Applied Energy: Wenjia Li, Yong Hao
Applied Energy: Wenjia Li, Yong Hao
Applied Energy: Wenjia Li, Yong Hao
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
h i g h l i g h t s
A PVT system integrating PV and thermochemistry is proposed, for the first time.
High efficiency, energy storage ability and temperature insensitivity are achieved.
Net solar-electric efficiency of 43% and overall efficiency of 55% are obtained.
Solar energy storage via syngas enables stable power supply all day long.
a r t i c l e i n f o a b s t r a c t
Article history: A hybrid solar power generation system integrating a solar photovoltaic (PV) module and a solar thermo-
Received 1 February 2017 chemical module is proposed based on methanol thermochemistry. Sunlight is concentrated by trough
Received in revised form 13 April 2017 mirror collectors and partially converted to electricity by PV cells overlain on the surface of a solar ther-
Accepted 9 May 2017
mochemical reactor. An endothermic chemical process of methanol (e.g., decomposition) within the reac-
Available online 1 June 2017
tor then absorbs the ‘‘waste heat” of the PV cells and simultaneously cools the cells. During this process,
the low-level thermal energy from the sunlight is upgraded to high-level chemical energy in syngas,
Keywords:
which is stored and burned to generate electricity when necessary. Analysis indicates that the theoretical
Solar
Photovoltaic
net solar-electric efficiency of the hybrid system could be as high as 45% and relatively insensitive to
Methanol thermochemistry operation temperature and pressure. Additionally, the system exhibit a good potential in solar energy
Hybrid storage, and capable of providing stable electricity supply around the clock. The PV–thermochemical
Power generation hybrid system might suggest a promising approach for efficient and stable power generation from solar
Efficiency energy.
Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction (PV) cells is relatively low. Take crystalline silicon solar cells for
example, although the entire visible part of the solar spectrum
Drastic increase in fossil fuel consumption and the consequen- and some parts of the infrared spectrum fall well within their con-
tial worldwide climate change have lead to a pressing demand on version capability, efficiency of commercially available silicon cells
low-cost, large-scale utilization technologies of renewable energy is usually 10–20% [1], while the majority of solar irradiation is
to gradually phase out fossil fuels in a foreseeable future. Renew- wasted in the form of low-temperature (e.g., 100 °C or lower)
able energy sources that are commensurate with or far beyond thermal energy [2]–longer wavelength radiation that falls below
the scale of human energy demand, such as solar energy, are rare the band gap of silicon does not carry enough energy to excite elec-
and yet carry particularly important historical missions. trons and thus entirely turns into dissipated heat; shorter wave-
Solar energy utilization still faces a major cost barrier that par- length radiation with photonic energy higher than the band gap
tially results from the relatively low conversion efficiency of sun- can only be partially converted to electricity by quantum principles
light to power. Among a portfolio of current solar power underlying photoelectric effects [3] and the rest is still converted to
technologies, the efficiency of practically applicable photovoltaic heat. Although much higher efficiencies have been achieved
through advances in materials and configuration of solar cells
⇑ Corresponding author at: Institute of Engineering Thermophysics, Chinese (e.g., 46.0% cell efficiency for III–V multi-junction solar cells [4]),
Academy of Sciences, Beijing 100190, PR China. waste heat still accounts for more than half of the solar energy
E-mail address: [email protected] (Y. Hao). input. Moreover, not only PV cells are practically used as modules
http://dx.doi.org/10.1016/j.apenergy.2017.05.086
0306-2619/Ó 2017 Elsevier Ltd. All rights reserved.
378 W. Li, Y. Hao / Applied Energy 202 (2017) 377–385
(instead of single cells), but high-efficiency PV cells commonly ORC (due to the low temperature of the heat source, 140 °C).
work in conjunction with concentrating optics (e.g., Fresnel lens; Tourkov et al. [31] expanded this concept aiming at the best sce-
CPV) to enhance power generation performance and lower gener- nario by combining the two technologies. As a result, efficiency
ation cost. These facts, however, lead to a clear distinction between of the system combining a triple-junction PV cell and an ORC is
cell efficiency and system efficiency; the module factor and optical as high as 45%, a significant improvement compared with the effi-
loss become inevitable factors that must be taken into account. For ciency of 34% of triple-junction PV cells at a solar concentration
example, if the optical efficiency of solar collection is 0.85 [5] and ratio of 200 and temperature of 25 °C. However, it must be pointed
the efficiency of a PV module to be 90% of that of a single cell [6], out that the efficiency of 45% did not take the optical and heat
the efficiency of a concentrated PV system consisting of 46.0% effi- losses of the solar collector (including solar concentrator and
cient single cells [4] then translates to 35.2% (practical efficiency), absorber) into account, which, however, significantly impacts the
which means that nearly two-thirds of the incident solar energy is performance of the system. If we took these factors into consider-
dissipated to the ambient. By comparison, concentrated solar ation, the promotion of the hybrid system efficiency would be less
power (CSP) exhibits similarly low or even lower efficiencies than 5% (rather than 11%).
