Renewable Energy 137 (2019) 2e9

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Performance evaluation of the senergy polycarbonate and asphalt

carbon nanotube solar water heating collectors for building
integration
Adrian Pugsley*, Aggelos Zacharopoulos, Mervyn Smyth, Jayanta Mondol
Ulster University, Centre for Sustainable Technologies, Jordanstown, Northern Ireland BT37 0QB, UK1

a r t i c l e i n f o a b s t r a c t

Article history: Senergy building integrated solar thermal collectors aim to reduce costs of solar water heating systems
Received 15 July 2017 by utilizing polymer materials and offsetting costs of conventional roofing construction components.
Received in revised form Carbon nanotubes can be added to polymer materials to improve their thermal, optical and mechanical
6 October 2017
properties. Two working prototypes; one Polycarbonate Carbon Nano-Tube (PCNT) collector and one
Accepted 25 October 2017
Asphalt Carbon Nano-Tube (ACNT) collector; have been tested using the solar simulator facility at Ulster
Available online 28 October 2017
University and their performances compared. The PCNT collector is single glazed and uses a twinwall
sheet to act as both the solar absorptive surface and the heat transfer fluid channeling element. The ACNT
Keywords:
Carbon nanotubes (CNT)
collector is unglazed and has an asphalt based absorber with embedded serpentine copper tubing to
Solar water heater channel the heat transfer fluid. Tests were conducted with 800 W m
2 illumination intensity and for
Polymeric solar collector water inlet temperatures between from 23  C and 47  C. The PCNT collector achieved 62% maximum
Polymer absorber collection efficiency compared to 45% for the ACNT collector. The heat loss coefficients were 6.0 and
Nanocomposite 8.1 W m
2 K
1 respectively. The performance of the PCNT collector was similar to benchmark values for
single glazed collectors with selective absorber surfaces. The ACNT collector responded very slowly and
performance was lower than a typical unglazed solar water heater with non-selective absorber due to
the high thermal resistance between the absorber surface and the serpentine tubing.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction mounted on roofs with minimal regard to potential adverse

architectural and aesthetic impacts [1]. Building Integrated Solar
Space heating and domestic hot water in buildings accounts for Thermal Systems (BISTS) utilise collectors to replace roof coverings,
about half of all energy use in Europe and generates significant façade cladding, or elements such as shading devices, to improve
carbon dioxide emissions. Solar thermal systems (STS) have a sig- economic benefits by offsetting construction material costs [2] and
nificant role to play in reducing fossil fuel consumption as they can reducing environmental impacts [3]. Several studies on BISTS [4,5]
be installed on the building and supply heat directly to meet a propose using polymer solar collectors in new architectural forms
proportion of its thermal energy demands. Most STS provide low which offer improved roof and façade integration opportunities.
temperature hot water (50e80  C) for domestic uses and space Most solar collectors are constructed from selectively coated
heating, or very low temperature hot water (30e50  C) for swim- metal absorbers covered by a glazing layer to reduce convective
ming pools, under floor heating, or preheated boiler feedwater. At heat losses. In conventional pressured solar water heating systems
the heart of all STS are the collectors which absorb solar energy and (with antifreeze working fluid) absorber temperatures vary from
convert it to generate heat that is typically dissipated into a around
20  C during winter up to as much as 200  C under
working fluid such as air, water or oil. Single glazed flat plate or stagnation conditions (ie low heat demand periods in summer).
evacuated tube collector types are cost effective but are often Hydraulic system pressures of up to 1.5 bar from water expansion
during normal operation and up to 5 bar during stagnation are
typical [6]. Low cost unglazed collectors made from commodity
polymers are increasingly popular for swimming pool heating.
* Corresponding author.
E-mail address: a.pugsley@ulster.ac.uk (A. Pugsley). Unglazed collectors become inefficient due to heat loss when
1
www.cst.ulster.ac.uk. ambient temperatures are significantly lower than the desired

https://doi.org/10.1016/j.renene.2017.10.082
0960-1481/© 2017 Elsevier Ltd. All rights reserved.
 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9 3

