Introduction

Insulation oil (INO), also referred to as dielectric oil, is a specialized type of oil intended for insulating electrical equipment. This type of oil finds common application in electrical systems and other high-voltage equipment, serving the purpose of electrical insulation and heat dissipation. Its primary role is to safeguard the integrity and efficiency of electrical systems by averting electrical breakdowns and offering protection against arcing. To avoid damage to the parts of high-voltage equipment and prevent performance degradation, it is crucial to cool this system. When the temperature of the INO exceeds a certain limit, the electrical system and other high-voltage equipment stop operating1. The dielectric coefficient, heat transfer coefficient, and AC BDV are critical parameters of INO that have a significant impact on its performance. For INO to be effective, it must possess both electrical insulating and heat transfer characteristics2. It is imperative to use additives that will not negatively affect the properties of INO to enhance its performance. While many additives studied thus far have been found to positively impact INO’s heat transfer, their usage often leads to a decrease in AC BDV3. Nanoadditives such as carbon nanotubes (CNTs), tungsten oxide (WO3), aluminum oxide (Al2O3), and graphene have been observed to have adverse effects on AC BDV4. To improve these properties, researchers have used the combination of this additive with metal oxide nanoparticles such as silicon dioxide (SiO2) + CNTs, Al2O3 + CNTs, and titanium dioxide (TiO2) + CNTs, which relatively improved the dielectric properties, but it was less than the pure dielectric oil of nanofluid5. However, some metal oxide nanoparticles with a high AC BDV, such as zinc oxide (ZnO) and TiO2, can be used to enhance AC BDV6.

Many nanoparticles are created using bottom-up processes from chemical precursors7,8. Co-precipitation9,10, sol–gel procedures11, microemulsion12,13, and solvothermal techniques14,15 are all included in this group. INOs are characterized primarily by their base oils, which are the key component. Oils contain far fewer additives, but additives are necessary for improving the qualities and compensating for the shortcomings of pure oils16,17,18. Therefore, in the present study, the co-precipitation process was used to synthesize ZnFe2O4 nanoparticles, and they were used for the first time in INO to enhance AC BDV and heat transfer properties.

Nanoparticles, on the other hand, have a high surface-to-volume ratio, which increases the surface energy of the particles and leads to the tendency for nanoparticles to aggregate19. Surface treatment of nanoparticles is employed to reduce surface energy and mitigate the potential for nanoparticle aggregation20,21,22. In the pursuit of enhancing stability and dispersion within nanofluids, a combination of techniques may be implemented23. Among these strategies, those incorporating surfactants such as gum arabic, oleic acid, sodium dodecyl sulfate, and Triton X100 have been widely employed24. Moreover, ultrasonics25,26, and alterations in pH levels27 have also shown efficacy toward the same end. With the goal of optimizing both stability and dispersion of nanoparticles in mind, this study aimed to improve their surface properties by treating them with oleic acid. This approach is known to be associated with enhanced nanofluid stability. Furthermore, ultrasonic waves were employed to further enhance the dispersion of the nanoparticles.

Several investigations have been carried out on the influence of nanoadditives on the thermal performance, thermophysical features, and AC BDV of INO, as described below. Siddique et al.6 explored how the presence of nanoparticles, specifically TiO2 and ZnO, influenced AC BDV in INO. According to the research, TiO2/INO and ZnO/INO nanofluids have a higher AC BDV than base oils, and with increasing TiO2 and ZnO additives, the maximum enhancements of AC BDV are 30% and 28%, respectively. Rajňák et al.28 conducted a study to examine how magnetic nano-additives influenced AC BDV in INO. Their findings from experiments indicate that AC BDV of the magnetic nanofluids studied did not improve over that of the basic INO. The effects of three nanoadditives, CNTs + TiO2, CNTs + SiO2, and CNTs + Al2O3, on the thermal properties and AC BDV of INO were investigated by Ghaffarkhah et al.5. By adding these nano additives, the INO’s thermal characteristics were enhanced. Nevertheless, AC BDV of all nanofluids was reduced with increasing nanoparticle concentration. Aberoumand et al.29 conducted a study on the influence of WO3 and silver nanoparticles on the thermal characteristics and AC BDV of INO. Their findings showed that the TCC improved by 41% when 4 wt% of the nanocomposite was added to INO. However, AC BDV of the nanofluids declined by 33.96%. Alizadeh et al.30 investigated how CNTs-OH nanoparticles affect heat transfer and AC BDV. Their conclusion showed that the addition of 0.01 wt% of the additive to INO increased the FCHT by 30.08%. However, AC BDV of the nanofluid decreased by 23.44% with the same amount of additive. Pourpasha et al.31 assessed the impact of TiO2-doped CNTs nanomaterials on the features of INO. In their study, it was discovered that the incorporation of 0.2 wt% of these additives into INO resulted in an 8.6% boost in the TCC and a substantial 23.08% increase in FCHT. Amiri et al.32 examined how hexylamine functionalized CNTs affected heat transfer and AC BDV of INO. With the addition of nanoparticles in the INO, the FCHTC increased by approximately 23%, and the TCC of the INO was boosted by about 10%. Nevertheless, AC BDV decreased by about 11% at 0.005 wt%. Marulasiddeshi et al.33 investigated the thermohydraulic behavior and the rate of thermodynamic irreversibility of a water-based Al2O3 and Al2O3 + CuO hybrid nanofluid flowing through a circular copper tube under a constant heat flux of 8625 W/m2. Their findings showed that the maximum increase in Nu was 51%, 59%, and 79.7% for 1 vol% nanofluid compared to water at temperatures of 30 °C, 45 °C, and 60 °C, respectively.

