Efficient solar-driven steam generation for clean water production using a low-cost and scalable natural rubber composite sponge

Abstract

Water scarcity is a global issue which might feasibly be addressed through the use of solar energy to produce uncontaminated steam from contaminated water. This technique would allow greater efficiency in purifying wastewater, or desalinating seawater, to produce an adequate supply of clean water. This work therefore presents a novel design for a solar receiver in the form of a composite sponge made from iron oxide black and natural rubber, prepared via the Dunlop process, which is commonly applied in the rubber sector. This composite sponge can absorb solar energy across a broad spectrum before focusing that energy directly on the interfacial surface. In tests using simulated seawater, and water containing organic dyes and heavy metals, the condensed steam produced met the required standards for potable water. The composite material involved exhibited durability, producing stable results beyond 20 cycles of evaporation and cooling. Furthermore, iron oxide black is cheap, abundant, and available in commercial quantities, while natural rubber latex and its associated technologies are widely established for large-scale usage. Therefore, solar receivers based on an iron oxide black/natural rubber composite sponge have significant potential in various applications which make use of solar steam generation, for instance, desalination for freshwater production, or even for sterilization.

---

Water impact

Utilizing an iron oxide/rubber composite sponge for solar distillation offers a sustainable method for purifying wastewater, or desalinating seawater into potable water. Iron oxide black is inexpensive and readily available in commercial quantities, while natural rubber and its associated technology are well-established for large-scale usage. So, these materials will soon make the transition from the laboratory to real-world and commercial applications.

1 Introduction

Among the natural resources, water and sunlight are perhaps the most abundant and readily available. Solar steam generation (SSG) is an approach that combines both resources and offers a means of addressing the challenges of inadequate water supplies in a sustainable and environmentally friendly manner.¹ By converting solar energy to heat, SSG can evaporate water from contaminated and non-volatile solutions. The water vapor is captured and condensed and pure water is collected. With SSG, clean water can be obtained from either polluted water supplies or seawater, and the use of solar power means there are no carbon dioxide emissions to consider.² Furthermore, the SSG technology does not consume large amounts of energy, nor is it operated at high pressure, while in addition, no membranes are required. Since sunlight is free and readily accessible, the cost of operating an SSG system is relatively low.³

Purifying water by solar distillation is not a new idea, but the poor efficiency of converting sunlight to heat makes the process generally impractical. Traditionally, the entire mass of water has been heated to produce water vapor, but this practice rather wastes solar energy.⁴ In contrast to this bulk heating method, the use of an interfacial solar evaporation process offers superior efficiency because only the water in contact with the evaporator is heated. This heat localization is supported by the design of the photothermal materials involved.⁵ More efficient light-to-heat conversion, better evaporation rates, and minimal heat loss can be achieved by obtaining an optimal balance between broadband light adsorption and heat localization. In addition, water must be transferred quickly to the surface of the evaporator and the evaporator must allow the vapor to escape efficiently. These features can support an effective solar-powered steam generation system.⁶

The SSG system relies upon the absorbers for its efficiency. The absorbers are responsible for the absorbance of light and the conversion of energy to produce steam. The light-to-heat conversion efficiency of absorbers is a crucial parameter and has been investigated in several reports. The absorbers have included semiconductors,⁷ metallic nanoparticles,^8–10 and various materials based on carbon.¹¹ However, absorbers with noble metal components face the problem of high prices,¹² while materials based on carbon, such as graphene, carbon nanotubes, or graphene oxide, require extensive preparation which makes them largely unsuitable for SSG systems on a large scale. Alternatively, solar absorbers could be based on metal oxides. When solar radiation is absorbed by metal oxide-based semiconductors, above-band gap electrons and holes are generated which produce substantial quantities of heat during the course of their non-radiated relaxation.¹³ One example is iron oxide black (IO), which is a widely used inorganic black pigment. It is a non-toxic heavy metal oxide, and therefore relatively environmentally friendly in comparison to black pigments such as lead sulfide, manganese oxide, or spinel copper chromite.¹⁴ Furthermore, in terms of global production quantities, it is second only to white titanium dioxide,¹⁵ and therefore it would appear to be a suitable source of black pigment for the SSG process.

Usually, solar absorbers are supported within a matrix to create a functional evaporator. In contrast to membrane-type two-dimensional evaporators, three-dimensional porous components provide unobstructed water-transport pathways and channels that allow vapor to escape freely, delivering superior evaporation performance.¹⁶ Furthermore, there is a larger surface area to aid evaporation, as well as greater efficiency in light trapping.⁵ An interesting form of rubber which is under consideration for numerous applications is the natural rubber latex sponge. It has open cell structures that can be useful for thermal insulators, energy absorption, and in many lightweight products.¹⁷ Although the literature contains few references to natural rubber latex sponge in the context of SSG, latex sponge does appear to have the appropriate chemical and physical properties to support a photothermal material for an interfacial solar evaporator. It also offers the advantages of low thermal conductivity, good elasticity, and strong resistance to solvents while remaining very durable.^18,19 Rubber is also a renewable plant-based material which can be considered environmentally friendly, especially when considering the fact that the process of harvesting rubber is actually beneficial to the trees.

Although natural rubber latex sponges have been used as supporting materials for polymer-based light absorbers such as polypyrrole ,¹⁸ polydopamine,¹⁹ and polyvinyl alcohol/MXene/protonated g-C₃N₄ nanosheet (PVA/MXene/p-g-C₃N₄) composites,²⁰ certain limitations have been observed. One of these is that when the pre-formed polymer-based light absorber solution is used to soak the rubber latex sponge, or is brushed onto the latex sponge surface, the resulting coating on that surface does not encourage entrapment between natural rubber and the polymer-based light absorber, and does not promote interpenetrating networks. As a consequence, the use of these materials over the longer term appears rather limited. Furthermore, while it is certainly not difficult to purchase polypyrrole and polydopamine, they are not available in sufficient quantities to support industrial-scale production, while the costs remain high. Meanwhile, MXene and protonated-g-C₃N₄ cannot be purchased, and their syntheses are complicated processes that involve caustic chemicals such as nitric acid, hydrochloric acid, and lithium fluoride. Accordingly, these materials would be difficult to use in practice.

