A Short Review on Electrochemical Sensing of Commercial Dyes in Real Samples Using Carbon Paste Electrodes

Synthetic dyes are commonly used in food products like soft drinks, vegetable sauces, jellies, etc. Most artificial dyes can cause cancer, therefore it is very important to develop sensors to detect them in food samples. Voltammetric methods with carbon paste electrodes (CPEs) are promising for this purpose. However, modification of CPEs is necessary to detect the commercial dyes in food samples in the presence of interferents. In the current review, we have discussed the different previous research in which detection of dyes is performed in real samples with good detection limits. The current review will be helpful for readers who are interested in developing low-cost electrodes for the effective determination of dyes in commercial products like soft drinks and vegetable sauces.


Introduction
Dyes play an integral part in the present world. They are responsible for imparting colors due to chromophoric (azo, azoxy, nitro, carbonyl, thiocarbonyl) and auxochromic groups. They are different from pigments as the former are organic compounds and become attached chemically to the substrate while the latter do not form any chemical bonds and can be organic or inorganic compounds [1,2].

Dye Classification
Broadly, dyes can be classified into natural and synthetic based on the source of origin. Natural dyes can be extracted from natural sources such as fruits, flowers, leaves, roots, barks, etc., whereas synthetic dyes are synthesized in laboratories by the use of chemicals. In addition, they can be classified based on general structure into anionic, cationic, and nonionic, and categorized based on application characteristics into ten different types (acidic, basic, mordant, reactive, direct, disperse, sulfur dye, pigment, vat, azo insoluble) [2].

Synthetic Dyes (Advantages and Disadvantages)
Dyes influence every sphere of human life, be it food, leather, cosmetics, or the drug industry. By 1900, with an increase in demand, synthetic dyes gained immense attention and replaced natural dyes due to a variety of reasons such as low-cost production, easy availability, easy application, more color stability, resistance to light, pH changes, oxygen, etc. However, as every coin has two sides, dyes also have pros and cons. Therefore, considering their importance and large-scale industrial use, we cannot neglect their drawbacks. Some of these are described below [2].

I.
They are not environmentally friendly as their synthesis involves extreme conditions such as high pH, high temperature, strong acids, and heavy metal catalysts. II.
The most common substrate for dye production is petroleum, which is a nonrenewable source of energy.

Industrial Applications
In the food industry [7,8], dyes act as color additives that impart appealing bright colors to the food products without giving any nutritional value. The first artificial food coloring was prepared from coal tar in 1856 and since then hundreds of dyes have been prepared, out of which only a few are permitted by food regulatory bodies (FDA, FSSAI, EEC) owing to health concerns. They have been reported as creating various health issues such as toxicity, cancers, allergies, hypersensitivities, etc., but still, research is ongoing to elucidate the harmful effects of food dyes on health.
The list of commonly used permitted dyes along with color imparted is given in Table 1. In addition, some non-permitted dyes that are added to foodstuffs to enhance the color of the food are Fast Red, Mentanil Yellow, Bromocresol Purple, Green-S, and Sudan1-4. Hence the detection of these dyes is necessary.
Textile industry: It has been observed that 7 × 10 5 tons of synthetic dyes are produced every year around the globe, out of which about 2 × 10 5 tons are lost after various processing processes as waste effluent into water bodies. This waste effluent disturbs various water parameters such as pH, salinity, COD, BOD, etc., as most of the dyes contain aromatic rings causing toxicity and mutagenicity, ultimately affecting aquatic life adversely [9][10][11].
The two important classes of dyes that are employed in the textile industry include azo and anthraquinone.
Azo dyes are characterized by the presence of -N=N-bonds mainly imparting yellow, orange, and red colors. Anthraquinone is used for violet-blue or green colors.
Pharmaceutical industry: This industry [12] also employs various food dyes in processing pharmaceutical products such as tablets, capsules, syrups, etc. Dyes provide appealing appearance and easy identification for similar drugs. However, considering various ill effects of dyes, certain permissible amounts are fixed for using dyes. Dyes often used in the manufacturing process are tartrazine, quinoline, Sunset Yellow, amaranth, Allura Red, erythrosine, and Brilliant Blue.
On similar grounds, dyes are used in other industries for their color attracting properties and to appeal to the customers as per their requirements.
Dye degradation methods are categorized into three types: chemical, physical, and biological. The physical method includes adsorption (using activated carbon, silica, fly ash, bio adsorbent, polymeric adsorbent), irradiation, and filtration; the chemical methods include oxidation (using UV, H 2 O 2 , O 3 ), coagulation, and precipitation; while biological methods include aerobic and anaerobic methods [6,13].

Methods for Dye Detection
Various chemical methods have been reported for the detection of dyes, such as chromatography (HPLC, TLC, RPLC), capillary electrophoresis, mass spectrometry, spectrophotometry, immunoassay techniques (ELISA), etc. Among these the most employed technique is HPLC. However, these techniques are time-consuming and are expensive using highly skilled labor. Moreover, capillary electrophoresis, ELISA, and mass spectrometry do not work on neutral molecules and require charged molecules to work [14,15].
Besides the above-mentioned techniques, electrochemical methods, especially voltammetric techniques such as cyclic voltammetry (CV), differential pulse voltammetry (DPV), square-wave voltammetry (SWV), polarography, etc., have gained immense attention. The voltammetric studies employ the use of a three-electrode system: working electrode, counter electrode (non-reactive), and a reference electrode (helps to monitor the potential of the working electrode). The working electrodes can be solid (mainly graphite, carbon, diamond, bismuth, platinum, and gold) or liquid (mercury) and can be easily modified using organic and inorganic ions, molecules, polymers, or nanoparticles (NPs) for enhancing the electron transport rate [16].
Compared with other detection methods, voltammetric methods offer many advantages [17]: (i) they decrease the capacitive current; (ii) they are best for organic and inorganic molecules, which are electroactive; (iii) they give lower detection limits as the preconcentration of the analyte is prevented; (iv) an easily renewable surface of the working electrodes is possible; (v) lightweight and small instruments are employed; (vi) they not only determine the concentration but also give the composition of different species; (vii) they can also be applied to solid analytes.
Apart from the above-mentioned properties, they are fast, cost-effective, user-friendly, and show high specificity and sensitivity with low-quantity sample usage [18,19].
The current article describes the electrochemical methods for the detection of the following dyes using carbon paste electrodes as working electrodes.

