Experimental and numerical study on the heat transfer performance of nanofluids in micro-channel and mini-enclosure

Nimmagadda, Rajesh (2017) Experimental and numerical study on the heat transfer performance of nanofluids in micro-channel and mini-enclosure. PhD thesis, Indian Institute of Technology Hyderabad.

Text
ME12P1006.pdf - Submitted Version
Restricted to Registered users only until 2 July 2022.
Download (4MB) | Request a copy

Abstract

Heat transfer performance of nanofluids in micro-channel and mini-enclosure has important applications in cooling of power transistors, mi crochips, microprocessors, laser diodes, aerospace thermal systems etc. It is also used for over clocking and high performance computer cooling. In conventional coolin g systems, the heat transfer performance is limited due to low thermal conducti vity of working fluids. Moreover, the rate of heat transfer from the heated surface i n such systems will be lower due to low surface area to volume ratio. These two criti cal limitations need to be eliminated in order to enhance the heat transfer performa nce of conventional cooling systems. The low thermal conductivity of convention al working fluids is enhanced by dispersing high thermal conductivity nanopart icles within the matrix of base fluid forming nanofluids. The lower rate of heat transf er from the heated surface is enhanced by forming rectangular micro/mini groo ves on the surface. These micro-channels and mini-enclosures will enhance the surface area to volume ratio and helps in providing large contact area with the circ ulating coolant. Thus, the flow as well as heat transfer characteristics of nanofluid s in micro-channel and mini-enclosure are investigated in the present study. The objective of the present study is to investigate the heat transfer performance of oxide, metallic and carbon nanofluids with particle as well a s base fluid hybridization. The effect of Reynolds number, Grashof number, nanoparticle volume concentration, nanoparticle size, hybrid particle and base fluid mixture co ncentrations on the flow as well as heat transfer characteristics are studied and rep orted. It is also important to select nanofluids based on their cost along with their heat transfer performance. Hence, it is essential to find different types of nanofluid pairs that exhibit similar heat transfer characteristics under same operating conditions . The present study considers the flow as laminar two dimensional with Reynolds number vary ing from 10 to 50 as well as from 200 to 600. In the case of mini-enclosure, buoyan cy driven heat transfer performance is investigated for various Grashof numbers ra nging from 500 to 10000. Two numerical models namely two dimensional conjugate homo geneous phase model as well as two phase mixture model are developed as a par t of the numerical study. Both the developed models are validated with the expe rimental and numerical results available in the literature. Two phase mi xture model imposes more accuracy on the evaluated heat transfer characteristics as it considers the relative slip between particle and continuous phases. Pure water, al uminium oxide ( Al 2 O 3 ), silver (Ag), gold (Au), copper (Cu), aluminium (Al), single wal led carbon nanotube (SWCNT) and hybrid ( Al 2 O 3 + Ag, Cu + Al, pure water + methanol) nanofluids vi (HyNF) are used in the numerical investigation for analyzing th e heat transfer performance. Experimental study is also carried out by usin g de-ionized (DI) water and SWCNT nanofluid at low Reynolds numbers. The conduction ph enomena of the solid region show a significant effect on the heat transfer cha racteristics of nanofluids. Hence, the channel is considered with finite thick ness on its bottom to accommodate heat source or electronic component and a unifo rm heat flux is applied to the solid region. Sphericity based effective ther mal conductivity evaluation is used for investigating the effect of cylindric al particle size on the heat transfer performance. This evaluation process considers v olume and surface area of particles in order to take in to account the length and diamet er of carbon nanotube. The results show that the average convective heat transfer c oefficient as well as average Nusselt number increases with increase in nanoparti cle volume concentration, Reynolds number and Grashof number. 3 vol.% Al 2 O 3 as well as 3 vol.% Ag nanofluids enhances the heat transfer coefficient by 17 % - 18% and 111% - 144% respectively in comparison with pure water. In the case of hybrid (0.6 vol.% Al 2 O 3 + 2.4 vol.