1. Introduction

Nanofluids are advanced colloidal systems composed of nanoparticles, typically metal oxides or carbon-based materials, dispersed in conventional base fluids such as water, ethylene glycol, or oil. They exhibit higher thermal conductivity and single-phase heat transfer coefficients than their base fluids, making them more suitable for high-performance cooling systems, heat exchangers, and industrial thermal processes. They demonstrate significant application potential in several fields including automotive radiators, heat exchangers, solar systems, and electronic cooling mechanisms^[1,2]. As a currently prominent fluid type, nanofluids have stimulated researchers’ interest in their synthesis techniques^[3], thermophysical properties, heat transfer mechanisms, and applications over the past decades^[4-6]. Currently, the widespread demand for more efficient fluids has led to the development of nanofluids as advanced thermal fluids^[7]. Among these, water-based and oil-based nanofluids have been extensively studied^[8-10]. Soylu et al.^[11] prepared three types of silica-coated core-shell nanoparticles and dispersed them in water to form a 2% mass concentration nanofluid. They systematically evaluated its thermophysical and optical properties, confirming the effectiveness of such fluids in high-temperature heat transfer and optical applications. Esfe et al.^[12] synthesized two ternary aqueous nanofluids with identical components but different ratios. Experimental comparisons of thermal conductivity at various temperatures (25-50 °C) and solid volume fractions revealed that the MWCNT(50%)-CuO(25%)-SiO₂ (25%) exhibited superior thermal conductivity under all tested conditions. However, their high vapor pressure, volatility, or flammability at elevated temperatures pose safety concerns, limiting their application in concentrated solar thermal energy utilization. Ionic liquids, with their extremely low vapor pressure and wide liquid phase range, were once considered ideal alternative base fluids. Yet their high cost, complex synthesis steps, and potential toxicity hindered large-scale industrial application. In recent years, deep eutectic solvents (DESs) have emerged as novel ion-liquid-like solvents. Combining the thermal stability advantages of ionic liquids with low cost, simple preparation, non-toxicity, environmental friendliness, and biodegradability, DESs have been hailed as “green solvents of the 21st century”^[13-15]. Therefore, DES-based nanofluids demonstrate significant potential in solar thermal energy utilization at medium and high temperatures.

Researchers have conducted studies on DES-based nanofluids. Jafari et al.^[16,17] prepared binary DES based on choline chloride (ChCl) and ethylene glycol (EG), along with their aqueous ternary systems. Using these as base fluids, they dispersed MgO nanoparticles to create nanofluids. The study systematically investigated the effects of nanoparticle concentration (1-10 wt.%), temperature, and water content on fluid stability, density, thermal conductivity, and the isothermal thermal expansion coefficient. Experimental data indicate that these fluids hold potential for superior performance in heat transfer applications compared to conventional fluids. Additionally, they analyzed the preparation process (e.g., ultrasonic treatment) and stability mechanisms (e.g., surfactant, pH control) of DES-based nanofluids, and explored the application of theoretical models in predicting their thermal behavior. They concluded that these fluids overcome the high cost and complex synthesis of ionic liquids, and exhibit superior thermal properties compared to pure DES. Dehury et al.^[18] predicted and synthesized a novel DES (diphenyl ether/DL-menthol) using the COSMO-SAC model. They compared it with another DES (DL-menthol/oil alcohol) as a base. They dispersed hexagonal boron nitride (h-BN) nanoparticles in this DES to prepare a nanofluid, tested its rheological and thermal properties, and selected the optimal fluid based on the Mo number for forced convection experiments and Aspen Plus process simulation. AN et al.^[19] synthesized four ChCl-based DESs, selected the system with the best desulfurization performance, and used it as the base liquid to prepare nanofluids by dispersing supported molybdate. Fang et al.^[20] synthesized multiple DESs using ammonium salts (ChCl) and methylisothiazolinone (MTPB) with different molar ratios and hydrogen bond donors (EG, triethylene glycol (TEG)). Walvekar et al.^[21] prepared eleven DESs based on varying ratios of MTPB/ChCl and EG/TEG. By dispersing 0.04 wt.% carbon nanotubes (CNTs) in these fluids, they demonstrated the feasibility of DES-based CNT nanofluids as highly efficient heat transfer media under extreme temperature conditions. Compared to the more commonly studied ChCl/EG DES, the EG/potassium acetate (PA) system selected in this work offers several advantages: (1) potassium acetate is significantly lower in cost and toxicity than choline chloride; (2) the EG/PA DES exhibits a wider liquid temperature range; and (3) the acetate ions in this system can coordinate with Zn²⁺ ions during the in situ synthesis, facilitating controlled nanoparticle nucleation and growth.