Wanatasanappan et al.34 investigated the effect of the Al2O3 + Fe2O3 nanocomposite on the viscosity of water-ethylene glycol. Their experimental data illustrated that the Al2O3 + Fe2O3 composition of 40/60 resulted in the highest viscosity value at all temperatures investigated, while the 60/40 composition recorded the lowest viscosity value. Besides, the increase in temperature of nanofluid shows a maximum viscosity reduction of 87.2% as the temperature is increased from 0 to 100 °C.

Previous studies conducted by researchers showed that most nanoparticles had a positive impact on the thermal properties of INO. However, these nanoparticles5,28,29,30,31,32,35,36 had a negative effect on the AC BDV of INO, except for TiO2 and ZnO6, which were studied in a few cases. Additionally, there were limited comprehensive studies that investigated various properties of INOs, such as viscosity, viscosity index, flash point, fire point, pour point, natural convection heat transfer, TCC, VHC, and AC BDV. It was crucial to select a nanoparticle that did not have a negative effect on the dielectric properties of INO, as it was one of the vital properties of INO. Furthermore, it was important to examine most properties of INO and its stability. To address these limitations, an extensive investigation was conducted to assess the influence of ZnFe2O4 additives on both the thermal properties, AC BDV characteristics, and stability of ZnFe2O4/INO nanofluids.

The novelty of this study lies in the comprehensive examination of the influence of the hydrophobicity of ZnFe2O4 nanoparticles on a range of crucial properties that are essential for the efficient and reliable functioning of INO in electrical equipment. By elucidating the relationship between the hydrophobicity of ZnFe2O4 nanoparticles and these properties, the study can contribute to the development of improved INO formulations with enhanced performance and tailored characteristics.

To improve AC BDV, thermal properties, and better dispersion of nanoparticles in INO, ZnFe2O4 nanoparticles were first synthesized. The ZnFe2O4 surface was treated with oleic acid to improve the stability. Then, the influence of the additive weight percentages on AC BDV and thermal properties such as TCC, VHC, flash point, pour point, VI, and kinematic viscosity were analyzed. In addition, the influence of the weight percentages of nanoparticles and various INPs of FCHTC was investigated. To the best of our knowledge, in most of the conducted studies, the most important and necessary parameters such as TCC, specific heat capacity, and viscosity were obtained by different theoretical equations. To reduce the error and obtain more accurate results, all parameters were determined by experimental tests, which is one of the new and strong sides of the present work. This study can establish the experimental basis for further research and optimization of the use of nanoparticles in INO to improve AC BDV.

Methods

Materials

The ZnFe2O4 nanoadditive was synthesized by a chemical technique using Zn(CH3CO2)2·2H2O (zinc acetate), Fe(NO3)3·9H2O (ferric nitrate), and NaOH (sodium hydroxide) purchased from Merck Co. The surface modification of the nanoparticles was carried out with oleic acid produced by Merck Co. INO is a product of APAR Co. whose properties are illustrated in Table 1. This oil is made from distillate (petroleum), including severely hydrotreated light naphthenic oil (75–85%) and severely hydrotreated light paraffinic oil (15–25%).

Table 1 Insulation oil features.
Full size table

Synthesis of nanoparticles and nanofluids

9 g of zinc acetate and 1 g of ferric nitrate were mixed in 150 ml of distilled water and 150 ml of ethanol for 1 h. After the addition of 10 g NaOH, the mixture was stirred for 2 h at room temperature37. Particles were separated by a centrifuge, and then washed with ethanol and distilled water several times. The nanoparticles were subsequently subjected to a drying process in an oven at 100 °C for a duration of 5 h, followed by calcination in a laboratory furnace at 400 °C for 3 h37. To increase the hydrophobicity of nanoparticles, the following method was used38. An ethanol solution containing 100 ml of ZnFe2O4 additives was mixed with 1 g of ZnFe2O4 additives at 40 °C using a magnetic stirrer. An oleic acid solution of 10 g and an ethanol solution of 50 ml were blended and added to the suspension, which was stirred for 24 h. Using centrifugation, the suspension was separated into nanoparticles. Several rinses of acetone and ethanol were conducted on the nanoparticles. Finally, the nanoparticles were dried for 9 h in a 70 °C oven.

