Fullerenes, Nanotubes and Carbon Nanostructures ...

2 downloads 0 Views 911KB Size Report
Aug 25, 2015 - In fact, YF is positive, ID/IG moderately increases, while IG0/IG remains nearly constant (Figure 3(a)) with respect to pristine NTs. Desorption of ...
This article was downloaded by: [Candida Milone] On: 27 August 2014, At: 14:24 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Fullerenes, Nanotubes and Carbon Nanostructures Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/lfnn20

Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Functionalised via a Novel EcoFriendly Approach to HNO3 Vapor Oxidation a

b

b

c

b

c

S. Santangelo , E. Piperopoulos , S. H. Abdul Rahim , G. Faggio , S. Ansari , G. Messina & C. b

Milone a

Dipartimento di Ingegneria Civile, dell’Energia, dell’Ambiente e dei Materiali (DICEAM), “Mediterranea” University, Reggio Calabria, Italy b

Dipartimento di Ingegneria Elettronica, di Chimica e di Ingegneria Industriale (DIECII), Messina University, Messina, Italy c

Dipartimento di Ingegneria dell’Informazione, delle Infrastrutture e dell’Energia Sostenibile (DIIES), “Mediterranea” University, Reggio Calabria, Italy Accepted author version posted online: 26 Jun 2014.Published online: 25 Aug 2015.

To cite this article: S. Santangelo, E. Piperopoulos, S. H. Abdul Rahim, G. Faggio, S. Ansari, G. Messina & C. Milone (2015) Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Functionalised via a Novel Eco-Friendly Approach to HNO3 Vapor Oxidation, Fullerenes, Nanotubes and Carbon Nanostructures, 23:1, 83-92, DOI: 10.1080/1536383X.2014.885956 To link to this article: http://dx.doi.org/10.1080/1536383X.2014.885956

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Fullerenes, Nanotubes and Carbon Nanostructures (2014) 23, 83–92 Copyright © Taylor & Francis Group, LLC ISSN: 1536-383X print / 1536-4046 online DOI: 10.1080/1536383X.2014.885956

Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Functionalised via a Novel Eco-Friendly Approach to HNO3 Vapor Oxidation S. SANTANGELO1, E. PIPEROPOULOS2, S. H. ABDUL RAHIM2, G. FAGGIO3, S. ANSARI2, G. MESSINA3 and C. MILONE2 Dipartimento di Ingegneria Civile, dell’Energia, dell’Ambiente e dei Materiali (DICEAM), “Mediterranea” University, Reggio Calabria, Italy 2 Dipartimento di Ingegneria Elettronica, di Chimica e di Ingegneria Industriale (DIECII), Messina University, Messina, Italy 3 Dipartimento di Ingegneria dell’Informazione, delle Infrastrutture e dell’Energia Sostenibile (DIIES), “Mediterranea” University, Reggio Calabria, Italy

Downloaded by [Candida Milone] at 14:24 27 August 2014

1

Received 15 October 2013; accepted 26 December 2013

Some crucial aspects of the functionalisation of carbon nanotubes by nitric acid vapors are investigated, namely selectivity of the process and thermal stability in air of the functionalised materials. Results obtained by varying duration (0.5–5.0 h) of the conventional treatment with vapors at azeotropic HNO3 concentration are compared with those obtained, for fixed duration (2.0 h), by using sub-azeotropic HNO3CH2OCMg(NO3)2 solution to generate acid vapors with acid concentration in the range 25–93 wt%, as recently proposed by us. A thorough picture is drawn based on evidences of high-resolution transmission electron microscopy, micro Raman spectroscopy, thermo-gravimetry, derivative thermo-gravimetry, and temperature-programmed desorption systematic analyses. Keywords: multi-walled carbon nanotubes, vapor phase functionalisation, carboxylation, micro-ramanspectroscopy, temperature programmed desorption

Introduction The very broad range of physical and chemical properties exhibited by its natural allotropes and engineered forms make of carbon a material with surprising potential in advanced technological applications (1,2). In the latest years, nanotubes and graphene, whose applications transversally involve the most diverse fields (3–9), have gathered the largest scientific interest among various nanocarbons. Natural tendency to agglomeration and hydrophobic nature of their surface are of great hindrance for the direct employment of nanotubes and graphene in applications. Dispersibility and wettability are mandatory for their processability. Chemical oxidation represents the most frequently utilized approach to improve their dispersibility and wettability via the introduction, over their surface, of polar oxygencontaining functional groups (10,11).

Address correspondence to Saveria Santangelo, Dipartimento di Ingegneria Civile, dell’Energia, dell’Ambiente e dei Materiali (DICEAM), Universit a “Mediterranea”, 89122 Reggio Calabria, Italy. E-mail: [email protected] Color versions of one or more of the figures in this article can be found online at www.tandfonline.com/lfnn.

