Optimized volumetric solar receiver: Thermal performance prediction and experimental validation

Accepted Manuscript Optimized volumetric solar receiver: Thermal performance prediction and experimental validation Raffaele Capuano, Thomas Fend, Hannes Stadler, Bernhard Hoffschmidt, Robert Pitz-Paal PII:

S0960-1481(17)30699-7

DOI:

10.1016/j.renene.2017.07.071

Reference:

RENE 9046

To appear in:

Renewable Energy

Received Date: 23 April 2017 Revised Date:

12 July 2017

Accepted Date: 14 July 2017

Please cite this article as: Capuano R, Fend T, Stadler H, Hoffschmidt B, Pitz-Paal R, Optimized volumetric solar receiver: Thermal performance prediction and experimental validation, Renewable Energy (2017), doi: 10.1016/j.renene.2017.07.071. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

3

Raffaele Capuano *ab, Thomas Fend c, Hannes Stadler a, Bernhard Hoffschmidt c, Robert Pitz-Paal c.

4

*Corresponding author.

5

a

German Aerospace Center DLR, Institute of solar Research, Professor-Rehm-Str. 1, Juelich, Germany.

6

b

Flemish Institute for Technological Research VITO NV, Thor Park, 3600 Genk, Belgium.

7

c

German Aerospace Center DLR, Institute of solar Research, Linder Hoehe, Cologne, Germany.

RI PT

2

Optimized volumetric solar receiver: thermal performance prediction and experimental validation.

1

SC

8 9

Abstract

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

In the last decade, different absorber geometries, such as foams and honeycombs, have been tested at laboratory or industrial scale in order to achieve high performance in the conversion of the solar radiation into usable heat, with the current state-of-the-art, the HiTRec-II monolithic honeycomb, characterized by a squarechannel section and made out of siliconized silicon carbide (SiSiC). Such geometry has been so far the best compromise for large-scale application thanks to the low production costs, easy manufacturability through extrusion procedure and overall acceptable performance. However, it does present some drawbacks, since the geometry is not able to contain the radiative heat losses, especially from the front surface. An optimized absorber geometry, capable to reduce overall thermal losses, is presented in this work, being able to increase the final thermal efficiency of more than 12% compared to the current state-of-the-art and showing the presence of the so-called volumetric effect, since the outlet fluid temperature is higher than the solid inlet temperature. A test sample has been produced for laboratory-scale experiments, in the form of a 3:1 scaled prototype through additive manufacturing procedure, using a titanium-aluminium alloy (Ti6Al4V) and the experimental results were in good agreement with the numerical calculation, with a deviation of 3%, computed considering a 3:1 Ti6Al4V scaled-up sample. As the manufacturing technology will progress and become cheaper in the near future, it will be possible to improve the overall Solar Power Tower (SPT) plants performance using advanced micro-geometry for open volumetric receivers.

AC C

EP

TE D

M AN U

10

Keywords: Concentrated Solar Power (CSP), volumetric solar receiver, porous microstructure, CFD numerical simulation, experimental validation.

1

ACCEPTED MANUSCRIPT 42

Nomenclature Latin symbols h

-2

-1

Heat transfer coefficient [W∙m ∙K ] 2

-3

Famb

Ambient view factor

Av

Specific surface area [m ∙m ]

n

Directional unit vector

p

Pressure [Pa]

Ḣ

Energy gained by the fluid [W]

-1

Velocity [m∙s ]

Qn

Incoming radiative power [W]

T

Temperature [K]

Ai

Projected inlet surface [m ]

cp

Specific heat capacity [J∙kg ∙K ]

ṁ

Mass flow rate [kg∙s ]

qr

Radiative heat density [W∙m ]

L

Sample length [m]

K1

Permeability coefficient [m ]

K2

Inertial coefficient [m]

εi

Solid surface emissivity

σB

Stefan-Boltzmann constant [W∙m ∙K ]

η

Thermal efficiency

Δ

Difference

i

Inlet

o

Outlet

exp

Experimental

-1

-2

-1

-1

k

Thermal conductivity [W∙ m ∙K ]

I

Irradiation [W∙m ]

-2

2

-1

2

SC

-1

RI PT

u

ε

Porosity -1

ξ

Extinction coefficient [m ]

µ

Dynamic viscosity [kg∙m ∙s ]

ρ

-3

-1 -1

Density [kg∙m ]

f

Fluid phase

s

Solid phase

m

Mutual

ext

External

43

TE D

Subscripts

M AN U

Greek symbols

-2

-4

1. Introduction

45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Open volumetric receivers represent the core of a particular type of solar power tower plants (SPT) where the reflected sunlight, coming from the heliostat field, is converted into thermal energy for further industrial uses. The volumetric receiver is characterized by a porous solid (absorber) crossed by atmospheric air that is heated by contact with the inner walls. In such technology, the volume of the receiver mainly has a thermal function. While with linear receivers the wall temperature is always higher than the fluid temperature, the use of volumetric receivers could allow the presence of the so-called volumetric effect. In this case, the outlet temperature of the air can reach values of temperature higher than the ones of the solid front surface. Several types of volumetric absorbers have been built and tested in the past decades, such as wire mesh, fibres, packed bed particles, honeycombs and foams, using both metals and ceramics as manufacturing materials. The entire variety of different structures can be characterized by parameters and coefficients, defined as effective properties, like the porosity (ε), the extinction coefficient (ξ) and the volumetric heat transfer coefficient (hAv). The first is the ratio between the void volume and the total volume of the porous absorber, while ξ defines the attenuation of the incoming sunlight in the absorber inner volume. The volumetric heat transfer coefficient is the product between the convective heat transfer coefficient (h), defined as the ratio of the heat flux and the thermodynamic driving force (that in this case is the temperature difference between the atmospheric flow and the inner surfaces of the absorber), and the specific surface area

