Nanofluids for Sustainable Heat Transfer Enhancement: Beyond Thermal Conductivity

Abstract

Nanofluids have long been explored for enhancing heat transfer, with early studies focusing primarily on improved thermal conductivity. However, in spray and droplet cooling applications, recent research indicates that conductivity alone cannot fully account for the observed performance gains. Additional mechanisms, such as Brownian-motion-induced convection, dynamic wetting, and nanoparticle-driven surface modification, significantly affect droplet impact dynamics, spreading behavior, boiling transitions, and transient heat transfer during impact and evaporation. This review critically synthesizes these effects, emphasizing how nanofluids interact with complex flow fields, steep thermal gradients, and heated substrates. It also examines emerging strategies for optimizing nanofluid design, including hybrid suspensions and phase-change-enhanced formulations. These developments open new avenues for high-efficiency cooling in electronics, renewable energy systems, and industrial spray processes. By moving beyond thermal conductivity as the sole performance metric, this review promotes a multi-scale, interdisciplinary framework for advancing nanofluid-based thermal technologies that align with sustainability, energy efficiency, and cost effectiveness.

1. Introduction

Enhancing heat transfer by adding small particles to coolant liquids has garnered considerable interest from both researchers and industry owing to its straightforward application and substantial potential without increasing operational costs. Engineered nanometer-sized particles, typically ranging from 1 to 100 nm, dispersed in a base fluid to improve heat transfer are termed nanofluids [1]. Several interrelated factors, including nanoparticle size, dispersion stability, surface chemistry, and temperature, govern the thermal performance of nanofluids. Smaller, well-dispersed particles with tailored surface treatments can remarkably enhance heat transfer, although challenges remain in independently controlling these properties and fully understanding their temperature-dependent behavior [2]. Early studies on nanofluids have demonstrated that their superior thermal properties, such as increased thermal conductivity, significantly improve their heat transfer performance [3]. Indeed, an abundance of papers attribute heat transfer enhancement primarily to increased thermal conductivity, and research continues to focus on the thermal properties and characteristics of nanofluids [4]. Nonetheless, practical challenges, inconsistent experimental results, and an incomplete understanding of the underlying physics have impeded the widespread application of nanofluids in real-world systems [5].

Depending on the experimental conditions and test setup, it is seen that nanofluids can enhance, have no effect on, or even deteriorate the heat transfer coefficient (HTC). This ambiguity stems from an incomplete understanding of the complex physical mechanisms introduced by nanoparticles and an overemphasis on increased thermal conductivity as the primary influencing factor. For instance, Bellerova et al. [6,7] reported a significant decrease, approximately 20% and 45%, in the HTC during spray and jet cooling, respectively, when the nanoparticle volume fraction increased from 0 to around 16 vol.% (40 wt.%). This reduction was attributed to nanoparticle sedimentation on the heated surface, which can adversely affect the heat transfer performance.

Given the diverse and sometimes contradictory findings in the literature, a comprehensive review is essential to clarify the mechanisms governing nanofluid behavior and to guide future research. This review focuses on the physical mechanisms underlying alterations in heat transfer induced by nanofluids, rather than presenting application-specific case studies. While the primary focus is on spray cooling, insights from other heat transfer applications are included when they provide transferable physical explanations relevant to spray systems. By synthesizing key findings and identifying knowledge gaps, this work aims to support the development of reliable, efficient, and scalable nanofluid systems for advanced heat transfer applications. Ultimately, this review seeks to guide researchers in advancing nanofluid-based spray cooling technologies toward practical and sustainable implementation.

Recent studies and reviews have addressed nanofluids from distinct perspectives. Buschmann et al. [8] emphasized the correction of thermophysical property measurements and the associated uncertainties in reported data. Motta and Sergis [9] applied meta-regression techniques to synthesizing results on pool boiling heat transfer, while Issa [10] systematically reviewed spray cooling experiments that excluded deposition effects. Pereira et al. [11] examined the cost and fouling implications of noble metal nanoparticles. In contrast, the present review integrates these strands into a broader framework that emphasizes droplet and spray cooling applications of nanofluids, with particular attention to transient wetting, the deposition dynamics, and the challenges in designing a self-maintaining system for long-term operation. This scope provides a unified perspective that has not been addressed in previous reviews.

A structured literature search was conducted in Scopus and Google Scholar using combinations of keywords related to nanofluids, spray/droplet cooling, and heat transfer enhancement. Peer-reviewed experimental studies that met the defined inclusion criteria were retained. For transparency and reproducibility, the full search strategy, inclusion/exclusion rules, and PRISMA flow diagram [12] are provided in the Supplementary Materials.

2. Models to Characterize the Thermophysical Properties of Nanofluids

The thermophysical behavior of nanofluids is often estimated using classical models that incorporate the particle concentration and fluid properties when direct measurements are unavailable [13,14]. One commonly used correlation for effective thermal conductivity is a Maxwell-based model (Equation (1)), which accounts for the nanoparticle volume fraction $\varphi$:

$$k_{\mathrm{nf}}=k_f\left(1+2.5\varphi +{\displaystyle \frac{0.75{\varphi }^2}{1-\varphi }}\right)$$

(1)

Einstein’s model (Equation (2)) describes the increase in viscosity due to suspended particles, assuming a dilute suspension (valid only below 2 vol.%) and negligible interactions between particles:

$${\eta }_{\mathrm{eff}}={\eta }_f(1+2.5\varphi )$$

(2)

The stability of nanofluids can be related to electrophoretic mobility $\mu$, which is connected to the zeta potential $\zeta$ through the Smoluchowski equation (Equation (3)). This relationship includes the fluid’s permittivity $\epsilon$ and viscosity:

$$\mu =\epsilon {\displaystyle \frac{\zeta \eta }{{\eta }_f}}$$

(3)

