Iridium Oxide-reduced Graphene Oxide Nanohybrid Thin Film Modified Screen-printed Electrodes as Disposable Electrochemical Paper Microfluidic pH Sensors

Device Setup

The pH sensor works by measuring the OCP between the working and reference electrodes, since it changes proportionally to the negative logarithm of H⁺ concentration. The measurements can be achieved both by a lab-based potentiostat such as CHI 660D and simple pH meter constructed on breadboard with reading by multimeter. Two different portable pH meters were built similarly on breadboards using two 9 V alkaline batteries, a digital multimeter, as-synthesized IrO₂-RGO-SPE and different OpAmps, which is two serial LF356N or one INA111. The entire cost of such pH meters is typically under $25 including around $1 apiece of disposable pH IrO₂-RGO-SPEs costing (cost and time to conduct chemical preparation and electrochemical depositions are not included). Figure 1C shows the circuit diagrams and connections with two LF356N or one INA111. Schematic of the entire device according to the connection pins of INA111 is shown in Figure 1D. Two 9 V batteries were used to create a dual supply of +/- 9 V to power the op amps by connecting to the input pins 2 and 3. A 1/4-W 4-band gain resistor (R_G) of 10 MΩ (5% tolerance) is connected between pins 1 and 8, resulting in a desired gain (G) close to 1, as calculated from Eq (1). Due to the comparatively low internal resistance of the multimeter, a sufficiently large load resistor is grounded and connected between shorted pins 5 and 6, which further connect the negative and positive probes of multimeters for stable OCP readings. As a result, the output voltage which can be calculated from Eq (2) is exactly equal to the potential difference between IrO₂-RGO working electrode and Ag/AgCl reference electrode, corresponding to the solution pH.

Where G is the gain of INA111, V_out is sensor output (V), and V⁺_で and V^–_で are voltage at the non-inverting and inverting inputs of OpAmps, respectively.

Surface Morphology

Figure 2A scanning electron microscopy (SEM) image shows the surface morphology of the intermittent RGO film formed with a typical wrinkled texture composed of flexible and ultrathin graphene sheets spreading through the rough screen-printed carbon electrode surface. The GO reduction occurs at potentials more negative than -0.65 V and is rapid, irreversible and controllable without reducing reagents⁷. After IrO₂ deposition, the black RGO film visually and notably turned dark purple, showing the characteristic electrochromic feature of anodically electrodeposited IrO₂. IrO₂-RGO nanohybrid thin film (Figure 2C) in µm-range thickness is observed with homogeneously dispersed nanoscale crystallites on the entire surface. The film appears to be uniform and smooth, without any noticeable cracks or mud structures. Moreover, an interesting finding in the current study is that IrO₂ films cannot form on the screen-printed graphitic carbon substrates without the assistance of RGO under the same conditions (Figure 2B). Also the good film forming ability of graphene is conceived the key component in the thin film deposition process. The oxalates in the deposition solution can trigger complex formation and prevent iridium precipitation in alkaline medium. The deposition is achieved by anodic oxidation of oxalated Ir(IV) compounds with CO₂ release and concomitant Ir(IV) oxide formation at the anode surface, as described by Yamanaka^8,9

[Ir(COO)₂(OH)₄]^2– ⇔ IrO₂ + 2CO₂ + 2H₂O + 2e^–

pH Measurement Performance

A general mechanism of metal oxides in pH sensing has been indicated by Fog and Buck to be associated with the ion-exchange processes within the OH group-bearing surface¹⁰. Different IrO₂-based electrodes exist in nature due to the differences in their synthesis mechanisms. A number of possible mechanisms are proposed involving pH-dependent redox intercalation equilibrium between two oxidation states of Ir oxides¹¹. As in the case of electrochemically-synthesized Ir oxides, the predominant state is the hydrated form⁸ which delivers super-Nernstian responses with a sensitivity superior than that of anhydrous forms at 59 mV/pH. The mechanism can be explained using the following equation^8,9:

2[IrO₂(OH)₂ · 2H₂O)]^2– + 3H⁺ + 2e^– ⇔ 2[IrO₂(OH)₂ · 2H₂O)]^2– + 3H₂O

As shown in Figure 3A (black), the IrO₂-RGO electrode showed a well-defined linear characteristic over a wide pH range from 2 to 12, with a slightly super-Nernstian slope of -61.71 mV/pH which is closer to anodically electrodeposited IrO₂ films with near-Nerstian responses^11,12 rather than those with super Nernstian^13,14. Possibly this happens due to the different electrodeposition conditions and substrates in each case. It is possible that a mixture of anhydrous and hydrated IrO₂ film is formed on graphene during the anodic electrodeposition. The negligible difference of measured pH between patterned paper microfluidic device (Figure 3A, red) and bulk solution indicates the fibrous cellulose paper matrix does not impede diffusion of hydrogen ions to any noticeable degree. Dissolved atmospheric oxygen exists ubiquitously in samples can sometimes greatly influence potential readings due to its redox processes, especially in biological systems¹². When the electrode is placed in N₂– or air-saturated buffer solution, there are only minor differences (Figure 3B). Meanwhile, since the ionic strengths and compositions are different depending on buffers, a number of pH buffers were tested including commercial pH calibration buffer, B-R buffer, phosphate buffered saline (PBS) and NaOH/HCl adjusted DI water (Figure 3C). The sensitivities (mV per unit pH) are almost identical in all buffers. However, an appreciable potential drift was observed in PBS buffer which has a relatively narrower pH range. This may be attributed to different conditional standard potentials (E^0' term in the equilibrium Nernst equation) of IrO₂-RGO pH electrode in PBS.

Response time, a key factor in any sensing application, is usually defined as the time required reaching certain percentages of equilibrium OCP. Typical response time is less than 250 sec under all pH but could be strongly pH-dependent¹¹. Hysteresis, or the so-called memory effect, is a well-known phenomenon with glass and metal oxide pH electrodes during repetitive uses of the same electrode. This phenomenon of hydrogen ion-selective electrodes is considered as the result of delayed responses to the pH change. The IrO₂-RGO pH electrode was tested in pH buffers from low to high, and in turn from high to low. This loop cycles of pH 2-12-2 and 12-2-12 were assessed by successively measuring OCP of different pH buffers in the cycles (Figure 4B). The hysteresis widths are calculated to be around 13 mV in both cycles, which are acceptable and accurate in routine pH measurements, particularly regarding the wide pH range studied here.

To further validate the IrO₂-RGO nanohybrid thin film pH sensor, the performance was tested in parallel with a standard glass electrode with a pH/mV/Ion/Conductivity Meter (Figure 5A). The results correlate well each other, suggesting the high accuracy and reliable performance of the developed pH sensor. To truly achieve on-site pH measurements featuring the portable electrochemical paper microfluidic device, two simple digital pH meters were built with slightly different configurations. Measured pH using both meters was highly consistent with the lab-based electrochemical analyzer. Reproducibility was tested multiple times with the same electrode in different pH B-R buffers, and also with different electrodes. The relative standard deviation (RSD) values were typically all <15%.

In conclusion, a rapid, controllable and green electrochemical method is developed for synthesis of uniform and smooth IrO₂-RGO nanohybrid thin films on rough SPE surface, assisted by the good film-forming and supporting capabilities of RGO. The resulting solid-state electrode exhibited a slightly super-Nernstian response with a high sensitivity of -62 mV/pH and with good linearity in a wide range of pH from 2 to 12. The electrodes also have small hysteresis, fast response time, reproducibility, and good agreements with commercial glass electrode equipped lab-based pH meter. A miniaturized paper microfluidic pad was fabricated by PDMS hydrophobic barrier patterning. A simple pH meter was fabricated using two 9 V alkaline batteries, a digital multimeter, and OpAmps. Measured pH results from the sensor correlated well with those obtained using lab-based techniques. Thus the sensor which combines advantages of both pH meters and pH strips is a promising platform for future on-site or point-of-care pH measurements, especially under limited-resources status.