AC C

EP

44

2

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

(Av), which is the index of the exchange surface in porous materials and is defined as the ratio of the wet surface and the corresponding volume. Volumetric receivers showed promising results in previous scientific works; however, they also showed a wide margin of improvement, since the volumetric effect has never been achieved. The shape of the solar absorber plays an important role in the utilization of the incoming sunlight, particularly at the inlet section where the radiation hits the volume. For this reason, the optimization of the porous structure to the different thermodynamic needs is a key aspect to improve the efficiency of the energy conversion. The first example reported of a ceramic volumetric absorber is the open cavity receiver developed by Sander Associates Inc. [1], in which the prototype, able to produce 250 kWth, was characterized by silicon carbide honeycomb structure and designed for 1100 °C air outlet temperature. A new concept of volumetric absorber, consisting of a mesh of thin metallic wires used to heat up the air flow was presented by Fricker et al. [2]. The prototype has been tested in a parabolic dish, producing hot air at more than 846 °C; however, the geometric design caused distortion in the structures while operating at such high temperatures, causing the formation of hot spots and hindering the achievement of the volumetric effect. Subsequently, a new absorber using a porous ceramic material was designed by the University of Colorado and produced as prototype by Sandia National Laboratories [3]. The absorber structure was a foam made out of aluminium oxide, coated with black paint in order to increase the solar absorption. During the test it delivered an average outlet air temperature at highest loads of 730 °C with a material temperature of 1350 °C, achieving a lower thermal efficiency compared to the calculated one. However, since it was the first attempt with ceramic foam, the results have been considered positive taking also into account that there was no degradation of the ceramic material. Following the development of volumetric receivers, a metallic wire-mesh based volumetric receiver was designed at the New Mexico State University, consisting of a multi-layered cylindrical structure [4] with whom a maximum mean air outlet temperature of 563 °C was reached during experimental sessions. This value resulted to be much lower than the one previously calculated in numerical analysis and this behaviour was due to the unbalanced air flow and the presence of severe hot spots, eventually leading to the end of the project. Hoffschmidt et al. [5] tested several ceramic material structures at the DLR solar furnace. One of them was the High Temperature Receiver (HiTRec), an extruded honeycomb structure made out of recrystallized silicon carbide. The absorber cross section was characterized by parallel square channels of 2 mm width, 0.8 mm wall thickness and 50 mm depth. The entire receiver surface, as foreseen for a 200 kW test on the PSA, was composed of 37 modules consisting of a hexagonal absorber structure and a siliconized silicon carbide (SiSiC) cup placed on a stainless steel supporting structure. The experiments showed good agreement with the calculated performance with an outlet air temperature of 800 °C. No hotspots were observed and eventually the receiver showed reduced start-up times and easy operability and maintenance. However, during the test the stainless steel structure was deformed, which made it not acceptable for industry-scale production. In 1998 DLR, Ciemat and Inabensa started the development of the evolution of the aforementioned HiTRec, realizing a new concept called HiTRec-II [6]. The receiver was characterized by 32 cups in hexagonal ceramic modules, both of the same material as the HiTRec I. The supporting structure was made of steel capable of very high working temperature and presenting a similar expansion coefficient of SiSiC. The improved receiver was tested for 155 hours without showing any sign of damage, operating with an outlet air temperature between 700 °C and 800 °C. The HiTRec II represents the current state-of-the-art for what concerns open volumetric air receivers in precommercial industrial scale. Nonetheless, several innovative concepts have been tested on experimental scale in order to improve the thermal performance of the absorber and to achieve the volumetric effect. Fend et al. investigated the performance of several absorber samples through experimental analysis. The complete set of porous structures included three different foam samples with different cell density and three different materials (SSiC, cordierite and clay-bound silicon carbide), SiC fiber mesh material and different configurations of improved metallic structures based on the Catrec technology (different surface densities and one sample modified introducing small chokes in each channel). Furthermore, a structure characterized by the

AC C

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110

3

ACCEPTED MANUSCRIPT combination of silicon carbide honeycomb with parallel channels and fiber mesh was also considered in the study. The experimental results showed high and promising performances for the foam, for the metallic structure with the largest surface density and for the combined honeycomb/fiber mesh structure. With those structures, more than 800 °C air outlet temperature has been achieved at highest radiative loads. In 2013, an advanced geometry based on the HiTRec II technology was investigated by Smirnova et al. [8]. The geometry was characterized by a parallel channel honeycomb with square cross section openings of 1.4 mm length and a wall thickness of 0.4 mm. Those dimensions characterized the lowest limit of the manufacturing process. The advanced sample was numerically analysed, showing an improvement of about 5% in terms of thermal performance. However, when the sample was tested at the DLR experimental facility, it showed a lack of agreement with the previous theoretical evaluation, mainly due to the occurrence of hot spots and inhomogeneous flow. A preliminary optimization procedure of the solar absorber structure for volumetric solar receivers has been presented in Capuano et al.[9]. In this work, an extensive parametric numerical analysis has been carried out, where effective quantities such as porosity, volumetric heat transfer coefficient and radiation extinction have been varied in order to obtain an optimized selection for future design processes. A combination of high porosity, high volumetric heat transfer coefficient and increasing extinction within the absorber volume resulted to be the very important to improve the overall conjugate heat transfer between the porous solid and the air flow in volumetric solar receivers, drastically reducing the radiative heat losses, thus obtaining high value of thermal performance and the achievement of the volumetric effect. Pfeifer et al [10] introduced a new concept of volumetric receiver using ultra-high temperature ceramics. Their concept combines the behaviour of spectral selectivity of selective coatings and the volumetric effect observed with refractory porous ceramics, showing improved performance compared to conventional porous ceramic foams with same shape characteristics. In the work of Shuja and Yilbas [11], different designs of volumetric solar receivers are presented. The structures are characterized by different configurations of absorbing blocks, arranged to obtain different behaviours concerning radiation absorption and pressure drops. The energy transfer is increased for certain arrangements in which pressure drops are minimized and absorption is distributed within the structure uniformly. Pabst et al. [12] also obtained very good results with the use of an advanced cellular metallic honeycomb, consisting of winded pairs of flat corrugated metal foils, widely used in the automotive industry for the aftertreatment of exhaust gases. As an outcome, air outlet with more than 800 °C has been reached in a 500 kW experiment, supporting the improved performance by theoretical calculations. In the present work, an innovative micro-geometry is presented, based on the numerical outcome of the work of the same author previously published [9]. The structure presents a particular pin-shaped inlet zone, coupled with an inner staggered-honeycomb geometry, allowing a better distribution of the incoming solar radiation, reducing the inlet radiative losses and optimizing the overall heat transfer process . The thermal performance of the new geometry has been predicted using a three-dimensional CFD numerical approach, in which a unit element of such geometry is used as control volume. The computational simulation has been carried out taking into account conjugate heat transfer between the irradiated porous solid and the ambient air flowing through the structure. Thus, the predicted results have been validated using a 3:1 scaled up demonstrator in small-scale laboratory experiments, showing a good agreement between each other. Improved thermal performance have been obtained compared to the current state-of-the-art, eventually achieving the volumetric effect.