Detection of Sunset Yellow (SY)
Sunset Yellow is a synthetic azo dye ( Figure 1) imparting a yellow to orange-yellow color. It is a permitted color additive and is widely employed in the food industry. It is commonly found as a disodium salt but can also be employed as potassium and calcium salts. The maximum permissible limit is 20 mg/kg in various food products.

Detection of Sunset Yellow (SY)
Sunset Yellow is a synthetic azo dye ( Figure 1) imparting a yellow to orange-yellow color. It is a permitted color additive and is widely employed in the food industry. It is commonly found as a disodium salt but can also be employed as potassium and calcium salts. The maximum permissible limit is 20 mg/kg in various food products. X. Chen et al. [20] detected SY in alumina-nanofiber-modified CPE in 2013 using DPV and CV studies. DPV of modified CPE showed an 8-fold increase in oxidation peak current as compared to bare CPE while the results for CV are shown in Figure 2. Thus, alumina nanofibers helped in detection precision with no influence of interfering molecules (<5%). Aluminum-nanofiber-modified CPE was applied to detect SY in soft drink samples and the obtained results were good with <4.5% deviation. The detection limit of the alumina-nanofiber-modified electrodes was 0.16 nM. The detection of SY was also performed for montmorillonitecalcium (MMT-Ca) functionalized cetyltrimethylammonium bromide (CTAB)-modified CPE by Y. Songyang et al. [21] in 2014. The oxidation behavior was studied for MMT-Ca/CPE and CTAB/MMT-Ca/CPE using CV. After the modifications, reversibility and peak currents were increased which was due to cationic exchange between CTAB and MMT-Ca and an increase in accumulation of dye at the electrode surface. Modified electrodes showed good sensitivity towards SY with a detection limit of 0.71 nM. In 2015, D. Sun et al. [22] modified the CPE by resorcinol-formaldehyde polymer films for the detection of SY using DPV in a phosphate buffer solution (PBS). After the modification, electrode sensitivity was improved and redox electron transfer reactions were fast because of the increased surface area and strong adherence of the resorcinol formaldehyde film at the modified electrode interface. They also performed real sample analysis using wastewater and soft drinks. Their results were good with a detection limit of 0.09 nM. In 2016 Z. Wang et al. [23] modified the CPE from molecularly imprinted polymers (MIPs) for determining SY. The sensitivity of the molecular imprinted polymer-modified electrodes was further improved by the addition of multi-walled carbon nanotubes (MWCNTs). Their results proved that a combination of multiwall carbon nanotubes and molecular-imprinted polymers increase the active surface area and conductivity at the modified CPE interface. Therefore, redox reactions of SY are easier and its effective detection is possible in the presence of metals like Ca 2+ , Cu 2+ , Cl − , and NO 3− . In 2017, Y. Ya et al. X. Chen et al. [20] detected SY in alumina-nanofiber-modified CPE in 2013 using DPV and CV studies. DPV of modified CPE showed an 8-fold increase in oxidation peak current as compared to bare CPE while the results for CV are shown in Figure 2. Thus, alumina nanofibers helped in detection precision with no influence of interfering molecules (<5%). Aluminum-nanofiber-modified CPE was applied to detect SY in soft drink samples and the obtained results were good with <4.5% deviation. The detection limit of the aluminananofiber-modified electrodes was 0.16 nM. The detection of SY was also performed for montmorillonitecalcium (MMT-Ca) functionalized cetyltrimethylammonium bromide (CTAB)-modified CPE by Y. Songyang et al. [21] in 2014. The oxidation behavior was studied for MMT-Ca/CPE and CTAB/MMT-Ca/CPE using CV. After the modifications, reversibility and peak currents were increased which was due to cationic exchange between CTAB and MMT-Ca and an increase in accumulation of dye at the electrode surface. Modified electrodes showed good sensitivity towards SY with a detection limit of 0.71 nM. In 2015, D. Sun et al. [22] modified the CPE by resorcinol-formaldehyde polymer films for the detection of SY using DPV in a phosphate buffer solution (PBS). After the modification, electrode sensitivity was improved and redox electron transfer reactions were fast because of the increased surface area and strong adherence of the resorcinol formaldehyde film at the modified electrode interface. They also performed real sample analysis using wastewater and soft drinks. Their results were good with a detection limit of 0.09 nM. In 2016 Z. Wang et al. [23] modified the CPE from molecularly imprinted polymers (MIPs) for determining SY. The sensitivity of the molecular imprinted polymer-modified electrodes was further improved by the addition of multi-walled carbon nanotubes (MWCNTs). Their results proved that a combination of multiwall carbon nanotubes and molecular-imprinted polymers increase the active surface area and conductivity at the modified CPE interface. Therefore, redox reactions of SY are easier and its effective detection is possible in the presence of metals like Ca 2+ , Cu 2+ , Cl − , and NO 3− . In 2017, Y. Ya et al. [24] detected SY in a ZnONano flower modified CPE. The electron transfer mechanism of SY is shown in Figure 3. The modified electrode had strong adsorption ability and a large surface area, which were attributed to the excellent electro-catalytic activity of the electrode (shown in Figure 4). Detection of SY was possible in the presence of various interfering molecules such as inorganic ions, organic molecules, citric acid, vitamin C, dyes (Sudan Red, Allura Red, Tartrazine) with a detection limit of 0.10 µg/L. Table 2 shows the detection limit of Sunset Yellow at the different modified carbon paste electrodes.

