Rapid In Situ Thermal Decontamination of Wearable Composite Textile Materials

Abstract

Pandemics stress supply lines and generate shortages of personal protective equipment (PPE), in part because most PPE is single-use and disposable, resulting in a need for constant replenishment to cope with high-volume usage. To better prepare for the next pandemic and to reduce waste associated with disposable PPE, we present a composite textile material capable of thermally decontaminating its surface via Joule heating. This material can achieve high surface temperatures (>100 °C) and inactivate viruses quickly (<5 s of heating), as evidenced experimentally with the surrogate virus HCoV-OC43 and in agreement with analytical modeling for both HCoV-OC43 and SARS-CoV-2. Furthermore, it does not require doffing because it remains relatively cool near the skin (<40 °C). The material can be easily integrated into clothing and provides a rapid, reusable, in situ decontamination method capable of reducing PPE waste and mitigating the risk of supply line disruptions in times of need.

Introduction

Most personal protective equipment (PPE) garments, including gloves and masks, are manufactured as disposable, single-use items. At the onset of a pandemic, the demand for PPE often increases significantly; consequently, the single-use, disposable nature of PPE results in a need for manufacturers to bolster production in order to replenish supplies and cope with this high-volume usage,^1–3 and also for organizations to establish rationing protocols for conservation of PPE to help combat supply chain disruptions.⁴ In cases of extreme demand, the consequences can be dire; for instance, due to shortages of PPE at the onset of the COVID-19 pandemic, many people—especially those working in medical professions—were forced to reuse disposable PPE and also to repurpose materials as makeshift PPE (e.g., homemade cloth masks), making them more susceptible to illness.^5–7 These shortages contributed to the death toll of COVID-19 by further stressing an already overburdened healthcare system and decreasing the availability and efficacy of treatment.⁸

Surges in demand have also caused an increase in the amount of disposable PPE waste,⁹ leading to environmental concerns.^10,11 Specifically, the World Health Organization (WHO) estimated that tens of millions of masks, gloves, and other PPE would be required on a monthly basis to effectively battle the COVID-19 pandemic worldwide.¹² This deployment of large quantities of PPE poses a threat to the environment due to the nonbiodegradable nature of many constituent plastics and the inclusion of biological contaminants after use.¹³ Proper disposal of PPE can mitigate environmental risk factors, such as local water contamination;¹⁴ however, accumulation of contaminated PPE in landfills remains undesirable and imposes health and safety risks on those in the sanitation industry.^15,16

Due to these environmental consequences and supply chain risks, we face an emerging need for reusable PPE. For example, isolation gowns—the second most used PPE after gloves—protect the body from surface contaminants and are typically discarded after use; however, reusable isolation gowns have been shown to provide comparable protection to single-use isolation gowns while being more cost-effective and environmentally friendly.^17–19 Similarly, reusable masks can slow PPE pollution and alleviate supply chain issues, with reusable alternatives typically made from cotton-based fabrics with varying levels of filtration efficiency.²⁰ Some fabrics have even been engineered to possess antiviral and antimicrobial properties.^21–23 However, all of these approaches either trap virions and other contaminants on the PPE material itself without inactivating them, allowing for an increased probability of fomite-mediated transmission of viruses, or require minutes to hours to achieve inactivation.

The effectiveness and lifetime of PPE can be extended through the use of adequate disinfection and decontamination methods. Many decontamination methods have been reported to safely decontaminate and allow reuse of PPE, including ultraviolet (UV) irradiation,²⁴ steam sterilization,^25,26 chemical disinfectants,^27,28 and dry heat decontamination.²⁹ However, among these methods, UV irradiation cannot effectively decontaminate within the crevices and folds of a material; the combination of high temperatures and moisture involved in steam sterilization can degrade delicate PPE; and chemical disinfectants can leave harmful residues. Alternatively, dry heat decontamination is accessible and scalable, and it reduces virus viability without compromising the integrity of PPE, but the process remains slow and the PPE is typically doffed and placed into external heating infrastructure (e.g., an oven).^23,30

Recent methods of dry heat decontamination have focused on integrating heat-generating elements into the PPE material itself. Nanomaterials, such as graphene, molybdenum disulfide (MoS₂), and titania (TiO₂), have been embedded into material layers to perform localized photothermal heating for the inactivation of viral and bacterial contaminants.^31–34 In another approach, a copper mesh was employed in a mask to perform Joule heating for the thermal inactivation of viruses.³⁵ Similarly, a layer of graphene was deposited onto a mask for use with a battery to provide portable Joule heating for the inactivation of viral contaminants.³⁶ Beyond the use of integrated heat-generating elements for decontamination,²³ heat generation in wearables in general presents a promising approach, especially as functional textiles have gained increasing interest.³⁷ Strategies for heated wearables have been developed using electrically conductive textiles; for example, conductive fabrics and threads can be transformed into sewable and weavable electric heaters.^38,39 Wearable heaters have also achieved localized heating through graphene-based electrothermal materials,⁴⁰ as well as thermofluidic actuation with phase-change fluids.⁴¹ While promising, most of these methods have been restricted to slow response times or low temperatures.

