotted in Fig. 9. As shown in the figure, the atomic ratio of both sp2-bonded carbon and sp3-bonded carbon remained almost the same at all the applied treatments. Therefore, no significant change in the ratio between sp2 and sp3-orbitals of the MWCNT was observed after treatment. It indicates that the initial structure of the MWCNT was preserved.

purchased MWCNT sample stayed stable with no weight loss by heating until the temperature reaches 150 °C which indicates no water was adsorbed in the bulk of pristine MWCNT due to its hydrophobic nature. With further heating, oxidation started and it resulted in 1.5% weight loss at 450 °C, which is attributed to the oxidation of carbonaceous impurities present in the as-purchased MWCNT [36,37]. After the oxidation of graphitic carbon was complete, the remaining sample quantity was 2.1%, which was the catalyst remaining in the as-purchased sample. The oxidation of the catalyst with further temperature rise resulted in an increase of mass to 3.6% at 1000 °C. Thermal degradation behavior of the treated samples was also observed. In the early stage of heating, until the temperature reaches 150 °C, an initial weight loss of nearly 1% was observed due to the evaporation of the adsorbed water in the case of more hydrophilic 15 M HNO3-treated sample [38]. The weight loss in the stage between 150 and 350 °C was due to the decarboxylation of carboxyl groups present on MWCNT surface [39]. The thermal degradation from 350 to 400 °C is attributed to the elimination of other few oxygenated surface groups in defects of the MWCNT structure. Consequently, the weight loss % between the temperatures 150 and 400 °C in the functionalized materials can be related to the decomposition of surface-attached moieties [40]. The disordered carbon was degraded during a temperature rise from 400 to 450 °C [2]. The oxidation of graphitic structure occurred at the temperature where the weight loss rate was maximum. After complete oxidation of graphitic structure, the remaining weight was attributed to the impurities due to the catalysts [38]. Following acid treatment, these impurities were significantly reduced from 2.1% to 1.6%, 1.4% and 1.2% by 10 M H2SO4/HNO3, 10 M HNO3, and 15 M HNO3, respectively. It can be inferred that concentrated HNO3 is an effective agent for purification of CNT [41]. The highest thermal stability was observed with pristine MWCNT due to the absence of the functional groups. 15 M HNO3-treated MWCNT exhibited the highest degradation in the temperature range of 150–450 °C indicating the presence of the functional groups. These findings are in good agreement with the TEM and XPS results.

3.5. Thermogravimetric analysis

4. Conclusions

TGA analyses were conducted to confirm the functionalization of the materials. The results of TGA are shown on a percent weight loss versus temperature graph in Fig. 10. In Fig. 11, the derivative weight loss curves (-dW/dT) directly shows the thermal occasions, such as the onset of burning (Fig. 11(b)), as a function of temperature. As-

This study presents the findings of an investigation on the oxidative acid treatment of MWCNT by refluxing HNO3 and H2SO4/HNO3 at different molar concentrations. The most severe chemical attack was observed with 15 M H2SO4/HNO3, which completely dissolved the MWCNT during the harsh refluxing process and caused MWCNT weight 7


N. Sezer, M. Koç

totally lost. 10 M H2SO4/HNO3 treatment also resulted in significant material weight loss (17%), however, it was only 5% and 6.8% after 10 M HNO3 and 15 M HNO3 treatments, respectively. According to the TEM observations, MWCNTs were exfoliated and curled after all treatments. The structure of the nanotubes after 10 M H2SO4/HNO3 treatment appeared to be more intact but after treatment with 15 M HNO3, the MWCNTs fragmented into smaller nanotubes of an average size of 900 nm. 15 M HNO3 treatment yielded well functionalized MWCNT with 5.15 at% oxygen on MWCNT surface which was followed by 3.2 at% with 10 M H2SO4/HNO3 and 2.2 at% with 10 M HNO3. According to the XPS results, carboxyl and carbonyl groups were identified as the main oxygen-containing moieties present on the treated MWCNT surface. Dispersion stability was highest with 15 M HNO3-treated MWCNT followed by 10 M H2SO4/HNO3 and 10 M HNO3-treated MWCNT in good agreement with the XPS results. TGA results under oxygen flow showed that H2O content of MWCNT due to water adsorption increased with the surface chemical functionality. 1% weight loss due to water evaporation was recorded for 15 M HNO3treated MWCNT that also confirmed the XPS result with highest surface oxygen concentration. Likewise, percent weight loss increased with functionality in a temperature range of 150 – 450 °C due to the newly formed functional groups on MWCNT surface. In summary, this paper comparatively demonstrated the performance of HNO3 and H2SO4/ HNO3 on surface functionalization of MWCNT by a systematic experimental study.