153

2. Methods and materials

154 155 156 157 158

The numerical results of the new geometry analysis have been compared to the ones characterizing the current state-of-the-art and the advanced honeycomb presented in the work of Smirnova et al. [8]. Afterwards, thermal and flow performances have been experimentally evaluated with the use of a solar simulator and an in-house developed experimental setup for the flow analysis. Thus, the outcome of the experimental campaign has been finally used as a validation of the previously predicted numerical results.

AC C

EP

TE D

M AN U

SC

RI PT

111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152

4

ACCEPTED MANUSCRIPT 159

3.1 Numerical model

160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211

The discrete numerical model presented in this study is characterized by a definite representation of the threedimensional solid geometry. Particularly, due to the complexity of such structures, unit elements and symmetry boundaries are used to represent the entire porous medium. For the numerical calculation of the velocity and pressure fields the Navier-Stokes equations have been used in the model, according to the following equations: − +  +    =  

RI PT

(1)

where p is the pressure, μ is the dynamic viscosity, u is the velocity and ρ is the density of the fluid and:   = 0; = , 

with Tf as the temperature of the fluid.

(2)

SC

The energy exchange has been treated using Fourier´s law for the solid and the fluid phases, through the following equations:

M AN U

  =  ∙   

(3)

where cp is the specific heat capacity at constant pressure of the fluid and kf is the thermal conductivity of the fluid; −  = 

(4)

where ks is the thermal conductivity of the solid and qr is the radiative heat density;  =   −  !

(5)

EP

TE D

where εi is the superficial emissivity of the solid, σB is the Stefan-Boltzmann constant and I is the direct radiative intensity source. The direct irradiation source I is set through a built-in ray-tracer algorithm, taking into account the setup of the HLS solar simulator and it has been distributed on different vectors on planes XZ and YZ, as shown in Fig. 1 and Fig. 2. Radiative heat transfer is treated by using the direct area integration method through view factor calculations. Thus, the irradiation I can be defined as follows: !  = " + #$"%  $"% + &'(

(6)

AC C

where Im represents the mutual irradiation coming from other volume boundaries; Famb is the ambient view factor, whose value is equal to the fraction of the field of view that is not covered by other boundaries. Therefore, by definition, the correlation 0 ≤ Famb ≤ 1 is valid for all points. Tamb is the assumed far away ambient temperature in the directions included in Famb, while Iext represents the sum of the products, for each external source, between their view factors and the corresponding source radiosity. The latter is defined as the sum of the reflected and emitted radiation that leaves the generic surface taking part in the heat transfer process as absorbing/emitting medium. The radiation is coupled to the heat conduction equation and the following boundary condition is set on the solid surfaces: )−  =   −  ! + ) 

(7)

where n is the directional unit vector and εi is the superficial emissivity of the solid surface. External walls are set as non-absorbing surfaces, while for the fluid phase fixed values of the mass flow rate and the ambient temperature have been considered at the inlet.

5

ACCEPTED MANUSCRIPT 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231

The key parameter for performance evaluation of the analyzed structures is the thermal efficiency, defined as the ratio between the power transmitted to the fluid and the incoming radiative power, according the following equation:

232

3.2 Experimental setups

233 234 235

The experimental campaign focused on the analysis of the thermal performance of the newly designed absorber sample and its flow characteristics. The experiments have been performed using two different experimental setups that will be presented in the next sections.

236

3.2.1 Thermal efficiency analysis – setup

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258

The High Flux Solar Simulator (HLS), coupled with a tubular test-bed, has been used to investigate the thermal performance characteristics of the absorber sample. With this configuration, it is possible to simulate boundary conditions similar to real-case applications in solar power plants for a receiver sample. The setup is -2 characterized by a set of 10 short-arc Xenon lamps with ellipsoid reflectors, providing up to 4.5 MW∙m of concentrated irradiation. The latter is focused on the investigated sample, placed in a test-bed at a distance of ca. 150 cm from the lamps [7]. Only 4 lamps have been used for the performance evaluation, in order to provide a mostly homogeneous irradiation on the small front surface of the demonstrator. In Fig. 3, the setup is shown: short-arc Xenon lamps are displayed on the right side of the picture, while on the left side there is the test-bed used for the experiments. In Fig. 4, a more detailed picture of the test-bed front zone is shown. The housing of the absorber sample is surrounded by the insulating material and a liquid circuit used to cool down the part of the structure nearby the sample when the lamps are operating. The concentrated radiation coming from the lamps (Iexp), hit the inlet surface of the absorber sample that is heated up. The blower is used to provide the pressure difference that generates the air flow in the circuit, as described in the flow-scheme in Fig. 5. Cold ambient air flowing through the sample and then in the circuit, is heated up and its temperature is measured by three thermocouples with an accuracy of ± 1%, as described in the work of Kalt et al. [13]. The thermocouples are placed in the positions TA and TB. A counter flow air/water heat exchanger has been used in order to protect auxiliary components, like the blower and the mass flow meter, from overheating. Additionally, surface temperature of the absorber has been measured with the use of an infrared camera while the radiation flux on the aperture of the sample has been measured using the Flux And Temperature Measurement System (FATMES) with an accuracy of ± 5% [13].

* = +,- ⁄.