(15% for solar thermal power generation systems with a central Primary reasons for modest efficiency enhancement of PVT sys-
tower receiver concentrator [7]) because significant losses (i.e., tems that combine PV modules and heat engines include negative
irreversibilities) typically occur during capture (e.g., from sunlight temperature coefficient of solar cells [8] and low efficiency of heat
to heat), transport (e.g., with heat transfer fluid), and conversion engines at low temperatures (e.g., 12.3% for 150 °C [32]). The
(e.g., from heat to work) processes of solar energy. potential contribution to efficiency promotion by heat engines is
Cascaded utilization of solar energy integrating solar PV and partially offset by the decline in PV efficiency. Most importantly,
CSP generation might achieve a synergistic effect between the stable power output is essential for solar energy utilization tech-
two approaches, aiming at higher efficiencies beyond the reach nologies, but conventional PVT systems often fall short in this
of using either one alone. The basic rationale is that thermal respect due to the lack of an energy storage component. A new
energy, which typically undermines solar PV generation due to PVT system that is not only capable of delivering high energy con-
adverse temperature effect of solar cells [8], could be utilized for version efficiency at relatively low temperatures, but can also
driving various thermal processes. A photovoltaic/thermal hybrid smooth out influences by both transient fluctuations and long-
solar system (i.e., PVT system) utilizes dissipated thermal energy term instabilities of solar irradiation through energy storage in
from PV modules to supply heat, either for direct heating purposes an appropriate form would be much more desirable.
or to drive thermal power cycles. From the perspective of the ther- In this study, we propose a hybrid PVT system, which is charac-
mal utilization module, PV cells are used as the thermal absorbing terized by high solar-electric efficiency and energy storage capac-
layer [9], with the capability of generating electricity while collect- ity. A temperature-tolerant, high-efficiency PV module and a
ing heat from the sun. The photonic energy of sunlight (at a rela- low-temperature, high-efficiency solar thermal power module,
tively high energy level) is first converted to electricity rather which is also capable of easily storing thermal energy, are both
than being converted directly to heat at a much lower energy level; indispensable for the hybrid system. For the former, GaAs PV cells
therefore, the exergy loss of the collection and conversion pro- meet the requirements; for the latter, a power generation module
cesses could be reduced. The energy level is defined as the ratio with solar methanol decomposition might be an appropriate
of the enthalpy change (Delta_H) to the exergy change (Delta_E) choice.
of a process [10]: The idea of power generation with solar methanol decomposi-
tion was first proposed by Hong and Jin; it combines a solar ther-
DE
A¼ ð1Þ mal driven, mid-temperature (200 °C) endothermic
DH decomposition reaction of methanol (CH3OH ? CO + 2H2,
Theoretical and experimental studies on PVT systems were DH200 C ¼ 97 kJ/mol) with a gas and steam combined cycle (GSCC)
reported as early as the 1970s [11,12]. In such systems, heat [33–35]. The net solar-electric (NSE) efficiency of the solar metha-
rejected from PV modules serves as a heat source for space- or nol decomposition system [33] is as high as 35% at a temperature
water-heating [13,14]. Analytical studies [15,16] and numerical of merely 230 °C. The high efficiency is mainly due to the compre-
modeling studies [17] on PVT systems were performed and verified hensive utilization of physical and chemical energy (Fig. 1). From
by experiments. Theoretical and experimental studies were carried the perspective of physical energy utilization, methanol decom-
out on the influences of PV cooling channel geometrical parame- poses at 200 °C and above; the low-energy level (0.36–0.40) of
ters [18,19], PVT collector orientation [20], cooling fluid flow rate solar thermal energy is converted to the high energy level (0.95)
[21–23], cooling fluid temperature [23], frame shadow [24], and
solar irradiation [21,22] on PV temperature, system energy effi-
ciency, and exergy efficiency. System efficiencies based on the first
law of thermodynamics could reach as high as 60–80% [25,26].
However, from the perspective of the second law of thermodynam- 1.0 B Methanol
ics, it would be much more desirable for such systems to have effi- D
Syngas C
cient power generation as energy output, because not only the
applicability of thermal energy is limited by many factors (e.g.,
Energy level
0.8
space, time), but the energy level of PV-rejected thermal energy Heat at 1300
A
is also quite low. Methanol decomposition: A, B
To this end, the concept of a hybrid power generation system 0.6 Syngas combustion: C
combining a heat engine and a photovoltaic module was put for- Methanol combustion: D
ward by Chubb et al. [27] and analyzed by other researchers later
[6,28,29]. Simulation of a hybrid system combining Si-based CPV
0.4
with organic Rankine cycle (ORC) was carried out and the effi- Heat at 200
ciency was enhanced from 9.81% to 11.83% [30]. The relatively
low efficiency is mainly due to the low efficiency of the PV module Fig. 1. Energy levels of methanol chemical energy, syngas chemical energy and
(Si solar cells with a cell efficiency of 12% at 25 °C) and that of the thermal energy at 200 °C and 1300 °C.
