Flexible and washable CNT-embedded PAN nonwoven fabrics for

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40...
0 downloads 0 Views 1MB Size
Subscriber access provided by Nottingham Trent University

Energy, Environmental, and Catalysis Applications

Flexible and washable CNT-embedded PAN nonwoven fabrics for solar-enabled evaporation and desalination of seawater Bo Zhu, Hui Kou, Zixiao Liu, Zhaojie Wang, Daniel K. Macharia, Meifang Zhu, Binhe Wu, Xiaogang Liu, and Zhigang Chen ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b12806 • Publication Date (Web): 30 Aug 2019 Downloaded from pubs.acs.org on August 30, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Flexible and washable CNT-embedded PAN nonwoven fabrics for solar-enabled evaporation and desalination of seawater Bo Zhu,† Hui Kou,† Zixiao Liu,† Zhaojie Wang,†Daniel K. Macharia,† Meifang Zhu,† Binhe Wu,*,‡ Xiaogang Liu §, Zhigang Chen*,† † State

Key Laboratory for Modification of Chemical Fibers and Polymer Materials, Shanghai

Belt and Road Joint Laboratory of Advanced Fiber and Low-Dimension Materials, College of Materials Science and Engineering, Donghua University, Shanghai 201620, China. ‡Department

of Applied Physics, Donghua University, Shanghai 201620, China.

§Department

of Chemistry, National University of Singapore, 117543, Singapore.

ABSTRACT : Nanostructured photothermal membranes hold the great capacity for solar-driven seawater desalination, however, their pragmatic applications are often limited by substantial salt accumulation. To solve this issue, we have designed and prepared flexible and washable CNT-embedded polyacrylonitrile (PAN) nonwoven fabrics by a simple electrospinning route. The wet fabric exhibits a strong photoabsorption in a wide spectral range (350-2500 nm), and it has a photoabsorption efficiency of 90.8%. When coated onto a PS foam, the fabric shows a high seawater evaporation rate of 1.44 kg m-2 h-1 under simulated sunlight irradiation (1.0 kW m-2). With a high concentration of simulated seawater as the model, the accumulation of solid salts can be clearly observed on the surface of the fabric, resulting in a severe decay of the evaporation rate. These salts can be effortlessly washed away from the fabric through a plain hand-washing process. The washing process has a negligible influence on the morphology/photoabsorption/evaporation performance of the 1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

fabric, demonstrating good durability. More importantly, a larger fabric can easily be fabricated, and the combination of washable fabrics with various parallel PS foams can facilitate the construction of large-scale outdoor evaporation devices, conferring the great capacity for efficient desalination of seawater under natural sunlight. KEYWORDS: Solar-driven seawater desalination; Photothermal fabric; Electrospinning; Polyacrylonitrile; Multiwall carbon nanotubes.

1. INTRODUCTION The shortage of fresh and clean water is one of the most serious and pressing global challenges1-2. To solve this issue, seawater desalination has become one of the most promising strategies, since approximately 75% surface area of the Earth is covered by seawater 3. Up to now, there are various methods of seawater desalination such as membrane distillation

4-5,

reverse osmosis

6-7,

electrodialysis

8-9

and solar-driven evaporation

10-27.

Among these methods, solar-driven evaporation has drawn tremendous attention because of its unique characteristics such as low-cost, free-energy consumption and non-pollution. The prerequisite for solar-driven evaporation is to develop photothermal materials with broadband and strong photoabsorption as well as high-efficiency devices for evaporation. In previous studies, several inorganic nanoparticles (Au nanoparticles

28,

Cu7S4

nanoparticles 11 and Fe3O4/C 29) have been developed as photothermal materials, and they are directly dispersed in water. Under the irradiation of sunlight, photothermal nanoparticles in the bulk seawater absorb the light and convert it to heat, conferring the uniform bulk heating and then the formation of watersteam bubbles. However, their evaporation rates are 2

ACS Paragon Plus Environment

Page 3 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

unsatisfactory, since the non-evaporative portion of the solution is also heated. To avoid uniform bulk heating, inorganic nanoparticles can be concentrated at seawater-air interface to form photothermal membranes. Photothermal membranes with single-layer structures have been firstly prepared, such as assembled films based on Au nanoparticle graphene

21

and semiconductor nanoparticles

12, 22.

