Optimization of nanofluid volumetric receivers for solar thermal energy conversion

Abstract

Improvements in solar-to-thermal energy conversion will accelerate the development of efficient concentrated solar power systems. Nanofluid volumetric receivers, where nanoparticles in a liquid medium directly absorb solar radiation, promise increased performance over surface receivers by minimizing temperature differences between the absorber and the fluid, which consequently reduces emissive losses. We present a combined modeling and experimental study to optimize the efficiency of liquid-based solar receivers seeded with carbon-coated absorbing nanoparticles. A one-dimensional transient heat transfer model was developed to investigate the effect of solar concentration, nanofluid height, and optical thickness on receiver performance. Simultaneously, we experimentally investigated a cylindrical nanofluid volumetric receiver, and showed good agreement with the model for varying optical thicknesses of the nanofluid. Based on the model, the efficiency of nanofluid volumetric receivers increases with increasing solar concentration and nanofluid height. Receiver-side efficiencies are predicted to exceed 35% when nanofluid volumetric receivers are coupled to a power cycle and optimized with respect to the optical thickness and solar exposure time. This work provides insights as to how nanofluids can be best utilized as volumetric receivers in solar applications, such as receivers with integrated storage for beam-down CSP and future high concentration solar thermal energy conversion systems.

Introduction

The use of concentrated sunlight as a thermal energy source for production of electricity promises to be one of the most viable options to replace fossil fuel power plants. However, the peak efficiencies of existing fossil fuel combined cycle power plants exceed 50% (Langston, 2009) while those of concentrated solar power (CSP) plants are below 20% (Pacheco, 2001, Romero et al., 2002). The relatively poor CSP performance is a result of a low solar-to-thermal efficiency (a combination of receiver and field efficiencies) and moderately low operational temperatures in the power cycle. Improving the conversion of incoming solar radiation to thermal energy at high temperatures is essential to improving the overall power conversion efficiency of CSP plants.

Most concentrated solar thermal technologies today use receivers with absorbing surfaces to convert solar energy from its radiative form into thermal energy. These surfaces are typically black or spectrally selective such that high absorptivity in the solar spectrum is coupled with low emissivity in the infrared (Bogaerts and Lampert, 1983). Although surface-based receivers are efficient at solar to thermal conversion, they are not well suited for transferring heat to a carrier fluid. In particular, at high levels of solar concentration, a large temperature difference between the absorber and the fluid arises. The temperature difference leads to significant emissive losses owing to the quartic dependence of thermal re-radiation on the absorber temperature, and correspondingly, a lowering of the overall conversion efficiency of solar energy. Moreover, the material stability of selective surfaces at temperatures above 800 K has not yet been demonstrated (Pitz-Paal and Trevor, 2008). Alternatively, in a volumetric receiver design, concentrated solar radiation is directly absorbed and more uniformly distributed in the surrounding fluid, which decreases the temperature difference between the absorber and the fluid.

Researchers have suggested various configurations for volumetric receiver designs, including: gas-particle suspensions (Bertocchi et al., 2004, Miller and Koenigsdorff, 1991), liquid films (Bohn and Wang, 1988, Caouris et al., 1978), and metal foams (Fend et al., 2004, Pitz-Paal et al., 1997). In this study, we focus on liquid-based volumetric receivers with integrated storage for central receiver CSP systems with beam-down optics (Epstein et al., 1999, Kribus et al., 1998, Yogev et al., 1998); an example configuration was recently described by Slocum et al. (2011) where hillside heliostats focus light onto a molten salt volumetric receiver. The potential advantage of such volumetric receivers (VR) compared to ideal selective surface receivers (SS) is illustrated in the representative schematic temperature profiles of Fig. 1. The exact temperature profiles will depend on the flow characteristics in the receivers, but for the same mean fluid temperature (T_f) and solar heat flux (CG_s), the temperature profile in the VR (Fig. 1a) can be favorable because the temperature associated with emissive loss is lower than that of the mean fluid temperature (by ΔT). This behavior is referred to as “thermal trapping” in solar thermal literature (Arai et al., 1984, Wijeysundera and Thevendran, 1988), but is physically similar to the “greenhouse effect” (Harries, 2000). On the other hand, the unfavorable temperature profile in the SS (Fig. 1b) leads to higher emissive losses. Fig. 1c highlights the difference in emissive loss (Δe) between an ideal selective surface with an ideal cutoff wavelength equal to 2 μm (purple) and a non-selective volumetric receiver (black) for the case when the mean fluid temperature is equal to 1000 K and ΔT is equal to 250 K. Thus, volumetric receivers, despite being non-selective, can trap thermal energy more effectively and lead to higher receiver efficiencies.

In particular, volumetric receivers with absorbing small particles in suspension have a high surface-to-volume ratio which minimizes the temperature difference between the absorber and the fluid (Hunt, 1978, Miller and Koenigsdorff, 2000). When the particle size is smaller than the characteristic wavelength of sunlight, less material is required to achieve the same amount of absorption (Hunt, 1978), and challenges related to clogging, sedimentation and erosion can be alleviated.

Past research on small-particle liquid suspensions as VRs has focused on thin liquid films and micro/mini-channel designs. Kumar and Tien (1990) developed a model for particle-laden falling liquid films (1–5 mm thick) incorporating the spectral and directional radiative properties of the particles, and provided a framework for future modeling studies. More recently, Tyagi et al. (2009) numerically investigated a low temperature nanofluid receiver inside a mini-channel; while, Otanicar et al. (2009b) extended this model to include multiple and dependent scattering, and size-dependent optical properties. Otanicar et al. (2010) also experimentally demonstrated the use of different nanofluids in a micro-channel solar collector. However, in these previous studies, the effect of increasing height of the absorbing liquid (H in Fig. 1a) beyond the millimeter-scale which can lead to lower emissive loss due to the favorable temperature profile was not considered. Arai et al. (1984) investigated transient radiative heating of a static semi-transparent liquid suspension in a taller receiver design (3 cm) and suggested that such a VR can be highly efficient. Nevertheless, an optimization of small-particle volumetric receivers with respect to particle loading, solar exposure time and nanofluid height has yet to be conducted. As the height of the nanofluid increases, the transient response, representing the thermal charging of these stationary volumetric receivers, becomes a significant portion of the daily operation because of their large thermal inertia.

In this paper, nano-sized (10–100 nm) solid particles are added to a liquid heat transfer fluid (i.e., nanofluid) to volumetrically absorb concentrated solar radiation. We investigate the design of these nanofluid volumetric receivers with nanofluid heights above 1 mm and develop a transient one-dimensional numerical model that examines the effect of solar concentration, nanofluid height, and nanofluid optical thickness on the temperature distribution inside the receiver (Section 2). We studied the effects of varying the optical thickness and validated the numerical model through experiments with a cylindrical nanofluid volumetric receiver, where the radiative and thermophysical properties of the nanofluid were experimentally characterized (Section 3). The model was subsequently used to determine the optimal exposure time and temperature at which point the nanofluid volumetric receiver should be thermally connected to a power generation cycle (Section 4). Throughout this study, a suspension of carbon-coated nanoparticles in Therminol® VP-1 is used as a model system for nanofluid receivers, but the results can also apply to other nanofluid VRs. The outcomes of the work suggest that nanofluids have significant potential as receivers with integrated storage for beam-down CSP systems and future high concentration solar thermal applications.