W. Li, Y. Hao / Applied Energy 202 (2017) 377–385 379
chemical energy of syngas. The solar thermal energy is then analysis of the hybrid system shows that the synergy between
released at the syngas combustion temperature (e.g., 1300 °C with the PV and thermochemical power generation components could
energy level of 0.81) for power generation. As for chemical energy lead to a win-win situation in terms of comprehensive generation
utilization, direct combustion of methanol for power generation efficiency. Implications of key features of the novel system are dis-
[36,37] leads to significant exergy losses due to the difference cussed shortly.
between chemical energy level of methanol (i.e., 1.03 [33]) and
thermal energy level (e.g., 0.81) of the combustion products (at 2. Description of the system
1300 °C), which is the largest exergy loss in power generation sys-
tems [38]. Splitting methanol into syngas before combustion can The schematic diagram of the hybrid system and flow diagrams
degrade the methanol energy level to the syngas energy level of the hybrid and reference systems are shown in Figs. 2 and 3,
(e.g., 0.95); the latter is closer to (and yet still above) the energy respectively. The major difference between the hybrid and refer-
level of thermal energy at combustion temperature and the exergy ence systems is that the absorbers of the former are covered by
loss during combustion is reduced by up to 7%, which accounts for PV modules. In the hybrid system (Fig. 3a), methanol is sequen-
up to 23.6% of the exergy loss during methanol combustion [39]. tially pumped from its storage tank into heat exchanger A, heater
Methanol solar thermochemistry makes it possible for low-level A, heat exchanger B, and heater B. Methanol is heated and/or
solar thermal energy to be upgraded to and utilized at an elevated vaporized by dissipated heat from PV modules and heat of reaction
energy level and to generate power at a significantly higher effi- products as well as unreacted methanol. Trickling out from heater
ciency. However, considerable irreversibility might still be B, methanol flows to the reactor, in which a part of the methanol is
incurred during key steps such as collection, absorption, and con- decomposed into syngas by the dissipated heat from PV modules.
version of solar energy. Reducing the irreversibility during such The mixture of methanol and syngas is cooled in heat exchangers
processes is important to further improve the efficiency of solar A and B and finally flows into the syngas storage tank, in which
power generation. One possible approach would be revisiting the the mixture is cooled to ambient temperature and methanol is con-
idea of PVT but replacing ‘‘thermal” with ‘‘thermochemical” for densed and recycled to the methanol tank. In the end, the syngas in
the ‘‘T” part to help the thermal module of PVT systems out of the tank is compressed and fed into the GSCC as fuel. To ensure the
the low-efficiency dilemma of heat engines driven by PV-rejected storage capacity of the hybrid system, the system is so designed
heat directly. In addition, the integration of thermochemistry into that syngas generated during daytime can last GSCC consumption
a PVT system essentially adds energy storage capability as well, a for up to 24 hours. The GSCC operates at rated operation conditions
feature very important for addressing the instability issue of solar unless the syngas supply is used up. Note that the GSCC employed
energy, which however was usually not emphasized or even in this system could be replaced by internal combustion engines,
ignored in previous PVT systems. The conceptual PVT system pro- fuel cells or other devices fueled by syngas. Thus, the hybrid sys-
posed above could use syngas-driven power generation tem might be suitable for a large-scale, centralized power genera-
approaches (e.g., internal combustion engines, or fuel cells) to tion plant with a GSCC and/or a distributed power generation
quickly offset fluctuations in PV outputs during daytime [40]; system with internal combustion engines/fuel cells.
while completely rely on syngas power generation at nighttime In the heaters (A and B) and reactor, sunlight is focused onto the
or whenever needed. surface of PV modules by trough concentrators. Temperature for
Given the complementary characteristics of PV and thermo- methanol decomposition in the hybrid system is 150–250 °C, the
chemical components with respect to temperature and energy same as that in Ref. [33]; the PV component employed in the
storage, we hereby propose a hybrid PVT system (Figs. 2 and 3a) hybrid system are InP/InGaAs/Ge triple-junction PV cells, which
in this study, integrating a GaAs PV module and a power genera- can tolerate temperatures up to 250 °C [41].
tion module with solar methanol decomposition. For comparison,
a solar thermochemical system similar to the system in Ref. [33] 3. Analytical methodology
is introduced as a reference system (Fig. 3b). Major differences
between them include the PV module and recycling of unreacted Thermodynamic modeling and simulation are carried out on the
methanol in the system proposed in this study. Thermodynamic hybrid and reference systems.
The absorber tube is discretized into control volumes (Fig. 4)
and a random one is selected for study; its temperature is denoted
as Ti.
Sunlight is first concentrated by a trough solar concentrator.
Due to optical losses during concentration, only a part of the sun-
light is absorbed by the PV modules; it is given as:
dQ abs ðT i Þ ¼ dQ tot ðT i Þ gopt ; ð2Þ
where dQ tot ðT i Þ is the total solar energy input of the point before
concentration and gopt is the collector optical efficiency. One part
of the absorbed sunlight is converted to electricity by the PV mod-
ule; the electrical power output is given as:
dW PV ðT i Þ ¼ dQ abs ðT i Þ gT i ;PV ; ð3Þ
Fig. 3. Flow diagrams of (a) the hybrid system, and (b) the reference system.
