Review

Progress and Expectation

of Atmospheric Water Harvesting
Yaodong Tu,1 Ruzhu Wang,1,* Yannan Zhang,1 and Jiayun Wang1

Even if people live in an arid desert, they know that plenty of water exists in the Context & Scale
air they breathe. However, the reality tells us the atmospheric water cannot help Airborne moisture is a potential
to slake the world’s thirst. Thus an important question occurs: what are the source of a plentiful amount of
fundamental limits of atmospheric water harvesting that can be achieved in freshwater that is accessible
typical arid and semi-arid areas? Here, through a thorough review on the pre- everywhere and can be easily co-
sent advances of atmospheric water-harvesting technologies, we identify the operated with a renewable energy
achievements that have been acquired and evaluate the challenges and barriers source (solar energy). This paper
that retard their applications. Lastly, we clarify our perspectives on how to presents a comprehensive and
search for a simple, scalable, yet cost-effective way to produce atmospheric critical review of state-of-the-art
water for the community and forecast the application of atmospheric water har- research on atmospheric water
vesting in evaporative cooling, such as electronic cooling, power plant cooling, harvesting. From the viewpoint of
and passive building cooling. applications, we are concerned
most about whether an
Introduction atmospheric water harvester can
Just as Leonardo da Vinci said, ‘‘water is the driving force of all nature.’’ Freshwater produce sufficient freshwater
scarcity is increasingly perceived as a globally systemic risk. Taking into account the under a wide range of weather
seasonal fluctuations in water consumption and availability, recent research1 found conditions in an energy-efficient
that two-thirds of the global population (4.0 billion people) live under conditions of way. Therefore, a variety of
modest water scarcity for at least one month in a year. Even worse, a half billion harvesting methods, including
people on Earth face severe water scarcity all year round. However, the atmospheric radiative cooling, solar distilling,
water, which is considered a huge renewable reservoir of water and enough to meet and sorption-based water
the needs of every person on the planet, is unfortunately ignored.2 Atmospheric collecting, are reviewed and
water usually exists in three basic types:3 clouds floating in the sky, fog close to discussed based on their capture
the land, and water vapor in the air. Cloud and fog are all made up of tiny drops materials, system designs, and
of water (typically with a diameter from 1 to 40 mm, compared with the size of rain thermodynamic cycles. The study
droplets varying from 0.5 to 5 mm), but the concentration of water droplets in fog also presents a systematic
is usually larger. When water is taken out of the air, different from desalination, little performance comparison of
influence may occur to disrupt the hydrological cycle or steal water away from impor- recently proposed atmospheric
tant critical sources nearby.4 In addition, since the source of the atmospheric water is water harvesters. Furthermore, we
normally clean, the water quality is good enough for drinking and for other domestic discuss four key problems that
and agricultural purposes.5 limit the cost-effectiveness and
provide some solutions as
Thus, today water harvesting from air shows a great promise in supplying water perspectives.
for community use in arid areas,6 portable water production by decentralized
systems,7,8 and emergency water supply in post-disaster times.9 The millennium Atmospheric water-harvesting
development goals10 set by the United Nations highlight the critical needs of self- technology has experienced
sustenance in potable water supply through small-scale decentralized water produc- significant progress in the past 20
tion in impoverished and developing regions of the world. On one hand, 1.7 million years. However, little research on
deaths per year are caused by poor water quality, sanitation, and hygiene.11 On atmospheric water harvesters is
the other hand, fresh water availability is essential for the economic development conducted with broad horizons,
in small communities in remote areas, where annual rainfall is negligible and and system integrations have
traditional water sources such as rivers, lakes, wells, and springs are also not

1452 Joule 2, 1452–1475, August 15, 2018 ª 2018 Elsevier Inc.

viable.12–15 Moreover, in recent years the frequency and intensity of natural disasters
induced by climate change have been dramatically increasing.16 One of the first pri- been poorly examined. More
orities after a disaster is to provide safe drinking water.9 In the past bottled water research is expected to deal with
and/or water tankers were delivered to the affected population. However, limited these issues to facilitate the efforts
water capacity and relatively expensive continuous delivery are the major draw- of turning decades of research on
backs. Thus, sustainable onsite water supply technology is more practical and prom- atmospheric water harvesting into
ising. Two techniques are usually applied to overcome such obstacles: desalination tangible benefits in our daily life.
of saline water (ground or underground) and extraction of water from the air. The
source of the water is usually chosen based on availability and cost. Current water
desalination methods include distillation, dew vaporation, reverse osmosis, electro-
dialysis, and electrosorption.17,18 However, for small-scale applications these tech-
niques both require accessible brackish water sources and expert labor for operation
and maintenance, which narrows the application conditions.18–20 Instead, atmo-
spheric water harvesting is accessible everywhere and can be easily co-operated
with a renewable energy source (e.g., solar energy) for local needs.21

Despite the significant value of the potentially extractable fresh water in the
world,22–24 few atmospheric water-harvesting systems are commercially operating
currently, leaving distribution of such systems a huge challenge. In general, any
viable atmospheric water-harvesting technology must satisfy five primary criteria:
It should be efficient, cheap, scalable, wide-band, and stable enough to operate
for a whole year or at last a monsoon season. Currently none of the existing commer-
cial atmospheric water generators (AWGs) meets all these five criteria. From the
point of view of thermodynamics, this is mainly due to the energy inefficiency of
the process.

As a first step for understanding different AWGs, we should comprehend the ther-
modynamic fundamentals and examine the performance indexes involved in the at-
mospheric water-harvesting technology (here we take the sorption-based AWG25 as
an example). The workflow of one AWG is usually described as follows: capture the
moisture from thin air, then condense the captured moisture into liquid water. The
separation process and the condensation process may consume energy, where
renewable energy, such as solar or wind energy, is a top priority. Just as with desa-
lination,19,20 three indexes are usually employed to evaluate the performance of
AWG, including the specific water production per day per unit collector area
(SWP), the specific energy consumption per unit mass water production (SEC), and
the recovery ratio of the feed air (RR). The SWP is usually used for evaluating the
water productivity of the passive AWG (without the need of energy input), while
SWP and RR are more often used to assess the energy efficiency and the water vapor
condensing effectiveness of the active AWGs. For water collection by direct cooling,
the value of SEC and RR can be defined as follows:
  
Qcond 3T Ti
Tcond
SEC = zCp + hfg ; (Equation 1)
m H2 O 3d d
dcond
i 
dcond
RR = 3 d 1
; (Equation 2)
di
where Ti and di are the temperature and humidity ratio of the inlet air of the
condenser, respectively. Tcond represents the condensation temperature. mH2 O is
the water production per unit mass dry air (kg/kg). 3 T and 3 d are the heat-exchange
1Institute
of Refrigeration and Cryogenics,
effectiveness and mass exchange effectiveness of the condenser, respectively. The
Shanghai Jiao Tong University, Shanghai 200240,
total cooling load of the moist air, Qcond, is the sum of the sensible heat load, asso- China
ciated with the temperature change of the moist air, and the latent heat load, asso- *Correspondence: rzwang@sjtu.edu.cn
ciated with the enthalpy of condensation, hfg. Obviously, a smaller sensible heat https://doi.org/10.1016/j.joule.2018.07.015

Joule 2, 1452–1475, August 15, 2018 1453

load may result in a smaller SEC, which means a higher relative humidity of the inlet
air is expected. A similar index (Moisture Harvesting Index [MHI]) was defined by
Gido et al.24 and a simple relationship between SEC and MHI can be obtained:
SEC 3 MHI = hfg. The definitions of SEC and RR indicate that a low Tcond and an inlet
air state with low Ti and high di are optimal.

Here, we first review the historical developments of the atmospheric water-harvest-

ing technologies in brief. We then categorize different kinds of AWGs according to
their moisture-capturing methods and comment on their strengths and weaknesses.
Based on this, we deduce that the sorption-based AWGs have the greatest potential
to be the cost-effective and scalable AWG technology. Thus in the following section,
we discuss the progress of the materials, devices, and system structures used in
sorption-based AWGs in detail. Lastly, we propose our perspectives on the future
trends on how to achieve a scalable sorption-based AWG.

Methods
Natural Harvesting
Nature has adapted different methods for surviving dry, arid, xeric conditions. One
of the first known studies of plants absorbing airborne moisture is carried out by
Hales in 1727.26 In 1957, a literature review was carried out by Stone27 concentrating
on dew and the absorption of water in plants. In 2014, a thorough review of the liter-
ature was carried out by Malik et al.26 which updated and extended the work,
focusing on direct dew, fog, and moisture harvesting in plants and animals, such
as beetles, frogs, lizards, spiders, Opuntia microdasys, Stipagrostis sabulicola,
and Trianthema hereroensis.

Early AWGs
The idea of collecting water from fog can be traced back to centuries ago. Several
accounts or even legends concern dew ‘‘springs’’ and ‘‘ponds,’’3 which essentially
come from the artificially harvested dew by using large stones or trees. Therefore,
it seems that mankind considered dew as a source of fresh water since time imme-
morial. Unfortunately, it was not until the 20th century that such condensers that
only appear in old-time accounts or tales came true. Historically it is argued in sci-
ence that the Early Greeks who founded Theodosia as early as the 6th century BC
used dew condensers to fulfill their water demands. This opinion first comes from
a Russian forester, F.I. Zibold, who built an experimental stone condenser in the
shape of a bowl during 1905 and 1912,4,21 following what he considered to be an
Early Greek condenser. However, some researchers challenged Zibold’s experi-
mental results and few positive responses were given. Therefore, until nowadays
the Early Greek dew condenser still remains a tale rather than a scientific story.3,28

During the first half of the 20th century, Zibold’s attempt inspired extensive experi-
ments with this type of water condenser in the South of France by L. Chaptal, M.
Goddard, and A. Knapen.3 These installations called ‘‘aerial wells’’ or (vapor) ‘‘cap-
tors’’ were analogous to the Crimean prototypes. Some of the constructions success-
fully produced condensed water, but the amount was less than expected. These
massive aerial condensers are designed to take advantage of the temperature vari-
ation during day and night. At night, the stone heap is cooled down by the chilled air.
During daytime, the warm air that comes from the ocean and thus is more or less
saturated with water vapor reaches this chilled surface and is consequently
condensed. However, due to the low thermal conductivity, the effective working
layer of the stone is believed to be very thin. Furthermore, the low heat capacity
of this thin active layer crucially limits the water condensation amount.

1454 Joule 2, 1452–1475, August 15, 2018

After the rosy hopes for massive aerial condensers failed, the publication by
Monteith29 in 1957 can be considered as a marker in dew research, shifting the focus
from observation and dew measurement attempts to understanding the energy and
heat balance mechanisms of dew formation and dew evaporation. Since then,
research on modern atmospheric water-harvesting technologies has continued,30
and a number of atmospheric water-collection methods,4,5,31 which were mainly
applied in arid and semi-arid areas, have been studied.

