Carbon nanomaterials: synthesis and applications to development of electrochemical sensors in determination of drugs and compounds of clinical interest

Abstract It is notorious that researches related to electrochemical sensors increased significantly due the promising characteristics that these devices present such as the possibility of obtaining information, with minimum manipulation of the studied system, in real time, and with low environmental impact. This article covers the carbon nanomaterials, presenting important aspects such as main properties, synthesis methods, and the application of these materials in the development of electrochemical sensors for the analysis of drugs and compounds of clinical interest. In this context, drug analysis is extremely important for quality control, to ensure that the medicine fulfills its role effectively without possible complications that could compromise the patient’s health and quality of life. In addition, analytical methods capable of determining compounds of clinical interest in biological fluids are extremely important for the indication of effective diagnoses. Thus, the versatility, selectivity, and portability of the electroanalytical techniques make the electrochemical sensors a favorite tool for the determination of drugs and compounds of clinical interest. It will be possible to follow in the present work that carbon nanomaterials have excellent thermal and electrical conductivity, strong adsorption capacity, high electrocatalytic effect, high biocompatibility, and high surface area. The possibility of formation of different composite materials based on carbonaceous nanomaterials that makes these materials promising for the development of analytical sensors, contributing to rapid, sensitive, and low-cost analyses can also be highlighted.


Introduction
Drugs have a medical purpose: for relief or treatment and used to alter or exploit physiological systems for the person in whom they are administered. Drug analysis is essential for quality control, which consists of a set of measures aimed at ensuring, at all times, the production of medicines and other products, which meet the standards of identity as content and purity among others (ANVISA, ). In this sense, the analysis of pharmaceutical compounds and of interest to clinical analysis is highly relevant, mainly to the health area. The analysis of compounds of clinical interest makes it possible to indicate a more accurate and effective diagnosis. Both investigations require sensitive, effective, and reliable analytical methods to ensure the quality of the drug and the good health of the patient.
Several analytical methods were used for the quantification of drugs and compounds of clinical interest, with special emphasis on spectroscopic and chromatographic techniques (Ferey et al., 2018;Østergaard, 2018). The literature shows that most of the methodology drugs for drug determination are based on chromatographic techniques. These techniques are well established and accepted by regulatory organizations; however, they have some disadvantages related to the relatively high cost, time of analysis, and pretreatment steps (Sena et al., 2007;Alam et al. 2017a;2017b). However, electroanalytical methods, especially those employing electrochemical sensors, showed to be promising to complement the existing traditional techniques because of their versatility, portability, and possibility of miniaturization of the system, without compromising their sensitivity 2 Carbon nanomaterials Nowadays, the use of different materials in the actions of our day-to-day life, such as driving a car, lighting a lamp, making calls on the cell phone, or watching a TV program became so commonplace and well established that we do not question what our life would be like if these materials were not developed. In general, any substance or mixture of substances, which have properties that make them useful for application in the development of devices, products, and machines, may be classified as materials. In the last years of the 20th century, the emergence of a new class of materials enabled a true scientific and technological revolution, leading to the consolidation of a whole area of knowledge: Nanoscience and Nanotechnology (N & N). This area of knowledge studies materials made up of particles in the order of nanometers, which are now recognized as nanomaterials (Song et al., 2019).
N & N became, by far, the fastest growing area in materials science today, pushing the boundaries of research, reaching industries, and rapidly reaching the general public. The effect of its development can already be seen in the products of our daily life that have had their aspects and properties renewed and improved, from computer chips and cell phones, to beauty products. The great N & N essence consists in the fact that the properties of materials depend not only on their composition and structure but also on their critical size. In this way, when the particles of this material differ below the critical size (in this case on the order of a few nanometers), the properties of the material form the properties observed for it on a macroscopic scale, thus, enabling the preparation of new materials only with the control of the size and shape of its particles (Gogotsi & Presser, 2014).
Among the new materials developed with the advancement of N & N, carbon nanomaterials deserve to be highlighted because they are considered highly versatile. Carbon nanomaterials presented thermal, electrical, chemical, and mechanical characteristics, besides the richness and diversity of carbon, which is the chemical element that considered the symbol of life and is present in thousands of compounds (Schroeder et al., 2019).
The discovery of fullerenes, followed by CNTs and the recent isolation and characterization of graphene in 2004 (Novoselov et al., 2004) provided new possibilities in the chemistry and physics of carbon and made interest in carbon nanomaterials grow still more. Figure 1 presents a short timeline containing information about the year of discovery and the main awards of work related to carbon nanostructures.

Graphene
Graphene (Figure 2A) is the carbon allotrope of the nanomaterials family that was isolated most recently. Before being isolated, graphene was considered a purely theoretical material only to explain the formation of other allotropic forms of carbon, such as graphite, CNTs, and fullerene ( Figure 2) (Syama & Mohanan, 2019). The term graphene came into being in 1962, a year in which this material was actually observed for the first time by German chemists Ulrich Hofmann and Hanns-Peter Boehm (Boehm et al., 1962), who named it graphene because of the graphite junction with the suffix-ene related to the double bonds existing.
