Magnetism for understanding catalyst analysis of purified carbon nanotubes

https://doi.org/10.1016/j.jmmm.2016.03.056Get rights and content

Highlights

  • Cl-gas treatment of Ni catalyst of carbon nanotubes leads to NiCl2 residue.

  • Magnetic measurements show the transformation of Ni0 in Ni2+through a purification process.

  • High temperature Cl treatment removes 75% of metallic impurities.

  • Cl-purification yields to an amount of metal of 1.5% in arc-discharge CNT samples.

Abstract

The precise quantification of catalyst residues in purified carbon nanotubes is often a major issue in view of any fundamental and/or applicative studies. More importantly, since the best CNTs are successfully grown with magnetic catalysts, their quantification becomes strictly necessary to better understand intrinsic properties of CNT. For these reasons, we have deeply analyzed the catalyst content remained in nickel–yttrium arc-discharge single walled carbon nanotubes purified by both a chlorine-gas phase and a standard acid-based treatment. The study focuses on Ni analysis which has been investigated by transmission electron microscopy, X-ray diffraction, thermogravimetry analysis, and magnetic measurements. In the case of the acid-based treatment, all quantifications result in a decrease of the nanocrystallized Ni by a factor of two. In the case of the halogen gas treatment, analysis and quantification of Ni content is less straightforward: a huge difference appears between X-ray diffraction and thermogravimetry results. Thanks to magnetic measurements, this disagreement is explained by the presence of Ni2+ ions, belonging to NiCl2 formed during the Cl-based purification process. In particular, NiCl2 compound appears under different magnetic/crystalline phases: paramagnetic or diamagnetic, or well intercalated in between carbon sheets with an ordered magnetic phase at low temperature.

Introduction

Carbon nanotubes (CNTs) and especially single walled carbon nanotubes (SWNTs) have remarkable intrinsic physical and chemical properties that make them as excellent candidates for fundamental studies and applications in many fields [1], [2]. However, whatever the synthesis methods, as-produced CNT samples contain both secondary carbon products and metallic-based impurities involved in the growth mechanism as catalyst. They are often used in high content (10–30 wt%) in order to improve the nanotube formation and limit that of carbon byproducts. These metal-based impurities are reported to dramatically modify both chemical and physical properties of CNT [3], [4], [5], [6], [7]. Without any post-synthesis purification treatment, SWNTs become unusable because their intrinsic properties are lost or hidden due to the presence of these metallic particles. For this very reason, the development of purification treatments is a long-standing and active research area. Standard processes are often multi-step and generally combine (i) a reflux in concentrated nitric acid solution in order to weaken the carbon shells and oxidize the metal-based impurities (ii) a gas-phase or wet oxidation and (iii) a final annealing treatment to restore the defects created through previous steps [8], [9], [10], [11], [12], [13], [14], [15], [16]. More recently, high temperature treatment under halogens were evidenced as an interesting alternative technique to selectively remove metal-based impurities [17], [18]. The assessment of the efficiency of such treatment indeed depends on the reliability of quantitative analysis of the remaining impurities. On one hand, local or surface analysis such as X-ray photoelectron spectroscopy or transmission electron microscopy (TEM) are used to determine their size and/or their chemical nature, however they are inefficient for a quantitative analysis [19]. On the other hand, macroscopic techniques such as thermogravimetry analysis (TGA) provide a quantitative analysis but without any discrimination regarding the chemical nature of the impurities in the purified samples. Among them, magnetic analysis (MAG) is a less commonly used technique [20], [21], [22]. It has been however shown to be very helpful to quantify magnetic residues from CNT samples [23], [24], [25]; especially because it is an ultrasensitive and non destructive method which can even detect very low amount of isolated paramagnetic impurities [19].

In this work, we have applied a Cl-based purification treatment on arc-discharged as-produced SWNTs under different temperature conditions. For comparison, a more standard purification method, based on an acid treatment has also been applied. Analysis and quantification of Ni-based impurities have been investigated by TEM, X-rays diffraction (XRD), TGA and MAG. A good agreement between TGA and MAG is found for the acid treatment whereas a strong deviation is observed between the results obtained from the different used techniques in the case of the Cl-based treatment. Thanks to magnetic measurements, these discrepancies are attributed to oxidation of Ni0 into Ni2+ by the halogen gas. This result proves that chemical treatment used for purification can modify the impurity.