As strictly concerns nanotubes (NTs), manifold methods are reported in literature (12–29), such as exposure to ozone (12,13), dinitrogentetraoxide (14), oxygen plasma (15), or fuming nitric acid (28). However, attack by acids in liquid phase (LP) is the most commonly used approach to generate oxygenated groups on their sidewalls and open ends (16– 25,29). Nonetheless, liquid phase functionalisation (LPF) shows a lot of important drawbacks (e.g., low efficiency and huge amount of liquid wastes) that strongly limit its application on large scale. In addition, the obtainment of high NT solubility requires prolonged exposure to concentrated acids and involves the loss of important fractions (up to 60–90%) of the initial material, as well as strong fragmentation of the oxidized NTs with amorphous carbon generation on their walls (29). Conversely, aggressive liquid oxidants can be exploited even to produce graphene nanoribbons (NRs) by oxidative unzipping of NTs (30–32). Involving no need of filtering, washing and drying oxidized materials vapor phase functionalisation (VPF) seems to be an efficient solution to overcome disadvantages of the LPF (12–15,27). On one hand, exposure to nitric acid vapors generated by boiling azeotropic solution allows introducing on the NT surface a total amount of oxygenated functionalities exceeding that obtained by means of the LPF with the same acid (27). On the other hand, high care is needed in

Downloaded by [Candida Milone] at 14:24 27 August 2014

84

Santangelo et al.

handling concentrated acid solution, and flexibility is extremely limited. Very recently, we have proposed the use of vapors generated by boiling sub-azeotropic HNO3CH2OCMg(NO3)2 solution (33) to combine advantages of VPF with the greater safety and flexibility deriving from handling more diluted acid solutions and tuning nitric acid concentration in the vapor phase (VP). Fundamentals of the method are very simple: it exploits the well-known “salt effect” (34), namely the ability of a solid salt to alter the vapor/liquid equilibrium enhancing relative volatility of the components of a liquid solution. Fixed the relative amount of magnesium nitrate, the concentration of HNO3 in the VP can be tuned up to values above the azeotropic one by properly choosing the composition of the HNO3CH2O solution, on the basis of the vapor/ liquid equilibrium curves available in literature (35). This paper intends to deepen some aspects of the functionalisation method proposed by us. Attention is focused on effects of acid concentration in the VP (25–93wt%) on morphology, crystalline arrangement, and oxidative resistance of NTs, and typology of the oxygenated groups introduced on their surface, because selectivity of the functionalisation process and thermal stability in air of the functionalised materials are crucial requirements for their subsequent processability (36). The influence of structural properties of the pristine NTs is also investigated in view of the extensive use of the method. For this purpose, a systematic analysis is conducted by means of high-resolution transmission electron microscopy (HRTEM), micro Raman spectroscopy (MRS), thermo-gravimetry (TG), derivative thermo-gravimetry (DTG), and temperature-programmed desorption (TPD).

Experimental Multi-wall nanotubes are synthesized by chemical vapor deposition at 600 C. Fe/Al2O3 catalysts (500 mg), reduced at 500 C upon 60 cc/min H2 flow, are used for reaction. Syntheses are carried out, upon 120 cc/min 1:1 i-C4H10CH2 flow, over catalysts with 17wt% and 29wt% nominal metal

loads. CNTs obtained in the two cases are respectively termed as A- and B-type. After synthesis, treatment with two separate solutions of NaOH and HCl removes from raw NTs residual catalyst support and metallic particles, respectively. Further details concerning preparation and processing of the catalyst, synthesis and purification of the nanotubes can be found in (37) and references therein cited. NTs are functionalised by exposure to nitric acid vapors under conditions reported in Table 1. In order to find the optimal operating conditions, the effect of VPF duration (tF D 0.5–5.0 h) at fixed (azeotropic) concentration of HNO3 in the VP (wV) is preliminarily evaluated on a given type of NTs (entries 1–4) by the use of the commercial (Carlo Erba) nitric acid solution. The equivalence of the two solutions (B D binary or T D ternary) is then ascertained by comparing results obtained, under the same conditions (tF D 2.0 h; wV D 68wt%), on a given type of NTs (entries 3 and 6). Carlo Erba magnesium nitrate hexahydrated is utilized to prepare HNO3CH2OCMg(NO3)2 solution with fixed salt amount (20wt%). The influence of structural properties of the pristine NTs is studied by comparing results of VPF with ternary solution at given conditions (tF D 2.0 h; wV D 68wt%) on different types of NTs (entries 1, 3, 8, and 9). Finally, the effect produced on a given type of NTs by the variation of xV (25–93wt%) is evaluated at fixed tF (entries 5–7). For this purpose, on the basis of the vapor/ liquid equilibrium curves available in literature (35), concentration of HNO3in the LP (wL, calculated as salt-free) is varied in the range 30–58wt%. In addition, the contribution due to water is investigated (entry 10). The experimental set-up utilized for VPF is the same as illustrated by other authors (27). Briefly, it consists of a round bottom flask (RBF), a reactor equipped with heater and temperature controller, and a condenser. A 400 mg of (A- or B-type) NTs are loaded into the reactor and heated up to 135 C. The RBF is filled with 300 g of (B- or T-) acid solution and heated, under magnetic stirring, up to the boiling point (120 C) in an ethylene glycol bath. The condenser is connected to the RBF providing for the solution recycling. Further details can be found in (33).

Table 1. Codes and oxidation conditions of the considered samples Entry 1 2 3 4 5 6 7 8 9 10

Code

Treated with

tF (h)

wL (wt%)

wV (wt%)

YF (wt%)

HNO3CH2O HNO3CH2O HNO3CH2O HNO3CH2OCMg(NO3)2 HNO3CH2OCMg(NO3)2 HNO3CH2OCMg(NO3)2

0.5 2.0 5.0 2.0 2.0 2.0

68 68 68 30 45 58

68 68 68 25 68 93

3.1 8.8 ¡10.5 ¡1.9 6.0 ¡0.2

HNO3CH2OCMg(NO3)2 H2O

2.0 2.0

45 0

68 0

1.9 ¡15.0

A

P B1 A B2 A B3 A T1 A T2 A T3 B P B T2 A W A

Notes: tF stands for duration of the functionalisation treatment; wL and wV indicate the concentrations of HNO3 in liquid and vapor phase, respectively. In the sample code, the superscript refers to the type of NTs (A or B); P stands for pristine NTs; B and T denote NTs functionalised with binary and ternary solutions, respectively; W indicates NTs functionalised with water vapor. Yield (YF) of the functionalisation process is also reported.

