Applied Energy 195 (2017) 414–425

Contents lists available at ScienceDirect

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

Solar steam generation through bio-inspired interface heating of

broadband-absorbing plasmonic membranes
Xinzhi Wang a, Yurong He a,⇑, Xing Liu b, Gong Cheng a, Jiaqi Zhu b
a
School of Energy Science and Engineering, Harbin Institute of Technology, Harbin, Heilongjiang 150001, People’s Republic of China
b
Center for Composite Materials and Structures, Harbin Institute of Technology, Harbin, Heilongjiang 150001, People’s Republic of China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

Bio-inspired solar steam generation

using floating plasmonic membranes
(PMs) proposed.

High steam generation efficiency of
85% achieved at illumination power
of 10 kW m2.

PMs enhanced the productivity of a
solar still for seawater desalination by
80%.

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

Article history: Efficient solar-enabled evaporation plays a critical role in solar power-based concentration systems,
Received 11 October 2016 photochemical plants, seawater desalination technologies, etc. However, traditional processes for solar
Received in revised form 15 March 2017 steam generation usually depend on high-temperature heating of the bulk liquid, which requires highly
Accepted 16 March 2017
concentrated solar power and suffers from high energy and optical losses. Therefore, the enhancement of
solar steam generation by bio-inspired interface solar heating is proposed in this work. In this study,
easy-to-prepare, flexible, and reusable plasmonic membranes (PMs) were fabricated for realizing the
Keywords:
bio-inspired interface solar heating and continuous steam transportation through the micropores of
Solar steam generation
Plasmonic membranes
the membranes. A solar steam generation efficiency of 85% was achieved at an illumination power of
Heat localization 10 kW m2. The effects of Au concentration in the membranes and optical power on the steam generation
Solar still efficiency were systemically studied. The observed high evaporation rate and efficiency were attributed
Photo-thermal conversion to three main factors: high (90%) and broadband solar absorption, efficient photo-thermal conversion
due to high plasmon dissipation losses, and fast capillary flow in the membrane micropores. Finally,
the application of PMs in a single basin solar still system for seawater desalination was investigated
and the PMs exhibited great performance on enhancing the productivity of clean water.
Ó 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.
E-mail address: rong@hit.edu.cn (Y. He).

