A Wearable Patch Sensor for Simultaneous Detection of Dopamine and Glucose in Sweat

: Achieving quantiﬁcation of biomarkers in body ﬂuids is crucial to the indication of the state of a person’s body and health. Wearable sensors could offer a convenient, fast and painless sensing strategy. In this work, we fabricated a wearable electrochemical patch sensor for simultaneous detection of dopamine and glucose in sweat. The sensor was printed on a ﬂexible PDMS substrate with a simple screen-printed method. This prepared four-electrode sensor integrated two working electrodes for dopamine and glucose electrochemical sensing, one Ag/AgCl reference electrode and one carbon counter electrode, respectively. Cyclic voltammetry, differential pulse voltammetry and chronoamperometry were used for the evaluation of the wearable electrochemical patch sensor. It exhibits good sensitivity, wide linear range, low limit of detection, good anti-interference and reproducibility toward dopamine and glucose sensing in PBS and sweat.


Introduction
The concentration of biomarkers (glucose, dopamine, uric acid, Na + , H + , etc.) in body fluids is a key indicator of the state of a person's body and health. E.g., the concentration of glucose in blood is a gold standard for glucose metabolism disorder evaluation. Chronic high blood glucose level is usually associated with diabetes, yet low blood glucose level is associated with hypoglycemia [1]. Dopamine is an important neurotransmitter. Several nervous system diseases, such as Parkinson's disease, dementia with Lewy bodies, attention-deficit/hyperactivity disorder, and schizophrenia, are associated with dysfunctions of dopamine [2].
Traditional biomarker detection methods, including gas chromatography, high-performance liquid chromatography, enzyme-linked immunosorbent assay, and so on, are time consuming, costly, requiring complex equipment and professional operators. Electrochemical sensing method offers a compact, low-cost, convenient and fast strategy for biomarker detection. Plenty of efforts, including those of our research group, have been devoted to developing various electrochemical biomarker sensors [3][4][5][6]. In recent decades, the field of research and applications of wearable electrochemical sensing devices in health and environment monitoring has been boosting [7][8][9]. Biomarkers in sweat exhibit some degree of correlation with components in blood [10,11], and there is a lot of excellent works trying to detect biomarkers electrochemically in sweat as an alternative to blood testing [12][13][14][15], such as the wearable electrochemical devices for glucose [16,17], uric acid [18,19], lactate [20][21][22], dopamine [23,24], cortisol [25] and electrolyte [26,27] sensing. Dual-function or multi-function electrochemical sensing devices can provide richer sweat biomarker Analytica 2023, 4 171 information, and many works have achieved good results [22,[28][29][30]. Overall, many efforts have been made in developing diverse wearable electrochemical sensing devices for biomarkers in sweat, and they have great application potential in realizing convenient biomarker detection and reducing the psychological burden of the subjects who need to pierce the skin for blood collection. Thus, a sensor capable of simultaneous electrochemical detection of dopamine and glucose can be a promising tool for the health management and diagnosis of these two-compound-related diseases. However, to the best of our knowledge, there is still no report on a wearable electrochemical sensor for the simultaneous detection of dopamine and glucose in sweat.
In this work, we fabricated a wearable patch sensor for simultaneous electrochemical detection of dopamine and glucose in sweat. In this protocol, one of the working electrodes was used for dopamine (DA) sensing, and the other working electrode was used for glucose sensing. The prepared four-electrode electrochemical sensor was screen-printed on flexible polydimethylsiloxane (PDMS) substrate, which allows the sensor to fit comfortably on curved skin surfaces. The prepared wearable patch sensor was finally attached to skin for simultaneous electrochemical detection of dopamine and glucose in sweat.

Apparatus
All the electrochemical experiments were carried out using a CHI-760e workstation (CH Instruments, Shanghai, China). The electrode printing stencil was cut out on a vinyl transfer film using the Roland desktop cutter GS-24. Then, the electrodes were printed on the flexible PDMS substrate.