Stokes’ law (Equation (4)) describes nanoparticle sedimentation, which estimates the settling velocity $v_s$ of particles under gravity (g), depending on the density difference (${\rho }_p-{\rho }_f$), particle size (R), and fluid viscosity (${\eta }_f$):

$$v_s={\displaystyle \frac{2}{9}}·{\displaystyle \frac{({\rho }_p-{\rho }_f)g{R}^2}{{\eta }_f}}$$

(4)

where

• $k_{\mathrm{nf}}$ is the effective thermal conductivity of the nanofluid;
• $k_f$ is the thermal conductivity of the base fluid;
• $\varphi$ is the volume fraction of nanoparticles;
• ${\eta }_{\mathrm{eff}}$ is the effective viscosity of the nanofluid;
• ${\eta }_f$ is the viscosity of the base fluid;
• $\mu$ is the electrophoretic mobility;
• $\zeta$ is the zeta potential;
• $\epsilon$ is the permittivity of the fluid;
• $\eta$ is the viscosity of the nanofluid;
• $v_s$ is the settling velocity;
• ${\rho }_p$ is the density of the particles,
• ${\rho }_f$ is the density of the fluid;
• g is the acceleration due to gravity;
• R is the particle radius.

These models provide a basic framework for estimating nanofluid properties related to heat transfer performance, stability, and dispersion. While they offer useful approximations, their accuracy is limited by factors such as particle shape, interparticle interactions, and experimental conditions, often necessitating more advanced or empirically corrected models [14].

3. The Effect of Nanoparticles on Thermal Conductivity

The concept of thermal conductivity enhancement using nanofluids was first introduced by Choi in 1995 [1], sparking extensive research into their applications in heat transfer. Building on this concept, Bansal and Pyrtle [15] explored the benefits of Al₂O₃ nanofluids in spray cooling systems. Since then, numerous studies have investigated the influence of nanoparticles on droplet behavior and spray system performance [13].

While Equation (1) provides a basic representation, a wide range of models have been proposed to capture the thermal conductivity of nanofluids better, accounting for variables such as particle shape, size distribution, and high volume fractions [16,17,18,19]. The foundational model by Maxwell [20] (Equation (5)) assumes a dilute suspension of spherical particles and neglects interparticle interactions:

$$k_{\mathrm{nf}}=k_f{\displaystyle \frac{k_p+2k_f+2\varphi (k_p-k_f)}{k_p+2k_f-\varphi (k_p-k_f)}}$$

(5)

where $k_p$ is the thermal conductivity of the solid phase.

To extend Maxwell’s model to non-spherical particles, Hamilton and Crosser [21] developed a formulation (Equation (6)) that incorporated the shape factor n, with $n=3$ for spheres and $n=6$ for cylinders:

$${\displaystyle \frac{k_{\mathrm{nf}}}{k_f}}={\displaystyle \frac{\frac{k_p}{k_f}+(n-1)-(n-1)(1-\frac{k_p}{k_f})\varphi }{\frac{k_p}{k_f}+(n-1)+(1-\frac{k_p}{k_f})\varphi }}$$

(6)

Later refinements incorporated additional parameters such as particle size effects and non-local interactions. For instance, Shaker et al. [19] proposed a model (Equation (7)) with a non-locality parameter h and the particle radius R:

$${\displaystyle \frac{k_{\mathrm{nf}}}{k_f}}={\displaystyle \frac{1+2\mathfrak{f}\varphi {\mathfrak{h}}^{-3}}{1-\mathfrak{f}\varphi {\mathfrak{h}}^{-3}}}$$

(7)

where $\mathfrak{h}=R/h$ and $\mathfrak{f}={\displaystyle \frac{k_p-k_f}{3({\mathfrak{C}}_1{k}_p-{\mathfrak{C}}_2{k}_f)}}$, with ${\mathfrak{C}}_1$ and ${\mathfrak{C}}_2$ being empirical coefficients dependent on R and h.

Thermal conductivity enhancement has been a key motivation behind the development of nanofluids, leading to growing interest in hybrid nanofluids, suspensions containing more than one type of nanoparticle. Babar et al. [22] reviewed their stability and thermophysical properties, concluding that hybrid nanofluids generally offer superior thermal conductivity over that of mono-nanoparticle systems. More recently, Kanthimathi et al. [23] further highlighted several critical parameters affecting thermal conductivity in these systems, including the nanoparticle volume fraction, particle shape and size, material type, temperature, surfactant use, pH, base fluid, preparation method, and colloidal stability. By optimizing these factors, hybrid nanofluids offer a flexible platform for tailoring thermal conductivity enhancements to meet specific application needs. Complementary reviews add further insights: Khoswan et al. [24] introduced a qualitative model specifically explaining how carbon nanotubes (CNTs) enhance thermal conductivity through factors such as the aspect ratio, alignment, surfactant effects, and functionalization. Meanwhile, Yasmin et al [25]. offered a state-of-the-art review of metal oxide nanofluids, highlighting the contributions from nanoparticle size, shape, surfactants, base fluid properties, and even electric or magnetic field alignment as powerful enhancers of thermal conductivity. These studies underscore that achieving precise predictions requires accounting for both nanoscale effects and external influencing factors beyond the predictions of the classic models.

A growing body of research has also explored how nanoparticle morphology, aggregation, ultrasonication, and chemical additives alter thermal conductivity [26]. As summarized by Buschmann et al. [8], thermal conductivity enhancement remains a reliable contributor to an improved heat transfer performance only when nanofluids maintain Newtonian behavior and the Reynolds and Prandtl numbers are held constant. Under these conditions, nanofluids can be modeled as homogeneous fluids, allowing for the use of standard single-phase correlations in practical designs.