Modern AWGs
According to the forms of airborne water, the atmospheric water-harvesting
technologies can be divided into three different categories: artificial rain collec-
tion,32–34 fog water collection,6,31 and dew water collection.35 Weather modifica-
tion (also known as cloud seeding or ‘‘artificial rain’’) may produce substantial pre-
cipitation but only in the troposphere where water-abundant clouds gather. No
evidence shows that the same process can be achieved at ground level in a routine,
controllable fashion.32,33 Fog collection, unlike weather modification, is a proven
technology for a substantial supply of potable water in certain arid regions.6,31
Dew collection can be achieved by passing the humid ambient air over a cooled
surface, and condensed liquid water is obtained if the surface temperature is lower
than the dewpoint temperature of air.5

Fog Water Collection. Fog harvest is feasible and technologically accessible to

slake the lack of fresh water, typically in arid coastal areas. The usual method to
collect fog water is placing a rectangular mesh perpendicular to the wind, which
traps fog droplets. When exposed to a foggy environment, water droplets carried
by the wind are pushed against the mesh and become trapped. After successive im-
pacts, the droplets grow by coalescence until they are large enough to fall by gravity,
and a gutter transports the water to a tank. In 1956, the Catholic University of the
North in Antofagasta in northern Chile conducted the first conclusive experiments
with nets.36 Since 1987, several experiments following such ideas but on larger
scales were conducted by Schemenauer and colleagues31 in many arid parts of
the world, for instance, the coastal deserts of West Africa (Namibia), South America
(Chile and Peru), and the Middle East (Saudi Arabia and Oman, monsoon season).
The successful results inspired similar fog-collection projects nowadays that have
been initiated and implemented in many parts of the world.37 Typical conditions
are liquid water content around 0.1–0.5 g/m3, 40% fog immersion time, and 50%
collection efficiency with 3 m/s wind speed.38 Most of the advanced systems using
high-elevation fog have the water production ability of 3–7 kg/day/m2,31,38 and
the literature review results are listed in Table S1. In most of the above projects,
the fog collectors were not assembled close to the residents. The installation of a
pipeline was needed to deliver the water to the users in the mountains. The pipe
costs from collectors to users were one of the major infrastructure costs that made
the system uneconomic and hydraulically difficult. Therefore, Abdul-Wahab
et al.39 examined the potential of residential-type fog collectors built in the vicinity
of the houses. As Schemenauer and Cereceda40 stated, a viable and efficient fog-
extraction project should meet the following criteria: (1) fog must occur frequently
throughout the year and should persist for a relatively long time; (2) high-elevation
fogs, with relatively high liquid water content, are of primary interest for fog water-
collection projects in arid lands; (3) fog collection must be accompanied by wind to
achieve a higher efficiency.

Today, the biggest challenge for fog collection is the low efficiency that is defined
as the ratio between the water reaching the collector’s gutter and the liquid water

Joule 2, 1452–1475, August 15, 2018 1455

Figure 1. Fog Collectors
(A) Traditional fine meshes, having problems of re-entrainment and clogging. As the small deposited water droplets coalesce, the growing droplets are
influenced by the competition between aerodynamic drag forces and surface adhesion forces. When the drag force overwhelms the adhesion force, the
droplets are re-entrained in the fog flow. In the clogging region, the hysteretic wetting force pinning a droplet in the interstices of the mesh exceeds the
gravitational draining force when the deposited water droplet size is less than a critical volume. Reproduced from Park et al. 41 with permission.
(B) Fog harps, utilizing ultrafine-scale and untreated metal wires to both capture and drain fog efficiently and durably. The vertical wires running parallel
to the drainage pathway serve to reduce the pinning force of captured droplets, which enables the efficient drainage of small water droplets even for
micrometric wire sizes. Reproduced from Shi et al. 42 with permission.
(C) Spider-web-like cavity-microfiber topological networks, where the cavity-microfiber is fabricated via the one-step microfluidic method. If a water
droplet is located on a knot near the intersection, it moves directionally toward the intersection and coalesce with the droplet at the intersection.
Reproduced from Tian et al. 43 with permission.

flux normal to the collector’s mesh.38 Collection efficiency is adversely affected by

two issues that depend on the surface wettability: re-entrainment of deposited
droplets and clogging of the mesh with pinned droplets (Figure 1A). The wire
meshes currently utilized for fog harvesting suffer from dual constraints: coarse
meshes cannot efficiently capture microscopic fog droplets, whereas fine meshes
suffer from clogging issues. While superhydrophobic surface treatments can pre-
vent clogging, they are often not durable enough for long-term use. Shi et al.42
have presented an approach to avoid these problems (Figure 1B); they used
small-scale, vertically arranged wires (‘‘fog harps’’) to replace the traditional
cross-like mesh. Another study demonstrated the high efficiency of spider-web-
like meshes (Figure 1C) assembled by low-cost cavity-microfibers in directional
water transportation.43 Frankly, these methods show great potential for collection
efficiency improvement, but these studies were only conducted under controlled
laboratory conditions and still need more field tests to verify their true
performance.

Dew Water Collection. Among alternative approaches, dew water collection5,44–47

has been widely recognized as the ideal candidate because it is minimally affected
by climatic and geographical constraints compared with fog water collection. It is
also more cost-effective than the cloud seeding in less cloudy areas. Initially research
mainly focused on passive radiative condensers,4,5,48,49 which do not need extra en-
ergy input. Such condensers normally work during the night since the solar radiation
in the daytime usually results in a higher surface temperature than the ambient tem-
perature. Researchers in Sweden and France3,4 developed a correct theoretical basis
to build efficient radiative condensers. The International Organization for Dew Utili-
zation has standardized the characterizations of dew collection in terms of method-
ology, instrumentation, and data obtained from field experimental tests. This

1456 Joule 2, 1452–1475, August 15, 2018

organization recommends the use of a standard material made of a special white
low-density polyethylene (PE) foil. This material provides hydrophilic properties
that lower the nucleation barrier at the onset of the condensation process together
with a high emissivity in the near infrared (wavelength range of 7–14 mm). These two
features are important in favoring dew formation. According to the radiative cooling
ability available for condensation, the upper limit of dew yield is 0.8 kg/day/m2.5
However, the maximum recorded yields of dew water in arid and semi-arid climates
typically fall within a range of 0.3–0.6 kg/day/m2 of surface area5 (for the literature
review results see Table S2). In general, the process is limited by the rate of radiative
heat exchange, the weather conditions, and the surface properties. In particular,
weather conditions dictate the ratio of latent to sensible heat exchange between
the surface and the air.

With the advance of chillers, active condensers24,44,45 were designed as an alterna-

tive method for dew collection that produces more water by using additional energy
inputs. Early active dew condensers were developed in the 1930s, but rapid and wide
innovations did not occur until the commercialization of mechanical refrigeration af-
ter the 1980s. Active condensers are now considered as an innovative option for
locally managed water supply systems in areas with water quality and/or quantity
problems. Active condensers work in a manner similar to that of a dehumidifier, which
extracts water from the air. It is reported that the unsatisfied high energy consump-
tion was due to the inefficient heat-exchange process and the low amount of water
obtained by small units.50 Several designs of active condensers have been patented,
of which the yields can reach 20 L/day by portable devices and up to 200,000 L/day by
larger agricultural water devices.5,51 The energy consumption mainly depends on the
system design. To date, most of the commercially available AWGs are using conven-
tional air-conditioning technology to generate water, and the tested efficiency rates
are within 650–850 Wh (electricity)/kg and among which the best efficiency is ca.
250 Wh (electricity)/kg.52 The literature review results are listed in Table S3. Note
that lowering the temperature of the moist air can be achieved by a much simpler
solution that directly expands the air in an expander. Following this idea, the
feasibility of the expander-based AWG has been explored in theory by Subiantoro53
and Shelton and Christiansen.54 Despite its advantages of compactness and
simplicity, few experimental prototypes have been reported.

To reduce the fossil fuel energy cost, researchers began to use solar energy to drive
the AWGs.55,56 Two methods have been reported and concluded by various inves-
tigators: one is the sorption-regeneration-condensation method,57 and the other is
dew water harvesting by a solar-powered sorption chiller.58

The sorption-regeneration-condensation method is an old wisdom that uses desic-

cants to capture the moisture from the air in the shade and then desorb the water out
of the desiccant under the sun. The generated water vapor will be liquefied and
removed while the reactivated desiccant will be cooled down in the shade again
for further water-capture cycles. The liquid desiccant was preferred because of its
system continuity. However, an additional large cooling surface is necessary to
cool the reactivated solution. So a modified version25,59 is proposed and applied
to simplify the system. The desiccant materials extract moisture from the humid air
during nighttime and the regeneration process occurs during daytime utilizing solar
heat. In this case, the desiccant in the sorption phase can be cooled by the feed air
itself during the night. There are three kinds of basic configurations of the sorption-
regeneration-condensation systems, as shown in Figure 2. Figures 2A and 2B show a
kind of glass-covered greenhouse (also called solar still).60–65 At night, the window or

Joule 2, 1452–1475, August 15, 2018 1457

Figure 2. Sorption-Based AWGs
(A and B) Glass-covered greenhouse sorber. Reproduced from Kumar et al.63 and Fathieh et al. 64 with permission.
(C) Sandwich plate sorber. Reproduced from Kim et al. 67 with permission.
(D) Packed columns sorber. Reproduced from Wang et al.70 with permission.
Their SWPs depend on the temperature difference between the desiccant and the condenser surface, thus the optimal designs need to maximize the
solar radiation heat received by the desiccant and keep the condenser surface cool enough. In general, (A) has a smallest average temperature
difference during daytime due to the high-temperature glass surface; (D) has the largest average temperature difference owing to water cooling. The
aims of (B) and (C) are to improve the desiccant temperature and to reduce the condensing temperature, respectively.

the glass cover is opened and the desiccants capture the moisture in the feed air by
natural or forced convection. During daytime, the sorber absorbs the incident solar
radiation and thus the desiccant temperature increases. At this time, the released
water vapor is condensed underneath the glass cover (Figure 2A) or in an additional
condenser only powered by ambient cooling (Figure 2B). Figure 2C is a sandwich
plate,66,67 which integrates the solar absorber and desiccant layer on two sides of
a metal plate, while another plate on the opposite side acts as the condenser.
Figure 2D shows a packed columns sorber and a separated condenser.68,69
The previous three (Figures 2A–2C) are passive AWGs, which have been studied
extensively due to their simple construction. The water productivity is in the range
of 1.0–2.5 kg/day/m2 of collector area, and the literature review results are listed
in Tables S4 and S5. The last AWG (Figure 2D) is an active sorber, more compact
but inevitably with more complex structure. This configuration needs an indepen-
dent solar heat collector and an additional condenser. The reported lowest SEC is
ca. 2.0 kWh (heat)/kg.70

Dew water harvesting by sorption chillers can work continuously in daylight and at
night if integrated with heat storage. This technology is one kind of dew water col-
lector driven by direct cooling. Some researchers58 claimed that this method was too
complicated and expensive to be applied in water production. However, the related

1458 Joule 2, 1452–1475, August 15, 2018

research is insufficient because only the simplest adsorption chillers were applied in
experiments.

Solar photovoltaic (PV)-powered thermoelectric condensers47,71–74 are usually cho-

sen among portable AWGs to meet the needs of low water production in some spe-
cial applications due to their small volume and low cost. For instance, Muñoz-Garcı́a
et al.47 built a mini AWG based on the thermoelectric effect to provide the minimum
amount of soil moisture needed by young trees for successful rooting when they are
transplanted from the nursery into open fields, especially in dry-climate areas. Joshi
et al.75 developed and experimentally investigated a thermoelectric fresh water
generator for individual person usage.

An interesting application concerns the HVAC (heating, ventilation, and air-condi-

tioning) systems. Cooling the air below its dew point (mostly in summer times) is a
dehumidification process, which produces a lot of condensed water that usually is
thrown away. Such condensed water from air-conditioner cooling coils usually
ends up in municipal sewerage systems and wasted. Therefore, an integration of
an air-conditioning system and a water-harvesting system could be a potential solu-
tion to achieve optimized energy consumption and sustainable use of resources. In
this perspective, the water production can be considered almost as a free benefit, or
in other words is obtained as the ‘‘by-product’’ of a smart HVAC system design.
Obviously, two fundamental conditions need to be fulfilled in order to obtain
good performance by applying such integrated system: firstly, a real indoor air-con-
ditioning process rather than on-purpose water harvest is necessary; secondly,
compatible external hygrometric conditions are required. Recently, an excellent
literature review conducted by Salem et al.76 has summed up the HVAC condensate
recovery technologies and their applications.

In brief, the technologies of AWGs summarized in Table 1 can be divided into three
main categories: direct harvesting by condensation, vapor concentration by mem-
brane or desiccant, and by-product collection from an integrated system, as shown
in Figure 3. Meanwhile, according to the dependence on external energy input, both
the vapor concentration process and the water vapor condensation can be catego-
rized as the passive one without extra energy input and the active one powered by an
external energy source. Selection of methods is an engineering decision depending
on both local climatic conditions and economic factors such as capital, operation,
and energy costs.