However, it was only four decades later in 2004 that graphene was isolated and first characterized by researchers Geim and Novoselov of the University of Manchester, through the mechanical exfoliation of graphene sheets in individual layers of small plates of pyrolytic graphite (Novoselov et al., 2004).
Novoselov, Geim, and coworkers increasingly improved the efficiency of the material obtained, improving its conductivity, and thinning it to the thickness of an atom (Geim & Novoselov, 2007). To the researchers' surprise, the ultrafine material presented not only a structure of flat monolayer of carbon atoms connected by means of a hexagonal organization but also excellent thermal and electrical conductivity, good mechanical resistance, being light, flexible, and presenting a large specific surface area.
All the excellent physical, chemical, mechanical, optical, and electrical properties that graphene presents make it one of the most promising and revolutionary nanomaterials today. However, all the prominence achieved by this material only became possible due to its two-dimensional structure and atomic organization, which is responsible and directly related to the properties that it presents.
Graphene is formed by carbon atoms with sp 2 hybridization in a conjugated π bond system, arranged in monolayers in a two-dimensional structure. The bond length between the carbon atoms in the graphene structure is approximately 1.42 Å. Each atom has an orbital p, oriented perpendicular to the molecular plane, which overlaps forming valence and conduction bands, which provide the phenomenon of planar conduction (Soldano, Mahmood & Dujardin, 2010;Ghany, Elsherif & Handal, 2017).
Some characteristics turn the graphene into a particular material, for example, the electrons of graphene monolayer can move at a velocity of up to 10 6 m s −1 , which turns this material into an excellent electrical conductor (up to 2 × 10 4 S cm −1 ). It possesses an electronic mobility up to 100 times larger than that of silicon (∼2 × 10 5 cm 2 V −1 s −1 ) and thermal conductivity at room temperature up to 5000 W m −1 K −1 (compared to 400 W m −1 K −1 ). Graphene also shows a very high surface area (∼2600 m 2 g −1 ), higher than those observed for materials such as graphite (10 m 2 g −1 ) and CNTs (1300 m 2 g −1 ) (Soldano, Mahmood & Dujardin, 2010;Syama & Mohanan, 2019).
However, the electronic properties of graphene change with the number of layers and with the relative position of atoms in adjacent layers determined by the order of stacking. Therefore, graphene should not be considered as a single material, but rather as a family of compounds that can be formed by a single layer of atoms (monoatomic sheet), to materials formed by two, three, four, etc. Graphene sheets are stacked in an organized way. Thus, the electronic structure of the graphene presents different couplings with the increase in the number of layers up to the limit of 10-20 layers (3D limit) when the material is considered graphite (Geim & Novoselov, 2007).
All these remarkable features and properties turn graphene into a promising material for applications ranging from polymer-composite materials for transistors, electrodes for electrochemical storage, supercapacitors, electrode material in lithium-ion batteries, photo application electrochemical sensors, and the development of electrochemical sensors and biosensors.
In recent years, several graphene-based sensors and biosensors, or graphene-related materials, such as graphene oxide and reduced graphene oxide, were reported in the literature for analyses of compounds from environmental sources. Some examples are the detection of metal ions such as Pb 2+ and Cd 2+ , analyses of pesticides, as well as food analyses, clinical applications, such as in the detection of glucose, ascorbic acid, uric acid, and drugs (Justino et al., 2017). Graphene is extremely sensitive to the changes in the environment to which it is inserted because of its high surface area and specific electronic properties. In this way, any molecular perturbation is easily detected, making this material contribute to the development of sensors that are enhanced and highly sensitive electrochemicals (Xiao-Mei et al., 2011).
In addition, graphene can form composites with other materials such as polymers, nanoparticles, and biomolecules, such as DNA and enzymes, by physical adsorption or chemical bonding. As a result, by combining the thermal, electrical, and mechanical properties of graphene with the properties and functions of several other components, new materials and innovative analytical systems can be created, providing new opportunities for the sustainable progress of chemical analyses.

Carbon nanotubes
CNTs are nanomaterials formed from one or more graphene sheets rolled cylindrically from their own axis, forming tubular structures with a diameter in the nanometer and length range that can range from micrometers to centimeters. The first evidence of the tubular nature of some nano-sized carbon filaments was reported in a Russian article in 1952 (Radushkevich & Lukyanovich, 1952). However, this article is not widely known or cited because of the difficulties of access by Western scientists to the Russian publications at that time because of the Cold War (Monthioux & Kuznetsov, 2006).
