Available online at www.sciencedirect.com
ScienceDirect Energy Procedia 49 (2014) 363 – 372
SolarPACES 2013
Lab-scale experimentation and CFD modeling of a small particle heat exchange receiver L. Fredericksona, M. Dordevicha, F. Millera,* a
Combustion and Solar Energy Laboratory, San Diego State University, 5500 Campanile Dr, San Diego, CA 92182, USA
Abstract Concentrating solar power currently relies on high temperature central receivers that utilize liquid cooling and operate in power steam cycles. However, highly efficient central receivers are being designed to operate at higher temperatures in a gas turbine power cycle. To address this, San Diego State University’s (SDSU) Combustion and Solar Energy Laboratory is experimenting with a lab-scale Small Particle Heat Exchange Receiver (SPHER) in order to understand performance and develop experience for designing and operating a full-scale 5 MW design. The full-scale design will be tested at the National Solar Thermal Test Facility at Sandia National Laboratories as part of the Department of Energy SunShot Initiative grant. The SPHER relies on carbon nanoparticles as an absorption medium and air as a working fluid. The carbon particles are generated onsite by the Carbon Particle Generator (CPG) and are mixed with dilution air prior to entering the SPHER. Lab scale on-sun testing is carried out with a 15kWe solar simulator. The lab scale experimental goal is to achieve an outlet flow of 650°C at 5 bar absolute operating pressure. To model the performance of the SPHER, CFD analysis is being used for comparison to lab scale testing. The lab scale SPHER is being modeled in ANSYS Fluent with coupled codes for oxidation and radiation input. In this paper, we present results of testing the lab-scale receiver and compare the measured outlet temperatures to predictions from the computer model. Finally, correlations are drawn for future experimenation and feasibility. © 2013 F. Miller. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2013 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and by by thethe scientific conference committee of SolarPACES 2013 under responsibility of PSE AG.of PSE AG. Selection andpeer peerreview review scientific conference committee of SolarPACES 2013 under responsibility Final manuscript published as received without editorial corrections. Keywords:Concentrated Solar Power; CSP; Small Particle Receiver; SPHER; Direct Absorption; High Temperature Receiver; Central Receiver; Brayton Cycle; SunShot; Gas Turbine; Solar Simulator
* Corresponding author. Tel.: +1-619-594-5791; fax: +1-619-594-3599. E-mail address:
[email protected]
1876-6102 © 2013 F. Millers. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and peer review by the scientific conference committee of SolarPACES 2013 under responsibility of PSE AG. Final manuscript published as received without editorial corrections. doi:10.1016/j.egypro.2014.03.039
364
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
1. Background The Small Particle Heat Exchange Receiver (SPHER) is a high temperature solar receiver that utilizes carbon particles which are suspended in air as an absorption medium. The full scale SPHER is to be installed as a central receiver in the tower of a concentrating solar power plant and operate within a Brayton cycle by preceding or replacing the combustor [1,2]. The full-scale SPHER is designed to produce a clear gas stream in excess of 1000°C at 5 bar, which will be used to power a gas turbine. This is possible by relying on the small carbon particles, with diameters of about 200 – 500 nm, to oxidize to carbon dioxide by the time they exit the receiver [3,4]. This eliminates the need for any downstream filtration or recycling, as particles are generated prior to entering the SPHER and oxidized prior to exiting the SPHER. The concept of utilizing small carbon particles in concentrating solar power research started in the 1970s. The idea was proposed separately by Hunt at LBL [5] and Abdelrahman et al