This study presents a simple solar steam generator fabricated using simple techniques that enable the production of a natural rubber sponge composite on a large scale. There are no reports detailing the use of a natural rubber latex sponge composite with IO in SSG systems. IO is a potentially excellent light-absorbing material whose low cost and availability offer advantages over polypyrrole, polydopamine, MXene, and protonated-g-C₃N₄. Moreover, blending IO with natural rubber latex through the Dunlop process strongly entraps the light-absorbing material, producing a durable and stable composite sponge. The rubber industry already makes frequent use of the Dunlop process, so the technology is readily available and well understood, allowing for production on a large commercially viable scale.

2 Materials and methods

2.1 Materials

The Rubber Authority of Thailand in Bangkok provided the following substances: high ammonia-concentrated natural latex (HA latex), a 50% aqueous solution of potassium oleate (KO), aqueous dispersions of sulfur (50%), zinc diethyldithiocarbamate (ZDEC) (50%), zinc-2-mercaptobenzothiazole (ZMBT) (50%), Wingstay L (50%), zinc oxide (ZnO) (50%), diphenylguanidine (DPG) (33%) and sodium silicofluoride (SSF) (12.5%). Industrial iron oxide black pigment (IO) was purchased from Civic Chemical LP, Thailand. Hydrochloric acid (HCl, 37%) and sodium hydroxide (NaOH, 99.0%) were from J.T. Baker. Simulated seawater was prepared from Red Sea Salt (RedSea Co., Ltd). Methylene blue (MB) was purchased from Unilab. Methyl orange (MO) was from Farmitalia Carlo Erba. Lead(II) chloride (PbCl₂) was from Fluka. Cadmium(II) chloride monohydrate (CdCl₂·H₂O) was from HiMedia.

2.2 Preparation of the natural rubber/iron oxide black composite sponge

Composite sponges were produced using the compounding formulations in Table 1. First, the ammonia present in 60% concentrated natural rubber latex was eliminated by employing a blending mixer (KMX750RD, Kenwood kMix) set at a speed of 80 rpm for 1 min. The exact quantities of IO (5, 10, 15, and 20 g) were evenly distributed in 20 mL of deionized water and then mixed into the prepared natural rubber latex with stirring for 1 min. A 10% solution of potassium oleate was then added, and the speed of the mixer was raised to 160 rpm, and mixing continued until the volume of the mixture grew by about three times. At this point, the mixer speed was reduced to 80 rpm, and sulfur, ZDEC, ZMBT, and Wingstay L were introduced and mixed for 1 min. ZnO and DPG were then added and mixing continued for 1 min. SSF was added to promote homogeneity and enhance the process of natural rubber gelation, and the mixture was then poured into a silicone mold and vulcanized in a hot air oven at 90 °C for 2 h. Before drying at 70 °C, the composite sponges were washed to remove any surplus chemicals. The obtained composite sponges were labeled as NRIO-X, with X denoting the quantity of iron oxide black in parts per hundred (phr). A reference natural rubber latex sponge was also made without using IO, and labeled NR.

Table 1 The compounding formulations of natural rubber/iron oxide black composite sponges

Ingredients	Total solid content (%)	Formulation^a (phr)	Weight (g)
a Ingredients are expressed as parts per hundred of rubber.
HA latex	60	100	167
Potassium oleate	10	1.5	15
Sulfur	50	2	4
ZDEC	50	2	4
ZMBT	50	2	4
Wingstay L	50	2	4
ZnO	50	5	10
DPG	33	1.4	4
SSF	12.5	1.66	13
Iron oxide black (IO)		5, 10, 15, 20	5, 10, 15, 20

---

2.3 Material characterization

The prepared composite sponges were examined by scanning electron microscopy coupled with energy dispersive spectroscopy (SEM–EDS, Quanta450, FEI) to elucidate morphologies and iron species distributions. The pore size distributions of macropores in the various sponges were determined from pore size data obtained with ImageJ software. μ-CT images of NRIO-20 were produced using a high-resolution μ-CT system (SKYSCAN 1173 High-Energy Spiral Scan Micro-CT, Bruker) at 70 kV and 200 μA with a pixel size of 7.0 μm, and a sample size of 1 × 1 × 1 cm³. The absorbance, transmittance and reflectance profiles of the prepared sponges were measured with a double beam UV-visible/NIR spectrophotometer (V-770, Jasco Inc.) equipped with an integrating sphere and fine BaSO₄ powder as a reference. Functional groups of the prepared sponges were analyzed with attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR, VERTEX 70, Bruker). Phase identification of the prepared samples was performed by using X-ray diffractometry (XRD, D8 Advance, Bruker). The thermal conductivity of the prepared samples was determined using a thermal conductivity analyzer (TPS2500S, HOT DISK). The water contact angle was measured by using a contact angle goniometer (Ossila).

2.4 Evaluation of solar-driven vaporization

A water container was overlaid with a composite sponge to absorb light. The sponge was exposed to simulated solar light by using a xenon arc lamp with an AM1.5G air mass filter (94011A, LCS-100, Newport, Oriel instrument, USA) (Fig. S1†). The change in water mass was recorded in real-time using a computer-controlled electronic analytical balance (ME204, Mettler Toledo) with a precision of 0.0001 g. The light intensity was calibrated to 1 kW m⁻² using a calibrated reference cell and meter (Newport Corporation). All investigations were conducted at 25 ± 1 °C and 60 ± 10% relative humidity. An infrared thermal imaging camera (FLIR-E63900, Teledyne) was used to produce the thermal images of the composite sponges during solar-driven vaporization and to monitor the surface temperature of the materials. The evaporation efficiency was calculated in terms of the conversion efficiency (η) using the following equations:

h_LV = λ + CΔT (2)

in which m denotes the water evaporation rate (kg m⁻² h⁻¹), h_LV is the total liquid–vapor phase-change enthalpy including sensible heat, P₀ is the nominal solar irradiation of 1 kW m² and C_opt refers to the optical concentration (1 sun irradiation). λ denotes the latent heat of water phase-change, C is the specific heat capacity of water (4.2 kJ kg⁻¹ K⁻¹), and ΔT is the difference in temperature between the steam and the surrounding environment.