Fig. 4. Diagram of energy transfer processes and energy balance on a selected control volume.
efficiency and individual PV cell efficiency (0.9) [6]. The individual shown in Tables 1 and 2, respectively. The rest of the absorbed sun-
PV cell efficiency at Ti is given as [6]: light is dissipated in the form of heat, and is given as:
Table 1 Table 2
Efficiencies of the PV cell in Ref. [6] at different temperatures. Efficiencies of the PV cell in Ref. [4] at different temperatures.
Temperature (°C) PV efficiency Temperature (°C) PV efficiency Temperature (°C) PV efficiency Temperature (°C) PV efficiency
25 0.322 150 0.275 25 0.389 150 0.342
50 0.313 175 0.266 50 0.380 175 0.333
75 0.304 200 0.257 75 0.370 200 0.323
100 0.294 225 0.247 100 0.361 225 0.314
125 0.285 250 0.238 125 0.351 250 0.304
W. Li, Y. Hao / Applied Energy 202 (2017) 377–385 381
ðT i þ DTÞ4 T 4sky 71:4 These equations describe how solar energy is absorbed and
dQ rad;loss ðT i Þ ¼ dQ tot ðT i Þ eab b ; ð8Þ stored in the syngas. The whole process, including methanol
DNI C
warm-up, methanol decomposition, methanol recycling, syngas
where eab is the emissivity of the absorber and Tsky is the sky tem-
storage, syngas combustion, and power generation is simulated
perature, which is given as [43]:
using Aspen Plus [44]. The schematic diagram of the simulation
T sky ¼ T a 8; ð9Þ is shown in Fig. 5.
The power generation efficiency of the hybrid system is defined
The heat loss parameters are: a = 1.92 10 W K m2, 2 1
as:
b = 2.02 109 W K4 m2, and c = 6.61 103 J K1 m3 [42].
The thermal energy absorbed by methanol is the difference W PVþME
ge ¼ ; ð16Þ
between the dissipated solar energy and the heat loss, which is Q tot þ n_ LHV
given as:
where WPV+ME is the overall work output of the hybrid system,
dQ ðT i Þ ¼ dQ diss ðT i Þ dQ con;loss ðT i Þ dQ rad;loss ðT i Þ ð10Þ which is the sum of the electricity from the PV modules and GSCC;
and LHV is the lower heating value of methanol.
The above-mentioned analysis describes how the sunlight is Similar to other solar hybrid power plants, the NSE efficiency is
utilized by the PV modules and methanol at one point with tem- defined as [45]:
perature Ti. To describe how the sunlight is utilized in the entire
W PVþME W ME
hybrid system, the point-wise analysis stated above is taken to tra- gse;PVþME ¼ ; ð17Þ
verse individual points of the system. The next analysis is based on Q tot
methanol flowing through the whole system. where W ME is the work output of GSCC with the same amount of
The thermal energy absorbed by methanol either raises metha- methanol as the input.
nol temperature and vaporizes methanol in the heaters or decom- To compare the performance of the hybrid system (Fig. 3b) and
poses methanol in the reactor. When the thermal energy absorbed that of the reference system (Fig. 3a), the NSE efficiency of the ref-
by methanol raises the methanol temperature from Ti to Ti + dT in erence system is also defined:
heaters A and B, the thermal energy absorbed by methanol is given
as: W ref W ME
gse;ME ¼ ; ð18Þ
8 Q 0tot
_
< n C P;T i dT
> ðT in;A ; T out;A Þ [ ðT in;B ; T out;B Þ=T v;P
where Wref is the overall work output of the reference system with
dQ ðT i Þ ¼ n_ r fT v;P g ; ð11Þ
>
:_ the same amount of methanol as the input. To compare the perfor-
n DHP;T out;B fT out;B g
mance of the hybrid system with that of the PV-only and methanol-
only power generation systems, the electricity increment ratio is
where n_ is molar methanol flow rate, C P;T i is heat capacity of metha-
defined as:
nol when methanol is at Ti, r is latent heat of methanol vaporization,
Tin,A and Tin,B are inlet temperatures of methanol in heaters A and B, W PVþME W PV W ME
rincr ¼ ; ð19Þ
respectively, Tout,B and Tout,B are outlet temperatures of methanol in W PV þ W ME
heaters A and B, respectively, T v;P is vaporization temperature of
where WPV is the work output of the PV-only system (its area is the
methanol at pressure P, and DHP;T out;B is enthalpy change when
same as that of the hybrid system).
methanol decomposes and reaches an equilibrium at pressure P
To explore the reason for NSE efficiency promotion in the
and temperature Tout,B.
hybrid system, the exergy loss reduction between the hybrid and
When methanol flows through heaters (A, B) and reactor, the
reference systems is defined as the exergy difference between
total thermal energy absorbed by methanol is given as:
(R the electricity and exergy of waste heat during sunlight collection.