http://dx.doi.org/10.1016/j.apenergy.2017.03.080
0306-2619/Ó 2017 Elsevier Ltd. All rights reserved.
 X. Wang et al. / Applied Energy 195 (2017) 414–425
1. Introduction
Solar energy is one of the most promising renewable energy
sources because it is abundantly available and no pollutants are
generated during its use [1–4]. Highly efficient steam generation
is a key solar-energy application, e.g., for small-scale water purifi-
cation [5], solar still [6], hygiene systems [7], large-scale solar-
power concentrating systems [8], chemical plants [9], and desali-
nation technologies [10]. Currently, solar steam generation sys-
tems are typically based on highly efficient solid surface
absorbers that first collect solar energy and then transfer it to a
bulk liquid or a thermal storage system [11–14]. Steam generation
usually occurs because of the high temperature to which the bulk
liquid gets heated [15]. However, complex artificial optical concen-
tration systems are necessary to concentrate the solar light, and
the use of these systems results in high energy and optical losses
[16]. To enhance the evaporation process, multiple advanced tech-
nologies involving facility improvement, system optimization,
thermal energy storage system, and development of highly effi-
cient absorber materials have been studied.
Recently, nanofluids, i.e., fluid suspensions containing solid
nanometre-sized particles (NPs), have been widely studied as
potential direct-absorption solar energy collectors [17], owing to
the strong coupling of NPs with light and their distinctive optical
absorption properties, especially in the case of plasmonic metallic
[18] and carbon-based [19] particles. The plasmonic NPs have the
extraordinary capability to convert photo energy to thermal energy
via localized plasmon resonance [20]. In real-time light-plasmon
NP interactions, the photon–electron interaction is initially domi-
nant, followed by the electron-lattice relaxation and, finally, heat
dissipation to the environment [21]. Plasmonic NPs have a wide
range of applications in sensing [22], photothermal therapy, and
energy-harvesting [23]. Localized steam generation around plas-
monic NPs can be achieved through irradiation of nanofluids with
high-intensity light, typically using lasers [24]. Neumann et al.
[25,26] demonstrated the use of Au-based nanofluids for direct
steam generation and achieved device efficiencies of 24% at solar
concentrations of 1000 suns (1 sun = 1 kW m2). Ni et al. [27] mea-
sured a steam generation efficiency of 69% at solar concentrations
of 10 suns using graphitized carbon black, carbon black, and gra-
phene suspended in water. Wang et al. [28] experimentally inves-
tigated the direct vapour generation performance of low-
concentration carbon-nanotube nanofluids and realized an evapo-
ration efficiency of 46.8%. In all of these applications using volu-
metric solar heating of nanofluids for steam generation, the bulk
liquid temperature increased rapidly and reached a high tempera-
ture owing to the high light absorption of the dispersed NPs. Steam
bubbles are usually generated around the NPs during steam gener-
ation using a nanofluid and these steam bubbles then move
through the sub-cooled water to release. The movement of these
steam bubbles in the sub-cooled water can result in heat losses
via energy transfer to the bulk fluid, resulting in heating of the bulk
liquid and reduction of the steam production. Notably, consider-
able heat dissipation losses from the hot solution to the surround-
ing environment are always observed, which is detrimental for
steam generation.
Evaporation is also a vital function in many biological systems,
such as the evaporation during sweating from human skin [29] and
transpiration at leaf surfaces [30]. In human system, the maximum
sweating rates for an adult can be up to 2–4 L/h. In the plant sys-
tem, the transpiration in plants is a vital part in terrestrial hydro-
logical cycle. There are two main factors affecting efficient water
evaporation in biological systems: (1) localized heating and (2)
capillary flow in porous structures, as shown in Fig. 1a. Inspired
by these types of biological systems, a new strategy for enhancing
415
the steam evaporation efficiency using floating porous materials
capable of light-to-heat conversion at the air-water interface has
been demonstrated [31]. In this bio-inspired solar steam genera-
tion method, the floating absorbers at the air-water interface col-
lect sunlight and convert it to thermal energy, enabling surface
heating at the interface and rapid evaporation while the bulk liquid
temperature stays low. The heat and mass transfer processes in
this system are illustrated in Fig. 1b for solar steam generation,
where the only energy source is the solar irradiation and the effec-
tive utilization of energy is via evaporation. The heat losses include
thermal radiation and convection losses from the absorbers to the
surrounding environment, heat conduction losses to the sub-
cooled water, and parasitic losses. Wang et al. [31] and Liu et al.
[32] demonstrated fast and highly-efficient steam generation
through localized plasmonic heating of gold NP (Au NP) films
and air-laid paper-supported Au films, respectively. Bae et al.
[33] fabricated black Au thin-film membranes that exhibited high
plasmon dissipation losses, high flexibility, and ultra-broadband
light absorption. They efficiently generated steam from water
and achieved a solar thermal conversion efficiency of 57% at a light
power of 20 kW m2. Zeng et al. [34] synthesized composite float-
ing magnetic NPs (Fe3O4/C NPs with average diameter of 500 nm),
and found that their application enhanced the evaporation rate by
a factor of 2.3, compared with that of pure water. Ghasemi et al.
[35] developed a porous volumetric receiver (double-layer struc-
ture consisting of an exfoliated graphite layer supported by hydro-
philic carbon foam), and obtained a high solar thermal efficiency
(85%) at a solar illumination power of only 10 kW m2. Ni et al.
[36] demonstrate a floating solar receiver capable of generating
100 °C steam under ambient air conditions without optical concen-
tration using thermal concentration and heat localization. In 2017,
Zhou et al. [37] prepared plasmonic absorbers with tunable broad-
band absorption and thermal stability through self-assembly. The
spectrum selective plasmonic absorbers demonstrated excellent
applicability for solar steam generation.
To date, only a few reports have been published regarding the
enhancement of solar steam through bio-inspired interface heat-
ing, specifically using broadband-absorbing plasmonic membranes
(PMs) [32,33,37]. In addition, most of these works focused on the
material properties and complicated methods were required to
fabricate the broadband absorber. The thermal and mass transfer
mechanisms of bio-inspired solar steam generation should be fur-
ther investigated. In this study, a facile and controllable method
was developed for preparing porous, high-efficiency, and reusable
plasmonic light-to-heat conversion membranes. The optical prop-
erties of PMs were optimised by varying the concentration of Au
NPs on the PMs; the Au NPs are the main absorbers that interact
with the solar light and realise the photo-thermal conversion.
The experiments for determining the solar steam generation prop-
erties were implemented under simulated solar light irradiation.
Firstly, the performance of different steam generation schemes
including volumetric solar heating, bottom solar heating, and air-
water interface solar heating were studied to demonstrate the
advantages of the bio-inspired solar heating of PMs. Then, the
effects of Au NP concentration, and optical power on the evapora-
tion rate and efficiency were investigated systemically. As the basic
design of the solar still used a water basin with black bottom to
absorb the incoming solar flux, the radiative thermal losses from
the hot water were high. Therefore, a solar still desalination exper-
iment was conducted to demonstrate the potential of application