(8)

. =  ∙ 0 where Ai is the projected surface at the inlet. ΔḢ can be calculated according to the following equation:

SC

+,- = 1- ℎ,3 − ℎ, 

RI PT

where ΔḢ is the energy gained by the fluid during the heat transfer process and Qr is the incoming radiative power and is equal to: (9)

(10)

(11)

AC C

EP

TE D

* = 1- ,3 ∙ ,3 − , ∙ ,  ⁄ ∙ 0

M AN U

where hf,o and hf,i are the fluid enthalpy evaluated at the outlet and inlet sections respectively. Thus, Eq. 8 becomes:

6

ACCEPTED MANUSCRIPT 259 260 261 262 263 264 265 266 267

For the evaluation of the thermal performance of the sample, the key parameter is the thermal efficiency, calculated according to Eq. 8. The irradiation power is calculated as follows:

268

3.2.2 Pressure drop analysis – setup

269 270 271 272 273 274 275 276 277 278 279 280 281 282 283

The analysis of the pressure drop inside the porous sample has been performed adapting the experimental setup according to Reutter et al. [14] and shown in Fig. 6. The setup is characterized by a tubular structure where a blower is used to generate the air stream inside the circuit (1). The tube is long enough to let the air flow be fully developed. Then, a standardized orifice (diameter: 22.7 mm) is placed perpendicular to the flow direction (2) and is used for the evaluation of the mass flow rate. The latter is calculated taking into account the pressure difference measured with an accuracy of ± 1% by two sensors that are placed before and behind the orifice. The pressure drop across the sample has been measured, together with the air temperature, using two additional sensors placed right next to the porous sample and presenting an accuracy of ± 1% [14]. For the comparison and the validation of the results obtained through numerical simulation, the specific pressure drop (Δp/L) will be considered. It is defined through the following equation:

284

4. Results

285 286 287 288 289

The thermal performance of the new absorber structure, in its original design dimensions and in the demonstrator form, has been evaluated through a discrete-based simulation and compared with the current state-of-the-art HiTRec-II honeycomb absorber and an improved honeycomb geometry [15]. Eventually, the outcome of the experimental campaign will be introduced, showing a final validation concerning thermal performance and flow behaviour.

290

4.1 Numerical test and comparison

291

4.1.1 Original design

292 293 294 295 296 297 298 299 300 301 302 303 304

Single unit elements and corresponding fluid zones have been used as control volume for this set of simulations. The unit element of the optimized geometry is shown in Fig. 7. It presents a channel width of 1.1 mm and a wall thickness of 0.2 mm. Moreover, it is characterized by a graded open porosity (ε) with the highest value reached in the first section of the pin zone (0.97), down to a minimum value of 0.71 for the 2 -3 honeycomb back section. The specific surface area (Av) ranges from the lowest value of 960 m ∙m in the first 2 -3 section of the pin zone, up to 2860 m ∙m in the honeycomb back zone. The unit elements of the structures considered for the numerical comparison are the HiTRec-II and an improved honeycomb geometry and their unit element are displayed in Fig. 8. The state-of-the-art is characterized by channels 2.0 mm wide, a wall thickness of 0.8 mm and 50.0 mm depth, resulting in ε = 0.51 2 -3 and Av = 1020 m ∙m . The improved honeycomb has a similar shape and depth compared to the HiTRec-II but 2 -3 the channels are 1.4 mm wide and the wall thickness is 0.4 mm, so that ε = 0.60 and Av = 2560 m ∙m . A tetrahedral mesh has been adopted for the discretization of the solid volume, as reported in Fig. 9. The mesh 5 consists of 8.5∙10 elements, presenting an average growth rate of the elements of 1.7 and a maximum growth

. = &' ∙ 0

(12)

M AN U

SC

RI PT

where Iexp is the homogeneous value of the irradiation source, coming from the Xenon lamps, while Ai is the aperture area of the cylindrical absorber sample. The power transmitted to the fluid is calculated according to Eq. 10. In the latter, the outlet air temperature is the average of the hot air temperatures detected by the thermocouples at position TA.

+⁄4 =  − 3 ⁄23 4

(13)

AC C

EP

TE D

where po and pi represents the air pressure evaluated at the outlet and at the inlet of the porous sample respectively, and L is the sample length.

7

ACCEPTED MANUSCRIPT

RI PT

-2

SC

5

In = 6.5∙10 W∙m ; -6 -1 -1 ṁ = 1.1∙10 kg∙s (corresponding to: U0 = 0.5 m∙s ); Tf,i = 318.3 K; 5 pf,o = pamb = 1.0∙10 Pa.

M AN U

The direct irradiation source In is set according to the setup of the HLS solar simulator of DLR, as previously described in section 3.1. Siliconized silicon carbide (SiSiC) has been considered for the different geometries and temperature dependent material properties have been implemented in the model according to Munro et al. [17]. Humid air with 60% relative humidity has been considered for the fluid volume. Mean temperature profiles of each case along the flow direction z are displayed in Fig. 11. They are the result of the interaction between the porous solid structure and the flowing air under conjugate radiation, convection and conduction. As the radiation hit the structure, the part which is not reflected and emitted from the inlet solid surface penetrates into it, generating a conductive heat flux and heating up the volume and the outer surfaces of the solid. This energy is then transmitted through convection to the fluid which is heated up to the highest value reached at the outlet. The higher the porosity, the higher the amount of radiation that enters the porous structure; thus, the thermal performance of the absorber is enhanced. On the other hand, the thermal performance is affected by higher optical losses from the inlet solid zone in case of low porosity structures. In fact, reflection and emission losses from the inlet are proportional to the amount of surface directly hit by solar radiation and to the temperature of the solid surface to the fourth power respectively. Moreover, when low porosity is joined by a low specific surface area, the convective heat exchange is also affected, causing a bigger temperature difference with thermal equilibrium achieved later in the structure between the solid and fluid phases. Material thermal conductivity also plays a role in the overall heat transfer process, as it directly affects the conduction of the heat into the solid. Whenever the volumetric effect is reached and the temperature of the solid phase is characterized by a monotone rising trend, a low thermal conductivity can be beneficial in the conjugate heat transfer process, since it slows down the transport of heat by conduction to the front. Because of this, the rise in temperature toward the front is prevented, keeping a low inlet solid temperature and hence improving the thermal performance [18]. The HiTRec-II honeycomb geometry is characterized by the lowest value of porosity and the highest initial solid temperature (1234 K). It also has the lowest specific surface area, therefore, thermal equilibrium is reached nearly at the end of the depth. The thermal conductivity of SiSiC also has an influence on this behaviour, as its value drops down as a consequence of the high initial solid temperature, increasing the thermal inertia in the initial section. The combination of all those aspects leads to the lowest value of the outlet air temperature (1015 K) and thermal efficiency (ηHiTRec-II: 0.70) of the examined structures. Performance raises when we consider the advanced honeycomb, as it presents a higher porosity and specific surface area compared to the current state-of-the-art (ηadvanced honeycomb = 0.74).