2.5 Desalination and clean water production

Freshwater production experiments were conducted in a custom-made plastic condensation chamber, as shown in Fig. S2.† This study tested pure water production from simulated seawater and solutions containing the organic dyes MB and MO and the heavy metals lead(II) and cadmium(II). The concentration of MB and MO before and after SSG was analyzed using a UV-VIS spectrophotometer (UV-2600i, Shimadzu Corp.) at λ_max 664 and 463 nm, respectively. Inductively coupled plasma-optical emission spectrometry (ICP-OES, Avio500, PerkinElmer) was used to determine the ion concentrations of seawater and heavy metals before and after SSG. The total dissolved solid (TDS) levels and electrical conductivity (EC) of the simulated seawater before and after condensation were measured using a TDS meter (Mi TDS Meter, Xiaomi), and a conductivity meter (SevenCompact™ Duo S213, Mettler Toledo), respectively.

3 Results and discussion

3.1 Preparation and characterization of composite sponges

The composite sponges were prepared using natural rubber latex and industrial-grade iron oxide black (IO) in a facile Dunlop process. The resulting sponges were grey to black in color, depending on the loading of IO (Fig. 1). At higher loadings of IO, the sponges were correspondingly darker. All samples were smooth with well-distributed pore channels.

Fig. 1 Digital image of composite sponges prepared in this study.

Physical morphologies and microstructures of the composite sponges were observed by SEM. All the samples had interconnected open macroporous networks throughout, with fine and uniform pores (Fig. 2A). Pore size distributions were determined using ImageJ 1.53k free software (Fig. S3†). All NRIO-X samples exhibited good pore size distributions in which most pores were in the range of 50 to 250 microns. Furthermore, the small pore fraction tended to be slightly greater at higher loadings of IO. The increased formation of smaller pores at higher loadings of IO was due to the stronger interfacial forces and entanglement between the natural rubber matrix and IO. The greater number of physical crosslinking points produced prevented the expansion of molecular chains during foaming.²¹ Also, the higher IO content increased the viscosity of the rubber, which limited bubble growth. These findings are supported by the classical nucleation theory, which proposes that a rise in polymer viscosity leads to an increase in the number of small pores formed in the foam.²²

Fig. 2 A) Cross-sectional SEM images and B) bulk densities of all prepared composite sponges. C) The photographs show the magnetic properties of the composite sponge. D) μ-CT images reveal the open-pore (grey color) and closed-pore (blue color) fractions of NRIO-20.

The 3D porous structure of the NRIO-20 sponge was analyzed using a high-resolution μ-CT system. The results confirmed the interconnected porous structure and open porosity of the composite sponge (Fig. 2D). The open-pore fraction of NRIO-20 was determined to be 90.2% with an average pore diameter of 230 microns. The open-pore structure of NRIO-20 had a surface-to-volume ratio of 101.6 mm⁻¹ and the closed-pore volume of NRIO-20 was less than 0.003%. The bulk density of each composite sponge (Fig. 2B) was significantly lower than the density of water (1 g cm⁻³) and increased with the loading of IO. The open-pore structure and buoyancy of the composite sponges presented in this study offer further compositional and structural advantages for SSG.

The distribution of IO within the composite sponges was analyzed using SEM/EDS. SEM/EDS mapping showed even distributions of iron species throughout the samples (Fig. S4A†). The amount of Fe increased in accordance with the IO loading. NRIO-20 had the highest atomic% of Fe (Fig. S4B†). To attain a more comprehensive understanding of the distribution of IO particles throughout the composite sponge, SEM/EDS analysis at higher magnification was performed. As illustrated in Fig. S5,† the iron oxide particles are mostly confined within the rubber matrix.

The phases of IO, NR and NRIO-20 were identified using XRD analysis. The XRD pattern of IO (Fig. S6†) exhibited distinctive peaks at 2θ of 18.33°, 30.14°, 35.48°, 43.10°, 57.00°, and 62.58°, which corresponded to the (111), (220), (311), (400), (333), and (440) planes, respectively, and matched the Fe₃O₄ reference data (JCPDS card no. 01-1111). There were some additional weak peaks of other iron oxide phases, including γ-Fe₂O₃ (maghemite, JCPDS No. 39-1346), α-Fe₂O₃ (hematite, JCPDS No. 80-2377), γ-FeOOH (lepidocrocite, JCPDS No. 76-2301), and α-FeOOH (goethite JCPDS No. 29-0713). These results indicated that IO comprised multiple phases of iron oxide, with the Fe₃O₄ (magnetite) phase being the predominant contributor. The XRD pattern of NRIO-20 presented typical peaks of amorphous natural rubber, with the addition of some shifted weak peaks corresponding to Fe₃O₄. The data suggested that the composite possessed a semicrystalline structure, wherein tiny crystallites were dispersed throughout the amorphous matrix phase.²³ The absence of other Fe₃O₄ peaks could be attributed to signal overlap resulting from the amorphous nature of the natural rubber.²⁴ It should also be noted that the presence of Fe₃O₄ particles can enable directed migration in a magnetic field, and that magnetic materials can produce magneto-optical phenomena when they interact with light. When sunlight interacts with a magnetic material, the properties of the output light may be different from the input light. These properties include the polarization state, spectrum, phase, and intensity of the light.²⁵ This interaction enables the manipulation of light absorption and its conversion into thermal energy²⁶ to enhance thermal energy production inside the material, more effectively elevating the temperature of the adjacent medium in an SSG system. According to previous reports, a buoyant solar interfacial evaporator is a key element of solar thermal conversion efficiency.²⁷ Therefore, the composite sponge developed here could be deployed in a self-floating solar generator in an open-water setting and be recovered with an external magnet, as illustrated in Fig. 2C.

ATR-FTIR spectroscopy was employed to evaluate the functional groups of IO and the composite sponges. The ATR-FTIR analysis of IO (Fig. S7†) supported the results of XRD analysis about the vibrational properties of the Fe–O bond found at both tetrahedral and octahedral sites of Fe₃O₄ and γ-Fe₂O₃.²⁸ These vibrational features were observed in a wavenumber range of approximately 441 to 699 cm⁻¹.²⁹ Furthermore, it is noteworthy that the intense vibration seen at 1014 cm⁻¹ corresponds to the bending vibration of hydroxyl (–OH) modes in γ-FeOOH.³⁰ The vibration characteristics of the composite sponges exhibited a strong resemblance to those of pure latex, with the exception of a notable increase in intensity at 1014 cm⁻¹, which was dependent on the loading of IO. In terms of intricate specifics, the composite sponges exhibited the symmetric stretching vibrations of the CH₂ groups in the methylene group and the asymmetric stretching vibrations of the CH₃ groups in the methyl group in the cis-1,4-polyisoprene unit. These stretching vibrations were detected from 2851 to 2920 cm⁻¹, and at 2958 cm⁻¹, respectively. All the ATR-FTIR spectra also presented absorption bands at 1657 cm⁻¹ and 833 cm⁻¹, which were attributed to the stretching of C=C bonds and the out-of-plane bending of =CH groups in the cis-1,4-polyisoprene moiety. In addition, the C–S and C=S functional groups in vulcanized rubber were indicated by relatively low-intensity and wide peaks at 570 and 1085 cm⁻¹, respectively.