Detection of Tartrazine (TR)
Tartrazine ( Figure 5) is a synthetic azo, water-soluble food dye. It imparts a yellow or greenish-yellow color. It is widely used in the food, pharmaceutical, and cosmetic industries. The maximum permissible limit is 7.5 mg/kg/day of body weight. This dye is also reported to be neurotoxic, genotoxic, and even causes allergic responses.

Detection of Tartrazine (TR)
Tartrazine ( Figure 5) is a synthetic azo, water-soluble food dye. It imparts a yellow or greenish-yellow color. It is widely used in the food, pharmaceutical, and cosmetic industries. The maximum permissible limit is 7.5 mg/kg/day of body weight. This dye is also reported to be neurotoxic, genotoxic, and even causes allergic responses.

Detection of Tartrazine (TR)
Tartrazine ( Figure 5) is a synthetic azo, water-soluble food dye. It imparts a yellow or greenish-yellow color. It is widely used in the food, pharmaceutical, and cosmetic industries. The maximum permissible limit is 7.5 mg/kg/day of body weight. This dye is also reported to be neurotoxic, genotoxic, and even causes allergic responses. S. M. Ghoreishi et al. [25] also detected TR in presence of Red 10B at Nano-Au modified CPE in 2013. The CV studies indicated that both the dyes show irreversible cyclic voltammetric behavior as displayed in Figure 6. This type of modified electrode gave a good sensitivity for detection of TR in the presence of Red 10B. Therefore, the modified electrode showed great conductivity and sensitivity with a detection limit of 0.017 µM. In 2016, K. Nezhad et al. [26] detected TR electrochemically at poly (p-amino benzene sulfonic acid)/poly (BSA) and ZnO NP modified CPE. The CV showed an irreversible oxidation peak with enhanced peak current due to increased conductivity and faster electron transfer rate; inorganic and organic molecules/ions do not affect the detection of TR at the modified electrode. The electrode gave a detection limit of 80 nM and with real sample analysis using processed foodstuffs gave a recovery rate of 98% indicating good accuracy S. M. Ghoreishi et al. [25] also detected TR in presence of Red 10B at Nano-Au modified CPE in 2013. The CV studies indicated that both the dyes show irreversible cyclic voltammetric behavior as displayed in Figure 6. This type of modified electrode gave a good sensitivity for detection of TR in the presence of Red 10B. Therefore, the modified electrode showed great conductivity and sensitivity with a detection limit of 0.017 µM. In 2016, K. Nezhad et al. [26] detected TR electrochemically at poly (p-amino benzene sulfonic acid)/poly (BSA) and ZnO NP modified CPE. The CV showed an irreversible oxidation peak with enhanced peak current due to increased conductivity and faster electron transfer rate; inorganic and organic molecules/ions do not affect the detection of TR at the modified electrode. The electrode gave a detection limit of 80 nM and with real sample analysis using processed foodstuffs gave a recovery rate of 98% indicating good accuracy and precision of the electrode. In 2018, C. Raril et al. [27] also detected TR at surfactant (TX-100)-modified CPE. The CV studies showed enhanced oxidation peak current at the modified electrode due to the presence of electro-active functional groups at the surface of the electrode, thus oxidation is adsorption controlled. This modified electrode gave a detection limit of 1.114 µM. In the same year, J. G. Manjunatha [28] detected TR at a CPE modified using poly(glycine). This modified electrode enhanced the electron transfer rate, due to which CV showed a well-defined peak at 910 mV and gave a detection limit of 0.283 µM. The real sample analysis using food samples showed great recovery rates (94.5-98.2%). In 2020, C. Raril et al. [29] used a surfactant, SDDS (sodium dodecyl sulfate)-modified CPE, for the de-tection of TR. They carried out CV and DPV studies and noticed a well-defined peak. This modified electrode gave a detection limit of 5.2 µM with great sensitivity (0.048 µA/µM) and stability. Analysis of this modified electrode in various food samples gave a recovery ranging from 97.1% to 104%. Table 3 shows the detection limits of Tartrazine at the different modified carbon paste electrodes.
(TX-100)-modified CPE. The CV studies showed enhanced oxidation peak current at the modified electrode due to the presence of electro-active functional groups at the surface of the electrode, thus oxidation is adsorption controlled. This modified electrode gave a detection limit of 1.114 µM. In the same year, J. G. Manjunatha [28] detected TR at a CPE modified using poly(glycine). This modified electrode enhanced the electron transfer rate, due to which CV showed a well-defined peak at 910 mV and gave a detection limit of 0.283 µM. The real sample analysis using food samples showed great recovery rates (94.5-98.2%). In 2020, C. Raril et al. [29] used a surfactant, SDDS (sodium dodecyl sulfate)-modified CPE, for the detection of TR. They carried out CV and DPV studies and noticed a well-defined peak. This modified electrode gave a detection limit of 5.2 µM with great sensitivity (0.048 µA/µM) and stability. Analysis of this modified electrode in various food samples gave a recovery ranging from 97.1% to 104%. Table 3 shows the detection limits of Tartrazine at the different modified carbon paste electrodes.