Beyond the ambiguity in the heat transfer performance due to different operational conditions, several physical parameters critically influence nanofluid behavior. Proper ultrasonication can enhance the dispersity and stability of nanofluids, increasing their potential for safe use across various applications. To achieve this, the sonication settings must be carefully optimized to promote effective cavitation and ensure a uniform nanoparticle distribution [27]. Improved colloidal dispersion is associated with a lower viscosity [28], and extended ultrasonication periods have been shown to positively influence thermal conductivity. However, the enhancement typically does not exceed 3% [29], which is advantageous for practical applications. Excessive ultrasonication may lead to re-agglomeration of the particles, whereas an inadequate sonication time may fail to break apart existing agglomerates [27]. Additionally, optimizing nanoparticle aggregates, potentially forming long conduction pathways, can enhance the thermal conductivity further [30,31].

Surfactants play a dual role: by enhancing particle dispersion, they improve thermal conductivity [32,33], and by reducing surface tension, they influence the spray characteristics. A lower surface tension promotes the formation of finer droplets upon injection [34], accelerating evaporation and enhancing the transition boiling heat transfer. Moreover, it improves the surface wettability by lowering the solid–liquid contact angle [35], further supporting efficient thermal management.

Sun et al. [36] investigated the thermal conductivity of CuO and Al₂O₃ nanofluids dispersed in YT198, a mineral oil commonly used in data center liquid cooling. Both types of nanoparticles enhanced the thermal conductivity, although the effect of increasing the Al₂O₃concentration was relatively limited. In contrast, CuO nanoparticles showed a consistent, monotonic increase in thermal conductivity with a rising volume fraction. This trend is illustrated in Figure 1.

Experimental evidence by Aksoy et al. [37] confirms that once factors such as the nanoparticle deposition [38], surface roughness changes [39], surfactant effects [40], material removal issues [5], and spreading behavior [41] are accounted for or excluded, comparable heat transfer rates are observed between nanofluids and base fluids. However, it is now clear that thermal conductivity alone does not fully explain the observed enhancements. These limitations highlight the need to consider additional mechanisms, which will be explored in subsequent sections.

4. The Effect of Nanoparticles on Droplet Behavior

To understand the physics behind nanofluid heat transfer enhancement in spray cooling, it is essential first to consider how nanoparticles affect droplet dynamics.

Droplet splashing is primarily governed by surface tension and viscosity, typically characterized by the Weber and Reynolds numbers, respectively [42,43]. In addition to fluid properties, the splashing threshold is shifted by the substrate temperature and surface roughness [44]. The splashing mechanism proceeds as follows [45]: upon impact with a solid surface, a rim starts to spread horizontally. The air beneath the rim is compressed, unable to escape quickly enough, which results in a buildup of pressure that pushes back against the spreading edge. This interaction produces undulations along the rim due to surface tension, eventually lifting the liquid into a corona. The thin liquid sheet forming this corona then retracts and fragments into smaller droplets. Thus, splashing should be viewed as a liquid–air interaction sensitive to changes introduced by nanoparticles.

Aksoy et al. [46] showed that even small amounts of Al₂O₃nanoparticles in droplets can trigger an earlier transition from spreading to splashing in high-viscosity liquids. In other words, splashing occurs at lower Weber numbers than those for the base liquid. Their non-dimensional analysis isolated the effects of the nanoparticles by canceling out the contributions from surface tension, density, and viscosity due to other factors. Supporting this, Thoraval et al. [47] observed a similar splashing promotion with Ag nanoparticle suspensions at 52 wt.%, although this splashing was suppressed at higher Weber numbers.

On the other hand, Abbot et al. [48] found that lower-density SiO₂ nanoparticles do not modify the splashing threshold in low-viscosity base liquids. The collective information about the droplet splashing due to nanoparticles is presented in Figure 2. In conclusion, these findings suggest that nanoparticle inertia may play a role in the splashing threshold, particularly in high-viscosity liquids where the splashing promotion is more pronounced. Therefore, the base liquid’s viscosity should not be overlooked when assessing the impact of nanoparticles on droplet dynamics.

Besides the addition of nanoparticles, the presence of surfactants also significantly alters splashing via surface tension modulation. The splashing behavior is mainly determined by surfactants that are adsorbed at the interface at the moment of impact, rather than those adsorbed afterward [49]. This emphasizes the importance of the surfactant selection in nanofluid preparation, introducing an additional implicit parameter considered in the heat transfer performance. Notably, surfactants have been shown to suppress receding, jetting, and splashing phenomena in nanofluid sprays, behaviors often observed in homogeneous liquids. Such suppression is beneficial in applications including inkjet printing, oil recovery, and biomedical diagnostics [50]. Improved spreading may also benefit spray cooling, where stable wetting and fluid–surface interactions are crucial.

Nanofluid droplets spread more extensively than the droplets of their base liquids at low Reynolds numbers [51,52]. As the nanoparticle concentration increases, the droplet behavior may deviate from Newtonian dynamics, displaying shear-thinning or yield-stress characteristics [53]. The nanoparticle concentration is particularly influential during the receding phase [54]. By tuning the impact velocity and the initial nanoparticle concentration, thereby influencing Marangoni flows driven by concentration gradients, a range of deposition patterns, from conventional ring-shaped residues to more complex mountain-like structures, can be observed [55].

Yet the type and wettability of nanoparticles can either decrease or increase the contact angle between nanofluid droplets and a solid surface depending on the particle surface chemistry [56]. This suggests that hydrophobic particles behave differently from hydrophilic ones: instead of dispersing uniformly, they tend to adsorb and accumulate near the three-phase contact line, forming a viscoelastic film on the droplet surface. This results in increased local viscosity and surface roughness, which in turn enhances the frictional dissipation and leads to stronger contact line pinning (see Figure 3). Under an increased surface tension, the droplets recede more due to a higher surface energy [55]. For a more detailed discussion on the influence of the nanoparticle material, size, and concentration on the droplet impact dynamics, we refer interested readers to reference [57].