Preparation of PDMS Substrate
For preparation of the flexible PDMS substrate, a cuboid groove with a length of 20 mm, width of 15 mm and height of 0.3 mm was first prepared. The groove was made of polyethylene terephthalate (PET) as the base and tape as the walls. Then, a mixture of part A (SYLGARD 184 silicone elastomer kit) mixing with part B (SYLGARD 184 silicone elastomer kit) at a mass ratio of 10:1 was poured evenly into the groove. Subsequently, the entire setup was put in vacuum oven at 75 • C for 40 min. Finally, a flexible PDMS substrate was obtained and the tape was removed before use. The schematic diagram of PDMS substrate preparation process can be seen in Figure 1.

Printing of the Electrodes
The electrode printing process is similar to that in our previous report but with a few modifications [3]. In detail, the electrode printing stencil was designed using Adobe Illustrator software before printing. Then, the vinyl stencil was cut out using the Roland desktop cutter. Subsequently, the vinyl stencil transfer films were progressively pasted and removed after the corresponding inks were progressively applied on the flexible PDMS substrate. It should be pointed out that, after each kind of ink was applied, the electrode setup was put in oven at 60 °C for 10 min. The printed four electrodes consisted of two working electrodes of Prussian blue (PB) modified carbon electrode (WE-PB) and unmodified carbon electrode (bare WE), reference electrode (RE) of Ag/AgCl, and counter electrode (CE) of unmodified carbon, respectively. The schematic diagram of the electrodes printing process can be found in Figure 1.

Preparation of Dopamine Biosensor (Sensor-DA)
The sensor for DA sensing was prepared by modifying MWCNT-COOH on screenprinting carbon electrode (SPCE). In detail, 1 mg MWCNT-COOH was mixed with 1 mL 2 wt% Nafion solution, which was previously prepared by diluting 5 wt% Nafion stock in deionized water. Then, the mixture was treated by ultrasound for half an hour. Finally, 3 µL MWCNT-COOH/Nafion composite was drop-cast on bare WE.

Preparation of Glucose Biosensor (Sensor-Glucose)
Firstly, the MWCNT-COOH/chitosan solution was prepared by dispersing 2 mg MWCNT-COOH in 1 mL 1 wt% acetic acid and 2 mL chitosan solution. Then, the MWCNT-COOH/chitosan solution was mixed with 40 mg/mL GOD solution, which was prepared by dissolving GOD in PBS, at a volume ratio of 2:1. Finally, 3 µL GOD/MWCNT-COOH/chitosan composite was drop-cast on WE-PB and stored in a fridge at 4 °C overnight.

Patch Sensor Design and Electrochemical Sensing Mechanism
The prepared wearable patch sensor can be tightly attached to the skin for continuous monitoring of biomarkers in sweat, as shown in Figure 2A. Figure 2B indicates the sensor's design sizes. This four-electrode patch sensing system was printed on a flexible PDMS membrane with a simple screen-printing technique. Figure 2C shows the digital photo of the patch sensor, and it resembles a "panda" shape, as shown in Figure 2D.

Printing of the Electrodes
The electrode printing process is similar to that in our previous report but with a few modifications [3]. In detail, the electrode printing stencil was designed using Adobe Illustrator software before printing. Then, the vinyl stencil was cut out using the Roland desktop cutter. Subsequently, the vinyl stencil transfer films were progressively pasted and removed after the corresponding inks were progressively applied on the flexible PDMS substrate. It should be pointed out that, after each kind of ink was applied, the electrode setup was put in oven at 60 • C for 10 min. The printed four electrodes consisted of two working electrodes of Prussian blue (PB) modified carbon electrode (WE-PB) and unmodified carbon electrode (bare WE), reference electrode (RE) of Ag/AgCl, and counter electrode (CE) of unmodified carbon, respectively. The schematic diagram of the electrodes printing process can be found in Figure 1.

Preparation of Dopamine Biosensor (Sensor-DA)
The sensor for DA sensing was prepared by modifying MWCNT-COOH on screenprinting carbon electrode (SPCE). In detail, 1 mg MWCNT-COOH was mixed with 1 mL 2 wt% Nafion solution, which was previously prepared by diluting 5 wt% Nafion stock in deionized water. Then, the mixture was treated by ultrasound for half an hour. Finally, 3 µL MWCNT-COOH/Nafion composite was drop-cast on bare WE.

