Voltammetric Determination of LSD with a Schiff Base – Chemically Modified Electrode in Aqueous Solution

In recent years, practical, inexpensive, and highly specific electroanalytical methods based on modified electrodes have been increasingly developed for forensic science. Simple modification of the carbon paste electrode with Schiff base complexes has become a promising strategy to detect and quantify narcotics. In this context, we aimed to develop voltammetric methods to quantify lysergic acid diethylamide (LSD) by using a carbon paste electrode modified with the complex [UO2(Ac-ophen)]·H2O. The use of an aqueous solution of KCl as supporting electrolyte makes the methodology less polluting, which contrasts with methods that still employ toxic solvents. The developed method for Differential Pulse Voltammetry provides a linear response at various concentrations of LSD and affords analytical curves with standard deviation, detection, and quantification limits around 2.45, 0.625, and 2.08 μmol L, respectively. The recovery values of 103 and 108% prove that the developed method is suitable for application in forensic science.

Although chromatographic methods are the most often applied, they are expensive and time consuming, usually due to the sample treatment step that must be performed for this equipment 7, [16][17][18] . As an alternative, voltammetric techniques have simpler, easy handling and low cost equipment, and can still be highly specific and provides limit of detection (LOD) as low as the LOD obtained in chromatographic techniques 19,20 .
Many studies have focused on voltammetric analysis applied in the field of forensic science, mainly in the determination of illicit drugs like cocaine 21,22 , LSD 3,23 , 25H-NBOMe 24,25 , MDMA (3,4-methylenedioxymethamphetamine) 26,27 , and natural cannabis in biological and seized samples 28,29 . Voltammetric analysis enables the development of sensors that provide specific response to low levels of illicit drugs, especially when these sensors are obtained by chemical modification of the surface of the electrode. The development of chemically modified carbon paste electrodes has attracted researchers' interest in recent years 30 . These sensors are low-priced and easy to manufacture, by mixing the conductive material, such as graphite powder or carbon nanotube, with an agglutinate, such as mineral oil or paraffin, and various types of modifiers such as transition metal complexes containing Schiff bases can be used [30][31][32][33] . Salen-type ligands (Salen = N,N-ethylene-bis(salicylideneiminato) have been applied in many electrochemical sensors because they display versatile stereo and electronic properties 30 . These modifiers can improve the selectivity/sensitivity of the devices, increase the electroanalytical response, or decrease the over-potential of the analytes [31][32][33] .
Recent works by our research group have demonstrated that uranyl (UO 2 ) complexes with Schiff bases of the Salen type can efficiently detect cocaine and have potential application in the forensic context [34][35][36] . However, there are no studies on the determination of LSD at the surface of electrodes chemically modified with such complexes.

Materials and methods
A potentiostat/galvanostat Autolab PGSTAT128N with NOVA 1.11 software was used for the voltammetric measurements with platinum wire as the counter electrode,  All the voltammetric measurements were performed in triplicates in a room under incandescent light, at a fixed temperature of 24 °C. These conditions were adopted because LSD is sensitive to light and degrades above 30 °C 7 .

Composition of the carbon paste
We determined the amount of paraffin in the carbon paste on the basis of the stability of paraffin in solution, predominantly when an organic solvent was added to the electrochemical cell. For this, the percentage of paraffin in the carbon paste was varied between 10 and 35 %. Then, each electrode was evaluated by cyclic voltammetry, applying 20 cycles with potential range of -1.5 and 1.5 V in the presence of 50 µL of acetonitrile. Thus, it was possible to evaluate the detachment of the carbon paste by observing the presence of particles in the electrolyte solution.
The tests indicated that at least 35% (w/w) paraffin was necessary to prepare a stable carbon paste electrode.
After fixing the quantity of paraffin, we tested the performance of the electrode at different proportions of Schiff base complex and graphite powder as well as in the presence and absence of about 20% of the modifier because we aimed to achieve the best voltammogram. We found that the optimum paste should consist of graphite powder, Schiff base complex, and paraffin at 45:20:35 (w/w/w) ratio, respectively. The resulting carbon paste electrode was stable even after 100 successive Cyclic Voltammetry scans without the need to renew the surface of the electrode. The electrode was stable even in the presence of acetonitrile, which was the solvent present in the standard solution of LSD.