Downloaded by [Candida Milone] at 14:24 27 August 2014

Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Oxidized by HNO3 Vapors The treatment is carried out for the desired time duration. Then, the ethylene glycol bath heating is turned off, and NTs are allowed to dry for 0.5 h maintaining the reactor heating at 110 C. Subsequently, the reactor is transferred in an oven and there kept at 110 C for 1h to complete the drying procedure. Finally, the NTs are discharged from the reactor, weighed for yield determination and analyzed. Yield of the VPF process is evaluated as percent variation of the NT mass, namely YF D 100(mF¡m0)/m0, where m0 and mF stand respectively for the NT mass before and after exposure to HNO3 vapors. In absence of mass consumption due to tube wall erosion by acid and/or carbon gasification by water vapor, mF would exceed m0 leading to positive YF values. Size and morphology of the pristine (p-NTs) and functionalised NTs (f-NTs) are investigated by HRTEM. A JEOL JEM 2010 analytical electron microscope (LaB6 electron gun and 200kV accelerating voltage), equipped with a Gatan 794 Multi-Scan CCD camera for digital imaging, is used. For each specimen at least 70 images with increasing magnification factors are recorded in order to have a reliable picture of the sample bulk. The effectiveness of chemical oxidation process is tested by measuring the mass loss caused by desorption of the oxygenated species upon inert environment. TG analysis is conducted in the temperature range 50¡1050 C, after preliminary sample stabilization at 100 C for water removal. A TA Instruments SDTQ 600 (balance sensitivity: 0.1 mg) is used. The same instrument is utilized to evaluate the resistance of NTs to the oxidation in air and to assess their content of metallic impurities (viz. the percent mass of non-burnt matter at the end of each TG measurement (38)). The changes in the NT crystalline arrangement are investigated by means of MRS. A Coherent Innova 70 ArC laser operating at 2.41 eV provides excitation source. Spectra are recorded in air using an Olympus BX40 microscope (X50 objective lens) coupled to a Jobin Yvon Ramanor U-1000 double monochromator and to a Hamamatsu R943-02 photomultiplier operating in photon-counting mode. Care is taken to prevent local heating and damage of the samples. A commercially available spectroscopic analysis software package is used for the spectra decomposition. Lorentzian bands, superimposed to a constant background, are used to reproduce the spectral profile. Speciation and quantification of the surface functional groups is carried out by means of TPD. Measurements are carried out using a flow reactor equipped with a quadrupole mass spectrometer (HPR 20 Hiden Analytical Instrument). Samples are placed in a U-shaped quartz tube inside an electrical furnace and heated up to 1000 C upon 30cc/min He flow. The amounts of CO, CO2 and NO released from NTs are investigated. At the end of each measurement a calibration with pure gases is performed. Further technical details concerning instrumentation utilized and its preliminary calibration, sample preparation, reproducibility check, and experimental data analysis are given elsewhere (37–39).

85

Results Table 1 reports yields of the VPF process under the considered conditions. Both positive and negative values are found, as result of the competition between two opposite mechanisms, namely formation of oxygenated groups causing mass increment and erosion of the NT walls responsible for mass diminishing. The well-known carbon gasification by water vapor, with hydrogen release (C C H2O ! CO C H2) accounts for the largely negative YF value in sample AW. Instead, under the remaining conditions, mass loss percent (if any) is extremely limited, different from the case of LPF with nitric acid, for which a mass loss up to 60–90% of the initial material is reported (29). The high functionalisation yield (common, indeed, to all the approaches to VPF) is a feature of remarkable importance, particularly as concerns cost effectiveness and advantage of present method of functionalisation in large scale over the other ones. The fact that functionalisation is mandatory for the subsequent processing in the most of NT-based applications and the cost of NTs may economically weigh on them so much as to compromise their feasibility in absence of an adequately-great added value by the NTs makes of the functionalisation yield maximization a crucial task. Figure 1 displays the effectiveness of the VPF process, as monitored by the mass loss (Dm) due to desorption of all the oxygenated species. Dm values measured in p-NTs (4wt% and 7wt%, respectively for the A- and B-type ones) reveal the presence of small amounts of surface oxygenated moieties (most likely generated during the post-growth purification step), in agreement with previous findings (12,37). In f-NTs Dm varies in the range 7–28wt% depending on the VPF conditions (Table 1). As known (36,39,40), the temperature of desorption in inert ambient allows assessing the nature of functional groups present on NTs. Decarboxylation occurs mainly below 400 C (21,36). Lactones, anhydrides, phenols, carbonyls, quinines, and ethers are more stable than carboxylic groups and decompose at higher temperatures. Hence, the mass loss (DmCOOH) in the temperature range 50¡400 C compared to Dm gives a coarse idea of the selectivity of the VPF process towards the formation of –COOH groups. It is found that under milder conditions (tF  2.0 h and wV  68wt%) oxidation by HNO3 vapors results in high selectivity towards the formation of carboxylic groups. Instead, selectivity is partly lost if the VPF is carried out upon harsher conditions (tF D 5.0 h or wV D 93wt%). Figures 2–4 show results of the systematic analysis by HRTEM (Figure 2), MRS (Figure 3) and DTG (Figure 4) aimed at assessing the changes produced by the acid attack in the morphology, crystalline arrangement and oxidative resistance of NTs. The analysis of the morphology of untreated samples indicates that A- and B-type NTs look as entangled filaments having external diameters in the range 13–20 nm (Figure 2 (a)) and 5–40 nm (Figure 2(g)–(h)), respectively. In sample, A P tubes are straighter than in sample BP. Higher magnification images prove the absence of amorphous phases and the presence of iron nanoparticles coated by carbonaceous layer

Downloaded by [Candida Milone] at 14:24 27 August 2014

86

Santangelo et al.