water temperature, especially when located in exposed sites sub- development is the selection of appropriate polymers which ach-
jected to wind. Whilst heat loss causes inefficiency, it also has the ieve sufficient robustness when subjected to stresses associated
benefit of protecting collectors from overheating and high pres- with fluctuating temperatures and pressures. Robustness must be
sures during tagnation. maintained throughout the whole service life, hence aging behav-
Most plastics have thermal conductivities of iour of polymers subjected to temperature fluctuations, prolonged
k z 0.1e0.5 W m
1K
1, radiative heat emissivities of ε z 0.9, and UV light exposure, and high humidity is a key issue [12,17e20]. This
solar absorptance of a z 0.9 when pigmented black. By compari- has significant bearing on the economics of polymeric collectors
son, copper and aluminium used in conventional solar collectors given that low cost commodity plastics generally have inferior
have much better thermal conductivities mechanical properties to more expensive engineering plastics.
(k z 200e300 W m
1K
1) and emissivities (ε < 0.1 when polished) Several authors have investigated fabricating absorbers from
but are reflective (a < 0.5). Metallic absorbers are therefore usually nanocomposites which can be engineered to improve optical,
treated with optically selective coatings which maximise absorp- thermal, and mechanical properties of base polymers. Kim et al.
tion whilst also minimising radiant heat emissions. Most [21] demonstrated that the solar thermal efficiency performance of
commercially available selective coatings use galvanic or high a transparent polycarbonate twinwall absorber could be signifi-
temperature application processes and are therefore unsuitable for cantly improved by adding 2%wt CNTs as a solar absorptive
polymer absorbers, although coatings with a ¼ 0.94 absorptivity pigment. Kurzbo €ck et al. [22] compared random copolymer poly-
and ε ¼ 0.09 emissivity have been successfully applied to extruded propylene samples pigmented with carbon black (1.5%wt) and
twinwall polycarbonate absorbers using sputtering techniques carbon nanotube (CNT) pigments (1.5%wt) and found that CNTs
[7,8]. achieved marginally higher solar absorption performances
A large body of work on glazed polymeric collectors has been (a ¼ 0.96) than carbon black samples (a ¼ 0.95). Abdelal and
undertaken in the last two decades, with considerable advances Taamneh [23] found that the performance of a solar distiller could
achieved under IEA-SHC Task 39 [9,10]. The most extensively be improved by adding CNTs or graphene platelets to epoxy resin
studied polymer absorber design concepts are extruded twinwall or based solar absorber plates, although the specific optical and/or
triplewall type arrangements [7,8,11e15] similar to Fig. 1. These thermal mechanisms underlying the performance increase are
designs utilize the upper surface of the twinwall as the absorber unclear. One possible mechanism is that CNT addition could be
and the internal channel structure as the working fluid contain- beneficial in reducing surface emissivity. Solar selective coatings
ment, which maximises heat transfer area. based on vertically aligned CNT forests on aerogel film substrates
Much of the work on single glazed polymeric collector design were developed and tested by Martinez et al. [24] who report solar
development has focused on preventing damagingly high stagna- absorptivity of a z 0.96 and emissivity of ε z 0.7 for their best
tion temperatures by employing methodologies such as absorber (densified) samples. Similar optical performance values are re-
venting, thermochromic glazing, and Integrated Collector-Storage ported by Mizuno et al. [25] for their roll-pressed sheet and filtrated
concepts which feature high thermal masses [6,15,16]. Several buckypaper CNT forest samples, although much higher emissivities
methods for reducing pressure related stresses in polymer collec- (around ε z 0.9) are reported for spray applied coatings.
tors have been developed, including the use of vented open-loop Several authors [26e28] have shown that carbon nanotubes
hydraulic circuits and also the drainback system concept which (preferably functionalized species) and other engineered graphene
has the added advantage of offering freeze protection without the particles such as nanoplatelets can be added to base polymers to
need for antifreeze chemicals. improve their thermal conductivity. Significant benefits (10-fold
One of the most challenging issues for polymeric solar collector increase bringing conductivities to k z 2e5 W m
1K
1) appear

Cross-section through twinwall absorber

Transparent cover

Air gap
Inlet Outlet
manifold manifold
Extruded polymer twinwall absorber

Insulation and casing

Fig. 1. Typical design of a twinwall polymer absorber solar collector.