R
dQ ðT i Þ ¼ n_ C P;T i dT þ n_ DHP;T out;B þ n_ r ðT v;P 6 T out;B Þ Furthermore, the exergy loss reduction ratio is defined as:
Q me ¼ R R Z
dQ ðT i Þ ¼ n_ C P;T i dT þ n_ DHP;T out;B ðT v;P > T out;B Þ T a þ 273:15
1 W PV ðT i Þ 1 1
ð12Þ Exheat T i þ 273:15
rlos;red ¼
DNI C A0 DNI C A
Accordingly, the total solar energy input before concentrating, Z
T a þ 273:15
when methanol flows through the system, is given as: 1 dA gopt gPV ðT i Þ
Z Exheat T i þ 273:15
¼ ; ð20Þ
Q tot ¼ dQ tot ðT i Þ ð13Þ DNI C A0 A
Table 3 are achieved using the PV cells from Ref. [6] when the reaction
Summary of key parameters. temperature is 225 °C and the reaction pressure is 7 bar.
Parameters Value To evaluate the benefits of the concept of integration between
DNI 800 W/m2
the PV-only and reference systems, a comparison between the
Emissivity of PV modules (eab) 0.5 hybrid system and the reference system is performed. The NSE effi-
Absorptivity of PV modules (a) 1 ciency of 40.2–42.9% of the hybrid system is approximately one-
Collector optical efficiency (gopt) 0.73 fifth higher than that of 32.0–35.9% of the reference system
Velocity of wind (Vwind) 4 m/s
(Fig. 6); if a 46.0% efficiency PV cell [4] were used instead, the effi-
Concentrating ratio (C) 80
Reaction pressure 3.0 105–2.3 106 Pa ciency of the hybrid system could be further boosted from 42.9% to
Reaction temperature 225 °C 45.4%. However, the seemingly marginal difference between the
Turbine inlet temperature 1250 °C NSE efficiency of 45.4% of the hybrid system and the efficiency of
Pressure ratio of gas turbine 17 46.0% of the PV cells disguises the real difference in practically
Fuel inlet pressure 2.5 106 Pa
Ambient temperature (Ta) 15 °C
achievable efficiencies at the system level; PV cells have to work
Exhaust temperature from HRSG 85 °C as modules and multi-junction cells have to work together with
Temperature of syngas storage tank 15 °C solar concentration systems, which inevitably incurs a modular
Minimum heat transfer temperature difference (DT) 10 °C efficiency coefficient of 0.9 and an optical efficiency of 0.85
[5] (both multiplied to single-cell efficiency). When taking both
factors into account, we reach a practically achievable (i.e.,
system-level) efficiency of 35.2% for the PV cells with 46.0%
4.1. Performance of the hybrid system single-cell efficiency. On the other hand, the 45.4% peak efficiency
of the hybrid system already takes both factors into account, and
4.1.1. Efficiency promotion resulting from integration the true difference therefore is that the NSE efficiency of 45.4% of
Numerical simulation is carried out with the model and param- the hybrid system is more than one-quarter higher than the PV-
eters described above. Fig. 6 shows the NSE efficiency of the hybrid only system efficiency of 35.2%.
and reference systems and various components of electricity out- The significant promotion in efficiency compared with the ref-
put. The NSE efficiency of the hybrid system first increases and erence system is mainly due to the fact that the absorber in the
then declines as the reaction pressure increases. The NSE efficiency hybrid system is covered by PV modules; hence, sunlight is con-
of up to 42.9% and the system electricity efficiency of up to 54.5% verted into electricity first and only the rest degrades to low-
energy level heat, which results in the alleviation of exergy loss
during sunlight collection and thus efficiency promotion in the
hybrid system. According to Eq. (20), the exergy loss reduction
0.80 0.46 ratio reaches up to 49%, indicating that the hybrid system obtains
PV[4]+Methanol almost one half more exergy from solar energy than the reference
Net solar-electric efficiency
PV[6]+Methanol 0.44 system. For the same amount of solar energy input, savings in
0.75 0.42 exergy always lead to more electricity output and thus higher
Electricity ratio
efficiency.
0.40 Compared with the PV-only system, the improvement is
0.70 obtained by taking advantage of the complementary nature
0.38
0.15 between the PV modules and solar thermochemical module and
0.36 the ‘‘leverage” effect (Fig. 1) of the latter, which promotes low-
0.10 level thermal energy to high-level chemical energy (e.g., syngas)
Methanol 0.34 at the cost of even higher-level chemical energy (e.g., methanol).
Solar thermal A comparison of the electricity generation between the tandem
0.05 0.32
Solar PV hybrid system and parallel system (i.e., a PV-only system and a
Methanol methanol-only system arranged in parallel), with the area of the
0.00 0.30
3 7 11 15 19 23 collector and amount of methanol being the same, is also carried
out. Selected as the indicator, the electricity increment ratio (Eq.
