High temperature and long-term stability of carbon nanotube nanofluids for direct absorption solar thermal collectors

Highlights

• Multi-walled carbon nanotubes suspensions are strong solar absorbers.

• Nanofluids exhibit long-term stability.

• Nanofluids exhibit high-temperature stability.

• Nanotube suspensions in Therminol VP-1 are not stable.

Abstract

Stable dispersions (nanofluids) are produced using plasma functionalized multi-walled carbon nanotubes (MWCNTs). To our knowledge, this study presents a first quantitative demonstration of nanofluid stability over extended periods of time (currently tested up to 8 months) and after intense heating. No agglomeration is found to occur in the water and glycol-based nanofluids after heating at 85 and 170 °C, respectively. Significant agglomeration does occur for suspensions produced using the non-polar Therminol® VP-1 heat transfer fluid. Optical characterization of the nanofluids demonstrates that the MWCNTs are highly absorbing over the majority of the solar spectrum, allowing for close to 100% solar energy absorption, even at low concentrations and small collection volumes. These absorption properties coupled with the stability of the nanofluids make them ideal candidates as direct solar thermal collectors.

Introduction

Solar thermal collectors, which capture the Sun’s rays as heat, are currently used world-wide in applications ranging from hot water heating to large scale electricity production. As demand for cleaner energy continues to grow, solar thermal processes are expected to play an increasingly significant role in our global energy future (Philibert, 2011). The concept of direct solar thermal collectors, in which solar radiation is absorbed and transported directly by the working fluid, was first introduced by Minardi and Chuang (1975). As most heat transfer fluids (water, glycols, oils, etc.) are transparent over the majority of the solar spectrum (Otanicar et al., 2009), additives, such as India Ink, can be added to improve the absorption characteristics of the working fluid. Unfortunately, India Ink, as well as other organic and inorganic chromophores have been shown to experience light- and temperature-induced degradation (Burke et al., 1982), which can lead to reduced absorption and fouling of surfaces over time, thus limiting their use, especially at high temperatures. Rather than using dyes, the use of nanoparticles, as dispersed and solar ray-absorbing phase in the heat transfer fluid, represents an interesting alternative. Once dispersed in a host heat transfer fluid, the nanofluid can pass through pumps and pipes without adverse effects, such as clogging and fouling. Such unique properties make the nanofluids ideal candidates for direct solar thermal energy absorption applications (Taylor et al., 2011a).

Nanofluids, which can be defined more formally as stable suspensions of nanoparticles (particles with one dimension <100 nm in size) in a host liquid, have gained a great deal of attention over the past 15 years. Initially considered as revolutionary in the heat transfer community, the actual heat transfer enhancements (heat conductivity and heat transfer coefficient) fell short of expectations. Recent works have demonstrated that more promising and scientifically-sound applications may stem out of nanofluids, in particular applications utilizing their highly-tunable optical properties (Taylor et al., 2013). Indeed, the use of nanofluids as both a volumetric solar collector and heat transfer fluid is now seen as a method of improving efficiencies and reducing costs in solar thermal devices (Mahian et al., 2013). Given the high surface-to-volume ratio of nanoparticles, quantities on the scale of mg/L can be used to obtain close to 100% absorption over relatively short distances (mm range). Furthermore, the concentration of nanoparticles can easily be controlled such that the incident radiation is absorbed over the entire volume of nanofluid, instead of over a thin surface layer, thus limiting heat losses to the surroundings (Otanicar et al., 2011). Several different types of nanoparticles in various base fluids have been modeled and tested experimentally for this purpose. These include metallic nanoparticles (He et al., 2011, Kameya and Hanamura, 2011, Lenert and Wang, 2011, Otanicar et al., 2010, Taylor et al., 2011a), carbon nanotubes (CNTs) (He et al., 2011, Meng et al., 2012, Otanicar et al., 2010, Taylor et al., 2011a), graphite (Otanicar et al., 2010, Taylor et al., 2011a, Taylor et al., 2011b), carbon black (Han et al., 2011, Sani et al., 2011), and carbon nanohorns (CNHs) (Sani et al., 2011). All of these materials have been shown to be effective solar absorbers, but graphitic particles, such as CNTs, are seen as the most promising for low to medium temperature applications (<400 °C) due to their high spectral absorptivity over the entire solar range (Taylor et al., 2011a). From an economic standpoint, cost is not seen as a prohibitive factor, as the CNT additive only adds around a dollar per liter of nanofluid produced.