The nanofluids were prepared in two steps. To prepare nanofluids, 0.05, 0.1, 0.15 and 0.2 wt% ZnFe2O4 nanoparticles were mixed with INO and dispersed for 15 min using an ultrasonic probe (Misonix Sonicator XL2020 Ultrasonic Liquid Processor, 550 Watts, 20 kHz, 1500 V rms). In the subsequent step, mechanical agitation was conducted at a speed of 1500 rpm for a duration of 2 h.

Characterization techniques

In this study, the composition of the additive was analyzed via X-ray diffraction (XRD). Fourier-transform infrared (FTIR) analysis serves the purpose of identifying a sample’s chemical composition and gaining insights into its molecular structure. Scanning electron microscopy (SEM) was employed to assess the nanoparticles’ shape, size, and morphology. Energy dispersive X-ray (EDX) analysis explores materials’ elemental makeup. Moreover, to analyze particle size distributions and stability of nanofluids, DLS and zeta potential were employed.

Thermophysical properties measurement methods

To evaluate the temperatures at the flash point and pour point, the open cup device PT-1015M and the digital instrument PT-1210 were used. The KD2 Pro thermal property analyzer (manufactured by Decagon) was used to investigate the thermal characteristics, including TCC and VHC. The nanofluid prepared earlier was placed in the testing chamber of the KD2 Pro instrument. The instrument applies a controlled heat pulse to the sample and measures the resulting temperature response over time. As the sample reaches thermal equilibrium, the instrument records the temperature response. The collected data includes the temperature change as a function of time. The collected data was then analyzed using the software provided with the KD2 Pro instrument. The software typically offers options to calculate VHC based on the recorded temperature response. The kinematic viscosity of the nanofluid was evaluated at different concentrations and temperatures of the nanofluid. ASTM D2270 and ASTM D445 test procedures were used to determine the kinematic viscosity and viscosity index (VI) of ZnFe2O4/INO nanofluids. Kinematic viscosities of both INO and nanofluids were determined using a Cannon Fenske Opaque viscometer. Each test was repeated at least twice to ensure that the findings were correct. The kinematic viscosity of nanofluids at various temperatures was measured using a precision temperature-controlled bath specifically designed to maintain a stable temperature. Additionally, a Cannon–Fenske Opaque viscometer was immersed in the temperature-controlled bath. To measure the kinematic viscosity at different temperatures, each desired temperature was set, and subsequent measurements were performed. Equation (1)39 was used to obtain the VI of the INO and nanofluids.

$$ {\text{VI}} = \frac{{{\text{K}} - {\text{U}}}}{{{\text{K}} - {\text{H}}}} \times 100 $$
(1)

The kinematic viscosity of the oil at 40 °C, denoted as “U”, is an essential factor in determining the oil’s viscosity index. Additionally, “K” represents the kinematic viscosity at 40 °C for oils with a viscosity index of 0, and “H” represents the kinematic viscosity at 40 °C for oils with a viscosity index of 100, which share the same kinematic viscosity as the oil being evaluated at 100 °C. For oils with a basic kinematic viscosity at 100 °C ranging from 2 to 70 cSt, the values of K and H can be obtained from ASTM D2270 tables. However, if the kinematic viscosity exceeds 70 cSt, K and H values are computed using the following equations39:

$$ K = 0.8353 Y^{2} + 14.6 Y - 216 $$
(2)
$$ H = 0.1684 Y^{2} + 11.85 - 97 $$
(3)

here Y represents the kinematic viscosity at 100 °C for the oil being evaluated for its viscosity index.

The IEC 60,156 test technique for nanofluids was used to determine the AC BDV. Two electrodes are placed in a chamber as part of the AC BDV measuring device. To measure AC BDV, the voltage is slowly increased. To determine the voltage at which the INO loses its dielectric property, the AC BDV experiment was repeated six times.

To study FCHT, a device was constructed that resembles the electrical system and other high-voltage equipment, with the difference that in electrical system and other high-voltage equipment the heat is generated by coils, but in the constructed device it is generated by a heater. Figure 1 shows a schematic representation of the setup used to determine the FCHTC of fluids. The experimental system consists of a liquid container, six thermocouples, six digital temperature displays, glass wool, a variac variable transformer (VVT), and a heating element.

Fig. 1
Fig. 1
Full size image

Schematic of the setup.