Pressure/bar
(19)) is in the range of 17–18% while the reaction pressure varies
Fig. 6. Net solar-to-electric efficiency of the hybrid and reference systems; from 3 to 23 bar, indicating that the hybrid system generates more
composition of electricity of the hybrid system. electricity than the parallel one. This is mainly due to two reasons:
W. Li, Y. Hao / Applied Energy 202 (2017) 377–385 383
(1) the solar energy in the hybrid system is utilized by the solar PV an optimization function of both parameters, is plotted in Figs. 7
module and thermal module (i.e., methanol decomposition) as and 8, respectively.
opposed to being only used in a cascaded way by the PV module As can be observed from the dashed lines in Figs. 7 and 8, the
in the PV-only system; (2) the methanol in the hybrid system is optimal NSE efficiencies for different reaction temperatures/pres-
decomposed before combustion, which results in a lower exergy sures remains quite stable (i.e., 42.9%). This implies that the influ-
loss during combustion compared with direct methanol combus- ence of temperature variation can be offset by changing the
tion system. pressure, which is essentially a system operation strategy to coun-
teract solar irradiation fluctuation. On the other hand, the optimal
4.1.2. Fossil fuel saving and energy storage reaction temperature for lower reaction pressure is also lower due
Analysis of the electricity composition reveals that the majority to heat absorption and volume expansion during methanol decom-
of the electricity originates from methanol, the share of which var- position. In other words, it is possible to keep the maximum NSE
ies around three-quarters (73.3–76.6% in Fig. 6), while the electric- efficiency of the hybrid system stable even though temperature
ity from solar energy accounts for approximately one quarter of the and pressure both change, which makes it possible to lower the
total electricity. By comparison, the electricity from solar energy in requirements on system operation, including temperature (related
the reference system (Fig. 3b) only accounts for 16% of the total to solar energy collection) and pressure (related to pump work),
electricity. This means that the hybrid system needs less fossil fuel without sacrificing system performance. According to Leest et al.
to generate the same amount of electricity. It is also noteworthy [46], with PV cells operating at 200 °C for 1320 h, no significant
that the electricity from the PV modules approximately equals that influences on the GaAs structure and PV efficiency are observed.
from solar thermal energy. This indicates that in the hybrid system, Thus, PV temperature in the range of 150–200 °C is reasonable
syngas carrying sufficient solar energy can be stored first during for present PV cells and PV temperature in the range of 200–
daytime (while the PV module simultaneously supplies electric- 250 °C might also be reasonable considering advancement in PV
ity); the energy is later released (e.g., by GSCC) for power genera- technologies. The major limiting factor of the system performance
tion. This means that syngas-driven power generation would then primarily shift to the kinetic side, which might inspire
approaches (e.g., internal combustion engines or fuel cells) could further studies on materials (e.g., catalysts) and reaction mecha-
be used to quickly offset fluctuations in PV outputs during daytime, nisms. For example, the latest studies by Ma et al. [47] demon-
while syngas power generation could be completely relied on at strated decomposition of methanol at a respectable rate with 2%
nighttime or whenever needed. Thus, the goal of steadily supplying Pt/a-MoC catalyst at 150 °C. It could therefore be envisaged that
electricity by solar energy all day could be achieved by using solar with advances in materials science and chemical engineering, the
energy storage via syngas. Furthermore, the solar energy stored in temperature for methanol decomposition to occur at respectable
syngas is relatively stable, which implies that the solar energy can rates could be further lowered. The developments of PV cell and
be stored for a relatively long time, transported to where it is catalyst shall provide a broader and more solid basis for the PV-
needed, and released when it is needed. In contrast, the storage thermal(chemical) hybrid concept presented to be widely applica-
of solar energy in previously described PVT systems based on heat ble with not only III-V solar cells, but other types of solar cells as
engines was not explicitly considered; even if considered, solar well.
energy storage in the form of low-level explicit or latent heat might A close inspection on the influence of temperature shows that
be inefficient and uneconomic. the efficiency of the hybrid system at the lowest temperature of
150 °C monotonically declines with the increase of pressure
(Fig. 7), while the efficiency curves start to peak at appropriate
4.2. Impact of reaction pressure and temperature pressures in the temperature range of 175–250 °C. The same data
set as in Fig. 7 is presented in a different manner in Fig. 8 in order
Major thermodynamic parameters for methanol decomposition to better illustrate the impact of reaction temperature on effi-
in the hybrid system, including reaction pressure and temperature, ciency. When pressure is in the range of 15–23 bar, the NSE effi-
influence methanol conversion rate to syngas, as well as the tem- ciency increases as temperature increases; when pressure is in
perature of the heaters and the reactor. To reveal the impact of the range of 2–11 bar, an optimum NSE efficiency is observed in
both parameters on system performance and determine optimum the range of 175–250 °C and the optimum remains virtually
operating conditions, the NSE efficiency of the hybrid system, as
0.429
Net solar-electric efficiency
Optimal efficiency
0.425 ar
3b
0.40 ar 2bar
0.400 7b
ar 1bar
b
0.375 11
0.36 r
ba
0.350 15
r
0.325 ba
0.32 19
150 r
0.300 175 225 ba
23
200 250
0.275 0.28
1 3 5 7 9 11 13 15 17 19 21 23 150 175 200 225 250
Pressure/bar Temperature/
Fig. 7. Net solar-to-electric efficiency of the hybrid system at different reaction Fig. 8. Net solar-to-electric efficiency of the hybrid system at different reaction
temperatures versus reaction pressure. pressures versus the reaction temperature.