In this work, we present a composite textile material that achieves viral decontamination at short timescales (<5 s) by supplying quick bursts of thermal energy without burning the user, even when used in situ on the body (Figure 1). This composite textile material consists of four layers: (i) an electrically insulating polymer film upper layer; (ii) a conductive textile layer that promotes Joule heating; (iii) a heat sealable textile (HST) support layer that anchors the conductive textile; and (iv) a thermally insulating backing layer to shield the user from the applied pulses of heat, all of which are composed of base materials used in synthetic textiles capable of being recycled—such as polyester, nylon, and thermoplastic polyurethane (TPU).^42–44 We combined analytical modeling of the electrical characteristics of the composite textile material, the thermal response of the material during Joule heating, and the inactivation of viruses under the resulting time-varying temperature profile⁴⁵ generated at the surface of the material to optimize the composite material design. We ensured decontamination to the US FDA-specified 3-log, or 99.9%, reduction in virions^46,47 by targeting a conservative 4-log, or 99.99%, reduction. We modeled the inactivation performance of the material with the SARS-CoV-2 and HCoV-OC43 viruses and demonstrated viral inactivation experimentally with HCoV-OC43, which was used as a surrogate for SARS-CoV-2 due to the chemical similarity between the viruses and HCoV-OC43 being safer to handle. The composite material was then demonstrated as a wearable in the form of a glove, showcasing its reusability, safety, dexterity, fast response time for decontamination, and durability over hundreds of cycles. We show two untethered use cases for the glove based on (i) an onboard battery and controller enabling hundreds of cycles of decontamination before recharging and (ii) a tap-to-decontaminate wall-mounted panel with a simple user interface.

Figure 1.

Thermally decontaminating wearable composite textile material design, fabrication, and operation. (a) The composite textile material is composed of four layers: an electrically insulating dielectric nylon film to prevent electrical shorting or permeation by biological fluids, a conductive textile for Joule heating, a plain-woven nylon HST with a TPU coating on the upper side to serve as a substrate during fabrication, and a thermally insulating textile to shield the user from high temperatures. (b) False-colored cross-section of the composite textile material obtained with a scanning electron microscope, with colors corresponding to those shown in (a) (thermally insulating layer not shown). (c) Schematic illustrating the fabrication process used to construct the composite material. (d) The thermally decontaminating material integrated into a glove architecture is shown as an example of a wearable application in which viral contaminants can be inactivated rapidly without doffing the garment.

Results

Fabrication

The composite textile material is composed of four individual wearable layers, depicted in Figure 1a, that are permanently laminated together in a thermal bonding process to form a fully wearable material that thermally decontaminates its surface by using Joule heating. The inclusion of each layer is essential for a safe and rapid thermally decontaminating textile material, and the thermal bonding approach enables a facile and scalable fabrication process. A textile coated on one side with TPU acts as the substrate HST layer of the composite material, upon which the rest of the material is architected. An electrically conductive textile layer is interfaced with the TPU-coated side of the substrate textile for the generation of heat via Joule heating and retains the mechanical compliance of a wearable fabric. A nylon film is then applied on top as an impermeable dielectric layer to prevent electrical shorting and the penetration of biological fluids. The thermal bonding process generates a composite material while maintaining the flexibility of its textile-based layers (Figure 1b). The addition of a thermally insulating textile layer composed of a polyester–spandex blend inhibits heat transfer through the lower side of the composite material (toward the user). The fabrication of our composite textile material uses a thermal bonding approach similar to other works describing multilayered wearable materials and devices (Figure 1c).^{41,48–49,50} Specifically, we align the conductive textile on top of the TPU-coated side of a similarly sized substrate textile and tack them together in a heat press (145 °C and 275 kPa) such that the TPU weakly adheres to the conductive textile layer to hold it in place yet still allows for separation by peeling the two layers apart. A serpentine path is then cut into the conductive textile layer, and the excess is peeled away, leaving only the serpentine path bonded to the substrate textile. A dielectric nylon film is adhered atop these layers in a heat press (185 °C and 415 kPa) with the ends of the conductive path exposed, forming a permanent bond which remains intact during Joule heating of the device to peak decontamination temperatures of 100–110 °C during use. Lastly, the thermally insulating polyester–spandex blend fabric is adhered to the underside of the already bonded layers. Our heat-sealing fabrication approach highlights a facile and scalable fabrication method for the combination of multiple layers for use in wearable devices and garments (Figure 1d). Additional details on the fabrication process are provided in Materials and Methods.

Electrical Performance

The properties of the composite textile material—and, more specifically, the serpentine conductive textile—control the electrical resistance, which in turn dictates the thermal response. Serpentine paths are used for electrical transport because they provide near-uniform heat generation, especially over larger areas, compared with point contacts on large conductive sheets. The overall electrical resistance of the conductive serpentine path can be determined based on its sheet resistance, R_s. We measured the sheet resistance using the 4-point probe method (additional information about the sheet resistance is provided in the Supporting Information Section S1 and in Figure S1).⁵¹ The measured values were compared to the documented values provided by the supplier to ensure the integrity of the conductive textiles (Table S1 and Figure S2) and showed good agreement. Based on the sheet resistance, the overall electrical resistance of a serpentine path can be found by multiplying the path aspect ratio (i.e., the overall length of the conductive path divided by the path width) by the sheet resistance, R_s. This overall path length can be approximated by multiplying the total number of paths by the length of the straight segments of the path to realize the electrical resistance of the material.

