Using Cyclic Voltammetry, UV-Vis-NIR, and EPR Spectroelectrochemistry to Analyze Organic Compounds

Cyclic voltammetry (CV) is a technique used in the analysis of organic compounds. When this technique is combined with electron paramagnetic resonance (EPR) or ultraviolet-visible and near-infrared (UV-Vis-NIR) spectroscopies, we obtain useful information such as electron affinity, ionization potential, band-gap energies, the type of charge carriers, and degradation information that can be used to synthesize stable organic electronic devices. In this study, we present electrochemical and spectroelectrochemical methods to analyze the processes occurring in active layers of an organic device as well as the generated charge carriers.


Introduction
Worldwide, researchers are continually searching for new organic materials that can be used in organic electronics with desirable performance or stability, which drops due to extended use. In the case of organic devices, it is important to understand the behavior of the charge carrier to fully know the rules driving the device behavior. Analysis of the effect of the molecular structure on the generation of the charge carrier and the dynamics and maintenance of the balance of injected charge carriers, both positive (holes) and negative (electrons), is crucial to improve the efficiency and stability of the organic devices. This ensures the effective recombination of these individual charges and consequently significantly improves the photoluminescence efficiency of the organic light emitting diodes (OLEDs) 1,2 . For organic photovoltaics (OPVs) 3,4 as well as organic field effect transistors (OFETs) 5,6 , it is necessary to have materials with high charge carrier mobility. In addition to the analysis of charge carriers, several important parameters of organic electroactive materials help in predicting where the material could be used: ionization potential (IP), electron affinity (EA) energy levels, and band-gap between them 7,8,9,10 .
In this work, we present a method for the efficient measurement of cyclic voltammetry (CV) that can be used in the analysis of all types of electroactive materials. This technique provides information about redox properties, the doping/dedoping mechanism, the stability, the conversion and storage of energy, etc. It also allows for the estimation of electron affinity and ionization energy of the test compounds in a much cheaper and faster way compared to other high vacuum methods. The aforementioned parameters correlate with the energy levels of highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO).
The method presented in this article can be used to analyze all types of conjugated compounds such as those with delocalized π-electrons in their structures. Conjugated compounds may be small molecules with large polymeric chains. Small molecules can also be monomers; during the initial reaction (photochemical, electrochemical, or chemical) monomers can form polymers. In OLED application, the energy level values are necessary to enable the use of the correct host for the emitter in a thermally activated delayed fluorescence (TADF) guest-host system or to decide with which compounds the exciplex donor-acceptor layer could be formed and what additional layers (electron transporting layer (ETL), hole transporting layer (HTL), electron blocking layer (EBL), and hole blocking layer (HBL)) will be necessary to synthesize stable efficiently charged balanced OLED devices 11,12,13,14,15,16,17 . Additional electrochemical measurements allow the investigation of possible side reactions during the process of degradation of the active layer and the formation of low mobile charge carriers (bipolarons) 1. In the section Export ASCII data, define a filename and choose a folder to save the data in, and push the START button (decrease or increase the lower and upper vertex potential to fit the electrochemical window or the sample electrochemical activity) 6. If any peaks are visible in the positive range of potential, then repeat the cleaning procedure. If there is a peak in the negative range potential, then put the argon (or nitrogen) pipe in the solution and start bubbling for an additional 5 min. 7. Add a drop of 1 mM ferrocene (in the solvent used to prepare the electrolyte) to the electrolyte solution by using a syringe. 8. Set the start potential to 0.00 V, the upper vertex potential to 0.85 V, the lower vertex potential to 0.00 V, the stop potential to 0.00 V, the number of stop crossing to 10, and the scan rate to 0.05 V/s. In the section Export ASCII data, change the filename and push the START button. 