Section snippets

Sample preparation

The SWNTs used in this study have been synthesized in an arc-discharge homemade reactor described elsewhere [16]. The catalyst mixture is prepared with a graphite powder SFG6 (synthesized flake graphite 6 µm) from Timcal, nickel particles of around 3 μm purchased from Sigma Aldrich and yttrium powder (40 mesh) from Acros Organics. Ni/Y/C is fixed at 4.2/1/94.8 at% (2.8/1/96.2 wt%). The as-produced SWNT powder (referred as “raw”) was purified using 2 different treatments: (i) a standard purification

Transmission electron microscopy: CNT and impurity observation

The as-produced SWNT powder (referred as “raw”) was purified using 2 different treatments: (i) a standard purification process that involves three main steps: air oxidation, acid reflux and annealing. The corresponding purified SWNT sample is referred as “sp”. (ii) An alternative halogen-based treatment that mainly consists in heating the SWNT powder under high chlorine partial pressure for 2 h at 800 °C, 950 °C, or 1100 °C [17], respectively assigned to “hp-800”, “hp-950”, “hp-1100”.

Fig. 1 shows

Conclusion

Quantitative analysis of metal-based residues in SWNT samples after chemical purification is an open and tricky question. In this work, we have investigated the commonly used characterization techniques combined with magnetization measurements. This latter allows to determine the exact state of the impurities that is of importance to know the chemical mechanisms involved in the purification treatments. Based on two different purification procedures, the results are discussed regarding the

Acknowledgment

The authors thank Jean-François Marêché for his technical help for the chemical treatments. The authors would like to thank Lionel Aranda for his help for TGA experiments and Pascal Franchetti for his valuable assistance for Raman spectroscopy.

References (37)

  • R. Brukh et al.

    Chem. Phys. Lett.

    (2008)
  • A.V. Ellis et al.

    J. Magn. Magn. Mater.

    (2006)
  • H.G. Cho et al.

    Carbon

    (2009)
  • B. Yu et al.

    Carbon

    (2010)
  • B. Vigolo et al.

    Carbon

    (2010)
  • E. Pellicer et al.

    Carbon

    (2009)
  • S. Osswald et al.

    Chem. Phys. Lett.

    (2005)
  • E. Remy et al.

    Carbon

    (2015)
  • P. Delhaes et al.

    Carbon

    (2006)
  • H. Kataura et al.

    Synth. Met.

    (1999)
  • P.M. Ajayan

    Chem. Rev.

    (1999)
  • H. Dai

    Acc. Chem. Res.

    (2002)
  • I.W. Chiang et al.

    J. Phys. Chem. B

    (2001)
  • J.P. Salvetat et al.

    Phys. Rev. B

    (2005)
  • M. Pumera et al.

    Nanoscale

    (2009)
  • G. Charron et al.

    New J. Chem.

    (2009)
  • A.C. Dillon et al.

    Adv. Mater.

    (1999)
  • M.T. Martinez et al.

    Carbon

    (2003)
  • Cited by (6)

    • Purification of carbon nanotubes produced by the electric arc-discharge method

      2021, Surfaces and Interfaces
      Citation Excerpt :

      A large number of purification methods have been reported in the literature [28, 50–53]. In general, these methods involve several processing steps that can combine: oxidation in gas or liquid phase (reflux in a solution containing oxidizing agents), physical separation processes such as filtration, centrifugation, chromatography, or selective functionalization methods [40, 50, 54–59]. Oxidative steps have been widely used in CNTs purification protocols for eliminating both unliked metal particles and carbons.

    • Transition metal impurities in carbon-based materials: Pitfalls, artifacts and deleterious effects

      2020, Carbon
      Citation Excerpt :

      This distribution shows directly which fraction of the metallic nanoparticles has been removed [200]. Vigolo et al. studied the Ni catalyst content in arc-discharge SWCNTs purified by two techniques, a Cl2 treatment at 800–1100 °C and a standard method including air oxidation at 350 °C followed by HCl reflux and an annealing under vacuum at 1100 °C using MAG techniques [201]. Thanks to magnetic analysis, they proved that a particular purification process can significantly alter the chemical state of a metallic impurity.

    • Dramatic enhancement of double-walled carbon nanotube quality through a one-pot tunable purification method

      2016, Carbon
      Citation Excerpt :

      Pure gas phase treatments can be preferred but nevertheless they have been poorly studied. They are either based on high temperature annealing [22,23], or assisted by halogen gases [24–29]. The advantages of these gas-phase methods are several, they are simple (one-step), they do not introduce defects in the CNT structure but they fail to remove carbon impurities.

    View full text