Preparation of Glucose Biosensor (Sensor-Glucose)
Firstly, the MWCNT-COOH/chitosan solution was prepared by dispersing 2 mg MWCNT-COOH in 1 mL 1 wt% acetic acid and 2 mL chitosan solution. Then, the MWCNT-COOH/chitosan solution was mixed with 40 mg/mL GOD solution, which was prepared by dissolving GOD in PBS, at a volume ratio of 2:1. Finally, 3 µL GOD/MWCNT-COOH/chitosan composite was drop-cast on WE-PB and stored in a fridge at 4 • C overnight.

Patch Sensor Design and Electrochemical Sensing Mechanism
The prepared wearable patch sensor can be tightly attached to the skin for continuous monitoring of biomarkers in sweat, as shown in Figure 2A. Figure 2B indicates the sensor's design sizes. This four-electrode patch sensing system was printed on a flexible PDMS membrane with a simple screen-printing technique. Figure 2C shows the digital photo of the patch sensor, and it resembles a "panda" shape, as shown in Figure 2D.
The electrochemical sensing mechanism of the wearable patch sensor is based on the enzymatic reaction of glucose oxidized on GOD-modified electrode in the presence of oxygen integrating with the electrochemical reaction of hydrogen peroxide reduced on a PBmodified carbon electrode. On one of these working electrodes (Sensor-glucose), hydrogen peroxide was produced enzymatically by GOD, and its concentration was proportional to glucose concentration. Then, hydrogen peroxide was electrochemically reduced on WE-PB. On the other working electrode (Sensor-DA), DA was electrochemically oxidized on the MWCNT-COOH-modified carbon electrode to produce dopaminequinone by a two-electron reaction process [31]. The modification process and electrochemical sensing mechanism of the wearable patch sensor are illustrated in Figure 3. The electrochemical sensing mechanism of the wearable patch sensor is based on t enzymatic reaction of glucose oxidized on GOD-modified electrode in the presence of o ygen integrating with the electrochemical reaction of hydrogen peroxide reduced on a P modified carbon electrode. On one of these working electrodes (Sensor-glucose), hydr gen peroxide was produced enzymatically by GOD, and its concentration was propo tional to glucose concentration. Then, hydrogen peroxide was electrochemically reduc on WE-PB. On the other working electrode (Sensor-DA), DA was electrochemically o dized on the MWCNT-COOH-modified carbon electrode to produce dopaminequino by a two-electron reaction process [31]. The modification process and electrochemic sensing mechanism of the wearable patch sensor are illustrated in Figure 3.  The electrochemical sensing mechanism of the wearable patch sensor is based on enzymatic reaction of glucose oxidized on GOD-modified electrode in the presence of ygen integrating with the electrochemical reaction of hydrogen peroxide reduced on a P modified carbon electrode. On one of these working electrodes (Sensor-glucose), hyd gen peroxide was produced enzymatically by GOD, and its concentration was prop tional to glucose concentration. Then, hydrogen peroxide was electrochemically reduc on WE-PB. On the other working electrode (Sensor-DA), DA was electrochemically o dized on the MWCNT-COOH-modified carbon electrode to produce dopaminequino by a two-electron reaction process [31]. The modification process and electrochemi sensing mechanism of the wearable patch sensor are illustrated in Figure 3.