Nanofluid sprays can outperform pure water sprays at intermediate Weber numbers and near the critical heat flux (CHF), but this advantage tends to diminish at very high Weber numbers. Under film boiling conditions, the relative performance of nanofluids versus water remains inconclusive. Near the CHF, water may surpass nanofluids if the particle concentration is low, while at the Leidenfrost point, the nanoparticle concentration becomes less significant. Nevertheless, increasing the concentration generally raises both the CHF and Leidenfrost temperatures [10].

Zhang et al. [58] compared the evaporation dynamics and deposition patterns of nanofluids containing multi-walled carbon nanotubes (MWCNTs) and multi-layer graphene (MLG). MLG droplets primarily evaporate in the constant contact radius mode and help reduce the temperature difference at the liquid–vapor interface. While both nanomaterials influence the final deposit, MWCNTs suppress the coffee-ring effect better. However, they produce smoother patterns than MLG, which forms deposits that are approximately 1.8 times rougher. In their experiments, an infrared camera was employed to monitor the temperature distribution at the liquid–vapor interface of nanofluid droplets, and an optical profilometer was used to quantify the sedimentary patterns after evaporation.

In sessile droplets of Al₂O₃–ethanol nanofluids, three distinct internal convection regimes were observed during evaporation. The presence of nanoparticles suppresses hydrothermal waves, and the interplay of Marangoni and buoyancy-driven flows eliminates the coffee-ring pattern, producing deposit structures that reflect the internal flow evolution [59]. Qian et al. [60] demonstrated that both rigid and elastic superhydrophobic surfaces can effectively repel nanofluid droplets from cold substrates, offering a novel strategy for anti-icing, deicing, and self-cleaning. In particular, an elastic polydimethylsiloxane (PDMS) surface, aided by trapped air pockets and surface elasticity, enables rapid droplet detachment and complete rebound without splashing or residual droplet formation, even at high impact velocities.

Further control over the droplet residue patterns can be achieved through nanoparticle chemistry. Yogita et al. [61] demonstrated that reduced graphene oxide (rGO) and graphene oxide (GO) yield distinct deposit morphologies after desiccation. While rGO followed an outward capillary flow to form classic coffee-ring patterns, the amphiphilic GO was captured by the descending liquid–vapor interface due to hydrogen bonding, resulting in uniform, saucer-like deposits (see Figure 4).

In a study on nanofluid droplet motion [62], the critical airflow velocity required to initiate movement in Au nanofluid droplets was found to be slightly higher than that for pure water droplets. This is attributed to the pinning of nanoparticles at the three-phase contact line, caused by their adsorption at the gas–liquid interface [63] and the presence of a capillary compensation flow within the droplet [64]. Additionally, at the same airflow velocity, the sustained motion velocity of Au nanofluid droplets was lower than that of water droplets. High concentrations (1–2 wt.%) of n-Al particles enhanced droplet spreading and significantly suppressed evaporation on aluminum surfaces. The contact line remained pinned throughout evaporation due to particle aggregation, leading to a larger equilibrium diameter and reduced edge recoil, contrary to earlier studies, where nanoparticles typically accelerated the evaporation on ammonium perchlorate surfaces [65].

5. Alternative Heat Transfer Mechanisms in Nanofluids

While increased thermal conductivity is a significant contributor to heat transfer enhancement by nanofluids, it does not fully explain the observed improvements in many applications. Several heat transfer mechanisms compete in the presence of nanofluids, often leading to inconsistent results unless all influencing parameters are considered. These parameters can be broadly categorized into two domains: fluid dynamics and heat transfer.

According to Chang et al. [66], spraying Al₂O₃–water nanofluids onto copper surfaces initially enhances the heat transfer; however, the HTC decreases with prolonged spraying or higher nanoparticle concentrations. The adsorption layer formed by Al₂O₃ increases the thermal resistance due to its lower conductivity than copper. This layer also increases the surface hydrophilicity. At low concentrations, nanoparticles enhance heat transfer by promoting disturbances in the liquid film through Brownian motion. In contrast, high concentrations increase the viscosity, surface tension, and particle agglomeration, which degrade nozzle atomization and reduce the cooling efficiency.

Nanoparticle deposition is a tricky phenomenon. They may deposit onto a heated surface, forming an insulating layer. The sedimentary region comprises void spaces and interconnected nanoparticles, which become denser as the nanoparticle volume fraction increases. In porous media, these spaces are filled by the working fluid in pipe flows. The effective thermal conductivity of this region depends on the porosity and is determined by the conductivities of both the fluid and the nanoparticles [67]. In spray cooling applications, these voids may be partially filled, depending on the spray density and substrate temperature.

On the other hand, the possibility of enhanced heat transfer via surface modification through nanoparticle deposition has garnered attention in various applications, including electronic device cooling, microelectronics, nuclear reactors, and compact heat exchanger systems [68]. Nanoparticle-coated surfaces exhibited an improved pool boiling HTC and CHF due to an increased nucleation site density, higher bubble release frequencies, greater porosity, enhanced wettability, and reduced wall superheats. However, beyond an optimal deposition thickness, the thermal resistance increases, leading to a decline in heat transfer [69].