3.2 Solar steam generation tests

The absorption of light by the composite sponges was studied using UV-VIS spectrophotometry within the range of 200 to 2500 nm. The absorption spectra of NRIO-X samples were compared to the spectrum of the NR sponge (Fig. 3A). The NR sponge demonstrated the lowest absorbance as a result of the absence of light-absorbing pigment. The light absorption capacity of NRIO-X increased with increments of IO loading. NRIO-20 demonstrated the highest level of absorbance, reaching 95%, across the entire solar spectral range. The increased light absorption of the composite sponges compared to the NR sponge can be attributed to the reduced reflectance and transmittance (Fig. S8A and B†). Typically, photothermal materials should exhibit a high degree of absorbance across a broad spectrum that includes ultraviolet (UV), visible (vis), and near-infrared (NIR) wavelengths, while minimizing reflection and transmission.³¹ NRIO-20 exhibited excellent light absorption and thus fully met the demands of solar thermal conversion for steam generation.

Fig. 3 A) Light absorption of composite and natural rubber-based sponges. B) Temperature response profiles of composite sponges with and without simulated solar illumination. C) Thermal mapping photographs of NRIO-20 in dry and wet states under 1 sun irradiation. D) Real-time mass change of water in the absence and presence of composite sponges. E) Evaporation rates and evaporation efficiency of NRIO sponges under 1 sun irradiation.

At the same time, in the context of SSG, photothermal materials must effectively manage heat. To understand the effect of local heating in the NRIO-20 sponge, the surface temperature profile was investigated and mapped using an infrared camera (Fig. 3B and C). In the presence of 1 sun irradiation, heat was concentrated on the surface of the material, significantly increasing the temperature of the surface of the dry sponge from 27 to 96 °C within 5 min. The temperature of the sponges remained constant until the exposure to light ceased. After the Xe light was turned off, the temperature dropped significantly, demonstrating that the sponge was capable of a rapid photothermal response. Similarly, the surface temperature of the wet NRIO-20 gradually rose from 27 to 44 °C, attaining equilibrium within 20 min. The temperature responses demonstrated the potential of the NRIO-X sponges to effectively convert solar energy into thermal energy to increase the temperature of water in the uppermost part of the sponge and provide the interfacial heating necessary for the production of steam. This increase in surface temperature was attributed to the significant light absorption capacity of the composite sponge. The localized heating effect observed in NRIO-20 was due to the influence of IO. When the semiconductor is exposed to light, the absorption of photons with a corresponding bandgap results in the excitation of molecules, leading to the generation of electron–hole pairs.³² This enables a wider spectrum of solar radiation to be absorbed and enhances light-to-heat conversion.³³ IO is the most representative narrow-bandgap semiconductor with a considerable capacity for photothermal conversion in the visible light spectrum.³⁴ IO is better than conventional semiconductors because they have larger bandgaps that permit a response only to UV light.³⁵ The properties of IO are therefore more suitable for a light-absorbing material applied in SSG.

The time-dependent mass changes under 1 sun irradiation during SSG were measured to determine the rate of evaporation of water. The mass change of water in the presence of the composite sponges ranged from 0.89 to 1.20 kg m⁻² h⁻¹ (Fig. 3D) and was directly correlated to the light absorption capacities and loading of IO. NRIO-20 produced the highest evaporation rate of 1.20 kg m⁻² h⁻¹, which was three times that of irradiated DI water in the absence of the composite sponge (0.36 kg m⁻² h⁻¹). The light-to-heat conversion efficiency of NRIO-20 reached 74.34% (Fig. 3E) and was attributed to its good light absorption. As a result, NRIO-20 was chosen for further investigation.

In addition to the excellent light absorption efficiency and highly interconnected pore channels of the evaporator, surface wettability and water uptake capacity are other key characteristics that contribute to the robust evaporation rate of photothermal materials. The wettability of the sponge was evaluated in terms of contact angle. As illustrated in Fig. S9A,† the average contact angle of NR is 78.5 ± 0.3°, indicating a relatively low hydrophilic surface. Filling the NR sponge with IO resulted in an 82.5 ± 0.3° contact angle (Fig. S9B†), demonstrating that the hydrophilicity of NRIO-20 reduced somewhat with IO loading. This might be attributed to the higher surface roughness and porosity of the composite sponge. A surface with a rougher texture and more porosity may trap a greater number of small spaces filled with air, which results in a smaller area of contact between the liquid and the material and eventually leads to a larger contact angle.³⁶ Initially, when the composite sponge is dry, the water contact angle changes slowly with time (Video S1†), suggesting that the sponge has a limited water absorption capacity. This is probably due to the air trapped inside the pores. When the rubber sponge is pre-wetted with water, the water contact angle can change rapidly (Video S2†). This indicates that pre-wetting the composite sponge results in a significant increase in its water transfer speed. This is because water is absorbed into the pre-wetted porous sponge by capillary action, or because hydrophilic components in the sponge migrate or segregate to the material surface.³⁶ This particular behavior has been commented on before in previous reports on polypyrrole-coated natural latex foam¹⁸ and natural rubber latex foam filled with micro- and nanofibrillated cellulose.³⁶ In order to enhance the amount of water at the interface of the evaporator, the composite sponge was saturated with a test solution prior to conducting the evaporation experiment. The water absorption ratio of the composite sponge was determined in terms of the amount of water absorbed by a material relative to its initial weight, according to ASTM D471. Fig. S9C† depicts the composite sponge's water absorption ratio following the pre-wetting step. This indicated that the sponge required around 5 minutes to become completely saturated. A sponge that is completely soaked with water would have an increased capillary force, which aids in maintaining the supply of water at the interface of the evaporator.³⁷

