I. Core Research Summary

• Complete integrated OECT fabrication system for all aerosols

Ultrasonic atomization of silver and PEDOT:PSS inks, pneumatic atomization of UV‑PDMS insulating ink, and single‑device layer‑by‑layer printing of multilayer structures; channel thickness of 560 nm, with a silver film resistivity as low as 5×10⁻⁶ Ω·cm, and device transconductance reaching 1.5 mS, exhibiting intrinsic signal amplification without the need for external amplification circuitry.

• Ultra-wide linear detection range, ultra-low limit of detection

In a pure water matrix, the detection limit is 0.1 nM, with a measurement range spanning 0.1 nM to 5 mM; in an artificial saliva environment, the detection limit is 1 nM, fully encompassing the physiological THC concentrations in human saliva (3.18–1039.88 nM), exhibiting two‑segment linear response across low and high concentration ranges, with a measurement error of less than 3%.

• Ultra-fast detection of trace samples, with extremely simple operation.

Only 2–10 μL of saliva is required, with stable signal acquisition lasting approximately 5 minutes and no complex sample preparation needed. Repeated measurements demonstrate high signal consistency, and a single device exhibits a cycle-to-cycle measurement deviation of less than 2.65% across seven cycles.

• Stable recognition of complex salivary matrices

Even in the presence of multiple ions and organic interferents, artificial saliva can accurately distinguish THC concentrations, with only brief baseline fluctuations; once equilibrium is reached, the current response remains stable, demonstrating suitability for on-site sample analysis.

• Low-cost, single-use sensor chips are scalable.

Digital maskless printing, combined with a flexible, lightweight, and portable substrate, enables rapid iteration of device architectures. Subsequent functionalization with THC antibodies can further enhance specificity, making it well-suited for the mass production of security‑screening consumables.

II. Complete Preparation Process and Core Advantages of Aerosol Spraying

(1) Complete OECT Aerosol Printing Process

• 25.4 μm Kapton flexible film substrate cleaning and pre-treatment;
• Ultrasonic atomization of silver nano‑ink, aerosol printing source / leakage electrode, followed by overnight annealing at 140°C to reduce resistivity;
• Ultrasonic atomization–modified PEDOT:PSS composite ink, with a double-layer printed sensing channel, cured and formed at 130°C.
• Pneumatically atomized and diluted UV‑PDMS insulating ink is used to precisely print the electrode passivation layer, followed by 405 nm UV‑induced simultaneous crosslinking to prevent electrolyte‑induced short circuits.
• An electrochemical detection platform was assembled using an external platinum wire grid, and a saliva sample was dropwise added to acquire the THC signal.

(II) Five Major Process Advantages (Compared with Inkjet, Screen Printing, and Photolithography for OECT Fabrication)

• Three-layer heterogeneous materials integrated into a single device via one-step molding.

Ultrasonic processing is compatible with conductive polymers and metal inks, while pneumatic processing works well with high-viscosity insulating adhesives. This enables the simultaneous fabrication of electrodes, sensing channels, and insulating protective layers, whereas photolithography requires multiple mask‑alignment steps, evaporation, and etching processes, significantly reducing the overall cycle time.

• Micron-level precise partitioned deposition, with precisely defined openings in the insulating layer.

The printing area is digitally constrained, with PDMS coating only the conductive leads while leaving the sensing channels fully exposed. The electrolyte contacts only the functional regions, thereby eliminating leakage currents and inter-channel crosstalk and enhancing the signal-to-noise ratio.

• Low-temperature, mild processing, compatible with flexible polymer substrates.

The maximum curing temperature is only 140°C, preventing warping and embrittlement of the PI film. The ultra-thin film can be laminated onto portable test‑strip carriers, making it ideal for disposable, adhesive‑type test cards.

• Maskless Rapid Iterative Sensing Configuration

With CAD, the channel aspect ratio and electrode dimensions can be adjusted with a single click, enabling rapid optimization of OECT transconductance for different toxicant molecules; new screening chip prototypes can be completed in just a few hours.

• Consumables are deposited on demand, resulting in low mass-production costs.

Only the sensing area is coated with noble metals and conductive polymers, resulting in a 70% material loss across the entire film; the flexible substrate is inexpensive and readily available, making it well suited for high‑volume, single‑use detection consumables in transportation and security applications.

III. Five Key Technological Innovations

Innovation 1: Innovative Preparation of an All-Aerosol Integrated THC Detection OECT

Innovation 2: Leveraging a novel label-free sensing mechanism based on electrochemical oxidation of THC.

Innovative 3 OECT endogenous signal amplification system, eliminating the need for an external amplification module.

Innovation 4: A Hierarchical Linear Sensing Model Adapted to Complex Saliva Matrices

Innovation 5: A Universal Aerosol-Printing Platform for Standardizing Toxicity Sensing

IV. Existing Process Challenges and Corresponding Solutions for Overcoming Them

Challenge 1: The layered fabrication of multi-material OECTs is cumbersome, and conventional processes require multiple pieces of equipment.