Bao et al. [70] reported only a marginal heat transfer enhancement, within measurement uncertainty, using water-based Al₂O₃, TiO₂, ZrO₂, and SiO₂ nanofluids to cool a copper block. Wang et al. [71] used water-based Cu, CuO, and Al₂O₃ nanofluids to cool an AISI 304 stainless steel (SS) plate from 720 °C. They observed an increase in the heat flux at the impinging central point with higher concentrations of Cu nanoparticles. Of this enhancement, 4% was attributed to the nanofluids. In contrast, an additional 13% increase was attributed to the use of a surfactant, which decreased the contact angle and improved the wettability; hindered nanoparticle agglomeration; and increased effective collisions between the particles and the SS surface. In another study, cooling a 700 °C SS substrate with nanofluids at concentrations below 0.1 wt.% showed no nanoparticle deposition, with the marginal heat transfer enhancement attributed mainly to increased thermal mixing rather than the conventionally understood improvement due to thermal conductivity [72]. Zhou et al. [73] evaluated Cu, CuO, and SiO₂ nanoparticles in an ethylene glycol–water mixture, finding that Cu nanoparticles yielded the highest enhancement in the heat transfer performance, exceeding 10%, within an optimal concentration range. Nevertheless, increasing the nanoparticle concentration beyond this range reduced the heat transfer efficiency due to a greater coating thickness and higher fluid viscosity. They also observed the dependency of the heat transfer on the surfactant.

The reduced surface tension of nanofluids, resulting from the addition of surfactants, is another key factor determining the instability wavelength of a vapor film. A shorter wavelength facilitates vapor film breakup, thereby improving the transition boiling performance [32]. A lower surface tension also increases the number density and coalescence of bubble nuclei on the heated surface, fostering more efficient nucleate boiling. Additionally, during nucleate boiling, some nanoparticles are deposited onto the heater surface, forming a porous layer, which considerably contributes to surface wettability [74]. This shift from a transition to a nucleate boiling regime leads to faster heat removal and an improved thermal performance. This phenomenon would be particularly pronounced when a liquid film accumulates on the cooled surface due to a dense spray.

In some studies, deposited nanoparticles have been shown to significantly enhance the heat transfer in spray cooling applications. Aksoy et al. [38] demonstrated that 0.2 wt.% water-based TiO₂ nanofluids can cool down an aluminum block from 190 °C to 65 °C approximately 25% faster than pure water. The enhancement in the heat transfer is even more pronounced in the boiling regime. This improvement is primarily attributed to the deposition of nanoparticles onto the heated surface, which introduces additional nucleation sites that facilitate a more efficient phase change. At lower concentrations (e.g., 0.1 wt.%), the nanoparticles are readily flushed away by the liquid film, limiting their effectiveness. In contrast, at higher concentrations (e.g., 0.2 wt.%), a continuous nanoparticle layer forms, maintaining the availability of nucleation sites and significantly enhancing the heat transfer. This behavior may be linked to the formation of nanoparticle agglomerates during droplet evaporation near the hot surface; these agglomerates, being larger and heavier, are less prone to removal by the flowing liquid. However, such an enhancement cannot be maintained in the non-boiling regime, where the heat transfer is dominated by convection and the role of the nucleation sites diminishes. Surface characterization after experiments using a 3D optical profiler further supports the presence of increased nucleation sites as a result of nanoparticle deposition [75].

The deposition of nanoparticles offers superior sustainability for heat transfer enhancement compared to conventional surface coatings [76]. Experimental results on repeatedly nanoparticle-deposited surfaces demonstrate that the deposits can self-renew through continuous nanoparticle spraying, maintaining an optimal coating over time. These nanoparticle layers inherently organize nucleation sites on the heated substrate, adapting to shifting hotspot locations under varying operational conditions. In closed-loop spray systems [77], even low-concentration nanofluids significantly enhance the heat transfer, providing a cost-effective solution. Moreover, multi-walled carbon nanotube (MWCNT) deposits exhibit markedly improved heat transfer properties compared to those of uncoated surfaces, attributed to their superior thermal conductivity, increased effective heat transfer area, and enhanced surface roughness [78]. Figure 5 illustrates the self-maintenance coating mechanism.

Nanofluids influence practical applications in several ways, notably through surface deposition. Wang et al. [79] reported that deposition correlates with the contact angle and decreases with an increasing Reynolds number. Assuming no multilayer deposition, the particle removal accelerates as the Reynolds number increases. Mao et al. [80] further showed that the removal rate also depends on the shear rate.

Copper surfaces coated with Cu–COOH–MWCNT hybrid nanoparticles exhibit a significantly enhanced boiling heat transfer coefficient compared to that of bare copper [81]. The increased surface roughness and superhydrophilic nature (contact angle < 15 °) facilitate the early onset of nucleate boiling, characterized by smaller bubble departure diameters and higher departure frequencies, even at a low heat flux. Figure 6 illustrates the visual differences between bare and coated copper surfaces; in fact, Figure 6b,c show SEM images of bare and coated samples, respectively, highlighting the morphological changes induced by the 1 wt.% Cu–COOH–MWCNT hybrid nanoparticle coating.

To summarize, nanoparticle-driven surface modifications primarily alter boiling regimes during droplet impacts by introducing additional nucleation sites, which promote earlier bubble inception and a transition to nucleate boiling. Beyond this dominant effect, several other mechanisms have been reported:

• Wettability alteration: Nanoparticle coatings can change the static and dynamic contact angle, thereby influencing liquid spreading and rewetting and shifting the Leidenfrost point;
• Surface roughness increase: Nanoparticle deposition increases micro- and nano-scale roughness, which increases the heat transfer area and creates vapor escape pathways that promote more efficient boiling;
• Thermal conductivity pathways: A nanoparticle layer may improve solid–liquid thermal coupling and reduce the interfacial thermal resistance.

Overall, these modifications not only increase the nucleation site density but also affect wettability, roughness, and interfacial thermal resistance, thereby reshaping the balance between different boiling regimes during droplet impacts.