Fig. 1. Effectiveness and selectivity of the VPF process, as monitored by results of TG analysis. In each plot, the percent losses of mass due to desorption of all the oxygenated species and to decarboxylation (below 400 C) are indicated.

Fig. 2. Morphology of the investigated NTs. TEM images refer to samples AP (a), AB2 (b), AB3 (c–d), AT2 (e–f) and BP (g–h). White circles in images (c) and (d) indicate open tube ends.

Downloaded by [Candida Milone] at 14:24 27 August 2014

Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Oxidized by HNO3 Vapors

87

Fig. 3. Raman spectra of the considered NTs. Spectra are normalized to the G-band maximum intensity (and the same vertical scale is used) for an easier comparison.

(as visible in Figure 2(a)) or encapsulated within the tube channels. Their relative amount, as resulting from TG measurements, is greater in sample BP than in AP (5.9wt% against 1.6 wt%). As for the treated samples (Figures 2(b)–2(f)), under milder conditions (tF  2.0 h and wV  68wt%), the acid attack produces no appreciable damage on the tube walls. Consistently, except sample AT1, YF is positive (Table1). On the contrary, upon harsher conditions (wV D 93wt% or, particularly, tF D 5.0 h) HRTEM evidences the occurrence of erosion and YF becomes negative. In sample, A B3 even some broken tubes (indicated by white circles in Figures 2(c)–(d)) are observed, together with increased agglomeration degree, in line with literature reports (19,25,39). In no case unzipped NTs are observed. This finding is in accord with the fact that the use of by far more powerful oxidizing agents (e.g., concentrated HNO3CH2SO4 mixture in a 1:3 volume ratio, or KMnO4 in case of singleand multi-walled NTs, respectively (30,31)) is required to obtain the strong oxidation of the NTs leading to the unzipping reaction of the rolled-up graphene sheets with production of graphene NRs. The negative value of YF in sample AT1, where no structural damage due to acid attack is observed, suggests that the high concentration of water (75wt%) and low concentration of acid (25wt%) in the VP cause mass consumption by C

gasification to occur at a faster rate than formation of oxygenated moieties. As for the crystalline arrangement, Raman D-, G-, and G0 -bands (at 1350 cm¡1, 1580 cm¡1, and 2700 cm¡1, respectively) are detected in all the samples (Figure 3). The in-plane optical G-mode, originating from the stretching of all C D C pairs (41), represents the fingerprint of the graphitic crystalline arrangement. The D- and G0 -bands are associated to second-order processes. Involving one iTO phonon and one defect (42), the in-plane D-mode is forbidden in perfect graphite and rendered Raman-active by finite size effects and presence of defects in the hexagonal network (e.g., substitutional hetero-atoms, vacancies, grain boundary, distorted hexagonal, and non-hexagonal rings) that break the crystalline translational symmetry of infinite graphene-layers (21,39,43). Conversely, the G0 mode involves two iTO phonons near the K point (42) and besides being, to a first approximation, defect-independent (17), requires long-range graphitic order for the needed coupling effect. Therefore, D/ G and G0 /G intensity ratios (ID/IG and IG0 /IG) give a measure of defect density and long-range ordering, respectively. Accordingly, normalizing Raman spectra to the G-band maximum intensity (Figure 3) allows directly visualizing the changes in crystalline arrangement of the NTs brought about by the VPF process at different conditions.

Downloaded by [Candida Milone] at 14:24 27 August 2014

88

Santangelo et al.

Fig. 4. DTG profiles of the investigated samples. In each plot the average thermal stability in air (expressed in  C) is reported.

Higher defectiveness (i.e., ID/IG) pertains to sample BP, where a larger amount of bended and twisted tubes is observed than in sample AP. In f-NTs defect density increases with increasing tF and/or wV. In addition, all the bands shift towards higher wavenumbers as an effect of the electron transfer from the p states in NTs to the oxygen atoms and of the strains induced on the nanotube walls (17,44). Nonetheless, long-range graphitic ordering (i.e., IG0 /IG) does not undergo dramatic changes, in line with the preservation of the tubular structure observed by HRTEM (Figure 2). Figure 4 displays results of DTG measurements for the assessment of the NT resistance to the oxidation in air. Both the untreated samples are featured by a doubly peaked profile, which hints at the co-existence of two phases having different thermal stability in air. This is probably related to the different NT diameter distributions (compare Figures 2(a) and 2(g)). A single peak is observed in the most of treated samples. In NTs functionalised upon harsher conditions (samples AB3 and AT3) two weak contributions appear below 450 C, where disordered graphitic phases burn (39,45), in line with the introduction of significant extents of structural disorder assessed through HRTEM and MRS analyses. The average thermal stability in air of the samples (TDTG) is evaluated by weighing the highest-oxidation rate temperature of each peak with the underlying area. Values obtained are reported in Figure 4. The higher density of defects (arising from tube

bending/twisting) makes sample BP thermally less stable than AP (621 C against 630 C) in spite of the thicker tubes. After the exposure to nitric acid vapors TDTG decreases. The treatment with 93wt% HNO3 vapors produces the most marked TDTG diminishing (from 630 C down to 555 C). On the contrary, exposure to water vapors does not significantly lower TDTG (from 630 C to 625 C). A trend inversion is noted for tF D 5 h.