at Sudan [6]. Hunt pursued further research on the topic by designing a receiver that operated at 1 bar and was tested at Georgia Tech test facility [7,8,9]. This receiver had a 30 kWth input, utilized particles produced from acetylene, and had a flat quartz window as an aperture. The receiver performed well, reaching outlet temperatures of 750°C. It is also important to note that, during testing, any particles that initially deposited on the window were burned off quickly and no permanent particle deposition was observed on the window. Extensive work was later done by the Weizmann Institute which looked at particle suspensions on a small scale along with window design for higher pressures [10,11]. The systems developed used pre-made commercial particles along with a different size and geometry than the SPHER. Testing was carried out with a solar furnace and outlet temperatures of 1727°C were achieved. This is a much higher temperature than the desired SPHER outlet temperature, but is useful for understanding the capabilities of the concept. Separate from using air as a working fluid was research done by Haag et al who used a particle receiver for hydrogen gas production [12]. The small scale, 5 kW receiver achieved an outlet temperature of 1227°C by using carbon particles suspended in natural gas for natural gas pyrolysis to hydrogen gas. Particle size is important to the success of the SPHER. Since carbon particles are generated within the system, there is some degree of flexibility during testing. The effect of mass loading and particle size on the receiver efficiency has been extensively researched by the use of numerical models which so far have not included the effect of particle oxidation. Conceptually, the radiation is absorbed by carbon particles inside of the SPHER, leading to >90% extinction across the SPHER path length. By numerous simulations with ANSYS Fluent, coupled with a Monte Carlo Ray Trace (MCRT) code, an optimal mass loading of 0.3 g/m3 has been shown in a full-scale SPHER with a 5 m path length [3,4]. These carbon particles are expected to oxidize by the entrance of the outlet because of the high oxidative rates at the desired receiver temperature. Comparison of an experimental setup with a numerical model was first done by Miller [13]. The numerical model attempted to predict the flux and temperature distributions inside the receiver, which were also a xenon arc lamp and a cylindrical particle receiver. This approach was not used for analysis in this study because of current technology that enables much more sophisticated flow and heat modeling through computational software. 2. Experimental setup The lab scale experimental equipment was designed and built at SDSU. The process diagram for the setup is shown in Fig. 1 [14,15,16]. The carbon particle generator (CPG) is a heated tube reactor that generates carbon particles from pyrolysis of a hydrocarbon fuel. Particles produced inside of the CPG are carried to the outlet by nitrogen, where they are met with a dilution flow of air. The gas-particle mixture flows to the SPHER, where the fluid is subjected to radiation from a solar simulator. The fluid heats as the carbon particles absorb radiation and transfer heat to the gas. The particles eventually oxidize and the mixture flows out as a clear gas stream at the exit of the SPHER. A cooling tower is installed for safety to exhaust gas at a low temperature. A back pressure valve maintains the system operating pressure at around 6 bar absolute pressure. The lab scale system is designed to mimic that of a large scale design, with the compressed air feed simulating the compressor outlet air, the solar simulator as the heliostat field, and the back pressure valve and cooling tower instead of a turbine. Important components are described below.
365
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
Solar Simulator MFC MFC MFC
CPG
Ext Tube
Air
Air
N2
HC
SPHER
Cooling Tower
Q
in
Fig. 1. Process flow schematic of the lab-scale experiment.