The system volume is equal to 0.2 × 0.1 × 0.22 = 0.0044 m3. Two thermocouples were installed on the central side walls, one thermocouple was installed on the central upper wall, and three thermocouples were placed on the upper wall with a distance of 2.5 cm between each. To prevent heat loss, the upper and lower walls were insulated with glass wool. The heat flux was obtained by heating an electric element with the power supply located in the center of the bottom wall. It should be noted that all measurements were made after the system reached a stable state. Steady state is when there is no noticeable change in wall temperatures.

The FCHT of ZnFe2O4/INO nanofluids was investigated at 0, 0.05, 0.1, 0.15, and 0.2 wt% of the nanoparticles and at different powers (50.5–124 W). After filling the reservoir, the VVT was utilized to control the electric current (I) and the input voltage (V) of the heater. All temperatures were recorded after they reached stable conditions and the corresponding calculations were performed. To calculate all the data, the hot and cold temperatures were recorded for each run of this test technique, which was carried out using nanofluids with four different mass fractions of additives and four various INPs. The INP to the heater was controlled by VVT. FCHTC was determined using Eq. (10) by recording the temperatures and INPs. Nu, Gr, and Ra were calculated by computing the FCHTC and Eqs. (4), (8), and (11). Each experiment was carried out three times.

Equations

In the FCHT, the Grashof number (Gr), the ratio between buoyancy force and viscous force, is determined through Eq. (4)36.

$$ {\text{Gr}} = \frac{{{\text{g}}\upbeta \left( {{\text{T}}_{{\text{h}}} - {\text{T}}_{{\text{c}}} } \right){\text{L}}^{3} }}{{\upupsilon ^{2} }} $$
(4)

The temperatures of the hot part (the center of the tank), the temperatures of the cold sides, and the gravitational acceleration are represented by Th, Tc, and g, respectively. The volume expansion coefficient (β) is 0.00075 (1/°C), the space between the heater and the wall is denoted by L, and the dynamic viscosity (µ) is determined as follows36:

$$\upmu = {\uprho \nu } $$
(5)

where \({\upnu }\) represents the kinematic viscosity and ρ is density of nanofluids. The calculation of Nu and Pr for FCHT is determined according to the following equations30:

$$ {\text{Nu}} = \frac{{{\text{hL}}}}{{\text{k}}} $$
(6)
$$ {\text{Pr}} = \frac{{{\text{C}}_{{\text{p}}} \cdot\upmu }}{{\text{k}}} $$
(7)

h is the FCHTC of fluids, k is the TCC of nanofluids, and Cp is the VHC. The Rayleigh number (Ra) is utilized in FCHT as a dimensionless quantity, calculated as follows40,41:

$$ {\text{Ra}} = {\text{Gr}} \cdot {\text{Pr}} = \frac{{{\text{g}}\uprho ^{2} {\text{C}}_{{\text{p}}}\upbeta \left( {{\text{T}}_{{\text{h}}} - {\text{T}}_{{\text{c}}} } \right){\text{L}}^{3} }}{{{\text{k}}\upmu }}. $$
(8)

The value of thermal element INP (Q) is calculated using Eq. (9)41:

$$ {\text{Q}} = {\text{V}} \cdot {\text{I}}{.} $$
(9)

The electric current is denoted by I, where V stands for the voltage applied to the thermal element. To determine the mean FCHTC under constant heat flux conditions, the following formula is employed30:

$$ {\overline{\text{h}}} = \frac{{\text{Q}}}{{{\text{A}}\left( {{\text{T}}_{{\text{h}}} - {\text{T}}_{{\text{c}}} } \right)}}. $$
(10)

The setup has a heat transfer area (A) of 0.129 m2. The following equation can be used to calculate the average Nu39:

$$ \overline{{{\text{Nu}}}} = \frac{{{\overline{\text{h}}} \cdot {\text{L}}}}{{\text{k}}}. $$
(11)

Considering the varied accuracy of each parameter assessed in this study, it is critical to examine the impact of measurement error on the outcomes. If R depends on the parameters x1, x2, …, xn, the effect of measurement error on the parameter xi can be calculated as illustrated in42:

$$ {\text{U}}_{{{\text{Ri}}}} = \frac{{{\text{x}}_{{\text{i}}} }}{{\text{R}}}\frac{{\partial {\text{R}}}}{{\partial {\text{x}}_{{\text{i}}} }}{\text{U}}_{{{\text{xi}}}} $$
(12)

where xi stands for a quantifiable factor, R for a value resulting from measurable quantities, Uxi is the error in measurement, and URi denotes the maximum possible error in the calculation of a quantity.

$$ {\text{U}}_{{\text{R}}} = \mp \left[ {\left( {{\text{U}}_{{{\text{R}}1}} } \right)^{2} + \left( {{\text{U}}_{{{\text{R}}2}} } \right)^{2} + \cdots + \left( {{\text{U}}_{{{\text{Rn}}}} } \right)^{2} } \right]^{1/2} $$
(13)

Table 2 demonstrates the amount of measurement error.