6. Sustainable Cooling Systems Incorporating Nanofluids

Nanofluids present a promising route for forming self-assembled layers and self-maintaining surface coatings via controlled nanoparticle deposition, eliminating the need for complex fabrication steps in traditional surface engineering. By adding different nanoparticles into coolants, it becomes possible to fabricate enhanced, patterned heat transfer surfaces with adaptive or self-healing properties in a scalable and cost-effective manner [9]. Recent studies have demonstrated that coatings using nanoparticles can sustain an effective cooling performance throughout their operation [76,78]. Moreover, nanoparticle deposition during boiling can be controlled by pre-conditioning the substrate—for example, through surface texturing, which alters the wettability, before cooling by the nanofluid [82].

Ensuring long-term nanofluid stability also depends on proper control over the pH, nanoparticle concentration, and preparation methods [83]. For example, water-based Al₂O₃ nanofluids show improved stability at lower nanoparticle concentrations due to reduced particle interactions and at an optimal pH around 4, where increased electrostatic repulsion limits aggregation. At elevated temperatures, the enhanced Brownian motion initially delays sedimentation but also increases the frequency of interparticle collisions, raising the risk of agglomeration over time [84]. Repeated use may thus lead to performance degradation due to nanoparticle aggregation, settling, or chemical breakdown, exposing the challenges that must be addressed to ensure long-term viability in industrial systems [85].

Evran et al. [86] conducted a numerical analysis to assess the contribution of the nanoparticle type and concentration to the heat transfer efficiency, using the heat transfer coefficient, wall shear stress, and friction factor as key metrics. They reported that the friction factor remained largely unaffected. Another numerical study by Wang et al. [87] investigated nanoparticle deposition and its implications for both heat transfer and pressure drops. Their findings underscore the importance of managing deposition to avoid excessive pressure losses, highlighting the need to optimize the nanoparticle usage when designing thermal systems. This optimization is equally critical for associated components in fluidic loops to ensure long-term nanofluid recovery and sustainable operation.

Noble metal nanofluids offer a compelling alternative by inherently reducing the risks of fouling, deposition, and sedimentation, likely due to their diamagnetic nature, which contributes to long-term colloidal stability and system cleanliness [11]. They typically require lower concentrations to achieve thermal enhancements, making them attractive for large-scale applications, thus reducing material costs and minimizing issues such as erosion, clogging, and pressure drops. Consequently, noble metal nanofluids enable more aggressive performance optimization without compromising the usual operational drawbacks. Many can also be synthesized using environmentally friendly processes, enhancing their sustainability appeal. Nevertheless, a careful economic analysis is essential to balance the initial investment against the expected gains in the heat transfer performance.

After exploring the potential of nanoparticles to enhance sustainable heat transfer in cooling systems, several practical challenges must be addressed to enable their effective deployment in real-world applications.

The presence of nanoparticles introduces a risk of clogging in narrow passages such as nozzles and microchannels, where particle accumulation can obstruct the flow and diminish the cooling efficiency. Over time, agglomeration can exacerbate these blockages and destabilize the nanofluid, compromising system reliability [88]. Furthermore, the abrasive nature of nanoparticles can accelerate the wear on mechanical components, including pumps and nozzles, thereby increasing maintenance demands and operational costs [89]. Inconsistent results across the literature can be attributed to nanoparticle clustering and tribological effects, such as abrasion, erosion, and corrosion, which cause material loss and alter the flow conditions, leading to changes in viscosity and channel dimensions over time [90].

A study on recycled Al₂O₃ nanofluids shows that the improvements in the CHF and the burnout heat flux (BHF) may decrease by 50% and 20%, respectively, after recycling, though the performance still exceeds that of the base liquid [91]. While this decline is anticipated, the long-term use of recycled nanofluids, beyond several cycles, should be investigated to determine the extent of the performance loss and identify precautions to extend their lifetime for sustainable operation. Figure 7 represents the CHF and BHF values for ethylene glycol–water mixture-based Al₂O₃ nanofluids.

Greater attention should be given to the long-term impacts of nanofluids in cooling systems, including potential issues like corrosion and clogging [92]. Conducting a life cycle analysis may be essential to assess the long-term suitability of nanofluids for such applications [36].

While material wear, recyclability, and maintenance have been discussed, it is important to note that quantitative sustainability indicators such as life cycle energy intensity, CO₂-equivalent emissions, market prices, and ecotoxicological endpoints for nanomaterials are highly system-dependent. The reported values vary widely with particle size, surface functionalization, synthesis route, and geographical energy mix, often differing by more than an order of magnitude across studies. Presenting a single “indicative” table may therefore create a misleading impression of a consensus. Instead, we emphasize this limitation explicitly and direct readers to recent systematic reviews that compile such data under controlled boundary conditions [93,94,95,96,97].

Table 1. A summary of experimental studies on nanofluid droplets and sprays with explanations of the underlying heat transfer mechanisms.