2.1. Carbon particle generator The CPG consists of an alumina tube with a Kanthal heating coil on the outer wall, wrapped in insulation and housed in a steel pressure vessel. Gases are delivered to the system from cylinders and regulated by mass flow controllers. Nitrogen and the hydrocarbon fuel are fed in a co-axial flow to the bottom of the vessel, where nitrogen is used as the carrier gas. The hydrocarbon undergoes pyrolysis as it is heated which produces carbon particles. The gas-particle mixture exits at the top of the CPG where it is mixed with incoming dilution air. This mixture then flows through the extinction tube and to the SPHER. 2.2. Solar simulator The solar simulator is a xenon arc lamp with an aluminum reflector sheltered inside a steel and aluminum housing. The xenon arc lamp is water cooled and powered by three 5-kWe power supplies. The solar simulator is designed for on-sun testing of the SPHER with the reflector to put a point focus on the quartz window aperture of the SPHER, much like a heliostat field would do. A radiometer is used to determine the output power within about a 5 cm diameter [14]. Flux at the window aperture is found by coupling radiometer data with camera imaging over a diffuse plate. The power input to the window can then be found with pixel intensity from the image. This is all done at a constant distance from the lamp, consistent with the SPHER aperture height. 2.3. Small particle heat exchange receiver The SPHER is a steel pressure vessel with a quartz window aperture connected by a water-cooled stainless steel flange. The inside of the SPHER is composed of a combination of ceramic fiber insulation and steel sheeting for maintaining flow paths. The gas-particle mixture enters the SPHER through an annular copper ring near the bottom of the bottom of the receiver in the outer insulating cavity. The mixture jets into the outer cavity through holes in the ring and flows upward along the outer wall. The mixture then flows into the inner cavity where it is exposed to radiation from the solar simulator. The particles absorb the radiation and quickly transfer heat to the fluid. The fluid heats up and the particles oxidize to carbon dioxide prior to exiting the receiver. A schematic of the SPHER can be found in previous papers [14,16]. The window of the SPHER is constructed of quartz and is attached to the water-cooled flange with silicone RTV. A graphite ring is set in between the window and the flange act as a lubricant during metal expansion. The current window shapes under research are an ellipsoidal shape and 60o spherical cap. 2.4. Data collection and instrumentation Experimental data is collected using a number of methods. Temperatures throughout the system are measured using Type-K thermocouples which are installed throughout the system. Pressure in the system is measured at the system inlet on the mass flow controllers and at the inlet of the SPHER using a pressure transducer. The pressure
366
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
drop across the inlet and inner cavity is displayed by a differential pressure gauge and is monitored visually. Temperatures, pressures, flow rates, and extinction tube detector voltage are all displayed and recorded in LabView. Particle mass loading is estimated using an extinction tube equipped with a laser and detector [16]. To obtain a size distribution for the particles produced by the CPG, scanning electron microscopy (SEM) and a diesel particulate scatterometer (DPS) are used. Images produced from SEM are analyzed using ImageJ software to obtain sizes for the particles captured in the image, from which a size distribution can be obtained [17]. These results can then be compared to results from the DPS, which is an instrument coupled with custom computer software that analyzes light scattering in a real time fluid flow [18,19]. 2.5. Design of experiment and testing Testing of the CPG at operating pressure (approximately 6 bar absolute) has been carried out using natural gas and propane, separately, as the fuel or carbon source. Natural gas is continuing to be further tested to develop operating condition correlations for particle size and mass loading. Results have shown that the upper temperature limit of the CPG heating coil (1100°C) has given the best results for mass loading with natural gas as the fuel. Therefore, the operating conditions during experimentation have been set to a pressure of around 6 bar absolute and a tube wall temperature of 1100°C. The setup of the experimentation is displayed in Table 1. The natural gas flow rate and nitrogen flow rate are varied during testing. The nitrogen flow rate varies the residence time in the reactor while the natural gas flow rate increases the mass of hydrocarbon available for pyrolysis. Table 1. Carbon particle generator experimental design conditions. Nitrogen mass flow (x10-5 kg/s)
Nitrogen Volume Flow (SLM)
Carbon mass flow (x10-6 kg/s)
Carbon Volume Flow (SCCM)
Velocity (m/s)
Residence time (s)
5.90 5.90 5.90 7.87 7.87 7.87 9.83 9.83 9.83
3 3 3 4 4 4 5 5 5
3.50 5.25 7.00 3.50 5.25 7.00 3.50 5.25 7.00
400 600 800 400 600 800 400 600 800
0.061 0.064 0.068 0.079 0.083 0.088 0.094 0.100 0.106
9.87 9.32 8.83 7.63 7.21 6.83 6.36 5.99 5.65
Since efforts have been made to obtain verifiable and relatively consistent results with the CPG, the SPHER has been on-sun tested for performance at these operating conditions. Previously, after the final stages of SPHER construction, the SPHER was tested to get quick results and, more importantly, to test the integrity of the vessel and window at operating pressure and thermal loading conditions. The latter proved to be successful; however, the performance of the SPHER was difficult to determine. This is due to issues with the CPG, including unknown mass loading and particle characteristics. Since this time, improvements have been made to the system and the CPG has been tested more extensively. Recent SPHER experimentation has been setup to first test the system without particles and then to introduce particles at known mass loading conditions. This will all take place with full power on-sun conditions to get results of how particles affect SPHER performance. 3. Computer model setup 3.1. Computer resources Computational models have been run leveraging parallel computing processes to increase efficiency. The computer simulation was performed on a Dell Precision T7600 running on 12 Intel Xeon E5-2620 threads using FLUENT v14.0.