Table 2 Parameters error values.
Full size table

To calculate standard deviation (SD) for AC breakdown voltage tests, Eq. (14) is used. x is each individual data point, \(\overline{x}\) is the data mean, and N is the total of data points.

$$ {\text{SD}} = \sqrt {\frac{{\sum \left( {{\text{x}} - {\overline{\text{x}}}} \right)^{2} }}{{{\text{N}} - 1}}} $$
(14)

Experimental results

Nanofluid and nanoparticle features

Figure 2a–c demonstrate SEM/EDX analyses of ZnFe2O4 nanoparticles. According to Fig. 2a,b, the nanoparticles’ diameter ranges from 32.09 to 53.50 nm. According to the EDX analysis of ZnFe2O4 (Fig. 2c), the components of ZnFe2O4 consist of 47.42% oxygen, 48.9 Zn, and 3.68% Fe. Figure 2d displays distinct crystal structures observed in the ZnFe2O4 nanoparticles during the XRD. The X Pert High Score software (version: 1.0d, produced by: PANalytical B.V Almelo, the Netherlands, license number:50000022) was used to analyze all the peaks. The samples were determined to have a hexagonal wurtzite structure existing in a single phase with the space group P63mc. The dominant peaks observed were at the (100), (002), (101), (102), (110), (103), (200), (112), (201), (004), and (202) positions, and there were no unusual peaks that would be associated with secondary iron oxide phases37,39. Figure 2e shows the results of the FTIR analysis performed on the zinc ferrite nanoparticles. A broad peak ranging from 3000 to 3700 cm−1 corresponds to the stretching vibration of hydroxyl groups. Peaks at 2927 cm−1 and 2850 cm−1, indicating the presence of CH2 and CH3 functional groups, respectively39. Two peaks, at 1629 cm−1 and 1706 cm−1 correspond to the asymmetric stretching frequencies of C=C and C=O bonds, respectively39. A series of peaks ranging from 1226 to 912 cm−1, which were attributed to the vibration of the ZnO-Fe local bonding environment. Typically, bond frequencies around 1000 cm−1 provide insights into the interactions involving mineral elements. Notably, a prominent peak was observed at 435 cm−1, which corresponds to the Zn–O stretching band, thereby confirming the formation of the wurtzite crystal structure39. The FTIR spectrum confirmed the composite nature of the zinc oxide sample, with an iron-related peak at 563 cm−1 and Fe–O vibration peaks from 700 to 912 cm−1, indicating good crystallinity in nanocrystalline form37,43,44,45.

Fig. 2
Fig. 2
Full size image

Analysis of ZnFe2O4 (a) and (b) SEM, (c) EDX, (d) XRD, and (e) FTIR39.

Figure 3a depicts the DLS analysis results of ZnFe2O4/INO nanofluids for varying concentrations of ZnFe2O4. Figure 3a manifests the size distribution (SD) of ZnFe2O4 nanoparticles at 0.05 wt% with 121.5 nm (67.91%) and 144.5 nm (32.09%). The SD at 0.1 wt% contains particles with the SD of 144.5 nm (14.59%), 171.9 nm (64.5%), and 204.4 nm (20.91%). At 0.15 wt%, the DLS gives 171.9 nm (5.4%), 204.4 nm (65.39%), 243 nm (27.92%), and 289 (1.29%). Finally, at 0.2 wt% of the nanofluid, the statistics from SD indicate 204.4 nm (41.21%), 243 nm (35.12%), 289 nm (17.04%), 344 nm (5.36%), and 409 nm (1.27%). Zeta potential analyses of ZnFe2O4/INO nanofluids at different concentrations of ZnFe2O4 are demonstrated in Fig. 3b. The ZnFe2O4/INO nanofluids exhibited reduced stability with an increase in nanoparticle concentration. The trend of zeta potential with nanoparticle concentration can vary depending on the specific system and factors involved. At low nanoparticle concentrations, individual particles may dominate the system, leading to a higher zeta potential due to the electrostatic repulsion between dispersed particles. As the concentration increases, the likelihood of particle interactions and collisions also increases. This can lead to the formation of particle aggregates or clusters, reducing the overall zeta potential due to decreased electrostatic repulsion between the aggregated particles. However, all the nanofluid samples reflected good stability with a zeta potential of less than − 39.2 mV.

Fig. 3
Fig. 3
Full size image

Size distribution and stability of ZnFe2O4 nanofluid in different concentrations: (a) DLS and (b) Zeta potential.