In the last 20 years, several modifications and improvements have been proposed in
the literature to improve the performance (e.g., water productivity or efficiency) of
the water-harvesting system, regarding different weather conditions, materials, or
collector design. However, until now the traditional AWGs have been restricted
only to small-scale portable water production and emergency water supply after a
disaster. It is still far away from people’s expectation of scalable water production
in an energy-efficient way. The fog water collection is efficient but limited to specific
locations;37 the passive radiative collector has a low SWP due to the low radiative
cooling capacity;5 the passive sorption-based AWG also has a low SWP due to its
low RR;77 and the chiller-based dehumidifier has an unexpected SEC associated
with the low COP of the chillers.23,24 In this case, the solar-powered sorption-based
AWGs with active condenser reveal a great potential for compact systems, efficient
operation, and wide adaptability. However, for reduction of the SEC according to
Equation 1, several remaining hurdles must be surmounted prior to the practical
application of sorption-based AWGs with active condenser for scalable atmospheric

Joule 2, 1452–1475, August 15, 2018 1459

Table 1. A Brief Summary of the Methods, Problems, and Perspectives of AWGs
Methods Features Capacity Problems Alternatives Challenges Applications
Cloud seeding hygroscopic NA considerable sky river proposed hard to predict weather
seeding skepticism by Wang cloud migration modification
Fog seeding hygroscopic salt NA hard to predict fog dew collection PCM: sorption microclimate
particles occurrence by using PCM solidification management
Fog mesh easy to construct 1.5–12 limited to locations nano-engineering low collection mount by the sea
kg/day/m2 mesh surface efficiency
Massive dew collector passive way very little thin valid layer PCM dew collector lack of high water-saving
output e* materials agriculture
Radiative dew collector passive way 0.3–0.6 large heat loss high-performance metamaterials low-T water
kg/day/m2 emitter production
Electric chiller high output 0.25 Wh/g great latent load DEHP durability portable water
Sorption chiller compact (electricity) and reliability production
Sorption-based AWG ambient cooling 1.0–2.5 high Tcond water-cooling Heat transfer scalable water
(solar distiller) kg/day/m2 condenser enhancement supply
Sorption-based AWG active condenser 2.0 kWh/kg large heat loss solar water-heating pump scalable water
(solar air heating) (heat) sorber supply
Sorption-based AWG air cooling 1.2 kg/day/m2 thickness high k desiccant desiccant electronic cooling
(sandwich plate) of desiccant layer
TEC dew collector small &compact COP < 0.1 low energy desiccant-enhanced proper meet minimum
(solar PV driven) efficiency heat sink application water load
Integrated system by-product depends on cooling NA NA offsetting water
the latent load dependence use of A/C

AWG, atmospheric water generator; COP, coefficient of performance; DEHP, desiccant-enhanced heat pump; DRH, deliquescence relative humidity; MHI, Mois-
ture Harvesting Index; RH, relative humidity; RR, recovery ratio of the feed air; SEC, specific energy consumption; SWP, specific water productivity; PCM, phase
change material.

water production: (1) the lack of cost-effective heat sinks to maximize the SWP; (2)
the energy-inefficient ways applied to desorb the desiccants; (3) the unreasonable
design that fails to reduce heat loss or to achieve heat recovery before and after
the condensing process; and (4) the ignorance of the kinetics in terms of material se-
lection and sorber design. For sorption-based AWGs, besides the challenges for
thermodynamic performance mentioned above, some current highly utilized desic-
cants, such as MOF-801,78 are still unstable and unhealthy, which may result in se-
vere safety issues. In addition, the methodologies on how to evaluate the feasibility
and energy requirement of atmospheric water harvesting are also very important
factors for location-specific and climate-specific design, but these still do not attract
sufficient attention from researchers.

Progress
Progress in Lowering the Condensing Temperature
About 50% of the energy consumed in harvesting portable water by cooling down
humid ambient air is wasted by producing unexpected cold air rather than water.24
Hence, lowering the condensing temperature can improve the RR and eventually
reduce the SEC. This can be achieved in two aspects: a cost-effective heat sink
and a high effective condenser.

Cost-Effective Heat Sinks. Obviously, the cold dark Universe is an ideal low-tem-
perature heat source. Radiative cooling technology utilizes the atmospheric trans-
parency window (wavelength range of 8–13 mm) to passively dissipate heat from
Earth into outer space (3 K).79,80 The technology utilized in radiative dew water col-
lector is relatively simple as it relies on exploiting the physical processes of dew

1460 Joule 2, 1452–1475, August 15, 2018

Figure 3. Technical Classification of AWGs

formation, without additional energy input. A well-performing radiative collector re-

quires not only high radiative cooling power but also a slippery surface that allows
the quick removal of condensate.5 However, the radiative cooling power of passive
dew collectors is a function of surface pattern (higher emissivity in infrared spectrum,
small thermal mass, high wetting properties, and good thermal insulation) and the
weather conditions (ambient temperature, relative humidity, wind speed, and cloud
cover), which affect the yield in a relatively complex way. A traditional radiative
cooler is designed for surface cooling at night whiling preventing warming up of
the condenser during the daytime. For cooling at night, the demonstrated temper-
ature reduction from ambient air is in the range of 15 C–20 C15 in typical populous
areas.

The influence of solar absorption during the day that causes radiative cooling
spread to a subambient surface seemed an insurmountable barrier until
2014.48,80 A thin polyethylene cover used to suppress convective exchange is
transparent to infrared radiation. Below this cover, a stack of two thin oxides is
deposited over a 200-nm-thick of silver layer on the silicon substrate, as shown
in Figure 4A. The polymer acts as an all-dielectric mirror and its performance of re-
flecting blue wavelengths is even better than metal, leading to a very high solar
reflectance. As a result, only 30 W/m2 out of 1,000 W/m2 incident solar energy
is absorbed. The maximum temperature reduction in the absence of a convective
barrier is 3 C at midday and 7 C at night in summer. About 2 years later,81 another
kind of radiative cooling device was reported as shown in Figures 4B–4D, which
demonstrated a temperature reduction record. In a 24-hr day-night cycle in winter,
the cooler is maintained at a temperature that is at least 33 C below ambient air
temperature of 5 C–15 C, and a maximal temperature reduction of 42 C is
observed. This maximum value occurs when the apparatus enclosing the cooler
is exposed to the peak solar irradiance. In recent years,49 with the emergence of
new classes of selective infrared emitters, the radiative cooler can provide a cool-
ing power that exceeds 100 W/m2 at reasonable temperatures.

Joule 2, 1452–1475, August 15, 2018 1461

Figure 4. Radiative Coolers
(A) The first selective IR emitters that can achieve daytime radiative cooling. Reproduced from
Raman et al. 48 with permission.
(B) The transmittance of the ZnSe emitter used in (C).
(C) The structure of the irradiative cooler achieving the record largest temperature reduction.
(D) The cooling performance of the cooler in (C).
(B)–(D) are reproduced from Chen et al. 81 with permission.

Of course, the Earth is also a good heat sink in summer considering the moderate
temperatures deep in the ground to boost condensing efficiency. Depending on
latitude, the temperature 6 m beneath the Earth’s surface maintains a nearly con-
stant value between 10 C and 16 C. Ground-coupled heat exchangers82 have
been widely used to warm or cool air or water and applied for residential, agricul-
tural, or industrial uses. However, few researchers have investigated the ground
heat exchanger to condense the water vapor for AWGs.

Besides radiative cooling, a noteworthy technique is cold storage at night and heat
absorption during the daytime. The obvious variation of temperature between day
and night in arid regions is very common,25 thus a cold storage tank that can be
charged during the night can work as a low-temperature heat source in the daytime.
With the rapid development of heat storage, quantities of novel materials with ultra-
high thermal diffusivity have been produced,83–85 which show a bright future for at-
mospheric water harvesting by using resonant ambient heat.

Condensation Enhancement. Dew formation on the condenser is influenced by the

properties (e.g., surface roughness and chemical heterogeneities) of the material
used as the surface cover. Properties such as initial nucleation of droplets, coales-
cence, and fast removal are expected.86 Obviously, a traditional hydrophilic or hy-
drophobic surface cannot satisfy these requirements at the same time. For the hy-
drophilic surface, a lower energy barrier may assist the initial nucleation of
droplets, making easier the coalescence of the droplet. However, large pinned
droplets with further droplet growth and coalescence will form an insulating liquid
film that may isolate the bulk air and the surface, alleviating or even stopping the
condensation process. Instead, dropwise condensation on a hydrophobic surface al-
lows the condensed droplets to roll off by gravity on a length scale comparable with

1462 Joule 2, 1452–1475, August 15, 2018

the capillary length of water. The surfaces may then refresh for renucleation and
growth of next-generation droplets, resulting in 5- to 7-fold higher heat transfer per-
formance. However, a large number of micro-defects inevitably exist on the current
hydrophobic surface, resulting in the formation of large pinned droplets that may
eventually impede the condensation. In the hope of solving this problem, a very
tempting solution obtaining spontaneous droplet jumping on nanostructured sur-
faces has been attracting increasing attention. For example, Zhai et al.49 presented
an approach to eliminate the micro-defects by using three-dimensional (3D) copper
nanowire networks and successfully demonstrated stable and efficient jumping
droplet condensation on a superhydrophobic surface. Similar studies bioinspired
from natural plants or animals offer several new routes. Two typical approaches
are: (1) 1D hydrophilic slippery rough surface to accelerate the droplet removal by
gravity before flooding occurrence;35 and (2) 2D hybrid surfaces with mixed wetta-
bility to delay uncontrolled flooding.87

For acceleration of droplet removal, a novel concept87 of integrating the growth and
transport of water droplets is derived from a combination of the strategies used by three
distinct biological examples—Namib desert beetles, cacti, and Nepenthes pitcher
plants—as summarized in Figures S1A–S1C. First, synthetic bumpy surfaces inspired
by beetles are designed to optimize fast and localized droplet growth by focusing vapor
diffusion flux at the apexes. The formed droplets then rapidly roll off the asymmetric
slope in a direction-guided structure, which is similar to cactus spines, to guide capil-
lary-driven transport of harvested water drops. Lastly, the molecularly smooth lubricant
is coated on the bumps to dramatically diminish the friction. Experimental results
confirmed that bumps that were rationally designed by integrating these mechanisms
are able to grow and transport large droplets even against gravity and overcome the
effect of an unfavorable temperature gradient.

Another interesting bioinspired design35 is a hydrophilic directional slippery rough

surface that is favorable for both droplet nucleation and removal, as shown in Figures
S1D–S1F. The slippery rough surfaces combine the unique surface functions of
pitcher plants (slippery interface) and rice leaves (micro-grooves parallel to the
sliding direction of the droplet and the nanotextured surface on microscale bumps),
which can repel liquids regardless of the wetness of surfaces. Molecular dynamics
simulations showed that the physical origin of this efficient droplet nucleation was
attributed to the hydrophilic surface functional groups, whereas the rapid droplet
removal was due to the significantly reduced droplet pinning of the directional sur-
face structures and slippery interface. Experimental results further demonstrated
that the slippery rough surfaces, owing to their larger surface area, more hydrophilic
slippery interface, and better directional liquid repellency, outperform conventional
liquid-repellent surfaces in water-harvesting applications.

The strategy of enhancing the condensation process by accelerating droplet removal

has wide applications in almost all atmospheric water-harvesting technologies. How-
ever, considering that the condensation heat eventually needs to be released through
the surface layer into the cooling media, the thickness and the thermal conductivity of
the surface material are also of great significance. Nevertheless, this information is still
mostly missing in current research. Besides, the effects related to the accumulation of
dust or other small solid particles on the duration and stability of these nanotextured
surfaces have not been carefully investigated. Therefore, investigations on scalable
nanostructured surfaces made from metals or plastics are still eagerly awaited to
improve condensation heat transfer in realistic industrial applications. Despite the
incompleteness, this research is still valuable because a bright future lies before us.

