Realization of 3D epoxy resin/Ti₃C₂T_x MXene aerogel composites for low-voltage electrothermal heating

Ti₃AlC₂ powders (purchased from Laizhou Kai Kai Ceramic Materials Co., Ltd), lithium fluoride (LiF, purchased from Alfa Aesar), hydrochloric acid (HCl, purchased from Sigma-Alrdrich), epoxy resin (Araldite, LY5052) and the hardener (Aradur, HY5052 purchased from Huntsman) were used.

Ti₃C₂T_x MXenes were prepared by in-situ HF etching of Ti₃AlC₂ powders and the experimental details can be found in our previous report [27]. Briefly, 3 M LiF were dissolved in 9 M HCl in a high-density polyethylene container at room temperature. Two grams of Ti₃AlC₂ powders were slowly added into the etching solution under vigorous stirring. The reaction was kept at 45 °C for 24 h to etch the Ti₃AlC₂. The etched MXenes were first washed with deionised water using centrifuge (at 10000 rpm for 5 min per cycle) for multiple cycles to remove the excess of acid. In between centrifuge cycles, vigorous shaking by hand was applied to delaminate the etched MXenes. The delaminated MXenes were collected by collecting the supernatants from multiple centrifuge cycles (at 3500 rpm for 5 min per cycle). The delaminated MXene suspension was concentrated via centrifuge (at 10000 for 1 h) to obtain a stock suspension which could later be used to prepare MXene suspensions for freeze-casting.

The MXene solution prepared above (120 mg cm⁻³) was poured into a square PTFE mould (with dimension of 2 cm × 2 cm × 2 cm) and frozen by unidirectional freeze-casting over a copper substrate. Freeze-casting was conducted from 20 °C to −100 °C at a cooling rate of 10 °C min⁻¹ and the solid structure then subsequently freeze-dried to obtain a Ti₃C₂T_x aerogel. To prepare the composite, hardener was added to epoxy resin (38 wt% with respect to the resin) and mixed by high shear mixing for 5 min. The mixture thereafter was kept in a vacuum oven for 10 min to remove any air bubbles. The Ti₃C₂T_x aerogel was immersed into the epoxy which was degassed and infiltrated by vacuum-assisted infiltration for 1 h (figure 1(a)). After an initial 24 h curing step at room temperature, the samples were then post-cured at 100 °C for 4 h in a conventional oven. The cured sample was polished to remove the excess epoxy resin that was not infiltrated into the aerogel to obtain the final epoxy resin/Ti₃C₂T_x MXene Aerogel composite. The mass loading of Ti₃C₂T_x in the epoxy resin/Ti₃C₂T_x MXene Aerogel composite was calculated by dividing the mass of the initial Ti₃C₂T_x aerogel by the mass of the final epoxy resin/Ti₃C₂T_x MXene Aerogel composite after polishing. The final epoxy resin/Ti₃C₂T_x MXene Aerogel composite was found to have 10 wt% loading of Ti₃C₂T_x . For comparison, Ti₃C₂T_x /epoxy composite with 10 wt% loading of Ti₃C₂T_x was prepared by dispersing delaminated Ti₃C₂T_x flakes in epoxy resin using a shear mixing method followed by the same degassing and curing process.

Zoom In Zoom Out Reset image size

The surface morphological images of the as-prepared samples were acquired by scanning electron microscope (Zeiss Ultra-55, Germany). X-ray micro computed tomography (µCT) imaging was performed using a Zeiss Versa 520, Oberkochen, Germany using a tube voltage of 60 kV and 5 W power in phase contrast mode. 3001 projections were taken at an exposure time of 12 s per projection. Source to sample and sample to detector distances were 26.0 and 43.5 mm respectively. A 4× magnification was used and voxel size was 1.264 µm. Data was reconstructed using x-ray microscopy (XRM) scout-and-scan control system (Zeiss, Oberkochen, Germany) and visualised using Avizo (version 2019.3; Thermo Fisher Scientific, Waltham, MA, US).

