Performance analysis of a photovoltaic-thermochemical hybrid system prototype

Applied Energy xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Performance analysis of a photovoltaic-thermochemical hybrid system prototype Wenjia Li a,b,1, Yunyi Ling a,b,1, Xiangxin Liu b,c, Yong Hao a,b,⇑ a

Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing 100190, PR China University of Chinese Academy of Sciences, Beijing 100049, PR China c Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, PR China b

h i g h l i g h t s
A modular photovoltaic-thermochemical hybrid system prototype is proposed.
Net solar-electric efficiency up to 41% is achievable.
Stable solar power supply is achievable via convenient energy storage.
The modular design facilitates the scalability of the hybrid system.

a r t i c l e

i n f o

Article history: Received 20 January 2017 Received in revised form 5 May 2017 Accepted 6 May 2017 Available online xxxx Keywords: Solar Photovoltaic Methanol thermochemistry Hybrid Power generation Efficiency

a b s t r a c t A solar photovoltaic (PV) thermochemical hybrid system consisting of a point-focus Fresnel concentrator, a PV cell and a methanol thermochemical reactor is proposed. In particular, a reactor capable of operating under high solar concentration is designed, manufactured and tested. Studies on both kinetic and thermodynamic characteristics of the reactor and the system are performed. Analysis of numerical and experimental results shows that with cascaded solar energy utilization and synergy among different forms of energy, the hybrid system has the advantages of high net solar-electric efficiency (up to 41%), stable solar energy power supply, solar energy storage (via syngas) and flexibility in application scale. The hybrid system proposed in this work provides a potential solution to some key challenges of current solar energy utilization technologies. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction As one of the few potential solutions to global energy and environmental problems, solar energy technologies are widely employed to supply heat [1], generate electricity [2] and produce solar fuels [3]. This is achieved separately or simultaneously by different solar energy technologies, such as solar photovoltaic (PV), thermal and photocatalytic [4] conversion approaches. Comprehensive utilization of solar energy by combining the strengths of multiple solar conversion approaches could enhance efficiency and reduce cost. It has become an intensively studied area of solar energy research in recent years. A typical example is the PVthermal (PVT) technique, which aims at recovering and utilizing ⇑ Corresponding author at: Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing 100190, PR China. E-mail address: [email protected] (Y. Hao). 1 Equal-contribution first authors.

thermal energy released from PV generation using a complementary thermal module. The PVT technique was first proposed by Wolf [5] in the 1970s. Since then, many studies on PVT systems have been performed [6], most of which focused on heating and cooling by the waste heat from PV cells. Performances of such systems have been studied theoretically and experimentally from thermodynamic [7], economic [8] and environmental [9] perspectives, and higher efficiency, better economy and lower emissions of greenhouse gases have been achieved. PVT systems aiming at direct utilization of thermal energy are severely constrained in space and time owing to challenges in thermal energy transport. Additionally, despite impressive first-lawbased total efficiencies of up to 80% [10], increasing the electricity ratio among the total energy output remains a considerable challenge, which, to a large extent, pivots on how thermal energy can be efficiently converted to power. A few research groups have attempted to address this issue by using organic Rankin cycle (ORC) or a thermoelectric generator (TEG) thermal module to

http://dx.doi.org/10.1016/j.apenergy.2017.05.077 0306-2619/Ó 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

suitable for HCPV cells. The performance of the reactor and the reactor-based system is analyzed, and implications for efficient utilization of solar energy are discussed using multidisciplinary coupled-field, experimental and thermodynamic analysis methods. 2. Description and modeling of the system 2.1. System description Fig. 1 shows the diagram of the proposed system, which consists of a methanol tank, a methanol pump, a solar PV hybrid heater, a solar PV-thermochemical hybrid reactor, a heat exchanger and a syngas storage tank. In the hybrid system, methanol is sequentially pumped from its storage tank into the heat exchanger and the heater; during this process, methanol is heated and/or vaporized by the dissipated heat from the PV modules and the recovered heat of the methanol and syngas mixture. Methanol vapor exiting the heater is transferred to the heat exchanger and then the reactor, where a part of the methanol is decomposed into syngas by dissipated heat from the PV modules. The methanol and syngas mixture is cooled in the heat exchanger and finally flows into the syngas storage tank, where the mixture is cooled to ambient temperature and methanol is condensed and recycled to the methanol tank. To facilitate the electricity generation efficiency evaluation, syngas from the tank is compressed and fed into the heat engine (i.e., gas and steam combined cycle, GSCC) for power generation. 2.2. Construction of the PV hybrid reactor Fig. 2 displays a schematic diagram of the modular solar PV hybrid reactor, which consists of a point-focus Fresnel concentrator, an optical prism, a PV cell and a methanol decomposition reactor. The operation of the hybrid reactor comprises the following