of this method for such an application. This work advances the
understanding of bio-inspired interface heating for solar steam
generation based on PMs, which promises to be a low-cost tech-
nique with the potential application in a wide range of solar-
energy-based technologies.
416 X. Wang et al. / Applied Energy 195 (2017) 414–425
Fig. 1. (a) Direct solar steam generation inspired by the sweating and transpiration in biological systems; (b) schematic of physical process in the solar steam generation
system.
2. Experimental section
2.1. Materials
Tetrachloroauric acid (HAuCl4; 49–50% Au basis), sodium citrate
dihydrate (HOC(COONa)(CH2COONa)22H2O, 99%, Analytical Reagent
(AR)), and poly(diallyl dimethyl ammonium chloride) (PDDA) solu-
tion (20 wt.% in water) were purchased from Aladdin Industrial
Inc., China and were used without further purification. Filter mem-
branes made of mixed cellulose esters (diameter: 47 mm; pore size:
220 nm) were obtained from Shanghai Xinya Co., Ltd., China.
2.2. Synthesis of Au NPs
An aqueous solution of 0.458 mM HAuCl4 was prepared by dis-
solving 180.0 mg of HAuCl4 in 950 mL of double deionised water
(DDI water, Sartorius Water Purification System, AriumÒ Mini;
18.2 MX). Meanwhile, a 34 mM trisodium citrate dihydrate aque-
ous solution was prepared by dissolving 510.0 mg of sodium
citrate dihydrate in 50 mL of DDI water. The prepared HAuCl4 solu-
tion was vigorously stirred and heated to its boiling point, and was
kept there for 20 min. Subsequently, 50 mL of the prepared sodium
citrate solution was added to the boiling HAuCl4 solution. The pale
yellow solution initially turned colourless, and then became wine
red. The mixture was boiled for 20 min until no further colour
changes were observed. Then, the solution was cooled to room
temperature for further use.
2.3. Preparation of PMs
The fabrication of the PMs followed a simple and straightfor-
ward process, involving vacuum-assisted flocculation (VAF) of an
aqueous dispersion of Au NPs on the surface of the filter
membranes. To disperse the Au NPs on the fibres of the filter mem-
branes, PDDA was used to pre-treat the paper. Uniform deposition
of Au NPs on PDDA-decorated microporous filter membranes was
achieved as PDDA carries a strong positive charge and the Au
NPs are negatively charged. In the first step, 3 mL of 1 wt.% PDDA
solution was filtered and washed with a large quantity of DDI
water to remove the excess PDDA. Then, the desired volume of
Au NP dispersion (0.09 mg mL1) was filtered through the
membrane. The target plasmonic membranes (PMs) were obtained
after washing several times with DDI water and air drying. The
concentration of Au NPs on the PM surface was controlled by
adjusting the filtered volume of the Au NP dispersion. The PM fab-
rication process is schematically shown in Fig. 2.
2.4. Characterization
The morphology of Au NPs was characterised using transmis-
sion electron microscopy (TEM; Tecnai G2 F30, FEI, US). Scanning
electron microscopy (SEM; Quanta 200FEG, FEI, US) was used to
examine the internal structure of the membrane and Au NP distri-
bution on the PMs. Ultraviolet–visible (UV–Vis) spectra of Au NP
dispersions were obtained using a double beam UV–Vis spec-
trophotometer (TU-1901, Persee, China). The total reflectance and
transmittance of the PMs were analysed using UV–Vis–Near-infra
red (NIR) spectroscopy (Lambda 950, PerkinElmer, US) with an
integrating sphere. The absorption efficiency was calculated using
the equation A ¼ 1  T  R, where T and R are the transmission and
reflection efficiencies, respectively. Fourier-transform infrared (FT-
IR; Frontier Optica PerkinElmer, US) spectra were obtained from
4000 to 500 cm1 at room temperature. The PM surface structures
were characterised using atomic force microscopy (AFM; Bruker
 X. Wang et al. / Applied Energy 195 (2017) 414–425
Fig. 2. (a) Schematic illustration of plasmonic membranes’ fabrication via vacuum-assisted flocculation and (b) photographs of PMs with a diameter of 40 mm.
Dimension Icon with ScanAsyst, Germany). The wetting properties
of the PMs were examined using a contact angle meter (JCY-2,
Shanghai Fangrui Instrument Co. Ltd, China). The mass concentra-
tions of the elements in test solution were assessed by using induc-
tively coupled plasma atomic emission spectrometry (ICP-AES,
OPTIMA7300DV, PerkinElmer, US).
2.5. Evaporation under simulated solar irradiation
A schematic of the experimental setup for solar steam genera-
tion is shown in Fig. 3. The variation in water weight during the
evaporation process was measured by placing a 50-mL beaker full
of water on an electronic precision balance with resolution of three
decimal places (Practum313-1CN, Sartorius, Germany). A PM
(40 mm in diameter) was left floating on the air-water interface.
The floating membrane was illuminated normal to the surface
using simulated solar light generated by a solar simulator (CEL
HXF300, CeauLight, China). An IR camera (Ti400, Fluke, USA) was
used to record the temperature distribution of the system. The
mass loss due to evaporation was recorded as a function of time
by a computer connected to the balance. The power density used
in this study was in the range of 1–10 kW m2. The same setup
was utilized to study the evaporation process using an empty filter
membrane, a PM submerged at the bottom of the beaker, and an
Au NP dispersion for comparison.
Fig. 3. Schematic diagram of the experimental setup for solar steam generation.
417
The solar thermal receiver efficiency (gth ) proposed by Chen
et al. [35] was employed to calculate the steam generation effi-
ciency of the PMs:
gth ¼ mðh
_ s þ hLV Þ=I ð1Þ
where m _ denotes the evaporation rate (m _ ¼ Dm=ðAtÞ), Dm is weight
change of the test system, A is the surface area of beaker, t is the
time of solar light illumination, hs is the sensible heat, hLV is the liq-
uid–vapour phase change enthalpy, and I is the power density of the
incident light.
The specific steam productivity (SSP) can be used to evaluate
the steam generation ability per unit mass of plasmonic Au NPs.
_
SSP ¼ mA=mAu ð2Þ
where mAu is the mass of Au NPs in the PM.
The evaporation behaviour of the PMs with different amounts of
Au NPs were examined and compared with those of blank filter
membranes and pure water under the same solar illumination
intensity (10 kW m2) over 60 min. Further investigation of the
PM evaporation behaviour under different light intensities was
conducted by varying the solar power density from 1 to
10 kW m2, keeping other parameters unchanged (the mass of
Au NPs on the PMs was 4.5 mg).
2.6. Seawater desalination experiment with a single basin solar still
The PMs prepared in this study could be a good choice for sea-
water desalination systems through bio-inspired solar heating. The
as-prepared PMs were used as floating absorbers in a commercially
available single basin solar still system, as illustrating in Fig. 4(a).
In the solar still system, the seawater was heated by the solar
absorber and converted into steam continuously. Subsequently,
the generated steam was condensed into water droplet at the inner
surface of the glass cover. Finally, the condensed fresh water was
collected through a rubber tube. In the present experiment, the
basin area of the still is 0.0625 m2 (0.25 m length  0.25 m width).
The height of the low side wall is 0.11 m and the height of high side
wall is 0.26 m. So, the glass cover could be inclined at nearly 30° as
the condensing surface of the system. The still was made of acrylic
and the inner surfaces were painted black to absorber the solar
light. To reduce the thermal loss of the system, the external and
bottom surfaces of the still were adiabatic coated by insulated
foams and then covered with a reflective aluminium film. A pho-
tograph of the solar still system was shown in Fig. 4(b). 36 pieces
of PMs were pasted together to construct a 6  6 array and it
played as the absorber. Before the experiment, 1000 ml seawater
(from the Bohai Sea, China; the average salinity 2.75 wt.%) was
418 X. Wang et al. / Applied Energy 195 (2017) 414–425