Discussion Effect of Treatment Duration and Type of Acid Solution As concerns the duration of the VPF process at fixed HNO3 concentration in the VP (68wt%), results relative to A-type NTs shown in Figures 1(a), 2(a)–2(d), 3(a), and 4(a) prove that, in the range of tF examined (0.5–5.0 h), the best compromise among abundance of oxygenated functionalities (i.e., Dm), selectivity towards carboxylic groups (i.e., DmCOOH) and tube integrity is achieved with the 2 h long treatment. In fact, YF is positive, ID/IG moderately increases, while IG0 /IG remains nearly constant (Figure 3(a)) with respect to pristine NTs. Desorption of oxygenated groups under annealing up to 1050 C produces a mass loss of 22.7wt%, 12.1wt% of which occurs below 400 C, i.e., in the

Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Oxidized by HNO3 Vapors range where the degradation of carboxylic groups takes place (21). No gain is achieved by prolonging VPF up to 5.0 h because tubes partly break (Figures 2(c)–2(d)), YF becomes negative, ID/IG markedly increases (Figure 3(a)), and Dm and DmCOOH reduce (down to 21.4wt% and 11.4wt%, respectively). For given tF and wV values, the use of ternary solution, which involves safer operating conditions with respect to the binary one (as wL lowers), leads to quite similar results. In fact, A-type NTs functionalised for 2.0 h with 68wt% acid concentration in vapor (i.e., samples AB2 and AT2) exhibit no appreciable morphologic difference, comparable defectiveness (Figure 3(b)) and desorption properties (Figure 1(b)).

Downloaded by [Candida Milone] at 14:24 27 August 2014

Effect of Pristine Nanotube Structure and Acid Vapor Concentration Under these conditions, functionalisation with ternary solution results to be more effective on A-type rather than on B-type NTs. Dm increases from 4.0wt% in sample AP to 23.7wt% in sample AT2, and from 7.0wt% in sample BP to 21.3wt% in sample BT2, with an increment with respect to pristine samples of 19.7wt% for NTs of type A and of only 14.3wt% in those of type B, originally featured by higher sp2 defect density (Figure 3(b)). The reason for this behavior might lie in the larger (highly reactive (38)) sp3 component associated to smaller NT diameters of AP sample. Also selectivity towards the formation of carboxylic groups worsens. DmCOOH increases from 0.9wt% in samples AP and BP to 11.4wt% in sample AT2 and to 8.1wt% in sample BT2, thus undergoing a larger increment in NTs of type A than in those of type B (10.5wt% against 7.2wt%). For fixed treatment duration (2.0 h), with increasing acid concentration in the VP from 25wt% to 93wt% the efficiency of VPF process improves (Figure 1(c)). Dm increases from 16.8wt% in sample AT1 to 27.9wt% in sample AT3. This corresponds to a linear increment with wV relative to the value (4.0wt%) measured in pristine sample (AP). An enhancement of the structural disorder level (Figure 3(b)) accompanies the increase of oxygenated functionalities formed on sidewalls and ends of the tubes.

89

The comparison between the spectra of NTs functionalised under harsher conditions (i.e., sample BT3 in Figure 3(a) and A T3 in Figure 3(c)) confirms that larger damage to the carbonaceous network is produced prolonging duration up 5 h than enhancing concentration of acid up to 93wt%, as observed by HRTEM. Conversely, a larger loss of selectivity seems to occur for wV D 93wt% than for tF D 5.0 h (compare DmCOOH and Dm of samples AB3 and AT3). Thermal Stability in Air of Functionalised Nanotubes Good correlation is found among results obtained by TG, DTG, and MRS (Figure 5(a)). The plot of TDTG as a function of Dm, shown in Figure 5(b), demonstrates that the thermal stability in air of f-NTs primarily depends on the amount of surface functionalities introduced, while the diverse typology of the oxygenated groups are responsible for details of the variation of TDTG with Dm. In particular, the less (more) than linear decrement of TDTG observed in correspondence of smaller (larger) Dm values hints at a greater (lesser) stability in air of the NTs treated upon milder (harsher) conditions. ID/IG ratio is commonly used to monitor the defect density (39,38,41,42), its reciprocal (IG/ID) is taken as graphitization index (39). The plot IG/ID of as a function of TDTG, shown in Figure 5(c), demonstrates that the graphitization degree of A-type NTs decreases linearly with decreasing thermal stability in air. The greater the amount of oxygenated moieties anchored on the NT surface, the larger the deviation from aromaticity of the carbon network, and the easier the oxidation of f-NTs in air results, in accord with literature (39). The only exception is the sample (AB3) damaged to the largest extent (Figure 2(c)–(d)). G-band weakening due to tube shortening might account for the different behavior of this sample. Quantification and Speciation of Oxygenated Moieties Thermal decomposition of oxygenated moieties leads to the release of CO2 (Figure 6(a) and (d)) and CO (Figure 6(b) and (e)), while thermal degradation of nitrogen-containing groups results in evolution of NO (Figure 6(c) and (f)). The different

Fig. 5. (a) Correlation among results obtained by TG, DTG, and MRS. (b) Dependence on mass loss (Dm) due to desorption of all oxygenated species of the NT resistance to oxidation (TDTG) as assessed by means of TDG. (c) Dependence on TDTG of the graphitization degree (of A-type NTs) as monitored by the G/D intensity ratio (IG/ID).

Downloaded by [Candida Milone] at 14:24 27 August 2014

90

Santangelo et al.