% Ag) nanofluid, the enhancement is found to be 126% - 1 48%. This shows that the use of hybrid nanofluids at high volume con centration reduces the cost of the working fluid and enhances the heat transfer ch aracteristics in comparison with that of metallic nanofluids. The interface t emperature between solid and fluid region increases along the length of the micro -channel. Hence, conventional approach of evaluating the heat transfer char acteristics directly within the fluid region using a constant surface temperature may not be appropriate for cooling of electronic devices. However, the interface tempe rature decreases with increase in Reynolds number and nanoparticle volume concen tration. The average Nusselt number increases with decrease in the diameter of sph erical nanoparticles. In the case of carbon nanofluid, sphericity based effective th ermal conductivity evaluation showed that increasing the length of SWCNT nanopa rticle has significant effect on the heat transfer performance concluding that axia l heat conduction dominates the radial heat conduction within the nanopartic le. Moreover, it is also observed that carbon nanofluid is identified as an optimized h eat transfer fluid with better heat transfer characteristics in comparison with go ld nanofluid. It also reduces the cost of the working fluid. The study presents two e quivalent nanofluid pairs namely [0.5 vol.% SWCNT and 2 vol.% Au with d p = 70 nm] as well as [1.5 vol.% SWCNT and 2 vol.% Au with d p = 50 nm] with similar heat transfer characteristics by using single phase homogeneous conjuga te heat transfer model. It is also observed that gold nanofluid with particle size less t han 90 nm is preferable vii for enhancing the heat transfer at a concentration of 2 vol.% . Further, it is also observed that multi phase mixture approa ch showed better enhancement in terms of heat transfer when compared with sin gle phase homogeneous model. The flow and heat transfer characteristi cs of working fluids at low Reynolds number are numerically evaluated by non-dimen sionalizing the pressure term in the momentum equation with viscous pressur e scale rather than dynamic pressure scale. The numerical study consider the eff ects of both inertial and viscous forces by solving the full Navier-Stokes equatio ns at low Reynolds numbers. The average Nusselt numbers evaluated experimenta lly and numerically for DI water as well as SWCNT nanofluid shows a difference at low R eynolds numbers. Four equivalent nanofluid pairs namely [1 vol.% SWC NT and 1 vol.% Au with d p = 50 nm], [2 vol.% SWCNT and 3 vol.% Au with d p = 70 nm], [3 vol.% Al 2 O 3 with d p = 30 nm and 2 vol.% Au with d p = 70 nm] as well as [3 vol.% HyNF (2.4% Al 2 O 3 + 0.6% Ag) and 3 vol.% Au with d p = 50 nm] with similar heat transfer characteristics are observed in the numerical inv estigation at low Re. However, in the case of laminar forced convection flows at high Reynolds number, three equivalent nanofluid pairs namely [1 vol.% Cu and 0.5 vo l.% SWCNT], [2 vol.% Cu, 1 vol.% SWCNT and 3 vol.% HyNF (0.6% Cu + 2.4% Al)] as well a s [2 vol.% SWCNT and 3 vol.% SWCNT (20% Me + 80% PW)] with similar heat transfer characteristics are identified. The study also sho ws that by dispersing SWCNT nanoparticles one can enhance the heat transfer charac teristics of base fluid containing methanol as antifreeze. The study on buoyan cy driven heat transfer performance of nanofluids shows that [1 vol.% Al 2 O 3 ( d p = 30 nm) and 1 vol.% Cu ( d p = 100 nm)] as well as [1 vol.% SWCNT and 3 vol.% Cu ( d p = 100 nm)] exhibit similar heat transfer characteristics. All these equivalen t nanofluid pairs with similar heat transfer characteristics provides a better switching option in choosing efficient working fluid with minimum cost based on cooling requirement . The study also investigated the effective comparison between three therma l conductivity models as well as three dynamic viscosity models. The study shows that sphericity based effective thermal conductivity evaluation exhibits higher heat transfer performance in comparison with other models. Furthermore, no significan t variation is observed with the use of different dynamic viscosity models. The prese nt results are in good agreement with the experimental and numerical results avai lable in the literature. Moreover, the results reported in the present study would be useful to understand the flow and heat transfer characteristics of nanofluids in mi cro-channel and provides guidance towards design of thermal systems for ele ctronics cooling

[error in script]

IITH Creators: