Stability of glycol nanofluids — The theory and experiment Sanjeeva Witharana ⁎, Ibrahim Palabiyik 1 , Zenfira Musina 1 , Yulong Ding 1 Institute of Particle Science and Engineering, School of Process, Environmental and Materials Engineering, University of Leeds, Woodhouse Lane, LS2 9JT Leeds, UK abstract article info Article history: Received 21 August 2012 Received in revised form 14 January 2013 Accepted 19 January 2013 Available online 26 January 2013 Keywords: Nanoparticles Nanofluids Glycols Formulation Stability Particle size Formulation of stable nanofluids containing ZnO, Al 2 O 3 and TiO 2 nanoparticles in propylene glycol (PG), ethylene glycol (EG) and 50 wt.% mixtures of PG and EG in water (WPG, WEG) was investigated, with and without the presence of surfactants. Nanofluid samples of particle concentrations 1–9 wt.% were prepared by dispersive method. Surfactant presence was in the range of 0–1 wt.%/wt.% of nanoparticles. Visual observation, particle size measurement and zeta potential analysis were performed to evaluate the dispersion stability. Overall the PG-based samples were found to be the most stable suspensions. The effect of base fluid on particle size and the effect of day light on nanofluid stability were also examined as a function of time. TiO 2 –PG samples showed a colour change when exposed to sunlight. Sunlight also caused the PG based TiO 2 and Al 2 O 3 nanofluid to in- crease their particle sizes by up to 45% in the course of 3 days. As for stability, the sedimentation velocity was ob- served to be a key parameter. Finally by comparison of settling theory with experiments, a stability boundary was demarcated to identify stable and unstable nanofluids. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Nanofluids, which are referred to as dilute liquid suspensions of nanoparticles in common fluids, have been a subject of great interest in the past decade due to their unique thermophysical properties and heat transfer behaviour. Experiments have shown that nanofluids were able to enhance the thermal conductivity and convective heat transfer by large margins [1–4], and critical heat flux by up to 300% [5–7]. In many instances, nanofluids also enhanced the pool boiling heat transfer [8,9]. Dispersion method, also called two-step method, is generally favoured for preparing nanofluids containing high volume fraction metals, oxides and carbon nanotubes. Here the dry nanopowder is dispersed in the liq- uid by application of one or many dispersion techniques [10]. This meth- od is more economical in comparison to one-step method, due to the low cost of nanopowders in the market. Decline of the price of nanopowders is a result of the rapid development of high-throughput nanoparticle pro- duction technologies over the years. Nanofluids prepared by dispersion method however commonly have shown a stability problem [11–16]. These nanofluids gradually start to settle after a period of time depending on the properties of base liquid, surfactant or dispersant used, type of nanoparticles, and the likelihood of nanoparticles to aggregate. The valid- ity of a nanofluid is as only long as it is stable. An agglomerated nanofluid is different in properties, and may cause operational problems similar to those encountered with micron-sized particulate suspensions; sedi- mentation and clogging of the system. Unstable nanofluids moreover are most likely a root cause for the wide discrepancies in literature data on their heat transfer behaviour. Therefore, the preparation of stable nanofluids is undoubtedly the first step in nanofluid research and applications. Colloids theory states that there is a critical radius below which the sedimentation of a particle ceases due to counterbalancing of gravity forces by the Brownian diffusion. Keeping the size of nanoparticles sufficiently small in the liquid should therefore be the focal point in the formu- lation exercise. On the other hand, smaller nanoparticles possess higher surface energies that causes higher tendency to build agglomerates among them. Furthermore, tiny particles causes higher electrolyte con- centration in the nano-suspensions; the reason is large surface area contains large amount of ionisable sites. In relation to this Jailani et al. [17] observed that high electrolyte concentration in nanofluids causes decrease in zeta potential. Keeping the particle sizes very small can hence be counterproductive for a stable nanofluid. The challenge of for- mulating stable nanofluids is to prevent coalescence of nanoparticles while keeping their size and concentration optimum in the base liquid. Particles suspended in liquids are in constant motion known as ‘Brownian motion’. As a consequence they randomly get closer to their neighbours. If the attractive forces between the two particles are strong enough, they form a cluster (or aggregate). DLVO theory explains this phenomenon to a greater depth [18]. Aggregation in a suspension can be delayed by using several tech- niques. In electrostatic stabilisation, the suspension stability is achieved by careful controlling of the solution chemistry, such as pH and ionic concentration. In steric stabilisation, the particle surface properties are modified by adding a surfactant or by adsorption of polymers. A third Powder Technology 239 (2013) 72–77 ⁎ Corresponding author at: Max Planck Institute for Solar System Research, Max Planck Str. 2, Katlenburg-Lindau, 37191, Germany. Tel.: +49 5556 979442; fax: +49 5556 979240. E-mail address: switharana@ieee.org (S. Witharana). 1 Tel.: +49 5556 979442; fax: +49 5556 979240. 0032-5910/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.powtec.2013.01.039 Contents lists available at SciVerse ScienceDirect Powder Technology journal homepage: www.elsevier.com/locate/powtec