Fig. 6. (a–f) TPD profiles relative to A-type NTs, (g) concentration of CO2, CO and NO released, and (h) CO/CO2 molar ratio. The changes in concentration of oxygenated groups with wV (i) are also shown.

thermal stability of the various oxygenated functionalities allows for their identification via TPD measurements (39,36,46). In fact, the decomposition of carboxylic acids and lactones, resulting in evolution of CO2, occurs at 150–450 C and 600–800 C, respectively; phenols and quinones/carbonyl decompose respectively at 560–750 C and 700–950 C, giving rise to release of CO; degradation of carboxylic anhydrides takes place at 350–650 C with release of CO and CO2 (36,39,46) to equal extent. From results of TPD measurements on A-type NTs functionalised at diverse tF (Figure 6(a)–(c)) and wV (Figure 6(d)–(f)) the concentrations of gases released (Figure 6(g)) and the CO/CO2 molar ratio (Figure 6(h)) are estimated. CO2 and CO mainly evolve, as usually reported (36,39). CO/ CO2 molar ratio largely exceeds 1 for sample AW, while it is less than 1 for all the acid treated samples. From profiles of Figure 6(b) it is argued that the main contribution to CO evolution is from quinones/carbonyl groups, while anhydrides are nearly absent. Therefore, at lower temperatures carboxylic acid groups prevailingly contribute to CO2 evolution (Figure 6(a), while lactones are responsible

for the contribution at higher temperatures. For tF  2.0 h, the formation of all oxygenated groups increases with increasing tF. Instead, for tF D 5.0 h a trend inversion occurs, which might be related to the total oxidation of carbon to CO2 under the harshest conditions in line with the structural modification observed by HRTEM (Figure 2(c)–(d)). This hypothesis is consistent with the drastic lowering of YF (Table 1). In addition, the amount of carboxylic acids decreases (Figure 6(a)) and that of quinones/carbonyl groups drops (Figure 6(b)), so as the VPF process selectivity is preserved and the CO/CO2 molar ratio approaches zero (Figure 6(h)). From the analysis of TPD profiles it further emerges that, at fixed wV (68wt%), prolonging the treatment favors both physisorption of nitrogen oxide and formation of chemically bonded nitrogen groups (lower and higher temperature peaks (47) in NO evolution profiles of Figure 6(c), respectively). Different from the increase of tF, the increase of wV does not produce appreciable changes in NO concentration (Figure 6(g)). Moreover, only the lower temperature contribution in NO evolution profiles is present indicating that, under the

Downloaded by [Candida Milone] at 14:24 27 August 2014

Surface Chemistry and Thermal Stability in Air of Carbon Nanotubes Oxidized by HNO3 Vapors latter conditions, HNO3 vapors are not able to nitrate the NTs but only to cause their oxidation, as previously reported by Cataldo et al (14) for the mixture of nitrogen dioxide/ nitrogen tetroxide. A possible reason for this behavior is the absence of NO2C ions (formed in presence of fuming nitric acid or concentrated H2SO4) that act as effective electrophiles and attack the p electron network in the aromatic rings of the NTs leading to nitration (28) and subsequent destruction by oxidation. Increasing wV at fixed tF (2.0 h) brings about a monotonic increase of CO2 and CO evolution (Figure 6(g)). TPD profiles (Figure 6(d)–(e)) reveal that the linear increase in the amount of carboxylic acid groups (33) is accompanied by a greater increment of the other oxygenated moieties (Figure 6 (i)), so as the selectivity decreases in line with indications coming from TG analysis (Figure 1(c)). Finally, treating NTs with water vapor mainly produces CO releasing functional groups according to typical carbon/ steam surface reaction C(s)CH2O(v)!C(O)CH2.

Conclusion NTs are functionalised by exposure to HNO3 vapors generated by using the conventional azeotropic HNO3CH2O solution and the alternative sub-azeotropic HNO3CH2OCMg (NO3)2 solution, recently proposed to render safer and more flexible the process. The NT functionalisation degree is varied by tuning the treatment duration (0.5–5.0 h) and the HNO3 concentration in the VP (25–93wt%) in the two cases, respectively. Selectivity of the process towards the formation of carboxylic groups and thermal stability in air of the functionalised materials are investigated. In order to thoroughly characterize pristine and oxidized NTs a systematic analysis is conducted by means of HRTEM, MRS, TG, DTG, and TPD. As for the conventional treatment with vapors at azeotropic HNO3 concentration, results show that the best compromise among abundance of oxygenated functionalities, selectivity towards carboxylic groups and tube integrity is achieved for a duration of 2 h. Prolonging the treatment up to 5.0 h leads to tube breaking, yield diminishing and nitrogen group formation increase. Instead, with the alternative method, no important damage to the carbonaceous network and yield reduction accompanies the linear increase of carboxylic groups produced by changing the acid concentration in the VP. Selectivity towards the formation of carboxylic groups slightly reduces at higher functionalisation degrees, but the formation of nitrogen containing groups remains limited. Regardless of the solution utilized to generate nitric acid vapors, the greater the amount of oxygenated moieties anchored on the NT surface, the larger the deviation from aromaticity of the carbon network, and the lower the resistance of f-NTs to the oxidation in air results. Nonetheless, f-NTs with the highest –COOH group concentration (256 mmol/g) are thermally stable up to 555 C. Amount and typology of native structural defects present in NTs seem to be influential on efficiency and selectivity of

91

the functionalisationprocess. However, this topic still requires deeper investigation.

Funding This work was partly financed by Progetto Operativo Nazionale Ricerca e Competitivita 2007–2013 (PON 01_01869) Tecnologie e Materiali Innovativi per la Difesa del Territorio e la Tutela dell’Ambiente (TEMADITUTELA).