Powder x-ray diffraction (XRD) was undertaken using a Proto AXRD θ–2θ diffractometer (284 mm diameter circle) with a sample spinner and Dectris Mythen 1K (5.01° active length) 1D-detector in Bragg–Brentano geometry employing a Copper Line Focus x-ray tube with Ni K_β absorber (0.02 mm; K_β = 1.392250 Å) K_α radiation (K_{α 1} = 1.540598 Å, K_{α 2} = 1.544426 Å, K_α ratio 0.5, K_{α av} = 1.541874 Å) at 600 W (30 kV 20 mA). X-ray photoelectron spectra (XPS) measurements were performed by a PHI Quantera SXM/AES 650 Auger Electron Spectrometer (ULVAC-PHI, Inc.) equipped with a hemispherical electron analyzer and a scanning monochromated Al Kα (hυ = 1486.6 eV) X-ray source. To measure the impedance, the electrical contact on the samples was made with silver paint to reduce the contact resistance. All impedances were tested using a PSM 1735 Frequency Response Analyzer from Newtons4th Ltd connected with Impedance Analysis Interface with two terminals at the range of frequencies from 1 to 10⁶ Hz. The specific conductivities (σ) of the materials were calculated from the measured impedances using the following formula:

$$\begin{equation}\sigma \left( \omega \right) = \,\left| {{Y^*}\left( \omega \right)} \right|\frac{t}{A} = \,\frac{1}{{{Z^*}}} \times \frac{t}{A}\end{equation} \tag{ 1 }$$

where ${Y^*}\left( \omega \right)$ is the complex admittance, ${Z^*}$ is the complex impedance, t and A are the thickness and cross-sectional area of the sample, respectively.

The Joule heating properties of all of the samples was conducted by applying the different voltages across the devices and measuring the current-induced temperature response. Samples were inserted with a custom-made clip and tightened enough to ensure a reliable and uniform electrical contact area. The electrical current and power applied to samples from two ends were controlled and monitored by the DC power supply. The maximum applied voltage was restricted up to 20 V and the maximum delivered current was restricted within 10 A for safety purposes. The surface temperature of each segment of the samples at steady-state condition was measured using an IR thermal camera with a recording function.

The surface morphologies of Ti₃C₂T_x and its epoxy composite aerogels are shown in figures 1(b) and (c). An anisotropic porous nature of the Ti₃C₂T_x aerogel with interconnected MXene flakes is evidenced from figure 1(b). During the freeze-casting process, MXene flakes are pushed to the entrapped regions between the anisotropically grown ice crystals. As a result, highly ordered, layered assemblies of 3D porous MXene aerogel are formed having uniform pores with an average size of around 45 µm. Such a microstructure, where each flake can serve as an nanoheater [28], may facilitate better electrical and thermal transportation during the Joule heating process compared to their randomly oriented counterparts [29]. A jagged crack pattern and the rough surface of the epoxy/aerogel composite can be seen in figure 1(c), confirming the effective infiltration of epoxy into the MXene aerogel.

The µCT image of epoxy resin/Ti₃C₂T_x MXene aerogel composite is shown in figure 2. The virtual cross-sectional image (left) shows a homogenous MXene sheet domain structure across the scanned area. The 3D microstructure within a domain is shown for a region of interest on the right hand side of figure 2. A 3D lamellae structure of MXene is confirmed, which serves as a scaffold for the epoxy resin/Ti₃C₂T_x MXene aerogel composite. Within the µCT scanned volume, no air filled pores were visible which confirms the excellent infiltration of epoxy throughout the aerogel structure.

Zoom In Zoom Out Reset image size

To validate the successful synthesis of Ti₃C₂T_x , XRD of all the samples are shown in figure 3(a). The (002) peak of Ti₃C₂T_x is found to have shifted towards a smaller angle around 7° and broadened compared to its MAX phase counterpart (∼10°), which indicates a successful extraction of Al-atoms from Ti₃AlC₂. Moreover, the characteristic peaks between 33° and 43° of Ti₃AlC₂ have vanished for both of the Ti₃C₂T_x samples. These facts show that Ti₃C₂T_x was successfully synthesised by the in-situ etching process. It should be noted that the XRD spectra for delaminated Ti₃C₂T_x and as-prepared Ti₃C₂T_x aerogel are similar, indicating the excellent stability of Ti₃C₂T_x flakes even after freeze-casting.