TE D

320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352

• • • •

EP

316 317 318 319

rate of 8.5. The growth rate factor is defined as the ratio between the volume of a single elements and the volume of the smallest elements characterizing the mesh. A combination of tetrahedral and hexahedral elements has been used for the discretization of the fluid volume, as reported in Fig. 10. Hexahedral elements have been used for the meshing of the fluid layers in the immediate vicinity of the bounding solid surfaces. Here the effects of viscosity are significant and a strong gradient of the velocity is present. Thus, since hexahedral elements provide more precise solutions and can be better adapted to the shape of the interface, 6 they are preferable compared to tetrahedral elements. The mesh consists of 1.6∙10 tetrahedral elements, 3 characterized by an average growth rate of 1.4 and a maximum growth rate of 51.1, and 7.3∙10 hexahedral elements, presenting an average growth rate of 1.2 and a maximum growth rate of 10.7. For this set of simulations, initial conditions are set according to the lay-out parameters of the Solar Tower in Jülich [16], as reported below:

AC C

305 306 307 308 309 310 311 312 313 314 315

8

ACCEPTED MANUSCRIPT The new geometry shows the best performance and also the presence of the volumetric effect, since the front solid temperature (842 K) is lower than the outlet air temperature (1203 K). Thermal equilibrium is reached in correspondence of 12.0 mm while the entire volume depth is 20.0 mm. Thanks to its graded porosity, the inlet radiative losses are minimized, allowing a good penetration of the incoming radiation. This, coupled with high values of porosity and specific surface area leads to the highest thermal efficiency achieved in this set of simulations (ηnew geometry = 0.90).

359

4.1.2 Demonstrator design

360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401

Electron Beam Selective Melting (EBSM) technique has been adopted for the prototype production. EBM foresees the use of a metal powder for the manufacturing of 3D parts that is selectively melted layer-by-layer by an electron beam under vacuum atmosphere [19]. However, the geometry numerically designed resulted to be too fine even for this process. Thus, an isotropic 3:1 scaled up version of the original geometry has been realized, taking into account that current technology is under development and finer structures are feasible in the mid-term future. Control volumes of the solid and fluid phases for the numerical evaluation of the thermal performance of the demonstrator have the same geometry of the simulation case presented before. However, in this particular case the dimensions are scaled by a factor of 3. The geometry has the same open porosity (ε) as the original design, as the upscaling does not affect its ratio, 2 -3 but it does affect the specific surface area (Av). In fact, this parameter ranges from 318 m ∙m in the entrance 2 -3 section of the pin inlet zone, up to 870 m ∙m reached in the back honeycomb zone. In Tab. 1, the characteristic geometric values of the demonstrator, the new geometry in the original design and the other structures used for the numerical comparison are reported. All the simulations herein presented have been carried out with the same initial and boundary conditions and material properties as the previous cases. Only the value of the mass flow has been adapted to the enlarged geometry, corresponding to the same value of incoming fluid velocity. The temperature profiles along the flow direction z are shown in Fig. 12, comparing the demonstrator geometry with the other structures. The new structure in the demonstrator form shows the presence of the volumetric effect, since the solid inlet temperature (1012 K) is lower than the fluid outlet temperature (1144 K), corresponding to the a thermal efficiency of 0.84. The solid temperature profile is characterized by a discontinuous trend: in the pin section (0.0 mm ≤ z ≤ 30.0 mm), the temperature decreases firstly, from a value at the inlet surface of 1012 K, down to a value of ca. 950 K at 10.0 mm. Then it rises again up to the peak value reached in correspondence of the end of the pin zone (ca. 1192 K). After that, the temperature decreases again down to the final value reached at the end of the structure (1144 K). On the other hand, the fluid temperature profile presents a continuous increase. The solid temperature profile is influenced by the porosity, the specific surface area and the corresponding radiation extinction of the demonstrator. The high porosity at the front is helpful for the distribution of the radiation into the porous volume, as the low extinction of the incoming sunlight in the inlet pin section reduces the initial optical losses. However, the demonstrator is also characterized by the lowest values of Av among the geometries due to the upscaling, affecting the convective heat exchange. In particular, the first section of the 2 -3 pin zone has the lowest value of Av (318 m ∙m ). In this part of the structure, the heat gained by the radiation which is then transmitted through conduction is larger than the one transmitted to the fluid through convection. For this reason, the solid temperature pattern shows a negative trend from the beginning to ca. 10.0 mm. Furthermore, in this section also the temperature of the fluid is influenced, as it is characterized by a slight increase. After this depth, since the cross section of the pins increases and hence also the value of Av becomes larger, the convective heat transfer is enhanced, resulting in a sharper increase of both solid and fluid temperatures. A peak characterizes the solid temperature profile at ca. 30.0 mm, where the honeycomb zone starts. Here, the porosity decreases and more radiation is absorbed due to the staggering at ca. 35.0 mm. Moreover, the specific surface area in the honeycomb section, even though it is larger than the initial pin 2 -3 section, is still characterized by a low value (870 m ∙m ). This leads once again to a condition where the heat gained through radiation and then transmitted through conduction is higher than the one transferred to the

AC C

EP

TE D

M AN U

SC

RI PT

353 354 355 356 357 358

9

ACCEPTED MANUSCRIPT fluid through convection, causing the peak of the solid temperature and a lower increase in the fluid temperature.

404

4.2 Experimental validation

405 406 407 408 409 410 411 412

In Fig. 13, the demonstrator sample used for the experiments is reported. It is characterized by a cylindrical outer shape, with a diameter of 70.0 mm and 60.0 mm depth. For its realization, titanium-aluminium alloy (Ti6Al4V) has been used, due to material availability constraints of the manufacturer. The variation of key parameters, such as the outlet air temperature and the corresponding thermal efficiency for different values of the ratio “power on aperture/mass flow rate” (P/ṁ) will be compared with the outcome of detailed simulations. The analysis of the pressure drops will be also displayed and the results compared with the ones obtained through the discrete numerical analysis and further values regarding the current state-ofthe-art.