Fig. 4. The experiment set up for seawater desalination. (a) Schematic illustration of a single basin solar still; (b) the photographs of the experiment set up (the insert images
are the photographs of the basin still and the PM array).

poured into the basin for desalination. The experiment was con-
ducted under a matrix solar simulator with power of 1 kW m2
(TRM-PD1, Jinzhou Sunshine Technology Co. Ltd, China) and the
ambient temperature was 18 °C.
3. Results and discussion
3.1. Structure and optical properties of PMs
The Au NPs used for the preparation of PMs had an
average diameter 11.3 nm and a surface plasmon (SP)
absorption peak at 518 nm (Fig. S2). After the filtration of
40 mL of the prepared Au NP dispersion (0.09 mg mL1), the
Au NPs formed a uniform and densely packed layer on the
microfibre surface (Fig. 5a). The Au layer became thicker as
more Au NPs were deposited on the PM surface. Owing to the
disordered hybrid cellulose fibrils of the filter membranes, the
PM exhibited sub-microscale surface roughness (Fig. 5b and c).
The rough surface could realise multiple-scattering of the
incident solar light and enhance the interaction between
the incident photons and the Au NPs. In addition, the
sub-microscale surface roughness increased the number of solar

Fig. 5. Structure of PMs. (a) SEM images of PMs at different scales (the inset image is the TEM image of Au NPs); (b) 2D and (c) 3D AFM images of plasmonic membrane.
 X. Wang et al. / Applied Energy 195 (2017) 414–425 419

Fig. 6. Optical properties of PMs. Experimental measurement of total absorption and reflection in the wavelength range of 400–2500 nm for (a) filter membranes and (b) PMs.
(c) Reflection and (d) absorption of PMs with different concentrations of Au NPs. (e) Transmittance measured by FT-IR spectra in the wavenumber range of 4000–400 cm1.
420 X. Wang et al. / Applied Energy 195 (2017) 414–425

collection surfaces and the air-water interface area, helping to
enhance the evaporation process.
The three-dimensional porous structure of the PM can also
improve its liquid wetting behaviour through capillary action,
resulting in a higher evaporation rate. The PM wetting properties
were examined by testing the dynamic contact angles of a 10-lL
droplet on the PM surface over 3 s (Fig. S3). After the 10-lL droplet
was dropped on the PM surface, the contact angle changed from
66° to 0° and the droplet was absorbed into the inner structure
of PM within 3 s, suggesting that the porous PMs had good wetting
ability due to its high capillary forces. Sufficiently high capillary
forces could be generated due to the microporous structure at
the air-water interface, enabling the supply of water during the
surface water evaporation.
Fig. 6a and b show the total absorption and reflection spectra in
the wavelength range of 400–2500 nm for a blank filter membrane
and a PM containing 3.6 mg Au NPs, respectively. The filter mem-
brane exhibited low absorption (<10%) and high reflection in the
studied wavelength range. Conversely, high absorption (90% in
the range of 400–1500 nm) and low reflection shown by the PM
indicated that it absorbed most of the optical light in the studied
range. The light reflection of the PMs in the wavelength range of
400–2500 nm decreased gradually with increasing amount of
deposited Au NPs (Fig. 6c). Meanwhile the light absorption of the
PMs in the 400–2500 nm wavelength range increased gradually
with increasing concentration of Au NPs and remained at a high
level when the mass of the Au NPs exceeded 3.6 mg (Fig. 6d).
The FT-IR transmission spectra (Fig. 6e) confirmed the low trans-
mission of the PM in the mid-IR region. Since most of the solar
energy is distributed in the visible and NIR range, the high optical
absorption of PMs over the 400–2500 nm range would make a
major contribution to solar energy collection.
3.2. Solar steam generation enhancement with PMs
All different solar steam generation experiments (① water; ②
Au NP dispersion; ③ submerged PM; ④ floating PM) were per-
formed under the same solar power of 10 kW m2 and the amount
of Au in both the tested Au NP dispersions and the PMs was 3.6 mg
(Fig. 7). In these different solar steam generation schemes shown in
Fig. 7, the Au NP dispersion represents the volumetric solar heating
of a bulk fluid, the submerged PM represents traditional solar heat-
ing from the bottom (like a solar still), and the floating PM repre-
sents the novel bio-inspired surface heating process. The weight
change due to solar steam generation is shown in Fig. 7a. When
Au NPs were introduced in the evaporation system, regardless of
whether they were dispersed in the water or deposited on the
PMs, the measured weight changes were much higher than that
obtained for the evaporation of pure water under solar illumina-
tion. The floating PM scheme achieved the highest weight change,
i.e. a large water steam yield was achieved. After the first 30 min of
illumination, the submerged PM system exhibited a higher weight

Fig. 7. Evaporation performance under different schemes (① water; ② Au NP dispersion; ③ submerged PM; ④ floating PM). (a) Water weight change due to evaporation
over time; (b) evaporation rate (red bars) and SSP (blue bars); (c) solar thermal receiver efficiency (red bars) and evaporation enhancement (blue bars). (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version of this article.)
 X. Wang et al. / Applied Energy 195 (2017) 414–425 421

change (i.e. higher evaporation rate) than the Au NP dispersion sys-
tem. The floating PM system exhibited the best solar steam gener-
ation performance among the studied schemes. The evaporation
rate and SSPs under the different schemes are shown in Fig. 7b.
The floating PM scheme had the highest evaporation rate
(11.8 kg m2 h1) and SSP (4.1 kg h1 g1) among the four
schemes. Fig. 7c shows that the gth of pure water was only
20.3%, while that of the Au NP dispersion reached 55.5%. Hence,
we observed a large enhancement in the gth when Au NPs were
introduced in the steam generation system. The floating PM
scheme clearly achieved the highest solar thermal receiver effi-
ciency of 85%, an enhancement of 320% compared to pure
water.
3.3. Effect of Au NP concentration on solar steam generation
The influence of the concentration of the Au NPs on the solar
steam generation was investigated. Compared with pure water,
the evaporation rate decreased in the presence of a floating blank
filter membrane (Fig. 8a). The blank paper hindered water evapo-
ration because of its low absorbance and high reflectance of solar
light. Interestingly, small amounts of Au NPs deposited on the
floating filter membranes enhanced water evaporation remark-
ably. Higher water weight changes due to evaporation were
achieved with increasing amounts of added Au NPs. The maximum
weight change was achieved by the PM with 3.6 mg of deposited
Au NPs; further increasing the amount of deposited Au NPs did
not significantly enhance evaporation. According to the optical
properties of the PM discussed above (Fig. 6), the addition of Au
NPs resulted in higher light absorption, leading to higher evapora-
tion rates; both light absorption and evaporation reached satura-
tion after adding 3.6 mg of Au NPs. The enhancement of the
evaporation with floating blank filter membranes and PMs with
different amounts of deposited Au NPs is shown in Fig. 8b. When
a blank floating filter membrane was used, the enhancement was
negative, while the evaporation enhancement of the PMs increased
gradually with added Au NPs and reached a maximum of 320%.
Furthermore, gth of the PMs increased gradually from 58.7% to
87.7% as the amount of deposited Au NP increased from 0.45 mg
to 14.4 mg (Fig. 8d).
The evaporation rate was in the range of 8.15–12.18 kg m2 h1,
as the amount of deposited Au NP increased from 0.45 to 14.4 mg,
under solar illumination with a power density of 10 kW m2. Con-
versely, the SSP values decreased with increasing Au NP concentra-
tions on the PM surface (Fig. 8c, blue line); SSP decreased from 22.7
to 1.1 kg h1 g1 when the Au NPs concentration increased from

Fig. 8. Effect of Au NPs mass concentration on the solar steam generation performance. (a) Water weight change due to evaporation over time; (b) evaporation enhancement;
(c) evaporation rate (red lines) and SSP (blue lines); (d) solar thermal receiver efficiency. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
422 X. Wang et al. / Applied Energy 195 (2017) 414–425
0.45 mg to 14.4 mg. The decrease in SSP could be due to the poor
light capture and absorbing ability of the individual Au NPs depos-
ited on the PMs.
Notably, the conversion efficiency of individual plasmonic Au
NPs on the PM decreased with increasing particle concentrations
in a specific region. One reason for this is that the Au NPs agglom-
erated during deposition. Au NPs can aggregate more easily when
higher amounts are deposited on the filter membranes. Another
possible reason is the thickness of Au film on the PM surface as
the absorption efficiency of each particle depends on its relative
depth within the Au film. The Au NPs on the surface that are
directly exposed to the incident light could fully absorb the solar
energy and convert it into thermal energy, while those located in
lower layers had limited absorption ability, resulting in less effi-
cient solar thermal conversion.
3.4. Effect of solar powers on the solar steam generation
The solar thermal efficiency and corresponding evaporation rate
at different optical concentrations are shown in Fig. 9a. For the
floating PM system, the evaporation rate reached 0.87 kg m2
after exposure to a 1 kW m2 light source for 60 min. As antici-
pated, the evaporation rate increased significantly with increasing
light intensity (Fig. 9a, blue squares). Under 10 kW m2 irradiation,
the floating PM exhibited an evaporation rate of 11.8 kg m2 h1
after 60 min. The PM gth increased from 62.5% at 1 kW m2 to 85%
at 10 kW m2 (Fig. 9a, red squares).
The efficiency curves obtained in this work have the same
trends as those shown in the work of Ghasemi et al. [35] and Bae
et al. [33]. The receiver efficiency decreased with increasing solar
intensity from 1 sun to 2 suns, then increased gradually to 10 suns.
Many factors could account for the observed changes in the evap-
oration efficiency, including thermal radiation losses and convec-
tion losses from the absorbers to the surrounding environment,
heat conduct losses to the sub-cooled water, and the mass transfer
resistance from the steam point to the air-water interface and so
on [36]. A possible explanation for the observed behaviour is that
the solar energy was utilized for evaporation at the surface of the
membranes at 1 kW m2, resulting in a relatively high efficiency.
As the solar light intensity increased from 1 to 2 kW m2, part of
the solar energy was utilized for the direct heating of the water
on the membrane surface and the rest was used for heating the
membranes, resulting in a thermal efficiency losses in the mem-
branes. As the solar light intensity increased from 2 to 10 kW m2,
Fig. 9. (a) The solar thermal receiver efficiencies (red points) and corresponding evaporation rates (blue points) of the evaporation process by PMs under different optical
concentrations; (b) evaporation weight change versus cycle number. (For interpretation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
there was sufficient energy for both membrane heating and steam
generation. Therefore, a high receiver efficiency was achieved at a
solar light intensity of 10 kW m2. Neumann et al. [25,26] observed
low efficiency (24%) with using Au-based nanofluids for solar
steam generation, which could be explained by two different
mechanisms. The first is that highly concentrated solar irradiation
(e.g. 1000 suns) could result in much higher thermal losses from
thermal radiation to the surrounding environment. The second is
that the volumetric solar steam generation method was utilized
in their work, where the concentrated solar light is only absorbed
by a small region of the working fluid. Even though steam bubbles
could be rapidly generated around the PNs, there were huge heat
losses as the steam bubbles moving through the sub-cooled water
to release at the air-water interface.
Since the recyclability of the PMs would favour future industrial
applications, evaporation experiments with PMs containing 3.6 mg
of Au NPs were repeated 15 times under the same conditions (opti-
cal density of 10 kW m2 for 60 min) to demonstrate the reusabil-
ity of the PMs (Fig. 9b). After each cycle, the wet PM was dried and
prepared for another cycle. Comparable performances were
obtained over the 15 cycles, with an average weight change of
11.8 kg m2.
3.5. Mechanism of solar steam generation enhancement
An IR camera was used to record the temperature distribution
during water evaporation experiments (Fig. 10). Before illumina-
tion, the system had the same temperature as the surrounding
environment (20 °C; Fig. 10a). After exposure to simulated sun-
light illumination (10 kW m2) for 60 min, temperature distribu-
tions were obtained for the pure water and floating PM systems.
The pure water system exhibited a uniform temperature distribu-
tion (Fig. 10b), while a hot zone was created in the floating PM sys-
tem (Fig. 10c) and the temperature of underlying water showed a
gradient from the top to the bottom of the beaker. This indicates
that this kind of bio-inspired interface heating using PMs could
realise less thermal losses compared to heating the bulk fluid
and more concentrated use of solar energy for steam generation.
Finally, the localized plasmonic heating induced by the PM
resulted in a high surface temperature and high evaporation rate.
In addition, the optical properties of the PMs have significant
effects on the solar steam generation performance. Usually, highly
efficient solar steam generation requires strong light absorption by
the Au NP structure along with high surface plasmon dissipation
 X. Wang et al. / Applied Energy 195 (2017) 414–425 423

Fig. 10. Temperature distribution recorded by IR camera. (a) A beaker full of water before solar illumination. After solar illumination for 60 min (b) water, (c) water with PM
floating at air-water interface.

Fig. 11. Seawater Desalination performance with PMs in a single basin solar still under one sun. (a) Comparison of hourly outputs of the fresh water between the still with
floating PM array and the conventional still (insert figure is the temperature distribution of the PM array after 5 hours’ working). (b) The concentrations of primary ions in
initial seawater and the condensed fresh water (the red dashed lines refer to the WHO standards for drinkable water). (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)