Voltammetric profile of the Schiff base
We evaluated the electrochemical behavior of the Schiff base by Cyclic Voltammetry in two steps. First, we only analyzed the structure of Ac-ophen in the composition of the carbon paste. Then, we analyzed the complete structure [UO 2 (Ac-ophen)]•H 2 O.  The uranyl complex is insoluble in aqueous medium, and the supporting electrolyte was prepared with water: then, the voltammetric response was due to the electrochemical behavior of the ion UO 2 2+ , as shown in Figure 2. The oxidative processes are observed with the uranyl complex, as a free base this behavior does not occur.
The electrochemical behavior of UO 2 + might be described as follows, Figure   3: firstly, there is a reduction of on electron and the product is unstable in aqueous solution 39,40 .
Then, this pentavalent uranium results in U(IV) and U(VI) by its disproportionation step. It was verified the pH value in solution may influence in the electrochemical reactions of uranium, once at low pH is possible to observe the signal corresponded to U(III) 41 . There is no electrochemical data of this uranyl complex in the literature, so because we performed all the measurements in aqueous medium, and on the basis of the literature for the uranium ion 41,42 we assumed that the following redox processes were involved in the electrochemical reaction of the UO 2 2+ ion: the first cathodic peak, c 1 , was due to reduction of U(VI) to U(V), whereas the second cathodic peak, c 2 , referred to reduction of U(V) to U(IV). As for the oxidation peaks, a 1 corresponded to oxidation of U(V) to U(VI), and a 2 was due to oxidation of U(IV) to U(VI).
According to literature 42 , when an aqueous solution containing uranium was electrochemically characterized, it was verified a high disproportionation rate of this U(V). In this reaction, U(V) is formed on the electrode surface and as a consequence, we observed the peak a 2 due to the oxidation reaction from U(IV) to U(VI).
A broader study about the potential range and the redox processes in uranium was performed (Figure 3a and 3b) to a better analysis of the electrochemical behavior exposed above: when various potential intervals are analyzed, it is possible to conclude that even when the scan is reverted to potentials lower than -0.1 V, the peak c 3from the reaction of U(IV) to U(III)-is still present in voltammogram.
Besides, when the potential interval is reverted to values higher than -0.4 V, the peak c 2 is observed; when potentials higher than -0.4 V are applied, the peak a 1 is verified.
There is a co dependence between the peaks a 1 and c 2 , and this fact indicates the oxidation to form U(V).
When the potential is reverted to higher values, peak a 1 disappeared, whilst peaks c 1 and a 2 arise. This behavior may be explained by the reduction reaction from U(VI) to U(V) and the oxidation reaction from U(IV) to U(VI). The disappearance of the signal a 2 proves the disproportionation process in U(V) stated 41,42 .
The scan rate determined the presence of the anodic peak a 1 . At high scan rates, this peak became unstable or even disappeared from the cyclic voltammogram. According to the literature 41 , a third cathodic peak may arise in more acidic solution and is related to reduction of U(IV) to U(III).