1

where N is the number of straight paths in the serpentine design (Figure 2a), L is the overall material length, W is the overall material width, w_p is the path width, L_arc is the radius of the arc-shaped region formed by w_p, w_s is the path separation (held constant at 1 mm in this work based on the resolution of the fabrication process), C_B is a correction factor accounting for the change in electrical resistance due to the TPU penetrating into the conductive textile during the thermal lamination process, and C_T is a correction factor accounting for the increase in electrical resistance of the material due to temperature rise (additional details on these correction factors are provided in the Supporting Information Section S2 and Figure S3).

Figure 2.

Geometric parameters of the heater design dictate the electrical performance. (a) Image showing all dimensions where L is the total length of the patterned heater area, W is the total width, w_p is the width of the conductive pathway, w_s is the separation between conductive pathways, and L_arc is the radius of an arc-shaped region in the path. (b) Comparison of the analytical resistance approximation from eq 1 and a numerical approximation with experimentally measured results across increasing aspect ratios, AR. The aspect ratio was made greater by increasing L. Error bars signify the standard deviation of the electrical resistance measurements across two trials (n = 2). (c) Comparison of the electrical resistance of the conductive path between the analytically calculated and experimentally measured values as the path width, w_p, increases, where L = 50 mm and the number of paths, N = 4. The error bars signify the standard deviation in the electrical resistance measurements across three trials (n = 3). (d) Comparison of the electrical resistance of the conductive path between the analytically calculated and experimentally measured values at various path widths, w_p, as the path length, L, increases and N = 4. (e) Comparison of the electrical resistance of the conductive path between the analytically calculated and experimentally measured values at various path widths, w_p, as the number of paths, N, increases and L = 50 mm.

The analytical approximation in eq 1 neglects the contribution to the overall electrical resistance from the arc-shaped regions in the serpentine path because the effective arc length could not be easily modeled analytically due to the current exhibiting a “bunching” nature near the inner portions of these regions. To validate this simplified modeling approach, we compared the analytical predictions generated by eq 1 to experimental data and to a numerical simulation that accounts for the resistance contribution from the arc-shaped regions of the path (Figure 2b; details on the numerical simulation are provided in the Supporting Information Section S7). Varying the aspect ratio, AR (where AR = L/W), we found that the full numerical solution was able to predict the electrical resistance, in good agreement with the experiments across all AR. Meanwhile, the analytical model only aligns well with the full numerical solution and the experimental results for AR > 2 and is therefore only suitable for system design in this range. All of the material configurations designed with the analytical model in this work and showcased in later demonstrations have AR > 4.

To better quantify the accuracy of eq 1 across various material geometries, we fabricated and characterized material samples of various path widths while maintaining a constant number of paths, N = 4, and path length, L = 50 mm. The samples were all fabricated with the same process and show an inversely proportional relationship between the electrical resistance and the path width, with higher resistance values corresponding to smaller path widths (Figure 2c). Similarly, we show proportional relationships between the electrical resistance, path length (Figure 2d), and number of paths (Figure 2e). These results highlight the ability of eq 1 to accurately predict resistance values for design purposes.

Thermal Performance

We approximated the heat dissipation due to Joule heating as heat transfer in one dimension, normal to the surface of the material (Figure 3a). We estimated the Biot number, Bi, of the upper three layers—HST, conductive textile, and electrically insulating nylon—of the composite material and found it to be small (≪1), within the acceptable range for use of the lumped capacitance model (additional details on the Biot number and lumped capacitance calculation are provided in the Supporting Information Section S3). We therefore analyzed the system as a thermal circuit with thermal resistances corresponding to the thermally insulating textile layer, R_th-ins, and convection heat transfer to the surrounding air, R_th-air, with subscript “th” denoting a thermal resistance as opposed to an electrical resistance (Figure 3b). The dielectric nylon layer, conductive textile heat-generating layer, and supportive HST layer all act as thermal capacitors because their temperatures increase and decrease together in response to the power input from Joule heating, Q̇ (t). The model can be further simplified by assuming the lower face of the substrate textile to be adiabatic based on the condition R_th-ins ≫ R_th-air. We modeled the transient temperature response through an energy balance of the electrical power input, the heat loss due to convection, and the change in the internal energy of the material over time (full derivation is provided in the Supporting Information Section S3).

Figure 3.

Thermal response of the wearable composite textile material. (a) Side-view of the integrated layers on the skin as they are constructed in a wearable glove architecture. (b) Thermal circuit analogy for the material system as the thermally insulating layer and air act as thermal resistors, and the dielectric layer, conductive textile heater, and substrate HST act as thermal capacitors, accumulating heat over time. (c) IR imaging during the Joule heating process (t_heat = 5 s) at 0, 5, and 15 s. (d) Lumped capacitance thermal model, T_model, with power density Φ = 19.6 kW m^–2 compared to experimentally gathered temperature data using thermocouples, T_thermocouple, and IR imaging, T_IR.