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
367
3.2. Geometry and mesh
Fig. 2. (a) SPHER 3D CFD model; (b) SPHER CFD 2D axisymmetric mesh.
To adequately and accurately represent the associated SPHER geometry, certain modifications to the complete model had to be made. At the base of the actual model, there is a sprinkler-like inlet system. This was removed because it would have inherently caused difficulty with the meshing process due to high curvatures and the close proximities of the components involved. It was assumed that as long as the mass flow rate was kept constant with what it would have been if the sprinkler system were included, neglecting this complex piece of geometry would have little or no impact on the overall flow profile and heat transfer mechanisms other than contributing to the inlet turbulence conditions; a representative turbulence value of 2% was therefore specified as an inlet boundary condition instead. Other geometries that were external to the flow were also removed. Through observation, it is obvious that the geometry is inherently symmetric about a central axis. Employing symmetry proved to be a pragmatic approach in this case because it allowed for the effects of turbulence (which is an inherently three dimensional phenomena and is very important to the heat transfer within the system) to be incorporated with a two dimensional axisymmetric model. The use of a two dimensional axisymmetric model was a good choice, in concert, because it drastically reduced the computational time (relative to a full three dimensional case) without sacrificing the intrinsic details and insights gained from a fully three dimensional treatment. This geometry was then meshed with great care and attention given to zones where high gradients existed. This required that the mesh density and cell to cell growth rate in these zones be dramatically increased and decreased, respectively. A boundary layer mesh was also included in an effort to capture the turbulent flows that would likely occur in these high gradient zones. The resulting mesh may be seen in Fig. 2. 3.3. Boundary conditions and simulation parameters Below are the most salient inputs and boundary conditions used in the simulation. Table 2. Salient input parameters and boundary conditions used in simulation. Model
Setting
Space
Axisymmetric
Operating Conditions
Gravity on; 6 bar
Time
Steady
Viscous
Realizable k-epsilon turbulence
Radiation Model
Discrete Ordinates with 50 angular divisions and 11 wavelength bands.
368
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
Material Properties
Type
Fluid
Air
Solids
Steel, Insulation, Quartz
Boundary Conditions
Specification Surface temperature: 600 °C; based on measurements yielded during steady state laboratory conditions. Irradiation: Normal Gaussian flux distribution based on ray tracing software VEGAS. Peak flux of 1.8E6 W/m2 occurring at the aperture center.
Aperture quartz
Beam width: ~6° Surface temperature: 70 ° C; based on measurements yielded during steady state laboratory conditions.
Steel shell Insulation Mass flow inlet
Coupled 2 g/s at 25 °C
Pressure outlet
1 bar absolute
4. Results 4.1. Experimental results of receiver testing In all of the testing presented in this section, operating conditions were kept constant with a dilution air flow rate of 2 g/s, a pressure of around 6 bar absolute, and a CPG wall temperature of 1100°C (excluding the no particle case). The CPG has been tested for particle size and mass loading at the various cases listed in Table 1. The particle size comparison is represented in Fig. 5. At this point, particle size can be controlled during system since deviations in particle size remain small (~10 nm standard deviation between samples). However, this does not hold true for particle loading, which can vary in the current CPG by about 0.5 g/m3 with an observed maximum of 2 g/m3.