Joule 2, 1452–1475, August 15, 2018 1463

Figure 5. Promising Desiccants for Sorption-Based AWGs
The moisture capture performance mainly depends on the ambient relative humidity (RH). The
region where the AWGs may be competitive can be divided into three categories according to the
ambient RH. In general, the RH in arid regions is usually less than 30%. When the ambient RH is
larger than 80%, the fog occurs relatively easily, particularly on a sunny morning. In the humid
regions, the RH is in the range of 30%–80%, but the precipitation is rare.

Progress in Vapor Concentration

Since AWG by direct cooling wastes a large portion of the energy on cooling the air,
vapor separation methods, only cooling the water vapor, may provide a technolog-
ical alternative to reduce the operational costs of AWGs. A way to reduce this sen-
sible heat load is to concentrate the water vapor.88 Vapor concentration can be
achieved by using desiccants that adsorb the vapor from the air, which can later
be recovered in a thermal-driven step.77

Desiccant materials can be either solid or liquid. Traditionally, it is believed that

solid desiccant systems can produce potable water but require large volumes of
desiccant. Significant operating costs are also expected for the blowers to circu-
late both the fresh air for adsorption and the hot air for desorption. Furthermore,
the efficiency of the desiccant bed will suffer a long-term reduction due to dust or
other impurities deposited into the pores. Although this problem can be solved
by adding air filters, this will lead to additional air pressure drop throughout
the system. These drawbacks make the solid desiccant system less attractive.
However, novel desiccants, such as nanoporous inorganic materials,89 metal-
organic frameworks (MOFs),90–92 and composite materials93,94 presented in Fig-
ure 5, show a much greater potential for water-harvesting systems than before.
Composite desiccant materials are most frequently utilized in sorption systems,
which are formed by impregnating hygroscopic salts into the porous desiccants.
Since the hygroscopic salts play a major role in water absorption, water uptake
of composite desiccants, such as ACF/LiCl94 (Figure 5) and SG/LiCl95 (Figure 5),
increase with the partial pressure of water vapor.94,96 For polymeric materials,
such as AlPO4-LTA97 (Figure 6A), Co2Cl2(BTDD)90 (Figure 6B), MIL-101,98,99
MOF-80166,67 (Figure 6C), MOF-30364,100(Figure 6C), PIZOF-2,91 nanoporous
material zeolites 13X,94 and mesoporous molecular sieves MCM-41,94 their water
absorption performance can be stepped within a very narrow relative humidity
(RH) range due to their high degree of consistency of microstructures. Other

1464 Joule 2, 1452–1475, August 15, 2018

Figure 6. Strategies for Designing Superior Sorbents
(A) AlPO4-LTA. Microporous AlPO 4 -34 is a very promising zeolite-like desiccant with highly
hydrothermal stability, large sorption capacity, and narrow RH interval. According to the database
of zeolite structure types, the accessible pore volumes for the LTA cages are expected to be
somewhat larger than that of AlPO 4 -34 with CHA topology. This inspires an idea that a microporous
AlPO 4 -n analogous to LTA-type Zeolite might have the even higher capacity and equally sudden
water uptake as AlPO 4 -34. This suggests that perhaps an even more capable material than AlPO 4 -
LTA can be found among the microporous AlPO 4 -n with even larger pores. Reproduced from
Andraz et al. 89 with permission.
(B) Co 2 Cl2 (BTDD). Co 2 Cl 2 (BTDD) illustrates a general strategy for designing superior sorbents.
Through open adsorbing metal sites binding water prior to pore filling, the pore diameter can be
effectively reduced to the critical diameter. Reproduced from Rieth et al. 90 with permission.
(C) MOF-303 and MOF-801. MOF-303 has the same water-uptake capacity as MOF-801, but uses
low-cost aluminum instead of toxic and expensive zirconium as the metal and uses water instead of

Joule 2, 1452–1475, August 15, 2018 1465

Figure 6. Continued
organic solvents (DMF) to wash the initial precipitates. It provides a novel approach to make MOFs
with enhanced water sorption properties potentially meeting the specification of industrial large-
scale production. Reproduced from Fathieh et al. 64 with permission.

candidates such as SG94 and UiO-6691 have increased water-uptake performances

with a higher RH.

Through comparing the isotherms (at 25 C) (Figure 5) of some very promising des-
iccants reported recently, it is easy to see that excluding the zeolites 13X,94 the
ACF/LiCl composites94 have the highest water uptake in the arid regions (RH <
30%) under the same RH. In other words, the ACF/LiCl composite94 can take place
of those expensive polymeric materials (labeled in red lines in Figure 5), including
AlPO4-LTA,97 Co2Cl2BTDD,90 MIL-101,98,99 and MOF-801,66,67 particularly consid-
ering that these materials usually have a relative small hysteresis. The same trend
for silica gel/LiCl composite is witnessed in the humid regions, whose water-uptake
capacity is much better than the UiO-66, MCM-41 and silica gel. However, a risk for
these two composites should be noted: the leakage of the salt solution perhaps oc-
curs because of overloading when the RH is larger than their deliquescence relative
humidity (DRH).101 This is why the scientists and engineers are still exploring other
high-performance physical sorbents.

For screening of the optimal physical sorbents, the material properties of stability,
hydrophilicity, and pore diameter are of critical importance. MOFs offer the flexi-
bility required to optimize all these parameters at once.90 To facilitate the applica-
tion of the desiccants in AWGs, the pore hydrophilicity will usually be sufficient
enough to allow the water nucleation and the pore-filling below an approximately
30% RH condition. Meanwhile, to avoid undesirable hysteresis upon water desorp-
tion, the pore size must be smaller than the critical diameter Dc = 4sTC/(TC
T)102 of
the adsorbates. This implies that an adsorbent with a pore diameter of approxi-
mately Dc will maximize the internal volume available for filling with water while
avoiding irreversible capillary condensation. The corresponding material design
strategies are shown in Figure 6. Otherwise, it is noteworthy that the material den-
sities are of great importance when comparing the different materials’ adsorption
capacity, because the adsorption capacity per unit volume is more important in prac-
tical application.

Of course, the vapor concentration can also be achieved by using polymeric electro-
lyte membranes (Figures 7A and 7B) or water vapor selective membranes (Figures
7C–7E) that allow the separation of water vapor from other molecules in air prior
to the cooling process. Theoretically, the use of water vapor selective membranes
can reduce the energy requirement for extracting water out of the humid air by
more than 50%,88,103 but until now related experimental investigation has not yet
been conducted. Instead, current experimental results show that only 30% of the to-
tal power input was effectively used by the polymeric electrolyte membrane to pump
humidity,104 while the rest was wasted, leading to a relatively low COP. Thus, with
regard to the limitation of the technology itself, the membrane separation is less
competitive for vapor concentration in the foreseeable future.

Progress in Heat Recovery

In fact, one of the factors that lead to the high energy requirement of the water va-
por condensers is the large temperature difference of heat transfer. A large portion
of the cooling capacity of the cold fluid is wasted on producing cold air. Recently,

1466 Joule 2, 1452–1475, August 15, 2018

Figure 7. Membrane Facilitated AWGs
(A and B) Principle (A) and structure (B) of the polymeric electrolyte membrane. Reproduced from Qi et al. 104 with permission.
(C–E) System illustration (C) of the AWG based on water vapor selective membrane, with microstructure photos (D and E). Reproduced from Bergmair
et al. 88 with permission.

two kinds of precooling design with heat recovery have been patented. The first
one105 achieves precooling of the inlet moist air of the evaporator by recovering
the cooling capacity of the exhaust air of the same evaporator. The second
one106 designs a novel refrigerant path to obtain a large subcooling and a minimum
heat transfer difference. These were very illuminating for several further
investigations.

Progress in Location-Specific and Climate-Specific Design

Naturally, different locations experience distinct microclimate due to their specific
altitudes, latitudes, distances to the coastline, surface albedo, vegetation cover,
and other land cover attributes. Meanwhile, the weather conditions in a fixed loca-
tion may also vary in a wide range in one day or as long as one year. Recently,
Gido et al.24 conducted a climate-specific analysis to assess the feasibility and en-
ergy requirements of atmospheric water harvesting by direct cooling in global
range. They analyzed the fraction of the time suitable for AWGs by direct cooling
in a total of 30 cities of 14 countries. Although this work is inspiring, it is used only
to make decisions when choosing the chiller-based AWGs at a certain location.
No information about alternatives is provided. Therefore, much more work on loca-
tion-specific and climate-specific design is expected in the future.

Perspectives
System Design Principle: Match Desorption Operation with Moisture
Condensation
During the desorption-condensation process, the inlet air of the condenser is at the
state of the outlet air of the desorber. For the condenser, an inlet air condition of low
temperature and high humidity is expected to achieve a low SEC, but an outlet air
condition of high temperature and low RH indicates that the desorber can release

Joule 2, 1452–1475, August 15, 2018 1467

Figure 8. System Design Principle
(A and B) The optimal condensing temperature (A) and the isotherms of an ideal desiccant (B).
Reproduced from Alayli et al. 25 with permission.

as much water vapor as possible to obtain a high SWP. For a given outlet air state of
the desorber, a lower condensing temperature can result in a higher SWP But also
causes a higher SEC. In this case, the desorption temperature and the condensation
temperature should be carefully preferred to balance the SEC and SWP. However, it
is still an open question as to how to determine the proper desorption temperature
and the condensing temperature. Through thorough psychrometric analysis, we find
a powerful tool to solve this problem. The inlet air state of the condenser is labeled
as point I (Ti, di), and the condensation states of the moist air are labeled by colorful
stars located on the saturated line in the psychrometric chart (Figure 8A), so that the
tangent of angle q can be expressed as follows:

Ti
Tcond
tan q = : (Equation 3)
di
dcond
Thus, SEC = Cp ð3 T =3 d Þ 3 tan q + hfg . For a given condenser, if the condensation
temperature is the only variable, the SEC is almost a linear function of the tangent
of angle q. Note that the tangent of angle q (qH / qopt / qL) will decrease first
and then increase with the reduction in condensation temperature. Therefore, there
exists an optimal condensation temperature that achieves the lowest tangent of the
angle qopt and eventually the smallest SEC. It is easy to understand when the line
through the point I (Ti, di) is tangent to the saturated line: the angle q is the smallest.
Therefore, the optimal condensing temperature Tcond,opt for a minimum SEC at
given inlet conditions I (Ti, di) can be obtained at the tangent point as shown in Fig-
ure 8A. Furthermore, we can find many iso-SEC or iso-Tcond,opt lines. Instead, for a
designed condensing temperature, appropriate condenser inlet air conditions can
be found along with the iso-SEC lines and then the optimal desorption operating
conditions can be obtained. For example, an appropriate heat source temperature
and airflow rate can be selected by using this tool to obtain the expected outlet air
conditions of the desorber.

Ideal Sorbents: Temperature-Sensitive Materials

Every specific application requires an ideal sorbent. For example, the optimal des-
iccants for an adsorptive chiller or heat pump are ones with S-type isotherms. Their
sudden uptakes mean that the desiccant can be regenerated by lower temperature
heat sources. Meanwhile, their narrow RH interval is well in accordance with the fixed
working conditions of the adsorptive chiller or heat pump. However, an open sys-
tem, such as AWGs, usually experiences a very wide range of weather conditions
during one day or throughout a whole year. Thus to maximize the capacity of the

1468 Joule 2, 1452–1475, August 15, 2018

Figure 9. Efficient Condenser Design
(A–D) Water cooling condenser assembly with heat recovery (A) and alternative cost-effective cold
water produced by ground source heat exchanger (B); cold storage tank (C); flat panel with radiative
cooling (D). Reproduced from Zhao et al.108 with permission.

AWG, a linear isotherm is expected for dehumidification, whose uptake will increase
along with the rising RH.