Author and Year	Nanofluid	Size	Concentration	Substrate	Temperature	Remarks
Shen et al. 2008 [52]	Au in tannic acid, Na3C6H5O7·2H2O, K2CO3, water	10–30 nm		Silicon	75–79 °C	Larger spreading rate, diameter, and early-stage dynamic contact angle observed with nanofluids.
Duursma et al. 2009 [98]	Al in water, ethanol, and dimethyl sulfoxide	20–50 nm	up to 3.2 wt.%	Copper	<120 °C	Increasing the nanoparticle concentration discourages receding breakup.
Mitra et al. 2012 [99]	TiO2, MWCNTs in water	20–70 nm, 100–500 nm	0.01–0.1 wt%	SS		The vapor film thickness is comparable to the nanoparticle deposition layer, suggesting that the shift in the boiling curve arises from vapor film instability induced by the surface-deposited nanoparticles.
Okawa et al. 2012 [100]	TiO2 in water	21 nm	0.2 kg/m3	AISI304	400 °C	Nanoparticle-coated surfaces enhance evaporation at low wall superheat by retaining more liquid. At high temperature, early vaporization reduces liquid–solid contact, degrading late-stage heat transfer.
Chang et al. 2012 [101]	Al2O3 in water	35 nm	0.001–0.05 vol.%	Copper		High-concentration of nanoparticles deposit on the surface, reducing nucleation points and hindering convective heat transfer.
Bellerova et al. 2012 [6]	Al2O3 in water	d50 = 80 nm	1–40 wt.%	AISI314 SS	200 °C	HTC decreases by 20% in spray cooling.
Bellerova et al. 2012 [7]	Al2O3 in water	d50 = 80 nm	1–40 wt.%	AISI314 SS	200 °C	HTC decreases by 45% in jet cooling.
Tseng et al. 2014 [102]	TiO2 in water	82 nm	1–40 wt.%	SS	200 °C	HTC decreased with higher nanoparticle fraction due to differing impact behavior of solid particles versus fluid droplets.
Ravikumar et al. 2015 [32]	Cu in water	34 nm	0.1 vol.%	AISI304 SS	900 °C	Nanoparticle deposition enhances nucleate boiling by increasing nucleation sites. It suppresses vapor layer formation during transition boiling, boosting heat transfer.
Chang et al. 2015 [103]	Al2O3 in water	25–40 nm	0.001–0.05 vol.%	Copper		A nano-adsorption layer forms on the sprayed surface, becoming thicker with higher nanoparticle concentration, increasing thermal resistance and reducing heat transfer. Heat transfer enhancement increases with heat flux due to intensified bubble generation. Although rougher surfaces generally improve heat transfer, this effect is suppressed at high concentrations by the uniform nano-layer. The porous layer enhances capillary forces and wettability, lowering the contact angle and increasing surface hydrophilicity.
Hsieh et al. 2015 [104]	AG, MWCNTs in water	15 nm, 10–250 nm	0.0025–0.0075 vol.%	Copper	300 °C	HT enhancement mainly due to increased mixing. Ag nanofluids outperform MWCNTs in the nucleate boiling regime despite lower conductivity, likely due to better dispersion and fewer agglomerations.
Jha et al. 2015 [105]	Al2O3-water + surfactant	5–50 nm	100 ppm	SS	900 °C	Alumina nanofluids cool faster than water; SDS enhances, while Tween-20 reduces, the cooling rate.
Nayak et al. 2016 [72]	Al2O3, TiO2-water	20 nm	0.01–0.07 wt.%	SS	700 °C	At 0.01 wt.%, the heat transfer was similar to that with DI water; the enhancement was mainly due to mixing, not conductivity. Al2O3 outperformed TiO2 due to better dispersion and less agglomeration. Minimal deposition observed.
Hsieh et al. 2016 [106]	Ag, Al, Al2O3, Fe3O4, SiO2, TiO2, MWCNTs in water	5–50 nm	0.04–0.1 vol.%	Copper	400 °C	Heat transfer improves with a higher volume fraction, the smallest nanoparticle size, and lower surface tension.
Chang et al. 2018 [66]	Al2O3 in water		0.001–0.05 vol.%	SS		Nanoparticles deposit a porous, hydrophilic layer that increases the thermal resistance and reduces the heat transfer efficiency.
Modak et al. 2017 [107]	CuO in water	50 nm	0.15–0.6 vol.%	AISI304	500 °C	Nusselt number enhanced with an increasing volume fraction: for $\varphi =0.15\%$, the enhancement was 14% ($l/d=6$) and 13.6% ($l/d=12$); for $\varphi =0.60\%$, the enhancement reached 62% and 90%, respectively.
Tiara et al. 2017 [108]	Al2O3 in water	9–50 nm	1–20 ppm	AISI304	900 °C	Surface roughness tests confirm increased nucleation sites due to nanoparticle deposition. A thin sorption layer formed after jet impingement enhances the heat transfer by promoting nucleate boiling.
Chakraborty et al. 2018 [109]	Cu-Al LDH nanofluid	40–240 ppm		AISI304	600–900 °C	Cu-Al Layered Double Hydroxide (LDH) (4:1, 120 ppm) gave the highest cooling rate and average heat flux, with 19% and 12.5% enhancements over water.
Chakraborty et al. 2018 [110]	Cu-Zn-Al LDH nanofluid	49 nm	40–240 ppm	AISI304	>900 °C	Deposition enhances the heat transfer by increasing nucleation sites via surface roughness, but excessive coverage (>160 ppm) hinders droplet contact, reducing the performance.
Chakraborty et al. 2019 [111]	Cu-Zn-Al LDH in water + surfactant			AISI304	600–900 °C	Both SDS and Tween 20 reduced surface tension and viscosity. SDS enhanced thermal conductivity; Tween 20 had a negative effect.
Wang et al. 2020 [71]	Cu, CuO, and Al2O3 in water + surfactant		0.1–0.5 vol.%	SS	700 °C	A higher surfactant concentration improves wettability, prevents agglomeration, enhances the particle–surface interaction, and increases the heat flux via accelerated nucleate boiling.
Pontes et al. 2021 [40]	Au/Ag in water	5–10 nm	0.1–1 wt.%	AISI304 SS	120 °C	Nanofluid droplets show a slightly higher heat flux and larger spreading but a lower heat flux during receding, reducing the cooling efficiency. Caused by increased surface tension and viscosity due to the local nanoparticle concentration during evaporation.
Bao et al. 2022 [70]	Al2O3, TiO2−, ZrO2−, SiO2 in water + surfactant	75 nm		Copper		Low nanoparticle concentrations enhance the heat transfer via Brownian motion; high concentrations cause agglomeration, higher viscosity, and a reduced spray performance. The surfactant improves the dispersion and modestly boosts the heat transfer.
Marseglia et al. 2022 [112]	Al2O3, Ag in water + surfactant	40–50 nm	0.5–2 % (m/m)	AISI304	85–140 °C	A higher specific heat capacity improved the spray thermal performance, while a higher thermal conductivity unexpectedly reduced the HTC under the studied conditions. Alumina nanofluids at a low concentration showed the best performance. Increased viscosity hindered the heat transfer; the particle shape had minimal effect.
Siddiqui et al. 2022 [113]	Cu-Al2O3 and Ag–graphene hybrid nanofluid		0.01–1 vol.%	Copper		Hybrid nanofluid sprays show higher critical surface temperatures, likely due to the improved wettability and wickability of porous residues.
Aksoy et al. 2023 [38]	TiO2 in water	30–50 nm	0.05–0.2 wt.%	Aluminum	190 °C	The enhancement is due to more nucleation sites, uneven nanoparticle deposition, and partial flushing, rather than thermal conductivity changes. Higher concentrations lead to clogging.
Aksoy et al. 2023 [37]	TiO2 in water	30–50 nm	0.2–1 wt.%	Sapphire	80 °C	Nanoparticles affect the heat transfer via droplet spreading during early-stage cooling.
Padiyaar et al. 2023 [114]	MWCNT/Al2O3 hybrid nanofluid	7 nm (Al2O3), 10–20 nm, 10–30μm (MWCNT)	0.025–0.15 vol.%	Copper	105 °C	The HTC and the Nusselt number increase with the nanoparticle volume fraction, but this enhancement diminishes beyond 0.05 vol.%, with 0.10 vol.% and 0.15 vol.% showing a similar performance.
Zhou et al. 2024 [73]	Cu, CuO, SiO2 in aqueous ethylene glycol + surfactant		0.01–0.25 wt.%	Copper	≈95 °C	Cu nanoparticles give the best enhancement, followed by CuO and SiO2. Higher concentrations reduce the performance due to increased coating thickness and fluid viscosity.
Aksoy et al. 2025 [76]	TiO2 in water	5–50 nm	0.05–0.2 wt.%	Aluminum	190 °C	Nanofluid sprays enhance the cooling by forming and self-maintaining a nanoparticle coating that increases the nucleation sites.