Among numerous nanoadditives, zinc oxide (ZnO) stands out as a wide-bandgap semiconductor material. Its properties, including high surface tension, strong magnetic properties, and potent ultraviolet absorption and scattering capabilities, have made it a focal material in current heat transfer fluid research. For example, Sahu et al.^[22] compared the performance of single-slope solar distillers using different nanofluids (ZnO, SiO₂, and their mixtures) versus those without nanofluids at varying water depths. Arun et al.^[23] investigated the impact of ZnO nanofluids combined with a dimpled tube and twisted helical tape (DTHTT) design on heat transfer and friction performance in parabolic trough solar water heaters, through experimental and computational fluid dynamics (CFD) simulation validation. Boroomandpour et al.^[24] conducted comprehensive experimental studies on the thermal conductivity of ternary mixed nanofluids containing MWCNTs, TiO₂, and ZnO, along with corresponding binary and monomeric nanofluids (based on an 80.20 water/ethylene glycol base fluid). Results indicate that thermal conductivity increases with rising temperature (25-50 °C) and volume fraction (0.1%-0.4%), with the highest enhancement observed in the MWCNT monomolecular nanofluids at 0.4% concentration and 50 °C. Nadooshan et al.^[25] measured the thermal conductivity of zinc oxide (ZnO)/ethylene glycol-water (30:70) nanofluids across varying temperatures (20-50 °C) and a broad volume fraction range (0-4%). Results indicate a monotonically increasing trend in thermal conductivity with rising particle concentration and temperature, peaking at a 20% enhancement at 50 °C and 4% volume fraction. While materials such as CuO and carbon-based nanostructures offer stronger visible-light absorption, ZnO was specifically selected in this study for three key reasons: (1) its high thermal conductivity (approximately 50 W/m·K for bulk ZnO) provides effective thermal transport enhancement; (2) its wide bandgap (3.37 eV) ensures excellent chemical stability and photo-corrosion resistance under prolonged illumination; and (3) the feasibility of its in situ synthesis from the zinc hydroxide (Zn(OH)₂) precursor via microwave-induced dehydration, which is compatible with the DES system without introducing additional impurities.

Currently, the primary method for preparing ZnO/DES nanofluids employs a “two-step process”. First synthesizing or purchasing ZnO powder, then dispersing it into DES via mechanical stirring or ultrasonication. However, due to the superior properties of nanofluids with large specific surface areas, the two-step method-prepared nanofluids are highly prone to agglomeration and difficult to disperse. This instability not only causes the thermal properties of the fluid to decay over time but can also lead to severe pipeline blockages, resulting in the failure of the entire heat exchange system. Since the structure of deep eutectic solvents depends on hydrogen bonding, adding dispersants is not feasible as they destroy these bonds. In contrast, the “one-step” in situ synthesis strategy enables direct surface modification using solvent molecules during nanoparticle nucleation and growth, achieving superior self-stability. However, studies on the in situ synthesis of ZnO in DES and its photothermal performance remain scarce. Previous studies have demonstrated that two-step methods for preparing ZnO nanofluids typically require surfactants to achieve short-term stability, but these surfactants can degrade the thermophysical properties of the base fluid^[26,27].

Currently, our group has systematically investigated the thermophysical properties of DES systems composed of ethylene glycol and various acetates, as well as the thermal conductivity and photothermal performance of nanofluids^[28-32], while a one-step method for ZnO filled DES nanofluids has not been intensively studied. In this work, we employed a eutectic solvent composed of ethylene glycol and potassium acetate as the base liquid, using Zn(OH)₂ as the precursor. By controlling the heating reaction, highly dispersible ZnO nanofluids were directly synthesized within the DES system. Comprehensive investigations were conducted on their thermophysical properties. Furthermore, this method requires no additional dispersants, as the components in the DES naturally form a solvation layer around the nanoparticles during the reaction, conferring exceptional long-term stability on the fluid. This work aims to provide theoretical and experimental support for developing novel, low-cost, and highly stable solar heat transfer fluids. Very recently, some studies have explored in situ nanoparticle synthesis in DES systems. For instance, Liu et al.^[32] reported the in situ formation of CuO nanoparticles in ChCl/EG DES for enhanced thermal conductivity. Our work extends this concept to ZnO nanoparticles and, more importantly, demonstrates for the first time the photothermal conversion performance of in situ synthesized ZnO/DES nanofluids, highlighting the unique advantage of ZnO’s strong UV absorption for solar thermal applications.

2. Experimental Section

2.1 Materials

Potassium acetate anhydrous (PA, CH₃COOK, analytical grade) and zinc hydroxide (Zn(OH)₂, analytical grade) were purchased from Aladdin Biochemical Technology Co., Ltd., Shanghai, China. Ethylene glycol (EG, HOCH₂CH₂OH, analytical grade) was purchased from McLean Biochemical Technology Co., Ltd., Shanghai, China.

2.2 Preparation of DES

DES was prepared using the melt blending method. A laboratory balance was used to precisely weigh a mixture of EG and PA at a molar ratio of 5:1. To facilitate subsequent nanofluid preparation and sample comparison, the required DES was prepared in batches in large quantities. The total mass weighed per batch was 120 grams, with weighing accuracy controlled within ±0.0025%. The weighed mixture was placed in a 200 mL flask and stirred at 80 °C and 300 rpm for 1 hour until the solvent became clear and transparent with no observable acetate solid particles. Subsequently, the mixture was cooled to room temperature to form a stable homogeneous system, yielding a 5:1 molar ratio ethylene glycol-potassium acetate copolymer.

