Elaine Lay Khim Chng

The Toxicity of Graphene Oxides: Dependence on the Oxidative Methods Used

Graphene, a class of two-dimensional carbon nanomaterial, has attracted extensive interest in recent years, with a significant amount of research focusing on graphene oxides (GOs). They have been primed as potential candidates for biomedical applications such as cell labeling and drug delivery, thus the toxicity and behavior of graphene oxides in biological systems are fundamental issues that need urgent attention. The production of GO is generally achieved through a top-down route, which includes the usage of concentrated H₂SO₄ along with: 1) concentrated nitric acid and KClO₃ oxidant (Hoffmann); 2) fuming nitric acid and KClO₃ oxidant (Staudenmaier); 3) concentrated phosphoric acid with KMnO₄ (Tour); or 4) sodium nitrate for in-situ production of nitric acid in the presence of KMnO₄ (Hummers). It has been widely assumed that the properties of these four GOs produced by using the above different methods are roughly similar, so the methods have been used interchangeably. However, several studies have reported that the toxicity of graphene-related nanomaterials in biological systems may be influenced by their physiochemical properties, such as surface functional groups and structural defects. In addition, considering how GOs are increasingly used in the field of biomedicine, we are interested to see how the oxygen content/functional groups of GOs can impact their toxicological profiles. Since in-vitro testing is a common first step in assessing the health risks related with engineered nanomaterials, the cytotoxicity of the GOs prepared by the four different oxidative treatments was investigated by measuring the mitochondrial activity in adherent lung epithelial cells (A549) by using commercially available viability assays. The dose-response data was generated by using two assays, the methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay and the water-soluble tetrazolium salt (WST-8). From the viability data, it is evident that there is a strong dose-dependent cytotoxic response resulting from the four GO nanomaterials tested after a 24 h exposure, and it is suggested that there is a correlation between the amounts of oxygen content/functional groups of GOs with their toxicological behavior towards the A549 cells.

Hydrogen-bonding interactions between water and the one- and two-electron-reduced forms of vitamin K1: applying quinone electrochemistry to determine the moisture content of non-aqueous solvents.

Vitamin K(1) (VK(1)) was shown by voltammetry and coulometry to undergo two chemically reversible one-electron reduction processes in acetonitrile (CH(3)CN) containing 0.2 M Bu(4)NPF(6) as the supporting electrolyte. The potential separation between the first and second electron-transfer steps diminished sequentially with the addition of water, so that at a H(2)O concentration of approximately 7 M (approximately 13% v/v) only one process was detected, corresponding to the reversible transfer of two electrons per molecule. The voltammetric behavior was interpreted on the basis of the degree of hydrogen bonding between the reduced forms of VK(1) with water in the solvent. It was found that the potential separation between the first and second processes was especially sensitive to water in the low molar levels (0.001-0.1 M); therefore, by measuring the peak separation as a function of controlled water concentrations (accurately determined by Karl Fischer coulometric titrations) it was possible to prepare calibration curves of peak separation versus water concentration. The calibration procedure is independent of the type of reference electrode and can be used to determine the water content of CH(3)CN between 0.01 and 5 M, by performing a single voltammetric scan in the presence of 1.0 mM VK(1). The voltammetry was also investigated in dichloromethane, dimethylformamide, and dimethyl sulfoxide. The reduction processes were monitored by in situ electrochemical UV-vis spectroscopy in CH(3)CN over a range of water concentrations (0.05-10 M) to spectroscopically identify the hydrogen-bonded species.

Solid-state electrochemistry of graphene oxides: absolute quantification of reducible groups using voltammetry.