presents an overview of experimental studies involving nanofluid droplets in both single and spray forms, along with the proposed mechanisms for the observed heat transfer modifications. Key experimental conditions are also included to provide researchers with a practical reference when designing or comparing similar studies and to support the identification of influential parameters affecting the thermal performance.

7. Insights and Future Directions for Nanofluids

This review synthesizes key insights to guide future research and development of nanofluids for enhanced heat transfer applications.

Nanofluids can significantly improve thermal systems’ performance by enhancing the thermal conductivity and introducing complex heat transfer mechanisms. However, their widespread adoption is constrained by challenges such as long-term stability [85], material wear [5,90,92,115], synthesis and dispersion difficulties [116], and cost effectiveness, compared to that of conventional coolants [117,118,119]. Addressing these practical issues is essential before focusing on their safe integration into sustainable technologies that account for environmental impacts like toxicity, biodegradability, and bioaccumulation. Establishing standardized reference nanofluids and clear regulatory frameworks that acknowledge their multiphasic nature will improve the data reliability and avoid misclassification, ultimately facilitating scalable adoption [120,121].

Many of the reported enhancements in the heat transfer coefficient are relatively small. In particular, several studies claim improvements of ≈5%, which fall within the typical apparatus uncertainty range (±7%) reported in the literature [8]. These cases should therefore be interpreted with caution, as the measured effect may not be statistically significant.

Polymer additives offer a promising strategy for modulating the droplet dynamics and flow behavior without sacrificing the thermal performance. Likewise, surfactants, commonly used to stabilize nanoparticle suspensions, can significantly influence the wettability and, consequently, the heat transfer, warranting careful consideration in system design [71]. The integration of machine learning methods holds strong potential for accelerating progress by enabling accurate prediction and optimization of nanofluid properties.

While hybrid nanofluids, comprising two or more distinct nanoparticle types, have recently attracted significant attention, their spray and boiling heat transfer performance cannot be directly compared to that of single-component nanofluids without accounting for the specific application, operating conditions, and interactions between the nanoparticle and substrate materials. These factors significantly the influence thermophysical properties, stability, and surface behavior, which in turn determine the overall enhancement in the heat transfer. Therefore, any evaluation of hybrid versus single-component systems should be conducted under well-defined experimental conditions.

Maintaining nanofluid stability in high-shear, high-temperature spray systems over prolonged operation remains a critical challenge. The conventional stabilization strategies, including ultrasonication, surfactant addition, and pH control, are often insufficient under conditions where strong shear forces and elevated surface temperatures accelerate nanoparticle agglomeration, sedimentation, and surface deposition. As a result, the long-term reliability of nanofluid-based spray cooling systems has not yet been demonstrated in practice. Future studies should therefore prioritize systematic investigations of stabilization techniques capable of withstanding extended operational times, with particular attention to the shear resistance, thermal durability, and evolution of the interfacial chemistry during repetitive spray–evaporation cycles.

Beyond heat transfer, nanoparticle-laden droplet dynamics have far-reaching implications in various fields, including inkjet printing, microarray fabrication, biosensing, and airborne virus detection, underscoring the interdisciplinary relevance of continued nanofluid research. This includes enhancing the latent heat storage in phase change materials by incorporating highly thermally conductive nanoparticles to overcome their low thermal conductivity barrier [122].