2.3 Sample preparation

Nanofluids with ZnO mass fractions of 0.5 wt.%, 1 wt.%, 2 wt.%, and 5 wt.% were prepared. The specific procedure is as follows. For each desired nanofiller mass fraction, the corresponding proportion of Zn(OH)₂ precursor (for pre-synthesized ZnO nanoparticles) was added to the DES solvent system, and the mixture was stirred at 75 °C under atmospheric pressure for 1 hour. The resulting mixture was subjected to ultrasonication in an ultrasonic bath (KQ-500DE, Kunshan Ultrasonic Instruments Co., Ltd., China; operating frequency: 35 kHz; power: 500 W) for 2 hours. During ultrasonication, the bath temperature was maintained below 40 °C by periodically replacing the water to prevent sample overheating and to form a uniformly dispersed suspension. Then, after 2,000 W microwave radiation for 5 minutes, the nanofluid with the target ZnO mass fraction was obtained. The sample volume for each batch was approximately 50 mL (mass about 60 g). During microwave irradiation, the temperature was monitored but not actively controlled; the rapid dehydration reaction was completed within 5 minutes without causing local overheating that would affect the DES structure, which has been confirmed in our previous work^[28-32]. Specifically, for preparing 50 g of nanofluid, the required masses of Zn(OH)₂ precursor were: 0.620 g for 0.5 wt.%, 1.245 g for 1 wt.%, 2.504 g for 2 wt.%, and 6.415 g for 5 wt.% ZnO (calculated based on the stoichiometric conversion: Zn(OH)₂ → ZnO + H₂O, where 1 g ZnO corresponds to 1.245 g Zn(OH)₂). The actual ZnO concentration in the prepared nanofluids was verified by thermogravimetric analysis (TGA). The residual mass at 300 °C, corresponding to the ZnO content, was compared with the nominal mass fraction, and the deviation was within ±5%, confirming the accuracy of the preparation method.

The nanoparticles in the one-step prepared nanofluids exhibited good static compatibility with the DES under microwave irradiation, which was verified by visual inspection (absence of visible precipitation). This favorable compatibility stems from a critical step in nanofluid preparation: microwave irradiation-induced dehydration of Zn(OH)₂ to form ZnO. The resulting ZnO nanoparticles are formed in situ. This in situ synthesis effectively avoids the agglomeration inherent in conventional nanoparticles, as evidenced by the long-term static stability observations (Figure 1c) and the consistent thermophysical property measurements over time, ensuring superior dispersion efficiency and significantly enhancing the nanofluids’ stability.

Display Full Size

Download

Figure 1. (a) Schematic illustration of the reaction mechanism for the in situ synthesis of ZnO nanoparticles within the DES system; (b) Schematic diagram of the preparation process for ZnO/DES nanofluids via the one-step microwave-assisted method; (c) Digital photographs of in situ synthesized ZnO/DES nanofluids with varying mass fractions (0.5-5 wt.%) after two weeks of static storage, showing minor sedimentation at the two-week mark; (d) Digital photographs of ZnO/DES nanofluids prepared by direct ultrasonic dispersion of commercial ZnO nanoparticles, taken after 2 hours of static storage, exhibiting rapid and severe aggregation and sedimentation. DES: deep eutectic solvent; EG: ethylene glycol.

Figure 1a,b schematically illustrate the reaction mechanism and the one-step preparation process of ZnO/DES nanofluids, respectively. To evaluate the long-term dispersion stability, nanofluids at varying mass fractions were stored under static conditions. As shown in Figure 1c, the in situ synthesized nanofluids maintained excellent dispersion with no visible precipitation during the first week, and only minor sedimentation occurred after two weeks. To further demonstrate the advantage of the in situ synthesis strategy, a comparative experiment was performed in which commercially available ZnO nanoparticles (50 nm, Aladdin) were directly dispersed into the DES by ultrasonic oscillation at the same mass fractions. Figure 1d reveals that nanofluids prepared by directly dispersing commercial ZnO nanoparticles via ultrasonic oscillation underwent rapid aggregation and substantial sedimentation within just 2 hours. This direct comparison unequivocally demonstrates that the one-step in situ synthesis strategy is essential for achieving the superior dispersion stability of ZnO/DES nanofluids.

2.4 Uncertainty analysis

Uncertainty analysis was considered. All measurements were performed at least three times, and the reported values represent the mean ± standard deviation. The experimental uncertainties were: ±0.1 °C for temperature, ±1% for viscosity, ±2% for thermal conductivity, ±3% for specific heat capacity, ±0.0005 g/cm³ for density, and ±5% for photothermal conversion efficiency. The reproducibility of the preparation procedure was verified by preparing three independent batches of the 1 wt.% nanofluid, which showed thermophysical property variations within ±3%.