Challenge 2: Conventional printed electrodes produce weak signals, and low-concentration THC is easily masked by noise.

Challenge 3: Saliva contains a large number of ionic impurities, which can easily lead to baseline drift and inaccurate quantification.

Challenge 4: The lithography-based OECT prototyping process can take several days, making rapid iterative screening of chips impractical.

Challenge 5: The unmodified bare-gate exhibits insufficient specificity and is easily interfered with by similar phenolic compounds.

V. Future Diversified Industrial Application Scenarios

1. On-site rapid screening for driving under the influence of alcohol or drugs

2. Mass screening at airports, train stations, and border checkpoints

3. Long-term medication monitoring at drug rehabilitation facilities

4. Rapid Preliminary Screening in Forensic Toxicology

5. A universal detection chip platform for multi-component body fluids

Figure 1. (a) Microscopic image of an aerosol‑jet‑printed OECT, with a scale bar of 500 µm. The channel region is shown in dark blue (PEDOT:PSS), covering the silver electrode tip. (b) Potential drop across the two electrical double layers (EDLs) at the platinum gate electrode surface and the PEDOT:PSS channel surface; capacitances CGE and CCE are used as electrical models to characterize the EDLs. (c) Schematic diagram of a D9‑THC sensor based on the OECT. (d) Chemical structure of Δ9‑tetrahydrocannabinol (D9‑THC) and the electrooxidation process of the phenolic group in its molecule (references 66, 67). (e) Cyclic voltammograms obtained using a water‑diluted D9‑THC solution. (f) Plot of the anodic peak current (Ip,a) as a function of D9‑THC concentration; Ip,a values were extracted from data acquired at a potential of 0.26 V for each D9‑THC concentration shown in (e).

Figure 2 shows the drain current (ID) response curves of aerosol‑jet‑printed OECTs under a fixed bias condition of VD = 0.1 V and VG = 1 V, following the addition of D9‑THC at different concentrations. These measurements were performed on three distinct OECT devices: (a) and (b) correspond to different devices, while (c) through (f) were carried out on the same OECT after each measurement underwent a cleaning step to remove residual D9‑THC, thereby confirming the device’s reproducibility. (a) presents the ID response curves for solutions containing 0.1 nM, 10 nM, 500 nM, and 5 mM D9‑THC in deionized water, with the corresponding real-time drain currents indicated at each concentration. (b) displays the ID responses upon adding 1 nM, 5 nM, 10 nM, 100 nM, 1 mM, and 5 mM D9‑THC solutions. (c) illustrates the ID responses when 0.1 nM, 0.5 nM, 1 nM, and 2.5 nM D9‑THC solutions are introduced. (d) shows the ID responses for 2.5 nM, 5 nM, 10 nM, and 50 nM D9‑THC solutions. (e) presents the ID responses for 50 nM, 100 nM, 500 nM, and 1 mM D9‑THC solutions. (f) depicts the ID responses when 1 mM, 2.5 mM, and 5 mM D9‑THC solutions are applied.

Figure 3. Normalized drain current responses of the OECT device upon addition of D9‑THC solutions at different concentrations. (a) Average normalized drain current response curves for two OECT devices measured using D9‑THC solutions at 0.1 nM, 10 nM, 500 nM, and 5 mM: the blue series corresponds to seven replicate measurements, while the black series reflects four replicates. Both curves exhibit similar slope characteristics: the blue curve has a slope of 0.0153 per decade in the 0.1–500 nM range and 0.1300 per decade in the 0.5–5 mM range; the black curve shows a slope of 0.0134 per decade from 0.1 to 500 nM and 0.1276 per decade from 0.5 to 5 mM. (b) Average normalized drain current response curve of the OECT device measured with D9‑THC solutions at various concentrations (1 nM, 5 nM, 10 nM, 100 nM, 1 mM, and 5 mM): both curves display comparable operating‑range features—slope of 0.02 per decade from 1 to 573 nM and 0.122 per decade from 573 nM to 5 mM; all data presented were obtained from four replicate measurements.

Figure 4. (a) ID response curves of aerosol‑jet‑printed OECTs to synthetic saliva buffer upon addition of the buffer. A pronounced peak in the gate current is also observed when the synthetic saliva buffer is added, owing to the high concentration of cations in the solution. (b) ID response curves of aerosol‑jet‑printed OECTs to D9‑THC at various concentrations diluted in synthetic saliva buffer. (c) Normalized ID values of the aerosol‑jet‑printed OECTs after adding D9‑THC at different concentrations. The operating range is divided into two regions: a slope of 0.003 per decade from 1 to 425 nM, and a slope of 0.02 per decade from 425 nM to 5 mM.

Table 1 Comparison of Detection Time, Required Analyte Volume, Limit of Detection, and Sensitivity for Different Types of D9-THC Electrochemical Sensors