The transient thermal response of the composite material depends on the applied voltage, V, path resistance, R, heat transfer coefficient (HTC) of air, h, area of the top surface of the material, A, mass of the material, m, specific heat of the material, c_p, and ambient temperature, T_air. We combine the voltage, V, resistance, R, and area, A, into a term called the power density, Φ = V²/(RA), which quantifies the input power per area of the material and can be used to realize a specific temperature profile given some heating duration, t_heat. We note that eq 2 assumes a constant HTC of air during heating and cooling despite greater natural convection occurring at higher temperatures due to the buoyant effects of warm air,⁵² and it assumes a constant electrical resistance that is independent of the temperature of the material and describes the convective cooling of the material from the peak temperature during Joule heating, T_max, at times t > t_heat as a piecewise function (further discussion of the codependence of the electrical resistance and temperature of the material is provided in the Supporting Information Section S6 and Figure S7).

2

During the Joule heating process, the surface temperature of the material, T_surface, was measured using thermocouples and infrared (IR) imaging (Figure 3c), where the thermocouples measured the temperature at specific points along the serpentine path (further modeling of the lateral heat transfer between regions of the path is discussed in the Supporting Information Section S7 and Figures S8 and S9), whereas the IR imaging captured the thermal performance across the entire surface of the material. We performed Joule heating of the material for 5 s, after which the electrical power was disconnected, and the sample remained exposed to air to cool via natural convection. This heating protocol, based on an input power density of Φ = 19.6 kW m^–2, is designed to achieve a 99.99% reduction in viable virus contaminants on the surface of the material, as described in detail in the following section. We then compared the experimentally measured temperatures—i.e., those obtained through thermocouples and IR imaging—to our transient thermal model from eq 2 and found good agreement between the modeled and measured temperature profiles and the maximum temperature achieved through Joule heating (Figure 3d).

Inactivation of Viruses

Prior work has shown that the temperature-dependent reaction kinetics of the inactivation of viruses can be modeled based on the rate law and the Arrhenius equation.^45,53,54 We combined our thermal model with this virus inactivation modeling framework to predict the change in viable virus titer as a function of the transient temperature response of our material, conservatively seeking a 99.99% (4-log) reduction to achieve, at minimum, a 99.9% (3-log) reduction in viable virus titer as recommended by the US FDA for adequate decontamination of PPE.^46,47 Because the temperature profile predicted by our model showed good agreement with the experimentally obtained temperature profiles with our assumption of a constant HTC of air of 50 W m^–2 K^–1 (Figure 4a), we used the temperature profile predicted by the model to calculate the rate of virus inactivation, k, over time using the Arrhenius equation to enable our predictive capability to extend to arbitrary profiles of temperature as a function of time. Based on this model, higher rates of inactivation occur at higher temperatures, as shown in Figure 4b (where A_f is the frequency factor of the inactivation reaction, E_a is the activation energy of the inactivation reaction, R* is the universal gas constant, and T is the temperature; further details on the calculation of virus inactivation and its corresponding temperature dependence are provided in Supporting Information Sections S3 and S4 and in Figures S4 and S5). We numerically integrated the rate of inactivation over time to account for the transient temperature profile and determined the overall level of virus inactivation for the three temperature profiles shown in Figure 4a. A reduction in viable virus titer of greater than 99.9% (3-log) was calculated for the temperature profiles obtained from both the experimentally recorded data and the transient thermal model (Figure 4c).

Figure 4.

Effect of the composite textile material’s surface temperature on the inactivation of viruses. (a) Predicted temperature profile from the thermal model, T_model, compared to temperature data observed experimentally along a straight portion of the serpentine path. T_thermocouple and T_IR are temperatures measured by thermocouples and IR imaging, respectively. (b) Rate of virus inactivation, k, as a function of the model temperature profile, where higher rates of inactivation are exhibited at higher temperatures. (c) Calculated reduction in infectious virus titer resulting from the transient thermal responses shown in (a) based on integration of the reaction rate over time determined from our temperature-dependent model of the reaction kinetics of virus inactivation. (d) Required power density input to achieve a 99.99% (4-log) reduction in viable virus titer as a function of the heating duration and energy density required for a duration of heating corresponding to a 99.99% reduction in viable virus titer, where shorter (yet more intense) pulses of applied heat require less overall energy to achieve decontamination. (e) Experimental procedure for the thermal inactivation of HCoV-OC43. (f) Experimentally determined level of viral inactivation due to Joule heating compared to the model prediction (i.e., conservatively targeting a 4-log reduction), reflecting more than a 3-log reduction in virus titer on material samples with and without the thermally insulating layer included. The control sample, referred to as “untreated virus,” shows a similar level of virus titer as the original virus stock when inoculated and subsequently recovered, whereas the samples (with and without thermal insulation) designed and fabricated through our modeling approach reflect the anticipated levels of viral inactivation on the surface of the material (>3-log reduction). Error bars represent the standard deviation [virus stock, n = 3; untreated virus, n = 5; heated, n = 3; heated (insulated), n = 4].