Simultaneous Detection of SY and TR
S. M. Ghoreishi et al. [30] used a modified CPE for the simultaneous detection of SY and TR using cyclic voltammetry (CV) and differential pulse voltammetry (DPV) (shown in Figure 7). For simultaneous detection of SY and TR, they used Au nanoparticles for the modification. Voltammetric experiments were conducted using KCl saturated Ag/AgCl as a reference electrode and platinum wire as a counter-electrode along with a CPE working electrode. The cyclic voltammograms gave a well-defined redox peak and they were pH-dependent at the modified CPE. After modifications from the gold, the nanoparticle (NP) electrode showed improved sensitivity and later same the electrodes were successfully applied to detect the analyte in soft drink samples. These works confirm that Au nanoparticles at the electrode interface are beneficial in SY and TR sensing. In 2018, CPE was modified using Nd2O3 [31] for the simultaneous detection of TR and SY. After the

Simultaneous Detection of SY and TR
S. M. Ghoreishi et al. [30] used a modified CPE for the simultaneous detection of SY and TR using cyclic voltammetry (CV) and differential pulse voltammetry (DPV) (shown in Figure 7). For simultaneous detection of SY and TR, they used Au nanoparticles for the modification. Voltammetric experiments were conducted using KCl saturated Ag/AgCl as a reference electrode and platinum wire as a counter-electrode along with a CPE working electrode. The cyclic voltammograms gave a well-defined redox peak and they were pH-dependent at the modified CPE. After modifications from the gold, the nanoparticle (NP) electrode showed improved sensitivity and later same the electrodes were successfully applied to detect the analyte in soft drink samples. These works confirm that Au nanoparticles at the electrode interface are beneficial in SY and TR sensing. In 2018, CPE was modified using Nd 2 O 3 [31] for the simultaneous detection of TR and SY. After the modifications, anodic peak currents were enhanced for both the dyes when compared to the bare CPE. This indicated improved sensitivity as less energy was required for the oxidation of the dyes with higher conductivity due to an increase in surface area by Nd 2 O 3 . The real sample analysis was performed using spiked unflavored gelatins, which gave a detection limit of 0.02 µM for TR and 0.09 µM for SY. In 2019 SY, TR, and AR were detected at a cobalt-complex-modified CPE using CV by J. Pinagos-Llanos et al. [32]. They modified the electrode using the cobalt complex. The modified CPE was stable and showed enhanced conductivity with reduced diffusion time. After the modifications, real sample analysis was performed using flavored gelatine powder and the electrode showed a detection limit of 0.9 µM for SY and 0.3 µM for TR. Table 4 shows the detection limit of Sunset Yellow and Tartrazine at the different modified carbon paste electrodes. modifications, anodic peak currents were enhanced for both the dyes when compared to the bare CPE. This indicated improved sensitivity as less energy was required for the oxidation of the dyes with higher conductivity due to an increase in surface area by Nd2O3. The real sample analysis was performed using spiked unflavored gelatins, which gave a detection limit of 0.02 µM for TR and 0.09 µM for SY. In 2019 SY, TR, and AR were detected at a cobalt-complex-modified CPE using CV by J. Pinagos-Llanos et al. [32]. They modified the electrode using the cobalt complex. The modified CPE was stable and showed enhanced conductivity with reduced diffusion time. After the modifications, real sample analysis was performed using flavored gelatine powder and the electrode showed a detection limit of 0.9 µM for SY and 0.3 µM for TR. Table 4 shows the detection limit of Sunset Yellow and Tartrazine at the different modified carbon paste electrodes.

Detection of Amaranth (AM)
Amaranth is a synthetic azo dye ( Figure 8) capable of causing dizziness, nausea, and damage to various other important organs of the body. It is commonly employed in industries for its color-imparting applications. Hence, owing to health concerns, a permissible limit has been set up by various regulatory bodies for its uptake, i.e., 0.15 mg/kg of body weight per day.

Detection of Amaranth (AM)
Amaranth is a synthetic azo dye ( Figure 8) capable of causing dizziness, nausea, and damage to various other important organs of the body. It is commonly employed in industries for its color-imparting applications. Hence, owing to health concerns, a permissible limit has been set up by various regulatory bodies for its uptake, i.e., 0.15 mg/kg of body weight per day. Electrochem 2021, 2, FOR PEER REVIEW 9 In 2008, M. Pandurangachar et al. [33] studied the adsorption behavior of AM at a surfactant (CTAB, Cetyl Trimethyl Ammonium Bromide)-modified CPE. It was seen using CV studies that the CTAB enhanced adsorption of AM at the surface of the modified electrode. At <10 µL concentration of the CTAB, dye formed monomers at the electrode surface while at >40 µL conc. of CTAB, monolayer formation was reported. Therefore, it was concluded that the surfactant enhanced the adsorption of dye at the surface of the electrode by increasing the surface area of the electrode and enhancing electroactive sites for dye interaction. M. L. Wang et al. [34] detected AM using square-wave adsorption stripping voltammetry (SWSV) on an expanded graphite paste electrode. SWSV showed a broad and well-defined anodic peak for AM at the modified electrode. This modified electrode gave a wide linear range with a detection limit of 0.005 µM. The real sample analysis was performed using soft drink samples giving a recovery >96.5%. These results were in consideration with those found using HPLC. In 2016, M.Sheikhshoaie et al. [35] detected AM at a CPE modified using RuO2 Nano road (NR) and dipropylimidaizolium bromide (DPIR). CV studies showed irreversible oxidation of AM at the modified electrode with a maximum oxidation peak at PH 7 (shown in Figure 9). As compared to bare CPE, the electron transfer rate was enhanced at modified electrodes due to which excellent conductivity results were obtained. In addition, real sample analysis using food samples showed good results and in the presence of various interfering ions/molecules, the modified electrode was able to show great sensitivity. The modified electrode gave a detection limit of 3 nM. In 2017, M.Bijad et al. [36] detected AM with 1-Methyl-3-butylimidazolium bromide and a CuO-SWCNT-modified CPE in the presence of nitrite using SWV. The SWV showed two well-defined oxidation signals. The electrode gave a sensitivity of about 0.038 µA/µM. Real sample analysis showed results less than the detection limit for in vitro studies (i.e., 0.001 µM). In 2020, AM was detected by S. Akbari [37] with graphene quantum dots/ionic liquid modified CPEs. There was an enhancement in peak current at the modified electrodes as shown by CV studies. Real sample analysis of AM using water samples from different sources in apple and orange juices showed good recovery rates (>95%). The modified electrodes gave a detection limit of 30 nM with a wide linear range. Table 5 shows the detection limits of Amaranth at the different modified carbon paste electrodes. In 2008, M. Pandurangachar et al. [33] studied the adsorption behavior of AM at a surfactant (CTAB, Cetyl Trimethyl Ammonium Bromide)-modified CPE. It was seen using CV studies that the CTAB enhanced adsorption of AM at the surface of the modified electrode. At <10 µL concentration of the CTAB, dye formed monomers at the electrode surface while at >40 µL conc. of CTAB, monolayer formation was reported. Therefore, it was concluded that the surfactant enhanced the adsorption of dye at the surface of the electrode by increasing the surface area of the electrode and enhancing electroactive sites for dye interaction. M. L. Wang et al. [34] detected AM using square-wave adsorption stripping voltammetry (SWSV) on an expanded graphite paste electrode. SWSV showed a broad and well-defined anodic peak for AM at the modified electrode. This modified electrode gave a wide linear range with a detection limit of 0.005 µM. The real sample analysis was performed using soft drink samples giving a recovery >96.5%. These results were in consideration with those found using HPLC. In 2016, M.Sheikhshoaie et al. [35] detected AM at a CPE modified using RuO 2 Nano road (NR) and dipropylimidaizolium bromide (DPIR). CV studies showed irreversible oxidation of AM at the modified electrode with a maximum oxidation peak at PH 7 (shown in Figure 9). As compared to bare CPE, the electron transfer rate was enhanced at modified electrodes due to which excellent conductivity results were obtained. In addition, real sample analysis using food samples showed good results and in the presence of various interfering ions/molecules, the modified electrode was able to show great sensitivity. The modified electrode gave a detection limit of 3 nM. In 2017, M.Bijad et al. [36] detected AM with 1-Methyl-3-butylimidazolium bromide and a CuO-SWCNT-modified CPE in the presence of nitrite using SWV. The SWV showed two well-defined oxidation signals. The electrode gave a sensitivity of about 0.038 µA/µM. Real sample analysis showed results less than the detection limit for in vitro studies (i.e., 0.001 µM). In 2020, AM was detected by S. Akbari [37] with graphene quantum dots/ionic liquid modified CPEs. There was an enhancement in peak current at the modified electrodes as shown by CV studies. Real sample analysis of AM using water samples from different sources in apple and orange juices showed good recovery rates (>95%). The modified electrodes gave a detection limit of 30 nM with a wide linear range. Table 5 shows the detection limits of Amaranth at the different modified carbon paste electrodes.