Heater 3

Pump

Valve

4 5

Reactor

7 9

Syngas

8

Heat Exchangers

6

+

2

1

-

Methanol

+

generate electricity from the waste heat of PV cells [11–14]. However, owing to opposite trends in the responses to the temperature of the PV part [15] and the thermal part, as well as their low heatelectricity conversion efficiencies, many (if not all) of the gains that the thermal parts bring about are cancelled to a large extent. Besides, difficulties in stable power supply (resulting from spacetime intermittency of solar energy) and/or scale-matching remain unresolved in these systems. To achieve the goals of high solar-electric efficiency and stable power supply, we have proposed an efficient solar power generation system that integrates a PV module and a low-temperature methanol decomposition module [16]. In the proposed system, sunlight is concentrated onto PV cells; part of the sunlight is converted to electricity directly by the PV cells and the rest is converted to heat, which is absorbed by the endothermic methanol decomposition at approximately 250 °C and stored in syngas in the form of chemical energy. The stored solar chemical energy is released in a combined cycle during combustion at approximately 1300 °C and eventually converted to electricity. During the entire process, solar energy is utilized in a cascaded way and the lowtemperature (250 °C) dissipated heat from the PV cells is upgraded to high-temperature (1300 °C) heat, partially assisted by chemical energy. Thermodynamic analysis indicated that the system was characterized by high net solar-electric (NSE) efficiency (43%; higher than the 35% efficiency of the solar-methanol thermochemical power generation system [16,17]) and easy energy storage. However, the detailed matching of the PV and methanol thermochemical modules and the design of the PV hybrid reactor have not been studied. For the methanol decomposition/reforming reactor without the PV module, Jin et al. [18] designed a 5 kW solar receiver/reactor prototype for solar methanol decomposition; the conversion ratio of solar thermal energy to chemical energy was in the range of 30%–60% with a concentration ratio of 70. Based on the same reactor, Liu et al. [19] further analyzed the performance of methanol steam reforming and obtained similar results. In contrast to the relatively large scale reactor of Jin and Liu, Zimmerman et al. [20] developed a micro solar methanol reforming reactor aiming at a high collector temperature without sunlight concentration. Theoretical analysis showed that a collector temperature of up to 250 °C was achieved by employing vacuum insulation and selective absorbing coating. Gu et al. [21,22] expanded further on this concept by incorporating a compound parabolic concentrator (CPC, concentration ratio of 1.75) in the reactor and achieved a thermal efficiency of 65–71% with a solar collector temperature of 250–300 °C. All of the above methanol thermochemical reactors were designed for a relatively low concentration ratio (lower than 100). However, the concentration ratio of a PV cell (i.e., efficient multi-junction GaAs), which can withstand temperatures higher than 200 °C [23], must be higher than 300 owing to economic considerations [24]. On account of the high energy flux resulting from this high concentration ratio, heat utilization and thermal management of PV cells become vital issues [23]. Furthermore, when the concentration ratio increases from below 100 to above 300, not only the energy flux increases, but also the type of concentrator changes from CPCs and line-focus trough concentrators to pointfocus Fresnel concentrators. Thus, the geometry matching and multi-field coupling between the high-concentration PV (HCPV) cell and the methanol decomposition reactor become important issues. In this work, we focus on the practical integration of a HCPV cell with a methanol thermochemical reactor and propose a hybrid system consisting of a point-focus Fresnel concentrator, a multijunction PV cell and a methanol thermochemical reactor. Furthermore, we design and manufacture a modular methanol reactor

-

2

To Heat Engine i Fig. 1. Diagram of the hybrid system.

Fig. 2. Schematic diagram of the PV hybrid reactor: a point-focus Fresnel concentrator, an optical prism, a PV cell, and a methanol decomposition reactor.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

3

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

Fig. 3. Construction of the methanol decomposition reactor; (a) exterior view and (b) interior view.

steps. First, sunlight is concentrated onto the PV cell by the Fresnel lens. Upon receiving concentrated light, the PV cell generates electricity and simultaneously dissipates heat. A part of this heat is lost to the environment and the rest is transferred to the reactor to provide the energy needed for the endothermic reaction, i.e., the conversion from methanol to syngas. The construction of the solar PV hybrid heater is similar to that of the solar PV hybrid reactor; as the heater is not important in this study, we will not repeat it here. A methanol decomposition reactor is designed as shown in Fig. 3a. The reactor consists of three parts: circular inlet and outlet tubes, a metal reactor box and a PV cell attached to one surface of the reactor. The structural parameters of the hybrid reactor are summarized in Table 1. It is worth noting that although the PV hybrid reactor examined in this study has a small size, its modular design facilitates scaling up, which makes this hybrid system potentially suitable for a wide range of applications. Fig. 3b displays the inner flow channel filled with catalyst. The cross section of the inner flow channel is a square with a side length of 7.5 mm. Properties of the catalyst and the bed are shown in Table 2. 2.3. Energy balance in the PV hybrid heater and reactor Development of the numerical models is presented in the following two sections, and experimental set-up is described in Section 2.5. As previously mentioned, sunlight is first concentrated onto the surface of the PV cells and converted into electricity and PVdissipated heat. The electricity and the dissipated heat are expressed, respectively, as: Table 1 Structural parameters of the hybrid reactor. Parameters

Value

Dimensions of the reactor box Side length of inner flow channel (one segment) Length of circular inlet or outlet tube Outer diameter of circular tubes Inner diameter of circular tubes Dimensions of the PV cell Material of the reactor box Type of the PV cell