4 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9

to be obtainable for relatively low nanoparticle concentrations (1e5

ðTin
Tamb Þ
%wt typically) if particle dispersion and alignment is carefully h ¼ FR ðtaÞ
FR UL (6)
controlled. In most practical applications where it is difficult to I
accurately control particle dispersion and alignment the achievable The key novelty of the Senergy solar collector concept is that the
improvements are only modest (2-fold or smaller increase, absorbers are fabricated from low cost polymer materials with
bringing conductivities to k z 0.2e0.8 W m
1K
1), unless very carbon nanotubes added to increase collection efficiency by
high concentrations (>30%wt) are added. improving the optical absorption and emittance characteristics and
In addition to increasing thermal conductivity, increasing by improving thermal conductivity between the absorbing surface
absorptance and decreasing emissivity in polymer solar absorbers, and the heat transfer fluid. A photograph of the two prototype
CNT addition may also yield other benefits in terms of mechanical collectors examined in this study is given in Fig. 2. The prototypes
strength and thermal stability. Work on PMMA-CNT nano- were fabricated by Queen's University Belfast School of Mechanical
composites by Du et al. [29] suggested increased elastic tensile and Aerospace Engineering in conjunction with Lawell Asphalt
modulus and reduced mass-loss in heated samples and work by Company Limited.
Ormsby et al. [30] indicated that CNTs could improve robustness by The PCNT collector design is based upon the use of poly-
retarding crack propagation. Somewhat contrary findings are carbonate twinwall sheeting to create the solar absorptive surface
however reported by both Kurzbo €ck et al. [22] and Kim et al. [21] and heat transfer fluid channeling elements in a similar manner to
who found that CNT addition reduced the ultimate mechanical that shown in Fig. 1. The absorber was made from two extruded
strength of polypropylene and polycarbonate samples respectively. carbon nanotube polycarbonate (2%wt) sheets which were bonded
The present study aims to complement the work of Kim et al. to spacer elements to form a twinwall structure. The twinwall
[21] and Abdelal and Taamneh [23] by presenting results of tests on absorber was connected to flow and return manifold tubes (see
prototype solar collectors featuring nanocomposite solar absorbers. Fig. 3) and mounted inside an insulated wooden tray base. A single
Two working prototypes; one Polycarbonate Carbon Nano-Tube transparent polycarbonate cover layer was added to provide aper-
(PCNT) collector and one Asphalt Carbon Nano-Tube (ACNT) col- ture protection and heat loss control. The finished active absorbing
lector; have been tested at the solar simulator facility at Ulster surface of the collector was 1480  500 mm.
University and their solar thermal collection efficiency perfor- The ACNT collector design is based upon a simple and robust
mances compared. Results are presented and compared with concept of using asphalt to create the solar absorptive surface. The
published data to demonstrate that collectors with polymer prototype consisted of a 30 mm deep layer of asphalt containing 2%
nanocomposite absorbers can achieve solar thermal efficiencies wt (dispersed within a Ethylene Vinyl Acetate carrier) carbon
approaching those of typical conventional solar collector nanotubes laid on a 50 mm insulation layer. A serpentine coil of
benchmarks. copper tubing was embedded within the asphalt to act as the heat
transfer fluid channeling element (see Fig. 4). The asphalt and
insulation were set into a wooden tray base without any cover
2. Collector design and theory (unglazed). The finished active absorbing surface of the collector
was 1605  600 mm.
The working principle of the Senergy solar water heating col-
lectors is essentially similar to other flat plate solar water heaters in
3. Experimental methodology
that solar radiation incident upon the planar absorbing surface is
converted to thermal energy and collected by the heat transfer fluid
All testing was conducted at the state-of-the-art indoor solar
[31,32]. The total amount of solar energy incident on the aperture
simulator facility at the Centre for Sustainable Technologies labo-
area over the test period can be expressed as:
ratory, Ulster University. The experimental set up consisted of
prototype collectors connected to a hydraulic circuit fed by a the
Q incident ¼ IDt ADt (1)
Julabo F33-MA Refrigerated/Heating Circulator that could provide a
where: closed flow loop under constant, controllable conditions. The
experimental set up is shown in Figs. 5 and 6. Tests were
0 1
Z
tstart ,
B C
IDt ¼ @ I ðtÞ dt A Dt (2)
tend

The useful energy gained by the collector is expressed by:

Q collected ¼ mcp ðTout
Tin Þ (3)

And thus the heat balance of the system (based on the Hottel-
Whillier-Bliss equation) that expresses the thermal performance
of a solar thermal collector under steady state conditions is given
by:

mcp ðTout
Tin Þ ¼ AFR ½IðtaÞ
UL ðTin
Tamb Þ (4)