The temperature profile during the Joule heating phase depends on the power density input, Φ (full derivation is provided in the Supporting Information Section S5). Based on our experimentally validated model of the time-dependent temperature profile during Joule heating, we calculated the necessary power density, Φ, required for a given duration of Joule heating to achieve a conservative target value of 99.99% reduction in viable virus titer (Figure 4d) for both SARS-CoV-2 and HCoV-OC43 and fabricated the requisite material; conversely, if the geometry of a material sample has already been established, any power density, Φ, could be applied over any duration of heating, t_heat, from which the level of viral inactivation could then be predicted. HCoV-OC43 was used as a surrogate virus for SARS-CoV-2 for experimental testing of the level of virus inactivation due to the similarities in their reaction kinetics⁴⁵ (additional details are discussed in the Supporting Information, Figure S5). From eq 2, the temperature at a given time during the heating phase scales proportionally with the power density; thus, greater reductions in virus titer are associated with higher power densities for a constant duration of heating. Additionally, as the duration of heating decreases, the required power density to achieve a 4-log reduction in the virus titer also increases. Despite the need for larger power densities at shorter heating timescales, the thermal inactivation process proves to be more energy efficient at these shorter timescales, as shown in Figure 4d (additional details are discussed in the Supporting Information, Figure S6).

To verify rapid and complete inactivation of viruses, we inoculated droplets of Dulbecco’s modified Eagle medium (DMEM) containing the HCoV-OC43 virus onto the surface of the composite material (Figure 4e) (additional details are discussed in the Supporting Information Section S8 and Figure S11). We performed Joule heating on the composite material and measured the viable virus titer on the heated samples—both with and without the thermally insulating layer of the material—and observed at least a 3-log reduction in virus titer in both heated cases. For our control samples, we recovered the virus after inoculation for the same duration of time but without heating the material. We show our results in Figure 4f, where the composite material exhibits greater than a 3-log reduction in virus titer to a value of 1.01 × 10⁵ TCID₅₀ mL^–1 (n = 3) compared to the initial virus titer of 1.35 × 10⁸ TCID₅₀ mL^–1 (n = 3) and the control sample titer of 1.21 × 10⁸ TCID₅₀ mL^–1 (n = 5). An even greater reduction is observed with the thermally insulating layer present, to a virus titer of 4.91 × 10⁴ TCID₅₀ mL^–1 (n = 4). The model predicted a 4-log reduction; although the experimental results did not achieve the anticipated 4-log reduction for which the samples were designed, this conservative overdesign still results in at least a 3-log reduction in virus titer, in alignment with the US FDA’s recommendation. These results demonstrate the capability of our material to thermally decontaminate the material surface in wearable textile architectures. Based on the performance with the surrogate virus HCoV-OC43, this material could reasonably be assumed to effectively decontaminate SARS-CoV-2 and other viruses with similar thermochemical properties.

Practical Use and Durability

We characterized the safety of this composite material for wearables by measuring the temperature of the material at the skin of a user during operation (Figure 5a). We show that the temperature profile, T_surface, on the surface of the composite material achieves the high temperatures necessary for adequate decontamination at short timescales, while the temperature inside of the glove, T_skin, increases to 35.9 °C but does not reach or exceed temperatures that would cause discomfort (T_pain ≈ 45 °C).⁵⁵ Furthermore, we examined the thermal safety of the composite material by interfacing the surface of the material with the skin on a user’s forearm after 1, 5, and 10 s following the completion of Joule heating for the thermal decontamination process (Figure 5b). T_surface increases greatly during the Joule heating process, while T_skin remains at its initial temperature because there is no contact with the material. When physical contact is made between the material surface and the skin, T_surface rapidly decreases to an equilibrium state with T_skin; however, the temperature of T_skin does not change at the same rate as T_surface due to the difference in thermal properties of the two objects, namely, that the arm of the user has a greater thermal mass than the material. These results confirm that the surface of the material is safe for a user to touch after the thermal decontamination process.

Figure 5.

Thermal and mechanical durability and sustainability of the composite textile material integrated in a glove. (a) Comparison of the temperature of the surface of the composite textile material, T_surface, to the temperature of the skin, T_skin, inside of the glove. (b) Temperature profiles, T_surface and T_arm, when the surface of the composite textile material physically contacts the arm of a user at a given duration after the completion of Joule heating. The horizontal red line denotes the temperature threshold at which humans begin to experience pain, T_pain = 47 °C. (c) Temperature profile over 900 operating cycles of 3 min each with t_heat = 5 s. (d) Maximum temperature achieved per cycle and the corresponding range in reduction of viable virus titer over 1000 cycles of use. (e) Change in electrical resistance of the composite textile material under mechanical deformation by bending the material over 5000 cycles. (f) Mass of waste produced through disposal of single-use nitrile gloves and the thermally decontaminating glove, showing the reusability of the thermally decontaminating glove results in 2 orders of magnitude less waste over its lifetime. The mass difference is visually depicted by comparing a photograph of 900 single-use disposable nitrile gloves to the thermally decontaminating glove, which can be used more than 900 times.