References 1. Burchell, T.D. (ed.). (1999) Carbon Materials for Advanced Technologies. Oxford: Elsevier Science Ltd. 2. Carbon: The Future Material for Advanced Technology Applications. (2006) Messina, G. and Santangelo, S. (eds.), Topics in Applied Physics, Springer: Heidelberg, Vol. 100. 3. Baughman, R.H., Zakhidov, A.A., and de Heer, W.A. (2002) Carbon nanotubes-the route toward applications. Science, 297: 787– 792. 4. Geim, A.K. (2009) Graphene: Status and prospects. Science, 324: 1530–1534. 5. Yang, W., Ratinac, K.R., Ringer, S.P., Thordarson, P., Gooding, J.J., and Braet, F. (2010) Carbon nanomaterials in biosensors: should you use nanotubes or graphene? Angew. Chem. Int. Ed., 49: 2114–2138. 6. Carbon Nanotubes: Advanced Topics in the Synthesis, Structure, Properties and Applications. (2008) Jorio, A., Dresselhaus, G. and Dresselhaus, M.S. (eds.) Topics in Applied Physics, SpringerVerlag: Berlin Heidelberg, Vol. 111. 7. Carbon Nanotube Electronics (2009) Javey, A. and Kong, J. (eds.), Springer: Berlin. 8. Carbon Nanotube Based Nanocomposites: Recent Development (2011) Kar, K.K. and Hodzic, A. (eds.), Research Publishing Services: Singapore. 9. DeLuca, L., Donato, A., Santangelo, S., Faggio, G., Messina, G., Donato, N., and Neri, G. (2012) Hydrogen sensing characteristics of Pt/TiO2/MWCNTs composites. Int. J. Hydrogen En., 37: 1842– 1851. 10. Kuila, T., Bose, S., Mishra, A.K., Khanra, P., Kim, N.H., and Lee, J.H. (2012) Chemical functionalization of graphene and its applications. Prog. in Mater. Sci., 57: 1061–1105. 11. Burghard, M. and Balasubramanian, K. (2005) Chemically functionalized carbon nanotubes. Small, 1: 180–192. 12. Cataldo, F. (2008) A study on the action of ozone on multiwall carbon nanotubes. Fullerenes, Nanotubes, and Carbon Nanostructures, 16: 1–17. 13. Byl, O., Liu, J., and Yates Jr., J.T. (2005) Etching of carbon nanotubes by ozone – A surface area study. Langmuir, 21: 4200–4204. 14. Cataldo, F., Compagnini, G., D’Urso, L., Mita, V., Strazzulla, G., Ursini, O., and Angelini, G. (2008) Adsorption of dinitrogen tetroxide (N2O4) on multi-walled carbon nanotubes (MWCNTs). Fullerenes, Nanotubes, and Carbon Nanostructures, 16: 154–164. 15. Bubert, H., Haiber, S., Brandl, W., Marginean, G., Heintze, M., and Br€ user, V. (2003) Characterization of the uppermost layer of plasma-treated carbon nanotubes. Diam. Rel.Mater., 12: 811–815. 16. Okpalugo, T. I. T., Papakonstantinou, P., Murphy, H., McLaughlin, J., and Brown, N. M. D. (2005) High resolution XPS characterization of chemical functionalised MWCNTs and SWCNTs. Carbon, 43: 153–161. 17. Murphy, H., Papakonstantinou, P., and Okpalugo, T. I. T. (2006) Raman study of multiwalled carbon nanotubes functionalized with oxygen groups. J. Vac. Sci. Technol. B, 24: 715–720.

Downloaded by [Candida Milone] at 14:24 27 August 2014

92 18. Osswald, S., Havel, M., and Gogotsi, Y. (2007) Monitoring oxidation of multiwalled carbon nanotubes by Raman spectroscopy. J. Raman Spectrosc., 38: 728–736. 19. Rasheed, A., Howe, J.Y., Dadmun, M.D., and Britt, P.F. (2007) The efficiency of the oxidation of carbon nanofibers with various oxidizing agents. Carbon, 45: 1072–1080. 20. Shen, J., Huang, W., Wu, L., Hu, Y., and Ye, M. (2007) Study on amino-functionalized multiwalled carbon nanotubes. Mater. Sci. Eng. A, 464: 151–156. 21. Datsyuk, V., Kalyva, M., Papagelis, K., Parthenios, J., Tasis, D., Siokou, A., Kallitsis, I., and Galiotis, C. (2008) Chemical oxidation of multiwalled carbon nanotubes. Carbon, 46: 833–840. 22. Wu, G., Zhou, L., Ou, E., Xie, Y., Xiong, Y., and Xu, W. (2010) Preparation and properties of hydroxylated styrene–butadiene–styrene tri-block copolymer/multi-walled carbon nanotubes nanocomposites via covalent bond. Mater. Sci. Eng. A, 527: 5280–5286. 23. Neelgund, G.M., Olurode, K., Luo, Z., and Oki, A. (2011) A simple and rapid method to graft hydroxyapatite on carbon nanotubes. Mater. Sci. Eng. C, 31: 1477–1481. 24. Hussain, S., Jha, P., Chouksey, A., Raman, R., Islam, S.S., Islam, T., and Choudhary, P.K. (2011) Spectroscopic investigation of modified single wall carbon nanotube. J. Mod. Phys., 2: 538–543. 25. Gong, H., Kim, S.T., Lee, J.D., and Yim, S. (2013) Simple quantification of surface carboxylic acids on chemically oxidized multiwalled carbon nanotubes. Appl. Surf. Sci., 266: 219–224. 26. Xia, W., Hagen, V., Kundu, S., Wang, Y., Somsen, C., Eggeler, G., Sun, G., Grundmeier, G., Stratmann, M., and Muhler, M. (2007) Controlled etching of carbon nanotubes by iron-catalyzed steam gasification. Adv. Mater., 19: 3648–3652. 27. Xia, W., Jin, C., Kundu, S., and Muhler, M. (2009) A highly efficient gas-phase route for the oxygen functionalization of carbon nanotubes based on nitric acid vapor. Carbon, 47: 919–922. 28. Kitamura, H., Sekido, M., Takeuchi, H., and Ohno M. (2011) The method for surface functionalization of single-walled carbon nanotubes with fuming nitric acid. Carbon, 49: 3851–3856. 29. Rosca, I.D., Watari, F., Uo, M., and Akasaka, T. (2005) Oxidation of multiwalled carbon nanotubes by nitric acid. Carbon, 43: 3124– 3131. 30. Cataldo, F., Compagnini, G., Patane, G., Ursini, O., Angelini, G., Ribic, P.R., Margaritondo, G., Cricenti, A., Palleschi, G., and Valentini, F. (2010) Graphene nanoribbons produced by the oxidative unzipping of single-wallcarbon nanotubes. Carbon, 48: 2596–2602. 32. Terrones, M. (2009) Materials science: Nanotubes unzipped. Nature, 458: 845–846. 31. Cataldo, F., Compagnini, G., D’Urso, L., Palleschi, G., Valentini, F., Angelini, G., and Braun T. (2010) Characterization of graphenenanoribbons from the unzipping of MWCNTs. Fullerenes, Nanotubes, and Carbon Nanostructures, 18: 261–272. 33. Milone, C., Piperopoulos, E., Ansari, S., Faggio, G., and Santangelo, S. (2013) Highly versatile and efficient process for CNT oxidation in vapor phase by means of Mg(NO3)2-HNO3-H2O ternary