Detection of Allura Red (AR)
Allura Red ( Figure 10) is a water-soluble synthetic azo dye. The permissible daily intake value is 7 mg/kg body weight per day. It is widely employed in the food industry but is banned in some countries. It is dark red and is also soluble in 50% ethanol.

Detection of Allura Red (AR)
Allura Red ( Figure 10) is a water-soluble synthetic azo dye. The permissible daily intake value is 7 mg/kg body weight per day. It is widely employed in the food industry but is banned in some countries. It is dark red and is also soluble in 50% ethanol. In 2015, J. Zhang et al. [38] detected AR at an IL (ionic liquid) expanded graphite In 2015, J. Zhang et al. [38] detected AR at an IL (ionic liquid) expanded graphite paste electrode (GPE/CPE) in the presence of Ponceau R dye. They conducted a CV and found enhancement in peak current at the modified electrode for both the dyes. The detection limit of 0.85 µg/L was found for AR with great sensitivity (with a deviation of <4.76%) in the presence of various interfering species. With real samples, the modified electrode gave a recovery ranging from 96% to 99.2%. In 2019, J.-P. Llanos et al. [39] detected AR using cobalt-complex-modified CPE. The CV showed the quasi-reversible process for the reduction of AR at the modified electrode with less energy requirement and faster electron transfer rate. The detection limit of 0.05 µM was given at the modified electrode. With real sample analysis using soft and isotonic drinks, this modified electrode gave good recovery limits (>89%). The detection of AR was performed using CPE modified using silica (SG) with cetylpyridinium chloride (CPC) by K. Pliuta et al. [40]. The optimal conditions were observed for SWV quantification: PH = 2, E ads = 300 mV, tags = 300 s, amplitude −40 mV, frequency −25 Hz, and potential scan rate 250 mV sec −1 . 1 × 10 −2 M of AR was dissolved in exact weight in double-distilled water for its preparation. The AR was determined in jelly sweets which were crushed and homogenized. A 1 g sample in 50 mL double-distilled water with heating was diluted up to 100 mL in the flask. Robinson buffer with pH 2 was mixed properly with the aliquot. For the CV using a CPE/SG/CPC, the oxidation current increased, and the peak shifted to the cathode at 950 mv. This decrease in oxidation overvoltage was explained by Allura adsorption on the electrodes. For the ranges 0.04-0.2 µM and 0.2-1.00 µM, the calibration plot had linearity. The values of a limit of quantification were 0.015 µM and the limit of detection was 0.005 µM. In 2019, AR was simultaneously detected along with SY and TR [32] with a detection limit of 0.08 µM. Recently, R. Darabi et al. [41] detected SY, TR, and AR simultaneously at a NiFe 2 O 4 -reduced graphene/IL-modified CPE. This modified electrode was also successful in determining Allura Red dyes in real sample analysis in cosmetic and food products (deviation < 5%). Table 6 shows the detection limits of Allura Red at the different modified carbon paste electrodes.