3. Results and Discussion

3.1 Characterization of nanoparticle

To further verify the morphology and size of the in situ synthesized ZnO nanoparticles, focused ion beam-scanning electron microscopy (FIB-SEM; Tescan GAIA3, Czech Republic) characterization was performed. The ZnO nanoparticles were extracted from the 5 wt.% nanofluid by filtration followed by drying at 160 °C for 6 h. As shown in Figure 2a,b, the ZnO nanoparticles exhibit a block-like (irregular polyhedral) morphology with an average particle size of approximately 120 nm. Compared with the Zn(OH)₂ precursor (Figure 2c,d), which displays a typical flaky/irregular morphology consistent with previously reported observations, the ZnO particles obtained after microwave-induced dehydration show a more compact and well-defined structure, confirming the successful conversion of the precursor. It should be noted that the drying process may introduce a certain degree of particle aggregation; nevertheless, the primary particle size observed by SEM is consistent with the in situ dispersed state during nanofluid operation, as supported by the stable thermophysical properties and long-term dispersion behavior.

Display Full Size

Download

Figure 2. SEM images of (a, b) ZnO nanoparticles and (c, d) Zn(OH)₂ precursor particles. SEM: scanning electron microscopy.

3.2 Viscosity

For nanofluids, the optimal size, shape, and properties of nanoparticles, along with their dispersion, significantly influence the enhancement of the thermophysical properties of nanofluids. Among these, viscosity determines the pumping power required for the fluid and is a crucial factor in thermophysical properties, playing a vital role in evaluating the performance of working media^[17]. A rotational viscometer (NDJ-5S, Shanghai Precision Instrument Co., Ltd.) was employed to measure the viscosity of the nanofluids at a fixed shear rate of 100 s^-1. Prior to measurements, the Newtonian behavior of all samples was verified by measuring the shear stress–shear rate relationship over the range of 20-200 s^-1 at 25 °C. All samples exhibited linear shear stress–shear rate profiles with correlation coefficients R² > 0.99, confirming Newtonian fluid behavior under the tested conditions. Test results are presented in Figure 2a. We measured the viscosity of DES and four different concentrations of the nanofluids across a temperature range of 25-60 °C. Findings show that increasing temperature significantly improved the nanofluids’ flowability, with viscosity exhibiting a decreasing trend as the test temperature rose. For instance, when the temperature increased from 25 to 60 °C, the viscosity of the 5% nanofluids decreased from 80 mPa·s to 33 mPa·s. This decrease may be attributed to enhanced molecular kinetic energy, which increases molecular motion and weakens the hydrogen bond network.

Conversely, the viscosity of the nanofluids exhibited a pronounced increase with rising ZnO nanoparticle mass fraction. Nanofluids with low mass fractions showed minimal viscosity enhancement, whereas those with a 5% mass fraction exhibited the most significant viscosity increase (Figure 3a), with the viscosity enhancement ratio relative to the base fluid shown in Figure 3c. At 25 °C, the base liquid viscosity was 36.8 mPa·s, while the viscosity of the 5% concentration nanofluids increased to 80 mPa·s. This can be attributed to the strong adsorption of acetate anions and polar ethylene glycol molecules from the DES onto the charged surfaces of ZnO nanoparticles. This adsorption increases the “effective hydrodynamic radius” of the nanoparticles, leading to greater resistance during particle motion and consequently higher viscosity. Additionally, as particle number density increases, the relative sliding between fluid layers is impeded by more solid particles, further contributing to the viscosity increase.

Display Full Size

Download

Figure 3. (a) Viscosity and (b) thermal conductivity of the DESs based nanofluids with various mass fraction of zinc oxide; (c)The viscosity enhancement ratio of nanofluids with different ZnO mass fractions relative to the base fluid as a function of temperature; (d)The thermal conductivity enhancement ratio of nanofluids with different ZnO mass fractions relative to the base fluid as a function of temperature; (e) schematic of the test platform of specific heat capacity; (f) specific heat capacity of the DESs based nanofluids with various mass fraction of zinc oxide. DESs: deep eutectic solvents.

Although nanoparticle concentration increases nanoparticle fluid viscosity, during steady state operation at high temperatures, even the highest-concentration nanoparticle fluids exhibit viscosity drops below 40 mPa·s, a reduction approaching 60%. This indicates that while nanoparticle fluids enhance thermal conductivity and photothermal efficiency, their pumping power consumption remains within an engineering-acceptable range. For context, typical viscosity limits for solar thermal heat transfer fluids in pumping systems are generally below 100 mPa·s at operating temperatures. The maximum viscosity of 80 mPa·s at 25 °C for the 5 wt.% nanofluid remains within this acceptable range, and the viscosity drops below 40 mPa·s at elevated temperatures (above 50 °C), which is well within the typical operating range of conventional solar thermal circulation pumps. Notably, as shown in Figure 2c, the viscosity enhancement ratio (the ratio of nanofluid viscosity to base DES viscosity) decreases with increasing temperature, indicating that the relative viscosity penalty of adding nanoparticles is mitigated at higher operating temperatures. This is particularly beneficial for solar thermal applications where the fluid typically operates at elevated temperatures.