Santangelo et al.

34. 35.

36.

37.

38.

39.

40.

41.

42.

43. 44.

45.

46. 47.

mixture. Fullerenes, Nanotubes, and Carbon Nanostructures, DOI: 10.1080/1536383X.2013.858132. Lei, Z., Li, C., and Chen, B. (2003) Extractive distillation: A review. Sep. and Purif. Rev., 32: 121–213. Shneerson, A.L., Miniovich, M.A., Filippova, Zh. M., Soroka, S.N., and Platanov, P.A. (1965) Liquid vapour equilibrium in the systems nitric acid-water-magnesium nitrate, nitric acid-watercalcium nitrate and nitric acid-magnesium nitrate-calcium nitrate. Russ. J. Phys. Chem., 39: 744–747. Kundu, S., Wang, Y., Xia, W., and Muhler, M. (2008) On the thermal stability and reducibility of oxygen-containing functional groups on multi-walled carbon nanotube surfaces: a quantitative high-resolution XPS and TPD/TPR study. J. Phys. Chem. C, 112: 16869–16878. Fazio, E., Piperopoulos, E., Abdul Rahim, S.H., Lanza, M., Faggio, G., Mondio, G., Neri, F., Mezzasalma, A.M., Milone, C., and Santangelo, S. (2013) Correlation between carbon nanotube microstructure and their catalytic efficiency towards the p-coumaric acid degradation. Curr. Appl. Phys., 13: 748–752. Santangelo, S., Lanza, M., and Milone, C. (2013) Evaluation of the overall crystalline quality of amorphous carbon containing multiwalled nanotubes. J. Phys. Chem. C, 117: 4815–4823. Santangelo, S., Messina, G., Faggio, G., Abdul Rahim, S.H., and Milone, C. (2012) Effect of sulphuric-nitric acid mixture composition on surface chemistry and structural evolution of liquid-phase oxidised carbon nanotubes. J. Raman Spectr., 43: 1432–1442.  ao, J.J.M., and Pereira, M.F. Goncalves, A.G., Figueiredo, J.L., Orf~ R. (2010) Influence of the surface chemistry of multi-walled carbon nanotubes on their activity as ozonation catalysts. Carbon, 48: 4369–4381. Ferrari, A.C., and Robertson, J. (2001) Interpretation of Raman spectra of disordered and amorphous carbon. Phys. Rev. B, 61: 14095–14107. Malard, L.M., Pimenta, M.A., Dresselhaus, G., and Dresselhaus, M.S. (2009) Raman spectroscopy in graphene. Phys. Rep., 473: 51–87. Dresselhaus, M.S., Dresselhaus, G., Saito, R., and Jorio, A. (2005) Raman spectroscopy of carbon nanotubes. Phys. Rep., 409: 47–99. Bae, J.H., Shanmugharaj, A.M., Noh, W.H., Choi, W.S., and Ryu, S.H. (2007) Surface chemical functionalized single-walled carbon nanotube with anchored phenol structures: Physical and chemical characterization. Appl. Surf. Sci., 253: 4150–4155. Lehman, J.H., Terrones, M., Mansfield, E., Hurst, K.E., and Meunier V. (2011) Evaluating the characteristics of multiwall carbon nanotubes. Carbon, 49: 2581–2602. Figueiredo, J.L. and Pereira, M.F.R. (2010) The role of surface chemistry in catalysis with carbons. Catal. Today, 150: 2–7.  atkowski, Szyma nski, G.S., Karpi nski, Z., Biniak S., and Swi ˛ A. (2002) The effect of the gradual thermal decomposition of surface oxygen species on the chemical and catalytic properties of oxidized activated carbon. Carbon, 40: 2627–2639.