Detection of Carmoisine (CR)
Carmoisine ( Figure 11) synthetic azo dye is widely used in the food industry, having two naphthalene units. It offers a color range from red to maroon with typical applications in food and beverages. It is used as a food colorant in sauces, candies, yogurt, and chewing gum, having permitted levels up to 5-50 mg/100 g or /100 mL depending on the type of food. The average daily intake that is acceptable is 0-4 mg/kg body weight. two naphthalene units. It offers a color range from red to maroon with typical applica in food and beverages. It is used as a food colorant in sauces, candies, yogurt, and che gum, having permitted levels up to 5-50 mg/100 g or /100 mL depending on the ty food. The average daily intake that is acceptable is 0-4 mg/kg body weight. In 2017, M. Bijad et al. [42] detected CR at a CPE modified using NiO/CNT w methyl-3-butyl imidazolium bromide (3BIBr) in the presence of TR. They used CV to pare the peak currents of bare and modified electrodes. Modified electrodes showe hanced peak currents and the electron transfer process was irreversible. The elec showed a detection limit of 20 nM in the presence of TR. Dried fruit and soft drink sam were employed for the real sample analysis of both the dyes (TR and CR) and g recovery >99.25%. In 2018, H. M. Nehzad et al. [43] detected CR at a CPE modified Cd/CNT& 1-ethyl-3-methylimidazolium tetrafluoroborate (EMITB). They studie electrochemical behavior of CR at the modified electrode using linear sweep voltamm (LSV) and found low oxidation current with enhanced peak current. This indicated electrical conductivity and long-term stability of modified electrodes with a dete limit of 40 nM. The real sample analysis using fruit juices showed a recovery >97% a electrode. In 2020 the detection of CR was performed by using CPE modified by In 2017, M. Bijad et al. [42] detected CR at a CPE modified using NiO/CNT with 3-methyl-3-butyl imidazolium bromide (3BIBr) in the presence of TR. They used CV to compare the peak currents of bare and modified electrodes. Modified electrodes showed enhanced peak currents and the electron transfer process was irreversible. The electrode showed a detection limit of 20 nM in the presence of TR. Dried fruit and soft drink samples were employed for the real sample analysis of both the dyes (TR and CR) and gave a recovery >99.25%. In 2018, H. M. Nehzad et al. [43] detected CR at a CPE modified using Cd/CNT& 1-ethyl-3-methylimidazolium tetrafluoroborate (EMITB). They studied the electrochemical behavior of CR at the modified electrode using linear sweep voltammetry (LSV) and found low oxidation current with enhanced peak current. This indicated good electrical conductivity and long-term stability of modified electrodes with a detection limit of 40 nM. The real sample analysis using fruit juices showed a recovery >97% at this electrode. In 2020 the detection of CR was performed by using CPE modified by silica with cetylpyridinium chloride (CPE/SG/CPC), which has high sorption ability compared to azo dye (A. Chebotarev et al. [44]). The authors discussed the electrochemical behavior of CR on the CPE, CPE/SG, and CPE/SG/CPCl, in a detailed manner using SWV (shown in Figure 12) and CV at pH 2. The peak current is higher at CPE/SG/CPCl when compared to other electrodes. The limit of detection for this modified electrode was found to be 0.01 µM. Their results prove that silica with cetylpyridinium chloride will improve the CR adsorption on the electrode surface. Table 7 shows the detection limit of Carmoisine at the different modified carbon paste electrodes. 3. Silica CPE 0.01 µM Non-alcoholic soft drinks [44] to azo dye (A. Chebotarev et al. [44]). The authors discussed the electrochemical behavior of CR on the CPE, CPE/SG, and CPE/SG/CPCl, in a detailed manner using SWV (shown in Figure 12) and CV at pH 2. The peak current is higher at CPE/SG/CPCl when compared to other electrodes. The limit of detection for this modified electrode was found to be 0.01 µM. Their results prove that silica with cetylpyridinium chloride will improve the CR adsorption on the electrode surface. Table 7 shows the detection limit of Carmoisine at the different modified carbon paste electrodes.

Detection of Rhodamine B (RHB)
Rhodamine B (Figure 13) is also a non-permitted color additive due to its carcinogenic and acute poisoning properties. In addition, the dye can cause various environmental issues. It belongs to the xanthene class of dyes.

Detection of Rhodamine B (RHB)
Rhodamine B (Figure 13) is also a non-permitted color additive due to its carcinogenic and acute poisoning properties. In addition, the dye can cause various environmental issues. It belongs to the xanthene class of dyes. In 2019, Noviandri et al. [45] detected RHB at a CPE modified with molecularly imprinted poly (3aminophenol) (PAP) in the presence of TR. After the modification the electrode showed a good concentration linearity range with a detection limit of 1.17 pM. Real sample analysis was also performed using molecularly imprinted CPE and the obtained results were good (with 7.21% deviation). They further verified the results using UV spectroscopy methods. M. Golestaneh et al. [46] detected RHB at a CPE modified with MWCNT. They noticed a single cathodic peak with enhanced peak current using CV studies at the modified electrode. This modified electrode gave a detection limit of 20 nM and a recovery ranging from 95-110% using real samples (hair color and water). Thus, MWCNT-modified electrode possesses electro-catalytic activity towards RHB with high stability and good repeatability (deviation of 3.9%). In 2021, H. Setiyanto et al. [47] detected RHB at a CPE modified using MIP (polymer of glutamic acid) (PGA). Their poten- In 2019, Noviandri et al. [45] detected RHB at a CPE modified with molecularly imprinted poly (3aminophenol) (PAP) in the presence of TR. After the modification the electrode showed a good concentration linearity range with a detection limit of 1.17 pM. Real sample analysis was also performed using molecularly imprinted CPE and the obtained results were good (with 7.21% deviation). They further verified the results using UV spectroscopy methods. M. Golestaneh et al. [46] detected RHB at a CPE modified with MWCNT. They noticed a single cathodic peak with enhanced peak current using CV studies at the modified electrode. This modified electrode gave a detection limit of 20 nM and a recovery ranging from 95-110% using real samples (hair color and water). Thus, MWCNT-modified electrode possesses electro-catalytic activity towards RHB with high stability and good repeatability (deviation of 3.9%). In 2021, H. Setiyanto et al. [47] detected RHB at a CPE modified using MIP (polymer of glutamic acid) (PGA). Their potentiometric studies showed that the modified electrode was highly sensitive (sensitivity coefficient < 10 −3 ) towards RHB with a detection limit of 8.91 µM. The results were in comparison to a t-test (0.12). Thus, modified electrodes show highly electrocatalytic behavior. Table 8 shows the detection limits of Rhodamine B at the different modified carbon paste electrodes.