In 1978, Wiles and Abrahamson also observed graphite fibers rolled in layers with a hollow center (Wiles & Abrahamson, 1978). However, it was only in 1991, in the work developed by Iijima that these fibers became known as carbon nanotubes (CNT) (Iijima, 1991). Therefore, even if there are previous reports, the scientific community considers Iijima's work to be the first description of the CNT, since a complete structural study was carried out for the first time.
Nowadays, the term "carbon nanotubes" corresponds to a family of compounds that may have different characteristics from each other. Conceptually, CNTs can be divided into two groups: single-walled carbon nanotubes (SWCNTs) ( Figure 3A) and multiple-walled carbon nanotubes (MWCNTs) ( Figure 3B). SWCNTs are made up of only one sheet of graphene and have a diameter ranging from 1 to 5 nm. However, the synthesis methods currently employed to produce only a small fraction of SWCNT increase its cost and make it difficult to be applied on a large scale. MWCNTs are formed by a set of two or more concentrically coiled graphene sheets, which may have diameters of 10-50 nm (Schroeder et al., 2019;Song et al., 2019). In order to form the nanotubes, a sheet of graphene is rolled up so that two sites of its hexagonal network, crystallographically equivalent, coincide. Depending on the way the graphene sheet wraps around, CNTs may have different geometries and properties. In this way, the electrical properties of the CNTs depend on how the hexagons are oriented along the axis of the tube (Terrones, 2003).
In general, the electronic properties of MWCNTs are similar to those observed in SWCNTs because of weak coupling between concentric carbon cylinders. Some studies describe that the electronic transport in these materials occurs in a ballistic way, that is, without spreading, allowing excellent electrical conduction through large extensions of nanotubes (Asadian, Ghalkhani & Shahrokhian, 2019).
CNTs have several attractive and promising properties for various applications because of their structure. The strength of the Csp 2 -Csp 2 bond gives CNTs high chemical and mechanical resistance. However, even presenting one of the most robust structures known, this material denotes high flexibility and can be tensioned and bent, without destroying its structure. In addition, CNTs have a tensile strength, tensile than steel, and excellent thermal and electrical conductivity (Terrones, 2003).
All these properties together in a single class of material consider the CNTs as one of the components most used in the development of N & N. These materials have a potential application in several systems, products, and devices such as, electron emitters for televisions; in composites with polymers, ceramics and metals, diodes, transistors, photovoltaic devices, computer memories, and development of electrochemical sensors and biosensors.
CNTs were employed in the development of highly sensitive electrochemical sensors to determine the most diverse analytes such as gases, toxins, DNA fragments, various biomolecules, drugs, among many others. High electron transfer velocity, high surface area, and the presence of functional groups are characteristics that make the CNTs a desirable material for the development of sensors, which can be modified with several types of species (enzymes, genetic, metal complexes, among others). These modifications can further improve the electron transfer between analyte and electrode, improving response in a short time.

Fullerene
Another example of a carbon allotrope at nanoscale is the fullerenes. Unlike the other allotropes, the fullerenes are in the molecular form, being considered spheroidal nanometric molecules constituted only by carbon atoms.
Since 1966, theoretical calculations demonstrated the possibility of the existence of "cages" formed only by carbon atoms. However, only in 1985 was the existence of fullerenes proven in the work done by Harold W. Kroto, Robert F. Curl, and Richard E. Smalley (Kroto et al., 1985).
The discovery of fullerenes was a result of Kroto's interest in the quantum mechanics study of certain carbon chains (polyines,…C≡C-C≡C...) and of attempting to reproduce interstellar conditions to prove the existence of these large chains of carbon in space. With the help of Curl and Smalley, Kroto performed an experiment in which a graphite plate was subjected to a pulsed high-frequency laser forming agglomerates, which were later analyzed by mass spectrometry (Ariga, Minami & Shrestha, 2016;Silveira et al., 2018).
At the end of this experiment, the formation of molecules with 60 carbon atoms was observed, leading researchers to wonder what molecular structure could give such stability to these molecules. After some suggestions, Kroto had the idea of organizing the atoms in the same form of the geodesic domes idealized by the architect Richard Buckminster Fuller, in which hexagonal faces construct a spherical structure due to the combination with pentagons. In this way, the structure of the C 60 was defined as a polyhedron of 32 faces, being 20 hexagonal and 12 pentagonal, and was baptized as buckminsterfullerene in honor of the American architect (Silveira et al., 2018).
Currently, other stable forms of fullerene were already isolated as, for example, C 20 , C 70 , C 78 , and C 84 . However, in all these forms, C 60 fullerene is the most abundant and most stable, with π bonds between the sixmembered rings and single bonds between the five-membered rings as shown in Figure 4. Through all the studies that were already done on fullerene, one of the most admirable discoveries is the ability of the C 60 molecule to receive from one to six electrons, even though it is already rich in electrons, forming corresponding anions. This fact is only possible because the non-binding LUMO molecular orbitals of the C 60 molecule are at a very low energy level. Other interesting properties that the fullerenes present are good chemical stability, large surface area, high mechanical resistance, and the ability to become a superconductor when combined with alkali metals. These properties make this material promising for several types of applications, such as photovoltaic (Li, Qi & Wang, 2013a, Li et al., 2013b, biomedical applications, transport of antivirals (Bosi et al., 2003), chemotherapeutics, antibiotics, and the development of sensors (Ariga, Minami & Shrestha, 2016).