413

4.2.1 Thermal efficiency analysis

414 415 416

The thermal efficiency measurements have been performed for different values of the air mass flow rate, keeping constant the power on aperture. The corresponding numerical simulations have been carried out taking into account the experimental conditions during the tests with the HLS, as stated below:

SC

5

-2

3

M AN U

Iexp = 7.6∙10 W∙m (corresponding power on aperture: 2.95∙10 W); -3 -1 ṁ = (33.2; 11.4; 8.3; 6.0; 5.2; 4.4; 4.1)∙10 kg∙s ; Tf,i = 309.7 K; 5 pf,o = pamb = 1.0∙10 Pa.

TE D

Furthermore, the thermal performances of HiTRec-II, calculated with the discrete numerical model in the same environmental conditions, are displayed for further comparison. Figure 14 shows the variation of the measured outlet fluid temperature for different values of the ratio “power on aperture/mass flow rate” (P/ṁ) and corresponding numerical results are reported for comparison. They are characterized by blue and red dots respectively, while the dark-grey dashed line represents the numerical results of the state-of-the-art. Vertical error bars for the experimental results take into account the accuracy of the thermocouples and the standard deviation of the results for each operational point, corresponding to a variation on the temperature value of 2%. Horizontal error bars consider a cumulated error of ± 5% on the P/ṁ values, taking into account the accuracy of the FATMES system used for the evaluation of the power-on-aperture and a negligible uncertainty on the evaluation of the mass flow rate [13]. The highest mass flow rate and hence the lowest ratio P/ṁ leads to the lowest value of the measured outlet fluid temperature in all the cases displayed on the chart. As the ratio increases, characterizing lower mass flow -1 rates, the outlet temperature rises up to the highest value, obtained for P/ṁ = 705 kJ∙kg , corresponding to -3 -1 ṁ = 4.1∙10 kg∙s . The correlation can be represented by a linear trend. Highest values of outlet fluid temperature are obtained with the demonstrator compared to the ones of the HiTRec-II, for each operational point. This is due to the lower radiative losses of the new geometry compared to the losses characterizing the state-of-the-art geometry. Furthermore, the numerical results of the demonstrator present a good agreement -1 with the measurements after the value of P/ṁ = 350 kJ∙kg . Before this point, their values are outside the error limit of the measurements. This condition is caused by a transient effect of the test-bed, which alters the measurements in correspondence of high mass flow rate and thus of low values of the ratio P/ṁ [7]. The trend of the corresponding thermal efficiency, calculated according to Eq. 8, is reported in Fig. 15. The propagation error has been calculated taking into account the accuracy of the instruments involved (thermocouples, mass flow and power on aperture measurement systems) and the standard deviation of the measured values of outlet fluid temperature. The variation of experimental results has been quantified in ± 5% on the calculated efficiency and the corresponding error bars are reported for the experimental results. The numerical trends of the demonstrator and the HiTRec-II are both characterized by a continuous decreasing trend, with the highest value obtained in correspondence of the lowest value of the ratio P/ṁ and thus the

EP

421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448

• • • •

AC C

417 418 419 420

RI PT

402 403

10

ACCEPTED MANUSCRIPT highest value of mass flow rate. In those cases the temperature of the solid is low and hence the re-emission of the radiation, representing the highest contribution of thermal losses, is negligible [21]. However, since the main aim of the volumetric absorber is to obtain a high temperature fluid, this condition is not applicable. Thus, decreasing the mass flow rate, the system is capable to produce a high temperature air flow. As a consequence, the temperature of the solid rises, as well as the thermal losses, affecting the value of thermal efficiency. The comparison with the HiTRec-II single channel geometry shows an overall better behaviour of the demonstrator, with an increased efficiency up to 13% thanks to the optimized inlet design that allows the minimization of radiative losses. Also in this case, the numerical results of the demonstrator present a good agreement with the measurements, characterized by the blue dots, after the value of -1 P/ṁ = 350 kJ∙kg , due to the transient effect of the test-bed on the measurements with low values of the ratio P/ṁ [7]

460

4.2.2 Pressure drop analysis

461 462 463 464 465 466 467

The fluid pressure at the outlet has been measured for the demonstrator geometry using the experimental setup shown in section 3.2. The results are shown in the chart of Fig. 16 and they have been used for comparison with the corresponding numerical results obtained through the discrete numerical simulation. A further comparison has been made with the results regarding the HiTRec-II, obtained through detailed numerical analysis. Measurements have been performed at constant temperature for different values of incoming mass flow rate, as reported below:

483 484 485 486 487 488 489 490

SC

EP

TE D

The specific pressure drop (Δp/L) has been calculated according to Eq. 31. Error bars for the experimental results show a variation of ± 3%, taking into account the accuracy of the instruments for the measurement of the outlet pressure and the standard deviation for each operational point. Experimental and corresponding numerical results are represented by blue and red dots respectively, while the dark-grey dashed line represents the results of the state-of-the-art. -3 -3 The measured specific pressure drop ranges between 660 N∙m and 1670 N∙m , corresponding to a total pressure drop of 20 Pa and 50 Pa, respectively. The numerical results calculated under the same experimental conditions show a good agreement with the experiments, with a maximum overestimation of 2%. As the mass flow rate increases, the pressure drop rises as well with a quadratic trend. This behavior follows the Darcy´s equation with the Dupuit/Forchheimer extension, where the specific pressure drop is expressed in relation to the flow velocity, dynamic viscosity (μ) and density (ρ), permeability (K1) and inertial (K2) coefficients [22], according to the following equation:

6 7

AC C

482

T = 298 K -3 -1 ṁ = (16.0; 14.0; 12.0; 8.0)∙10 kg∙s .