57.6 mm  71.6 mm  80 mm 7.5 mm 80 mm 10 mm 8 mm 10 mm  0.1 mm  10 mm Aluminum InP/InGaAs/Ge

Table 2 Properties of the CNZ-1 catalyst and the bed. Parameters

Value

Properties of catalyst Mass fraction of CuO Mass fraction of ZnO Mass fraction of Al2O3 Density of the catalyst

40% 40% 20% 5767 kg/m3

Parameters of the catalyst bed Porosity of the catalyst bed Equivalent diameter of the catalyst bed Permeability of the catalyst bed

0.49 1 mm 3  109 m2

_ PV ¼ W Q_ diss ¼

Z Z

_ PV ðT i Þ ¼ dW dQ_ diss ðT i Þ ¼

Z Z

dQ_ tot ðT i Þ  gopt  gT i ;PV

ð1Þ

dQ_ tot ðT i Þ  gopt  ð1  gT i ;PV Þ

ð2Þ

where dQ_ tot ðT i Þ is the total solar energy input rate at a point of the PV cell with temperature Ti before concentration, gopt is the optical efficiency of the collector, and gT i ;PV is the efficiency of the PV modules at Ti. The efficiency of the InP/InGaAs/Ge triple-junction PV module employed in this work is given by [25]:

h

gT i ;PV ¼ 0:298 þ 0:0142 ln C þ ð0:000715 þ 6:97  105 ln CÞ i  ðT i þ DT  298:15Þ  gmod

ð3Þ

where C is the concentration ratio and gmod is the ratio of the efficiency of the entire PV module to that of an individual PV cell, which is considered as 0.9 [25]. As mentioned, part of the dissipated heat is lost to the environment, and the rest is absorbed by methanol during its heating or decomposition. The convective and the radiative heat loss rate are given, respectively, by:

Q_ con;loss ¼ Q_ rad;loss ¼

Z Z

dQ_ con;loss ðT j Þ ¼ dQ_ rad;loss ðT i Þ ¼

Z hj  dAj  ðT j  T 0 Þ

ð4Þ

ePV  dAi  ðT 4i  T 40 Þ

ð5Þ

Z

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

4

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

where hj is the convective heat transfer coefficient at a point on the surface of the heater, the reactor or the PV cells, which is obtained by minimizing the difference between the experimental and the numerical heat loss; ePV is the PV emissivity; dAi is the area of the point on the PV cells; and dAj is the area of the surface of the heater, the reactor or the PV cells. The remaining dissipated heat is expressed as:

Q_ me ¼ Q_ diss  Q_ con;loss  Q_ rad;loss

ð6Þ

This part of the heat is absorbed by methanol for heating, evaporation and decomposition, and it is provided by:

Q_ me;heat ¼ n_ 

Z



C p;T i  dT þ r eva

þ Q_ chem

ð7Þ

where n_ is methanol molar flow rate, C p;T i is methanol heat capacity at temperature Ti, reva is the latent heat of methanol vaporization, and Q_ chem is the heat absorption rate corresponding to methanol decomposition, which is calculated with the following kinetic model (details in Section 2.4). 2.4. Kinetic model of the reactor Coupled-field analysis is employed to analyze methanol decomposition reaction in the reactor. The steady-state continuity equation is:

r  ðq~ uÞ ¼ 0

ð8Þ

where q and ~ u represent density and velocity of the gas mixture, respectively. The flow inside the inlet and outlet circular tubes is considered laminar owing to a low Reynolds number (lower than 500). The reactor filled with catalyst can be considered as a packed bed reactor [26], which is homogeneous in the model. The steady-state momentum equation inside the reactor is described by:

r  ðq~ u~ uÞ

e

2 p

 !   2l ! l ~ T ¼ r  P I þ ru þ ðr~ ðr  ~ uÞ I  lj1~ uÞ  u ep 3ep

ð9Þ where ep is porosity of the catalyst bed, P is pressure of the gas mixture, l is dynamic viscosity of the gas mixture, and j is permeability of the catalyst bed described by [26]: 2

j¼

dp e3p

ð10Þ

150ð1  ep Þ2

where dp represents the characteristic diameter of the catalyst bed. The steady-state energy equation, assuming no body forces or viscous dissipation, is:

u  rT ¼ r  ðkeff rTÞ þ q qC p~

ð11Þ

keff ¼ hp kp þ ð1  hp Þk

ð12Þ

where xk is the mass fraction of gas component k and subscript k represents H2, CO or CH3OH; Mn represents the average molar mass of the gas mixture; Rk denotes the production or destruction rate of species k; Dk is the binary diffusion coefficient from the FullerSchettler-Giddings theory [27], choosing the air at 500 K and 1 bar as the reference. The reaction mechanism of methanol decomposition has been studied extensively [28–30]. According to Jin’s kinetic model [30], methanol reaction rate during methanol decomposition can be formulated as:

RMe ¼ ð1  ep Þqs r Me rMe ¼ 43:5a exp

ð14Þ

  5856 Pk ð1  XÞ T Rg T k ð1 þ 2XÞ

ð15Þ

where qs is the density of the catalyst; rMe is methanol reaction rate per kilogram of catalyst; a is the correction factor of methanol reaction rate for the reactor, which is obtained by minimizing the difference between the experimental and numerical methanol conversion ratios; Rg is universal gas constant; and X represents the conversion ratio of methanol. The heat absorbed by methanol during decomposition in the reactor is:

Q_ chem ¼ DH 

Z Z Z

RMe  dV

ð16Þ

where DH is the decomposition enthalpy of methanol and dV is a differential volume of the inner flow channel of the reactor. 2.5. Experimental set-up description Experiments are performed on the reactor described above to verify the feasibility of the reactor and the accuracy of the kinetic model. Fig. 4 shows the experimental system, which mainly consists of a methanol tank, a methanol metering pump, a heater, a reactor, electric heating plates, thermocouples, a mass-flow meter and a data acquisition and control module. In the system, methanol is pumped into the preheater, where it is heated, vaporized and superheated by electric heating plates and then flows into the reactor. Inside the reactor, methanol further absorbs heat from the electric heating plates and decomposes into syngas. When leaving the reactor, the mixture of reactants and products is cooled in the condenser, where unreacted methanol is separated. The catalyst is a commercial CuO/ZnO/Al2O3 product from the Southwest Research & Design Institute of Chemical Industry and its properties are shown in Table 2. The temperatures of the heater, the reactor and their inlets and outlets are measured by K-type thermocouples. The dissipated heat from the PV cell is simulated by electric heating plates. The thermal power of the electric heating plates is measured with a power meter. Pictures of the reactor and the apparatus employed in the experiment are shown in Fig. 5.

where C p is heat capacity of the gas mixture; T represents temperature of the gas mixture; q denotes heat flow rate, including thermal heat flow and chemical reaction heat flow; keff represents the equivalent thermal conductivity of the porous medium; hp is volumetric fraction of the catalyst; kp is thermal conductivity of the catalyst; and k is thermal conductivity of the gas mixture. The mass balance equation, including Fick diffusion and convective diffusion, accounts for mass transport and conversion of all gas-phase species, and is expressed as:



r  qk Dk rxk  qk xk Dk

rMn Mn



þ qk ð~ u  rÞxk ¼ Rk

ð13Þ Fig. 4. Diagram of the experimental system.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

5

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

Fig. 5. Pictures of (a) the reactor with an electric heating plate, (b) mass-flow controllers, (c) methanol metering pump, (d) temperature controller, (e) power meter and (f) data acquisition module.

Experimental investigations were conducted in the following sequence. Firstly, reduction of the catalyst was performed by generating a H2/N2 mixture flow through the reactor channel with a volumetric concentration of H2 ranging between 0.5% and 10% and a temperature of 250 °C. Secondly, the thermal efficiency (Eq. (18)) of the reactor was measured. Without any methanol input to the reactor, the heat loss of the reactor at different temperatures was measured by the electricity input of the electric heating plates while maintaining the temperature of the reactor. Thirdly, methanol conversion ratios (X in Eq. (15)) for an electric heating plate power of 35 W and a liquid methanol flow rate (a key parameter for the system operation) of 0.50–1.50 ml/min were obtained from the energy balance and verified by the flow rate of the products. The energy balance is given as

Q_ me;heat ¼ n_ 

Z

C p;T i  dT þ n_  X  DH þ Q_ loss

ð17Þ

where Q_ loss is the total heat loss of the hybrid reactor, obtained from the second step. During the experiments, the inlet temperature of the reactor was set at 160 °C. The experimental results are used to calibrate the parameter values presented in Sections 2.3 and 2.4. Then, the calibrated numerical model of the reactor is employed to simulate the reactor and system performance in a wide range of input parameters, as described in Sections 3.2–3.4. 2.6. Thermodynamic performance criteria of the system

Q_

Q_ con;loss þ Q_ rad;loss Q_ diss

P

ð18Þ

ð19Þ Q_

P

gte ¼ _chem ¼ _chem _ chem ¼ gtc gce Q diss Q diss Q chem P

gNSE ¼ _ net ¼ Q tot

ð20Þ

PPVþME  Pref ¼ gopt  ½gPV þ ð1  gPV Þ  gtc  gce  Q_ tot ð21Þ

where P chem is the additional electricity generation, compared with that from GSCC with the direct combustion of the same amount of methanol, in the thermochemical module due to methanol decomposition; gce is the chemical-electric efficiency; Pnet is the net electricity generated by solar energy in the hybrid system, which deducts the electricity generated by methanol from the total electricity; PPVþME is the overall electricity output of the hybrid system, which is the sum of the electricity from the PV modules and the GSCC; and P ref is the electricity output of the GSCC with the same amount of methanol as the input. To evaluate the energy storage capacity of the hybrid system, the energy storage ratio is defined as the ratio of the additional electricity in the thermochemical module, which can be stored via the storage of the syngas, to the net electricity generated by solar energy in the hybrid system:

rstore ¼

Besides the above models, the rest of the process, composed of the methanol warm-up, methanol recycling and syngas separation, is simulated using Aspen Plus [31]. For the system electricity generation performance evaluation, the syngas is fed into a GSCC as fuel and its combustion and mechanical power generation are also simulated with Aspen Plus [31]. For the evaluation of performances of the reactor and the system, we define reactor thermal efficiency, solar thermal-chemical efficiency, solar thermal-electric efficiency and NSE efficiency as follows, respectively:

gheat ¼ 1  _ loss ¼ 1  Q diss

Q_

gtc ¼ _chem Q diss

Pchem P net

ð22Þ

3. Results and discussion The kinetic and thermodynamic performances of the reactor and the hybrid system are studied via experiment and simulation. The operating parameters of the hybrid reactor used in the simulation are summarized in Table 3. It is worth noting that the thermal energy input of the reactor (i.e., dissipated heat from the PV cells), which is positively correlated with the solar energy input, is used as the indicator of the solar energy input of the reactor and the entire system, and the solar energy input can be easily obtained with Eq. (2).