9. After the measurement, finish by repeating the cleaning procedures as mentioned in steps 1.2 and 1.3. 10. Prepare 4 mL of 1 mM test compound in the prepared electrolyte (step 1.1). 11. Fill the electrochemical cell with the test compound solution (1.5 mL); put all three electrodes in a cell and connect to the potentiostat (step 2.4). To investigate the process of reduction, remove the oxygen (step 2.5). 12. Set the start potential to 0.00 V, the upper vertex potential to 0.50 V (or 0.00 V in case of investigation of reduction process), the lower vertex potential to 0.00 V (−0.50 V in case of investigation of reduction process), the stop potential to 0.00 V, the number of stop crossing to 10, and the scan rate to 0.05 V. In the section Export ASCII data, change the filename and push the START button. 13. Repeat step 2.13 by increasing upper vertex potentials (or by decreasing the lower vertex potential) by 0.1 V until peak registration ( Figure  2a). If the successive scan is shifted in potential (Figure 2b), then clean the RE and leave it in the electrolyte solution. Then, repeat the measurement. 14. Measure both oxidation and reduction processes on the same CV for the determination of IP and EA: set the start potential to 0.00 V, the upper vertex potential to 1.00 V, the lower vertex potential to −2.70 V, and the stop potential to 0.00 V. Choose the upper and lower vertex potential in a way that registers full reduction and oxidation peaks and avoids further redox steps (if any exists) (Figure 2a). NOTE: Estimate IP and EA by measuring the onset potential. There are a couple of ways to calculate these parameters from CV but the most useful is to calculate using the onset potential. Sometimes, it is not possible to observe reversible redox couple, especially for both n and p doping. Mixing of two different techniques is also undesirable as it may provide wrong results and conclusions. 1. To measure the onset potential, mark the tangent line to the CV peak (Figure 3). Calculate the desired values from the onset potential of the reduction or oxidation process using equations (Equations 1 and 2) 10,18,19,36,37,38 . (1) where is the onset of the oxidation potential calibrated with ferrocene * (2) where is the onset of the reduction potential calibrated with ferrocene * * Potential calibrated with ferrocene means that the value of potential from the measurement is reduced by the oxidation potential of ferrocene.
15. Pull out the electrodes and pour out the test solution into the proper waste container. 16. Repeat the cleaning procedure (steps 1.2 and 1.3). 17. Repeat steps 2.9-2.12 with 0.25, 0.10, and 0.20 V/s scan rates. 18. Add one drop of 1 mM ferrocene (in the solvent used to prepare the electrolyte solution) to the test solution using a syringe. Set the start potential to 0.00 V, the upper vertex potential to 0.60 V, the lower vertex potential to 0.00, the stop potential to 0.00 V, the number of stop crossing to 6, and the scan rate to 0.05 V. Push the START button. After the measurement is completed, change the upper vertex potential to a value that registers the ferrocene couple with the first peak of oxidation of the test compound. 19. Pull out the electrodes and pour out the test solution into the proper waste container. 20. If the electropolymerization occurs during electrochemical oxidation, wash the WE carefully with the electrolyte solution using the syringe (3 0 .5 mL). Clean the RE, AE, and electrochemical cell using the standard procedure (steps 1.2-1.3). 21. To investigate electrochemical products deposited on WE, proceed to step 2.23. To investigate a solution, proceed to step 1.2.1 and start again from step 2.3. 22. Fill the electrochemical cell with an electrolyte solution. Then, put all the three electrodes in a cell and connect them to a potentiostat (step 2.4). To investigate the reduction process, deoxidate the solution (step 2.5). 23. Repeat steps 2.10-2.12 and then step 2.16.  1. Apply a neutral potential (i.e., 0.00 V). This spectrum will be the starting spectra. Increase potential by 0.1 V and wait ca. 10 s until the process stabilizes and records absorption spectra (steps 3.1.13-3.1.16). Continue the measurement to the first change in UV-Vis-NIR spectra; wait for 10 s and then save the spectra (steps 3.1.13-3.1.16). 2. Increase potentials by 0.05 V; wait for 10 s and save (steps 3.1.13-3.1.16). 3. Repeat 3.2.2 until achieving the potential of the first or second oxidation potential (these potentials are known from the CV measurement); then, reverse the potential and go back to the starting potential.