Voltammetric profile of LSD
We conducted three different types of voltammetric measurements to explore the interaction between the Schiff base complex and LSD: Cyclic Voltammetry, Square Wave Voltammetry, and Differential Pulse Voltammetry.
In presence of LSD, the carbon paste electrode without modification (65:35 (w/w) graphite:paraffin) didn´t show any oxidative process for the drug, as well as the carbon paste modified with the Schiff base. Because of this absence, and the previous works of the group using the uranyl complexes whit Schiff base for cocaine detection [34][35][36] , the modification of the electrode with [UO 2 (Ac-ophen)]•H 2 O was applied.
Cyclic Voltammetry gives an overview of the behavior of the molecule and provides valuable information about the kinetic mechanism. In the presence of LSD, only peaks c 1 and a 2 arose in the cyclic voltammogram, which indicated interaction between the Schiff base complex in the modified electrode and the drug. This could be a catalytic response of the LSD to the oxidation of U(V). Similar voltammetric response was also observed in the analyses of cocaine using another uranyl Schiff base complex as modifier of the working electrode, whereas the presence of the drug increases the current signal for the oxidative processes of the uranyl group. 35 The exact mechanism of the reaction between the complex and the drugs, LSD and cocaine, has not been elucidated yet, but both drugs have a tertiary amine in their structure, which could explain the similarity in the voltammetric behavior.
The uranyl ion can be easily coordinated by more than 6 ligands (as in the pentagonal bipyramidal geometry) 43 . Also, Schiff base uranium complexes present great variety of oxidation states of the metal center, which are dependent of the coordination sphere. When spatial configuration of the molecule is considered, LSD can be coordinated to the uranyl on the electrode surface as the eighth donor atom, coordinated to the metal center, or either substitute the water ligand in the coordination sphere. This observation proves the feasibility of uranium as a chemical modifier for other drugs, such as cocaine as it was mentioned before [34][35][36] .
The linear relationship between the square root of the scan rate and the peak current for both peaks c 1 and a 2 pointed to a controlled diffusion system, and an irreversible electrochemical behavior was evident.

Square wave voltammetry
The two oxidative processes observed in the oxidative direction of Square Wave Voltammetry resembled the processes we had verified during Cyclic Voltammetry

Study about the interference of acetonitrile
Before studying the response to LSD in more detail, we analyzed how acetonitrile interfered in the electrochemical system because this solvent was present in the standard solution of the drug. Figure 5 illustrates how presence of acetonitrile affects the analysis of LSD.

Validation and comparison of techniques
After establishing the conditions for the quantification of LSD, we verified whether the results were reliable by validating Square Wave Voltammetry and Differential Pulse Voltammetry. We fixed the concentration of acetonitrile at the same value, and we analyzed the analytical curve by correlating the concentration of LSD to the anodic peak a 1 for Square Wave Voltammetry and Differential Pulse Voltammetry ( Figure 6). Due to the similar behavior of the linear increase in the a 1 peak current, the limits of detection (LOD) and quantification (LOQ) were also similar for both techniques. The lowest concentration of LSD found in blotters (containing 10 µg of LSD) after extraction of the analyte with an organic solvent 3 would correspond to a concentration of 5.18 10 -6 mol L -1 . This value was higher than the LOQ of both Square Wave Voltammetry and Differential Pulse Voltammetry, proving that these techniques are suitable strategies to determine even low concentrations of LSD within the scope of forensic routine.
Due to the unavailability of real samples of LSD, the recovery assay was performed considering two different concentrations of LSD (obtained by adding the LSD standard solution in supporting electrolyte) in the middle of the determined linear range, one for the square wave voltammetry and other for the differential pulse voltammetry. Table 2 lists these results.

Conclusions
The present work presents an overview of the electrochemical behavior of the hallucinogenic drug LSD by different techniques, namely Square Wave Voltammetry and Differential Pulse Voltammetry, on the basis of the interaction of LSD with the chemical modifier [UO 2 (Ac-ophen)]•H 2 O. Although the mechanism reaction is not completely elucidated, an innovative and pioneering way of using the Schiff base complex in the forensic context was presented.
The use of carbon paste as a working electrode also stands out due to its ease manufacture and low cost. Besides, the methods developed herein require aqueous medium instead of organic medium, like the work of Merli et al. 3 , which uses dimethylformamide/tetrabutylammonium perchlorate as supporting electrolyte, making this work more eco-frendely and sustainable.
These advantages make the tested voltammetric techniques suitable sensors for use in the forensic area. The figures of merit of the methodologies developed in this study attest that they can be used to detect and quantify even the smallest concentration of LSD reported on blotters.