One of the main obstacles to the large-scale use of nanofluids, in all applications, is the stability of the nanoparticles in suspension (Taylor et al., 2013; Yu and Xie, 2012). The unique properties of the suspension, attributed to small size and homogeneous dispersion of the particles, are lost if the nanoparticles agglomerate and settle out over time. The stability of nanofluids has, up to now, been essentially described qualitatively with little consistency for defining a “stable nanofluid”. More often than not, studies claim that stable nanofluids are produced without any quantitative evidence being reported. Photographs of the nanofluids are often provided as proof of stability. However for concentrated nanofluids, particularly those containing CNTs, the suspension is often too dark to determine if settling has occurred (Yu and Xie, 2012). In addition, even if sedimentation is not present, agglomeration of nanoparticles may still have taken place (Fedele et al., 2011). Agglomerated particles that remain in suspension may provide the appearance of a stable suspension over a short period of time, but the available surface area of the particles to achieve the specific purpose (absorb light, transfer heat, etc.), may have decreased significantly. For direct solar thermal energy collectors, agglomeration of nanoparticles translates directly into a decrease in absorption. Thus, it is essential that nanofluid stability (i.e. nanoparticles dispersion) be maintained over extended periods of time (years).

Analytical techniques involving spectral absorbance (Hwang et al., 2007, Jiang et al., 2003; Yu et al., 2007), zeta-potential (Huang et al., 2009, Jiang et al., 2003), particle-sizing (Fedele et al., 2011), and analytical centrifugation (Harel et al., 2013, Krause et al., 2009) have been used as ways to quantify nanofluid stability. All of these techniques can provide initial indications if agglomeration has occurred, however testing must be conducted over long time scales if stability quantification on the order of months or years is needed. Using an analytical centrifuge to examine the transmittance of the nanofluid over time can provide accelerated stability measurements, as larger agglomerates will settle quicker than under normal gravitational forces (Krause et al., 2009). However, the extent of agglomeration, unlike sedimentation, is not directly proportional to the applied gravitational force (Melik and Fogler, 1984). Long-term testing may still be required. In addition, suggesting that low sedimentation velocities are an indication of nanofluid stability, as has been recently proposed by Lamas et al. (2012), is somewhat questionable. A nanofluid that contains well-separated supernatant (particle-free fluid) and a sediment phase prior to testing should not be considered a stable and uniform system. If enough agglomeration has already occurred to create a heterogeneous fluid, then most likely the suspension that remains no longer contains particles that are “nano” in size.

Temperature stability of the nanofluid is another key aspect for all heat transfer applications. Typically, nanoparticles in a host fluid are stabilized through the use of a surfactant (Yu and Xie, 2012). Unfortunately, most surfactants decompose upon modest heating and can lose effectiveness at temperatures as low as 70 °C (Wen and Ding, 2004). Direct solar thermal collectors will require operating temperatures from 60 °C for solar water heating, to values as high as 400 °C for concentrated solar power (Taylor et al., 2011b). To date, very few studies have tested the stability of nanofluids at high temperatures and those that have made such tests reported significant agglomeration and irreversible nanofluid degradation (Tavares and Coulombe, 2011, Taylor et al., 2011b). Sani et al. (2011) did report good stability of ethylene glycol-based nanofluids up to 150 °C, but no evidence was provided to support their claim. To the best of our knowledge, no study has been done to quantify the temperature stability of a nanofluid.

The purpose of this work is to quantitatively examine both the long-term and high-temperature stability of CNT nanofluids for use in direct solar absorption. The optical properties of three base fluids that are commonly used in solar thermal applications, namely ethylene glycol, propylene glycol and Therminol® VP-1, are characterized along with a range of concentrations of corresponding nanofluids. Aqueous based nanofluids are also included for comparison, given that nanoparticle–water dispersions are by far the most common in the literature.