Average Particle Diameter [nm]
Average Diameter vs. Natural Gas Flow Rate 400 380 360 340 320 300 280 260 240 220 200
SEM 3N2 SEM 4N2 SEM 5N2 DPS 3N2 DPS 4N2 DPS 5N2
300
400
500
600
700
800
Natural Gas Flow Rate [SCCM] Fig. 3. DPS and SEM particle size comparison.
900
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
369
Outlet Temperature Comparison
Temperature (Celcius)
900 800 700 2.0 with particles
600
2.0 without particles
500
1.5 with particles
400 300 0
10
20
30
Run Time (min) Fig. 4. SPHER experimentation outlet temperature comparison.
Testing with the SPHER has been carried out separately from the CPG testing shown in Fig. 3. Data from recent testing of the lab-scale SPHER has been limited by a malfunctioning analog input module. This has led to incorrect thermocouple readouts in LabVIEW. However, the temperatures were recorded manually during testing with a handheld thermocouple processing unit. Fig. 4 displays SPHER testing outlet temperature for tests with particles (1.5 and 2.0 g/s dilution air) and without particles. As seen from the plot, the outlet temperature difference between the with and without particle cases is around 50°C. We would like to see a much larger difference than this along with an outlet temperature of over 1200°C for the case with particles. After these tests, we noticed that significant particle settling occurs around the inlet sprinkler injector, which also becomes clogged with particles. This particle injector into the receiver, along with several other system components, has since been redesigned. System improvements include adding insulation to the inner cavity of the receiver to get a better flow profile and adding ports across the receiver to measure particle loading. During this test period, we did not have a method to determine the particle concentration inside of the receiver or at the outlet. This is because the only way to measure particle loading is at the entrance of the outlet rather than at the exit of the outlet tube. As fluid flows through the outlet tube, particles could still be oxidizing, hence the reason for adding ports across the receiver. 4.2. Simulation results The flow profile, seen in Fig. 5 and Fig. 6, is based on a mass flow at the inlet of 2 g/s of air. This study has elucidated some of the expected flow patterns and has shown where one can expect zones of relatively high velocity, and equally zones which will have a proclivity to be stagnant. These dead zones (i.e. regions of slow moving fluid) will surely need to be addressed in later design modifications. The associated flow patterns also indicate that buoyancy effects are considerable and contribute to the overall formation of the stagnation zones both inside the main body of the receiver and near the quartz window. The velocity profiles associated with the no particle and particle cases are shown below with their associated vectors.
370
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
Fig. 5. (a) SPHER 2D axisymmetric velocity profile; no particles; (b) SPHER 2D axisymmetric velocity profile; with particles.
Fig. 6. (a) SPHER 2D axisymmetric velocity vectors profile; no particles; (b) SPHER 2D axisymmetric velocity vectors; with particles.
The generation of heat within the SPHER is the direct result of the incident irradiation falling on the surfaces within the receiver and the volumetric heating occurring within the absorbing medium. In order to capture the incident irradiation, the ray tracing software VEGAS was utilized with an ellipsoidal reflector representing the exact specifications of laboratory equipment used in experimentation. This boundary condition was then applied at the aperture plane, which was located in the region above the curved glass window; this area of application is reflected by the topmost blue region in Fig. 5 and Fig. 6. Although this region seems anomalous at a first pass, it is a computational method that employs a “dead zone” with solid air allowing for the irradiation boundary condition to be correctly implemented while not contributing to any constitutive equations governing the dynamics within the system. Implementation of this boundary condition at the aperture was a very involved undertaking and is outside the scope of this journal entry. Once this boundary condition was correctly implemented, the correct heat flux and source terms could be applied within the system through the Discrete Ordinates model. The associated steady state temperatures of the no particle and particle cases may be seen below.