Zoom In Zoom Out Reset image size

Increasing the resistive features of Ti₃C₂T_x by incorporating epoxy is evidenced in figure 3(b). The room temperature electrical conductivity for Ti₃C₂T_x aerogel/epoxy is found to be 2.1 S cm⁻¹ at 1 Hz, which is lower than the bare Ti₃C₂T_x aerogel (3.1 S cm⁻¹) and much higher than the epoxy resin (∼10⁻¹¹ S cm⁻¹). The relative reduction in electrical conductivity in the composite aerogel is due to the epoxy resin incorporated into the aerogel thereby separating the flakes slightly. It is noteworthy that response of both the Ti₃C₂T_x aerogel and epoxy resin/Ti₃C₂T_x MXene aerogel composite are quite independent of the applied frequency and hence the resistive component dominates in this case. The impedance of the comparison sample where Ti₃C₂T_x flakes were directly mixed into epoxy is also shown (figure 3(b)). This sample was highly resistive [28], showing the importance of the percolated, connected nature of the aerogel for imparting good electrical conductivity.

The electrical conductivity of the Ti₃C₂T_x aerogel was almost completely independent of temperature, whereas a drastic drop in conductivity occurred for the epoxy resin/Ti₃C₂T_x MXene aerogel composite (figure 3(c)). Note that the measurement of electrical conductivity of the Ti₃C₂T_x aerogel was restricted to 50 °C since MXenes are very sensitive to temperature under ambient conditions due to the attached functional groups. In contrast to the Ti₃C₂T_x aerogel, the electrical conductivity of epoxy resin/Ti₃C₂T_x MXene aerogel is measured at a relatively high temperature to ensure the stability and integrity of epoxy in the Ti₃C₂T_x aerogel.

The excellent Joule heating capability of the composite was validated by IR image inspection at different applied voltages (figures 4(a)–(c)). The steady-state temperature of epoxy resin/Ti₃C₂T_x aerogel composite was found to increase from 43 °C to 127 °C as the applied voltage was increased from 1 to 2 V. At 3 V applied voltage with 7.8 A current, the steady-state temperature of the composite was raised to 166 °C. The thermal IR images of the samples at 1.25, 1.75 and 3 V are shown in figure S1 (see supplementary materials (available online at stacks.iop.org/tdm/8/025022/mmedia)). The photographic images of bare MXene aerogel and epoxy resin/Ti₃C₂T_x aerogel composite before and after Joule heating are shown in figure S2 (see supplementary materials). The obtained result is impressive compared to the electrothermal materials reported in the literature (table 1). Our intention in table 1 is to show the importance of infiltrating the polymer into the 3D interconnected skeleton over the composite film such that best performance from the composite can be obtained. Essentially, 3D structures are well known to offer excellent electrical and thermal conducting pathways [26]. The steady-state temperature of Ti₃C₂T_x aerogel/epoxy is higher than the bare Ti₃C₂T_x aerogel at the same input voltage which can be visualised from figure 4. For instance, at the same input voltage of 2 V, the Ti₃C₂T_x aerogel surface can only heat up to 48.3 °C with 6.7 A current whereas the epoxy resin/Ti₃C₂T_x aerogel composite can provide a much higher steady-state temperature of 123 °C at 5.1 A current. Thermal IR images of the Ti₃C₂T_x aerogel at different voltages are shown in figures 4(d)–(f). The Ti₃C₂T_x MXene aerogel heater also outperforms the Ti₃C₂T_x MXene thin film and thread heater [17].

Zoom In Zoom Out Reset image size

It should be noted that we did not observe any rise in temperature from the epoxy resin/Ti₃C₂T_x MXene composite made from simple shear mixing with any application of external voltage up to 20 V. This fact is due to the discontinuous MXene sheets in the composite which result in incomplete electrically and thermally conducting pathways. Although Ti₃C₂T_x has been found to be exhibit promising and impressive Joule heating features [17, 18], we demonstrate that the combination of epoxy and Ti₃C₂T_x aerogel can be a potential candidate due to better electrothermal behaviour.