Electrochemical Sensing Performance of Sensor-DA toward Dopamine
To investigate the electron-transfer property of SPCE after the modification of MWC-NTs, electrochemical impedance spectroscopy (EIS) was carried out in 0.1 M KCl solution containing 5 mM KFe II [Fe III (CN) 6 ]. The potential was controlled at the open circuit potential (OCP) of 0.17 V. As Figure 4A shows, two classic EIS Nyquist plots of SPCE and Sensor-DA were obtained, and the semicircle plots in high-frequency region corresponded to the dynamic process. It is obvious that, after the modification of MWCNTs on SPCE, the interfacial electron-transfer resistance (R et ) increases. This is due to the formation of different kinetic barriers after the modification, resulting in the increase in R et [5]. This indicates the successful modification of MWCNTs on SPCE. Moreover, the electrochemical sensing performance of the Sensor-DA toward DA was investigated using cyclic voltammetry (CV), which was performed in 0.1 M PBS, pH 7.4, at a scan rate of 50 mV/s, potential ranging from −0.35 V to 0.8 V (vs. Ag/AgCl). The electrochemical cyclic voltammograms of bare WE and Sensor-DA are shown in Figure 4B. The inset showed the cyclic voltammograms of the bare WE in absence and presence of 80 µM DA. In the absence of DA, the CV of bare WE showed classic cyclic voltammogram of carbon in PBS with no obvious redox peak. After the addition of 80 µM DA, an anodic peak at 0.5 V and a cathodic peak at −0.1 V were found. These redox peaks should be ascribed to the electrochemical redox reactions between DA and dopaminequinone of a two-electron reaction process [31,32]. Moreover, compared to bare WE, after the modification of MWCNT-COOH, Sensor-DA showed an increased double layer capacitance in PBS without the presence of DA. It revealed the successful modification of MWCNT-COOH on bare WE. In the presence of 80 µM DA, Sensor-DA showed a pair of well-defined redox peaks, where the anodic peak was at 0.14 V and the cathodic one was at 0.01 V. The above result showed that, compared to bare WE, Sensor-DA characterized lower redox overpotential and more comparable redox peaks current value toward the electrochemical reaction of DA in PBS. It showed the enhanced electrocatalytic activity of MWCNT-COOH compared to unmodified carbon electrode. Figure 4C Figure 4D illustrated, the DPV anodic peak current increased with the increase in DA concentrations. The inset showed the calibration curve of the relationship between response peak current against DA concentration. The limit of detection (LOD) of Sensor-DA toward DA was calculated as 0.043 µM (S/N = 3). Moreover, the curve could be divided into two parts. At low concentration of DA, the electrocatalytic oxidation mechanism is dominant; yet at high concentration of DA, the ability of surface electro-oxidation is crucial for current response [33]. The linear ranges for these two calibration curves were from 0 to 70 µM with the linear correlation coefficients (R 2 ) of 0.9938, and 70 µM to 180 µM with the R 2 of 0.9961, respectively. The equations for these two linear fitting curves are as follows:

Electrochemical Sensing Performance for Glucose
The electrochemical sensing performance of Sensor-glucose was investigated in 0.1 M PBS, pH 7.4. The electrochemical sensing mechanism of Sensor-glucose is based on the enzymatic reaction of glucose by GOD. Glucose can be enzymatically oxidized in the presence of oxygen and GOD to produce hydrogen peroxide. Then, hydrogen peroxide can be further electrochemically reduced on the PB-modified electrode [34]. The current value of the final response is proportional to the glucose concentration. Thus, the electrochemical sensing performance of the PB-modified carbon electrode was investigated first. As Figure  5A shows, WE-PB showed a pair of well-defined redox potential of PB in PBS solution. The anodic peak was at 0.13 V and the cathodic one was at 0.06 V. It is needed to be pointed out that the redox peak difference (around 0.7 V) is close to the theoretical reversible value of one electron transport. This should be ascribed to the reversible electrochemical redox reaction for interconversion between Prussian blue and Prussian white [35]. As hydrogen peroxide concentration increased from 0.4 to 1 mM, the electrochemical reduction peak current increased. This showed the electrochemical sensing ability of PB-modified carbon electrode toward hydrogen peroxide. Figure 5B shows the chronoamperometry response of WE-PB to the different concentration of hydrogen peroxide from 0 to 5.1 mM in 0.1 M PBS, pH 7.4. A constant potential of 0.1 V was applied for 60 s, where the current response value was read. As hydrogen peroxide concentrations increased, the electrochemical reduction current increased. Correspondingly, the calibration curve was obtained as shown