M AN U

• •

468 469 470 471 472 473 474 475 476 477 478 479 480 481

RI PT

449 450 451 452 453 454 455 456 457 458 459

=

8

9:

 +

;

9<



(14)

where u is the fluid velocity. Thus, permeability and inertial coefficients have been calculated for the innovative geometry and are listed below: • •

-7

2

K1 = 3.0∙10 m ; -3 K2 = 4.9∙10 m.

In comparison to the HiTRec-II structure, the innovative geometry has a higher pressure drop. This is due to the staggered characterization of the demonstrator, as section variation leads to higher flow velocities and, thus, to higher values of Δp/L.

11

ACCEPTED MANUSCRIPT

5. Conclusions

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507

In the study presented in this article, a new geometry has been presented to be used as solar absorber in volumetric receivers. The structure design is based on the previous results obtained by the author concerning the optimization of such structures for high temperature solar applications [9]. The new geometry has been numerically tested with the use of a discrete CFD simulation and a first performance comparison has been carried out taking into account the current state-of-the-art of volumetric receivers and an advanced honeycomb geometry produced at the DLR. Afterwards, a 3:1 scaled up demonstrator has been produced, taking into account current technology limits in manufacturing. Experimental validation of the numerical methodology and results has been performed showing good agreement between each other. Improved thermal performance and the achievement of the volumetric effect have been obtained with the innovative structure presented, resulting in a big step forward for volumetric solar receivers. However, current manufactory technology limits have forced the analysis on a scaled-up structure that still showed improved performance compared to the current state-of-the-art, but with lower performance in comparison to the original structure developed. Those performance can be enhanced even more in the midterm future with the use of finer structures and this will be feasible thanks to the potential of manufacturing procedures in the mid-term future, like the electron beam sintering (EBM) used in the present study.

508

Acknowledgment

509 510

This work was carried out with financial support from the Ministry of Innovation, Science and Research of the State of North Rhine-Westphalia (MIWF NRW), Germany under contract 323-2010-006 (Start-SF).

511

References

512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537

[1].

[5].

[6].

[7].

[8]. [9].

[10].

M AN U

TE D

[4].

EP

[3].

Sander Associates Inc, 1979. 1/4-Megawatt Solar Receiver. Final Report. INC. DOE/SF/90506-1, October. H.W. Fricker, Proposal for a novel type of solar gas receiver. Proceedings of the International Seminar on Solar Thermal Heat Production, Stuttgart, German Aerospace Center (DLR), Pfaffennwaldring, 3840, D-70569 Stuttgart (1983). R.D. Skocypec, R. F. Boehm, J. M. Chavez, Heat Transfer Modeling of the IEA/SSPS Volumetric Receiver. Journal of Solar Energy Engineering 111 (1989) 138-143. T.E. Hellmuth, L. K. Matthews, J. M. Chavez, C. A. Hale, Performance of a wire mesh solar volumetric air receiver. Proceeding of the 1994 ASME/JSME/JSES International Solar Energy Conference (1994) 573-578. B. Hoffschmidt, R. Pitz-Paal, M. Böhmer, T. Fend, P. Rietbrock, 200 kWth open volumetric air receiver (HiTRec) of DLR reached 1000 °C average outlet temperature at PSA - Journal of Physics IV France 9 (1999). B. Hoffschmidt, F. M. Téllez, A. Valverde, J. Fernández, V. Fernández, Performance Evaluation of the 200-kWth HiTRec-II Open Volumetric Air Receiver, Journal of Solar Energy Engineering 125 (2003) 8794. T. Fend, B. Hoffschmidt, R. Pitz-Paal, O. Reutter, P. Rietbrock, Porous materials as open volumetric solar receivers: Experimental determination of thermophysical and heat transfer properties. Energy 28 (2004) 823-833. O. Smirnova, T. Fend, P. Schwarzbözl, D. Schöllgen, C. Jakob, Numerical investigation of flow and heat transfer in a volumetric solar receiver. Renewable Energy 60 (2013) 655-661 R. Capuano, T. Fend, B. Hoffschmidt, R. Pitz-Paal, Innovative Volumetric Solar Receiver Micro-Design Based on Numerical Predictions. ASME 2015 International Mechanical Engineering Congress and Exposition, Volume 8B: Heat Transfer and Thermal Engineering (2015). T. Pfeifer, J. Matyáš, P. Balaya, D. Singh, J. Wei, Tuning the Spectral Selectivity of Sic-Based Volumetric Solar Receivers With Ultra-High Temperature Ceramic Coatings, in Ceramics for Energy Conversion,

AC C

[2].

SC

RI PT

491

12

ACCEPTED MANUSCRIPT

[16].

[17]. [18]. [19]. [20]. [21]. [22].

RI PT

[15].

SC

[14].

M AN U

[13].

TE D

[12].

EP

[11].

Storage, and Distribution Systems. Ceramic Transactions - 255, John Wiley & Sons, Inc., Hoboken, NJ (2016). S. Z. Shuja, B. S., Yilbas, Innovative design of a solar volumetric receiver: Arrangements of absorbing block configurations. Solar Energy, 146 (2017) 105-112. C. Pabst, G. Feckler, S. Schmitz, O. Smirnova, R. Capuano, P. Hirth, T. Fend, Experimental performance of an advanced metal volumetric air receiver for Solar Towers. Renewable Energy, 106 (2016) 91-98. A. Kalt, M. Becker, G. Dibowski, U. Groer, A. Neumann A., The new solar furnace of the DLR, Cologne, th Germany, specifications and first test results. Proceedings of the 7 International Symposium on Solar Thermal Concentrating Technologies. September 26-30, (1994) 1023-1035. O. Reutter, O. Smirnova, J. Sauerhering, S. Angel, T. Fend, R. Pitz-Paal, Characterization of Air Flow through Sintered Metal Foams, ASME Journal of Fluids Engineering, 130 (2008). P. Schwarzboezl, U. Hack, M. Ebert, Improvement of ceramic absorber material for open volumetric th receviers. 17 SolarPACES Conference, Granada, Spain (2011). G. Koll, P. Schwarzboezl, K. Hennecke, T. Hartz, M. Schmitz, B. Hoffschmidt. The solar tower Jülich - a research and demonstration plant for central receiver systems. Proceedings of the 15th SolarPACES Conference, Berlin, Germany (2009). R. G. Munro, Material Properties of a Sintered alpha-SiC. Journal of Physical and Chemical Reference Data, 26 (1997) 1195-1203. A. Kribus, Y. Gray, M. Grijnevich, G. Mittelman, S. Mey-Cloutier, C. Caliot, The promise and challenge of solar volumetric absorbers. Solar Energy, 10 (2014) 465-481. B. Kloeden, Principle workflow of the EBM process, Laser Technik Journal, 2 (2012) 33-38. R. Capuano, T. Fend, H. Stadler, Innovative porous microstructure for solar thermal high temperature application. German Patent and Trade Mark Office(DPMA), DE202016003017.6 (2016). J. R. Howell, M. P. Menguc, R. Siegel, Thermal radiation heat transfer, 6th Edition, Taylor and Francis, 2015. M. Kaviany, Principles of Heat Transfer in Porous Media? 2nd edition, Springer, 1999.