Another prominent feature of the thermal images of all the samples is the spatial variation in temperature (over an approximate area of 1.3 × 1.3 cm²). It is noteworthy, that the central uniform part of the epoxy resin/Ti₃C₂T_x MXene aerogel composite is observed to be around 40% hotter than its peripheral region (figures 4(a)–(c)). On the contrary, a non-uniform temperature distribution over the surface has been observed from the Ti₃C₂T_x aerogel (figures 5(a)–(b)). In addition, the central part shows lower surface temperature than the two sides of the bare Ti₃C₂T_x aerogel. This is due to the porosity of the Ti₃C₂T_x aerogel which allows heat convection and radiation to the surrounding air and the thermally isolating nature of the air in the aerogel structure that restricts the heat transfer [33]. However at the sides of the sample, lower air density and direct contact with the clamp at the sides of the sample give rise to a locally higher temperature field (figures 4(d)–(f)). On the other hand, epoxy resin is uniformly incorporated throughout the Ti₃C₂T_x aerogel and hence able to maintain an uniform surface temperature upon application of the external voltage.

Zoom In Zoom Out Reset image size

As seen from the figure 6(a), the Joule heating profile of the sample follows three stages: the initial increase in surface temperature with time (0–160 s), a steady-state zone (160–800 s) and a regime where it recovers to its original condition (800–1000 s). The rise in temperature is directly proportional to the square of the applied voltage and inversely proportional to the resistance of the materials. It has also been seen that the electrical conductivity reduces linearly with the temperature (figure 3(c)). Hence, at a higher applied voltage, a better and quicker response in the temperature distribution is observed for the epoxy resin/Ti₃C₂T_x aerogel composite (figures 6(b)–(c)). The response time, which is defined as the time required to attain 90% of the steady-state temperature from room temperature, is another deciding factor for evaluating the Joule heating performances (see table 1). The aerogel composite shows a heating rate of 3.5 °C s⁻¹cm⁻³ in the initial stage under the applied voltage of 3 V (figure 6(c)). It is also important to see from figure 6(c) that the cooling profile of the aerogel composite follows similar trends with respect to the applied voltage as did the heating rate. A greater dissipation takes place at a higher temperature and it can maintain the steady-state temperature for the desired time, indicating its 'self-regulating' behaviour. As a higher voltage is applied, the power delivery is increased and hence the surface temperature of epoxy resin/Ti₃C₂T_x aerogel composite is increased up to 166 °C at 3 V. The drastic enhancement of specific power (power density) from 1.7 to 13.9 W cm⁻² (5.7–46.3 W cm⁻³, considering a height of 3 mm) is observed as the input voltage increased from 1 to 3 V, as shown in figure 6(d). The energy densities of the studied materials are estimated using the relation: specific energy = specific power × heating time (see table 1). This result confirms the significant benefits of using our composite as an effective heater.

Zoom In Zoom Out Reset image size

To gain insight into the electric heating behaviour of the epoxy resin/Ti₃C₂T_x aerogel composite, the temperature–time profile (figure 6(a)) was further analysed. In the heating zone, the temperature–time profile can be expressed by the following equation [25, 34]

$$\begin{equation}\left( {\frac{{{T_{\text{t}}} - { }{T_0}}}{{{T_{\text{m}}} - { }{T_0}}}} \right) = 1 - \exp \left( { - { }\frac{t}{{{\tau _{\text{g}}}}}} \right)\end{equation} \tag{ 2 }$$

where T₀, T_m, and T_t, are the initial temperature, maximum temperature and temperature at any arbitary time (t), respectively. The characteristic rate constant ($\tau$ _g) values for the composite could be evaluated by fitting data in the heating zone of the temperature–time plots, as summarised in table 2. A low $\tau$ _g value represents a faster thermal response to the applied voltage. It is clearly seen from figure 6(a) that the surface temperature of the composite is higher and found to be stable over 10 min without any deterioration at higher input voltage (V₀) and steady-state current (I_c). In this zone, the net heat gain is transferred to the surroundings by radiation and convection (h_r+c) via the following equation

$$\begin{equation}{h_{{\text{r}} + {\text{c}}}} = \frac{{{I_{\text{c}}}{V_0}}}{{{T_{\text{m}}} - { }{T_0}}}.\end{equation} \tag{ 3 }$$

Table 2. Extracted characteristic parameters ($\tau$ _g, $\tau$ _d, and h_r+c) for the Joule heating performance of epoxy resin/Ti₃C₂T_x aerogel composite at different applied voltages.