The images in Fig. 4 depict samples with varying mass fractions of ZnFe2O4 prepared for 1 h (Fig. 4a–d), 7 days (Fig. 4e–h), and 14 days (Fig. 4i–l). These samples proved to be stable for a period of 7 days. After 7 days, precipitation occurred in the nanofluids with 0.15 and 0.2 wt%. The sedimentation process of nanofluids with 0.15 and 0.2 wt% ZnFe2O4 over a period of 14 days is illustrated in Fig. 4k,l, respectively. It was noted that nanofluids containing lower concentrations of 0.05 wt% and 0.1 wt% displayed satisfactory stability for a duration of up to 14 days, with only minimal precipitation observed (as depicted in Fig. 4i,j). The findings suggest that treating ZnFe2O4 with oleic acid can enhance its surface area by minimizing the nanoparticles’ surface energy and agglomeration.

Fig. 4
Fig. 4
Full size image

Visual images illustrating sedimentation quality.

Breakdown voltage and thermophysical properties

AC BDV of INO is defined as the lowest voltage at which a portion of the dielectric fluid conducts electric current. A crucial feature of INO is the AC BDV, which necessitates the use of nanofluids with higher values in electrical system and other high-voltage equipment. Figure 5 illustrates the variation in mean AC BDV of six test repetitions for ZnFe2O4/INO nanofluids under identical conditions. As shown in Table 3 and Fig. 5, adding ZnFe2O4 to INO enhanced the AC BDV of nanofluids. According to the IEC 60,156 standard, the AC BDV should be in the range of 30 to 70 kV. The mean AC BDV values of the nanofluids, which contained 0 to 0.2 wt% were found to vary from 65.15 to 74.02 kV when analyzed. The addition of ZnFe2O4 to INO increased the AC BDV value of the nanofluid due to its low electrical conductivity. This resulted in acceptable AC BDV data that complied with the IEC 60,156 standard. The nanofluid with 0.1 wt% nanoparticles showed the maximum improvement in the mean AC BDV, which was 76.42 kV, an increase of 17.3%. According to Table 3, the maximum standard deviation was 5.08, which was deemed acceptable. The dielectric properties of nanofluids are determined by the contact surface between nanoparticles and oil. According to the concept of the electrical double-layer (EDL), nanoparticles in the INO carry free charges. These charged nanoparticles attract counter-ions while repelling co-ions46. The counter-ions form a stationary layer around the nanoparticles known as the compact layer, which is rigid. Beyond this compact layer, a diffuse layer is formed, extending towards the electrically neutral base INO. The mobile ions in the diffuse layer are easily influenced by electrostatic forces. The EDL reduces the energy of fast electrons in an electrical field and traps them. Due to the uniform distribution of nanoparticles in the oil, interfacial volumes play a significant role in the system, resulting in the creation of numerous traps. Higher concentrations of nanoparticles in the base oil increase the interfacial volume, leading to a greater number of traps. This improvement in the nanofluid’s properties enhances the AC breakdown strength, dissipation factor, and volume resistivity of the oil46. According to Fig. 5, with an increasing concentration of nanoparticles from 0.1 to 0.2 wt%, the mean AC BDV of the nanofluid decreased by 3.14%. This trend may depend on flowability issues, interfacial interactions, reduced nanoparticle dispersion, and agglomeration of nanoparticles. This is because with the addition of nanoparticles exceeding 0.1 wt%, the zeta potential value decreased, and these issues may occur, causing a reduction in AC BDV. Compared to various studies on AC BDV of INO, the results obtained in this study are the best achievement in this field. For instance, according to Table 4, the use of most nanoparticles, especially carbon-based nanoparticles, leads to a reduction in the AC BDV of INO because these nanoparticles have high electrical conductivity. Among the nanoparticles, TiO2 and ZnO nanoparticles have a high dielectric coefficient than pure oil, increasing the voltage coefficient of all types of oil. According to Siddique et al.6, the suspension of TiO2 and ZnO nanoparticles in INO, specifically ester oil, has resulted in an enhanced AC BDV for the oil.

Fig. 5
Fig. 5
Full size image

Mean AC breakdown voltage of ZnFe2O4/INO nanofluids.

Table 3 Average AC breakdown voltage of ZnFe2O4/INO nanofluids.
Full size table
Table 4 Previous studies on the effect of various nanoparticles on the BDV of insulation oil.
Full size table