To achieve superior photothermal efficiency, it is imperative that the solar interfacial evaporator has a high capacity for light absorption for efficient conversion of light into heat, and a low thermal conductivity to minimize the dissipation of heat by conduction, convection, and radiation.^38,39 The thermal conductivity of NRIO-20 was investigated in both dry and wet states and compared with the thermal conductivity of the NR sponge (Fig. 4A). The thermal conductivity of the photothermal material in its hydrated state must be known in order to be sure that performances will be consistent throughout the whole solar steam generating process. The thermal conductivity of NRIO-20 in its dry state was 0.048 W m⁻¹ K⁻¹. This value is relatively low compared to other materials (Table S1†) due to the inherent insulating properties of natural rubber⁴⁰ and the porosity of the sponge that impedes heat transfer and produces some degree of thermal insulation.⁴¹ Furthermore, it is worth noting that the NRIO-20 sponge exhibited a low thermal conductivity of 0.108 W m⁻¹ K⁻¹, even when fully saturated with water. This value is considerably lower than the thermal conductivity of pure water, which is 0.600 W m⁻¹ K⁻¹. This thermal characteristic of the NRIO-20 sponge is particularly advantageous as it assists in minimizing the dissipation of heat generated during photothermal conversion.

Fig. 4 A) Thermal conductivity of pure latex and NRIO-20 sponge in dry and wet states. B) The recyclability of NRIO-20 was tested over 20 cycles of water evaporation. C) Mass change of water due to evaporation under 1 sun irradiation of NRIO-20 before and after ozone or acid treatment. D) Real-time mass changes of simulated seawater and DI water under 1 sun irradiation of NRIO-20.

The photothermal materials prepared in this study not only demonstrate remarkable performance in light-to-heat conversion, but also exhibit exceptional stability in terms of water evaporation rate over 20 consecutive runs, as depicted in Fig. 4B. This observation suggests a high level of cycling stability. In practical terms, the degradation of polymer-based materials exposed to water, air, and particularly ozone can lead to a rapid decline in their photothermal conversion performance and reduce the operational lifespan of solar steam generators.⁴² Therefore, a study was carried out to examine the structural attributes and SSG performance of NRIO-20 after treatment with ozone. NRIO-20 was subjected to a 72 hour ozone treatment in a chamber with an ozone concentration of 50 pphm and an interior temperature of 40 °C.⁴³ The findings (Fig. S10†) indicated that the porosity of NRIO-20 was preserved, and that no surface cracking of the composite sponge was present following ozone treatment. In addition, the ozone-treated sample exhibited a similar vapor production capacity to the untreated sample (Fig. 4C). The results indicated that ozone curing had no statistically significant impact on the parameters of interest and implied that NRIO-20 could be used for an extended period without significant complications. Furthermore, the corrosion resistance test was carried out as shown in Fig. S11.† The NRIO-20 sponge was submerged in an acid solution (pH 1) for 7 days. The results shown in Fig. S12† demonstrate that the acid-treating process did not cause any surface cracking on the composite sponge and that the porosity of NRIO-20 was maintained. Additionally, Fig. 4C shows that the acid-treated composite sponge maintains a comparable evaporation rate as the untreated sample. This implies that the acid treatment did not have any apparent effects on the material's performance.

3.3 Solar-driven clean water production

Numerous solar-driven steam generators have been documented in the literature. The reports have showcased a diverse range of uses, including desalination, and water purification and sterilization.⁴⁴ Desalination stands out as the predominant utilization of high-performance solar steam generators. To explore the performance of NRIO-20 in this application, simulated saltwater was utilized with the average salinity of the global oceans. The mass change of simulated seawater after 1 h of exposure to 1 sun irradiation was identical to that of DI water (Fig. 4D). Under the same conditions, the water evaporation rates of NRIO-20 in seawater and DI water were around 1.20 kg m⁻² h⁻¹ without discernible signs of salt crystals on the surface of the sample. NRIO-20 completely rejected 1 g of NaCl on the composite surface within a short time (Fig. S13†). The rejection of salt may be associated with the hierarchical micron-scale porous network of NRIO-20. The distinctive configuration of this structure provides interconnected pathways that facilitate the effective transfer of water. When salt particles are deposited on the surface of the solar evaporator, a rapid dissolution process ensues, resulting in the formation of a concentrated salt zone on top of the solar evaporator. The disparity in concentration between this highly saline zone and the surrounding water prompts diffusion and convection processes that reduce the concentration of salt on the solar evaporator, preventing the persistence of salt deposits.⁴⁵ The findings provided confirmation of the possible applications of NRIO-20 as an evaporator in desalination systems.

An on-site solar still was established to examine the efficacy of water purification and desalination processes using the NRIO-20 evaporator. Following the completion of the solar-powered desalination process, the condensed water was collected, and the levels of ions of principal interest were analyzed using ICP-OES. The considerable reduction in salinity complied with the World Health Organization (WHO) drinking water guidelines. The % removals of Na⁺, Ca²⁺, K⁺, Mg²⁺, and B³⁺ were 99.98, 99.96, 99.97, 99.99 and 99.45, respectively (Fig. 5A). Traditional methods of desalination and water purification typically employ techniques such as reverse osmosis (RO) and ultrafiltration. Water treatment with RO is recognized for its efficiency, but the systems involve substantial upfront investment, power consumption, effluent handling, and regular replacement of filters and membranes. However, RO systems also use a significant amount of water, with around 75% or more of the water being discarded along with the contaminants.⁴⁶ In this study, interfacial solar-driven vapor production presents a viable and environmentally friendly solution which runs without the release of greenhouse gases, making the technology sustainable and energy-efficient.⁴⁷ The condensed water satisfies the WHO drinking water standards for electrical conductivity (EC) and total dissolved solids (TDS). It is also comparable to the permeate water obtained from conventional seawater desalination by reverse osmosis (SWRO).⁴⁸ The comparative performances (Table S2†) suggest that the NRIO-20 composite sponge has the potential for use as a solar evaporator for the production of clean water.

Fig. 5 A) Salinity of simulated seawater before and after evaporation. B) Absorbance spectra show the absence of organic dyes in condensed water produced from contaminated water by solar steam generation with the NRIO-20 interfacial evaporator.