3.3 Thermal conductivity

Following the viscosity investigation, we analyzed another critical thermophysical property for evaluating nanofluid performance. thermal conductivity. Thermal conductivity was measured using a transient hot-wire instrument (TC 3000E, Xiaxi Technology, China) with an accuracy of ±2%. The transient hot-wire method was selected for its rapid measurement speed and minimal convective interference. Initial testing conditions involved maintaining the actual temperature within ±0.05 °C for ten minutes in a water bath. As shown in Figure 3b, all ZnO nanofluids exhibited significantly enhanced thermal conductivity compared to the base DES (Figure 3b), with the thermal conductivity enhancement ratio shown in Figure 3d. This improvement amplified with increasing temperature and particle concentration. For instance, at 65 °C, the 5% ZnO nanofluid demonstrated optimal thermal performance, achieving approximately 12% higher thermal conductivity than the base DES. These results demonstrate the superior heat transfer advantages of this nanofluid in high-temperature applications such as solar thermal collectors. The enhanced thermal conductivity after nanoparticle addition stems from multiple factors. First, as discussed previously, the viscosity of the DES base liquid decreases significantly at high temperatures, greatly reducing constraints on particle motion. Consequently, the vigorous Brownian motion of ZnO nanoparticles directly carries heat for diffusion. Simultaneously, based on the micro convection effect, these nanoparticles induce random micro-scale convection within the surrounding liquid, generating intense disturbances within the fluid and significantly enhancing heat exchange efficiency at the microscopic level. Second, as a typical wide-bandgap semiconductor, ZnO possesses an intrinsic thermal conductivity far exceeding that of the organic-based DES. Consequently, the introduction of this highly thermally conductive solid phase establishes an efficient thermal conduction pathway. For reference, the absolute thermal conductivity of the base DES ranged from approximately 0.22 W/m·K at 25 °C to 0.24 W/m·K at 65 °C. While a 12% enhancement in thermal conductivity with 5 wt.% ZnO loading may appear moderate, it is comparable to or exceeds values reported for other DES-based nanofluids at similar loadings^[16,17]. The moderate enhancement is offset by the significant advantages in dispersion stability and the anomalous specific heat capacity enhancement, which together contribute to superior overall thermal performance.

3.4 Specific heat capacity

Specific heat capacity determines a working fluid’s ability to store thermal energy per unit temperature rise, directly influencing the heat storage density of solar thermal systems. To further investigate the specific heat capacity of ZnO nanofluids, we conducted measurements using a custom-built platform. The specific heat capacity testing setup is illustrated in Figure 3e. The setup specifically includes: a direct current (DC) power supply, resistance wire, thermal insulation cotton, and thermocouples. The resistance wire heats the nanofluids, the insulation cotton minimizes heat exchange between the test system and the environment, and the thermocouples monitor the liquid’s temperature in real time. Using this platform, we recorded the voltage and current across the heating resistance wire terminals.

To evaluate the energy storage capacity of the nanofluids, the heat exchanged between the electrically heated plate and the external environment is given by Equation (1).

(1)$\Delta Q=P\Delta t=UI\Delta t$

where U and I represent the voltage across the resistive wire and the current flowing through it, respectively. Based on the definition of specific heat, this is expressed by Equation (2).

(2)$\Delta Q=\mathrm{cm}\Delta T$

where m is the mass of the liquid being measured, and c is the specific heat capacity of the liquid being determined. Based on Equations (1) and (2), the formula for calculating specific heat capacity is given by Equation (3).

(3)$c=\frac{UI\Delta t}{m\Delta T}$

Based on the above formulae and fundamental principles, we tested the specific heat capacity of the prepared nanofluids and analyzed its variation with temperature, as shown in Figure 3f. First, we observed that the specific heat capacity of all samples exhibited an increasing trend with rising temperature. Interestingly, we found that after adding nanoparticles, the specific heat capacity did not decrease as predicted by classical theory with increasing nanoparticle mass fraction. Instead, it demonstrated a significant anomalous enhancement phenomenon. For instance, at 25 °C, the specific heat capacity of 5% nanoparticle-dispersed nanofluids reached 2.43 J/g·K, representing an approximately 14.6% increase over the pure DES base liquid. This trend persisted at elevated temperatures. At 65 °C, the specific heat capacity of the 5% nanoparticle-dispersed nanofluids reached 2.61 J/g·K, showing an approximately 13.9% improvement over the DES base liquid.