AC C

538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563

13

ACCEPTED MANUSCRIPT

Table

Table 1 – Characteristic geometric values of the structures analysed in the numerical comparison

Wall thickness

Channel depth

[mm]

[mm]

[mm]

HiTRec-II

2.0

0.8

50.0

0.51

Advanced honeycomb

1.4

0.4

50.0

0.60

New geometry – original design

1.1

0.2

20.0

ca. 0.97 – 0.71

ca. 990 – 2860

Demonstrator

3.6

0.6

60.0

0.97 ca. – 0.71

ca. 318 – 870

Specific surface area

ε

Av - [m2∙m-3]

AC C

EP

TE D

SC

M AN U

Structures

Porosity

RI PT

Channel width

1020

2560

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

Figures

AC C

EP

TE D

Figures 1. Incoming irradiation settings for the discrete model – Comsol built-in ray-tracer algorithm set on plane XZ according to the experimental setup HLS Solar Simulator, DLR Cologne.

Figures 2. Incoming irradiation settings for the discrete model – Comsol built-in ray-tracer algorithm set on planes YZ according to the experimental setup HLS Solar Simulator, DLR Cologne.

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

Figure 3. HLS setup – Right: Used lamps highlight. Adapted from: Smirnova et al. (2013).

Figure 4. Test bed – Sample housing.

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

Figure 5. HLS flow scheme. 1 – absorber sample, 2 – hot gas tube, 3 – thermal insulation, 4 – heat exchanger, 5 – mass flow measurement, 6 – air blower. Adapted from: Smirnova et al. (2013).

Figure 6. Experimental setup for the pressure drop analysis: 1 – air blower, 2 – orifice, 3 – thermocouples and pressure sensors, 4 – porous sample.

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

Figure 7. Single unit element new geometry. Patent pending (Capuano et al., 2016).

Figure 8. Single unit elements characterizing the geometries used for comparison.

RI PT

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

Figure 9. Tetrahedral mesh used for the discretization of the unit element solid volume.

AC C

Figure 10. Complex mesh used for the discretization of the unit element fluid volume.

ACCEPTED MANUSCRIPT

1200

RI PT

1000 800

T Solid

600

T Fluid 400

HiTRec-II

200

SC

Temperature profiles - T [K]

1400

Advanced honeycomb

0 0.0

10.0

20.0

M AN U

New geometry

30.0

40.0

50.0

60.0

Unit element depth - z [mm]

TE D

Figure 11. Temperature profiles along the flow direction for the new geometry and structures used for comparison.

1200

EP

1000 800

AC C

Temperature profiles - T [K]

1400

T Solid

600

T Fluid HiTRec-II

400

Advanced honeycomb

200

Demonstrator

0 0.0

10.0

20.0

30.0

40.0

50.0

60.0

Unit element depth - z [mm] Figure 12. Temperature profiles of the demonstrator and the comparison cases along the flow direction.

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

TE D

Figure 13. Ti6Al4V demonstrator. Patent pending (Capuano et al., 2016).

1000

800

EP

700 600 500

AC C

Outlet air temperature - Tf,o [K]

900

400 300

Experimental results

200

Numerical results

100

HiTRec-II - numerical

0 0

100

200

300

400

500

600

700

Power on aperture / Air mass flow rate - P/ṁ [kJ∙kg-1] Figure 14. Outlet air temperature – experimental and numerical results of the Demonstrator and further comparison with numerical results of Hitrec-II state-of-the-art.

ACCEPTED MANUSCRIPT 1.0 0.9 0.8

Efficiency - η

0.7

RI PT

0.6 0.5 0.4 0.3

Experimental results

0.2

SC

Numerical results

HiTRec-II - numerical

0.1 0.0 100.0

200.0

300.0

400.0

500.0

M AN U

0.0

600.0

700.0

800.0

Power on aperture / Air mass flow rate - P/ṁ [kJ∙kg-1] Figure 15. Thermal efficiency – experimental and numerical results of the Demonstrator and further comparison with numerical results of Hitrec-II state-of-the-art.

TE D

1600 1400

EP

1200 1000 800

AC C

Specific pressure drop - Δp/L [N∙m-3]

1800

600

Experimental results Numerical results HiTRec-II - numerical

400 200 0

0.0

1.0

2.0

3.0

4.0

5.0

Mass flow rate - ṁ [kg ∙ s-1] Figure 16. Specific pressure drop – experimental and numerical results of the Demonstrator and further comparison with numerical results of Hitrec-II state-of-the-art.

ACCEPTED MANUSCRIPT

Optimized volumetric solar receiver: thermal performance prediction and experimental validation. Raffaele Capuano*a, Thomas Fendb, Hannes Stadler a, Bernhard Hoffschmidtb. *Corresponding author. German Aerospace Center DLR, Institute of solar Research, Professor-Rehm-Strasse 1, Juelich.

b

German Aerospace Center DLR, Institute of solar Research, Linder Hoehe, Cologne.

RI PT

a

SC

Highlights

M AN U

 Volumetric receiver presents high margin of improvement concerning the absorber shape.  Inlet heat losses reduction is of utmost importance to enhance the performance.

TE D

 An optimized absorber geometry, reducing the heat losses, is presented in this work.  Thermal performance results to be higher than the current state-of-theart.

AC C

EP

 Manufacturability of the geometry can be improved in the near future.