losses. The sub-microscale structured surface roughness described
earlier (Fig. 5) helped achieve broadband absorption by the PMs.
The incident light on the surface of the PMs could be scattered
multiple-times by the rough structure and multiple-light absorp-
tion steps could occur, explaining the broadband absorption of
PMs [38]. Moreover, the rough structures minimised dissipative
losses, leading to low surface plasmon reflection [39], which corre-
sponds to high power dissipation losses and a short plasmon prop-
agation length [40]. The plasmons experienced substantially low
reflection efficiency with short propagation lengths on the PM
structures, and the solar energy eventually dissipated into thermal
energy.
3.6. Applications in solar still for seawater desalination
To evaluate the performance of PMs in a single basin solar still
system for seawater desalination, a comparison between the mod-
ified method with PMs and a conventional method was investi-
gated. Fig. 11(a) shows the hourly output of the fresh water with
two different conditions: (1) a conventional solar still; (2) a solar
still with floating PM arrays. Compared to the conventional solar
still, the floating PM arrays could lead to higher yield of fresh
water, and the productivity of still enhanced by about 80%. The
inset picture in Fig. 11(a) is the temperature distribution of the
floating PM arrays after 5 h illumination. It can be seen that the
424 X. Wang et al. / Applied Energy 195 (2017) 414–425
temperature of the PMs would increase to 25 °C. The thermal
energy in the PMs would be utilized for the localized heating of
the attached seawater and generating water steam. Furthermore,
the quality of the collected fresh water was investigated by com-
paring the concentrations of the elements in the initial seawater
and the condensed fresh water, as shown in Fig. 11(b). The concen-
trations of the primary ions (Na+, K+, Ca2+, Mg2+ and B3+) in the con-
densed fresh water were obviously reduced and far below the
World Health Organization (WHO, less than 200 ppm) standards
for drinking water [41].
Thus, the bio-inspired solar steam generation using PMs has
great potential application in the seawater desalination or water
purification. For other applications, such as the solar power gener-
ation, where high temperature and pressure steam could be used
to produce electricity using a turbine, it would be interesting to
further increase the temperature and pressure of the generated
steam in the future. As organic Ranking cycles only require work-
ing fluid temperatures of 100–200 °C [42–44], these could be suit-
able power application for our small-scale device right now.
4. Conclusion
The enhancement of a solar steam generation process based on
bio-inspired interface solar heating of PMs is presented here.
Owing to the multiple-scattering and absorption of surface plas-
mons by the PMs, a high light absorption of 90% in the broadband
range of 400–1500 nm was achieved. The bio-inspired floating PMs
were superior to conventional evaporation systems because of
localized plasmonic heating that only occurred at the air-water
interface (while the bulk liquid temperature remained low) yield-
ing a solar thermal conversion efficiency of 85% under the solar
illumination intensity of 10 kW m2. In addition, comparison with
solar steam generation using pure water showed that the relative
efficiency increased up to 320%. This is because a hot zone was
generated near the floating PMs under solar light irradiation,
enhancing evaporation. In addition, the PMs exhibited excellent
recyclability and maintained a high solar steam generation rate
even after repeated use. In the application of PMs in a single basin
solar still system for seawater desalination, the fresh water pro-
ductivity could be enhanced 80% with using PMs. In the future,
these PMs can be utilized as new components for bio-inspired
solar-energy-driven steam generation systems that in turn could
have many potential applications such the seawater desalination,
solar energy harvesting, sterilization, and hygiene systems.
Acknowledgment
This research work has been financially supported by the
National Natural Science Foundation of China (Grant No.
51676060), the Natural Science Founds of Heilongjiang Province
for Distinguished Young Scholars (Grant No. JC2016009) and the
Fundamental Research Funds for the Central Universities (Grant
No. HIT. BRETIV. 201315).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.apenergy.2017.
03.080.
References
[1] Duić N, Guzović Z, Kafarov V, KlemeŠ JJ, vad Mathiessen B, Yan J. Sustainable
development of energy, water and environment systems. Appl Energy
2013;101:3–5.
[2] Lewis NS. Toward cost-effective solar energy use. Science 2007;315:798–802.
[3] Gao X, Liu J, Zhang J, Yan J, Bao S, Xu H, et al. Feasibility evaluation of solar
photovoltaic pumping irrigation system based on analysis of dynamic
variation of groundwater table. Appl Energy 2013;105:182–93.
[4] Yan J, Desideri U, Chou SK, Li H. Energy solutions for a sustainable world. Int J
Green Energy 2016;5075:20–2.
[5] Shannon MA, Bohn PW, Elimelech M, Georgiadis JG, Mariñas BJ, Mayes AM.
Science and technology for water purification in the coming decades. Nature
2008;452:301–10.
[6] Elfasakhany A. Performance assessment and productivity of a simple-type
solar still integrated with nanocomposite energy storage system. Appl Energy
2016;183:399–407.
[7] Neumann O, Feronti C, Neumann AD, Dong A, Schell K, Lu B, et al. Compact
solar autoclave based on steam generation using broadband light-harvesting
nanoparticles. Proc Natl Acad Sci USA 2013;110:11677–81.
[8] Gupta MK, Kaushik SC. Exergy analysis and investigation for various feed water
heaters of direct steam generation solar-thermal power plant. Renew Energy
2010;35:1228–35.
[9] Andasol T, Nevada S, Goswami Y, Florida S, Fabrizi F. Saving for a rainy day.
Science 2011;334:922–4.