In seeking durability of this composite material, we characterized the thermal performance through 900 cycles of Joule heating, where each cycle lasted 180 s—i.e., 5 s of Joule heating followed by 175 s of convective cooling to ambient conditions (Figure 5c). We also characterized the bond strength of the lamination between the dielectric nylon film and the rest of the composite material as a function of the number of Joule heating cycles and found little change—i.e., a standard deviation of 4.59% across tests at different numbers of thermal cycles—in the mechanical performance of the heat-sealed regions (additional details are discussed in the Supporting Information Section S10 and Figure S14). The temperature profiles in the first five and last five cycles exhibit similar behavior and achieve the necessary temperatures for adequate decontamination during a 5 s heating period. Because the thermal and mechanical properties of the composite material used in this work remain nearly constant, the shape of the temperature profile per cycle remains constant as well, allowing us to predict the level of viral inactivation based on the maximum temperature achieved per cycle (Figure 5d). The material consistently achieves between a 3-log and 4-log reduction in viable virus titer over many cycles; however, the performance degraded after 940 cycles due to the material deforming and folding onto itself, resulting in areas of heat concentration that caused failure.

We further characterized the mechanical robustness of the material by imposing a bending motion onto it, similar to that of finger bending. We measured the electrical resistance of the conductive serpentine path over 5000 cycles of bending (Figure 5e) because an increase in the resistance would inhibit the thermal performance of the material under a constant-voltage input. The material exhibited a 3% increase in its electrical resistance, indicating a negligible change in its electrical and thermal performance of the material.

To realize a path toward creating reusable and sustainable PPE in an effort to mitigate the risks imposed upon sanitation workers and reduce the land area covered by PPE waste,^15,16 we compared the mass of waste produced by a wearable glove incorporating our material to that of nitrile gloves over intervals of 900 uses (Figure 5f) because our material approaches sub-3-log reductions in virus titer based on our results after 900 uses. We show that the mass of waste produced by nitrile gloves is two orders of magnitude greater than that produced by gloves incorporating our composite material, and we show a break-even point in cost effectiveness at 457 uses of our composite material compared to nitrile gloves (additional details are discussed in the Supporting Information Section S11 and Figure S16).

We also investigated the practicality of our material as a wearable. In addition to characterizing the robustness and durability, we show through multiple hand positions—including opening and closing the hand and forming and rotating a fist—that our material can undergo natural hand movements and remain operable while also allowing unimpeded movement of the wearer (Figure 6a).

Figure 6.

Practicality and usability of the material in wearable applications. (a) Natural hand movements demonstrate the wearability and mobility of the composite textile material in a glove architecture and its negligible effect on thermal performance (the color bar corresponds to a temperature range of 23–110 °C). (b) Portable battery-powered system for supplying Joule heating for thermal decontamination in an untethered on-the-go scenario. After the user contacts a contaminated surface, they begin the decontamination process by pressing a button located on the wrist-mounted onboard controller (color bars correspond to a temperature range of 23–110 °C). (c) Wall-mounted panel system used to supply Joule heating to achieve decontamination. The user approaches the panel and initiates electrical contact between the panel and the glove via electrical contact pads that are magnetized for ease of alignment. The wall-mounted panel system measures the electrical resistance of the composite textile, calculates the time required to heat the material for adequate decontamination, and then supplies the requisite electrical power to the material. The user breaks electrical contact with the panel upon completion of the decontamination process (indicated by LED lights) and proceeds; this system eliminates the need for a battery and controller on the wearable, allowing it to be lightweight and fully compliant (color bars correspond to a temperature range of 27–112 °C).

Untethered Demonstrations

Recognizing the promise of our approach for wearable applications, we show the use of our material in an on-the-go decontamination scenario (Supporting Information, Movie 1). Other methods of decontamination require the user to doff their PPE and oftentimes pause what they are doing for several minutes or hours. Our material can decontaminate itself quickly without substantially interrupting ongoing tasks. This application makes use of the composite material as a wearable by incorporating it onto a glove; a custom circuit board allows the user to press a button to control the dissipation of electrical power from a rechargeable battery used to power the circuit controller and supply the electrical power for Joule heating (Figure 6b). The user initiates contact with a commonly contaminated surface (Supporting Information Movie 1) and then presses a button on the custom circuit board to perform the thermal decontamination process through Joule heating of the composite material (additional details are discussed in the Supporting Information Section S9 and Figure S12).

Alternatively, we can eliminate the control circuitry and power supply from the user altogether by implementing a tap-to-decontaminate wall-mounted panel to control the application of electrical power. The dissipated electrical power is modulated by calculating the time necessary for decontamination, t_calc, through measuring the electrical resistance of the composite glove material because the resistance may increase due to material imperfections in fabrication and potential degradation during use (additional details are discussed in the Supporting Information Section S9 and Figure S13).

A green LED indicates that the system is actively waiting to detect the presence of electrical contact, which is made when the user taps the magnetic electrical leads of the glove to the wall-mounted panel (Figure 6c). Once electrical contact is made, circuitry mounted behind the panel then measures the electrical resistance of the composite textile material to account for potential variability in resistance across the individual material samples due to fabrication tolerances or routine wear and tear. The circuit then calculates t_calc (i.e., the time necessary for a 4-log reduction in viable virus titer based on our results) and supplies power to perform Joule heating for some time, t_calc, indicated with a red LED. Upon completion of the decontamination process, a yellow LED begins to blink, indicating that the user may remove their hand, and the process is able to restart (Supporting Information Movie 2). We expanded upon these practical uses of our composite textile-based material by demonstrating its ability to be washed with minimal changes (<10% change in electrical resistance) to its electrical properties (additional details are discussed in the Supporting Information Section S10 and Figure S15).