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

6

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

Table 3 Operating parameters of the hybrid reactor. Parameters

Value

Solar collector optical efficiency Thermal energy input of the reactor Inlet temperature of the reactor Inlet flow rate

0.8 35 W, 45 W and 55 W 160 °C 0.50–1.50 ml/min at 35 W 0.75–1.75 ml/min at 45 W 1.25–2.25 ml/min at 55 W 1 bar

Outlet pressure

3.1. Calibration of the numerical model The numerical model is calibrated by minimizing the difference between the simulation and the experimental results in terms of thermal and chemical performances. The convective heat transfer coefficient (hj) and the correction factor of the methanol reaction rate (a) are 1.7 W/(m2K) and 0.37, respectively. The results of the numerical model calibration process are shown and analyzed below. Fig. 6 shows the heat loss, average temperature and methanol conversion ratio of the reactor obtained from numerical simulation and experiments. The heat loss difference between the numerical and experimental results is lower than 0.3 W, and the relative difference is lower than 3%; the difference between average reactor temperature acquired by the two approaches is lower than 3K; methanol conversion ratio difference is lower than 0.01 and the relative difference is 3%. The good agreement between simulation

15.0 1.0

13.5 13.0 12.5

0.6

12.0 11.5

0.4

11.0 10.5 10.0 0.4

0.2 0.6

0.8

1.0

1.2

1.4

1.6

280

Methanol conversion

Temperature Methanol conversion Heat loss 0.8

14.0

Heat loss (W)

Sim

260 240 220 200 180

Average temperature (°C)

300 Exp

14.5

160

Methanol flow rate (ml/min) Fig. 6. Comparison between experiment and simulation: heat loss, average temperature, and methanol conversion ratio of the reactor.

and experimental results verifies the accuracy of the numerical model. Based on the calibrated numerical model, the kinetic and thermodynamic analysis of the system is performed as described in Sections 3.2–3.4. 3.2. Temperature distribution of the hybrid reactor A reactor energy input of 35 W and a methanol flow rate of 0.75 ml/min are chosen for the studies on heat transfer performance and kinetic performance of the reactor. Fig. 7 shows the temperature distribution inside the methanol decomposition reactor and over the PV cell. The main body of the reactor (Fig. 7a) is at approximately 236 °C with a maximum temperature of 248 °C in the PV cell and a minimum temperature of 233 °C, indicating a relatively uniform temperature distribution, which results from the small size and outstanding thermal conductivity of the reactor. To examine the PV cell heat transfer performance, the PV cell temperature distribution is illustrated in Fig. 7b. As the only heat source of the entire system, the PV cell is the hottest part of the reactor, reaching 248 °C. The temperature decreases towards the edge of the PV cell, which results from heat transfer from the sides. The maximum temperature difference over the PV cell is approximately 5 K, which is acceptable for operation. Among all the operating conditions studied, the temperature difference is in the range 5 K–10 K, with 10 K observed when the reactor power is 55 W and the inlet flow rate is 2.25 ml/min. Considering the relatively uniform temperature distribution in the PV cell, the average PV temperature is used to calculate the PV-module efficiency using Eq. (3). In summary, the PV cell and the methanol decomposition reactor both have relatively uniform temperature distribution, which benefits the PV-module efficiency and the methanol decomposition reaction. 3.3. Reaction rate and conversion ratio In this study, the reaction (i.e., methanol decomposition) is assumed to occur only inside the reactor, where the catalyst is present. Therefore, to determine the reaction rate, we focus on the inner part of the reactor, as shown in Fig. 8. The distribution of the reaction rate along the flow channel (Fig. 8a) shows a maximum of approximately 4.0 mol/(m3s) at the inlet, followed by a monotonic decrease along the channel, and a minimum of approximately 1.7 mol/(m3s) at the outlet owing to the consumption of methanol along the flow path. Fig. 8b displays the reaction rate from an x-y cross-section view. In addition to the general trend shown in Fig. 8a, the figure also reveals an interesting

Fig. 7. Temperature distribution of: (a) the hybrid reactor and (b) the PV cell.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

7

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

Fig. 8. Reaction rate distribution inside the PV-thermochemical hybrid reactor; (a) side view of the reaction rate along all flow channels and (b) cross-section view.