This anomalous enhancement may result from multiple factors. On one hand, ZnO nanoparticles possess an extremely large specific surface area. Atoms on their surfaces exhibit higher potential energy and more intense thermal vibrations, providing additional energy storage pathways during heat absorption, manifesting as increased specific heat capacity. On the other hand, within the DES system, ethylene glycol molecules and acetate ions may form an ordered interfacial layer on the ZnO surface through strong electrostatic interactions and hydrogen bond rearrangement. This process absorbs substantial heat, contributing to the observed increase in specific heat capacity. This anomalous specific heat capacity enhancement has also been observed in other nanofluid systems, such as molten salt-based nanofluids with SiO₂, Al₂O₃, and TiO₂ nanoparticles, where the high specific surface area of nanoparticles and the formation of interfacial liquid layers with ordered molecular structures were identified as key mechanisms^[17,18]. The enhancement observed in our ZnO/DES system (up to 14.6%) is particularly significant because DES has an inherently higher specific heat capacity than molten salts, making the relative enhancement more noteworthy.

This research finding indicates that under identical volume and temperature difference conditions, ZnO/DES nanofluids can transport greater thermal energy, demonstrating the dual advantages of “rapid heat transfer” and “high heat storage capacity”. This positions them as a highly promising next-generation working fluid for solar thermal-to-photovoltaic conversion.

3.5 Density characteristics

To investigate the effect of ZnO nanoparticle addition on the density of ethylene glycol/potassium acetate DES, we employed a densimeter to measure the density of pure DES and four ZnO/DES nanofluids at different mass fractions across a temperature range of 25 to 65 °C. The results are presented in Figure 4a. The density of all samples exhibited a linear decrease with increasing temperature, which is in line with the general law of liquid thermal expansion.

Display Full Size

Download

Figure 4. (a) Density of DES-based nanofluids with different zinc oxide mass fractions; (b) Thermogravimetric curves of DES-based nanofluids with varying ZnO mass fractions. DES: deep eutectic solvent; EG: ethylene glycol.

At 35 °C, the density of the base DES solution is 1.1931 g/cm³. Upon adding 0.5 wt.%, 1 wt.%, 2 wt.%, and 5 wt.% ZnO nanoparticles, the densities increase to 1.1973, 1.2023, 1.2045, and 1.2092 g/cm³, respectively. That is, at the same temperature, the density of all nanofluids was significantly higher than that of the pure DES base liquid and monotonically increased with the mass fraction of ZnO nanoparticles. The direct cause of this phenomenon is that the intrinsic density of ZnO nanoparticles is much higher than that of the DES base liquid. Their addition is equivalent to introducing a higher-density solid phase component into the system, thereby increasing the overall density.

Although density increases with nanoparticle addition, the absolute increase remains relatively limited within the mass fraction range (0-5 wt.%) investigated here. This result indicates that within this concentration range, the density of the fluid system remains predominantly governed by the low-density continuous phase of the DES with nanoparticles contributing only a small proportion to the overall density as a dispersed phase. This characteristic is beneficial to practical applications. It means that while adding nanoparticles can improve photothermal performance, stability, and other properties, the original flow inertia characteristics of the working fluid can be largely maintained. This avoids imposing excessive additional loads on circulation pumping systems.

3.6 Thermogravimetric analysis

Our study investigated the influence of ZnO nanoparticles on the thermal decomposition behavior of ethylene glycol/potassium acetate DES through thermogravimetric analysis. Figure 4b displays the TG curves of four nanofluid samples with different ZnO mass fractions (0.5 wt.%, 1 wt.%, 2 wt.%, 5 wt.%).

As shown in Figure 4b, all samples exhibited observable weight loss between 80 and100 °C and entered a rapid decomposition phase at 150 to 250 °C. This indicates that the ZnO/DES nanofluid maintains structural integrity below 150 °C, meeting the fundamental thermal stability requirements for medium-to-high-temperature solar thermal systems. The intended operating temperature range for the proposed solar thermal system is 100-250 °C, which aligns well with the thermal stability window observed in TGA. The fluid maintains structural integrity below 150 °C, and the rapid decomposition phase (150-250 °C) still allows for short-term operation at elevated temperatures, with the upper limit corresponding to the onset of significant mass loss.

Notably, the weight loss curves of samples at different concentrations were nearly similar throughout this process, indicating substantially consistent thermal decomposition rates. This characteristic signifies that the thermal stability of the system is primarily governed by the DES matrix, which means that adjusting the ZnO concentration does not induce abrupt changes in thermal stability, enhancing engineering flexibility. It provides crucial evidence for the controllable design and stable operation of ZnO/DES nanofluids in solar medium-to-high temperature collection systems.

3.7 Photothermal conversion effect

Photothermal conversion is a key physical process that directly and efficiently converts solar energy into utilizable thermal energy, holding central significance in medium-to-high temperature solar utilization systems. High-efficiency photothermal conversion implies that the working fluid can absorb more solar radiation in a shorter time and convert it into effective thermal energy that can directly drive thermal cycles or thermal storage systems. This directly enhances the instantaneous output power of the collector and the overall system efficiency. Besides, the photothermal conversion performance directly affects the operating temperature and application range of the system. Nanofluids with high photothermal conversion capability can support stable system operation at higher working temperatures, enabling applications such as medium-to-high temperature power generation.