Therefore, the collector efficiency is defined as:

mcp ðTout
Tin Þ

h¼ (5)
IA
Or substituting Equation (4): Fig. 2. Senergy PCNT (left) and ACNT (right) solar water heating collectors.
 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9 5

temperature. Thermocouples supplied with the ACNT collector

were embedded within the asphalt to measure absorber core
temperature. All thermocouples were connected to a stand-alone
Delta-T DL2e data logger to continuously monitor temperatures.
The logged data was transferred to a computer via a data trans-
mission cable for storage.
Manual gravimetric flow meters were employed to monitor
water flow rates in the hydraulic circuit. Tests were undertaken
simultaneously on both the PCNT and ACNT with flow rates of
0.99 ± 0.2 L/min and 1.31 ± 0.3 L/min respectively, corresponding to
Fig. 3. Twinwall absorber and inlet manifold of the PCNT prototype. 0.02 kg s
1 m
2 through each collector throughout the measure-
ments, except during the stagnation test. The flow through each of
the collectors was checked periodically using measuring jugs and a
stopwatch.
All tests were conducted under a constant 800 W m
2 solar flux
(±6% uniformity) simulated by a state-of-the-art solar simulator.
This irradiance level was chosen to be broadly representative of
typical global (direct beam plus diffuse) radiation incident on a
horizontal surface during midday hours on a clear sunny summer
day in Northern Europe [32].
The simulator lamp array consists of 35 high power metal halide
lamps arranged in 7 rows of 5 lamps each. Each lamp is equipped
with a rotation symmetrical paraboloidal reflector to provide a light
beam of high collimation. In order to achieve uniform distribution
of light intensity on the test area, a lens is inserted in each lamp to
widen the illumination of light. The combination of reflector-
characteristics, lens and lamps ensures a realistic simulation of
the beam path, spectrum and uniformity [33].
Fig. 4. Wooden base tray of ACNT prototype before pouring asphalt to form absorber Water supply temperatures at the collector inlets were varied
and encapsulate the serpentine copper tube. from 23  C (equal to ambient temperature) in steps of 4e5  C, up to
a final inlet temperature of 47  C. Stagnation tests were undertaken
at the end of the test to estimate the maximum absorber surface
undertaken using water as the working fluid, with no antifreeze temperature reached by each collector in the absence of water flow.
agents added. The test setup was operated for 2½ hours prior to taking any
Pairs of T-type copper-constantan thermocouples, which had a measurements in order to allow for stabilisation of the irradiance
standard uncertainties of ±0.3  C between 0  C and 100  C were levels, ambient temperatures and the initial thermal mass of the
used to measure the temperature of solar collector inlet and outlet ACNT collector. Further stabilisation periods (typically about 30 min
flows, the absorber surface temperature and the ambient air tem- per temperature step) were factored into the test programme to
perature. Thermocouples supplied with the PCNT collector were ensure that all step measurements would be representative of
attached to the absorber in order to measure absorber surface reasonably steady state conditions. Axial fans were used to provide

Fig. 5. Schematic detail of test facility and instrumentation (two collectors, side by side).
6 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9

4. Results and discussion

Figs. 7 and 8 illustrate the inlet, outlet, absorber and ambient

temperatures measured for both collectors, based on arithmetic
averages of results from the thermocouple pairs. Step changes to
the inlet water temperature are apparent, together with the sub-
sequent stabilisation periods.
Unfortunately, a leak occurred in the PCNT collector sometime
after 16:30 (Fig. 8) so results corresponding to inlet water tem-
perature of 47  C are unreliable for this collector due to the dis-
rupted flow rate. After 16:45 the collectors were largely emptied of
water and the subsequent data therefore relates to a stagnation
condition, characterised by a rapid rise in absorber temperatures
(particularly apparent for PCNT collector data denoted by red
dashed lines on Fig. 8). Problems caused by the leaking collector
meant that the experiment was terminated prior to steady-state
stagnation conditions being achieved and hence estimated stag-
nation temperatures shown on the graphs are based on observed
trends.
The comparatively low rate of temperature change for the ACNT
collector comparted to the PCNT collector is indicative of the large
thermal mass associated with the asphalt absorber. This results in a
very slow response to changes in inlet water temperature. A simi-
larly slow response would be expected to changes in solar irradi-
ation conditions which may result in a poor diurnal performance,
especially in climates typified by intermittent cloud cover. Whilst a
degree of thermal inertia can be beneficial, the current ACNT design
appears to carry excessive mass.
Using Equations (1)e(6), the solar thermal collection efficiencies
and heat loss coefficients of the prototypes were determined (Fig. 9
Fig. 6. Photograph showing the PCNT (left) and ACNT (right) collectors being tested and Table 1) for a variety of operating conditions and compared
under the solar simulator. with published benchmark performances for typical solar water
heating collector types. The benchmark performances for unglazed,
low speed airflow across the external surfaces of the prototype single glazed and double glazed flat plate collectors, along with
collectors in the direction from outlet towards inlet, throughout the evacuated tube collectors, were determined from data presented by
entire test period. Twidell & Weir [32], Kalogirou [31], Liang et al. [34] and