UV-Vis-NIR Spectroelectrochemical Analysis
3. When the measurements are completed, measure the CV of the test solution. Add 1 mM ferrocene (10 μL) and measure the CV again. The value of ferrocene will help estimate the potential of the process and unify the data. 4. This analysis provides information regarding the optical band-gap, maximum wavelength of undoped and doped states, and isosbestic points of the electrochemical process. Calculate the optical band-gaps from onset values of π-π * bands on the UV-Vis spectra and using Equation 3 18,19,23,24 .
where h is Planck's constant, λ onset is compound absorption onset, and c is the speed of light in vacuum. 5. Time-Resolved Potentistatic Analysis 19,32,33 1. Prepare the measurement similar to the method described in 3.1.1-3.1.16 and put the argon (or nitrogen) pipe (through an additional hole in the electrode holder) and bubbling the solution at least 5 min prior to the measurement. After this, move the argon pipe above the solution level and keep the gas flow running for the measurement. 2. Run the potentiostat software, choose the chronoamperometry procedure with the following settings: potential start potential to 0.00 V, upper potential to 1.0 V, lower potential to −1.00 V, duration as 5 s, interval time as 0.010 s, number of repeats to 10, and delay time to 10 s. In the section Export ASCII data, define a filename and choose a folder to save the data. 3. In the spectrometer software, choose File | New | High-speed Acquisition. A new window will appear; set up the following parameters: Integration Time-10 ms, Scans to Average-1, Boxcar Wirth-0, Number of Scans-10000. Do not click the Go button. 4. In the potentiostat software, push the START button. When the countdown drops to 1 s, press the Go button in the spectrometer software. 5. Note that it is not possible to see absorption results until the process finishes. 6. Analyze the data to provide parameters. Use the information saved in the file such transmittance, current, voltage, and current density to provide working parameters: 1. Calculate Optical Density (ΔOD), the absorbance of the test compound during doping (dedoping) with Equation 4: where T b is transmittance in doped form, a "bleach" state [%], and T c is transmittance in undoped form, a "colored" state [%]. 2. Calculate Coloration Efficiency (CE, η), the amount of electrochromic color formed by the used amount of charge and as the connection between the injected/ejected charge as a function of electrode area (Q d ) and the change in the optical density (ΔOD) value at a specific wavelength (λ max ) with Equation 5. Its value depends on the wavelength chosen for study, so the value is measured at λ max of the optical absorption band of the colored state.
where T b is transmittance in doped form, "bleach" state [%], and T c is transmittance in undoped form, "colored" state [%].

EPR Spectroelectrochemical Analysis
where l mer -number of repeating units in the polymer, Q polym is the polymerization charge, F is the Faraday constant, and N A is the Avogadro number. 6. To calculate doping charge, use the CV of deposited polymer recorded in a spectroelectrochemical cell before the EPR measurement, using Equation 8: where Q d is the doping charge, i is the current, E is the set potential, E 0 is the starting potential, v is the scan rate, and e is the elemental charge. 7. Use calculated doping charge to calculate doping level, using Equation 9: where ∆D-doping level, Q d is the doping charge, and l mer -number of repeating units in the polymer 8. To calculate the number of bipolarons, assume an initial doping level equal to zero. Use the number of polarons per unit from the concentration of spins, so the number of bipolarons should be calculated using Equation 10. In the calculations, assume that the initial amount of bipolarons is zero and that faradaic current is much higher than that of non-faradaic current.