The removal of organic dyes and heavy metals from water by using SSG was investigated. MB and MO were selected as typical cationic and anionic organic dyes. The impacts of the respective positive and negative charges on the SSG process were evaluated. Neither MB nor MO was detected in the condensed water (Fig. 5B), implying the absence of any charge effect. The result indicated that NRIO-20 could be applied to treat dye-contaminated wastewater. Lead(II) and cadmium(II) in wastewater are known health hazards, originating from industrial processes such as metallurgy, electroplating, dyeing, textile printing, chemical processing, battery production, and lead mining.^49,50 In this study, the concentrations of both heavy metals in the condensed water were found to be below the detection limit. Consequently, these concentrations were much lower than the permissible values for human safety established by the WHO. The WHO guidelines specify that the acceptable levels of lead and cadmium in drinking water are 0.01 ppm and 0.003 ppm, respectively.⁵¹ These results confirm the potential for clean water production and water treatment using the low-cost and scalable natural rubber composite sponge NRIO-20. An iron oxide nanofluid (Fe₃O₄, IONF) photothermally increased the surrounding water temperature, but the condensed water obtained was contaminated with nanoparticles of IONF.⁵² That type of problem could be eliminated by incorporating IO into a rubber matrix as presented in this work.

Table S1† displays the comparative characteristics of semiconductor-based solar absorbers when subjected to an irradiation level of 1 sun. In contrast to previous fabricated semiconductor-based evaporators, the proposed evaporator is characterized by ease of fabrication and use, affordability and environmental sustainability. Moreover, IO and natural rubber latex are readily available in large quantities. The proposed composite sponge is self-buoyant, provides a network of linked channels for efficient water and vapor transfer, and offers good thermal insulation properties. When considering factors such as material renewability, sustainability, and environmental friendliness, natural rubber latex outperforms synthetic polymers as a supporting material. Although there are some reports of rubber sponges absorbing coating materials, this is the first study of the incorporation of iron oxide black pigment into a natural rubber latex matrix to produce an evaporator.

In order to conduct a more comprehensive analysis of the economic advantages associated with the NRIO-20 composite sponge, the manufacturing cost of the evaporator was calculated. This cost was then compared to the costs of other photothermal materials chosen from the recent literature. The cost per exposed area of NRIO-20 was calculated at $5.28 per m², which is much lower in comparison to the other materials in Table 2. The present study places particular emphasis on the industrial iron oxide pigment and the well-established Dunlop process, both of which are known to enable large-scale production.

Table 2 Cost estimation of NRIO-20 in comparison to other photothermal materials

Material	Cost per exposed area (USD per m^2)	Ref.
NRIO-20	5.28	This study
3D transformed PPy-decorated jute-cord	55	53
Binary gel based on MXene and montmorillonite	100	54
Heat-treated wood with Au coating	17 509	55
Al–Ti–O PVDF hybrid membrane	11.17	56

---

4 Conclusions

An interfacial composite sponge evaporator was fabricated from iron oxide black and natural rubber for use in solar steam generation (SSG). The composite sponge was produced with relative ease using existing technology for the production of latex sponge. Iron oxide black acted as a solar absorber, and was incorporated while the sponge was forming. No additional coating step was necessary. The attributes of the composite sponge in its optimal form included good solar light absorption (>95%), low thermal conductivity (0.048 W m⁻¹ K⁻¹), an excellent water evaporation rate (1.20 kg m⁻² h⁻¹), and a high level of light-to-heat conversion efficiency (74.34% under 1 sun irradiation). When simulated seawater underwent solar evaporation using the composite sponge, the condensate revealed an almost complete removal of ions. Since natural rubber is cheap and sustainable, and the production process is well-established, it can be argued that the composite sponge developed in this study shows potential in commercial and portable SSG technologies.

Author contributions

PO: investigation, validation, formal analysis, writing – original draft. PT: investigation. PP: investigation. WS: investigation. ST: writing – review & editing, visualization, supervision. LC: conceptualization, methodology, formal analysis, writing – review & editing, visualization, funding acquisition. DD: conceptualization, methodology, validation, writing – original draft, writing – review & editing, funding acquisition, project administration.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was funded by the National Research Council of Thailand (NRCT) and the Kasetsart University Research and Development Institute (grant no. FF(KU)27.67). P. Onsri gratefully acknowledges allowance support from the Science Achievement Scholarship of Thailand (SAST). L. Chuenchom gratefully acknowledges partial support from the Center of Excellence for Innovation in Chemistry (PERCH-CIC), Ministry of Higher Education, Science, Research, and Innovation, Thailand. Thomas Duncan Coyne, Faculty of Science, Prince of Songkla University, is recognized for English proofreading and editing.

