Thermal cooling is one of today’s most challenging technological problems, and there has been a constant effort by the scientific community to improve the heat transfer capabilities of cooling fluids. The notion that thermal conductivity of suspensions increases with the total surface area of the particles in the suspension triggered the concept of using nanosized particles in suspensions. This introduced a new class of engineered fluids called nanofluids: colloidal suspensions of nano-sized particles (approximately 5 to 100 nm) in a base fluid [1
]. The prevalent popularity of nanofluids in the heat transfer community has been due to its reported anomalous thermal conductivity enhancement. Following the initial development of nanofluids, several experimental reports have shown the great potential of these fluids to be used for heat transfer applications [2
]. However, there have been several recent reports questioning the heat transfer enhancement of nanofluids [10
]. These reports assert that nanofluids behave as a homogeneous mixture and that their thermal properties can be successfully predicted by classical effective medium theories [13
Several theoretical models and explanations have been proposed to explain the anomalous heat transfer characteristics of nanofluids. Brownian motion of nanoparticles in combination with aggregation and diffusion theories have been claimed to be the major justifications for the observed anomaly in thermal conductivity [17
]. Flattening of velocity profile, shear thinning, and thermophoretic forces in the near-wall region have been asserted to be the probable causes for enhanced convective heat transfer characteristics of nanofluids [9
]. However, it may be noted that most of the theoretically proposed models have not been directly validated by experimental results. Gaining an understanding of the near-wall flow field can help in better comprehending the phenomenon and assists in improving the theoretical models used in predicting the enhanced heat transfer characteristic of nanofluids. Walsh et al. [20
] used microparticle image velocimetry (μPIV) [21
] to obtain the velocity profile of nanofluids flowing inside a microchannel, d
1,085 μm. However, the spatial, out-of-plane resolution of their work was from 4 to 30 μm and could not capture the flow velocities in the region very close to the wall. To the best of the authors’ knowledge, there have been no other experimental studies that have investigated the near-wall flow fields in nanofluids.
The evanescent wave-based PIV technique, also known as nanoparticle image velocimetry (nPIV), is an extension of nPIV which can substantially improve the out-of-plane resolution of the measurements close to the wall [22
]. Evanescent wave-based techniques, namely total internal reflection microscopy and total internal reflection fluorescence (TIRF), have been widely used in biological and surface science fields in the past for near-wall visualization [23
]. In nPIV, evanescent wave illumination generated by the total internal reflection (TIR) of a laser beam at the fluid–solid interface is used to illuminate particle tracers in the flow field with a spatial resolution on the order of O(100 nm). This technique has been successfully implemented in the past in fluid velocimetry for studying electro-osmotic flows through microchannels [24
], near-wall Brownian diffusion [26
], and the effect of the near-wall forces [27
Present work demonstrates applicability of nPIV to investigate flow behaviors very near to the wall while using nanofluids. Near-wall measurements are reported for the first time for silicon dioxide (SiO2)-water nanofluid flow inside a microchannel at varying particle concentrations and flow rates with an out-of-plane spatial resolution of less than 300 nm. The results are then compared with those obtained for the base fluid. These results, along with the rheological characterization of the bulk nanofluids, are used to investigate occurrence of any non-homogeneity in the flow characteristics of the nanofluids in the near-wall region.
Similar to PIV, both μPIV and nPIV ‘track’ naturally buoyant fluorescent tracers to measure fluid velocity with the assumption that they follow fluid flow faithfully. As mentioned earlier, in nPIV, the evanescent wave generated at the glass water interface is used to illuminate particles only in the near-wall region. A brief introduction into the working principle behind nPIV follows.
When a light beam travels through a medium with a refractive index n1
into another transparent medium with a lower refractive index of n2
at an angle exceeding the critical angle θc
), it is totally reflected at the interface. However, the electromagnetic field penetrates into the lower refractive index region and propagates for a small distance parallel to the interface creating what is called an evanescent wave. This evanescent wave is capable of exciting fluorescent particles in this region, while the large numbers of particles farther away in the bulk liquid remain unexcited. One distinct characteristic of the evanescence wave is its nonuniform intensity in the direction normal to the interface, where the intensity, I
, decays exponentially with distance, z
, normal to the wall as follows:
I0 is the maximum intensity at the wall, and zp is the penetration depth:
where λ0 is the wavelength of the light, and θ is the incident angle. For visible light at a glass water interface, zp is on the order of O(100 nm) and is independent of the incident light polarization direction. It can be seen that in addition to the incident angle of the light, penetration depth depends on the refractive indices at the interface and the wavelength of light. Figure shows the schematic of a TIRF setup used in a nPIV experiment where only the near-wall fluorescent particles in the fluid are excited and viewed from the bottom of the microscope plate. The emission intensity of the tracer particles in this region is also an exponential function of the distance from the wall with a decaying trend as stated by Equation 1. However, depending on the optical characteristics of the imaging system, ultimate depth of visible region, zv, depends on the intensity of the incident laser beam, fluorescent particle characteristics, camera, and the background noise of the imaging system. In practice, this depth is usually more than the estimated penetration depth.
In a μPIV experiment, the whole flow field is illuminated, and the focal depth of the microscope objective sets the out-of-plane resolution of the measurement. The emitted light from the unfocused particle tracers acts as background noise for the measurement, reducing the signal-to-noise ratio of the measurement. However, with nPIV, the focal depth of the objective lens is larger than the penetration depth of the evanescence wave; therefore, all the particles in the image are in focus, and there is no background light. The brightness (size) of the particle images is a function of their distance from the wall, where particles near the wall look bigger and brighter than those further away. The effect of this nonuniformity in the tracer brightness combined with the effect of Brownian motion and the near-wall velocity gradient is discussed in detail in a recent publication [28
]. More details on TIRF, nPIV characterization, and its implementation can be found in the literature [22