Sample	Voltage (V)	${\mathbf{\tau }}$ g (s)	hr+c (W °C−1)	${\mathbf{\tau }}$ d (s)
	1	38.7 $\pm$ 0.5	0.050	28.0 $\pm$ 1.3
Epoxy resin/Ti3C2Tx	1.25	64.5 $\pm$ 1.0	0.035	86.8 $\pm$ 6.5
aerogel composite	1.5	66.9 $\pm$ 1.8	0.031	72.4 $\pm$ 1.1
	1.75	72.3 $\pm$ 0.8	0.027	67.0 $\pm$ 3.2
	2	44.0 $\pm$ 2.6	0.027	55.0 $\pm$ 4.0
Ti3C2Tx aerogel	2	10.22 $\pm$ 2.1	0.348	24.4 $\pm$ 7.8

As given in table 2, this value of h_r+c highlights the good electric heating efficiency of the epoxy resin/Ti₃C₂T_x MXene aerogel composite [25]. In the cooling zone, the surface temperature of epoxy resin/Ti₃C₂T_x MXene aerogel composite drops very rapidly as the input voltage is turned off. To find out the characteristic decay time constant (${\tau _{\text{d}}}$) the cooling profile was fitted with equation (4) and the extracted value is tabulated (see table 2).

$$\begin{equation}\left( {\frac{{{T_{\text{t}}} - { }{T_0}}}{{{T_{\text{m}}} - { }{T_0}}}} \right) = \exp \left( { - { }\frac{t}{{{\tau _{\text{d}}}}}} \right).\end{equation} \tag{ 4 }$$

A low ${\tau _{\text{d}}}$ value at a higher applied voltage indicates faster recovery of the composite. Overall the aerogel composite shows a faster response with excellent heat dissipation along the plane of MXene alignment. Significantly, the cooling profile of the composite is found to be a mirror image of the heating characteristics being in good agreement with equations (2) and (4).

To examine the stability of the materials, the Joule heating test was repeated for a prolonged steady-state phase at 2 V applied voltage. Figure 7(a) shows the prolonged steady-state phase of bare MXene aerogel and epoxy resin/Ti₃C₂T_x MXene aerogel composite for 4 h. Moreover, figure 7(b) shows the Joule heating cycles of the epoxy resin/Ti₃C₂T_x MXene aerogel composite and bare MXene aerogel for several cycles at an applied voltage of 2 V. The temperature profile of bare MXene aerogel and epoxy resin/Ti₃C₂T_x MXene aerogel composite for repeated cycles is shown in figure S3 (see supplementary materials). The cycle stability of epoxy resin/Ti₃C₂T_x aerogel composite at different applied voltages is shown in figure 7(c) for each input voltage. The trapped water molecules between MXene layers could be evaporated during the rapid local heating, leading to crack formation and hence it may lead to a deterioration in performance. Since we cured the composite at the temperature of 100 °C over a long time of 4 h, such a possibility can be neglected here. Most importantly, the results shown in figure 7 offer direct proof of the structural stability of the aerogel composite as an electrothermal heater.

Zoom In Zoom Out Reset image size

To strengthen the statement, we carried out an XPS study of the studied materials before and after Joule heating (figures S4–S6 in supplementary materials). The extracted elemental composition is listed in table 3. As seen from table 3, epoxy resin/Ti₃C₂T_x MXene aerogel composite do not show any significant structural changes. However, slight changes in the Ti/C ratio from 1.29 to 1.53 have been observed for the bare Ti₃C₂T_x MXene after the Joule heating (table 3). This change can be attributed to the formation of TiO₂ on the surface. It is important to note that, Ti/C ratio of epoxy resin/Ti₃C₂T_x MXene is relatively lower than the bare Ti₃C₂T_x MXene due to the carbon content of the epoxy. Although the epoxy resin/Ti₃C₂T_x MXene aerogel composite reaches a much higher surface temperature compared to the bare MXene aerogel, the existing epoxy resin can protect the MXene nanofillers in the composites from oxidation and hence the Ti/C ratio remains unchanged even after Joule heating. Thus, our result confirms that the both MXene aerogel and epoxy resin/Ti₃C₂T_x MXene aerogel composite have excellent structural stability even after several Joule heating cycles and after prolonged steady-state thermal exposure.