phenomenon: the reaction rate is always much greater near the channel wall of the reactor than in its interior. This is a major advantage of an all-metal reactor, as its superior thermal conductivity guarantees a relatively uniform and high temperature along the channel walls in a very compact space, which makes the

endothermic reaction proceed at reasonably high rates. On the other hand, the energy transport from the channel walls to its interior is significantly limited by the fact that the gas flow is laminar and by the much lower thermal conductivity of the gas mixture. Additionally, given the relative uniformity of the temperature, the methanol concentration becomes the dominant factor affecting the reaction rate. Fig. 9 depicts the conversion ratio of methanol inside the reactor. During the methanol flows through the inner channel of the reactor, the conversion ratio increases monotonically and reaches a maximum of 67% at the exit of the inner flow channel. 3.4. Thermodynamic performance of the hybrid system The comprehensive utilization of solar energy enables high power generation efficiency. Fig. 10a shows that a maximum NSE efficiency of 41% is obtained when the thermal energy input of the reactor is 55 W (the total solar energy input is 119 W) and the inlet flow rate of methanol is 1.25 ml/min. The primary reason for this high efficiency is the cascaded utilization of solar energy. Part of the solar energy is first converted to electricity by the PV cells, and the rest is converted into low-temperature thermal energy as the energy input for methanol decomposition. The

Fig. 9. Methanol conversion ratio distribution inside the reactor.

45 W

55 W

35 W

0.41

0.40 0.35

0.36

0.30

Efficiency

0.25

(a)

0.20 NSE efficiency 0.45 0.40 0.35 0.30 0.25

(b)

0.20 PV efficiency 0.45 0.40 0.35 0.30 0.25

(c)

0.20 Thermal-electric efficiency

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Methanol flow rate (ml/min)

1.8

2.0

Conversion Thermal efficiency TEMP (°C)

35 W 0.45

45 W

55 W

340 320 300 280 260 240 220 0.80

(d)

0.75 0.70 0.65

(e)

0.60 1.0 0.8 0.6 0.4 0.2

0.4

(f) 0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Methanol flow rate (ml/min)

Fig. 10. Simulation results from the calibrated numerical model: (a) NSE efficiency of the hybrid system; (b) average efficiency of the PV module of the hybrid system; (c) thermal-electric efficiency of the thermochemical module in the hybrid system; (d) average PV temperature of the reactor; (e) thermal efficiency of the reactor; and (f) methanol conversion ratio of the hybrid system at different simulated solar-thermal energy input levels of the reactor.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

second reason is the relatively good match between the operating temperature ranges of the PV and the thermal modules, while considerable efficiencies are achievable by each module alone. The third reason for the high efficiency, which is related to the thermal modules mentioned above, is the upgrade from low-grade thermal energy (e.g., 250 °C from the PV cells) to high-grade chemical energy in the syngas at the expense of the work availability decrease of methanol as it decomposes. In Fig. 10a, it is also observed that the NSE efficiency first increases and then decreases as methanol flow rate increases. Although a similar trend is observed for different reactor energy input levels, the occurrence of the peak efficiency shifts to higher methanol flow rates as the energy input increases. Because the generated electricity originates from the PV and thermodynamic modules (Eq. (20)), the following analysis investigates the factors affecting these two components of the NSE efficiency to determine the causes of these trends. The PV-module efficiency (Fig. 10b) is selected as the performance indicator of the PV module, and is influenced by the PV temperature. For example, with the PV temperature increasing from 25 to 200 °C, the PV-module efficiency decreases from 35% to 28% and simultaneously the PV-system efficiency decreases from 28% to 22% with an optical loss of 20%, which is unavoidable for a concentrated PV system. Since the heater temperature is constant at 75 °C, the PV-module efficiency is mainly affected by the reactor temperature. Fig. 10d shows that for a constant reactor energy input, the average PV temperature decreases as methanol flow rate increases, which results from the cooling effect due to methanol heating and decomposition. The lower temperature of the PV cells causes the PV-module efficiency to increase as the methanol flow rate increases. The thermal-electric efficiency (Fig. 10c) is selected as the performance indicator of the thermochemical module. Because the chemical-electric conversion process is performed in the same heat engine with the same fuel (i.e., syngas), the chemical-electric efficiency (gc-e) is constant and the thermal-electric efficiency (gt-e) is defined by the thermal-chemical efficiency (gt-c). Considering the energy balance during the thermal-chemical conversion, the thermal energy from the PV cells is divided into three parts: (1) heat losses to the ambient from the heater and the reactor, (2) heat loss during the recycling of unreacted methanol in the heat exchanger, and (3) a portion of the chemical energy of the syngas due to absorption of solar thermal energy by methanol decomposition. Owing to the constant temperature of the heater, the heat loss and thermal efficiency of the heater remains the same. In contrast, Fig. 10e shows that the heat loss of the reactor decreases and the thermal efficiency increases as methanol flow rate increases because the reactor temperature is reduced (Fig. 10d). The heat loss during methanol recycling (i.e., part 2) is positively correlated to the amount of methanol recycled, which in turn depends on the methanol conversion ratio. In Fig. 10f, we observe that this ratio drops rapidly as methanol flow rate increases due to increasingly insufficient reaction time and the decrease in reactor temperature. Therefore, the heat loss during methanol recycling increases with an elevated methanol flow rate. In summary, the increase in methanol flow rate leads to a decrease in heat loss during methanol decomposition and an increase in heat loss during methanol recycling, resulting in the trend observed in Fig. 10c, where the thermal-electric efficiency first increases and then decreases. The increase of the PV-module efficiency and decrease of the thermal-electric efficiency lead to the peaks of the NSE efficiency shown in Fig. 10a. Subsequently, the influence of the reactor energy input on NSE efficiency is analyzed. As the energy input level increases from 35 W to 55 W, the PV-module efficiency curve (Fig. 10b) drops slightly and shifts towards higher methanol flow rates, which