[14] B. Smith, K. Wepasnick, K.E. Schrote, H.-H. Cho, W.P. Ball, D.H. Fairbrother, Influence of surface oxides on the colloidal stability of multi-walled carbon nanotubes: a structure−property relationship, Langmuir 25 (17) (2009) 9767–9776 Sep.. [15] F. Avilés, J.V. Cauich-Rodríguez, L. Moo-Tah, A. May-Pat, R. Vargas-Coronado, Evaluation of mild acid oxidation treatments for MWCNT functionalization, Carbon N. Y. 47 (13) (2009) 2970–2975 Nov.. [16] S. Haydar, et al., Regularities in the temperature-programmed desorption spectra of CO2 and CO from activated carbons, Carbon N. Y. 38 (9) (2000) 1297–1308. [17] A.G. Osorio, I.C.L. Silveira, V.L. Bueno, C.P. Bergmann, H2SO4/HNO3/ HCl—Functionalization and its effect on dispersion of carbon nanotubes in aqueous media, Appl. Surf. Sci. 255 (5) (2008) 2485–2489 Dec.. [18] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Characterization of active sites on carbon catalysts, Ind. Eng. Chem. Res. 46 (12) (Jun. 2007) 4110–4115. [19] H. Wilson, et al., Electrical monitoring of sp3 defect formation in individual carbon nanotubes, J. Phys. Chem. C 120 (3) (2016) 1971–1976. [20] P. Ma, N.A. Siddiqui, G. Marom, J. Kim, Composites : part a dispersion and functionalization of carbon nanotubes for polymer-based nanocomposites : a review, Compos. Part A 41 (10) (2010) 1345–1367. [21] J. Zhang, H. Zou, Q. Qing, Y. Yang, Q. Li, and Z. Liu, “Effect of chemical oxidation on the structure of single-walled carbon nanotubes,” pp. 3712–3718, 2003. [22] A.G. Osorio, I.C.L. Silveira, V.L. Bueno, and C.P. Bergmann, “H2SO4/HNO3/ HCl—Functionalization and its effect on dispersion of carbon nanotubes in aqueous media,” vol. 255, pp. 2485–2489, 2008. [23] T.A. Saleh, The influence of treatment temperature on the acidity of MWCNT oxidized by HNO3 or a mixture of HNO3/H2SO4, Appl. Surf. Sci. 257 (17) (2011) 7746–7751. [24] M. Zhang, L. Su, L. Mao, Surfactant functionalization of carbon nanotubes (CNTs) for layer-by-layer assembling of CNT multi-layer films and fabrication of gold nanoparticle/CNT nanohybrid, Carbon N. Y. 44 (2) (2006) 276–283 Feb.. [25] B. Ruelle, et al., Functionalization of carbon nanotubes by atomic nitrogen formed in a microwave plasma Ar + N 2 and subsequent poly(ε-caprolactone) grafting, J. Mater. Chem. 17 (2) (2007) 157–159. [26] J. Zhu, J. Kim, H. Peng, J.L. Margrave, V.N. Khabashesku, E.V. Barrera, Improving the dispersion and integration of single-walled carbon nanotubes in epoxy composites through functionalization, Nano Lett. 3 (8) (2003) 1107–1113. Aug.. [27] M.N. Tchoul, W.T. Ford, G. Lolli, D.E. Resasco, S. Arepalli, Effect of mild nitric acid oxidation on dispersability, size, and structure of single-walled carbon nanotubes, Chem. Mater. 19 (23) (2007) 5765–5772. [28] M.V. Naseh, A.A. Khodadadi, Y. Mortazavi, O.A. Sahraei, F. Pourfayaz, S.M. Sedghi, Functionalization of carbon nanotubes using nitric acid oxidation and DBD plasma, World Acad. Sci. Eng. Technol. 27 (2009) 567–569. [29] I.D. Rosca, “Oxidation of multiwalled carbon nanotubes by nitric acid,” vol. 43, pp. 3124–3131, 2005. [30] J. Yang, S. Wang, X. Zhou, and J. Xie, “Electrochemical behaviors of functionalized carbon nanotubes in LiPF 6 / EC + DMC Electrolyte,” vol. 7, pp. 6118–6126, 2012. [31] H. Ago, et al., Work functions and surface functional groups of multiwall carbon nanotubes, J. Phys. Chem. B 103 (38) (Sep. 1999) 8116–8121. [32] Z. Zhang, L. Pfefferle, G.L. Haller, Comparing characterization of functionalized multi-walled carbon nanotubes by potentiometric proton titration, NEXAFS, and XPS, Chinese J. Catal. 35 (6) (2014) 856–863. [33] K.A. Wepasnick, B.A. Smith, K.E. Schrote, H.K. Wilson, S.R. Diegelmann, D.H. Fairbrother, Surface and structural characterization of multi-walled carbon nanotubes following different oxidative treatments, Carbon N. Y. 49 (1) (2010) 24–36. [34] S. Osswald, E. Flahaut, H. Ye, Y. Gogotsi, Elimination of D-band in Raman spectra of double-wall carbon nanotubes by oxidation, Chem. Phys. Lett. 402 (4–6) (2005) 422–427. [35] K.R. Moonoosawmy and P. Kruse, “Ambiguity in the characterization of chemically modified single-walled carbon nanotubes : a Raman and ultraviolet - visible - nearinfrared study,” vol. 113, pp. 5133–5140, 2009. [36] H. Hu, P. Bhowmik, B. Zhao, M.. Hamon, M.. Itkis, R.. Haddon, Determination of the acidic sites of purified single-walled carbon nanotubes by acid–base titration, Chem. Phys. Lett. 345 (1-2) (2001) 25–28 Sep.. [37] A. Hirsch, Functionalization of single-walled carbon nanotubes, Angew. Chemie Int. Ed. 41 (11) (2002) 1853 Jun.. [38] W.M. Silva, et al., Surface properties of oxidized and aminated multi-walled carbon nanotubes, J. Braz. Chem. Soc. 23 (6) (2012) 1078–1086. [39] M. Tang, H. Dou, K. Sun, One-step synthesis of dextran-based stable nanoparticles assisted by self-assembly, Polymer (Guildf) 47 (2) (2006) 728–734 Jan.. [40] M.E. Lipińska, S.L.H. Rebelo, M.F.R. Pereira, J.A.N.F. Gomes, C. Freire, J.L. Figueiredo, New insights into the functionalization of multi-walled carbon nanotubes with aniline derivatives, Carbon N. Y. 50 (9) (2012) 3280–3294. [41] H. Hu, B. Zhao, M.E. Itkis, and R.C. Haddon, “Nitric acid purification of singlewalled carbon nanotubes,” pp. 13838–13842, 2003.