Synthesis of graphene by mechanical and chemical exfoliation
As previously mentioned, graphite is composed of several layers of graphene sheets bonded by Vander Waals forces. Consequently, graphene can be produced from the graphite after the breakage of these bonds, which can be accomplished by two main methods: mechanical exfoliation or chemical exfoliation.
Mechanical exfoliation is the method that allowed the isolation of graphene for the first time and is used until today. In this method, an adhesive tape is used repeatedly to perform exfoliation of high-purity graphite. Next, some captured flakes are transferred to the surface of a silicon/silicon oxide disk, and it is possible to verify the existence of graphene by an atomic microscope (Novoselov et al., 2004). This approach presents good reliability and provides graphene samples that are practically free of defects. However, its main drawback is that it does not allow the production of graphene on a large scale.
Chemical exfoliation, or chemical reduction, involves the conversion of graphite to graphene by the introduction of functional groups containing oxygen, by reacting graphite with strong oxidizing agents such as sulfuric acid or nitric acid. This reaction forms graphite oxide, which is chemically reduced to graphene using reducing compounds such as hydrazine or sodium borohydride. The chemical exfoliation from graphite to graphene is possible because of the addition of groups such as OH and COOH that by chemical oxidation reduces the forces of Van der Waals, responsible for keeping the graphene sheets in the structure of the graphite, and the separation takes place ( Figure 5). The main advantage of this method is the possibility of obtaining graphene with good uniformity and excellent yield, when compared to the mechanical methods, allowing its use to obtain graphene on a large scale (Gilje et al., 2007;Zhang et al., 2010;Oliveira et al., 2018).

Synthesis of graphene and carbon nanotube by chemical vapor deposition
It is possible to obtain thin films of various materials, such as ceramics, metal, or semiconductors on the surface of different substrates using chemical vapor deposition (CVD). For the preparation of graphene by this method, pyrolysis of precursor compounds, such as hydrocarbons, occurs on the surface of metal catalysts at high temperatures up to the formation of the graphite structure from the previously dissociated carbon atoms (Son & Ham, 2017). Thus, by thermal decomposition of hydrocarbons, or segregation of carbon atoms on the surface of metallic catalysts, such as Ni, Pt, Ru, Ir, for example, followed by a cooling process, carbon species precipitate on the surface of the metal leading to nucleation and growth to form the graphene. The number of the graphene layers formed depends directly on the cooling rate (Zhao & Gou, 2018). This method is schematized in Figure  6. CVD is also one of the most widely used methods for the synthesis of CNTs because it enables large-scale production and is relatively inexpensive. Thus, in the synthesis of CNTs by CVD, carbon precursors (hydrocarbons, alcohols, etc.) are also used in atmospherically controlled kilns with a temperature high enough to provide the energy required to perform the decomposition of the hydrocarbons and form CNTs on the surface of a metal catalyst. The role of the catalyst is fundamental in obtaining the CNTs, as it will guarantee the process of nucleation and growth of the bundles of nanotubes. The advancement of characterization techniques (mainly electron microscopy techniques) enabled a better understanding of the growth stages of CNTs and, consequently, better control of the homogeneity of the prepared samples, as well as the control in the production of SWCNTs or MWCNTs (Shah & Tali, 2016).

Synthesis of fullerene by electric arc discharge
Even with the discovery of fullerenes in 1985, the development of a method for the synthesis of C60 in macroscopic quantities occurred only in 1990 with the work done by Krätschmer and collaborators, making it possible to obtain sufficient quantities so that the chemical properties of fullerenes could begin to be explored (Krätschmer et al., 1990). The method developed by Krätschmer and collaborators was very similar to the electric arc discharge method, which was also used by Iijima during the first synthesis of CNT (Iijima, 1991).
The arc-discharge method consists of applying a potential difference between two graphite electrodes at a sufficiently small distance within a steel chamber in an inert atmosphere (usually Ar or He) and at low pressure, so that an electric arc between the electrodes is formed, ionizing the surrounding gas, and generating a plasma between the electrodes (Figure 7). Thus, as the electrons are accelerated against the anode, collision occurs with carbon atoms from the surface of the electrode transferring enough energy to pull them out, so that carbon atoms pass through the plasma and lose kinetic energy due to collisions with the gas atoms of the inert atmosphere. This procedure allows the formation of new graphite chains on the surface of the other electrode (cathode) (Ariga, Minami & Shrestha, 2016). The electric arc discharge method is still one of the most widely used methods for the synthesis of fullerene.