Discussion

We fabricated a wearable composite textile material with integrated thermal decontamination capabilities for the inactivation of viral contaminants at fast timescales (<5 s). This process is governed by the interplay between electrical, thermal, and biochemical subsystems; specifically, the geometry and material properties of the serpentine conductive textile path dictate the electrical resistance, which, in tandem with the power density, can specify a constant voltage input used for Joule heating to produce a temporally varying temperature profile, and the resulting temperature profile subsequently determines the level of viral inactivation. Through detailed analytical modeling, we have generated a design process in which a specific level of viral inactivation can be selected, a requisite temperature profile can be determined, and the necessary geometrical parameters for the material can then be calculated based on this temperature profile, allowing high customizability of the composite textile material.

The customizability of the composite textile material allows for even higher temperatures at shorter decontamination times, making it more energy efficient and therefore longer-lasting when untethered; however, we are limited by the thermal degradation that would occur at higher temperatures due to the material properties of the individual layers and also by the thermal resistance of the thermally insulating layer, where maintaining a cool skin temperature is a constraint. Our approach also assumes a constant convective HTC of air and does not account for greater natural convection at higher temperatures due to the buoyant effects of warm air.⁵² We determined the specific heat of the material system used in the lumped capacitance thermal model through fitting the model to experimental data rather than conducting a direct experimental measurement of the specific heat, and we assumed one-dimensional heat transfer normal to the surface of the composite textile material; this assumption does not account for lateral heat transfer in the areas separating the conductive paths, although more detailed modeling of heat spreading within these areas reveals that decontamination still occurs (additional explanation is provided in the Supporting Information Section S7 and Figures S9 and S10). We used this more detailed numerical thermal model to characterize the extent of viral inactivation in the gap areas between the serpentine path as a function of the heat spreading by varying the path width, w_p, and the width of the gaps, w_s. Under a constant Φ, and as w_p decreases, the maximum temperature achieved in the gap areas decreases, revealing a critical path width, w_pc = 2.86 mm, that allows for at least a 3-log reduction in virus titer in the areas between the paths (additional details are discussed in the Supporting Information, Section S7 and Figure S10). We also recognize that our material only covers the fingers and not the palm of the hand in our demonstrations. We made this selection because the fingers are more likely to contact various surfaces than the palm, but we note that our design approach also allows customizability in terms of the area covered by the thermally decontaminating material, such as covering the palm of the hand or other parts of the body.

Viruses can be transmitted from host to host through multiple pathways, including contact transmission (via direct exposure to an infected person), airborne transmission (via suspended aerosols that contain the virus), and fomite-mediated transmission (via contaminated surfaces like gloves).^56–58 Healthcare and other essential workers are at an inherently higher risk of exposure to these different modes of transmission due to their interaction with large groups of people daily.⁵⁹ Prior work has shown that the SARS-CoV-2 virus (like other coronaviruses) can remain viable on the surface of PPE materials for periods of days at room temperature, especially when inoculated with highly concentrated virus solutions,^56,60,61 where its long persistence time could potentially increase the probability of fomite-mediated transmission and allow it to spread through human contact with contaminated surfaces. Previous work has investigated ways to shorten this persistence time,⁶² and we expanded on that work by demonstrating the practical use of this rapidly thermally decontaminating material as a wearable in the form factor of a glove in which the geometry of each composite material sample shown here has been tailored to achieve the necessary power density for a 4-log reduction in viable virus titer with 5 s of heating.

Future directions include (i) the implementation of this material in other practical, wearable applications, such as reusable PPE garments like gowns and masks, that are also made from textile materials for ease of integration;^6,17,19 (ii) the implementation of a thinner, but still thermally resistive, insulating layer, which could prove useful in improving the dexterity of the material when adhered to other base textiles (e.g., gloves, gowns, masks); (iii) covering larger areas (e.g., the entire hand) without significantly increasing the necessary electrical power input, which we expect will also increase the usability of this material in more areas on the body aside from the hands; and (iv) the extension and experimental validation of this work with other virus families as well as bacteria, given their reaction kinetics can be characterized via the Arrhenius equation.

Conclusions

In summary, we demonstrated a composite textile material with integrated thermal decontamination capabilities for the rapid inactivation of viruses in situ on the body. This material shows promise for waste reduction and protection from supply chain disruptions because it is reusable over many cycles, with minimal degradation in performance observed through 900 cycles of use. By conservatively designing the material to achieve a 4-log (99.99%) reduction in virus titer, we were able to experimentally demonstrate a reduction in virus titer between 3-log and 4-log, in accordance with US FDA guidelines. We demonstrated the ability of the material to thermally decontaminate within a wearable architecture, both (i) with a battery supplying power so the material can thermally decontaminate while on-the-go, without interrupting tasks, and (ii) with a “tap-and-go” wall-mounted power station to perform Joule heating for a specified duration based on the measured electrical resistance, without doffing the material. This highly customizable composite textile material can serve as a pathway toward reusable thermally decontaminating materials for PPE to better combat shortages due to supply chain issues during the next pandemic and to reduce the environmental impact of single-use disposable PPE.