The quantification of various oxygen-containing groups in graphene oxides is of very high importance as these groups have strong influences on the physical, chemical, and material properties of graphene oxides. All current methods provide only relative quantification. Herein, we present a powerful electrochemical technique for the absolute quantification of selected oxygen-containing groups on graphene oxide. Since its discovery in 2004, graphene has captured the imaginations of many scientists. Likewise, graphene-related materials have also attracted the attention of chemists, physicists, and materials scientists alike owing to their interesting properties. Graphene-related materials can be prepared by various methods. The mainstream methods mainly consist of 1) mechanical exfoliation of graphite, 2) Chemical-vapour deposition (CVD) growth of graphene, and 3) chemical oxidation of graphite to graphite oxide followed by consequent exfoliation and modifications leading to the formation of various types of graphene oxides. The third preparation procedure is the most-popular method used for the manufacturing of bulk quantities of graphenerelated materials. This procedure typically involves the oxidation of graphite by a mixture of nitric acid and strong oxidants (i.e. KClO3 or KMnO4), which yields graphite oxide. The structure of graphite oxide consists of graphite layers with significant increment in their interlayer spacing owing to the presence of functional groups on the basal planes of the individual graphene sheets that are introduced through oxidative treatment. The functional groups are a mixture of various oxygen-containing groups (C/O ratio of graphite oxide is ca. 1.9:1), such as hydroxy, carbonyl, aldehyde, carboxy, epoxide, peroxy, ether, and ester groups. Consequently, graphite oxide is exfoliated by thermal or sono-exfoliation to give graphene oxide (GO). With the use of reducing agents such as hydrazine or sodium tetrahydroborate, GO is often reduced to yield “graphene”. Such chemically reduced graphene possesses different structural and chemical features when compared to the pristine graphene obtained by methods (1) or (2). To distinguish pristine graphene from graphene oxide and chemically reduced graphene oxide (CRGO), graphene oxide and CRGO are generally termed as chemically modified graphenes. The exact quantification of various types of oxygen-containing groups on graphene-related materials is a very challenging issue. Whilst the total amount of oxygen in graphene oxides is easily accessible by chemical analysis, the absolute quantification of these particular oxygen-containing groups, such as hydroxy, aldehyde, carbonyl, carboxy, peroxy, ether, or ester groups is currently practically impossible. Spectroscopic methods are often used in an attempt to elucidate the relative ratio of the particular types of oxygencontaining groups. However, the most widely used spectroscopic methods, such as the Fourier transform infrared spectroscopy (FTIR) or X-ray photoelectron spectroscopy (XPS), suffer from the fact that graphene is actually a heterogeneous material where there is no homogenous distribution of these oxygen-containing groups on its surface. This problem is worsened by the fact that graphene oxides are capable of forming multilayer structures when their suspensions in solutions are dried prior to spectroscopic analyses. In addition, the FTIR and XPS data are often misinterpreted as the vibration/rotation modes of several functional groups overlap each other (in the case of FTIR) or their binding energies are too close that artificial deconvolution is needed (in the case of XPS). This misinterpretation resulted in the assignment of the signals that belong to the same energy levels to different groups, in different journals (for FTIR and XPS, see Ref. [14] and Ref. [13], respectively, and references within). This confusion consequently raises doubts regarding the reliability of these spectroscopic techniques in the identification and quantification of specific oxygen-containing groups. In addition, because there is no standard material at the moment, all of these methods can [a] E. L. K. Chng, Prof. M. Pumera Division of Chemistry & Biological Chemistry School of Physical and Mathematical Sciences Nanyang Technological University Singapore 637371 (Singapore) Fax: (+65) 6791-1961 E-mail : pumera@ntu.edu.sg

Toxicity of graphene related materials and transition metal dichalcogenides

The dramatic rise in the development and application of graphene related materials (graphene, graphene oxide and reduced graphene oxide) as well as of nanosized layered transition metal dichalcogenides gives a strong incentive to study the toxicity of these nanomaterials. It was found that size, surface area, shape, number of layers and amount and type of oxygen containing groups strongly influence toxicity of the nanomaterials. Important toxicity studies are reviewed here with a focus on the above mentioned materials.

Cytotoxicity profile of highly hydrogenated graphene

Graphene and its graphene-related counterparts have been considered the future of advanced nanomaterials owing to their exemplary properties. An increase in their potential applications in the biomedical field has led to serious concerns regarding their safety and impact on health. To understand the toxicity profile for a particular type of graphene utilized in a given application, it is important to recognize the differences between the graphene-related components and correlate their cellular toxicity effects to the attributed physiochemical properties. In this study, the cytoxicity effects of highly hydrogenated graphene (HHG) and its graphene oxide (GO) counterpart on the basis of in vitro toxicological assessments are reported and the effects correlated with the physiochemical properties of the tested nanomaterials. Upon 24 h exposure to the nanomaterials, a dose-dependent cellular cytotoxic effect was exhibited and the HHG was observed to be more cytotoxic than its GO control. Detailed characterization revealed an extensive C-H sp(3) network on the carbon backbone of HHG with few oxygen-containing groups, as opposed to the presence of large amounts of oxygen-containing groups on the GO. It is therefore hypothesized that the preferential adsorption of micronutrients on the surface of the HHG nanomaterial by means of hydrophobic interactions resulted in a reduction in the bioavailability of nutrients required for cellular viability. The nanotoxicological profile of highly hydrogenated graphene is assessed for the first time in our study, thereby paving the way for further evaluation of the toxicity risks involved with the utilization of various graphene-related nanomaterials in the real world.

The hydrogen-bonded dianion of vitamin K1 produced in aqueous-organic solutions exists in equilibrium with its hydrogen-bonded semiquinone anion radical.

When the quinone, vitamin K1 (VK1), is electrochemically reduced in aqueous-acetonitrile solutions (CH3CN with 7.22 M H2O), it undergoes a two-electron reduction to form the dianion that is hydrogen-bonded with water [VK1(H2O)y(2–)]. EPR and voltammetry experiments have shown that the persistent existence of the semiquinone anion radical (also hydrogen-bonded with water) [VK1(H2O)x(–•)] in aqueous or organic–aqueous solutions is a result of VK1(H2O)y(2–) undergoing a net homogeneous electron transfer reaction (comproportionation) with VK1, and not via direct one-electron reduction of VK1. When 1 mM solutions of VK1 were electrochemically reduced by two electrons in aqueous-acetonitrile solutions, quantitative EPR experiments indicated that the amount of VK1(H2O)x(–•) produced was up to approximately 35% of all the reduced species. In situ electrochemical ATR-FTIR experiments on sequentially one- and two-electron bulk reduced solutions of VK1 (showing strong absorbances at 1664, 1598, and 1298 cm(–1)) in CH3CN containing <0.05 M H2O led to the detection of VK1(–•) with strong absorbances at 1710, 1703, 1593, 1559, 1492, and 1466 cm(–1) and VK1(H2O)y(2–) with strong absorbances at 1372 and 1342 cm(–1).