Salehi et al., tested the system at air velocities 2–6 m/s, coolant flows 3–8 L/min, and coolant inlet 90°C. Combined experiments and CFD to demonstrate that installing passive flow distributors in the radiator plenum reduced flow maldistribution by 18% and boosted effectiveness from 0.63 to 0.71, corresponding to a 12% increase in heat rejection [1]. Feleke et al., varied fin pitch (1.2–1.8 mm), louver angle (25–35°) and edge radius ratio (r/p = 0.01–0.05) at air-side Re no 400–1200 to make cores more compact [3]. The experimental comparison tested flat-tube, circular-tube and elliptical-tube using the ε–NTU method. revealed elliptical tubes provide 14% higher heat transfer coefficients and 11% higher effectiveness than flat tubes [10] many authors further validated elliptical tubes as the most balanced option, The authors present empirical correlations per tube shape and recommend elliptical tubes for compact radiator cores where air-side resistance dominates[11–13].

Basir et al. demonstrated that a map-controlled thermostat (MCT) reduced warm-up time by 18%, and achieved 3–4% lower fuel consumption and CO₂ emissions compared to wax thermostats. Confirming that thermostat control strategy significantly affects radiator loading, warm-up, and emissions [2]. Becker et al. showed that active grille shutters combined with staged fan control shortened warm-up time by 22% and prevented frost accumulation in winter conditions, highlighting how radiator control logic directly affects energy efficiency [14]. Recent diagnostic and prediction studies emphasize high-resolution mapping and data-driven modeling. Gan et. al. developed an infrared thermography plus inverse heat conduction method capable of reconstructing core temperature fields with less than 2°C accuracy, detecting hot spots with 12°C gradients enabling real-time monitoring of maldistribution. Machine learning has also entered radiator design [4].

Other works investigate wet operating conditions and non-traditional enhancement structures, confirmed that inlet straighteners and tailored louvers could reduce multi-row maldistribution from 25% to 8%, recovering 10–14% lost capacity [15, 16]. Dika & Dika reported that wavy fins improved Nusselt numbers by 8–12% .while other s found that metal foam inserts deliver 20–35% volumetric heat transfer improvement but often double or triple ΔP, restricting their use to highly space-constrained cores[18, 19]. Habibian & Abolmaali highlighted coolant–fin interaction, showing that wavy fins increased Nusselt number by 9–13% over straight fins but raising EG content from 20% to 50% reduced heat rejection by up to 6%. (due to lower cp) while improving freeze protection[20].

Table-1 shows different types of nano-particles used in a radiator coolant. A nanofluid is a mixture of a base fluid (like water, ethylene glycol) and nanoscale particles (1–100 nanometers in size) that are suspended uniformly in it. Thermal conductivity Increases with nanoparticle addition gives better cooling and higher convective performance. Viscosity Slightly increases and can affect pump power. Specific heat slightly decreases .This requires surfactants or special mixing to avoid settling. These nanoparticles are typically metals, metal oxides, or carbon-based materials like:

Table-1: Types of Nano-particles

Over the past decade, extensive research has explored the use of nanofluids to enhance automotive radiator cooling performance, primarily focusing on the effects of particle type, concentration, base fluid, flow conditions, and fin/tube geometry. Ramesh Kumar et al., used 3D CFD simulations to analyze a radiator with composite fins and nanoparticle-based coolant at a velocity of 1.61 m/s. Incorporating nanoparticles led to a 12% increase in heat transfer [21]. Sivakumar et al., experimentally investigated hybrid Al₂O₃–Cu/water nanofluids in commercial car radiators at 1.0 vol%, the heat transfer coefficient rose 22% [22]. Sarafraz and Safaei tested a flat-tube automotive radiator operating with TiO₂/water nanofluids at %, the Nusselt number improved 19% and effectiveness increased 14%[23],. Similarly, Al-Rashed and Chamkha simulated graphene–water nanofluids showing 20–27% higher heat transfer. The pumping power requirement increased slightly due to higher viscosity. The study emphasizes graphene’s very high thermal conductivity, showing its superior potential for next-generation radiator coolants compared to oxide-based nanofluids [24]. Sajid et al., tested SiO₂/water nanofluids, finding 15–18% improvement in heat transfer with only 3–4% higher ΔP. SiO₂ nanofluid as a low-cost, effective solution for enhancing radiator cooling performance [25].

Hybrid nanofluids were prepared by dispersing SiO₂ and MWCNT nanoparticles (0.1 vol% total) in different ratios (80:20, 50:50, 20:80) in distilled water. The best composition (20:80 SiO₂:MWCNT) achieved 15.6% Nusselt number increase compared to the base fluid. The results confirmed that mixing oxide particles with MWCNT in proper ratios can significantly enhance radiator air-side heat transfer, even at very low nanoparticle concentrations [26]. Mamat et al.,tested hybrid CuO–TiO₂ (50:50) in 60:40 water–EG ,the overall heat transfer coefficient improved 83% compared to base fluid, demonstrating strong enhancement at optimized loading. Microchannel designs have further amplified nanofluid benefits [27]. The nanofluid increased thermal conductivity by 40% and achieved a heat transfer coefficient of 5366 W/m²K, a 116% improvement over water. Results show that CuO nanofluid microchannels can nearly double cooling capacity, enabling compact radiator designs with superior heat rejection[28, 29].In some Experiments used a 60:40 water–EG base fluid with Al₂SiO₅ nanoparticles at 0.1–0.4 vol% and coolant Reynolds numbers from 500–1500. Also revealed Nusselt number improvements of 174% and 162%, respectively, with convective HTC reaching 37,021 W/m²K, highlighting nanofluid suitability for compact radiator designs[30, 31].Experimental studies also demonstrated that performance scales sharply with concentration. Adhikari et al., reported Al₂O₃–EG/water nanofluids improved HTC by 33.6%, 65.4%, and 83.5% at 0.2%, 0.5%, and 1.0% volume fraction, respectively[32, 33, 34]. Subhedar et al., showed that 0.2–0.6 vol% Al₂O₃–EG/water nanofluids allow frontal area reduction up to 65% without loss of cooling, supporting radiator downsizing for compact vehicles [35]. at 60–80°C produced 28–29% higher HTC, 23–24% higher Nusselt number, and 22.5% higher effectiveness than water, though ΔP increased 24%, highlighting the trade-off between heat transfer enhancement and pumping requirements[36].

In summary, these studies converge on three main findings. First, geometry refinements, tube cross-section, louver edges, and fin design can consistently yield 8–15% higher heat transfer with minimal ΔP penalty. Second, system-level control strategies such as map-controlled thermostats and active shutters improve warm-up, reduce overcooling, and contribute to 3–4% fuel savings. At the same time, nanofluid technology offers substantial potential to enhance automotive radiator efficiency.

Once the engine reaches its normal operating temperature, usually around 82C, the thermostat opens gradually and allows coolant partially to flow through the radiator circuit. When it reaches to 90–95°C, thermostat opens fully now the water pump pushes all the hot coolant from the engine into the radiator. The temperature sensor constantly monitors the coolant and signals the ECU to control the fan as needed, ensuring stable temperature. The radiator cap keeps the system pressurized, raising the boiling point of the coolant, and directs excess coolant (1.1bar) into the reservoir when expansion occurs. Once the engine cools down, this coolant is drawn back into the system, maintaining a closed and balanced loop. This process keeps the engine within a safe operating range, typically between 90 and 105°C, preventing overheating and ensuring efficient performance.

Fig-1: Cooling and Heating circuit