Synthesis of carbon nanotube and fullerene by laser ablation
The laser ablation method is used both for the synthesis of CNTs and for the synthesis of fullerenes. In this method, a solid graphite stick is used as a carbon source and is placed inside a quartz tube that is taken to a temperature-controlled oven. The tube is then filled with an inert gas, and the laser is focused on the graphite. Then, successive laser pulses are used as an energy source to uniformly vaporize the carbon, while the inert gas stream entrains the generated species, depositing them in a copper collector, located at the opposite end of the tube (Figure 8). After, the system is cooled by water to obtain the material (Terrones, 2003;De Bonis et al., 2015;Sajid et al., 2016). Laser ablation can also be performed in solution. This liquid technique gained prominence and attracted the interest of the scientific community due to its ability to produce stable nanostructures whose physical structure and chemical properties can only be altered by changing the experimental conditions used, such as the liquid medium or the laser parameters. Furthermore, when compared to the laser ablation technique performed in an inert atmosphere, the thermodynamic conditions in the ablation technique in solution result in more extreme temperature and pressure values, higher than that obtained for the gases. These conditions are necessary so that these conditions make the active species, generated by the laser, react more with each other, or with the molecules of the liquid medium and present greater stability (De Bonis et al., 2015).

Use of carbon nanomaterials in the development of electrochemical sensors
As previously seen, carbon nanomaterials gained increasing prominence, and their use in the development of sensors and biosensors for various applications in electroanalysis was continuously described in the literature due to the remarkable chemical and physical characteristics that these materials have, such as high surface area and excellent electrical conductivity, which facilitate the electronic transfer promoting the electrocatalysis.
In recent years, several graphene-based sensors and biosensors, or graphene-related materials, such as doped graphene, graphene oxide, and reduced graphene oxide, were reported in the literature. Several applications were described, such as environmental analysis for detection of metal ions such as Pb 2+ and Cd 2+ and pesticide analysis. Khan et al. (2015) utilized a sensor based on polyaniline-grafted graphene oxide-WO 3 nanocomposite for application as a chromium(III) chemisensor. The sensor showed a detection limit of 0.031 nmol l −1 . According to the authors, the method is promising for potential applications in the environmental and healthcare fields (Khan et al., 2015).
In another paper, Alam et al. (2017aAlam et al. ( , 2017b described that the use of a sensor of glassy carbon electrode (GCE) was fabricated using reduced graphene oxide/zinc oxide composite for 2-nitrophenol detection. The sensor presented a low detection limit of 0.27 nmol l −1 (Alam et al., 2017a,Alam et al., 2017b. These authors also developed a selective chemical sensor based on reduced graphene oxide/hydroxyapatite nanocomposites for the analysis of bis-phenol A. Their proposed sensor presented a detection limit of 60.0 pmol l −1 and also proved suitable for analysis in environmental samples (Alam et al. 2017a, Alam et al. 2017b).
Hussein et al. (2019) developed a selective Ag + ionic sensor based on polyvinyl chloride/mixed graphene-CNT nanocomposites. A detection limit of 14.78 pmol l −1 was obtained, with the possibility of future applications in environmental samples (Hussein et al., 2019).
Devices such as electrochemical sensors were also widely employed in food analysis and clinical applications, such as detection of glucose, ascorbic acid, uric acid, and in drugs.
In order to guarantee the quality of the results obtained and to validate the analytical method, the sensors and biosensors developed are usually evaluated for some parameters that determine their analytical performance, such as sensitivity, selectivity, limit of detection and quantification, repeatability, and reproducibility. Different modifications of the same base electrode can generate sensors with different specificities for a particular analyte. Graphene is extremely sensitive to the changes in the chemical environment to which it is inserted. This allows that any molecular perturbation can be easily detected, which contributes to the development of improved and highly sensitive electrochemical sensors. In addition, graphene can form composites with other materials such as polymers, nanoparticles, and biomolecules, such as DNA and enzymes, by means of physical adsorption or chemical bonding. As a result, combining the thermal, electrical, optical, and mechanical properties of graphene with the properties and functions of several other components, new materials and innovative analytical systems can be created, providing new opportunities for the sustainable progress of chemical analyses (Xiao-Mei et al., 2011).
In 2014, Jiang and Du developed an electrochemical sensor based on a glass-carbon electrode modified with a nanocomposite of gold (Au), palladium (Pd), and reduced graphene oxide (RGO) for individual and simultaneous determination of dopamine (DA), ascorbic acid (AA), and uric acid (UA), which are important molecules for physiological processes in human metabolism. The analysis of these species in body fluids (serum and urine) is a very valuable diagnostic indicator. As a result, for individual detection, the proposed sensor presented a wide linear response range, 0.1-1000, 0.01-100, and 0.02-500 μmol l −1 , and detection limits of 0.02, 0.002, and 0.005 μmol l −1 for AA, DA, and UA, respectively. In the simultaneous detection, the sensor presented separate and well-defined oxidation peaks. In addition, the sensor was applied in urine samples, showing good sensitivity and selectivity for the determination of AA, UA, and DA. The good response in terms of sensitivity and selectivity of Au-Pd-OGR-modified ECV can be explained by the presence of some oxygenated functional groups on the surface of reduced graphene oxide, thus, increasing its electrical conductivity and its biocompatibility, in addition to excellent electrocatalytic activity exerted by Au and Pd nanoparticles (Jiang & Du, 2014).