Detection of Erythrosine (ER)
Erythrosine ( Figure 14) is a permitted color additive belonging to the xanthene class of dyes but is toxic, causing inhibition of various functions in the body such as various types of allergies, anemia, thyroid alteration, DNA damage, and cancer. The permissible limit for the usage of the dye given by the FDA is 1.4 mg per person daily. It is a water-soluble organoiodine pink synthetic dye widely used as a food colorant, for printing inks, drugs, cocktails, tinned fruits, cosmetics, scotch egg, and snacks. In 2014, M. Golestanch et al. [48] used a MWCNT-modified CPE. This modified electrode possessed electro-catalytic activity as it enhanced peak currents (CV studies) and gave a detection limit of 5 nM. Real sample analysis in the food and cosmetic industry showed satisfactory results. D. S. Nayak et al. [49] detected ER at a glucose-modified CPE. The CV at the modified electrode showed two anodic peaks and enhancement of peak current as compared to bare CPE (displayed in Figure 15); the electron transfer process is an irreversible oxidation process, and the detection limit of the modified electrode is 21.6 nM. Real sample analyses were performed using dye-spiked human urine samples and results showed good recovery (91.6%-98%). M. Arvand et al. [50] also modified CPE using bimetallic Ni-Sn oxide hollow nanosphere (NS) in 2017. Electrochemical studies at this electrode showed a reversible peak with enhanced peak currents due to an increase in the In 2014, M. Golestanch et al. [48] used a MWCNT-modified CPE. This modified electrode possessed electro-catalytic activity as it enhanced peak currents (CV studies) and gave a detection limit of 5 nM. Real sample analysis in the food and cosmetic industry showed satisfactory results. D. S. Nayak et al. [49] detected ER at a glucose-modified CPE. The CV at the modified electrode showed two anodic peaks and enhancement of peak current as compared to bare CPE (displayed in Figure 15); the electron transfer process is an irreversible oxidation process, and the detection limit of the modified electrode is 21.6 nM. Real sample analyses were performed using dye-spiked human urine samples and results showed good recovery (91.6-98%). M. Arvand et al. [50] also modified CPE using bimetallic Ni-Sn oxide hollow nanosphere (NS) in 2017. Electrochemical studies at this electrode showed a reversible peak with enhanced peak currents due to an increase in the number of active sites at the surface of the modified electrode. The interfering species gave an error of ±10% in the determination of ER at this electrode with a detection limit of 2.1 nM. Real sample analysis results for gelatins, candies, and smarties at this electrode were confirmed by F-test (1.56-1.79) and t-test (2.02-1.78). Table 9 shows the detection limit of Erythrosine at the different modified carbon paste electrodes.

Detection of Sudan
Sudan dyes ( Figure 16) are fat-soluble synthetic azo dyes having wide applicability ranging from industry to scientific research. Sudan dyes are aromatic synthetic chemical dyes having (N=N-) azo groups. There are four kinds of Sudan, named Sudan I, II, III, and IV. These are used for coloring substances like oil, fats, waxes, hydrocarbon solvents, plastics, printing inks, textiles, leather, and wood. However, their use is banned in the food industry in various countries as they poses a threat to life due to their mutagenic, genotoxic, and carcinogenic properties. They are placed as category 3 toxins by the ICARC.
In 2008, H. Lin et al. [51] detected Sudan I at a montmorillonite Ca modified CPE (Shown in Figure 17). The CV analysis showed a single well-defined, enhanced oxidation peak at the modified electrode. The modified electrode was also applied to examine the electrochemical behavior of other Sudan dyes (Sudan II, Sudan III, and Sudan IV) as shown in Figure 18. Each type of Sudan dye can be detected simultaneously at the Ca modified CPE. The Ca modified CPE showed good accuracy (>88%) to detect Sudan dyes for real samples. In 2013, M. M. Elaysi et al. [52] detected Sudan I in food samples using Pt and carbon nanotube (CNT)-modified CPEs. Electrode performance was analyzed using CV. At the modified electrode interface CNT increased the oxidation peak current. Obtained CV results prove that CNT enhances the charge transfer kinetics between the modified electrode and Sudan I. The authors also studied the effect of various interferents like amino acids, organic biomolecules, cations (Ca 2+ , Mg 2+ ), and anions (F − , ClO4 − , SO4 2− , etc.) for the sensing of Sudan I. Later, the authors used the modified electrode to detect Sudan I in different food samples (chili sauce, chili powder, tomato sauce, and strawberry sauce). The CNT-modified electrode was able to detect the Sudan dyes in real samples with the detection limit <0.003 µM. In 2014, Sudan I was also detected by J. B. Raoof et al. [53] using ZnO NPs and ionic-liquid-modified CPEs in the presence of vitamin B6. They