Electrochemical Sensing Performance for Glucose
The electrochemical sensing performance of Sensor-glucose was investigated in 0.1 M PBS, pH 7.4. The electrochemical sensing mechanism of Sensor-glucose is based on the enzymatic reaction of glucose by GOD. Glucose can be enzymatically oxidized in the presence of oxygen and GOD to produce hydrogen peroxide. Then, hydrogen peroxide can be further electrochemically reduced on the PB-modified electrode [34]. The current value of the final response is proportional to the glucose concentration. Thus, the electrochemical sensing performance of the PB-modified carbon electrode was investigated first. As Figure 5A shows, WE-PB showed a pair of well-defined redox potential of PB in PBS solution. The anodic peak was at 0.13 V and the cathodic one was at 0.06 V. It is needed to be pointed out that the redox peak difference (around 0.7 V) is close to the theoretical reversible value of one electron transport. This should be ascribed to the reversible electrochemical redox reaction for interconversion between Prussian blue and Prussian white [35]. As hydrogen peroxide concentration increased from 0.4 to 1 mM, the electrochemical reduction peak current increased. This showed the electrochemical sensing ability of PB-modified carbon electrode toward hydrogen peroxide. Figure 5B shows the chronoamperometry response of WE-PB to the different concentration of hydrogen peroxide from 0 to 5.1 mM in 0.1 M PBS, pH 7.4. A constant potential of 0.1 V was applied for 60 s, where the current response value was read. As hydrogen peroxide concentrations increased, the electrochemical reduction current increased. Correspondingly, the calibration curve was obtained as shown in the inset. The linear range of WE-PB sensor toward hydrogen peroxide was from 0 to 3.  To investigate the electrochemical sensing performance of Sensor-glucose, CVs of Sensor-glucose in 0.1 M PBS, pH 7.4 were performed. As Figure 5C shows, the electrochemical reduction current increased when the glucose concentration increased from 0 to 1.9 mM. This showed the electrochemical sensing ability of Sensor-glucose toward glucose. To further investigate its sensing performance, chronoamperometry was carried out in 0.1 M PBS, pH 7.4, as shown in Figure 5D. A constant potential of 0.1 V was applied for 60 s, where the current response value was read. As glucose concentrations increased from 0 to 13 mM, the electrochemical reduction current increased. The corresponding calibration curve was obtained as shown in the inset. The linear range of Sensor-glucose toward glucose was from 0 to 11.

Anti-Interference Ability, Reproducibility and Stability
Sweat contains a rich variety of biomarkers, such as glucose, dopamine, uric acid, urea, lactic acid, ascorbic acid, and so on. Many of these compounds are also electrochemically active. Anti-interference ability is also a key indicator for evaluating sensor To investigate the electrochemical sensing performance of Sensor-glucose, CVs of Sensor-glucose in 0.1 M PBS, pH 7.4 were performed. As Figure 5C shows, the electrochemical reduction current increased when the glucose concentration increased from 0 to 1.9 mM. This showed the electrochemical sensing ability of Sensor-glucose toward glucose. To further investigate its sensing performance, chronoamperometry was carried out in 0.1 M PBS, pH 7.4, as shown in Figure 5D. A constant potential of 0.1 V was applied for 60 s, where the current response value was read. As glucose concentrations increased from 0 to 13 mM, the electrochemical reduction current increased. The corresponding calibration curve was obtained as shown in the inset. The linear range of Sensor-glucose toward glucose was from 0 to 11.5 mM with the LOD of 40.36 µM (S/N = 3). The linear fitting equation is as follows: I (µA) = −0.7117×c(glucose) (mM) − 0.0322(R 2 = 0.9989).