384 W. Li, Y. Hao / Applied Energy 202 (2017) 377–385
0.8 5. Conclusions
decrease of exergy losses during solar energy collection and con- [19] Zhou J, Zhao X, Ma X, Qiu Z, Ji J, Du Z, et al. Experimental investigation of a solar
driven direct-expansion heat pump system employing the novel PV/micro-
version by way of both the power generation of solar cells and
channels-evaporator modules. Appl Energy 2016;178:484–95.
endothermicity of methanol decomposition. The second reason is [20] Kostić LT, Pavlović TM, Pavlović ZT. Optimal design of orientation of PV/T
the ‘‘leverage” effect of energy levels between high energy level collector with reflectors. Appl Energy 2010;87(10):3023–9.
fossil fuel (i.e., methanol) and low energy level thermal energy [21] Chow TT, Pei G, Fong KF, Lina Z, Chana ALS, et al. Energy and exergy analysis of
photovoltaic–thermal collector with and without glass cover. Appl Energy
from solar cells; the latter is upgraded to the energy level of syngas 2009;86(3):310–6.
chemical energy, at the cost of the decrease of methanol energy [22] Kaiser AS, Zamora B, Mazón R, García JR, Vera F. Experimental study of cooling
level during methanol decomposition. The hybrid system also BIPV modules by forced convection in the air channel. Appl Energy
2014;135:88–97.
exhibits its capability of uninterrupted and stable electricity sup- [23] Tiwari A, Dubey S, Sandhu GS, Sodhac MS, Anward SI. Exergy analysis of
ply due to the combination of solar PV cells (for daytime power integrated photovoltaic thermal solar water heater under constant flow rate
generation) and solar energy storage through syngas (for nighttime and constant collection temperature modes. Appl Energy 2009;86
(12):2592–7.
power generation). Sensitivity analysis of key operation parame- [24] Wang Y, Pei G, Zhang L. Effects of frame shadow on the PV character of a
ters of the hybrid system indicates its relative insensitivity to the photovoltaic/thermal system. Appl Energy 2014;130:326–32.
operation pressure and temperature, which brings tremendous [25] Coventry JS. Performance of a concentrating photovoltaic/thermal solar
collector. Sol Energy 2005;78(2):211–22.
benefits in system design and system operation, due to opposite [26] Henning H, Bett AW, Jürgen P, Carsten A. Modeling of concentrating
responses of PV cells and methanol thermochemistry to tempera- photovoltaic and thermal systems. Prog Photovolt Res Appl 2014;22
ture. Based on all above-mentioned advantages, the PV–thermo- (4):427–39.
[27] Chubb L, Donald. Performance characteristics of a combination solar
chemical hybrid system proposed in this work could provide a
photovoltaic heat engine energy converter. In: Proceedings of the 22nd
promising approach for effective solar energy utilization in the intersociety energy conversion engineering conference, Philadelphia, 10–14
near future. August 1987. p. 254–63.
[28] Vorobiev Y, González-Hernández J, Vorobiev P, Bulat L. Thermal-photovoltaic
solar hybrid system for efficient solar energy conversion. Sol Energy 2006;80
Acknowledgments (2):170–6.
[29] Han X, Xu C, Ju X, Yang Y. Energy analysis of a hybrid solar concentrating
This study is partially supported by the National Natural photovoltaic/concentrating solar power (CPV/CSP) system. Sci Bull 2015;60
(4):460–9.
Science Foundation of China under award number 51590904 and [30] Kosmadakis G, Manolakos D, Papadakis G. Simulation and economic analysis
51236008, partially by the Chinese Academy of Sciences Innovative of a CPV/thermal system coupled with an organic Rankine cycle for increased
and Interdisciplinary Team Award, and partially by the Chinese power generation. Sol Energy 2011;85(2):308–24.
[31] Tourkov K, Schaefer L. Performance evaluation of a PVT/ORC (photovoltaic
Government Award of the Recruitment Program of Global Experts. thermal/organic Rankine cycle) system with optimization of the ORC and
evaluation of several PV (photovoltaic) materials. Energy 2015;82:839–49.
References [32] Lecompte S, Huisseune H, van den Broek M, Vanslambrouck B, Paepe MD.
Review of organic Rankine cycle (ORC) architectures for waste heat recovery.
Renew Sustain Energy Rev 2015;47:448–61.
[1] International Energy Agency. Technology Roadmap: Solar Photovoltaic Energy;
[33] Hong H, Jin H, Ji J, Wang Z, Cai R. Solar thermal power cycle with integration of
2014.
methanol decomposition and middle-temperature solar thermal energy. Sol
[2] Makki A, Omer S, Sabir H. Advancements in hybrid photovoltaic systems for
Energy 2005;78(1):49–58.
enhanced solar cells performance. Renew Sustain Energy Rev 2015;41:658–84.
[34] Hong H, Liu Q, Jin H. Operational performance of the development of a 15 kW
[3] Hirst LC, Ekins-Daukes NJ. Fundamental losses in solar cells. Prog Photovolt Res
parabolic trough mid-temperature solar receiver/reactor for hydrogen
Appl 2011;19(3):286–93.
production. Appl Energy 2012;90(1):137–41.