results from the slowly increasing temperature of the PV cells. On the thermochemical side of the system, the behavior of the methanol conversion ratio curves (Fig. 10f) is similar, i.e., their shape generally remains stable but their position shifts to the right as the energy input increases. This is because higher methanol flow rates are necessary to match the increased thermal energy input. Furthermore, although the heat loss during methanol decomposition increases with the operating temperature, the thermal efficiency curve rises with an elevated energy input level (Fig. 10e) because the energy input increase is faster than the heat loss during methanol decomposition. Consequently, the thermal-electric efficiency increases with elevated reactor power. The plateauing decrease of the PV-module efficiency and the faster increase of the thermal-electric efficiency lead to slightly smaller changes of the NSE efficiency with the energy input variation. This relatively stable NSE efficiency (36–41%) could signify design and operation advantages. For system designs, the reactor input energy is influenced by the total solar energy input (Eq. (2)), which in turn is defined by the concentration ratio, assuming a constant PV area and direct nominal irradiation (DNI). This indicates that a wide range of concentration ratio adjustment (e.g., 500–1000) is permitted. In practice, concentration ratio of a system is usually fixed, and reactor energy input changes with DNI. That is, the system can maintain high and stable NSE efficiencies by quickly adjusting methanol flow rates in response to changes in solar radiation (i.e., DNI), thus providing a possible solution to one of the key challenges to solar energy technologies, i.e. instability of solar irradiation. In addition to its stable and high NSE efficiency, the hybrid system can also store a significant amount of solar energy via syngas. As shown in Fig. 11, the energy storage ratio (Eq. (22)) is in the range of 30–45%, indicating that more than one-third of the net electricity generated in the hybrid system originates from chemical energy stored via syngas, which can be fed into heat engines to supply electricity when sunlight is absent (e.g., night or overcast day). Note that the energy storage ratio calculation does not consider the electricity from the methanol chemical energy. If this part was included, the ratio could reach 80%. Fig. 11 also shows that the energy storage ratio decreases with increasing methanol flow rate and increases with enhanced reactor energy input. This is because both the increase in methanol flow rate and the decrease in reactor energy input lead to the decrease in reactor temperature, which is favorable for the PV module but unfavorable for the thermochemical module. Therefore, a high reactor energy input level and a reasonably low methanol flow rate are likely the optimum operating conditions for achieving high NSE efficiencies and high energy storage ratios simultaneously.

0.50

Electricity storage ratio

8

0.45

0.40

0.35

35 W 45 W 55 W

0.30

0.25 0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Methanol flow rate (ml/min) Fig. 11. Energy storage ratio of the hybrid system at different simulated solar thermal energy input levels of the reactor.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077

W. Li et al. / Applied Energy xxx (2017) xxx–xxx

4. Conclusions We have proposed and studied a hybrid solar PVthermochemical energy conversion device combining a pointfocus Fresnel concentrator, a III–V multi-junction PV cell, and a methanol decomposition reactor. Numerical and experimental results show that the net solar-electric efficiency could reach 41%, which results partially from the cascaded utilization of solar energy and partially from the leverage effect that elevates the energy level of low-grade solar heat at the expense of the work availability of methanol. In addition, the hybrid system can operate with a relatively high and stable efficiency range for varying solar irradiation, which results from opposite responses of the PV cells and the thermochemical module to solar irradiation and temperature variations, respectively. Furthermore, the hybrid system exhibits a satisfactory capability to store solar energy at a relatively high percentage (45%) through the solar-driven endothermic reaction of methanol decomposition, which might allow the system to supply electricity from solar energy throughout the day. Lastly, the prototype device presented in this work is easily scalable, which means that the system is suitable for both small-scale scenarios (e.g., household equipment) and much greater power generation needs. The high efficiency, power supply stability, energy storage capability, and flexibility in application scale could make the proposed hybrid device a promising means for effective solar energy utilization in the future. Acknowledgments This study is partially supported by the Chinese Academy of Sciences Innovative and Interdisciplinary Team Award, the National Natural Science Foundation of China under award numbers 51590904 and 51236008, and the Chinese Government Award of the Recruitment Program of Global Experts. References [1] Colangelo G, Favale E, Miglietta P, et al. Innovation in flat solar thermal collectors: a review of the last ten years experimental results. Renew Sustain Energy Rev 2016;57:1141–59. [2] Khan J, Arsalan MH. Solar power technologies for sustainable electricity generation – a review. Renew Sustain Energy Rev 2016;55:414–25. [3] Ganesh I. Solar fuels vis-à-vis electricity generation from sunlight: the current state-of-the-art (a review). Renew Sustain Energy Rev 2015;44:904–32. [4] Lewis NS. Introduction: solar energy conversion. Chem Rev 2015;115 (23):12631–2. [5] Wolf M. Performance analyses of combined heating and photovoltaic power systems for residences. Energy Convers 1976;16(1):79–90. [6] Chow TT. A review on photovoltaic/thermal hybrid solar technology. Appl Energy 2010;87(2):365–79. [7] Gholampour M, Ameri M, Yan J. Energy and exergy analyses of photovoltaic/ thermal flat transpired collectors: experimental and theoretical study. Appl Energy 2016;164:837–56.