70
Ambient (Tamb)
Inlet (Tin)
65
PCNT outlet (Tout)
PCNT absorber surface (Tabs)
60
ACNT outlet (Tout)
ACNT absorber surface (Tabs)
55 ACNT absorber core (Tabs)
Temperature (°C)

50

45

40

35

30

25

20
11:45 12:00 12:15 12:30 12:45 13:00 13:15 13:30 13:45 14:00
Time (hh:mm)

Fig. 7. Measured temperature time histories under lowest inlet temperature conditions.
 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9 7

Fig. 8. Measured temperature time histories under highest inlet temperature and stagnation conditions.

Fig. 9. Comparison of instantaneous efficiencies of Senergy prototypes against other solar collector benchmarks.

Table 1
Comparison of Senergy prototypes against performance benchmarks for other solar collector types.

Collector Maximum efficiency Typical efficiency at Heat loss

(FRta) at [Tin
Tamb]/I ¼ 0 m2 K,W
1 [Tin
Tamb]/I ¼ 0.05 m2 K,W
1 (FRta, W$m
2 K
1)

ACNT 45% 5% 8.1

PCNT 62% 32% 6.0
Unglazed (matt black) 90% 0% 18.0
Single glazed (matt black) 83% 42% 8.3
Single glazed (selective) 81% 56% 5.1
Double glazed (matt black) 78% 55% 4.6
Evacuated tube (selective) 59% 51% 1.7
8 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9

Giovannetti et al. [35]. components. Given the early stage of development of the PCNT
The measured zero temperature rise efficiencies (the h0 ¼ FRta prototype it is not yet possible to offer reliable comments on the
values apparent from the y-axis intercepts on Fig. 9) of the PCNT anticipated cost of a finished product. The above analysis will
and ACNT collectors are broadly similar to the benchmark for however aid the establishment of a suitable price-point against
evacuated tube collectors but considerably lower than achieved by which design development options can be evaluated in future.
most other flat plate collectors. The ACNT collector appeared to
have reasonably good light absorption characteristics (black 5. Conclusions
bitumen absorber surface with a slight sheen) and is unlikely to
have suffered from significant reflection losses (as it was unglazed) The performance of the Senergy polycarbonate (PCNT) and
which suggests that the low measured efficiency (h0 ¼ 45% asphalt (ACNT) carbon nanotube solar water heating collectors was
compared to h0 ¼ 90% for the unglazed collector benchmark) is investigated experimentally under controlled solar simulated
primarily attributable to a low heat removal factor. The relatively conditions. The single glazed PCNT collector achieved higher
low measured efficiency for the PCNT collector (h0 ¼ 62% compared collection efficiencies than the unglazed ACNT collector under all
to h0 ¼ 83% for the single glazed matt black collector) may also be tested conditions. Test results have been compared to published
partially attributable to a low heat removal factor, although optical benchmark performances for flat plate and evacuated tube solar
losses (absorber and cover reflectivity and low optical trans- water heaters. Both collectors achieved measured zero temperature
missivity of the semi-translucent cover) seem likely to be the more rise efficiencies (h0 ¼ 45% for ACNT and h0 ¼ 62% for PCNT) of
significant causes. Absorbers made of polymeric materials often similar magnitude to the benchmark for evacuated tube collectors.
have lower heat removal factors than conventional metal sheet- The measured heat loss coefficient for the ACNT collector
and-tube absorbers owing to inherently lower thermal conductiv- (8.1 W m
2 K
1) is significantly better than the benchmark for an
ities. The very low apparent heat removal factor for the ACNT is unglazed collector. The measured heat loss coefficient for the PCNT
likely to be caused by the heat exchanger pipework being buried collector (6.0 W m
2 K
1) is equivalent to that achieved by a con-
too deeply within the asphalt and perhaps also by the pipe length ventional one-cover liquid heater with moderately selective
being too short, which introduces very high thermal resistances absorber. The ACNT collector had a very slow thermal response rate
between the absorber and the working fluid. This is evidenced by due to its high thermal mass and a very low heat removal factor due
the significant difference (10e20  C) between the ACNT's absorber to high thermal resistances within the asphalt absorber.
surface and outlet temperatures and the 5e7  C temperature dif- The ACNT collector would be ineffective for domestic hot water
ference between the core and absorber surface temperatures. production and, even in low temperature applications, its low ef-
The thermal insulation performances (the FRUL values apparent ficiency would most likely render it uncompetitive relative to
from the trendline gradients on Fig. 9) of the prototype collectors existing unglazed polymeric collectors unless it could be produced
are considerably lower (better) than the corresponding bench- at a significantly lower cost.
marks. The ACNT achieved a heat loss coefficient of 8.1 W m
2 K
1 The PCNT collector can achieve 32% solar thermal efficiency
which is less than half the unglazed collector benchmark value when operating under conditions relevant to summertime do-
(18 W m
2 K
1). The PCNT achieved a heat loss coefficient of mestic hot water production (0.05 m2 K W
1). Whilst this perfor-
6.0 W m
2 K
1 which is 28% lower than the benchmark value for mance is lower than that typically achieved by conventional metal-
single glazed collectors with matt black absorbers (8.3 W m
2 K
1) glass flat plate or evacuated tube collectors the PCNT concept could
and approaches the benchmark value of 5 W m
2 K
1 for single be economically attractive if the use of polymer materials enabled it
glazed collectors with selective absorbers (high absorptivity and to be produced at much lower cost, particularly if also designed to
low emissivity). The cause of the low heat loss coefficients cannot replace conventional roofing components.
be reliably determined from the test data but, particularly in the The study indicates that carbon nanotube addition in polymeric
case of the ACNT, this is thought to be primarily due to the low heat solar absorbers may be of benefit, but minimising the thickness of
removal factor. the absorber-to-fluid layer remains a crucial parameter affecting
It is clear from Table 1 that the ACNT collector would be inef- solar thermal collection efficiency. The degree to which the carbon
fective for domestic hot water applications since the efficiency is nanotubes increase the polymer thermal conductivity and benefi-
only ~5% under a solar thermal condition of 0.05 m2 K W
1. This cially affect the optical properties of the absorber surfaces will be
solar thermal condition was chosen to be representative of a determined through further testing as part of the ongoing research.
“typical” summertime domestic hot water heating scenario where
the collector is used to heat water to >60  C when the ambient Acknowledgements
temperature is 20  C and the irradiance is at 800 W m
2. Whilst the
ACNT collector could in principle be useful for applications The authors would like to thank Queens University Belfast and
requiring very low temperatures (eg swimming pool heating or low Lawell Asphalt Company Limited for designing and fabricating the
temperature source for heat pump systems at ~0.01 m2K$W
1) its collector prototypes. This paper has been prepared as part of the
low efficiency would most likely render it uncompetitive relative to Senergy Demonstrator Project which is funded by Invest Northern
existing unglazed collectors unless it could be produced at a Ireland as part of the Centre for Advanced Sustainable Energy
significantly lower cost. (CASE) research initiative. Project partners include Ulster Univer-
By contrast, the PCNT collector would (assuming that leakage sity, Senergy Innovations Ltd, Lawell Asphalt Company Ltd, and
problems were resolved) achieve an efficiency of 32% under a solar Semple & McKillop Consulting engineers.
thermal condition of 0.05 m2 K W
1. This performance is less than
two-thirds that of a conventional collector but the PCNT could Nomenclature
become an economically attractive option if it could be produced at
a cost which was half that of a conventional metal-glass collector
(eg evacuated tubes or flat plate types). Even if it was only possible Abbreviations
to produce the PCNT at two-thirds the cost of a conventional metal- ACNT Asphalt with Carbon Nano Tubes
glass collector, it could still be economically attractive if the design BISTS Building Integrated Solar Thermal System
enabled replacement of conventional roofing construction CNT Carbon Nano Tubes
 A. Pugsley et al. / Renewable Energy 137 (2019) 2e9 9