Anti-Interference Ability, Reproducibility and Stability
Sweat contains a rich variety of biomarkers, such as glucose, dopamine, uric acid, urea, lactic acid, ascorbic acid, and so on. Many of these compounds are also electrochemically active. Anti-interference ability is also a key indicator for evaluating sensor performance. To evaluate the anti-interference ability of the patch sensor, chronoamperometry was carried out. Figure 6A shows the Sensor-DA's chronoamperometric response toward combinations of dopamine and different species. A constant potential of 0 V (vs.Ag/AgCl) was controlled for 60 s, and the current response against time was recorded. It can be found that the Sensor-DA provided a current response around 0.3 µA to 20 µM DA. However, compared to a single 20 µM DA, the Sensor-DA provided a similar current response to the compositions containing 20 µM DA and different interferences. This indicated that the Sensor-DA has a good anti-interference ability. Similarly, the interference ability of the Sensor-glucose was investigated by chronoamperometric test. A constant potential of 0.1 V (vs.Ag/AgCl) was controlled for 60 s. As Figure 6B shows, the Sensor-glucose provided a 1.23 µA current response to 300 µM glucose, but a similar response to compositions containing 300 µM glucose and different interferences. This also indicated that the Sensor-glucose has a good anti-interference ability.
Analytica 2023, 4, FOR PEER REVIEW 8 performance. To evaluate the anti-interference ability of the patch sensor, chronoamperometry was carried out. Figure 6A shows the Sensor-DA's chronoamperometric response toward combinations of dopamine and different species. A constant potential of 0 V (vs.Ag/AgCl) was controlled for 60 s, and the current response against time was recorded. It can be found that the Sensor-DA provided a current response around 0.3 µA to 20 µM DA. However, compared to a single 20 µM DA, the Sensor-DA provided a similar current response to the compositions containing 20 µM DA and different interferences. This indicated that the Sensor-DA has a good anti-interference ability. Similarly, the interference ability of the Sensor-glucose was investigated by chronoamperometric test. A constant potential of 0.1 V (vs.Ag/AgCl) was controlled for 60 s. As Figure 6B shows, the Sensor-glucose provided a 1.23 µA current response to 300 µM glucose, but a similar response to compositions containing 300 µM glucose and different interferences. This also indicated that the Sensor-glucose has a good anti-interference ability.  To investigate the reproducibility of the patch sensor, 15 chronoamperometric tests were performed on Sensor-DA and Sensor-glucose in PBS solutions containing 70 µM DA and 5 mM glucose, respectively. The applied constant potentials were controlled at 0 V and 0.1 V, respectively. Figure 6C shows the current response percentage of Sensor-DA of the 15 tests relative to the first one, and the inset shows the recorded 15 chronoamperometric tests. The relative standard deviation (RSD) of the 15 test response values was 1.184%. Similarly, Figure 6D shows the current response percentage of Sensor-glucose of the 15 tests relative to the first one, and the inset shows the recorded 15 chronoamperometric tests. The response revealed an RSD of 1.443%. These results showed that there was no significant difference in the 15 chronoamperometric response values of the patch sensor, demonstrating good reproducibility.
The stability of the wearable electrochemical patch sensor was also evaluated for five days. As shown in Figure 6E,F, stability tests of the patch sensor were performed daily for five days, in a PBS solution containing, respectively, 20 µM DA and 500 µM glucose. The insets show the corresponding chronoamperometric responses. It can be found that, compared to the first day, the current response of the Sensor-DA has almost no attenuation. This is ascribed to the stability of the inorganic material of carbon ink and carbon nanotubes. However, the current response of the Sensor-glucose decreased at day 5, but remained at 91.9%. This is due to the intrinsic instability of biomaterials of glucose oxidase.