To evaluate the practical application potential of the prepared ZnO/DES nanofluids in direct solar energy absorption and utilization, their photothermal conversion performance was systematically tested. In this study, as shown in Figure 5a, three thermocouples (upper, middle, and lower) were positioned at different depths within the fluid to monitor the dynamic evolution of the internal temperature field during illumination. During the photothermal experiments, no vapor bubble formation was observed in any of the tested samples, confirming that the fluid temperature remained below the boiling point and that the photothermal conversion process occurred under stable single-phase conditions.

Display Full Size

Download

Figure 5. (a) Photothermal conversion experimental platform schematic and temperature measurement point layout. Temperature changes at different measurement points under 1,400 W/m² light intensity; (b1) upper layer temperature (T1); (b2) middle layer temperature (T2); (b3) lower layer temperature (T3). Temperature changes at different measurement points under light intensity of 1,700 W/m²; (c1) upper layer temperature (T1); (c2) middle layer temperature (T2); (c3) lower layer temperature (T3). Temperature changes at different measurement points under an irradiance of 2,000 W/m²; (d1) upper layer temperature (T1); (d2) middle layer temperature (T2); (d3) lower layer temperature (T3). DES: deep eutectic solvent; EG: ethylene glycol.

Figure 5b,c,d show the temperature profiles of the base DES and ZnO nanofluid layers as a function of illumination time under different incident light intensities of 1,400 W/m², 1,700 W/m², and 2,000 W/m². At all tested light intensities, the incident light energy is primarily absorbed by the fluid’s surface layer and converted into thermal energy, causing the surface temperature to rise. So, the temperature of the upper layer (T1) of the nanofluids responds first and reaches the highest equilibrium temperature. The middle layer temperature (T2) exhibits a noticeable lag in response compared to the upper layer, while the lower layer temperature (T3) shows the slowest heating rate and the smallest temperature increase, thereby forming a top-to-bottom temperature gradient. This layered response characteristic is more pronounced at higher light intensities. Under 2,000 W/m² illumination, the temperature difference between the upper and lower layers can exceed 10 °C.

The incorporation of ZnO nanoparticles significantly enhances the fluid’s light absorption and thermal conversion capabilities. Compared to pure DES base liquids, all concentrations of nanofluids exhibit elevated equilibrium temperatures under identical illumination. Under 1,400 W/m² illumination, comparing data from the initial stage (first 1,000 s) of the experiment, the upper layer temperature change (ΔT) of the 0.5 wt.% ZnO nanofluid reached the highest value of 15.538 °C, while the base DES liquid only reached 12.032 °C. This indicates that introducing a small number of nanoparticles can effectively enhance the photothermal effect.

However, the enhancement of photothermal performance does not exhibit a simple positive correlation with nanoparticle concentration. When the concentration increased from 0.5 wt.% to 5 wt.%, the overall heating rate and final equilibrium temperature of the upper layer did not continue to rise. This phenomenon was particularly evident in the temperature changes of the lower layer. Under 1,400 W/m² irradiation, the temperature change in the lower layer of the nanofluid showed a decreasing trend with increasing concentration. The temperature change in the lower layer of the 5 wt.% sample was only 2.67 °C, significantly lower than that of low-concentration samples and the base fluid. This is likely attributed to the light-shielding effect caused by excessive nanoparticles in the upper layer at higher concentrations, which reduces the light intensity absorbed by the lower layer. Furthermore, since the fluid becomes overheated near the surface while there is inadequate heat accumulation in the lower layer, this non-uniform temperature distribution partially restrains downward heat transfer driven by natural convection, further weakening the overall temperature elevation of the fluid.

To quantitatively evaluate photothermal conversion efficiency, calculations were performed using Equations (4) to (6). This experiment assessed the photothermal performance of nanofluids by measuring their temperature rise under simulated illumination. Based on the principle of energy conservation, the heat absorbed by the fluid during illumination, Qabs, can be expressed as:

(4)$Q_{\mathrm{abs}}=\mathrm{mc}\Delta T$

where m is the fluid mass, c is the specific heat capacity, and ΔT is the temperature rise during the illumination period.

The total incident light energy Ein reaching the fluid surface is:

(5)$E_{in}=IAt$

where I is the light intensity, A is the illuminated area, and t is the irradiation duration.

Therefore, the photothermal conversion efficiency η is defined as the ratio of absorbed energy to incident energy.

(6)$\eta =\frac{Q_{\mathrm{abs}}}{E_{\mathrm{in}}}\times 100\%=\frac{\mathrm{mc}\Delta T}{IAt}\times 100\%$

The calculated results for photothermal conversion efficiency are shown in Figure 6. Overall, within the tested concentration range, the addition of ZnO nanoparticles enhanced the photothermal conversion efficiency of the fluid. Notably, light intensity significantly influenced efficiency. At lower light intensities, the photothermal conversion efficiency remained relatively high, and the efficiency enhancement effect of adding ZnO nanoparticles was more pronounced. As light intensity increased to 2,000 W/m², although the fluid’s final equilibrium temperature rose, the photothermal conversion efficiency generally showed a decreasing trend. This may be because of the gradual saturation of nanoparticle light absorption capacity under high irradiance, resulting in partial loss of energy that could not be effectively utilized.

