Unruled Notebook

Like Heat Capacity, the coinage Nanofluids is a misnomer.

The phrase Heat Capacity suggests the capacity of a substance for storing Heat, which it is not. Heat cannot be stored; it is energy in transit, a boundary phenomenon; substances store enthalpy or energy. Similarly, the word Nanofluid suggest a fluid that is nano-sized or, at least, made of nano sized matter, which at first glance, every fluid (air, water) is. Obviously Nanofluid should mean something more special.

Nano-meter is “one by ten power nine” th of a meter (that makes a nanopolitan is “one by ten power nine” th of multi colored tan). Typical atom size is “one by ten power ten” th of a meter – called an Angstrom – so, nanoscale objects are very small and invisible to the human naked eye (but nanoscale views is definitely bigger in size and very readable to the naked eye).

A Nanofluid is a fluid with a colloidal dispersion of nano-sized particles of another substance.

A colloid is typically not a single chemical compound made of two substances. In a colloid the two substances are distinguishable but can interact through weak surface molecular forces. Just after a shower, we apply oil and comb to settle the slightly wet hair nicely. The mix of oil and water is a colloid. We want such qualifiers as colloidal dispersion or suspensions. Otherwise coffee with sugar could suit our nanofluid definition.

Three methods are followed to prepare a nanofluid: nanoparticles dispersed in powder form in the base fluid, synthesis of nanoparticle by chemical precipitation and synthesis by organic reduction. The first method is simple which involves in principle, buying commercially available nanoparticles and dispersing it in a base fluid and keep shaking it like James Bond’s martini so as not to allow the particles to settle down. A crude explanation, but in reality the preparation is expensive.

A digression. Nanofluidics is not about making nanofluids. Nanofluidics deals with confining fluids in nano-sized objects and regions and studying their behavior (which gets peculiar). It is completely different from nanofluids, which has an ordinary base fluid with suspended nano-particles of (usually) metals. Given the history of secure knowledge that we have about colloidal solutions (remember Tyndall effect from our high schools), we could have called these nanofluids, say, metal colloids. But it doesn’t probably sound zany and representative in the era of nanotechnology.

So why every nano-researcher (big researchers who work in nano) and nano-researcher (yet-to-make-their-name researchers) is interested in nanofluid?

In a nanofluid, nano-sized particles of one substance – usually a solid – is dispersed or suspended in a base fluid – usually a liquid. Such a nanofluid is reported to exhibit properties that are remarkably different suggesting exciting applications. Steve Choi and Jeff Eastman of the Argonne National Labs in 2001 showed with preliminary experiments using copper nano-particles dispersed in ethylene glycol that nanofluids can enhance heat transfer by forty percent [see links in reference 1].

A recent news item in the Hindu reported about Dr. John Philip and his team at the IGCAR perfecting a method in which a small magnetic field is applied across a nanofluid to increase its thermal conductivity by three hundred percent. Usually such news items don’t clarify certain basic things. For instance, three hundred percent increase when compared to what? When compared to the base fluid – in this case, hexadecane, a viscous fluid.

The work done by John Philip and his group (and published in a journal paper) uses a magnetically polarizable nanofluid that is made of magnetite () nanoparticles with average diameter of 6.7 nm coated with Oelic acid and colloidally suspended in hexadecane. When such a magnetic nanofluid is tested for its thermal conductivity, it showed only about 25 percent increase from its base fluid (hexadecane) thermal conductivity. But when a magnetic field is passed through the nanofluid, the nano-sized magnetite particles aligned themselves in a linear sequence of micro scale length. This led to better thermal conduction through them, better by 216 percent over the base fluid conductivity for an applied magnetic field of about 100 Gauss, as reported by John Philip et al. in their 2008 paper published in the Applied Physics Letters [see reference 2]. The relevant result from their paper is given below under fair copyright use.