Detection of Sudan
Sudan dyes (Figure 16) are fat-soluble synthetic azo dyes having wide applicability ranging from industry to scientific research. Sudan dyes are aromatic synthetic chemical dyes having (N=N-) azo groups. There are four kinds of Sudan, named Sudan I, II, III, and IV. These are used for coloring substances like oil, fats, waxes, hydrocarbon solvents, plastics, printing inks, textiles, leather, and wood. However, their use is banned in the food industry in various countries as they poses a threat to life due to their mutagenic, genotoxic, and carcinogenic properties. They are placed as category 3 toxins by the ICARC. The modified CPE showed a detection limit of 0.2 µM (K. Maleh et al. [54]). The authors proved that NiO NPs and DEDED act as electro-catalysts at the modified CPE interface. Heydari et al. [55] detected Sudan II at a CPE modified using ZnO NPs in 2019. They used CV to study the behavior of Sudan II at the ZnO NP-modified electrode and they found that NPs increase peak current 2.4 fold when compared to bare CPE. The increased peak currents at the modified electrodes are due to the increase in active surface area and charge transfer kinetics by NPs. NPs are also able to detect Sudan II in the presence of interference like different ions (Ca 2+ , Mg 2+ , Zn 2+ ), molecules (glucose, caffeine, phenol), and acids (ascorbic acid). Interferents had almost a ±5% effect on peak currents. The detection limit of the ZnO NP-modified CPE was 0.0017 µM. In the same year M. Heydari et al. [56] used the ZnO NP-modified CPE for the determination of Sudan III. Like Sudan II, ZnO NP-modified electrode increases the peak currents of Sudan II when compared to bare CPE. The ZnO NP-modified CPE showed detection of 2.56 nM for Sudan III. The authors also tested the proposed electrode to detect the Sudan III in chili sauce and ketchup, for which the results were good with a recovery limit >95%. Table 10 shows the detection limits of Sudan at the different modified carbon paste electrodes. In 2008, H. Lin et al. [51] detected Sudan I at a montmorillonite Ca modified CPE (Shown in Figure 17). The CV analysis showed a single well-defined, enhanced oxidation peak at the modified electrode. The modified electrode was also applied to examine the electrochemical behavior of other Sudan dyes (Sudan II, Sudan III, and Sudan IV) as shown in Figure 18. Each type of Sudan dye can be detected simultaneously at the Ca modified CPE. The Ca modified CPE showed good accuracy (>88%) to detect Sudan dyes for real samples. In 2013, M. M. Elaysi et al. [52] detected Sudan I in food samples using Pt and carbon nanotube (CNT)-modified CPEs. Electrode performance was analyzed using CV. At the modified electrode interface CNT increased the oxidation peak current. Obtained CV results prove that CNT enhances the charge transfer kinetics between the modified electrode and Sudan I. The authors also studied the effect of various interferents like amino acids, organic biomolecules, cations (Ca 2+ , Mg 2+ ), and anions (F − , ClO 4 − , SO 4 2− , etc.) for the sensing of Sudan I. Later, the authors used the modified electrode to detect Sudan I in different food samples (chili sauce, chili powder, tomato sauce, and strawberry sauce). The CNT-modified electrode was able to detect the Sudan dyes in real samples with the detection limit <0.003 µM. In 2014, Sudan I was also detected by J. B. Raoof et al. [53] using ZnO NPs and ionic-liquid-modified CPEs in the presence of vitamin B 6 . They found that the combination of ZnO NPs and ionic liquid will enhance charge transfer rate and sensitivity (0.5446 µA/µM), and prevents fouling of the CPE. In the same year, Sudan I was also detected simultaneously along with ascorbic acid using CPE modified from-NiO NPs and DEDED ((9,10-dihydro-9,10-ethanoanthracene-11,12-dicarboximido)-4-ethylbenzene-1,2-diol).       The modified CPE showed a detection limit of 0.2 µM (K. Maleh et al. [54]). The authors proved that NiO NPs and DEDED act as electro-catalysts at the modified CPE interface. Heydari et al. [55] detected Sudan II at a CPE modified using ZnO NPs in 2019. They used CV to study the behavior of Sudan II at the ZnO NP-modified electrode and they found that NPs increase peak current 2.4 fold when compared to bare CPE. The increased peak currents at the modified electrodes are due to the increase in active surface area and charge transfer kinetics by NPs. NPs are also able to detect Sudan II in the presence of interference like different ions (Ca 2+ , Mg 2+ , Zn 2+ ), molecules (glucose, caffeine, phenol), and acids (ascorbic acid). Interferents had almost a ±5% effect on peak currents. The detection limit of the ZnO NP-modified CPE was 0.0017 µM. In the same year M. Heydari et al. [56] used the ZnO NP-modified CPE for the determination of Sudan III. Like Sudan II, ZnO NP-modified electrode increases the peak currents of Sudan II when compared to bare CPE. The ZnO NP-modified CPE showed detection of 2.56 nM for Sudan III. The authors also tested the proposed electrode to detect the Sudan III in chili sauce and ketchup, for which the results were good with a recovery limit >95%. Table 10 shows the detection limits of Sudan at the different modified carbon paste electrodes.

Conclusions
It is concluded that the various types of dyes used in the food, cosmetic, plastic, textile, and pharmaceutical industries for imparting colors represent a large group of organic compounds that have an undesirable effect on the environment. Some of the dyes can create difficulties and risks to humans if used and consumed within the undesired limit. Therefore, it has become necessary to perform a uniform search and generate new methods that are effective and economically manageable for the determination of the dyes. Voltammetric methods are among these and are highly economical to both environment and humans, and carbon paste electrodes are widely employed in these methods. A carbon paste electrode is a heterogeneous system in which graphite powder is mixed with a pasting liquid (paraffin wax/silicone oil). They can be modified easily with various modifiers (e.g., MIP, organic molecules, inorganic molecules, CNTs, IL, etc.), so are equipped with great advantages such as low production cost, precise results, ready to use, and good stability. Due to this, they have a wide scope of applications in the detection of dyes even at very low concentration levels. Modifying the electrode also helps in improving specificity and increasing electroactive sites for interaction of the analyte for easy detection. The current review can form a basis for extensive research for using various other biomolecules for modifying CPEs apart from the above-mentioned, which can further improve the detection limits with improved accuracy and precision for a specific dye. These carbonbased modified sensors are not only helpful in dye detection but are being employed in discovering various other pollutants of wastewater such as byproducts, metal ions, organic molecules, etc.