Recently, Pothipor et al. (2019) described the use of a biosensor based on graphene-poly (3-aminobenzoic acid)-modified electrodes and porous-hollowed silver-gold nanoparticle labeling for prostate cancer detection. The developed biosensor presented a highly satisfactory performance, with a detection limit of 0.13 pg ml −1 , better than the clinically relevant level in human serum. The sensor is promising for the diagnosis of prostate cancer and other medical applications (Pothipor et al., 2019).
Another example of the use of electrodes modified with graphene was the work performed by Kesavan and John (2016), in which a sensor was prepared by the electrochemical reduction of graphene oxide (GO) with 2,4-diamino-1,3,5-triazine (aminotriazine-AT) on the surface of a vitreous carbon electrode. This electrode was applied in the determination of paracetamol (PA) in the presence of UA, which is an important interference agent for the determination of PA in human urine. At usual therapeutic doses, PA is completely metabolized and is eliminated in the urine. However, the consumption of high doses of PA can cause serious liver damage, inflammation of the pancreas, and renal failure. The modified electrode (GCE-GO-AT) proposed by Kesavan and John (2016) presented excellent electrocatalytic activity for simultaneous determination of PA and UA, successfully separating both their voltammetric signals. It was possible to determine PA even at higher concentrations of UA (up to 50 times more concentrated). The linear response range found for the proposed sensor was 0.04-100 μmol l −1 , and the detection limit was 0.68 nmol l −1 . The practical application of this sensor was also demonstrated when simultaneously determining the concentrations of PA and UA in human urine samples (Kesavan & John, 2016).
Many studies showed that electron conductivity, electron-donor properties, and the binding ability of graphene could be significantly increased by controlling its morphology and tailoring its electronic structure (Tang, Zhou & Chen, 2013). Nitrogen-doped graphene (NG) is an example of material that has been applied in electrocatalysis and development of electrochemical biosensor due to its capability to enhance the sensitivity and biocompatibility of graphene in biosensing applications . Yang et al. (2017) proposed, for the first time, an efficient and universal streptavidin-functionalized nitrogendoped graphene (NG) and used this material to develop a highly sensitive electrochemical immunosensor for tumor marker detection. The streptavidin-functionalized NG was characterized by transmission electron microscopy (TEM), electrochemical impedance spectrum, and cyclic voltammetry (CV). The immunosensor showed high electrochemical activity, larger specific surface area, and was successfully applied in human serum samples with a wide linear range of 0.02-12.0 ng ml −1 , and detection limit of 0.01 ng ml −1 , using carcinoembryonic antigen (CEA) as model analyte, providing a promising platform for electrochemical immunosensing applications (Yang et al., 2017).
Nitrogen-doped graphene is also reported to be functionalized with metal nanoparticles such as platinum nanoparticles (PtNPs). The interest in PtNp-functionalized NG (PtNPs@NG) for electrochemical sensor development steadily increased due to its high electrocatalytic activity and free electron mobility. Yang et al. (2015a), Yang et al. 2015b) reported a novel electrochemical glucose biosensor modified with PtNPs@NG and glucose oxidase (GOx). The fabricated biosensor was characterized using transmission electron microscopy, high-resolution transmission electron microscopy, X-ray photoelectron spectroscopy, static water contact angle, ultraviolet-visible (UV-vis) spectroscopy, electrochemical impedance spectra, and cyclic voltammetry, re-spectively. The GOx immobilization on PtNPs@NG was performed in a way that retained its bioactivity. The proposed glucose biosensor was applied in five human serum samples as well as the reference method (enzyme catalytic spectrophotometry) and showed a quasi-reversible and fast electron transfer process with a linear range from 0.005 to 1.1 mmol l −1 and a detection limit calculated to be 0.002 mmol l −1 . The electrochemical glucose biosensor also exhibited excellent selectivity, sensitivity, good reproducibility, and acceptable stability, which demonstrates that it is a promising platform for the development of excellent electrochemical biosensors (Yang et al. 2015a(Yang et al. , 2015b. The determination of l-Dopa (or Levodopa) is of great importance because it is the immediate precursor of DA and is used in the treatment of Parkinson's disease. DA cannot be administered orally, as it does not pass through the bloodstream barrier. Thus, l-Dopa is administered and converted into dopamine by the action of the enzyme dopa-decarboxylase, thereby, stimulating the production of dopamine in the body. Leite et al. (2012) developed a sensor using a pyrolytic graphite electrode (PGE) modified with chloro(pyridine)bis(dimethylglyoximato)cobalt(III) (Co(DMG) 2 ClPy) adsorbed on MWCNTs for the determination of l-Dopa in pharmaceutical formulations. At the end of the studies, this sensor presented a linear response in the concentration range of 3-100 μmol l −1 and a detection limit of 0.86 μmol l −1 , demonstrating that the proposed sensor is a viable alternative for the determination of l-Dopa in pharmaceutical samples (Leite et al., 2012).