Materials and Methods

Materials and Fabrication

The electrically insulating layer is a nylon film (Stretchlon 800, Fibre Glast). The conductive textile is a plain-woven Ni/Cu metal-plated polyester fabric (Nora LX PW, V Tech Textiles). The substrate textile is a 400-denier plain-woven heat-sealable packcloth (FHSP-BLACK, Seattle Fabrics). The thermally insulating layer is an 88% polyester, 12% spandex textile (B0777LDYZK, Aegend). The wires used in the demonstrations are insulated 22 AWG tinned copper (extension cable LED welding wire, AOTOINK). Electrical connection was made between the ends of the serpentine paths and the wires with tin-plated brass quick disconnect female single crimps (69525K11, McMaster-Carr). The thermally bonded layers—the electrically insulating layer, conductive textile, and substrate HST—were adhered to the thermally insulating layer with a Peel-n-Stick Fabric Fuse (B001A36I2A, iCraft).

Resistance Measurements

The 4-point probe method was used to measure the sheet resistance, R_s, of the conductive textile. Four jumper wires were spaced 1 cm apart in the middle of a conductive textile sheet with dimensions of 1.5 m × 1.5 m. The outer two wires supplied a constant current of 1 A with a Riden RD6018 DC power supply, and the inner two wires measured the voltage drop with a Fluke 177 True-RMS Digital Multimeter. The sheet resistance, R_s, was then calculated using Equation S8 (full derivation is provided in the Supporting Information Section S1). A Fluke 177 True-RMS Digital Multimeter was used to measure the electrical resistance of the conductive serpentine paths. A NI-6002 DAQ measured the voltage drop across the composite textile material with 1 V being applied during the bending cycle experiment, and MATLAB was used to perform voltage division and calculate the resistance of the composite textile material over each cycle.

Temperature Measurements

Temperature measurements at points along the serpentine path were obtained using 0.002-in diameter T-type thermocouples from Omega. The thermocouples were secured to the composite textile material with 0.05 mm thick high-temperature polyimide tape. Temperature data were processed using MATLAB with a NI-9211 DAQ. Thermal imaging of the composite textile material was conducted with a FLIR One Pro-iOS camera.

Virus Inactivation

Virus infection was performed on the HEK293T cell culture, which was maintained routinely at 37 °C with 5.0% CO₂ in DMEM (Gibco) supplemented with 10% fetal bovine serum (Sigma-Aldrich). The HCoV-OC43 samples were amplified by infection in HEK293T cells. Cell supernatants containing HCoV-OC43 were harvested and filtered through 0.45 μm filters (VWR). All virus samples were stored at −80 °C for long-term usage. The infectious titers of viral stocks or heat-decontaminated viral samples were determined by the 50% tissue culture infectious dose (TCID₅₀) assay. All experiments involving HCoV-OC43 viruses were conducted in a biosafety cabinet in a biosafety level 2 (BSL-2) laboratory, following all approved notification-of-use and safety protocols. To quantitate the virus inactivation efficiency, samples of the composite textile material were first inoculated with 4 μL virus droplets. Thereafter, the heat-treated samples, together with untreated samples and the original viral stock aliquots, were subjected to TCID₅₀ assays in 96-well plates to determine the titers of the infectious virus. The virus titers of individual samples were expressed as log₁₀ TCID₅₀ per milliliter of media. At least three replicates were performed under each experimental condition.

Circuit Electronics

The on-the-go application was controlled with an Arduino Nano Every and powered by a 3 S type, 11.1 V, 25 C, and 2200 mAh lithium polymer (LiPo) battery. The tap-to-decontaminate wall-mounted panel application was controlled by MATLAB using an NI-6002 DAQ and a Riden RD6018 DC power supply.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.3c09063.

• Characterization of the conductive textile; calculation of the electrical resistance and relevant parameters; derivation of the transient analytical thermal model; relation of the power density, Φ, and virus inactivation pertaining to the material design; derivation of the transient numerical electrothermal model; analytical thermal modeling accounting for the thermal mass of the virus-laden droplet; design of the circuits for the wearable demonstrations; and T-peel tests and washability of the material (PDF)

• On-the-go demonstration of the composite textile material being used (MP4)

• Wall-mounted panel used for Joule heating the material in a tap-to-decontaminate demonstration (MP4)

Author Contributions

M.D.B. and D.J.P. conceived the research. M.D.B., K.Y., A.R., Y.J.T., and D.J.P. designed the experimental setups and collected data. M.D.B., K.Y., T.F.Y., and D.J.P. analyzed the data. Z.L. conducted SEM imaging to visualize the cross-section of the composite textile material. M.D.B., T.F.Y., and D.J.P. contributed to writing the original draft. All authors contributed to writing and reviewing the final draft and have given their consent to the final version of the manuscript. Y.J.T. and D.J.P. supervised the research.

The authors declare no competing financial interest.