IEA-SHC International Energy Agency Solar Heating and Cooling polymeric components for a new generation of solar thermal energy systems,
Energy Procedia 48 (2014) 130e136.
programme
[11] C. Cristofari, G. Notton, P. Poggi, A. Louche, Modelling and performance of a
PCNT Polycarbonate with Carbon Nano Tubes copolymer solar water heating collector, Sol. Energy 72 (2) (2000) 99e112.
PMMA Poly(methyl methacrylate) polymer resin [12] A. Olivares, J. Rekstad, M. Meir, S. Kahlen, G. Wallner, A test procedure for
STS Solar Thermal System extruded polymeric solar thermal absorbers, Sol. Energy Mater. Sol. Cell. 92
(2008) 445e452.
UV Ultra Violet [13] G. Martinopoulos, D. Missirlis, G. Tsilingiridis, K. Yakinthos, N. Kyriakis, CFD
modeling of a polymer solar collector, Renew. Energy 35 (2010) 1499e1508.
Latin symbols and subscripts [14] D. Missirlis, G. Martinopoulos, G. Tsilingiridis, K. Yakinthos, N. Kyriakis,
Investigation of the heat transfer behaviour of a polymer solar collector for
Aap (m2) Collector aperture area different manifold configurations, Renew. Energy 68 (2014) 715e723.
cp (J$kg
1K
1) Specific heat capacity of the working fluid [15] A. Piekarczyk, M. Volk, M. Heck, S. Wiesmeier, K. Weiß, M. Ko €hl, Field test for
FR (dimensionless) Heat removal factor polymeric collector models in different climatic locations, Energy Procedia 91
(2016) 384e391.
I (W$m
2) Solar radiation flux incident on collector [16] A. Weber, K. Resch, Thermotropic glazings for overheating protection, Energy
aperture Procedia 30 (2012) 471e477.
m (kg$s
1) Working fluid mass flow rate [17] S. Kahlen, G. Wallner, R. Lang, Aging behavior of polymeric solar absorber
materials e Part 1: engineering plastics, Sol. Energy 84 (2010a) 1567e1576.
Q (J) Thermal energy, where subscripts denote energy [18] S. Kahlen, G. Wallner, R. Lang, Aging behavior of polymeric solar absorber
“incident” on aperture, and energy “collected” by working materials e Part 2: commodity plastics, Sol. Energy 84 (2010b) 1577e1586.
fluid [19] S. Kahlen, G. Wallner, R. Lang, M. Meir, J. Rekstad, Aging behavior of polymeric
solar absorber materials: aging on the component level, Sol. Energy 84
t (s) Time
(2010c) 459e465.
T ( C) Temperature, where subscripts denote water flowing [20] M. Povacz, G. Wallner, M. Grabmann, S. Beißmann, K. Grabmayer,
“in” and “out” of the collector, absorber surface (“abs”), W. Buchberger, R. Lang, Novel solar thermal collector systems in polymer
and ambient environment (“amb”) design e Part 3: aging behavior of PP absorber materials, Energy Procedia 91
(2016) 392e402.
UL (W m
2 K
1) Heat loss coefficient which quantifies [21] S. Kim, J. Kissick, S. Spence, C. Boyle, Design, analysis and performance of a
thermal conductance between the working fluid bulk and polymerecarbon nanotubes based economic solar collector, Sol. Energy 134
the ambient environment (2016) 251e263.
[22] M. Kurzbo € ck, G. Wallner, R. Lang, Black pigmented polypropylene materials
for solar absorbers (SHC 2012), Energy Procedia 30 (2012) 438e445, https://
Greek symbols and subscripts doi.org/10.1016/j.egypro.2012.11.052.
a (dimensionless) Solar absorption coefficient [23] N. Abdelal, Y. Taamneh, Enhancement of pyramid solar still productivity using
absorber plates made of carbon fiber/CNT-modified epoxy composites,
Dt (s) Duration of the test period Desalination 419 (2017) 117e124.
h (dimensionless) Solar thermal collection efficiency. [24] P. Martinez, V. Pozdin, A. Papadimitratos, W. Holmes, F. Hassanipour,
Subscript “0” denotes the zero temperature rise condition G. Dover, A. Zakhidov, Carbon nanotube dry spinnable sheets for solar se-
lective coatings by lamination, Eurasian Chem.-Technol. J. 18 (2016) 241e249.
t (dimensionless) Solar transmission coefficient [25] K. Mizuno, J. Ishii, H. Kishida, Y. Hayamizu, S. Yasuda, D. Futaba, M. Yumura,