Figure 6 illustrates thermophysical attributes and viscosity features of the ZnFe2O4/INO nanofluids, such as TCC, VHC, flash point, fire point, pour point, kinematic viscosity, and viscosity index. The addition of ZnFe2O4 to INO increases TCC and VHC of the nano lubricant because the TCC and VHC of the ZnFe2O4 are greater than those of INO. The highest enhancement in TCC of the nanofluid at a concentration of 0.2 wt% corresponded to 0.358 (W/m K) by a 6.23% increase, as shown in Fig. 6a. The heat transfer efficiency of nanofluids is strongly influenced by TCC. As a result, nanofluids act as smart fluids that can dissipate more heat at higher temperatures. When heated or cooled, nanofluids demonstrate a faster rate of heat transmission. Based on the observed enhancements in thermal performance, the following controlling mechanisms are proposed: (1) the transportation of phonons through nanoparticles; (2) the random movement of nanoparticles known as Brownian motion; and (3) interactions between particles influenced by the electric double layer47. Phonon transport can be understood as the transmission of lattice vibrations. The efficiency of phonons is determined by their mean free path (MFP) relative to the thickness of the nanoparticles. For efficient heat transport through ballistic phonons, the phonon MFP should exceed the thickness of the nanoparticles47. This condition ensures improved heat transport47. Various factors such as the TCC of the base fluid and solid particles, particle concentration, size, shape, and thickness, as well as temperature play a crucial role in determining the TCC of nanofluids. Moreover, the mechanics of the particles in the base fluid can lead to Brownian motion, thermophoresis and diffusion, which can also influence the TCC of nanofluids32,35. Figure 6b demonstrates that the VHC of a nanofluid containing 0.2 wt% nanoadditives improved by 1.874 (MJ/m3K), representing a 26.53% increase. This work’s noteworthy finding is the substantial enhancement in VHC of the nanofluid compared to the base fluid, with a slight increase in TCC. Additionally, Fig. 6c illustrates that increasing the nanoparticle concentration to 0.2 wt% raised the flash point and fire point to 7 °C and 6 °C, respectively. In general, the increased resistance of the oil to burning can be attributed to an increase in TCC due to the existence of nanoparticles that retard the evaporation of ignition vapors. Figure 6d indicates the pour point of ZnFe2O4/INO nanofluids with various mass fractions of ZnFe2O4 nanoparticles. The addition of ZnFe2O4 to INO lowers the pour point of the nanofluids because the TCC and VHC of the ZnFe2O4 are larger than the INO. At a concentration of 0.15 wt% nanoparticles, the pour point of the nanofluid was enhanced by 6.25%, resulting in a maximum decrease of − 42.5 °C. The kinematic viscosity of ZnFe2O4/INO nanofluids with varying mass fractions of ZnFe2O4 nanoparticles at temperatures of 40 °C, 60 °C, 80 °C, and 100 °C is shown in Fig. 6e. The fluids’ viscosity of the fluids decreased by 76% on average when the temperature increased from 40 to 100 °C. This phenomenon occurs because as the temperature increases, the Brownian motion of nanoparticles in the oil intensifies. Consequently, the intermolecular interactions between the oil and the nanoparticle surfaces decrease due to the heightened random motion of the nanoparticles30. As temperature and molecular energy rise, the intermolecular forces within the liquid diminish, resulting in an expansion of the distance between molecules. Raising the temperature decreases the viscosity of the nanofluids. At a temperature of 100 °C, the addition of 0.2 wt% ZnFe2O4 to INO resulted in a 3.47% increase in the kinematic viscosity of the nanofluid. To investigate the VI of fluids with different mass fractions of additives, the kinematic viscosity values at 40 °C and 100 °C were analyzed, and the findings are displayed in Fig. 6f. The amount of VI is directly proportional to the viscosity at both 40 °C and 100 °C. A higher VI indicates that as temperature increases there’s no much increase in the oil’s viscosity hence maintaining a constant lubrication property. Conversely, low VIs indicate high change in oil viscosity with rise or drop in temperatures thus resulting in performance problems sometimes. The nanofluid with a 0.1 wt% additive had a maximum VI increase of 73.61, which corresponds to an increase of 6.41%. The interactions between particles or molecules in a suspension can significantly impact the VI. With increased concentrations there can be more interactions among them resulting into variations in the flow characteristics of this system which eventually affects its VI.

Fig. 6
Fig. 6
Full size image

Thermophysical attributes and viscosity features of the ZnFe2O4/INO nanofluids, such as (a) thermal conductivity coefficient, (b) volumetric heat capacity, (c) flash point and fire point, (d) pour point, (e) kinematic viscosity, and (f) viscosity index.