Recently, Sooraj and collaborators also developed a sensor for l-Dopa determination in human urine and pharmaceutical samples using copper nanoparticle grafted with molecular imprinted polymer on MWCNTs (CuNPs/MWCNT-MIPs). The non-covalent interaction between l-Dopa and the functional groups present in the selective binding sites of the polymer composite sorbent is mainly responsible for the recognition property. The electrochemical investigation shows that the imprinted (CuNPs/MWCNT-MIP) material has good recognition capacity toward l-Dopa with a detection limit of 7.23 nmol l −1 . The developed sensor also showed a good selectivity, being able to discriminate between l-Dopa and structurally related compounds such as DA, UA, 3,4-dihydroxyphenylacetic acid, and homovanillic acid (Sooraj et al., 2020). Anojčić et al. (2019), aiming at the determination of dopamine, developed carbonaceous nanomaterialmodified carbon paste electrodes (CPEs) with MWCNTs in their native and functionalized (ox-MWCNT) forms. Under optimized conditions, the method presented a linear concentration range of 16.15-192.70 ng ml −1 DA and a low detection limit of 4.89 ng ml −1 . The sensor was successfully applied for DA determination in injection (Anojčić et al., 2019).
Another work using CNTs in the modification of electrochemical sensors was performed by Majidi et al. (2016) who developed a sensor for the determination of tryptophan (Trp) in biological samples (urine, saliva, and blood serum). In order to carry out this study, the researchers used a golden screen-printed electrode (SPEAu) modified with MWCNTs (SPEAu-MWCNT) with Trp aptamers immobilized on the surface of this sensor. Sensitive and effective analytical methods for the determination of Trp in clinical studies and medical studies, particularly in neuroscience and oncology, are extremely important, as Trp is one of the essential amino acids for human life. The sensor showed excellent sensitivity, selectivity, and precision for the detection of Trp in real samples presenting a linear response range of 1.0 × 10 −11 to 1.0 × 10 −4 mol l −1 and a detection limit of 4.9 × 10 −12 mol l −1 . This sensor is extremely promising as it is a simple, efficient, portable, and miniaturized system (Majidi et al., 2016).
Upadhyay and Srivastava (2019) describe an extremely relevant work, an enantioselective electrochemical sensor that, for the first time, discriminates atorvastatin isomers (ATS). The sensor was develop based on a glassy carbon electrode (GCE) modified with a functionalized MWCNT with hydroxypropyl-b-cyclodextrin (HBC). According to the authors, the method developed demonstrated outstanding advantages of chiral nanocomposite-modified electrodes such as excellent enantioselectivity, high stereospecificity, and good reproducibility. Under optimized experimental conditions, the anodic oxidation peak current ratio (ISS/IRR) of SS-ATS to RR-ATS was found to be 3.62 (Upadhyay & Srivastava, 2019).
Rahimi-Nasrabadi and collaborators developed a glass-based carbon-modified sensor modified with a composite based on fullerene (C 60 ) and CNTs for the determination of diazepam in real samples, such as blood serum, urine, and pharmaceutical formulations. After the development of the glass-carbon electrode modified with a nanocomposite based on fullerene (C 60 ), CNTs, and ionic liquid (1-butyl-3-methylimidazolium tetrafluoroborate), it showed good electrocatalytic activity in relation to reduction of diazepam in one linear response range of 0.3-700 μmol l −1 and a detection limit of 0.087 μmol l −1 . This sensor presented good stability and repeatability, and its application for the determination of diazepam in real samples was satisfactory (Rahimi-Nasrabadi, Khoshroo & Mazloum-Ardakani, 2017).
Prasad, Kumar, and Singh described in their work a pencil graphite electrode modified with a molecularly imprinted polymer (MIP) using functionalized fullerene (C 60 ) as a nanomediator for ultratrace analysis of primaquine (PQ). The functional monomer employed by the authors was the 2,4,6-trisacrylamido-1,3,5-triazine, and the antimalarial drug (PQ) was used as a template for the molecular imprinting with the "surface grafting-from" approach. In this composite, the fullerene assists the electron transfer between the recognition sites and the electrode, inducing electrocatalytic action and thereby greatly extending the electrode kinetics. The importance of PQ determination in different matrices (blood plasma, urine, and pharmaceutical formulations) is related to the fact that PQ is the only anti-malarial drug that is active against latent liver forms of relapsing malaria and useful for the treatment of pneumocystis infections. Therefore, Prasad and collaborators validated the proposed sensor for complex matrices of human blood plasma, urine, and pharmaceutics, without any matrix effect and cross-reactivity. The linear response ranges were found to be: 2.7-848.5, 4.2-827.4, 3.4-795.2, and 4.8-803.2 nm for aqueous, blood plasma, urine, and pharmaceutical samples, respectively, with detection limits lower than 0.80 nm (Prasad, Kumar & Singh, 2016).