Conflict of interest The authors declare that there is no conflict of interest regarding the publication of the article entitled “Oxidative Acid Treatment of Carbon Nanotubes”. References [1] E.T. Thostenson, Z. Ren, T.-W. Chou, Advances in the science and technology of carbon nanotubes and their composites: a review, Compos. Sci. Technol. 61 (13) (Oct. 2001) 1899–1912. [2] V. Datsyuk et al., Chemical oxidation of multiwalled carbon nanotubes, vol. 6, pp. 2–9, 2008. [3] M. Meyyappan, Carbon Nanotubes: Science and Application, CRC Press, 2005. [4] A. Mews, Y. Gogotsi (Ed.), Nanomaterials Handbook, 46 CRC Press, 2006. [5] N. Sezer, M. Koç, Dispersion stability of CNT and CNT/metal-based nanofluids, International Conference on Thermal Engineering: Theory and Applications, 2018, pp. 1–4. [6] N. Sezer, M. Koç, Stabilization of the aqueous dispersion of carbon nanotubes using different approaches, Therm. Sci. Eng. Prog. 8 (2018) 411–417 June. [7] Y.C. Chiang, W.H. Lin, Y.C. Chang, The influence of treatment duration on multiwalled carbon nanotubes functionalized by H2SO4/HNO3 oxidation, Appl. Surf. Sci. 257 (6) (2011) 2401–2410. [8] C. Prado-Burguete, A. Linares-Solano, F. Rodríguez-Reinoso, C. Salinas-Martínezde Lecea, The effect of oxygen surface groups of the support on platinum dispersion in Pt/carbon catalysts, J. Catal. 115 (1) (Jan. 1989) 98–106. [9] S. Goyanes, G.R. Rubiolo, A. Salazar, A. Jimeno, M.A. Corcuera, I. Mondragon, Carboxylation treatment of multiwalled carbon nanotubes monitored by infrared and ultraviolet spectroscopies and scanning probe microscopy, Diam. Relat. Mater. 16 (2) (2007) 412–417. [10] S.Y. Yang, et al., Effect of functionalized carbon nanotubes on the thermal conductivity of epoxy composites, Carbon N. Y. 48 (3) (2010) 592–603. [11] C.H. Lau, et al., The effect of functionalization on structure and electrical conductivity of multi-walled carbon nanotubes, J. Nanopart. Res. 10 (Suppl. 1) (2008) 77–88. [12] S. Banerjee, T. Hemraj-Benny, and S.S. Wong, “Covalent surface chemistry of singlewalled carbon nanotubes,” Advanced Materials, no. 1, pp. 17–29, 2005. [13] A.K. Cuentas-Gallegos, R. Martínez-Rosales, M.E. Rincón, G.A. Hirata, G. Orozco, Design of hybrid materials based on carbon nanotubes and polyoxometalates, Opt. Mater. (Amst). 29 (1) (2006) 126–133 Oct..

8