[10] Elimelech M, Phillip WA. The future of seawater desalination: Energy,
technology, and the environment. Science 2011;333:712–8.
[11] Wang W, Yang X, Fang Y, Ding J, Yan J. Preparation and thermal properties of
polyethylene glycol/expanded graphite blends for energy storage. Appl Energy
2009;86:1479–83.
[12] Wang W, Yang X, Fang Y, Ding J, Yan J. Enhanced thermal conductivity and
thermal performance of form-stable composite phase change materials by
using b–Aluminum nitride. Appl Energy 2009;86:1196–200.
[13] Wang W, Guo S, Li H, Yan J, Zhao J, Li X, et al. Experimental study on the direct/
indirect contact energy storage container in mobilized thermal energy system
(M-TES). Appl Energy 2014;119:181–9.
[14] Wang W, Li H, Guo S, He S, Ding J, Yan J, et al. Numerical simulation study on
discharging process of the direct-contact phase change energy storage system.
Appl Energy 2015;150:61–8.
[15] Roldán MI, Valenzuela L, Zarza E. Thermal analysis of solar receiver pipes with
superheated steam. Appl Energy 2013;103:73–84.
[16] Weinstein LA, Loomis J, Bhatia B, Bierman DM, Wang EN, Chen G.
Concentrating solar power. Chem Rev 2015;115:12797–838.
[17] Colangelo G, Favale E, Miglietta P, de Risi A, Milanese M, Laforgia D.
Experimental test of an innovative high concentration nanofluid solar
collector. Appl Energy 2015;154:874–81.
[18] Duan H, Xuan Y. Enhanced optical absorption of the plasmonic nanoshell
suspension based on the solar photocatalytic hydrogen production system.
Appl Energy 2014;114:22–9.
[19] Karami M, Akhavan Bahabadi MA, Delfani S, Ghozatloo A. A new application of
carbon nanotubes nanofluid as working fluid of low-temperature direct
absorption solar collector. Sol Energy Mater Sol Cells 2014;121:114–8.
[20] De Sio L, Placido T, Comparelli R, Lucia Curri M, Striccoli M, Tabiryan N, et al.
Next-generation thermo-plasmonic technologies and plasmonic nanoparticles
in optoelectronics. Prog Quantum Electron 2015;41:23–70.
[21] De Sio L, Placido T, Serak S, Comparelli R, Tamborra M, Tabiryan N, et al. Nano-
localized heating source for photonics and plasmonics. Adv Opt Mater
2013;1:899–904.
[22] Liu C, Fuh AY, Chen Y, Chen S, Lin J, Lee C, et al. Plasmonic photoheating of gold
nanorods in thermo-responsive chiral liquid crystals. J Opt 2016:125005.
[23] Chen M, He Y, Zhu J, Shuai Y, Jiang B, Huang Y. An experimental investigation
on sunlight absorption characteristics of silver nanofluids. Sol Energy
2015;115:85–94.
[24] Govorov AO, Richardson HH. Generating heat with metal nanoparticles. Nano
Today 2007;2:30–8.
[25] Neumann O, Urban AS, Day J, Lal S, Nordlander P, Halas NJ. Solar vapor
generation enabled by nanoparticles. ACS Nano 2013;7:42–9.
[26] Lombard J, Biben T, Merabia S. Nanobubbles around plasmonic nanoparticles:
Thermodynamic analysis. Phys Rev E 2015;91:43007.
[27] Ni G, Miljkovic N, Ghasemi H, Huang X, Boriskina SV, Lin C-T, et al. Volumetric
solar heating of nanofluids for direct vapor generation. Nano Energy
2015;17:290–301.
[28] Wang X, He Y, Cheng G, Shi L, Liu X, Zhu J. Direct vapor generation through
localized solar heating via carbon- nanotube nanofluid. Energy Convers Manag
2016;130:176–83.
[29] Wingo JE, Low D a, Keller DM, Brothers RM, Shibasaki M, Crandall CG. Skin
blood flow and local temperature independently modify sweat rate during
passive heat stress in humans. J Appl Physiol 2010;109:1301–6.
[30] Wheeler TD, Stroock AD. The transpiration of water at negative pressures in a
synthetic tree. Nature 2008;455:208–12.
[31] Wang Z, Liu Y, Tao P, Shen Q, Yi N, Zhang F, et al. Bio-inspired evaporation
through plasmonic film of nanoparticles at the air-water interface. Small
2014;10:3234–9.
[32] Liu Y, Yu S, Feng R, Bernard A, Liu Y, Zhang Y, et al. A bioinspired, reusable,
paper-based system for high-performance large-scale evaporation. Adv Mater
2015;27:2768–74.
[33] Bae K, Kang G, Cho SK, Park W, Kim K, Padilla WJ. Flexible thin-film black gold
membranes with ultrabroadband plasmonic nanofocusing for efficient solar
vapour generation. Nat Commun 2015;6:10103.
[34] Zeng Y, Yao J, Horri BA, Wang K, Wu Y, Li D, et al. Solar evaporation
enhancement using floating light-absorbing magnetic particles. Energy
Environ Sci 2011;4:4074.
 X. Wang et al. / Applied Energy 195 (2017) 414–425
[35] Ghasemi H, Ni G, Marconnet AM, Loomis J, Yerci S, Miljkovic N, et al. Solar
steam generation by heat localization. Nat Commun 2014;5:4449.
[36] Ni G, Li G, Boriskina SV, Li H, Yang W, Zhang T, et al. Steam generation under
one sun enabled by a floating structure with thermal concentration. Nat
Energy 2016;1:16126.
[37] Zhou L, Zhuang S, He C, Tan Y, Wang Z, Zhu J. Self-assembled spectrum
selective plasmonic absorbers with tunable bandwidth for solar energy
conversion. Nano Energy 2017;32:195–200.
[38] Xu L, Ma W, Wang L, Xu C, Kuang H, Kotov NA. Nanoparticle assemblies:
dimensional transformation of nanomaterials and scalability. Chem Soc Rev
2013;42:3114–26.
[39] Vogel MW, Gramotnev DK. Adiabatic nano-focusing of plasmons
by metallic tapered rods in the presence of dissipation. Phys Lett A
2007;363:507–11.
425
[40] Søndergaard T, Bozhevolnyi SI, Beermann J, Novikov SM, Devaux E, Ebbesen
TW. Resonant plasmon nanofocusing by closed tapered gaps. Nano Lett
2010;10:291–5.
[41] Safe Drinking-Water from Desalination (WHO, 2011); http://who.int/water_
sanitation_health/publications/2011/desalination_guidance/en/.
[42] Delgado-Torres AM, García-Rodríguez L. Analysis and optimization of the low-
temperature solar organic Rankine cycle (ORC). Energy Convers Manag
2010;51:2846–56.
[43] Bruno JC, López-Villada J, Letelier E, Romera S, Coronas A. Modelling and
optimisation of solar organic rankine cycle engines for reverse osmosis
desalination. Appl Therm Eng 2008;28:2212–26.
[44] Tchanche BF, Papadakis G, Lambrinos G, Frangoudakis A. Fluid selection for a
low-temperature solar organic Rankine cycle. Appl Therm Eng
2009;29:2468–76.