Similarly, according to the above analysis, the ideal sorbents for air-water harvesting
should have features as follows: (1) in the adsorption process (<25 C), their water
sorption capacity should increase linearly with RH; (2) in the desorption process
(>35 C), their water sorption capacity should drop steeply with increased tempera-
ture, shown as S-type isotherms. Although it is difficult to fully satisfy these features
with the current commercial or laboratory desiccants, which are either temperature-
insensitive or RH-sensitive, Alayli et al.25 had reported a composite materials, a
mixture of a natural wool and polyester doped with CaCl2, which had typical linear
isotherms at low temperatures and S-shaped isotherms at high temperatures, as
shown in Figure 9B. Obviously this is an ideal desiccant, but unfortunately little
attention has been attracted and scarce research regarding the material perfor-
mance can be found in the past 30 years. Recently, Kallenberger and Fröba107 re-
ported another composite material that incorporated calcium chloride into an algi-
nate-derived matrix. The water sorption isotherms of the fully dehydrated material at
28 C, 65 C, and 85 C show that its water uptake almost linearly increases with RH
from 0% to 79% at 28 C. However, when the adsorption temperature rises to
65 C, the water uptake is very little, meaning the desiccant can be regenerated at
a temperature as low as 65 C. This work thus provides a useful approach to design
high-performance temperature-sensitive desiccants for AWGs. Of course, if the har-
vested water by AWGs is used for electronic cooling, an S-type isotherm is necessary
and a lower RH at the onset of the isotherm is better because the ability to collect
sufficient water under any climate condition is of primary importance.

Joule 2, 1452–1475, August 15, 2018 1469

Efficient Condenser Design
Given that water-cooling condensation has a much higher heat-exchange effective-
ness than air cooling, a water-cooling condenser with heat recovery (Figure 9A, the
same as the evaporator used in the direct expansion chiller105) may have a higher
condensing efficiency. This is particularly helpful to the sorption-based AWGs where
the regenerated moist air has a very high temperature. Meanwhile, three facts are
worthy of note: (1) the earth collectors have been successfully proven, but no
research has investigated the potential of dew water harvesting using the cold water
from ground heat exchangers (Figure 9B); (2) the massive dew collectors have been
proven ineffective due to the low specific heat capacity and the low thermal conduc-
tivity of the stone, although high-performance phase change materials (PCMs)
recently have been able to solve these two problems83 (Figure 9C); (3) with the rapid
development of radiative cooling technology,108 the radiative cooling panel not
only can be directly used to condense the dew, but also can work as a cost-effective
heat sink to improve the performance of the sorption-based AWGs (Figure 9D).

Efficient Sorber and System Designs

Sorption-based atmospheric water harvesting is a vital technology for cost-effective
portable water production in arid areas. However, a key obstacle to its deployment
on a large scale remains its cost, both capital and operating. In this context, the
development of high-performance desiccants is a key research priority. Conse-
quently, a great effort has been made to develop new materials for this purpose,
with thousands of new materials having been proposed in the literature. One com-
mon feature of these studies is that they mainly focus on the equilibrium water-up-
take capacity of the desiccants and frequently ignore other key factors such as kinetic
properties.109 To date, the majority of this effort has consumed plentiful time and re-
sources and has almost exclusively focused on developing desiccants with increased
water-uptake capacity and reduced heat of regeneration. This would appear rational
given that desiccant regeneration largely affects operational cost. The transport and
kinetic properties of the desiccants directly determine the size of the sorption-based
AWGs and, thus, their capital costs. Therefore, in order to develop AWGs that will
result in an obvious cost reduction, it is essential to move beyond the equilibrium-
based metrics of the desiccants.

In the following section, we propose two new system designs for mass transfer
enhancement to improve energy efficiency and reduce the system size. Our ap-
proaches use a kind of water-sorbing, desiccant-coated heat exchanger, the
detailed parameters of which are referred to in several previous papers.110–112 The
water-sorbing heat exchanger can independently handle the sensible and latent
loads at the same time. The desiccants coated absorb moisture almost isothermally
because the heat of adsorption is directly taken away by the thermal fluid flowing
through the heat exchanger. Meanwhile, due to a thin desiccant layer (less than
1 mm), the adsorptive and desorption kinetics are very fast and the desiccant also
can be regenerated by low-temperature source. Therefore, the water-sorbing heat
exchanger has great potential to efficiently concentrate the water vapor.

Solar Water Heating + Cost-Effective Water-Cooling Condensation. Through the

comparison among the aforementioned different sorber configurations, it is clear
that both the traditional greenhouse-type sorber and the solar air heating sorber suf-
fer from a slow adsorption, thus taking a long time to become saturated. This is
because the adsorption processes are adiabatic: the adsorption heat can only be
taken away by the feed air. The temperatures of the desiccant and feed air rise
quickly and remain at a high level for a long time, which eventually retards the

1470 Joule 2, 1452–1475, August 15, 2018

Figure 10. Efficient Sorber and System Designs
(A) Concept of the solar water-heating-powered AWG with heat recovery.
(B) Concept of the solar PV-driven AWG based on the desiccant-enhanced heat pump.

sorption process. Therefore, a new system configuration based on water-heating

desiccant-coated heat exchangers,110,111 as shown in Figure 10A, may be attractive,
with the following features:

(1) Water-cooling adsorption is applied to maximize the RR. The adsorptive

process will be almost isothermal, which is beneficial in enlarging the water
uptake.
(2) Water-cooling condensation is adopted to reduce the SEC. The cold water
comes from ground-source heat exchangers, radiative coolers, or cold stor-
age tank (night storage and day discharge).
(3) The sorber made by a water-sorbing heat exchanger is introduced to inter-
nally cool or heat the desiccants, which further decrease the required temper-
ature of the heat source and the outlet air temperature of the desorber.
(4) Heat recovery is involved between the outlet air of desorber and the outlet air
of condenser to reduce the SEC.

Apparently, it is optimal for locations where the RH differs largely between day and
night.

Solar PV-Driven Desiccant-Enhanced Heat Pump with Passive Heat-Recovery

Condenser. Another very promising technology is desiccant-enhanced heat pump in-
tegrated with a passive heat-recovery condenser as shown in Figure 10B. The fresh air
flowing through the system will be divided into two parts; one part is cooled and dehu-
midified and the other is heated and humidified. The dehumidification effectiveness and

Joule 2, 1452–1475, August 15, 2018 1471

energy efficiency of the solar PV-driven desiccant-enhanced heat pump have been also
investigated and proved in previous papers.110,112 When the two air flows come into the
heat-recovery condenser, the moisture in the humid and hot air will be condensed. Prin-
cipally it is optimal for locations where continuous water production is needed or a con-
stant high RH in a whole year or at least a monsoon season.

A New Vision of Water-Energy-Food Nexus

Water is precious. The water-food-energy nexus is central to sustainable develop-
ment. Agriculture is the largest consumer of the world’s freshwater resources. Cool-
ing in air-conditioning systems, power plants, refining petroleum, producing fuel,
and extracting energy resources from the earth make up a significant fraction
of water use. Thus, the solar-powered scalable atmospheric water-harvesting
technologies can inspire us to create new possibilities of energy-efficient water,
water-efficient energy, and water-saving agriculture. For example, it is a promising
opportunity to bloom in the desert by integrating the solar-powered atmospheric
water-harvesting technologies with water-saving agriculture. Besides, utilizing the
accessible water produced by such technologies can promote the application of
evaporative cooling in overheating protection of outdoor electronic devices, wet
cooling of the power plants in arid areas, and passive cooling in buildings, which
may obviously lower the energy consumption. In this case, the atmospheric water-
harvesting technologies will be adapted to the whole world in the future, far beyond
the arid and semi-arid areas as summarized in Table 1.

SUPPLEMENTAL INFORMATION
Supplemental Information includes one figure and five tables and can be found with
this article online at https://doi.org/10.1016/j.joule.2018.07.015.

ACKNOWLEDGMENTS
The authors gratefully acknowledge Dr. Quanwen Pan and Dr. Zhengyuan Xu for dis-
cussions and Mr. Audun Bull Kristiansen, Mr. Biye Cao, and Ms. Linji Hua for manu-
script modification. This work was supported by the Key Program of National Natural
Science Foundation of China (grant no. 51336004) and the Foundation for Innovative
Research Groups of the National Natural Science Foundation of China (grant no.
51521004).

AUTHOR CONTRIBUTIONS
Proposal, R.W.; Conceptualization, Y.T., R.W., and J.W.; Original Draft, Y.T., Y.Z.,
and J.W.; Review & Editing, Y.T., Y.Z., and R.W.

REFERENCES
1. Mekonnen, M.M., and Hoekstra, A.Y. (2016). 5. Khalil, B., Adamowski, J., Shabbir, A., Jang, production. Renew. Sustain. Energy Rev. 14,
Four billion people facing severe water C., Rojas, M., Reilly, K., and Ozga-Zielinski, B. 1187–1201.
scarcity. Sci. Adv. 2, e1500323. (2016). A review: dew water collection from
radiative passive collectors to recent 8. Peter-Varbanets, M., Zurbrügg, C., Swartz, C.,
2. Wahlgren, R.V. (2001). Atmospheric water developments of active collectors. Sustain. and Pronk, W. (2009). Decentralized systems
vapour processor designs for potable water Water Resour. Manag. 2, 71–86. for potable water and the potential of
production: a review. Water Res. 35, 1–22. membrane technology. Water Res. 43,
6. Fessehaye, M., Abdul-Wahab, S.A., Savage, 245–265.
3. Beysens, D., and Milimouk, I. (2000). The case M.J., Kohler, T., Gherezghiher, T., and Hurni,
9. Loo, S.L., Fane, A.G., Krantz, W.B., and Lim,
for alternative fresh water sources. http:// H. (2014). Fog-water collection for community
T.T. (2012). Emergency water supply: a review
rexresearch.com/dewharvest/Secheresse. use. Renew. Sustain. Energy Rev. 29, 52–62.
of potential technologies and selection
pdf. criteria. Water Res. 46, 3125–3151.
7. Narayan, G.P., Sharqawy, M.H., Summers,
4. Nikolayev, V.S., Beysens, D., Gioda, A., E.K., Lienhard, J.H., Zubair, S.M., and Antar, 10. UN Country Team in China. (2004).
Milimouka, I., Katiushin, E., and Morel, J.P. M.A. (2010). The potential of solar-driven Millennium Development Goals: China’s
(1996). Water recovery from dew. J. Hydrol. humidification–dehumidification desalination Progress (Office of the United Nations
182, 19–35. for small-scale decentralized water Resident Coordinator in China).