Display Full Size

Download

Figure 6. Photothermal conversion efficiency at different measurement points under various light intensities. 1,400 W/m² light intensity. (a1) upper layer photothermal conversion efficiency; (a2) middle layer photothermal conversion efficiency; (a3) lower layer photothermal conversion efficiency. 1,700 W/m² light intensity; (b1) Upper layer photothermal conversion efficiency; (b2) middle layer photothermal conversion efficiency; (b3) lower layer photothermal conversion efficiency. 2,000 W/m² light intensity; (c1) upper layer photothermal conversion efficiency; (c2) middle layer photothermal conversion efficiency; (c3) lower layer photothermal conversion efficiency.

Further analysis revealed that photothermal conversion efficiency is closely related to the dynamic stages of the irradiation process, exhibiting a characteristic decline over time. The sample was irradiated under the corresponding light intensity for 6,000 seconds, with the first 5,000 seconds of data taken for analysis. Throughout the 5,000-second illumination, the photothermal conversion efficiency of the fluid does not remain constant but decreases as the fluid temperature rises. Taking typical data at a 1,400 W/m² illumination intensity as an example, during the initial low-temperature phase (first 1,000 seconds), the nanofluids exhibit the highest instantaneous conversion efficiency; as the temperature rises to the intermediate-temperature phase (middle 1,000 seconds), the efficiency value shows a significant decline; and when the system enters the high-temperature phase (final 1,000 seconds), efficiency further declines to approximately 20% of the initial value. This efficiency decay is attributed to intensified thermodynamic loss mechanisms. According to the Stefan-Boltzmann law, surface radiation losses increase four times with rising fluid temperature. Additionally, elevated temperatures significantly enhance convective heat transfer between the nanofluids and their surroundings. As the system gradually approaches thermal equilibrium, the net energy accumulation rate naturally slows.

Comprehensive analysis of photothermal performance and potential economic viability reveals an optimal range for nanoparticle mass fraction. Under the conditions of this study, the 0.5 wt.% ZnO nanofluid demonstrated relatively balanced and outstanding photothermal conversion performance in most test scenarios.

From an economic perspective, the one-step in situ synthesis method offers cost advantages by eliminating the need for: (1) separate nanoparticle synthesis or purchase; (2) surfactants or dispersants; and (3) extended ultrasonication processing. The primary raw material costs are estimated at approximately $15-20 per kilogram of nanofluid, compared to an estimated $30-45 per kilogram for conventionally prepared DES nanofluids using pre-synthesized ZnO nanoparticles with dispersants. The microwave-assisted synthesis also significantly reduces processing time (from hours of ultrasonication to minutes of microwave irradiation), further lowering energy consumption and manufacturing costs.

4. Conclusions

In summary, we successfully employed a “one-step in situ generation method” to prepare novel nanofluids using an ethylene glycol-potassium acetate DES as the base liquid and ZnO as the nanofiller. Through microwave-induced dehydration, Zn(OH)₂ was converted into ZnO nanoparticles within the DES, achieving excellent long-term dispersion stability without external dispersants. The addition of ZnO resulted in a slight increase in density and a concentration-dependent rise in viscosity, although the latter significantly decreases at elevated temperatures, facilitating practical engineering applications. Notably, the nanofluids demonstrated enhanced thermal conductivity—increasing by approximately 12% at 65 °C with a 5 wt.% concentration due to Brownian motion and microconvection—alongside an anomalous 14.6% increase in specific heat capacity, indicating simultaneous heat transfer and storage capabilities. While photothermal efficiency tends to decline at higher intensities and temperatures due to heat loss, considering stability, viscosity, and overall efficiency, the 0.5 wt.% sample is recommended as the optimal choice, although 1 wt.% and 2 wt.% samples showed comparable photothermal conversion efficiency in certain test stages. Consequently, considering stability, cost, and overall performance, the 0.5 wt.% ZnO/DES nanofluids is identified as the superior choice for medium-to-high temperature solar systems, confirming this green synthesis strategy as a feasible pathway for developing high-efficiency heat transfer fluids.

Acknowledgments

The authors thank Dr. Hua Wei and Dr. Rui Zhou at Advanced Analysis & Computation Center of CUMT for their assistance with chemical analysis.

Authors contribution

Zhang X: Methodology, investigation, writing-original draft.

Du X: Data curation, formal analysis, visualization.

Wang X: Investigation, validation, visualization.

Liu C: Conceptualization, sSupervision, project administration, funding acquisition, writing-review & editing.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and materials

All data and materials could be obtained from the corresponding author upon reasonable request.

Funding

This work was supported by the National Natural Science Foundation of China (51906252).

Copyright

© The Author(s) 2026.