Is this increase in thermal conductivity of nanofluids useful? Definitely. In applications where aqueous medium is used to transfer energy (as heat), if “nanofluids” could be used, they could in principle conduct/convect away the heat three times quicker.

So what is the catch? Why don’t we see it, for instance, in our computer heat sinks (yet)?

One reason is the colloidal state is short lived (Update Jun 2, 2008: this is not always true. see the comment by John Philip below). After a while the nano-sized particles agglomerate and settle down into the base fluid as a separate compound. This is a major hindrance a work-around for which is a research field in itself. A typical lifetime for ethylene glycol well stirred with nano-sized copper nanofluid could be about twenty to thirty minutes during which period, the nano-particles of metal loose their surface charge to the aqueous medium. A temporary solution is to electrically charge the nano-sized particles and make them repel each other and the base fluid molecules to prolong their suspension state. For instance, in the APL paper about magnetic nanoparticles we mentioned Oelic acid coating of magnetite nanoparticles. This coating is to retain the surface charge on the nanoparticles so that they stay repelling each other and doesn’t agglomerate and settle down. Coating copper nanoparticles with Zinc stearate is another option.

In case they agglomerate, hitting the nano particle agglomerate with ultra sound (ultrasonification) is a workable solution, which breaks up the particles and re-suspends them in the base fluid. However, once the charge dissipates, the particles settle down and the nanofluid looses its remarkable characteristics. Such drawbacks are to be overcome for a commercially viable nanofluid heat sink for the thermal management of electronics. Should be possible in the near future.

Before we close, a quick explanation on how does nanofluid enhance heat transfer. It does by means of the high surface to volume ratio possible with nano-sized particles of a good conductor (such as copper), when compared to their micro or macro sized particles. About twenty percent more of the atoms of copper, for instance, would be near the surface of a nano-sized copper particle than a micro-sized copper particle. This configuration, when suspended in a base fluid (such as water), allows heat to be absorbed (via heat capacity of copper) and transfered (through thermal conductivity) much quicker in nano-sized particles. An explanatory schematic from the Argonne National Labs, carrying this argument, is given below.

[Image source: ANL Media Center Website]

A theoretical analysis using similar arguments has been performed by Prof. Peter Vedasz and published in the ASME Journal of Heat transfer in 2005 [see 3]. Based on theoretical modeling, this paper shows why the heat transfer between the nano-particle surface and the fluid should be accounted for while calculating the thermal conductivity of nanofluids. I shall explain this in a separate write-up as it would require equations, which I am keeping out from this post meant for my general readers.

Other theoretical reasonings exist for the high thermal conductivity of nanofluids. These include explanations using Brownian motion [see 4] and micro convection [see 5]. Although after 2005 there seems to be a consensus that Brownian motion cannot be discounted (see recent paper 6), a single conclusive explanation is yet to be arrived to categorically explain the increase in thermal conductivity of nanofluids as suggested by experimental results.

References

1. [ANL nanofluid - Eureka Alert link | ANL nanofluid picture link ]

2. Philip, J., Shima, P., & Raj, B. (2008). Nanofluid with tunable thermal properties Applied Physics Letters, 92 (4) DOI: 10.1063/1.2838304

3. Vedasz, P., Heat Conduction in Nanofluid Suspensions, ASME Journal of Heat Transfer, May 2006, vol. 128, 5, pp. 465-477

4. Jang, S. P., and Choi, S. U.-S., 2004, “Role of Brownian Motion in the Enhanced Thermal Conductivity of Nanofluids,” Appl. Phys. Lett., 84(21), pp. 4316–4318.

5. Kumar, D. H., Patel, H. E., Rajeev Kumar, V. R., Sundararajan, T., Pradeep, T., and Das, S. K., 2004, “Model for Heat Conduction in Nanofluids,” Phys. Rev. Lett., 93, p. 144301.

6. Ratnesh K. Shukla and Vijay Dhir, Effect of Brownian Motion on Thermal Conductivity of Nanofluids, J. Heat Transfer 130, 042406, 2008.