A Versatile Platform of Poly(acrylic acid) Cryogel for Highly Efficient Photothermal Water Evaporation

Photothermal water evaporation, benefiting from solar energy, is one hot topic concerning with seawater desalination, sewage treatment and water recovery. Commonly, photothermal evaporation system consists of photothermal material supported by certain matrix with fundamental necessities. Poly(acrylic acid) (PAAc) cryogels, prepared at the temperature below the freezing point of polymerization mixture, are hierarchically macro-porous and highly hydrophilic with rapid water uptake, speedy water transport and low thermal conductivity. As well, PAAc has the acid-doping ability for polyaniline and polypyrrole (PPy), which faciliates the well incorporation photothermal materials in cryogel matrix. Thus, PAAc cryogel is an excellent platform for phtothermal water evaporation. In this research, we prepared PAAc cryogels, incorporated different photothermal materials and constructed the modular water evaporation systems. The photothermal evaporation rate of pure water with PAAc-PPy cryogel system reached as high as 1.819 kg∙m -2 ∙h -1 while its apparent efficiency exceeded surprisingly 100 %. These modular systems exhibited excellent long-term and recyclable utility. As for the photothermal evaporation of salty water, the precipitated salt solid could be spontaneously dissolved during the light turnoff due to the quick water transport, maintaining the high water-to-vapor conversion. This study represents an ideal matrix platform, PAAc cryogel, for the photothermal evaporation and opens up one new route to improve photothermal evaporation technique for water purification and seawater desalination.


Introduction
The global area covered by water is as high as 70.8 %, but freshwater resource is extremely limited and unevenly located. Of the total water resource, 99.7 % is salty and polluted water, being unable to be consumed directly 1,2 . Therefore, the techniques and materials for environment-friendly and efficient wastewater purification and seawater desalination are urgently needed. Solar energy is a green, inexhaustible and recyclable resource. With the assistance of photothermal materials, solar energy is converted into heat for water evaporation, offering a green and promising route in water treatment [3][4][5] . To attain highly efficient photothermal evaporation performance, there still exist many challenges to meet, including the photothermal material with excellent photo-physical properties and the evaporation system with high mass transfer ability 6 . Photothermal evaporation system is commonly composed of two components: photothermal material and supporting matrix. Various photothermal materials such as carbon materials, metallic nanoparticles, semiconductors and conjugated polymers have been widely studied 7,8 . They should have the full-spectrum light absorbance and high light-to-heat conversion effieicency, but their actual performance varies from each other, being dependent on the mechanism of photothermal conversion and light absorption capacity 9,10 . Among them, carbon materials such as carbon nanotube 11 , graphene 12,13 , carbon black 14 and carbon aerogels as well as conjugated polymers like polypyrrole (PPy) and polyaniline (PANI) [15][16][17] have broadband absorption spectrum, high photothermal stability and low density, making them as ideal candidates. Additionally, many carbon materials and conjugated polymers are of low cost and can be well incorporated into polymeric matrices to produce photothermal evaporation systems through various routes including in-situ polymerization 18 . Polymeric matrices not only provide the support for photothermal materials but also take the responsibilities of light, heat and mass transfer in the evaporation systems. Therefore, one ideal polymeric matrix should have some substantial and structural requirements to eliminate the light loss from the sunlight reflection and transmittance, disregard the heat loss into the environmental body, uptake easily the surrounding water and deliver water rapidly inside the system. For examples, the rough surface weakens light reflection and the thermal insulation eliminates heat loss while the hydrophilicity and micro-channel facilitate the uptake and transport of water through capillary action. Therefore, porous and hydrophilic matrixes have been extensively utilized in photothermal evaporation 19,20 and selffloating devices have been achieved due to low density [21][22][23] . So far, photothermal evaporation performance has been improved significantly through the joint efforts of researchers [23][24][25][26][27] . Xiao et al. prepared an evaporation system which exhibits a double-layer cooperative effect with efficient light-absorbing and water transportation, leading to a high evaporation rate of 1.37 kg•m -2 •h -1 with a photothermal conversion efficiency of 88.7% under one-sun illumination 28 . Chen at al. prepared a dual-functional material with single component by a simply carbonizing biowaste platanus fruit, which can achieve a high evaporation rate up to 2.00 kg•m -2 •h -1 under one-sun irradiation by 3D spherical evaporator 29 . Ma et al. produced a device made from paulownia wood covalently bonded with MXene on the top, achieving an evaporation rate of 1.465 kg•m -2 •h -1 under one-sun irradiation with 96 % conversion efficiency 30 . More recently, the photothermal desalination research gradually concentrates on the structural modulation of photothermal systems, including the buildup of hierarchical porosity 31 , the vertical alignment of micro-channel 32 , the two-layer arrangement of porous substrate 33,34 and the mushroom-like device 35 . Since most of the current systems are of two components, photothermal materials should be uniformly dispersed insides the polymeric matrices to exert photothermal activity as efficiently as possible and the integration should be durable to avoid the escape of photothermal materials during the long-term application. However, little concerns have been devoted into this important issue. As for the photothermal desalination, lots of salt solid accumulate, which clogs the porthole of water transport and weakens light absorption. Thus, Kou et al. developed low-cost and washable photothermal fabrics by dyeing traditional cotton fabrics with carbon nanotube based ink 36 . Cryogel, one newly developed macroporous material, has been introduced as the matrix of photothermal evaporation systems 37 . Compared with common (hydro)gels, cryogels are synthesized through cryo-polymerization of monomers and cryo-crosslinkage of polymeric precursors at the temperature below the freezing point of reaction mixture. They have interconnected macropores, high pore volume fraction and sufficient mechanical strength. As well, they can be hydrophilic or hydrophobic varying with the monomeric and polymeric precursors while their porosity morphology can be modulated through different routes 38 . The most important is that cryogels take up the solvent very rapidly and hold the absorbed solvent firmly. Therefore, cryogels have been used as the matrix to incorporate various functional materials 39 . In our previous research 40 , polyacrylamide cryogel was introduced as the matrix for mushroom-like photothermal evaporation device, whose evaporation rate and efficiency reached 1.63 kg•m -2 •h -1 and 89.2 %, respectively. In this work, poly(acrylic acid) (PAAc) cryogel was chosen as the matrix to incorporate polypyrrole (PPy) and polyaninile (PANI) since those conjugated polymers could be acid-doped by PAAc, facilitating their uniform distribution in the matrix. As well, graphene oxide was also integrated during the preparation of PAAc cryogel to testify the versatility of PAAc cryogel as the matrix, considering the possible polar interaction and hydrogen bonding. Evaporation system was designed to have the composite cryogel sheet on the top of mushroom-shaped PAAc cryogel substrate, being able to assemble or disassemble upon desired. Due to high sunlight absorbance of those photothermal materials along with sufficient water transport and heat insulation of the substrate, the evaporation system presented admirable performance of solar-to-heat water evaporation with the rate of pure water as high as 1.819 kg•m -2 •h -1 under one-sun irradiation and the high durability even for salty water evaporation within 6 days.

Materials
Graphite power was purchased from Beijing Jin-Long Technology Ltd. Graphene oxide (GO) was prepared from graphite power according to Hummers' method 41 . Acrylic acid (AAc), pyrrole (Py) and aniline (ANI) were purchased from Aladdin as AR reagent and used after distillation. (NH 4 ) 2 S 2 O 8 (APS, AR) purchased from Sinophamn Chemical Reagent was used after recrystallization. N, N′-Methylenebisacrylamide (MBAm), L-ascorbic acid (Vc), FeCl 3 and other chemicals were of AR grade, purchased from Sinopharm Chemical Reagent and used as received. Distilled water was used in all experiments.

Synthesis of PAAc cryogels and hydrogels without photothermal materials
PAAc cryogel was synthesized according to literature 42 with modification. AAc (20 mmol), MBAm (1 mmol) and Vc (0.24 mmol) were dissolved in distilled water (9 mL) while APS aqueous solution (0.2 mmol/mL, 1.0 mL) was prepared separately. After both aqueous solutions were pre-cooled in an ice bath for 10 min, they were mixed homogeneously. Then, the mixture was instantly transferred into a petri dish with 6 cm diameter and kept at -18 o C for 24 h to complete cryopolymerization. The obtained PAAc cryogel was immersed in distilled water for 48 h with exchanging distilled water every 6 h to remove the unreacted monomer and un-crosslinked polymer. Afterwards, the dried cryogel sheet was obtained through lyophilization at -44 o C for 48 h. Based on the left mass of cryogel sheet, the monomer conversion and gel fraction were estimated. The preparation of PAAc cryogel substrate was carried out through the same procedure as above in one mushroomshaped mould ( Figure S1a). One cryogel sheet and one mushroom-shaped cryogel were assembled within polystyrene (PSt) foam ( Figure S1b), resulting in the photothermal evaporation system. With the same polymerization recipe and vessels, PAAc hydrogel substrate was prepared through common polymerization at room temperature.

Synthesis of PAAc composite cryogels and hydrogels with photothermal materials
As for the preparation of PAAc composite cryogels with PPy or PANI (PAAc-PPy or PAAc-PANI cryogel), a dried sheet of PAAc Please do not adjust margins Please do not adjust margins cryogel (6 cm in diameter) was kept in 1 M HCl solution (140 mL) containing aniline or pyrrole (0.25 g) at 0 o C for 60 min for the monomer absorption. Afterwards, APS (0.785 g for aniline) or FeCl 3 (3.0 g for pyrrole) in HCl solution (10 mL) was added under magnetic stirring. The mixture of oxidative coupling polymerization stood at 5 o C for 6 h. Finally, the composite cryogels were washed with ethanol/water mixture (50:50 by volume) for several times and lyophilized. The preparation of composite cryogel with graphene oxide (PAAc-GO cryogel) was quite similar to that of pure PAAc cryogel, except that AAc and MBAm were beforehand dissolved in GO aqueous dispersion (2 mg/mL, 9 mL). The obtained PAAc-GO cryogel was kept in a Vc solution (3 mg/L, 100 mL) at 60 o C for 12 h to convert graphene oxide into reduced GO (rGO) 43,44 . Then, PAAc-rGO cryogel was washed with distilled water and lyophilized. The colour change from brown to black indicated the reduction of GO ( Figure S2)

Swelling behaviour and porosity fraction of cryogels/hydrogels
One dried cryogel sheet was soaked in 100 mL distilled water. At a regular interval, the sample was taken out, the surface water was removed with filter paper and the swollen cryogel was weighted. The above step was repeated until the mass no longer changed. The swelling degree (SD, g/g) is obtained according to Eq. 1.
where m s and m 0 is the mass of swollen and dried cryogel, respectively. The apparent density () of dried cryogel was determined by its mass and dimensional size. The diameter and thickness of cryogel sheet are shown in Figure S1c and S1d. The actual density (  ) of cryogel substance excluding the pore was calculated according to Eq. 2. The related density values are collected in Table S1. Thus, the porosity (, vol%) of cryogel was determined according to Eq. 3. 0 = 0,PAAc × PAAc + 0,tm × tm Eq.2 Eq.3 where w PAAc and w tm is the mass fraction of PAAc and photothermal material in composite cryogel, respectively. The parameter of ρ 0,PAAc and ρ 0,tm represents the density of pure PAAc and photothermal material, respectively. The density value is 1.200, 0.967, 1.022 and 2.100 g/cm 3 for PAAc, PPy, PANI and graphene, respectively.

Photothermal evaporation with different cryogels
The measurement setup of photothermal evaporation is demonstrated in Scheme S1. The experiment was performed at room temperature (25 o C) and humidity (45-55 %) using a solar simulator (CEL-HXF300/CEL-HXUV 300) equipped with an AM 1.5G filter and the incident light was adjusted to be 1 kW/m 2 (one-sun irradiation) with an optical power meter (PL-MW2000). In each test, after one mushroom-shaped PAAc cryogel was inserted through PSt foam button, one composite cryogel sheet (cut to the diameter of 3.0 cm) was kept in touch with the top of mushroom-shaped cryogel and entrapped inside PSt foam. Then, the integrated system was placed across a beaker filled with water, with the tip of mushroom-shaped cryogel contacting with water ( Figure S1b). The temperature of evaporation system was recorded with an infrared camera (Fluke TiS40). The mass loss was measured on a high accuracy balance (FA2204, 0.1 mg in accuracy) and a computer installed with a software (ShuDaXia) was connected to the balance to record the mass data in real time.

Characterizations
The chemical structure of cryogel was analysed by Fourier transform infrared spectroscopy (FT-IR, Nicolet 6700, Thermo Fisher, USA) in KBr pellets with the wavenumber range of 4000-500 cm -1 , scanning step of 4 cm −1 and 16 scanning times. Raman spectra of PAAc-GO/rGO cryogels were collected by laser confocal scanning Raman microscope (LABRAM-HR, France) with a laser source of 532 nm. Thermogravimetric analysis (TGA, Q5000IR, USA) was carried out from 25 to 800 o C with a heating rate 10 o C/min. Solar reflectance and transmittance measurement from 250 to 2500 nm was performed through UV-vis-NIR spectrophotometer (SOLID3700, USA) equipped with an integrating sphere, where the absorption efficiency (A) was calculated by A = (1 -R -T) (R and T is the reflectance and transmission efficiency, respectively). Tungsten filament scanning electron microscopy (TFSEM, EVO18, Germany) was used to study the morphology of dried cryogel and hydrogel. The porosity, channel size and pore area of cryogel were evaluated with a mercury intrusion porosimeter (AutoPore IV 9500 2.03.01, USA). To confirm the distribution of PANI and PPy in PAAc cryogel, an energy disperse X-ray spectroscopy (EDX, GeminiSEM 500, Germany) coupled to SEM was used. Thermal diffusion coefficiency was determined by a laser thermal conductivity meter (LFA-467, Germany). Specific heat capacity was characterized through differential scanning calorimetry (DSC, Q2000, USA). The contents of mental ions in collected water after photothermal evaporation were measured by inductively coupled plasma-mass spectrometry (ICP-MS PGS-2, Zeiss, USA).

Preparation of different cryogels
Pure PAAc cryogels were prepared through redox-initiated cryopolymerization with MBAm as the crosslinking monomer at -18 o C (Scheme 1). Based on the mass of the obtained cryogel, monomer conversion was nearly complete and gel fraction exceeded 95 %. FTIR, Raman and TGA results confirmed the chemical composition of all cryogels, referring to Figure S3 and S4. The carboxylic group of PAAc has not only strong acid-base interaction with Py and ANI units in monomer and polymer, but also sufficient hydrogen bonding with GO having oxygencontaining groups, which facilitated the uniform distribution and tight integration of photothermal materials insides PAAc cryogel.

Porous morphology of cryogels and the distribution of photothermal component
The real density of individual component in composite cryogel (  ) and the apparent density of cryogel/hydrogel ( app ) are listed in Table S1. According to Eq. 3, the porosity in volume fraction () is obtained and also summarized in Table S1. All the cryogels have low apparent density (0.136 -0.171 g/cm 3 ) and high porosity (86 -89 vol%), which effectively helps water transport and lessens heat loss. On the contrary, PAAc hydrogel has higher density (0.360 g/cm 3 ) and lower porosity fraction (70 vol%), which limits rapid water transport inside PAAc hydrogel. As shown in Figure S5, the porosity of the cryogel and hydrogel tested by the mercury porosimeter is consistent with the result by density calculation. The total area per unit mass of cryogel is more than ten times that of hydrogel. TFSEM was used to observe the pore morphology of different cryogels with regards to PAAc hydrogel and their images are shown in Figure 1. From Figure 1a, open macro-pores are not obviously observed for PAAc hydrogel and some bulb-like pores are present in the inset with high magnification. As suggested by Figure 1b-f, all the cryogels contrarily exhibit the macroporous morphology and the pores are open and interconnected, which is important for easy water transport inside cryogels. As for pure PAAc cryogel, the macro-pore is much and the pore-wall is thin. After the polymerization of ANI or Py insides PAAc cryogel, the pore-wall of both PAAc-ANI and PAAc-PPy cryogel becomes much thicker and the macro-pore seems less as seen in Figure 1c and 1d, suggesting that the formed PANI and PPy are tightly and uniformly attached onto the pore-wall. SEM coupled with EDX was used to further confirm the uniform distribution of PANI and PPy inside the cryogels. As shown in Figure S6 and S7 are the element mapping images of PAAc-ANI and PAAc-PPy cryogel, respectively. The images of C, O and N elements reveal the same pattern from vertical and lateral viewing. As indicated by Figure 1e and 1f, PAAc-GO and PAAc-rGO cryogels display lamellar-like macro-porous morphology, induced by the presence of graphene nano-sheets. Compared other cryogels, their macro-pores are much larger and porewalls are loosely piled up. The uniform distribution of photothermal materials inside the cryogels is indirectly supported with FTIR and TGA analysis. As indicated by the local FTIR spectra ( Figure S8), the signal of C=O stretching vibration for the composite cryogels displays the redshift to different extent, which follows the sequence of PAAc-PPy > PAAc-PANI > PAAc-GO > PAAc-rGO. This red-shift is caused by hydrogen-bonding between the carboxylic group of PAAc and the polar groups of photothermal components. As well, the signal of C-O stretching vibration of COOH also identify the above interaction. This signal is present at ~1252 cm -1 in FTIR spectra of PAAc, PAAc-GO and PAAc-rGO cryogels, with distinguishable red-shift for PAAc-PANI cryogel (1242 cm -1 ) due to acidic doping. This signal is absent for PAAc-PPy cryogel, suggesting the acid doping of PPy by PAAc is much extensive. Such a strong interaction also caused the difference of TGA thermogram between PAAc-PPy cryogel and others ( Figure  S4b).

Light absorption, thermal insulation and water uptake of different cryogels
While the -conjugated structure of photothermal materials facilitates the sunlight absorbance over widespread spectrum 10 , the photothermal systems should eliminate the light transmittance and reflection. Thus, UV-vis-NIR reflectance and transmittance spectroscopy was used to evaluate sunlight absorbance, as demonstrated in Figure 2. The transmittance and reflectance of PAAc cryogel in the wavelength range of 300-1400 nm are very high up to 17 and 76 %, respectively. As shown in Figure 2a, after introducing photothermal materials, the transmittance of PAAc-GO cryogel drops below 4 % and others are nearly zero in the whole spectrum. Meanwhile, the reflectance of PAAc composite cryogels (Figure 2b) decrease obviously compared with PAAc cryogel. The reflectance decrease for PAAc-GO cryogel is the weakest with the reflectance above 20 %, due to its lamellar morphology. After GO is reduced, the reflectance of PAAc-rGO cryogel falls to about 17 % in visible/near-infrared region. Both PAAc-PPy and PAAc-PANI cryogel has the reflectance less than 5 %. Therefore, compared with the sunlight absorption by PAAc cryogel (less than 60 %, especially as low as 8 % in visible region) in Figure 2c, composite cryogels with photothermal materials have significant increase of sunlight absorption, especially for PAAc-PPy and PAAc-PANI cryogels. It should be mentioned that the sunlight absorbance of PAAc-PPy cryogel is almost 100 % in the whole spectrum and the light intensity in visible region is much high as indicated by solar spectral radiance. Based on the intensity distribution of sunlight over wavelength, the overall efficiency of sunlight absorption ( abs ) by different cryogels are calculated in the range of 200~2000 nm according to Eq. 4 and the results are listed in Table S2.
where   and   is the absorbance and intensity at each wavelength (), respectively. The sunlight absorption absolutely increases the cryogel temperature due to solar-to-heat conversion. As shown in Figure S9 are IR camera images of different dry cryogels after one-sun irradiation for 20 min.  Figure 3 offers the temperature outlook during steady photothermal evaporation of pure water in different cases after one-sun irradiation for 20 min. Since there exists the temperature in-homogeneity over the cryogel surface under irradiation as indicated in Figure 3, the average temperature over cryogel sheet (T ave ) under one-sun irradiation has been calculated. The temperature difference between cryogel sheet and water reservoir is taken as (T ave -T room ). All the temperature data are summarized in Table S2.  (Figure 3b), being close to that of dried PAAc cryogel. On the other hand, with the incorporation of photothermal materials, the surface temperature of cryogel sheet during photothermal evaporation exceed 43 o C for all composite cryogels and the elevated temperatures are much lower than those for dried cryogels, confirming that the light energy absorbed by photothermal materials is mostly consumed for water evaporation. It should be noted that the surface temperature of photothermal system plays two opposite influences on water evaporation. One is the less evaporation enthalpy of water at higher temperature, which prompts the water evaporation. Another is the less heat loss from cryogels to environment at lower temperature. Photothermal evaporation system should require low thermal conductivity to prevent heat transfer to the surrounding environment and water reservoir. The thermal conductivity is determined to be 0.177, 0.056, 0.049, 0.043 and 0.038 W•m -1 •K -1 for PAAc-PPy, PAAc-PANI, PAAc-GO, PAAc-rGO and PAAc cryogel, respectively. Those values are much lower than that of water (0.592 W•m -1 •K -1 at standard state) and comparable to that of PS foam (~0.043 W•m -1 •K -1 ). The excellent thermal insulation of all cryogels can effectively minimize heat loss. Considering a little higher thermal conductivity of PAAc-PPy cryogel, all the composite cryogel sheet is mounted on the mushroom-shaped PAAc cryogel substrate to further eliminate the heat transfer and maintain the same evaporation condition. Water uptake by cryogels and water transport inside cryogel is also important to achieve excellent photothermal evaporation. Thus, the water absorption behaviours of PAAc hydrogel and different cryogels are compared through swelling kinetics study, as shown in Figure 4a. It takes about 4 h for PAAc hydrogel (H-PAAc) to reach swelling equilibrium and the swelling ratio is about 4.6 g/g. On the contrary, all the cryogel reaches swelling equilibrium within 10 s and the swelling ratio of PAAc cryogel (C-PAAc) is about 7.3 g/g, as shown in Figure 4b. To verify capillary action, one rectangle sheet of PAAc cryogel or hydrogel (1 cm × 6 cm) was used to absorb thinned solution of red ink, as recorded in Video S1 and S2 (Supporting Information), respectively. It is seen that the cryogel transfers red ink to the uppermost end in only 25 sec while the hydrogel does only 0.5 cm after 120 sec. The rapid water uptake for cryogels and speedy transport of water inside the cryogels is attributed to their interconnected and hierarchical macro-porosity. Later comparison of photothermal evaporation of pure water between PAAc hydrogel and cryogel as the substrate provides additional support.

Photothermal evaporation of pure water with different cryogels and energy consideration
To evaluate photothermal evaporation performance of cryogels, one mushroom-shaped PAAc cryogel is used as a pedestal covered with one composite cryogel sheet. Although the apparent density of each cryogel is lower than water, the completely swollen cryogels are still unable to self-float on water. Additionally, elevated temperature from sunlight absorption might bring out the heat loss. Therefore, all the tested cryogels are entrapped in PSt foam with the mushroom tip keeping in touch with water reservoir (Scheme S1 and Figure  S1b). The steady photothermal evaporation of pure water without and with the cryogels is shown in Figure 5a. In the dark, the evaporation rate of pure water is 0.350 kg•m -2 •h -1 . Under onesun irradiation, the evaporation rate of pure water without and with pure PAAc cryogel is almost the same (0.553 and 0.555 kg•m -2 •h -1 , respectively). This result proposes that PAAc cryogel does not affect the water transport and evaporation. After one composite cryogel sheet is used, the evaporation rate increases obviously with more than three folds.  During steady photothermal evaporation, incident sunlight is absorbed by composite cryogel sheet and the absorbed solar energy is converted into heat to accelerate water evaporation. According to UV-vis-NIR absorbance spectrum (Figure 2c), both PPAc-PPy and PAAc-PANI cryogel has almost 100 % absorption of incident light but the absorption percent by both PAAc-GO and PAAc-rGO cryogel is relatively low. Thus, the high evaporation rate for PAAc-PPy and PAAc-PANI cryogel could be assigned to their efficient ability of sunlight absorption.
Considering that PPAc-PPy and PAAc-PANI cryogel has similar sunlight absorption efficiency but different evaporation rate as well as that PAAc-GO cryogel has lower light absorption efficiency than PAAc-rGO cryogel but their evaporation rates are fairly close, there are other factors influencing the photothermal evaporation rate. Within certain period, the absorbed sunlight energy ( abs ) during steady photothermal evaporation is consumed by water-tovapour conversion (Q evp = R real × Δ H evap ), heating evaporated water from room temperature to the temperature of cryogel sheet (Q heating, water = C water × ΔT) and possible heat loss (Q loss ), as indicated with the following equations based on energy balance.
abs inc = abs = real × (Δ evap + water × ∆ ) + loss Eq.5 where R real is the real light-driven evaporation rate, being equal to the evaporation rate under light irradiation (R light ) subtracted by the evaporation rate in dark (R dark ).  inc is the incident light intensity (one-sun as 1 kW•m -2 ).  abs is the light absorbance, i.e. the light absorption efficiency. C water is the specific heat capacity of water and taken as a constant of 4.18 J•g -1 •K -1 . ΔT is the temperature increment of water during the photothermal evaporation and taken as the temperature difference between the cryogel surface under irradiation and the water reservoir. H evap is the evaporation enthalpy of water, which decreases indeed with the increase of temperature. H evap is about 2250 and 2400 J/g at 100 and 40 o C under standard pressure, respectively. Since ΔT is about 15 o C and C water × ΔT is about 3 % of H evap (Table S3), lessening H evap is an important and efficient approach to increase evaporation rate. As mentioned above, the decrease in H evap can be achieved with the increase of evaporation temperature. Additionally, considering that H evap is attributed to the breakdown of intermolecular interaction among water molecules, especially the hydrogen bonding, the appropriate interaction between water and Please do not adjust margins Please do not adjust margins cryogel sheet could prompt the decrease of H evap . This assumption is experimentally supported with the phenomenon that PAAc-GO with oxygen-containing groups exhibits an apparent evaporation efficiency ( evap,app ) of 123 % (Table S3), considering no influence of cryogel structure on H evap . As proposed by the literature 24,45 , there are three states of water molecules in the composite cryogel, namely free water, transitional water and bound water. The molecule of free water is connected with each other through hydrogen bond and confined in dynamic and stable clusters, making the water evaporation more difficult. Since there are still hydrogen bond and other interaction between water molecules and the cryogel pore-wall, the molecules of bound water are also confined and difficulty to evaporate. As for transitional water, water molecules are loosely confined and more likely to escape from water cluster and cryogel network. Therefore, the matrix as well as the photothermal material should be delicately chosen to balance the interactions of water-to-water and water-tomatrix, which could lessen the real evaporation enthalpy of water.
Because much low thermal conductivity of cryogels and the entrapping of cryogel sheet inside PSt foam, the heat loss (Q loss ) in Eq. 5 is neglected. Thus,  evap and total evaporation efficiency ( total ) are calculated according the following equations.
Taking H evap as 2400 J/g (40 o C), all the total evaporation efficiency for different composite cryogel is more than 82 %, being comparable with the literature (Table S4). However, it is questionable that the total evaporation efficiency for PAAc-PPy cryogel is nearly 100 %, considering the possible heat loss and the deviation of H evap from its real value at higher temperature than 40 o C. Although H evap for photothermal evaporation has been measured with differential scanning calorimeter (DSC) by previous reports 24 , those obtained values do not reflect the real evaporation enthalpy at the operation temperature since the wide temperature scan of DSC.

Photothermal evaporation of pure water and water transport consideration
As discussed beforehand, PAAc cryogels have excellent ability to transport water due to their high hydrophilicty and hierarchical porosity. Excellent water transport ensure the instant supply from water reservoir to cover the evaporated water in composite cryogel sheet, especially in the case with high evaporation rate. Therefore, photothermal evaporation under four-sun irradiation has also been performed with composite cryogel sheets and mushroom-shaped PAAc substrate. Upon light irradiation, water mass loss was immediately recorded. As shown in Figure 6 is the variation of mass loss versus time within 1 h. In the first 10 min, all cases exhibit a slower mass loss because the absorbed light energy is partially consumed to heat cryogel sheet. After 20 min, photothermal evaporation goes steady for all cryogel sheets and the evaporation rate within 20~60 min is determined to be 6.492, 5.663, 5.652 and 5.407 kg•m -2 •h -1 for PAAc-PPy, PAAc-PANI, PAAc-rGO and PAAc-GO cryogel sheet, respectively. With mushroom-shaped PAAc cryogel as the substrate, water transport insides is sufficiently high to satisfy the surficial water evaporation. After the replacement of the substrate with PAAc hydrogel, the water mass loss behaviors differently as suggested by the hollow data points of PAAc-PPy@hydrogel. From 40 min, the fitting line slopes up, indicating the decrease of evaporation rate. Based on the slopes within 20-40 min and 40-60 min, average evaporation rate in the related duration is 4.871 and 2.962 kg•m -2 •h -1 , respectively. This result is caused by the slow water transport inside PAAc hydrogel, leading to the shortage of water in PAAc-PPy cryogel sheet. It can be observed that PAAc-PPy cryogel sheet starts to distort at 40 min and finally becomes warped as shown in Figure S10a. The surplus heat has to increase the temperature of PAAc-PPy cryogel sheet. After being irradiated with four-sun light for 1 h, T ave and T max for PAAc-PPy@cryogel remains 57.0 and 60.1 o C, respectively. However, due to insufficient water supply when the hydrogel is used as the substrate, the light-to-vapour cannot happen fluently, and the light energy absorbed by the photothermal layer is mostly converted into heat. T ave and T max for PAAc-PPy@hydrogel increases to 150.0 and 174.7 o C, respectively. All the data are offered in Table S5. Long-term photothermal evaporation has been done under one-sun irradiation for 4 h. The mushroom-shaped PAAc cryogel and composite cryogel sheet were assembled and the starting moment was taken as that upon light irradiation. The results are demonstrated as the solid points in Figure 6b. Over the duration of 4 h, the fitting lines seems linear for PAAc cryogel case, indicating the constant evaporation rate and sufficient water transport inside PAAc cryogel. As well, PAAc-PPy cryogel sheet has the highest evaporation rate and PAAc-GO the lowest.  Figure S10b, PAAc-PPy cryogel sheet at the top of hydrogel substrate becomes short of water and the dehydration makes the sheet gradually warp upwards. At the same time, the surface temperature of different cryogel sheets was measured every half an hour and the average values are demonstrated in Figure S11a. After 30 min, the temperatures for all cryogel cases quickly rise to a level-off value with the highest for PAAc-PPy@cryogel. Extraordinarily, the temperature of PAAc-PPy@hydrogel rises to 45.3 o C at 150 min, then keeps increasing up to 67.2 o C at 240 min. In spite of the temperature increase in this case, evaporation rate drops obviously after 120 min ( Figure S11b). Based on the above evaporation results, it is established that PAAc cryogel as the substrate could efficiently transport water from water reservoir to evaporation sheet surface, while PAAc hydrogel would fail. The morphology observation by SEM, the measurement of swelling rate and the capillary test provide the credible supports.
To clarify the long-term durability and reusability of current photothermal evaporation system, different composite cryogel sheets and mushroom-shaped PAAc cryogel substrate have been assembled to test photothermal evaporation within three months. Each test was performed occasionally and the recorded durations were accidentally chosen. As shown in Figure 7a-d are the results of ten tests for four composite cryogel sheets with mushroom-shaped PAAc cryogel substrate. All the photothermal evaporation systems maintain the high evaporation rate.

Photothermal evaporation of salty water with different cryogels
To evaluate the application of cryogels in seawater desalination, simulated seawater (3.0 × 10 4 mg/L for Na + , 3.5×10 3 mg/L for Mg 2+ and 1.76×10 3 mg/L for K + ) was used as water reservoir to perform photothermal evaporation with PAAc-PPy cryogel sheet and PAAc cryogel substrate. Under one-sun irradiation, photothermal evaporation lasted for 12 h. Then, the light was turn off and the tip of mushroom-shaped cryogel kept the same status for another 12 h. Afterwards, this procedure was repeated in the following 4 days to follow the mass loss and the salt precipitation on cryogel sheet. As shown in Figure 8a is the mass loss at every one hour within the first day. Except an unreliable data point (at 10 h), others are in linear correlation. Within each one-hour duration, the mass loss lies in the range of 1.343-1.648 kg•m -2 •h -1 and the average evaporation rate in the first 12-hour is 1.560 kg•m -2 •h -1 , which is a little lower than that of pure water. The decrease in vapour pressure of solution compared with pure solvent is responsible for this result. As shown in Figure 8b, the mass loss of salty water reservoir in the following days keeps decreasing and exhibits the tendency of levelling-off at the fifth day. Considering water evaporation, the concentration of salty water keeps increasing, which is one reason for the further decrease of evaporation rate. Another reason is the salting out during the evaporation, which might clog the pathway of water transport. The surface appearance of PAAc-PPy cryogel sheet was recorded as shown in Figure 8c. The salt separated out on the surface keeps increasing in amount at the night beginning of each day. It hinders the solar light absorption and water volatilization, and thus the evaporation rate of water might be reduced. However, due to rapid water transport inside PAAc cryogel, some salt was dissolved during the night, which is evidenced by the less salt on the surface at the next morning. At 20:00pm of the fifth day, the salt on the surface of PAAc-PPy cryogel reached maximum amount, due to the edgepreferential crystallization 46  With a self-made device (Figure 9 right), photothermal evaporation of simulated seawater was performed under onesun irradiation and the evaporated water was collected. As shown by the recorded video (Supporting Information, video S3), water vapour was generated under light irradiation and gradually condensed on the collector top. The water drops flowed along the inner surface and accumulated at the button. ICP-MS was used to determine the ion concentration. The ion concentrations of simulated seawater were pretty high. After photothermal evaporation, the concentrations of those ions were reduced by at least 4 orders of magnitude, being 0.015 mg/L for Na + , 0.102 mg/L for Mg 2+ and 0.063 mg/L for K + ( Figure  9). All of the three ions are far below than the fresh water taste threshold defined by World Health Organization (WHO).

Figure 5
The concentrations of three ions in a simulated seawater before and after photothermal desalination (Left) with one self-made vapour collection device (right).

Conclusions
Through cryopolymerization, different PAAc cryogels with photothermal materials have been prepared to assemble photothermal evaporation systems with composite cryogel sheet and mushroom-shaped PAAc cryogel substrate. The hydrophilicity and hierarchical porosity facilitate the rapid water absorption and efficient water transport compared with PAAc hydrogel prepared at room temperature. The porous morphology is also helpful to lesson heat loss. Since the strong interaction between PAAc and photothermal materials, composite cryogels with PPy, PANI, GO and rGO exhibit uniform distribution of photothermal materials inside PAAc cryogel. Thus, light absorbance of composite cryogel sheet is almost 98% for PAAc-PPy and PAAc-PANI cryogels. In steady photothermal evaporation under one-sun irradiation, the evaporation rate of pure water for PAAc-PPy cryogel sheet can reach 1.819 kg•m -2 •h -1 with evaporation efficiency of about 100 % if possible heat loss and the deviation of evaporation enthalpy are neglected. However, the abnormal efficiency for PAAc-GO sheet suggests the uncertainty of such assumption. Increasing the incident light to four-sun, evaporation rate increases obviously for all composite cryogel sheet. PAAc-PPy cryogel sheet still exhibits the highest evaporation (6.492 kg•m -2 •h -1 ), but after replacing the substrate with PAAc hydrogel, its evaporation rate drops noticeably (2.962 kg•m -2 •h -1 ) due to the poor water transport of PAAc hydrogel. All photothermal evaporation systems with composite cryogel sheet and PAAc cryogel substrate demonstrate long-term availability and reproducibility of photothermal evaporation under one-sun irradiation but PPAc-PPy sheet with PAAc hydrogel substrate displays slow-down evaporation. As for photothermal evaporation of salty water with PPAc-PPy sheet with PAAc hydrogel substrate, the evaporation rate under one-sun reaches to 1.560 kg•m -2 •h -1 within the first 12-hour. The accumulated salt could be partially removed with renewed water through cryogel capillary and the performance was recovered by washing with salty water. The ion concentrations in condensed water reduced significantly. Therefore, the present PAAc cryogel is an excellent platform for photothermal evaporation to incorporate different active materials in water purification and seawater desalination.

Author Contributions
Wei

Conflicts of interest
There are no conflicts to declare.