References

1. C. J. Chen, Y. D. Kuang and L. B. Hu, Challenges and opportunities for solar evaporation, Joule, 2019, 3, 683–718.
2. Y. Ma and J. Cao, Facile preparation of magnetic porous carbon monolith from waste corrugated cardboard box for solar steam generation and adsorption, Biomass Convers. Biorefin., 2022, 12, 2185–2202.
3. Y. Shi, R. Y. Li, Y. Jin, S. F. Zhuo, L. Shi, J. Chang, S. Hong, K. C. Ng and P. Wang, A 3D photothermal structure toward improved energy efficiency in solar steam generation, Joule, 2018, 2, 1171–1186.
4. M. W. Higgins, A. R. S. Rahmaan, R. R. Devarapalli, M. V. Shelke and N. Jha, Carbon fabric based solar steam generation for waste water treatment, Sol. Energy, 2018, 159, 800–810.
5. S. L. Loo, L. Vásquez, M. Zahid, F. Costantino, A. Athanassiou and D. Fragouli, 3D photothermal cryogels for solar-driven desalination, ACS Appl. Mater. Interfaces, 2021, 13, 30542–30555.
6. J. Yuan, X. Lei, C. Yi, H. Jiang, F. Liu and G. J. Cheng, 3D-printed hierarchical porous cellulose/alginate/carbon black hydrogel for high-efficiency solar steam generation, J. Chem. Eng., 2022, 430, 132765.
7. H. Liu, C. Chen, H. Wen, R. Guo, N. A. Williams, B. Wang, F. Chen and L. Hu, Narrow bandgap semiconductor decorated wood membrane for high-efficiency solar-assisted water purification, J. Mater. Chem. A, 2018, 6, 18839–18846.
8. C. Kim, Y. Ryu, D. Shin, A. M. Urbas and K. Kim, Efficient solar steam generation by using metal-versatile hierarchical nanostructures for nickel and gold with aerogel insulator, Appl. Surf. Sci., 2020, 517, 146177.
9. L. Zhou, Y. L. Tan, D. X. Ji, B. Zhu, P. Zhang, J. Xu, Q. Q. Gan, Z. F. Yu and J. Zhu, Self-assembly of highly efficient, broadband plasmonic absorbers for solar steam generation, Sci. Adv., 2016, 2, e1501227.
10. L. Zhou, Y. L. Tan, J. Y. Wang, W. C. Xu, Y. Yuan, W. S. Cai, S. N. Zhu and J. Zhu, 3D self-assembly of aluminium nanoparticles for plasmon-enhanced solar desalination, Nat. Photonics, 2016, 10, 393–398.
11. L. Hao, N. Liu, B. Zhang, R. Niu, J. Gong and T. Tang, Waste-to-wealth: Sustainable conversion of polyester waste into porous carbons as efficient solar steam generators, J. Taiwan Inst. Chem. Eng., 2020, 115, 71–78.
12. G. Xie, X. Wang, M. Mo, L. Zhang and J. Zhu, Photothermal hydrogels for promoting infected wound healing, Macromol. Biosci., 2023, 23, 2200378.
13. M. M. Zhu, A. D. Xia, Q. Q. Feng, X. S. Wu, C. Y. Zhang, D. X. Wu and H. T. Zhu, Biomass carbon materials for efficient solar steam generation prepared from carbonized enteromorpha prolifera, Energy Technol., 2020, 8, 1901215.
14. D. R. Swiler, Pigments, inorganic, Kirk-Othmer Encyclopedia of Chemical Technology, 2000.
15. G. Pfaff, The world of inorganic pigments, ChemTexts, 2022, 8, 15.
16. B. Peng, Q. Lyu, M. Li, S. Du, J. Zhu and L. Zhang, Phase-separated polyzwitterionic hydrogels with tunable sponge-like Structures for stable solar steam generation, Adv. Funct. Mater., 2023, 33, 2214045.
17. N. Tangboriboon, W. Pannangpetch, K. Aranyik, K. Petcharoen and A. Sirivat, Embedded eggshells as a bio-filler in natural rubber for thermal insulation composite foams, Prog. Rubber, Plast. Recycl. Technol., 2015, 31, 189–205.
18. Y. Xu, J. Wang, F. Yu, Z. Guo, H. Cheng, J. Yin, L. Yan and X. Wang, Flexible and efficient solar thermal generators based on polypyrrole coated natural latex foam for multimedia purification, ACS Sustainable Chem. Eng., 2020, 8, 12053–12062.
19. H. Yu, Y. Shi, A. Ding, J. Liao, H. Gui and Y. Chen, Polydopamine-coated natural rubber sponge for highly efficient vapor generation, Polymer, 2022, 14, 1486.
20. R. T. Ginting, H. Abdullah and V. Fauzia, Facile preparation of MXene and protonated-g-C₃N₄ on natural latex foam for highly efficient solar steam generation, Mater. Lett., 2022, 313, 131779.
21. P. Pongmuksuwan, K. Salayong, T. Lertwiriyaprapa and W. Kitisatorn, Electromagnetic absorption and mechanical properties of natural rubber composites based on conductive carbon black and Fe₃O₄, Materials, 2022, 15, 6532.
22. E. Rostami-Tapeh-Esmaeil, A. Vahidifar, E. Esmizadeh and D. Rodrigue, Chemistry, processing, properties, and applications of rubber foams, Polymers, 2021, 13, 1565.
23. F. Kanwal, S. A. Siddiqi, A. Batool, M. Imran, W. Mushtaq and T. Jamil, Synthesis of polypyrrole–ferric oxide (Ppy–Fe₂O₃) composites and study of their structural and conducting properties, Synth. Met., 2011, 161, 335–339.
24. J. C. Ríos-Hurtado, E. M. Múzquiz-Ramos, A. Zugasti-Cruz and D. A. Cortés-Hernández, Mechanosynthesis as a simple method to obtain a magnetic composite (activated carbon/Fe₃O₄) for hyperthermia treatment, J. Biomater. Nanobiotechnol., 2016, 7, 19–28.
25. T. Lan, B. Ding and B. Liu, Magneto-optic effect of two-dimensional materials and related applications, Nano Sel., 2020, 1, 298–310.
26. J. Han, A. Zhu and H. Wang, Nonreciprocal absorption properties of a magneto-optical microstructure with suppressed photothermal effects, Opt. Commun., 2023, 545, 129666.
27. L. Zhang, X. Xu, J. Feng, B. Bai, N. Hu and H. Wang, Waste semi-coke/polydopamine based self-floating solar evaporator for water purification, Sol. Energy Mater. Sol. Cells, 2021, 230, 111237.
28. B. S. Yadav, R. Singh, A. K. Vishwakarma and N. Kumar, Facile synthesis of substantially magnetic hollow nanospheres of maghemite (γ-Fe₂O₃) originated from magnetite (Fe₃O₄) via solvothermal method, J. Supercond. Novel Magn., 2020, 33, 2199–2208.
29. C. Silva, R. e Silva, A. de Figueiredo and V. Alves, Magnetic solid-phase microextraction for lead detection in aqueous samples using magnetite nanoparticles, J. Braz. Chem. Soc., 2020, 31, 109–115.
30. L. Mei, L. Liao, Z. Wang and C. Xu, Interactions between phosphoric/tannic acid and different forms of FeOOH, Adv. Mater. Sci. Eng., 2015, 2015, 250836.