1472 Joule 2, 1452–1475, August 15, 2018

11. Macedonio, F., Drioli, E., Gusev, A.A., 28. Beysens, D., Milimouk, I., Nikolayev, V.S., 43. Tian, Y., Zhu, P., Tang, X., Zhou, C., Wang, J.,
Bardow, A., Semiat, R., and Kurihara, M. Berkowicz, S., Muselli, M., Heusinkveld, B., Kong, T., Xu, M., and Wang, L. (2017). Large-
(2012). Efficient technologies for worldwide and Jacobs, A.F.G. (2006). Comment on ‘‘The scale water collection of bioinspired cavity-
clean water supply. Chem. Eng. Process. 51, moisture from the air as water resource in arid microfibers. Nat. Commun. 8, 1080.
2–17. region: hopes, doubt and facts’’ by Kogan
and Trahtman. J. Arid Environ. 67, 343–352. 44. Sharan, G., Roy, A.K., Royon, L., Mongruel, A.,
12. Schemenauer, R.S., and Cereceda, P. (1991). and Beysens, D. (2017). Dew plant for bottling
Fog-water collection in arid coastal locations. 29. Monteith, J.L. (1957). Dew. Q. J. Royal water. J. Clean. Prod. 1, 83–92.
Ambio 20, 303–308. Meteorol. Soc. 83, 322–341.
45. Sharan, G. (2008). Harvesting dew water using
13. Olivier, J., and Rautenbach, C.J.D. (2002). The 30. Gindel, I. (1965). Irrigation of plants with radiative-cooled condenser to supplement
implementation of fog water collection atmospheric water within the desert. Nature drinking water supply in hot arid coastal area
systems in South Africa. Atmos. Res. 64, 207, 1173. of north-west India. In International
227–238. Conference on Agricultural Engineering.
31. Klemm, O., Schemenauer, R.S., Lummerich, Hersonissos, Greece.
14. Abdul-wahab, S., and Lea, V. (2008). A., Cereceda, P., Marzol, V., Corell, D.,
Reviewing fog water collection worldwide and Heerden, J.O., Reinhard, D., Gherezghiher, 46. Kotzen, B. (2014). Novel ideas for maximising
in Oman. Int. J. Environ. Studies 65, 487–500. T., Olivier, J., et al. (2012). Fog as a fresh-water de55w collection to aid plant establishment
resource: overview and perspectives. Ambio to combat desertification and restore
15. Guan, H., Sebben, M., and Bennett, J. (2014). 41, 221–234. degraded dry and arid lands. In EGU General
Radiative- and artificial-cooling enhanced Assembly Conference. Vienna, Austria.
dew collection in a coastal area of South 32. DeFelice, T.P., and Axisa, D. (2017). Modern
Australia. Urban Water J. 11, 175–184. and prospective technologies for weather 47. Muñoz-Garcı́a, M.A., Moreda, G.P., Raga-
modification activities: developing a Arroyo, M.P., and Marı́n-González, O. (2013).
16. Bathiany, S., Dakos, V., Scheffer, M., and framework for integrating autonomous Water harvesting for young trees using Peltier
Lenton, T.M. (2018). Climate models predict unmanned aircraft systems. Atmos. Res. 193, modules powered by photovoltaic solar
increasing temperature variability in poor 173–183. energy. Comput. Electron. Agr. 93, 60–67.
countries. Sci. Adv. 4, eaar5809.
33. Bruintjes, R.T. (1999). A review of cloud 48. Raman, A.P., Anoma, M.A., Zhu, L., Rephaeli,
17. Bao, W., Tang, X., Guo, X., Choi, S., Wang, C., seeding experiments to enhance E., and Fan, S. (2014). Passive radiative cooling
Gogotsi, Y., and Wang, G. (2018). Porous precipitation and some new prospects. Bull. below ambient air temperature under direct
cryo-dried MXene for efficient capacitive Am. Meteorol. Soc. 80, 805–820. sunlight. Nature 515, 540–544.
deionization. Joule 2, 778–787.
34. Wang, G.Q., Zhong, D.Y., Li, T.J., Wei, J.H., 49. Zhai, Y., Ma, Y., David, S.N., Zhao, D., Lou, R.,
18. Gude, V.G. (2016). Desalination and Huang, Y.F., Fu, X.D., Li, J.Y., and Zhang, Y. Tan, G., Yang, R., and Yin, X. (2017). Scalable-
sustainability—an appraisal and current (2016). Sky River: discovery, concept, and manufactured randomized glass-polymer
perspective. Water Res. 89, 87–106. implications for future research (in Chinese). hybrid metamaterial for daytime radiative
Sci. Sin. Tech. 46, 649–656. cooling. Science 355, 1062–1066.
19. Pinto, F.S., and Marques, R.C. (2017).
Desalination projects economic feasibility: a 35. Dai, X., Sun, N., Nielsen, S.O., Stogin, B.B., 50. Nebbia G. (1961). The problem of obtaining
standardization of cost determinants. Renew. Wang, J., Yang, S., and Wong, T.S. (2018). water from the air. In Proceedings of the
Sustain. Energy Rev. 78, 904–915. Hydrophilic directional slippery rough Conference on Solar and Aeolian Energy.
surfaces for water harvesting. Sci. Adv. 4,
20. Mezher, T., Fath, H., Abbas, Z., and Khaled, A. eaaq0919. 51. Khalil, B., Adamowski, J., Rojas, M., and Reilly,
(2011). Techno-economic assessment and K.. (2014). Towards an independent dew
environmental impacts of desalination 36. Gultepe, I., Tardif, R., Michaelides, S.C., water irrigation system for arid or insular
technologies. Desalination 266, 263–273. Cermak, J., Bott, A., Bendix, J., Müller, M.D., areas. In: American Society of Agricultural and
Pagowski, M., Hansen, B., and Ellrod, G. Biological Engineers International Meeting,
21. Kogan, B., and Trahtman, A. (2003). The (2007). Fog research: a review of past pp. 2827–2835.
moisture from the air as water resource in arid achievements and future perspectives. Pure
region: hopes, doubts and facts. J. Arid Appl. Geophys. 164, 1121–1159. 52. Bui, D.T., Chua, K.J., and Gordon, J.M. (2017).
Environ. 53, 231–240. Comment on ‘‘Water harvesting from air with
37. Batisha, A.F. (2015). Feasibility and metal-organic frameworks powered by
22. Vuollekoski, H., Vogt, M., Sinclair, V.A., sustainability of fog harvesting. Sustain. Water natural sunlight’’. Science 358, eaao0791.
Duplissy, J., Järvinen, H., Kyrö, E.M., Qual. Ecol. 6, 1–10.
Makkonen, R., Petaja, T., Prisle, N.L., 53. Subiantoro, A. (2017). Expander-based
Raisanen, P., et al. (2014). Estimates of global 38. Montecinos, S., Carvajal, D., Cereceda, P., atmospheric water harvesting in the tropics.
dew collection potential. Hydrol. Earth Syst. and Concha, M. (2018). Collection efficiency Asian J. Water Environ. Pollut. 14, 1–8.
Sci. Discussions 11, 9519–9549. of fog events. Atmos. Res. 209, 163–169.
54. Shelton, S.V., and Christiansen, P.J. (1999).
23. Beysens, D. (2016). Estimating dew yield 39. Abdul-Wahab, S.A., Hilal, A.H., Al-Najar, K.A., Analysis of a nozzle condensation drying
worldwide from a few meteo data. Atmos. and Al-Kalbani, M.S. (2007). Feasibility of fog cycle. Appl. Therm. Eng. 19, 831–845.
Res. 167, 146–155. water collection: a case study from Oman.
J. Water Supply Res. Technol. Aqua 56, 55. Hamed, A.M., Aly, A.A., and Zeidan, E.S.B.
24. Gido, B., Friedler, E., and Broday, D.M. (2016). 275–280. (2011). Application of solar energy for
Assessment of atmospheric moisture recovery of water from atmospheric air in
harvesting by direct cooling. Atmos. Res. 182, 40. Schemenauer, R.S., Cereceda, P., and Osses climatic zones of Saudi Arabia. Nat. Resour. 2,
156–162. P. (2003). The complementary aspects of 8–17.
projects to collect rain, fog and dew. In XIth
25. Alayli, Y., Hadji, N.E., and Leblond, J. (1987). A IRCSA Conference. 56. Mahal, S.K., and Alimin, A.J. (2017).
new process for the extraction of water from Advancement in the technology of solar
air. Desalination 67, 227–229. 41. Park, K.C., Chhatre, S.S., Srinivasan, S., powered liquid desiccant systems for fresh
Cohen, R.E., and Mckinley, G.H. (2013). water production from atmospheric humidity.
26. Malik, F.T., Clement, R.M., Gethin, D.T., Optimal design of permeable fiber network Int. Rev. Mech. Eng. 11, 191–199.
Krawszik, W., and Parker, A.R. (2014). Nature’s structures for fog harvesting. Langmuir 29,
moisture harvesters: a comparative review. 13269. 57. Edmund, A. (1938). Method for gaining water
Bioinspir. Biomim. 9, 031002. out of the atmosphere. US patent US2138689.
42. Shi, W., Anderson, M.J., Tulkoff, J.B.,
27. Stone, E.C. (1957). Dew as an ecological Kennedy, B.S., and Boreyko, J.B. (2018). Fog 58. Scrivani, A., and Bardi, U. (2008). A study of the
factor: I. a review of the literature. Ecology 38, harvesting with harps. Acs Appl. Mater. use of solar concentrating plants for the
407–413. Interfaces 10, 11979–11986. atmospheric water vapour extraction from

Joule 2, 1452–1475, August 15, 2018 1473

 ambient air in the Middle East and Northern 73. Milani, D., Qadir, A., Vassallo, A., Chiesa, M., 87. Park, K.C., Kim, P., Grinthal, A., He, N., Fox, D.,
Africa region. Desalination 220, 592–599. and Abbas, A. (2014). Experimentally Weaver, J.C., and Aizenberg, J. (2015).
validated model for atmospheric water Condensation on slippery asymmetric
59. Hamed, A.M. (2000). Absorption– generation using a solar assisted desiccant bumps. Nature 531, 78–82.
regeneration cycle for production of water dehumidification system. Energy Build. 77,
from air-theoretical approach. Renew. Energy 236–246. 88. Bergmair, D., Metz, S.J., de Lange, H.C., and
19, 625–635. van Steenhoven, A.A. (2015). A low pressure
74. Milani, D., Abbas, A., Vassallo, A., Chiesa, M., recirculated sweep stream for energy efficient
60. Kabeel, A.E., Hamed, A.M., and El-Agouz, and Bakri, D.A. (2011). Evaluation of using membrane facilitated humidity harvesting.
S.A. (2010). Cost analysis of different solar still thermoelectric coolers in a dehumidification Sep. Purif. Technol. 150, 112–118.
configurations. Energy 35, 2901–2908. system to generate freshwater from ambient
air. Chem. Eng. Sci. 66, 2491–2501. 89. Andraz, K., Varlec, J., Mazaj, M., Ristic, A.,
61. Muftah, A.F., Alghoul, M.A., Fudholi, A., Logar, N.Z., and Mali, G. (2017). Superior
Abdul-Majeed, M.M., and Sopian, K. (2014). 75. Joshi, V.P., Joshi, V.S., Kothari, H.A., Mahajan, performance of microporous
Factors affecting basin type solar still M.D., Chaudhari, M.B., and Sant, K.D. (2017). aluminophosphate with LTA topology in
productivity: a detailed review. Renew. Experimental investigations on a portable solar-energy storage and heat reallocation.
Sustain. Energy Rev. 32, 430–447. fresh water generator using a thermoelectric Adv. Energy Mater. 7, 1601815.
cooler. Energy Procedia 109, 161–166.
62. Elnaby, K.A. (2015). Water recovery from 90. Rieth, A.J., Yang, S., Wang, E.N., and Dincǎ,
atmospheric air using wick desiccant solar 76. Algarni, S., Saleel, C.A., and Mujeebu, M.A. M. (2017). Record atmospheric fresh water
still. Environ. Eng. Manag. J. 14, 2365–2372. (2018). Air-conditioning condensate recovery capture and heat transfer with a material
and applications—current developments and operating at the water uptake reversibility
63. Kumar, M., and Yadav, A. (2015). Experimental challenges ahead. Sustain. Cities Soc. 37, limit. ACS Cent. Sci. 3, 668–672.
investigation of design parameters of solar 263–274.
glass desiccant box type system for water 91. Furukawa, H., Gándara, F., Zhang, Y.B., Jiang,
production from atmospheric air. J. Renew. 77. William, G.E., Mohamed, M.H., and Fatouh, J., Queen, W.L., Hudson, M.R., and Yaghi,
Sustain. Energy 7, 033122. M. (2015). Desiccant system for water O.M. (2014). Water adsorption in porous
production from humid air using solar energy. metal-organic frameworks and related
64. Fathieh, F., Kalmutzki, M.J., Kapustin, E.A., Energy 90, 1707–1720. materials. J. Am. Chem. Soc. 136, 4369.
Waller, P.J., Yang, J., and Yaghi, O.M. (2018).
Practical water production from desert air. Sci. 78. Kim, S.I., Yoon, T.U., Kim, M.B., Lee, S.J., 92. Trapani, F., Polyzoidis, A., Loebbecke, S., and
Adv. 4, eaat3198. Hwang, Y.K., Chang, J.S., Kim, H.J., Lee, H.N., Piscopo, C.G. (2016). On the general water
Lee, U.H., and Bae, Y.S. (2015). Metal-organic harvesting capability of metal-organic
65. Liu, Y.F., and Wang, R.Z. (2003). Pore structure frameworks with high working capacities and frameworks under well-defined climatic
of new composite adsorbent cyclic hydrothermal stabilities for fresh water conditions. Micropor. Mesopor. Mater 230,
SiO2$xH2O$yCaCl2 with high uptake of water production. Chem. Eng. J. 286, 467–475. 20–24.
from air. Sci. China 46, 551–559.
93. Permyakova, A., Wang, S., Courbon, E.,
79. Sun, Y., Zhou, Z., Ashraful Alam, M., and
66. Kim, H., Rao, S.R., Kapustin, E.A., Zhao, L., Nouar, F., Heymans, N., D’Ans, P., Barrier, N.,
Bermel, P. (2017). Radiative sky cooling:
Yang, S., Yaghi, O.M., and Wang, E.N. (2018). Billemont, P., Weireld, G.D., Steunou, N.,
fundamental physics, materials, structures,
Adsorption-based atmospheric water et al. (2017). Design of salt–metal organic
and applications. Nanophotonics 6, 997–
harvesting device for arid climates. Nat. framework composites for seasonal heat
1015.
Commun. 9, 1191. storage applications. J. Mater. Chem. A 5,
80. Gentle, A.R., and Smith, G.B. (2015). A 12889–12898.
67. Kim, H., Yang, S., Rao, S.R., Narayanan, S., subambient open roof surface under the mid-
Kapustin, E.A., Furukawa, H., Umans, A.S., 94. Zheng, X., Ge, T.S., and Wang, R.Z. (2014).
summer sun. Adv. Sci. 2, 1500119.
Yaghi, O.M., and Wang, E.N. (2017). Water Recent progress on desiccant materials for
harvesting from air with metal-organic 81. Chen, Z., Zhu, L., Raman, A., and Fan, S. solid desiccant cooling systems. Energy 74,
frameworks powered by natural sunlight. (2016). Radiative cooling to deep sub-freezing 280–294.
Science 356, 430–434. temperatures through a 24-h day–night cycle. 95. Zheng, X., Ge, T.S., Jiang, Y., and Wang, R.Z.
Nat. Commun. 7, 13729. (2015). Experimental study on silica gel-LiCl
68. Wang, J.Y., Wang, R.Z., Wang, L.W., and Liu,
J.Y. (2017). A high efficient semi-open system composite desiccants for desiccant coated
82. Yu, X., Wang, R.Z., and Zhai, X.Q. (2011). Year
for fresh water production from atmosphere. heat exchanger. Int. J. Refrig. 51, 24–32.
round experimental study on a constant
Energy 138, 542–551. temperature and humidity air-conditioning 96. Grekova, A.D., Girnik, I.S., Nikulin, V.V.,
system driven by ground source heat pump. Tokarev, M.M., Gordeeva, L.G., and Aristov,
69. Wang, J.Y., Liu, J.Y., Wang, R.Z., and Wang, Energy 36, 1309–1318. Y.I. (2016). New composite sorbents of water
L.W. (2017). Experimental investigation on
and methanol ‘‘salt in anodic alumina’’:
two solar-driven sorption based devices to 83. Cottrill, A.L., Liu, A.T., Kunai, Y., Koman, V.B.,
evaluation for adsorption heat
extract fresh water from atmosphere. Appl. Kaplan, A., Mahajan, S.G., Liu, P., Toland,
transformation. Energy 106, 231–239.
Therm. Eng. 127, 1608–1616. A.R., and Strano, M. (2018). Ultra-high thermal
effusivity materials for resonant ambient 97. Krajnc, A., Bueken, B., De Vos, D., and Mali, G.
70. Wang, J.Y., Liu, J.Y., Wang, R.Z., and Wang, thermal energy harvesting. Nat. Commun. 9, (2017). Improved resolution and simplification
L.W. (2017). Experimental research of 664. of the spin-diffusion-based NMR method for
composite solid sorbents for fresh water the structural analysis of mixed-linker MOFs.
production driven by solar energy. Appl. 84. Han, G.G.D., Li, H., and Grossman, J.C. (2017). J. Magn. Reson. 279, 22–28.
Therm. Eng. 121, 941–950. Optically-controlled long-term storage and
release of thermal energy in phase-change 98. Ehrenmann, J., Henninger, S.K., and Janiak,
71. Jradi, M., Ghaddar, N., and Ghali, K. (2012). materials. Nat. Commun. 8, 1446. C. (2011). Water adsorption characteristics of
Optimized operation of a solar-driven MIL-101 for heat-transformation applications
thermoelectric dehumidification system for 85. Veerakumar, C., and Sreekumar, A. (2016). of MOFs. Eur. J. Inorg. Chem. 2011, 471–474.
fresh water production. Energy Power Eng. 6, Phase change material based cold thermal
878–891. energy storage: materials, techniques and 99. Wittmann, T., Siegel, R., Reimer, N., Milius,
applications—a review. Int. J. Refrig. 67, W., Stock, N., and Senker, J. (2015).
72. Jradi, M., Ghaddar, N., and Ghali, K. (2011). 271–289. Enhancing the water stability of Al-MIL-101-
Experimental and theoretical study of an NH2 via postsynthetic modification. Chem. A
integrated thermoelectric–photovoltaic 86. Wen, R., Xu, S., Ma, X., Lee, Y.C., and Yang, R. Eur. J. 21, 314–323.
system for air dehumidification and fresh (2017). Three-dimensional superhydrophobic
water production. Int. J. Energy Res. 36, nanowire networks for enhancing 100. Kalmutzki, M.J., Diercks, C.S., and Yaghi
963–974. condensation heat transfer. Joule 2, 1–11. Omar, M. (2018). Metal-organic frameworks

1474 Joule 2, 1452–1475, August 15, 2018

 for water harvesting from air. Adv. Mater. electrolyte membrane-based electrochemical 109. Motamartinez, M.T., Hallett, J.P., and Dowell,
https://doi.org/10.1002/adma.201704304. air dehumidification system. Appl. Energy N.M. (2017). Solvent selection and design for
208, 1174–1183. CO2 capture – how we might have been
101. Yu, N., Wang, R.Z., Wang, L.W., and Lu, Z.S. missing the point. Sustain. Energy Fuels 1,
(2014). Development and characterization of 105. Kohavi, A., and Dulberg, S. (2016). Planar 2078–2090.
silica gel-LiCl composite sorbents for thermal element for forming heat exchanger. US
energy storage. Chem. Eng. Sci. 111, 73–84. patent US20160010930A1. 110. Tu, Y.D., Wang, R.Z., Ge, T.S., and Zheng, X.
(2017). Comfortable, high-efficiency heat
102. Canivet, J., Bonnefoy, J., Daniel, C., Legrand, 106. Dieckmann, J.T., and Westphalen, D. (2014).
pump with desiccant-coated, water-sorbing
A., Coasne, B., and Farrusseng, D. (2014). Dehumidification. US patent US8640472.
heat exchangers. Sci. Rep. 7, 40437.
Structure-property relationships of water
adsorption in metal-organic frameworks. New 107. Kallenberger, P.A., and Fröba, M. (2018).
Water harvesting from air with a hygroscopic 111. Tu, Y.D., Wang, R.Z., Hua, L.J., Ge, T.S., and
J. Chem. 38, 3102–3111.
salt in a hydrogel–derived matrix. Commun. Cao, B.Y. (2017). Desiccant-coated water-
103. Bergmair, D., Metz, S.J., de Lange, H.C., and Chem. 1, 28. sorbing heat exchanger: weakly-coupled heat
van Steenhoven, A.A. (2014). System analysis and mass transfer. Int. J. Heat Mass Transf.
of membrane facilitated water generation 108. Zhao, D., Martini, C.E., Jiang, S., Ma, Y., Zhai, 113, 22–31.
from air humidity. Desalination 339, 26–33. Y., Tan, G., Yin, X., and Yang, R. (2017).
Development of a single-phase thermosiphon 112. Tu, Y.D., Wang, R.Z., and Ge, T.S. (2018). New
104. Qi, R., Li, D., and Zhang, L.Z. (2017). for cold collection and storage of radiative concept of desiccant-enhanced heat pump.
Performance investigation on polymeric cooling. Appl. Energy 205, 1260–1269. Energy Convers. Manag. 156, 568–574.

Joule 2, 1452–1475, August 15, 2018 1475

JOUL, Volume 2

Supplemental Information

Progress and Expectation

of Atmospheric Water Harvesting
Yaodong Tu, Ruzhu Wang, Yannan Zhang, and Jiayun Wang
Table S1. A summary of literature review on fog water collectors.
Operation Fog Elevation SWP
Reference Material Location year Method
days/year m kg/day/m2
2001 Raschel 65% 2001-
Nepal 122 3750 3.7 Exp.
Macquarrie [36] shadow net 2010
2014 1995-
/ Peru 210 800 11.8 Exp.
Fessehaye [37] 1999
2002
1999-
Olivier and Rautenbach Polypropylene yarn South Africa 184 1600 4.6 Exp.
2001
[1]
2011 Canary 2000-
/ 354 700 10 Exp.
Marzol et al. [2] Islands 2010
2002 Chile (Alto 1997-
Raschel mesh type 365 700 7.8 Exp.
Larrain [3] Patache) 2000
2014 Guatemala 2006-
/ 210 3300 6 Exp.
Fessehaye [37] (tojquia) 2010
2014 Chile
/ 2010 365 550 2 Exp.
Fessehaye [37] (padre Hurtado)
Table S2. A summary of literature review on passive radiative condensers.

SWP
Reference Material Location Method
kg/day/m2
Israel (Revivim desert, Eshtaol
1965 0.86-3.63
Polyethylene foil sub-humid hills, and Exp.
Gindel [4] kg/month/m2
Rehovot semi-arid plain)
Condensing foil
2003
(TiO2 and BaSO4 microspheres Grenoble (France) 0.17 (Max) Exp.
Beysens et al. [5]
embedded in polyethylene)
Condensing foil
2006
(TiO2 and BaSO4 microspheres Ajaccio Gulf (France) 0.25 Exp.
Muselli et al. [6]
embedded in a polyethylene sheet)
2007
Corrugated galvanized iron roofs Kothara (NW India) 0.09 Exp.
Sharan et al. [7]
2008 Polyethylene foil The Netherlands (a grassland Exp.
0.19
Jacobs et al. [8] Polyvinyl chloride area) &Cal.
Condensing foil Mountainous
2008 (TiO2 and BaSO4 micro spheres Island (Punaauia, Tahiti Island)
0.23 Exp.
Clus et al. [9] embedded in low density polyethylene Atoll (Tikehau, Tuamotu
with a food surfactant, 0.35 mm) Archipelago)
Condensing foil 0.14 (Zadar, after
2009 (TiO2 and BaSO4 microspheres Zadar scraping)
Exp.
Muselli et al. [10] embedded in a matrix of low-density Komiža (Vis Island) 0.11 (Komiža, after
polyethylene ) scraping)
2011 White hydrophilic polyethylene foil
0.10 (WSF)
Maestre-Valero et al. (WSF) Southern Spain Exp.
0.13 (BF)
[11] Black polyethylene foil (BF)
Active substrate (foil)
2012 Mirleft, Id Ouasskssou 0.11 (Mirleft)
(TiO2 and BaSO4 embedded in a film of Exp.
Lekouch et al. [12] (Southwest Morocco) 0.14 (Id Ouasskssou)
low-density polyethylene)
Table S3. A summary of literature review on sorption-based AWGs.

