How can chip technology realize electronic sensing of viruses?

Regrettably, COVID-19 taught us how significant the portable virus sensing technology is. Before it, the cost issues and different regulations in various countries have hindered the widespread use of the Point-of-Care Testing (POCT). Suppose we can electronically control it using Si nanotechnology, named e-POCT. Following the concept of IoT sensors, we can expect that a plenty of deployed e- POCTs may retrieve digital data about where, when, how much, and which virus is being spread. The collected digital data (to be BigData) can be analyzed by AI machines to optimize the global distribution of medical resources. The core technology for e-POCT must be an electronic biosensing. It is composed of a controlling system to selectively collect biomolecules (specific to targeted virus) in specimen solutions, oxide films (or membranes, polymers) to adhere specific biomolecules, and a sensing system to convert the adhered biomolecules to electrical signals. The efficiencies and reliabilities of the components are critical to the performance of the virus sensing. The performance can be characterized by the Limit-of-Detection (LoD) and detection time. We propose a concept of electronic sensing systems to substantially improve both LoD and detection time.

Introduction

A typical size of viruses is 80 - 100 nm in diameters, which is much larger than advanced technology nodes (far beyond 10 nm) using nanowires (NWs) or nanosheets. In Figure 1, viruses are dressed in glycoproteins (named, spike proteins or spikes) with the size being less than 10 nm. Each virus dissolved by detergent releases 500-1,000 spikes in the average. In the specimen solution where viruses are dissolved, the floating spikes specific to the virus have the targeted charge by tuning PH from the isoelectric point. If we can collect and immobilize a sufficient number of spikes near NWs, then the transconductance (G) of NW-FETs may be affected. We can thus define the sensitivity as |ΔG/G| if we can exclude contributions of non-targeted biomolecules on transconductance. This reminds us of the possibility of silicon-based electronic virus sensors. It can be a key component of electronic Point-of-Care Testing (e-PoCT), which is cost-effective and feasible to mass-production. If we deploy it worldwide like IoT sensors, then it may co-work with AI/BigData and thus help not only disease control systems but also the optimization of medical resources distribution, as illustrated in Figure 2. The virus density in a specimen solution is limited during incubation period, although the virus detection during incubation period is critically important for the aims --- the disease control and the optimization of medical resources distribution. Longer detection time is inconvenient. A best scenario is that it is easily usable as anywhere as possible, and able to detect spikes specific to targeted virus even in incubation period within a predetermined detection time, and then automatically forward digital data of viruses to AI machines and return diagnostic result to necessary parties at once. From such a viewpoint, there are two factors to judge the function of electronic virus sensors. One is the Limit-of-Detection (LoD) [1]. It is the lowest density of specific biomolecules (i.e., spikes), at which density the sensors can sense. The other is detection time, which should be less than one minute or less as possible. In Figure 3, we gathered published data of detection times and LoDs with various structures of sensing area. Our aim is to reach the left-bottom corner by using the silicon nanotechnology. The spike sensing may have the advantage by 2-3 digits than virus sensing.