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
371
Fig. 7. (a) SPHER axisymmetric steady state temperature profile, no particles; (b) SPHER axisymmetric steady state temperature profile, particles
The temperature fields for the no particle and particle cases were distinct and different, with steady state outlet temperatures being 784°C and 801°C, respectively. It is evident in the particle case that there is a larger volume of hot fluid in the center of the receiver and the walls are cooler relative to the no particle case where the hot fluid tends to concentrate around the insulation walls. The results yielded so far are encouraging. In regards to the no particle case, the deviation between model and experiment does not exceed 5% per results taken from Fig. 4. As it stands, there is confidence in the no particle case as the metrics analyzed between simulation and experiment agree very well. Greater deviation between experiment and simulation, however, was expected for the particle case. This greater expected deviation is due to the fact that the model has some computational shortcomings, specifically in regards to wavelength dependent scattering. To work around this issue, a method was devised using a weighted average based upon the incoming irradiation and percentage of power emitted in each band. This value was then used to approximate the band dependent scattering. While not perfect, the approximation is good and further development may be undertaken to address this issue in future simulations. 5. Conclusion The lab-scale system is undergoing testing at an operating pressure of around 6 bar absolute with various mass loading conditions and dilution air flow rates. The lab-scale SPHER is being modeled in ANSYS Fluent to compare with experimentation and to predict performance. Particle sizing has been carried out with the SEM and DPS which has shown that the CPG particle range is approximately 250 – 400 nm. Testing with the lab-scale SPHER has yielded promising results, with outlet temperatures in excess of 800°C. From the computational results, it can be seen that there are large areas of recirculation from gas buoyancy effects which are difficult to overcome in a vertically situated receiver. This is only significant on the lab scale, since the full scale SPHER will be downward facing. Therefore, modifications need to be made to the SPHER to reduce these effects. These include changing the inner cavity geometry and possibly the injection locations, along with testing at higher flow rates to increase velocity and reduce residence time within the SPHER. Based on the results presented, performance needs to be improved and the SPHER has since been modified. These modifications include allowing for particle loading measurements at various points throughout the SPHER, an improved particle injection setup, and added insulation to reduce flow recirculation and “dead zones” within the inner cavity of the receiver. Further improvements will be made based on testing with these modifications. Plans for future testing include experimentation at various air mass flow rates, mass loading, and particle size conditions,
372
L. Frederickson et al. / Energy Procedia 49 (2014) 363 – 372
which will coalesce with CFD simulations. Comparisons can then be drawn between the refined model and experimental testing in hopes of accurately predicting SPHER performance. Acknowledgements The authors gratefully acknowledge the support by the California Energy Commission's Energy Innovations Small Grant Program (Award 56626A/10-02) to build the solar simulator and solar receiver, and by the United States Department of Energy with its SunShot program under the Award # DE-EE0005800 to conduct some of the experiments. The authors acknowledge the assistance of Dr. Steve Barlow, and use of equipment at the San Diego State University Electron Microscopy Facility acquired by NSF instrumentation grant DBI-0959908. The authors would like to thank Dr. Arlon Hunt for his collaboration and insight on the project. We would like to thank all of our current and former colleagues in the SDSU Combustion and Solar Energy Laboratory who have helped tremendously