Fan/ SWP
Reference Desiccant Sorber Condenser Working conditions Method
Pump kg/day/m2
1997
35% CaCl2 Solution flowrate:
Abualhamayel and Greenhouse Glass cover Pump 1.92 Exp.
solution 5 kg/hr/m2
Gandhidasan [13]
2001 Greenhouse Fan Solution concentration
CaCl2 solution Cold plate 1.50 Exp.
Gad et al. [14] Cloth bed (ad) 41.7%→31%
30% CaCl2 Greenhouse Fan 1.10 (Exp.) Cal.
2006 Kabeel [15] Glass cover Arab desert climate
solution Sandy bed (ad) 1.20 (Cal.) &Exp.
2007 30% CaCl2 Cloth bed glass pyramid
Glass cover NO 2.50 Exp.
Kabeel [16] solution Saw wood bed Multi-shelf bed
2007 CaCl2/ Desorption temp.
Greenhouse Glass cover NO 1.20 Exp.
Ji et al. [17] MCM-41 >80 oC
2015 Fan 34 oC (day)
CaCl2 solution Greenhouse Glass cover 1.40 Exp.
Kabeel [18] (ad) 25 oC/55%RH (night)
2015 30% CaCl2 Cloth bed glass pyramid 2.32 (cloth)
Glass cover NO Exp.
William et al. [19] solution Sandy bed Multi-shelf bed 1.23 (sand)
2016 37% CaCl2/ Fan Material temperature in night:
Greenhouse Glass cover 1.00 Exp.
Kumar [20] floral foam (ad) 78.8 oC
2017 Improved Fan
MOF-801 Back plate >20%RH 1.20 Exp.
Kim et al. [21] greenhouse (ad)
Table S4. A summary of literature review on solar still AWGs.
Dayligh
Fan SWP
Reference Sorber Condenser location t Method
/Pump kg/day/m2
hours
2003
Single-slope Glass cover pump Egypt 6-18 5.8 Exp.
Fath et al. [22]
2007
Single-slope Glass cover No Pakistan 10-19 2.3 Exp.
Samee et al. [23]
2009
Single-slope Glass cover pump India 9-17 1.3 Exp.
Kumar and Tiwari [24]
2009 With solar
Glass cover pump India 9-17 4.6 Exp.
Kumar and Tiwari [25] collector
2005 With solar
Glass cover pump Jordan 8-18 3.1 Exp.
Badran and Tahaineh [26] collector
2007 With solar
Glass cover pump Egypt 9-19 3.8 Exp.
Abdel-Rehim and Lasheen [27] concentrator
2008 With sun
Glass cover No Jordan 7-18 0.96 Exp.
Abdallah and Badran [28] tracking
Table S5. A summary of literature review on integrated AWGs.

Fan SWP Metho

Reference System Condenser Working conditions COP
/Pump kg/day/m2 d
LiBr-H2O Solar collecting
1989 absorption cooling Fan, temperature: 140 oC
Open absorber 5.8 1.63 Cal.
ALY [29] machine (open and pump Cooling water
closed) temperature: 30 oC
2011 Solar-powered Cooling coil Beirut humid climate
Fan,
Audah ea liquid desiccant submerged in (room air of 25 oC and 2.3 / Cal.
pump
al. [30] system cold sea water 50% RH)
2012
Licina and large dedicated air Pre-cooling and 27.5 oC and 84% RH,
Fan 1.3 / Cal.
Sekhar handling units cooling coil 10000 m3/h
[31]
2014 Solar PVT powered Moderate climates (like
Air cooled heat Cal. &
Milani DOAS air- Fan Sydney, average 14.2 oC 0.6 /
exchanger Exp.
[32] conditioner (10 m2) and 64.8% RH)
2015
HVAC system Inlet air:
Magrini et Cooling coil Fan 3.1 / Cal.
chillers 20°C and 60% RH
al. [33]
2016 Fresh air: 0.66
Solar powered LiBr- Fan,
Ibrahim et Fan coil unit 30~45oC and 30~70% 3.8 (45°C/50% Cal.
H2O chiller pump
al. [34] RH RH)
Air-cooling 0.878
2017
LiCl-H2O condenser Fan, Inlet air: (Treg=70 oC)
Ahmed et 0.96 Cal.
absorption chiller (cooling stream: pump 30 oC, 74% RH 0.898
al. [35]
seawater) (Treg=90 oC)
Fig.S1 Condensation enhancement surfaces. (a~c) a slippery asymmetric bump. Reproduced from Park et al.38 with permission.
(d~f) a hydrophilic slippery rough surface. Reproduced from Dai et al.39 with permission.
Reference
[1] Olivier, J., & Rautenbach, C. J. D. (2002). The implementation of fog water collection systems in south
africa. Atmospheric Research, 64(1), 227-238.
[2] Marzol, M. V., Sánchez, J. L., & Yanes, A. (2011). Meteorological patterns and fog water collection in
morocco and the canary islands. Erdkunde,65(3), 291-303.
[3] Larrain, H., Velásquez, F., Cereceda, P., Espejo, R., Pinto, R., & Osses, P., et al. (2002). Fog measurements
at the site “falda verde” north of chañaral compared with other fog stations of chile. Atmospheric Research, 64(1–
4), 273-284.
[4] Gindel, I. (1965). Irrigation of plants with atmospheric water within the desert. Nature, 207(5002), 1173-1175.
[5] Beysens, D., Milimouk, I., Nikolayev, V., Muselli, M., & Marcillat, J. (2003). Using radiative cooling to
condense atmospheric vapor: a study to improve water yield. Journal of Hydrology, 276(1), 1-11.
[6] Muselli, M., Beysens, D., & Milimouk, I. (2006). A comparative study of two large radiative dew water
condensers. Journal of Arid Environments,64(1), 54-76.
[7] Sharan, G., Beysens, D., & Milimouk-Melnytchouk, I. (2007). A study of dew water yields on galvanized iron
roof in kothara (north-west india). Journal of Arid Environments, 69(2), 259-269.
[8] Jacobs, A. F. G., Heusinkveld, B. G., & Berkowicz, S. M. (2008). Passive dew collection in a grassland area,
the netherlands. Atmospheric Research, 87(3–4), 377-385.
[9] Clus, O., Ortega, P., Muselli, M., Milimouk, I., & Beysens, D. (2008). Study of dew water collection in humid
tropical islands. Journal of Hydrology, 361(1–2), 159-171.
[10] Muselli, M., Beysens, D., Mileta, M., & Milimouk, I. (2009). Dew and rain water collection in the dalmatian
coast, croatia. Atmospheric Research,92(4), 455-463.
[11] Maestre-Valero, J. F., Martínez-Alvarez, V., Baille, A., Martí n-Górriz, B., & Gallego-Elvira, B. (2011).
Comparative analysis of two polyethylene foil materials for dew harvesting in a semi-arid climate. Journal of
Hydrology,410(1–2), 84-91.
[12] Lekouch, I., Lekouch, K., Muselli, M., Mongruel, A., Kabbachi, B., & Beysens, D. (2012). Rooftop dew,
fog and rain collection in southwest morocco and predictive dew modeling using neural networks. Journal of
Hydrology, 448-449(27), 60-72.
[13] Abualhamayel HI, Gandhidasan P. (1997). A method of obtaining fresh water from the humid atmosphere.
Desalination, 113(1), 51-63.
[14] Gad, H. E., Hamed, A. M., & El-Sharkawy, I. I. (2001). Application of a solar desiccant/collector system for
water recovery from atmospheric air. Renewable Energy, 22(4), 541-556.
[15] Kabeel, A. E. (2006). Application of sandy bed solar collector system for water extraction from
air. International Journal of Energy Research,30(6), 381-394.
[16] Kabeel, A. E. (2007). Water production from air using multi-shelves solar glass pyramid system. Renewable
Energy, 32(1), 157-172.
[17] Ji, J. G., Wang, R. Z., & Li, L. X. (2007). New composite adsorbent for solar-driven fresh water production
from the atmosphere. Desalination,212(1), 176-182.
[18] Kabeel, A. E. (2015). Water recovery from atmospheric air using wick desiccant solar still. Environmental
Engineering & Management Journal,14(10).
[19] William, G. E., Mohamed, M. H., & Fatouh, M. (2015). Desiccant system for water production from humid
air using solar energy. Energy, 90, 1707-1720.
[20] Kumar, M., & Yadav, A. (2016). Solar-driven technology for freshwater production from atmospheric air by
using the composite desiccant material “cacl 2 /floral foam”. Environment Development & Sustainability,18(4),
1151-1165.
[21] Kim H, Yang S, Rao SR, Narayanan S, Kapustin EA, Furukawa H, et al. (2017). Water harvesting from air
with metal-organic frameworks powered by natural sunlight. Science, 356(6336), 430-434.
[22] Fath HES, El-Samanoudy M, Fahmy K, Hassabou A. (2003). Thermal-economic analysis and comparison
between pyramid-shaped and single-slope solar still configurations. Desalination, 159(1), 69-79.
[23] Samee, M. A., Mirza, U. K., Majeed, T., & Ahmad, N. (2007). Design and performance of a simple single
basin solar still. Renewable & Sustainable Energy Reviews, 11(3), 543-549.
[24] Kumar, S., & Tiwari, G. N. (2009). Life cycle cost analysis of single slope hybrid (pv/t) active solar
still. Applied Energy, 86(10), 1995-2004.
[25] Kumar, S., & Tiwari, G. N. (2009). Thermal modelling, validation and exergetic analysis of a hybrid
Photovoltaic/Thermal (PV/T) active solar still. International Journal of Exergy, 6(6), 567-591.
[26] Badran, O. O., & Al-Tahaineh, H. A. (2005). The effect of coupling a flat-plate collector on the solar still
productivity. Desalination, 183(1–3), 137-142.
[27] Abdel-Rehim, Z. S., & Lasheen, A. (2007). Experimental and theoretical study of a solar desalination system
located in cairo, egypt. Desalination,217(1), 52-64.
[28] Abdallah, S., & Badran, O. O. (2008). Sun tracking system for productivity enhancement of solar
still. Desalination, 220(1-3), 669-676.
[29] Aly, S. E. (1989). Humidity water vapour collection by absorption cooling machines. Desalination, 73(1-3),
139-165.
[30] Audah, N., Ghaddar, N., & Ghali, K. (2011). Optimized solar-powered liquid desiccant system to supply
building fresh water and cooling needs. Applied Energy, 88(11), 3726-3736.
[31] Licina, D., & Sekhar, C. (2012). Energy and water conservation from air handling unit condensate in hot and
humid climates. Energy & Buildings,45(1), 257-263.
[32] Milani, D., Qadir, A., Vassallo, A., Chiesa, M., & Abbas, A. (2014). Experimentally validated model for
atmospheric water generation using a solar assisted desiccant dehumidification system. Energy & Buildings,77(7),
236-246.
[33] Magrini, A., Cattani, L., Cartesegna, M., & Magnani, L. (2015). Integrated systems for air conditioning and
production of drinking water – preliminary considerations. Energy Procedia, 75, 1659-1665..
[34] Ibrahim, N. I., Al-Sulaiman, F. A., & Saidur, R. (2016). Performance assessment of water production from
solar cooling system in humid climate. Energy Conversion & Management, 127, 647-655.
[35] Ahmed, M. A., Gandhidasan, P., Zubair, S. M., & Bahaidarah, H. M. (2017). Thermodynamic analysis of an
innovative liquid desiccant air conditioning system to supply potable water. Energy Conversion &
Management, 148, 161-173.
[36] Macquarrie KIA,Pokhrel A,Shrestha Y,Osses P,Schemenauer R.S,Vitez F, et al. (2001). Results from a high
elevation fog-water supply project in Nepal. In: Proceedings of the 2nd international conference on fog and fog
collection, 227–9.
[37] Fessehaye, M., Abdul-Wahab, S. A., Savage, M. J., Kohler, T., Gherezghiher, T., & Hurni, H. (2014). Fog-
water collection for community use. Renewable & Sustainable Energy Reviews, 29(7), 52-62.
[38] Park, K. C., Kim, P., Grinthal, A., He, N., Fox, D., & Weaver, J. C., et al. (2015). Condensation on slippery
asymmetric bumps. Nature, 531(7592), 78-82.
[39] Dai, X., Sun, N., Nielsen, S. O., Stogin, B. B., Wang, J., & Yang, S., et al. (2018). Hydrophilic directional
slippery rough surfaces for water harvesting. Sci Adv, 4(3).