31. Y. W. Lin, H. Xu, X. L. Shan, Y. S. Di, A. Q. Zhao, Y. J. Hu and Z. X. Gan, Solar steam generation based on the photothermal effect: from designs to applications, and beyond, J. Mater. Chem. A, 2019, 7, 19203–19227.
32. L. Zhu, M. Gao, C. K. N. Peh and G. W. Ho, Solar-driven photothermal nanostructured materials designs and prerequisites for evaporation and catalysis applications, Mater. Horiz., 2018, 5, 323–343.
33. M. Gao, L. Zhu, C. K. Peh and G. W. Ho, Solar absorber material and system designs for photothermal water vaporization towards clean water and energy production, Energy Environ. Sci., 2019, 12, 841–864.
34. J. X. Yao, K. Zhang, W. Wang, X. Q. Zuo, Q. Yang, M. Z. Wu and G. Li, Great enhancement of electrochemical cyclic voltammetry stabilization of Fe₃O₄ microspheres by introducing 3DRGO, Electrochim. Acta, 2018, 279, 168–176.
35. Y. Xu, J. Yin, J. Wang and X. Wang, Design and optimization of solar steam generation system for water purification and energy utilization: A review, Rev. Adv. Mater. Sci., 2019, 58, 226–247.
36. S. Phomrak, A. Nimpaiboon, B.-m. Z. Newby and M. Phisalaphong, Natural rubber latex foam reinforced with micro- and nanofibrillated cellulose via Dunlop method, Polymer, 2020, 12, 1959.
37. Y. Xia, Q. Hou, H. Jubaer, Y. Li, Y. Kang, S. Yuan, H. Liu, M. W. Woo, L. Zhang, L. Gao, H. Wang and X. Zhang, Spatially isolating salt crystallisation from water evaporation for continuous solar steam generation and salt harvesting, Energy Environ. Sci., 2019, 12, 1840–1847.
38. Y. Wang, X. Wu, P. Wu, J. Zhao, X. Yang, G. Owens and H. Xu, Enhancing solar steam generation using a highly thermally conductive evaporator support, Sci. Bull., 2021, 66, 2479–2488.
39. R. T. Ginting, H. Abdullah, D. A. Barus and V. Fauzia, Extremely high-efficiency solar steam generation, robust and scalable photothermal evaporator based on ZIF-67@MXene/rGO decorated rock wool, J. Mater. Chem. A, 2023, 11, 5296–5308.
40. H. Niu, Y. Ren, H. Guo, K. Małycha, K. Orzechowski and S.-L. Bai, Recent progress on thermally conductive and electrical insulating rubber composites: Design, processing and applications, Compos. Commun., 2020, 22, 100430.
41. P. Ruckdeschel, A. Philipp and M. Retsch, Understanding thermal insulation in porous, particulate materials, Adv. Funct. Mater., 2017, 27, 1702256.
42. M. Morciano, M. Fasano, U. Salomov, L. Ventola, E. Chiavazzo and P. Asinari, Efficient steam generation by inexpensive narrow gap evaporation device for solar applications, Sci. Rep., 2017, 7, 11970.
43. B. C. Zhong, Z. X. Jia, Q. F. Shi, C. Yang, Y. F. Luo and D. M. Jia, Characterization of effects of thermal-oxidative aging on styrene-butadiene rubber/silica composites with vitamin C-lanthanum complex, Int. J. Polym. Anal. Charact., 2016, 21, 1–10.
44. H. M. Wilson, D. J. Ahirrao, S. Raheman Ar and N. Jha, Biomass-derived porous carbon for excellent low intensity solar steam generation and seawater desalination, Sol. Energy Mater. Sol. Cells, 2020, 215, 110604.
45. S. Wang, Y. Fan, F. Wang, Y. Su, X. Zhou, Z. Zhu, H. Sun, W. Liang and A. Li, Potentially scalable fabrication of salt-rejection evaporator based on electrogenerated polypyrrole-coated nickel foam for efficient solar steam generation, Desalination, 2021, 505, 114982.
46. A. Malan and H. R. Sharma, Assessment of drinking water quality and various household water treatment practices in rural areas of Northern India, Arabian J. Geosci., 2023, 16, 96.
47. L. Zang, C. Finnerty, S. Zheng, K. Conway, L. Sun, J. Ma and B. Mi, Interfacial solar vapor generation for desalination and brine treatment: Evaluating current strategies of solving scaling, Water Res., 2021, 198, 117135.
48. F. Pontius, Treatability of a Highly-impaired, saline surface water for potential urban water use, Water, 2018, 10, 324.
49. P. Onsri, D. Dechtrirat, P. Nooeaid, A. Eiad-ua, P. Amornpitoksuk, S. Techasakul, A. Taufiq and L. Chuenchom, Novel magnetically interconnected micro/macroporous structure of monolithic porous carbon adsorbent derived from sodium alginate and wasted black liquor and its adsorption performance, J. Renewable Mater., 2021, 9, 1059–1074.
50. K. H. Nelke, M. Mulak, K. Luczak, W. Pawlak, J. Nienartowicz, D. Szumny, M. Kochman and H. Gerber, Occurrence and exposure to lead and cadmium and their environmental influence on eyesight, Pol. J. Environ. Stud., 2015, 24, 1491–1496.
51. B. Maleki, M. Baghayeri, M. Ghanei-Motlagh, F. Mohammadi Zonoz, A. Amiri, F. Hajizadeh, A. Hosseinifar and E. Esmaeilnezhad, Polyamidoamine dendrimer functionalized iron oxide nanoparticles for simultaneous electrochemical detection of Pb²⁺ and Cd²⁺ ions in environmental waters, Meas.: J. Int. Meas. Confed., 2019, 140, 81–88.
52. E. T. Ulset, P. Kosinski, Y. Zabednova, O. V. Zhdaneev, P. G. Struchalin and B. V. Balakin, Photothermal boiling in aqueous nanofluids, Nano Energy, 2018, 50, 339–346.
53. H. Lim and S. K. Kim, An easily scalable, durable, and highly efficient three-dimensional solar evaporator inspired by a rice paddy field, Desalination, 2023, 548, 116251.
54. Z. Ai, Y. Zhao, R. Gao, L. Chen, T. Wen, W. Wang, T. Zhang, W. Ge and S. Song, Self-assembly hierarchical binary gel based on MXene and montmorillonite nanosheets for efficient and stable solar steam generation, J. Cleaner Prod., 2022, 357, 132000.
55. M. M. Ghafurian, H. Niazmand, E. Ebrahimnia-Bajestan and R. A. Taylor, Wood surface treatment techniques for enhanced solar steam generation, Renewable Energy, 2020, 146, 2308–2315.
56. L. Yi, S. Ci, S. Luo, P. Shao, Y. Hou and Z. Wen, Scalable and low-cost synthesis of black amorphous Al-Ti-O nanostructure for high-efficient photothermal desalination, Nano Energy, 2017, 41, 600–608.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ew00094c

---