on this project, including Kent Kurashima and Adnan Ansar who helped design and fabricate multiple pieces of equipment and Sofia Gomez who helped with particle sizing. We would also like to thank Mike Lester and the Machine Shop staff for their help and insight on fabrication. References [1] K. Kitzmiller and F. Miller, "Thermodynamic Cycles for a small particle heat exchanger used in concentrating solar power plants," Journal of Solar Energy Engineering, vol. 133, 2011. [2] F. Miller and A. Hunt, “Developing the Small Particle Heat Exchange Receiver for a Prototype Test,” ASME 2012 6th International Conference on Energy Sustainability, San Diego, CA, July 2012, Paper 91337. [3] A. Crocker and F. Miller, “Coupled Fluid Flow and Radiation Modeling of a Cylindrical Small Particle Solar Receiver,” ASME 2012 6th International Conference on Energy Sustainability, San Diego, CA, July, 2012, Paper 91235. [4] Steve Ruther, “Radiation Heat Transfer Simulation of a Small Particle Solar Receiver Using the Monte Carlo Method,” Master’s Thesis, Department of Mechanical Engineering, San Diego State University, CA, 2010. [5] A. Hunt, "Small Particle Heat Exchange Receivers," Lawrence Berkeley Laboratory Report, 1978. [6] M. Abdelrahman, P. Fumeaux, and P. Suter, “Study of solid-gas-suspensions used for direct absorption of concentrated solar radiation,” Solar Energy, Volume 22, Issue 1, Pages 45-48, 1979. [7] Arlon Hunt, “A New Solar Thermal Receiver Utilizing a Small Particle Heat Exchanger,” 14th Intersociety Energy Conversion Engineering Conference, Boston, Aug. 5-10, 1979 (also LBL Report 8520). [8] Arlon Hunt and David Evans, “The Design and Construction of a High-Temperature Gas Receiver Utilizing Small Particles as the Heat Exchanger (SPHER), Phase 1 Report, LBL-13755, 1981 [9] A.J. Hunt and C.T. Brown, Phase II Report “Solar Testing of the Small Particle Heat Exchanger Receiver,” February 1983. LBNL-15807 [10] R. Bertocchi, J. Karni and A. Kribus, "Experimental Evaluation of a non-isothermal high temperature solar particle receiver," vol. 29, pp. 687-700, 2004. [11] H. H. Klein, R. Rubin and J. Karni, "Experimental Evaluation of Particle Consumption in a Particle Seeded Solar Receiver," Journal of Solar Energy Engineering, vol. 130, no. 1, 2008. [12] G. Maag, G. Zanganeh and A. Steinfeld, "Solar thermal cracking of methane in a particle flow reactor for the co-production of hydrogen and carbon," Internation Journal of Hydrogen Energy, vol. 34, no. 18, pp. 7676-7685, 2009. [13] Fletcher Miller, An Experimental and Theoretical Study of Heat Transfer in a Gas-Particle Flow Under Direct Radiant Heating, Ph.D. Dissertation, Department of Mechanical Engineering, University of California, Berkeley, 1988. [14] Kyle Kitzmiller, “Introduction to Concentrating Solar Power and the SPHER Experiment at SDSU,” Master’s Thesis, Department of Mechanical Engineering, San Diego State University, CA, 2013. [15] K. Kitzmiller, F. Miller, and L. Frederickson, “Design, Construction, and Preliminary Testing of a Lab-Scale Small Particle Solar Receiver” in Proceedings of SolarPACES, 2012, Marrakech, Morocco, 2012. [16] L. Frederickson, F. Miller, and K. Kitzmiller, “Carbon Particle Generation and Preliminary Small Particle Heat Exchange Receiver Lab Scale Testing,” ASME 2013 7th International Conference on Energy Sustainability, Minneapolis, MN, July 2013, Paper 18215. [17] Emily Mitchell, “Generation and Characterization of Carbon Nanoparticles”, Senior Thesis, Department of Physics, San Diego State University, CA, 2012. [18] Paul Schroeder, “Measurements and Analysis of Light Scattered by Small Carbon Particles”, Senior Thesis, Department of Physics, San Diego State University, CA, 2010. [19] A. Hunt, I. Shepard and J. Storey, "Diesel Particle Scatterometer," Lawrence Berkeley Laboratories, 2001. [20] Fernández, P. “Numerical Modeling, Simulation and Design Optimization of a Small Particle Solar Receiver for Concentrated Solar Power Plants”, Proyecto Fin de Carrera, University of Valladolid, Spain, 2013. [21] Crocker, A. “Coupled Fluid Flow and Radiation Modeling of a Small Particle Solar Receiver”, Master’s Thesis, San Diego State University, 2012.