9

[8] Herrando M, Markides CN. Hybrid PV and solar-thermal systems for domestic heat and power provision in the UK: techno-economic considerations. Appl Energy 2016;161:512–32. [9] Gazda W, Stanek W. Energy and environmental assessment of integrated biogas trigeneration and photovoltaic plant as more sustainable industrial system. Appl Energy 2016;169:138–49. [10] Qiu Z, Ma X, Zhao X, et al. Experimental investigation of the energy performance of a novel micro-encapsulated phase change material (MPCM) slurry based PV/T system. Appl Energy 2016;165:260–71. [11] Vorobiev Y, González-Hernández J, Vorobiev P, Bulat L. Thermal-photovoltaic solar hybrid system for efficient solar energy conversion. Sol Energy 2006;80 (2):170–6. [12] Kosmadakis G, Manolakos D, Papadakis G. Simulation and economic analysis of a CPV/thermal system coupled with an organic Rankine cycle for increased power generation. Sol Energy 2011;85(2):308–24. [13] Tourkov K, Schaefer L. Performance evaluation of a PVT/ORC (photovoltaic thermal/organic Rankine cycle) system with optimization of the ORC and evaluation of several PV (photovoltaic) materials. Energy 2015;82:839–49. [14] Kwan TH, Wu X. Power and mass optimization of the hybrid solar panel and thermoelectric generators. Appl Energy 2016;165:297–307. [15] Wysocki JJ, Rappaport P. Effect of temperature on photovoltaic solar energy conversion. J Appl Phys 1960;31(3):571–8. [16] Li W, Yong H. Efficient solar power generation combining photovoltaics and mid-/low-temperature methanol thermochemistry. Appl Energy, http://dx.doi. org/10.1016/j.apenergy.2017.05.086. [17] Hong H, Jin H, Ji J, Wang Z, Cai R. Solar thermal power cycle with integration of methanol decomposition and middle-temperature solar thermal energy. Sol Energy 2005;78(1):49–58. [18] Jin H, Sui J, Hong H, et al. Prototype of middle-temperature solar receiver/ reactor with parabolic trough concentrator. J Sol Energy Eng 2007;129 (4):378–81. [19] Liu Q, Jin H, Hong H, et al. Performance analysis of a mid- and low-temperature solar receiver/reactor for hydrogen production with methanol steam reforming. Int J Energy Res 2011;35(1):52–60. [20] Zimmerman R, Morrison G, Rosengarten G. A microsolar collector for hydrogen production by methanol reforming. J Sol Energy Eng 2010;132(1):011005. [21] Gu X, Taylor RA, Morrison G, et al. Theoretical analysis of a novel, portable, CPC-based solar thermal collector for methanol reforming. Appl Energy 2014;119(15):467–75. [22] Gu X, Taylor RA, Li Q, et al. Thermal analysis of a micro solar thermal collector designed for methanol reforming. Sol Energy 2015;113:189–98. [23] Zhao J. Solar radiation transfer and performance analysis of an optimum photovoltaic/thermal system. Energy Convers Manage 2011;52(2):1343–53. [24] Theristis M, O’Donovan TS. Electrical-thermal analysis of III–V triple-junction solar cells under variable spectra and ambient temperatures. Sol Energy 2015;118(7):533–46. [25] Kribus A, Mittelman G. Potential of polygeneration with solar thermal and photovoltaic systems. J Sol Energy Eng 2008;130(1):555–67. [26] Fogler HS. Elements of chemical reaction engineering. New Jersey: Prentice Hall; 1999. [27] Fuller EN, Schettler PD, Giddings JC. New method for prediction of binary gasphase diffusion coefficients. Ind Eng Chem 1966;58(5):18–27. [28] Amphlett JC, Creber KAM, Davis JM, et al. Hydrogen production by steam reforming of methanol for polymer electrolyte fuel cells. Int J Hydrogen Energy 1994;19(2):131–7. [29] Hou Z, Zheng D, Jin H, et al. Performance analysis of non-isothermal solar reactors for methanol decomposition. Sol Energy 2007;81(3):415–23. [30] Jin H, Sui J, Hong H, et al. Experimental investigation of a middle-and-low temperature solar receiver/reactor prototype. Proc ISES World Congr 2007;2008:2206–10. [31] Aspen PlusÒ. Aspen technology, Inc., version 8.2, http://www.aspentech.com/.

Please cite this article in press as: Li W et al. Performance analysis of a photovoltaic-thermochemical hybrid system prototype. Appl Energy (2017), http:// dx.doi.org/10.1016/j.apenergy.2017.05.077