[4] Green MA, Keith E, King DL, Warta W, Dunlop ED. Solar cell efficiency tables
[35] Liu X, Hong H, Jin H. Mid-temperature solar fuel process combining dual
(Version 48). Prog Photovolt Res Appl 2016;24(7):905–13.
thermochemical reactions for effectively utilizing wider solar irradiance. Appl
[5] Buonomano A, Calise F, Palombo A. Solar heating and cooling systems by CPVT
Energy 2017;185(Part 2):1031–9.
and ET solar collectors: a novel transient simulation model. Appl Energy
[36] Olah GA. Beyond oil and gas: the methanol economy. Angew Chem 2005;44
2013;103(1):588–606.
(18):2636–9.
[6] Kribus A, Mittelman KG. Potential of polygeneration with solar thermal and
[37] Wang X, Ge Y, Zhang C, Tan J, Hao L, et al. Effects of engine misfire on
photovoltaic systems. J Sol Energy Eng 2008;130(1):555–67.
regulated, unregulated emissions from a methanol-fueled vehicle and its
[7] Gupta MK, Kaushik SC, Ranjan KR, Panwar NL, Reddy VS, et al. Thermodynamic
ozone forming potential. Appl Energy 2016;177:187–95.
performance evaluation of solar and other thermal power generation systems:
[38] Hong H, Jin H, Sui J, Ji J. Mechanism of upgrading low-grade solar thermal
a review. Renew Sustain Energy Rev 2015;50:567–82.
energy and experimental validation. J Sol Energy Eng 2008;130(2):247–66.
[8] Wysocki JJ, Rappaport P. Effect of temperature on photovoltaic solar energy
[39] Jin H, Hong H, Cai R. A chemically intercooled gas turbine cycle for recovery of
conversion. J Appl Phys 1960;31(3):571–8.
low-temperature thermal energy. Energy 2006;31(10):1554–66.
[9] Chow TT. A review on photovoltaic/thermal hybrid solar technology. Appl
[40] Mazzola S, Astolfi M, Macchi E. The potential role of solid biomass for rural
Energy 2010;87(2):365–79.
electrification: a techno economic analysis for a hybrid microgrid in India.
[10] Ishida M, Kawamura K. Energy and exergy analysis of a chemical process
Appl Energy 2016;169:370–83.
system with distributed parameters based on the enthalpy-direction factor
[41] Takamoto T, Agui T, Kaneiwa M. Annual output estimation of concentrator
diagram. Ind Eng Chem Process Des Develop 1982;21(4):690–5.
photovoltaic systems using high-efficiency InGaP/InGaAs/Ge triple-junction
[11] Wolf M. Performance analyses of combined heating and photovoltaic power
solar cells based on experimental solar cell’s characteristics and field-test
systems for residences. Energy Convers 1976;16(1):79–90.
meteorological data. Sol Energy Mater Sol Cells 2006;90(1):57–67.
[12] Florschuetz LW. Extension of the Hottel-Whillier model to the analysis of
[42] Odeh SD, Morrison GL, Behnia M. Modelling of parabolic trough direct steam
combined photovoltaic/thermal flat plate collectors. Sol Energy 1979;22
generation solar collectors. Sol Energy 1998;62(6):395–406.
(4):361–6.
[43] Zhao Y. Thermodynamics investigation on integrating mid-temperature solar
[13] Sun W, Ji J, Luo C, He W. Performance of PV-Trombe wall in winter correlated
thermal energy and coal-fired power system [diss.]. University of Chinese
with south façade design. Appl Energy 2011;88(1):224–31.
Academy of Sciences; 2012.
[14] Chen H, Zhang L, Jie P, Xiong Y, Xu P, et al. Performance study of heat-pipe solar
[44] Aspen PlusÒ. Aspen technology, Inc., version 8.2. <http://www.aspentech.com/>.
photovoltaic/thermal heat pump system. Appl Energy 2017;190:960–80.
[45] Buck R, Bräuning T, Denk T, Pfänder M, Schwarzbözl P, et al. Solar-hybrid gas
[15] Gaur A, Tiwari GN. Analytical expressions for temperature dependent
turbine-based power tower systems (REFOS). J Sol Energy Eng 2002;124
electrical efficiencies of thin film BIOPVT systems. Appl Energy
(1):2–9.
2015;146:442–52.
[46] Leest RH, De KK, Bauhuis GJ, Mulder P, Cheun H, et al. Degradation mechanism
[16] Dubey S, Sandhu GS, Tiwari GN. Analytical expression for electrical efficiency
(s) of GaAs solar cells with Cu contacts. Phys Chem Chem Phys 2016;18
of PV/T hybrid air collector. Appl Energy 2009;86(5):697–705.
(15):10232–40.
[17] Sarhaddi F, Farahat S, Ajam H, Behzadmehr A, Adeli MM. An improved thermal
[47] Lin L, Zhou W, Gao Rui, Yao S, Zhang X, et al. Low-temperature hydrogen
and electrical model for a solar photovoltaic thermal (PV/T) air collector. Appl
production from water and methanol using Pt/a-MoC catalysts. Nature
Energy 2010;87(7):2328–39.
2017;544(7648):80–3.
[18] Teo HG, Lee PS, Hawlader MNA. An active cooling system for photovoltaic
modules. Appl Energy 2012;90(1):309–15.