Copyright © 2018 Creative Commons Attribution 3.0 License October 2018 | 140 | e56656 | Page 6 of 13 (10) where z i is the charge carrier's elemental charge, n i is the number of charge carriers per polymer unit, (p) is the polarons, and (b) is the bipolarons. 9. Extract the g-factor value from the EPR spectra. Determine using Mn as an internal standard (steps 4.2.3 and 4.3.8), knowing that the fourth manganese line in it has a g-factor of 2.03324. Calculate the line width of EPR signal as the distance in mT between minimum and maximum of the EPR signal. This value is important as it may tell where (on which atom) the radicals are formed.

Representative Results
The most common application of CV analysis is the estimation of IP and EA. Even though there are a couple ways to obtain CV data, it is strongly recommended to calculate them based on the onset of redox peaks (Figure 3). This approach allows unifying the calculation procedure. Not all of the tested materials undergo reversible oxidation/reduction processes (Figure 3); in such situations, it is not possible to calculate based on average potentials (average from potentials correspond to a maximum of cathodic-reduction and anodic-oxidation peaks). However, it is almost always possible to run the tangent to a peak as shown in There are many articles published regarding the analysis of CV. Herein, we show when the process is not going to be as it is expected. The analysis is based on thiophene derivative: NtVTh (structure shown in Figure 4), which undergoes degradation upon oxidation ( Figure 5).
NtVTh has two oxidation potentials: the first at 0.70 V and the second at 0.84 V (Figure 5). During the first cycle, the reduction peak is not observed, indicating an irreversible process. Electrochemical characteristics of NtVTh show that polymerization does not occur and after the first oxidation potential, some electro-inactive layer of reaction products deposits at the electrode surface, thus hindering the polymerization process. What is visible is the reaction with the radical cation on the vinyl bond, where the molecule is losing its conjugation and form dimers on the electrode.
While it is difficult to extract information about charge carriers from the CV, it is possible to distinguish between polarons and bipolarons when supported by a UV-Vis-NIR spectrometer. The neutral poly(OiPrThEE) was characterized by two wide absorption π|π* transitions bands with peaks maxima at λ max1 = 363 nm and λ max2 = 488 nm, related to the aromatic form of the undoped polythiophene derivative. During the oxidative doping, new polaronic and bipolaronic bands are generated. The UV-Vis spectrum obtained during the poly(OiPrThEE) oxidation revealed the diminishing the neutral polymer absorption band (300-550 nm) (Figure 6) together with the formation of a new absorption band (550-950 nm) of the radical cations of bithiophene and p-phenylenevinylene with maxima at 692 nm. The isosbestic point of the oxidation process was located at 604 nm. The bipolaronic band appeared between the 950 nm and 1700 nm with a maximum located at 1438 nm.
EPR spectroscopy is the technique that detects materials with an unpaired electron, this includes organic radicals 39 . There are several parameters that could be extracted from EPR spectra, but one of the most interesting is to estimate where the radicals are localized. Electrons, similar to protons, possess spin. By placing an electron in an external magnetic field, this spin can be split two ways: parallel and antiparallel to the magnetic field, giving two energy levels. This phenomenon is known as the Zeeman effect 40,41 . In case of organic radicals, the unpaired electron interacts not only with the external magnetic field but also with magnetic nuclei (nuclei which have a nonzero spin; I≠0). A number of degenerate energy levels are equal to 2I + 1, where I is the spin quantum number of the nucleus with which the unpaired electron interacts 42 .
The interaction of the unpaired electron with a larger number of magnetic nuclei leads to further splitting of the energy levels and to hyperfine structure of EPR spectra registration 43 (Figure 7).
For molecules where the unpaired electron interacts with an even larger number of nuclei, the individual spectral line could overlap, which results in registration of a single, broad signal 44,45,46 (Figure 8). This is typical for conjugated polymers, where the generated radical ion during a redox process is delocalized 47,48 .