Heat transfer evaluation

Figure 7a depicts the FCHTC of fluids with different concentrations (0.05–0.2 wt%) of ZnFe2O4 at different INPs of 50.5, 75.6, 100, and 124 (W). The FCHTC for INO and nanofluids increased with increasing INP, and so did the heat flux supply. As the FCHTC is directly related to power, and the increase in heat flux was greater than the increase in temperature difference, consequently, the ratio of these two parameters increased. The FCHTC improved with the addition of ZnFe2O4 to oil. The highest FCHTC is related to the mass fraction of 0.1 wt% of ZnFe2O4 and an INP of 124 (W) of ZnFe2O4, which equals 89.85 (W/m2 K). The maximum increment in FCHTC occurred with an increase in the mass fraction of ZnFe2O4 to 0.1 wt% at an INP of 50.5 (W), and the FCHTC of the nanofluid improved by 14.15% compared to INO. The value of FCHTC increases as the free heat transfer increases; thus, this parameter is critical in determining the rate of heat transfer. According to the results, FCHTC is a temperature-dependent property. In Fig. 7b, the impact of ZnFe2O4 on INO’s Nu under various inlet powers is exhibited. The addition of ZnFe2O4 to INO resulted in an improvement of Nu. Nu is determined by the values of h and k, where h is significantly higher than k. The results indicate that the addition of 0.1 wt% ZnFe2O4 to INO at an INP of 50.5 (W) led to an 11.17% increase in Nu. At a mass fraction of 0.1 wt% and an INP of 50.5 W, the h/k ratio was the highest. Figure 7c shows that Gr varied with input power at different ZnFe2O4 mass fractions. For all ZnFe2O4 mass fractions, Gr improved with increasing INP due to the temperature difference between zones. As the mass fraction of ZnFe2O4 increases due to enhanced heat transfer, the concentration of ZnFe2O4 also increases, leading to a decrease in the temperature difference between hot and cold regions and thus a lower Gr value. At an INP of 50.5 (W), the Gr number of INO is 0.154 × 106, but when 0.15% ZnFe2O4 is added to INO, the Gr value decreases by 56.34% to 0.0985 × 106. Figure 7d demonstrates the effect of ZnFe2O4 on the Ra value of INO at different input powers. As expected, increasing the INP to the heater resulted in higher Ra. This relationship can be explained by Eq. (8), which states that Ra is directly proportional to the temperature difference. Addition of 0.05 wt% and 0.1 wt% ZnFe2O4 to INO at various INPs between 75.6 (W) and 124 (W) increased the temperature difference and thus Ra. However, at concentrations of 0.15 wt% and 0.2 wt%, a downward trend in Ra was observed at INPs between 75.6 (W) and 124 (W). At an input power of 50.5 (W), increasing the mass fraction of ZnFe2O4 caused a significant decrease in Ra. In particular, the addition of 0.15 wt% ZnFe2O4 to INO led to a 14.11% decrease in the Ra of the nanofluids compared to INO alone. Table 5 indicates that the suspension of carbon, metal, and metal oxide-based particles in INO leads to an improvement in convective heat transfer. Among the nanoparticles, carbon nanotubes had a greater impact on the thermal performance of the oil, even at low weight percentages, owing to their high TCC.

Fig. 7
Fig. 7
Full size image

Variations in (a) free convection heat transfer coefficient, (b) Nusselt number, (c) Grashof number, and (d) Rayleigh number of ZnFe2O4/insulation oil at various input powers.

Table 5 Previous studies on the effect of various nanoparticles on the heat transfer of various insulation oils.
Full size table

Conclusion

The study commenced with the synthesis of hydrophobic ZnFe2O4 nano-additives, followed by their characterization through XRD, FTIR, SEM, and EDX analyses. The nanofluid stability was monitored using DLS, zeta potential, and visual observations. Subsequently, the impact of ZnFe2O4/INO nanofluids with varying mass fractions and different INPs on FCHTC, Nu, Gr, and Ra was investigated. Key findings from the research include:

  1. 1.

    The diameter of ZnFe2O4 nanoparticles was determined by SEM analysis and varied between 32.09 nm and 53.50 nm.

  2. 2.

    Nanofluids with low concentrations remained stable for 14 days, with minimal sedimentation.

  3. 3.

    The nanofluid exhibited fine stability, as reflected by a zeta potential below − 39.2 mV.

  4. 4.

    The maximum increase in TCC of the nanofluid at 0.2 wt% ZnFe2O4 was 0.358 W/m K, which was equal to 6.23%.

  5. 5.

    The substantial increase in Nu for fluids is associated with an INP of 50.5 W and 0.1 wt% ZnFe2O4 at an improvement rate of approximately 11.17%.

  6. 6.

    The maximum increase in FCHTC occurred with increasing the mass fraction of nanoparticles to 0.1 wt% at an INP of 50.5 (W), and the FCHTC the of nanofluid improved by 14.15% compared to INO.

  7. 7.

    The AC BDV experienced a 17.3% increase, equivalent to 11.27 kV increase in breakdown voltage, upon the addition of 0.1 wt% of nanoparticles.

This research underscores the promising results of ZnFe2O4 nanoparticles in improving the thermal and dielectric properties of INO, particularly through the use of nanofluids. The incorporation of these nanoparticles holds significant implications for the industry by enhancing transformer efficiency and reliability. Notable advancements in TCC and AC BDV offer potential solutions for overheating concerns and bolster the insulation capabilities of transformer systems, leading to increased energy efficiency, higher power transmission capacity, and prolonged equipment lifespan.