Electrochemical Sensing in Sweat
The evaluation of real sample sensing performance of the patch sensor was carried out in sweat, which was collected by a volunteer after exercise. DPV and chronoamperometry were applied for Sensor-DA and Sensor-glucose, respectively. The electrochemical parameters of the evaluations were the same with the ones for the patch sensor in PBS. Figure 7A shows that the DPV peak current of Sensor-DA increased with the increase in DA concentrations, and the inset shows the corresponding calibration curve. Similar to the case in PBS, it can be found that the calibration curve was split into two parts within the linear ranges from 0 to 50 µM and 50 µM to 120 µM, respectively. An LOD of 0.065 µM was obtained (S/N), and these two linear equations followed: Analytica 2023, 4, FOR PEER REVIEW 9 To investigate the reproducibility of the patch sensor, 15 chronoamperometric tests were performed on Sensor-DA and Sensor-glucose in PBS solutions containing 70 µM DA and 5 mM glucose, respectively. The applied constant potentials were controlled at 0 V and 0.1 V, respectively. Figure 6C shows the current response percentage of Sensor-DA of the 15 tests relative to the first one, and the inset shows the recorded 15 chronoamperometric tests. The relative standard deviation (RSD) of the 15 test response values was 1.184%. Similarly, Figure 6D shows the current response percentage of Sensor-glucose of the 15 tests relative to the first one, and the inset shows the recorded 15 chronoamperometric tests. The response revealed an RSD of 1.443%. These results showed that there was no significant difference in the 15 chronoamperometric response values of the patch sensor, demonstrating good reproducibility.
The stability of the wearable electrochemical patch sensor was also evaluated for five days. As shown in Figure 6E,F, stability tests of the patch sensor were performed daily for five days, in a PBS solution containing, respectively, 20 µM DA and 500 µM glucose. The insets show the corresponding chronoamperometric responses. It can be found that, compared to the first day, the current response of the Sensor-DA has almost no attenuation. This is ascribed to the stability of the inorganic material of carbon ink and carbon nanotubes. However, the current response of the Sensor-glucose decreased at day 5, but remained at 91.9%. This is due to the intrinsic instability of biomaterials of glucose oxidase.

Electrochemical Sensing in Sweat
The evaluation of real sample sensing performance of the patch sensor was carried out in sweat, which was collected by a volunteer after exercise. DPV and chronoamperometry were applied for Sensor-DA and Sensor-glucose, respectively. The electrochemical parameters of the evaluations were the same with the ones for the patch sensor in PBS. Figure 7A shows that the DPV peak current of Sensor-DA increased with the increase in DA concentrations, and the inset shows the corresponding calibration curve. Similar to the case in PBS, it can be found that the calibration curve was split into two parts within the linear ranges from 0 to 50 µM and 50 µM to 120 µM, respectively. An LOD of 0.065 µM was obtained (S/N), and these two linear equations followed:    Figure 7B shows the chronoamperometric current response of Sensor-glucose to different concentrations of glucose in sweat. In addition, the electrochemical reduction current increased with the glucose concentration increasing. The inset shows the corresponding calibration curve, and the linear range was from 0 to 8 mM. The LOD was calculated as 55.65 µM. The linear fitting equation is as follows: I (µA) = −0.5124×c(glucose)(mM) + 0.0428(R 2 = 0.9941).
It is worth mentioning that the linear detection range of this wearable electrochemical patch sensor for glucose covers well its physiological concentration range in sweat from 0.01 to 1.11 mM [36]. The comparison of the electrochemical sensing performance of the wearable patch sensor with other reported ones for dopamine and glucose sensing in sweat is listed in Table 1. It demonstrates good electrochemical sensing performance of the wearable patch sensor.

Conclusions
In this work, we fabricated a wearable electrochemical patch sensor for simultaneous detection of dopamine and glucose in sweat. The prepared sensor was printed on flexible PDMS substrate, which can fit comfortably on curved skin surfaces. The Sensor-DA was prepared by the modification of MWCNT-COOH on screen-printed carbon electrode, exhibiting an enhanced dopamine electrochemical sensing performance compared to the unmodified electrode. The Sensor-glucose was prepared by immobilizing glucose oxidase on Prussian blue modified screen-printed carbon electrode, exhibiting high sensing sensitivity and specificity toward glucose electrochemical sensing. Moreover, the wearable patch sensor's electrochemical performance was fully evaluated in PBS, and it exhibited good sensitivity, wide linear range, low limit of detection, good anti-interference ability and reproducibility. We believe that the wearable electrochemical patch sensor's sensing performance, such as the limit of detection, can be further improved after integrating other functional materials. Finally, the wearable patch sensor was evaluated for dopamine and glucose electrochemical sensing in real sweat, demonstrating its potential application in health management and diagnosis of these two-compound-related diseases.