K. Hata, A black body absorber from vertically aligned single-walled carbon
References nanotubes, Proc. Natl. Acad. Sci. 106 (15) (2009) 6044e6047.
[26] Z. Han, A. Fina, Thermal conductivity of carbon nanotubes and their polymer
nanocomposites: a review, Prog. Polym. Sci. 36 (2011) 914e944, https://
[1] S. Kalogirou, Chapter 1.1 description of the action TU1205: building integrated
doi.org/10.1016/j.progpolymsci.2010.11.004.
solar thermal systems (BISTS), in: S. Kalogirou (Ed.), COST Action TU1205
[27] W. Hong, N. Tai, Investigations on the thermal conductivity of composites
BISTS e Design and Applications Handbook, COST office, Brussels, Belgium,
reinforced with carbon nanotubes, Diam. Relat. Mater. 17 (2008) 1577e1581.
2017, ISBN 978-9963-697-22-9.
[28] Y. Jia, H. He, Y. Geng, B. Huang, X. Peng, High through-plane thermal con-
[2] Maurer, Smyth, Chapter 2.11 economics, in: S. Kalogirou (Ed.), COST Action
ductivity of polymer based product with vertical alignment of graphite flakes
TU1205 BISTS e Design and Applications Handbook, COST office, Brussels,
achieved via 3D printing, Compos. Sci. Technol. 145 (2017) 55e61.
Belgium, 2017, ISBN 978-9963-697-22-9.
[29] F. Du, J. Fischer, K. Winey, Coagulation method for preparing single-walled
[3] C. Lamnatou, D. Chemisana, R. Mateus, M. Almeida, S. Silva, Review and per-
carbon nanotube/poly(methyl methacrylate) composites and their modulus,
spectives on life cycle analysis of solar technologies with emphasis on
electrical conductivity, and thermal stability, J. Polym. Sci. Part B Polym. Phys.
building-integrated solar thermal systems, Renew. Energy 75 (2015)
41 (2003) 3333e3338.
833e846.
[30] A. Ormsby, T. McNally, C. Mitchell, A. Dunne, Incorporation of multiwalled
[4] M. Buker, S. Riffat, Building integrated solar thermal collectors e a review,
carbon nanotubes to acrylic based bone cements: effects on mechanical and
Renew. Sustain. Energy Rev. 51 (2015) 327e346.
thermal properties, J. Mech. Behav. Biomed. Mater. 2 (2010) 136e145.
[5] C. Cappel, T. Kuhn, C. Maurer, Research and Development Roadmap for
[31] S. Kalogirou, Solar Energy Engineering, Academic Press (Elsevier), London,
Façade-integrated Solar Thermal Systems, Available at:, Fraunhofer ISE, Frei-
2009, ISBN 978-0-12-374501-9.
burg, Germany, 2015 http://publica.fraunhofer.de/documents/N-349494.html.
[32] J. Twidell, T. Weir, Renewable Energy Resources, second ed., Taylor Francis,
[6] T. Ramschak, R. Hausner, C. Fink, Polymeric materials in solar-thermal systems
London, 2006, ISBN 978-0-419-25330-3.
- performance requirements and loads, Energy Procedia 91 (2016) 105e112.
[33] A. Zacharopoulos, J. Mondol, M. Smyth, T. Hyde, V. O'Brien, State of the art
[7] A. Kudish, E. Evseev, G. Walter, T. Leukefeld, Simulation study of a solar col-
solar simulator with flexible mounting, in: Proceedings ISES Solar World
lector with a selectively coated polymeric double walled absorber plate, En-
Congress, 11-14 October 2009, Johannesburg, South Africa, 2009,
ergy Convers. Manag. 43 (2002) 651e671.
pp. 854e863.
[8] M. Rommel, M. Koehl, W. Graf, C. Wellens, F. Brucker, K. Lustig, P. Bahr,
[34] R. Liang, M. Ma, J. Zhang, D. Zhao, Theoretical and experimental investigation
Corrosion-free collectors with selectively coated plastic absorbers, Desalina-
of the filled-type evacuated tube solar collector with U tube, Sol. Energy 85
tion 109 (1997) 149e155.
(2011) 1735e1744.
[9] M. Koehl, M. Meir, P. Papillon, G. Wallner, S. Saile (Eds.), Polymeric Materials
[35] F. Giovannetti, S. Foste, N. Ehrmann, G. Rockendorf, High transmittance, low
for Solar Thermal Applications, Wiley-VCH, Weinheim, Germany, 2012, ISBN
emissivity glass covers for flat plate collectors: applications and performance,
978-3-527-33246-5.
Sol. Energy 104 (2014) 52e59.
[10] M. Koehl, S. Saile, A. Piekarczyk, S. Fischer, Task 39 exhibition e assembly of