Recently, Zang and Li evaluated the electrochemical activity of dopamine (DA) using a glassy carbon electrode modified with non-covalent self-assembly of porphyrin-diazocine-porphyrin (PDP) and fullerene (C 60 ). Several works in the literature reported that porphyrin derivatives have exceptionally high affinity with fullerene, forming efficient electron transfer supramolecular assemblies. In the work described by Zhang and Li, the PDP-C 60 /GCE composite-modified electrode exhibited high electrocatalytic activity for DA detection. The anodic peak current increased linearly with increasing concentration of DA in the range of 0-200 μm. The detection limit was found to be 0.015 μm. Thus, the proposed sensor showed high sensitivity, satisfactory selectivity, excellent repeatability, and stability, showing that PDP-C60/GCE is a potential electrode material for dopamine analysis in real samples (Zhang & Li, 2020). Rather et al. (2016) reported the first-time synthesis of the fullerene-graphene oxide nanocomposite (C 60 -GO), conjugate by 1,3-dipolar cycloaddition reaction between fullerene (C 60 ), and azide-functionalized graphene oxide (GO-N 3 ). The C 60 -GO nanocomposite was characterized by UV-vis spectroscopy, electrochemical and field emission electron microscopy (FESEM), and Fourier transform infrared spectroscopy (FTIR). Then, using surface-bound diazonium salts to give a phenyl-modified conjugate interface, fullerene-graphene oxide nanocomposite was immobilized on the surface of glassy carbon electrode (GCE) forming C 60 -GO-Ph-GCE. After that, an electrochemical reduction (ER) was performed in an aqueous solution of potassium hydroxide (1.0 m KOH) to produce highly conductive films of C 60 GO-Ph at the electrode surface with electrocatalytic properties. In this step, the sensor was named ERC 60 -GO-Ph-GCE and was applied in the determination of homovanilic acid (HVA), a biomarker for Parkinson's disease, over a concentration range from 0.1 to 7.2 μm. The detection limit of the proposed sensor was found to be equal to 0.03 μm, which indicates good efficiency, high sensitivity, and makes this device suitable for the detection of HVA in biological fluids. The monitoring of HVA in patients suffering from Parkinson's disease is extremely important as DA undergoes catabolism by monoamine oxidase and catechol-O-methyl transferase to give HVA and 3,4-dihydroxyphenylacetic acid (DOPAC). Thus, electrochemical methods are promising tools that showed good results in providing quick and sensitive analysis of biomarkers in biological fluids (Rather et al., 2016).
The development of biosensors, enzymatic or non-enzymatic, for the detection of glucose in biological fluids gained prominence because of the interest in the application to the clinical diagnosis of diabetes, as they are relatively low-cost devices with fast and accurate responses, good stability, sensitivity, and selectivity. Sutradhar and Patnaik (2017) recently developed a gold nanoparticle-functionalized fullerene (C 60 )-modified vitreous carbon electrode using thiol-capped gold nanoparticle-based nanocomposite using 3-amino-5-mercapto-1,2,4triazole as the ligand for glucose analysis. This sensor showed electrocatalytic activity, effective electron transfer capacity, linear response range of 0.02-0.8 mmol l −1 , detection limit of 22.0 μmol l −1 , good stability, besides good selectivity in relation to possible interferents (Sutradhar & Patnaik, 2017).
Numerous studies involving electrochemical sensors and biosensors using graphene, CNTs, and fullerene were constantly described in recent literature for the determination of drugs and compounds of clinical interest. In this article, certain trends were highlighted instead of comprehensively covering all the published literature. Table 1 shows some examples highlighting the analyte of interest, the electrode, the modification used, and the limit of detection obtained.

Final considerations
From this review, it was possible to verify that the chemistry of nanomaterials is very dynamic. Applications such as determination of pharmaceutical compounds, disease diagnosis, and other medical applications justify the growing scientific interest in these materials. Particularly, in relation to electrochemical sensors, it was possible to verify that carbonaceous materials are often used in the form of composites with polymeric materials or metallic nanoparticles in the development of sensors, presenting excellent analytical performance. In this way, the electrochemical sensors became increasingly promising because of their relatively low cost compared to other conventional analytical methods, besides less reagent expense, quick and easy execution, making it possible and feasible to apply these devices in various analyses of environmental interest, food, and, in particular, analyses of drugs and compounds of clinical interest.