The combination of EPR spectroscopy with electrochemical methods allows the characterization of charge carriers (radical ion) generated during the redox process as well as the determination of the mechanism of these processes 49,50 . If well-resolved (peaks are separated; not in the form of one broad peak) spectra are registered, as in the case of electrochemical reduction of s-tetrazine derivative (Figure 7), then the analysis of the hyperfine structure of spectra leads to conclusions about the localization of unpaired electron. One way to analyze this kind of spectra is to conduct simulation with special software and to fit simulated spectra with the experimental one 51 . This is especially helpful when the hyperfine structure is complex due to the interaction of the unpaired electron with large numbers of protons. In case of the s-tetrazine derivative shown in Figure 7, simulation of EPR spectra (red line) indicates the interaction of the unpaired electron with four nitrogen atoms of s-tetrazine.

Discussion
Electrochemical and spectroelectrochemical techniques have no limitations; one can analyze both solid state and liquid solutions in a broad range of temperature and other conditions with these techniques. The important thing in all of these cases is that compounds/materials are analyzed under the applied potential, replicating real world conditions for working organic electronics devices. The only difference is that in electrochemistry, the formation of charge carriers, is observed.
The methods presented here show the usefulness of the analysis of charged carriers generated in organic compounds that correlate with their applicability in organic electronics. Moreover, the electrochemical and spectroelectrochemical techniques are cheaper and less demanding than that of typical methods used in charge carrier analysis, but there are some critical steps and modifications to the protocol that are needed depending on the obtained results.
During electrochemical characterization, always start with a particular concentration. If a set of the compounds is compared, then all materials need to have the same molar concentration. The best is to start with 1 mM concentration and 50 mV/s scan rate as indicated in the protocol in this study, but it is good to know the concentration of the sample on the observed electrochemical behavior. Always try to measure at least three scans. The first two scans are usually different because the starting conditions (equilibrium) are different. The second and the third scans should be the same. If the second and third scans are the same, then there are probably no side reactions observed in this system (Figure 2a). In an oxidation process, a new peak at a lower potential appears showing that the conductive material was deposited on the WE 18,19,24,25,29,30,31,32 .
If the height of the lower peak increases in successive scans, then probably the conjugated polymer was deposited 18,19,24,25,29,30,31,32 . If all the currents decrease in successive scans, then the nonconductive product of degradation was deposited on the electrode. If a very small peak is observed before the main oxidation or reduction peak (especially for polymers), then this is probably charge-trapping process 19,23,31,34 . If a very sharp dedoping peak of oxidation or reduction is observed, then this is probably caused by the decomposition of crystalline structures on an electrode formed through the electrocrystallization process during oxidation 35 .
Always check the behavior of the test compound before, during, and after redox peaks. It means that at least three CV scans should be registered: with upper (in the case oxidation) or lower vertex potential lower or higher, respectively, then the potential of peak maximum, with upper or lower vertex potential set to exactly on the peak maximum and with vertex potentials higher (oxidation) and lower (reduction) than potential of the peak maximum. The observed process may vary and sometimes two processes may be observed under one peak theoretically. Always compare the collected cyclic voltammograms of the electrolyte (step 2.6), the ferrocene (step 2.9), the compound (step 2.13), and the ferrocene with compound (step 2.19); there are several issues to be taken into account.
Always compare the CV signals of the electrolyte and the test compound, if any signals from the electrolyte is visible on the cyclic voltammogram of the measured compound, then the electrolyte must be changed because its electrochemical window is too low, or the electrolyte is contaminated. If the signal (redox couple) of ferrocene (step 2.9) and ferrocene with compound (step 2.19) are at the same position, then everything is performed properly. If the peaks are shifted between each other, then check the RE and repeat the measurement. If the signal (oxidation, reduction, or redox couple potential) of the test compound with added ferrocene (step 2.19) is at a higher potential than that of the pure compound (step 2.13), then consider the values (oxidation, reduction, or redox couple potential) from the cyclic voltammogram of the pure compound. The shift is caused by the higher amount of ferrocene in the solution. When two oxidation processes are observed, the first process (oxidation or reduction) which is always on the WE may affect the active surface; this may cause an increase in the oxidation potential of the second process (Figure 9).

Disclosures
The authors have nothing to disclose.