15,

polymer

20, 27,

These membranes can be floated on the

water surface, and they exhibit obviously enhanced evaporation rates under sunlight irradiation compared to the bulk heating model. Even so, there are still severe and unavoidable heat-losses from these single-layer membranes to bulk water. To reduce the heat loss, double/multi-layer photothermal membranes have been proposed, where the upper layer is the photothermal part and the down layer is thermal insulator part. For example, several bilayer membranes and multi-layer membranes (graphene/glass fiber/polystyrene foam 23, and graphene oxide/ cellulose/polystyrene 18, polystyrene foam/cermet/transparent bubble wrap 13 and exfoliated graphite/carbon foam 25) have been well developed, and they exhibit excellent solar utilization efficiency and evaporation rate, conferring a great progress in solar water evaporation. It should be pointed out that during most evaporation processes, the deionized water is used as the model to evaluate the evaporation rate of photothermal membranes, without salt separation/accumulation phenomenon10, 12, 30. Only in a few reports

21, 23,

seawater have been

evaporated by the photothermal membranes, but the salt separation/accumulation issues are not considered or addressed in details. In practical thermal desalination systems, apparent salt separation/accumulation can be found, which shields the incident sunlight and damage the nanostructure of the photothermal membranes. These solid salts may be removed by washing 3

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

the membranes, but many photothermal membranes (such as Ti2O3/cellulose membrane graphene/cellulose membrane

19,

and graphene aerogel

21, 24)

12,

are easy to be damaged during

the washing process due to the exfoliation of nanoparticles and the absence of large micro-sized channels. To facilitate the washing and regeneration, it is very necessary to develop novel photothermal membranes with large micro-sized channels and without the possible exfoliation of the photothermal nanomaterials. As we all know, clothes have the adjustable texture and micro-sized pores, and the soiled clothes can easily be washed with good durability for the removal of contaminants. These features triggered our interest in the development of flexible, washable photothermal fabrics with micro-sized channels and with the encapsulation of photothermal nanomaterials. These fabrics were prepared by electrospinning the precursor solution containing polymer and photothermal nanomaterials, where the polyacrylonitrile (PAN) was used as the polymer matrix, and carbon nanotubes (CNTs) was used as the model of photothermal materials. CNTs-embedded PAN non-woven fabric consisted of nanofibers (diameters: 200-300 nm) and hierarchical pores, and the wet fabric exhibited strong (>90%) photoabsorption in a wide range (350-2500 nm). When the fabric with PS foam as the heat-insulation layer was floated on the surface of seawater, the evaporation rate was 1.44 kg m-2 h-1 under simulated sunlight irradiation (1 kW m-2), coupled with the accumulation of solid salts which subsequently deteriorated the evaporation performances. These solid salts were quickly removed through a plain hand-washing process, while fabrics retained their excellent color/structure/performance stability even after multiple washing processes. 2. MATERIALS AND METHODS 4

ACS Paragon Plus Environment

Page 5 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

2.1. Preparation of non-woven fabric. Pristine CNTs were oxidized by using mixture solution containing concentrated sulfuric acid and nitric acid according to a previous report

31

(the detail is shown in section 1.2 in

supporting information), resulting in the obvious improvement of the dispersion in N,N-dimethylformamide (DMF) (Figure S1). The oxidized CNTs with different weights (3.5, 7 or 17.5 mg) and PAN (350 mg) were added into DMF (5 mL), and the DMF dispersion was stirred at 60 °C for 12 h. By using DMF dispersion (5mL) with different CNT/PAN ratio (1, 2, 5wt%) as the precursor solution, PAN-CNT non-woven fabrics were prepared by an electrospinning method (voltage: 12 kV; spinning distance: 15 cm; feeding rate: 0.5 mL/h), and they are respectively denoted as CNT@PAN-1%, CNT@PAN-2%, CNT@PAN-5%. Taking DMF dispersion with 2wt% CNT/PAN as a model, the fabrics with different thickness were prepared by increasing volume from 5 to 15 mL. To investigate the effects of electrospinning process, different electrospinning parameters (spinning distance: 10-20 cm, feeding rate: 0.25-0.75 mL/h) were used to prepare CNT@PAN-2% fabrics, and the details are shown in Table S1. During the electrospinning process, the ambient temperature was kept at about 40 °C and the relative humidity was 30 ~ 35 %. For comparison, pure PAN non-woven fabric was also prepared in the absence of CNTs under the other identical conditions. 2.2. Characterizations. CNT and nonwoven fabric samples were analyzed by field electron scanning microscope (FE-SEM, Hitachi S-4800), transmission electron microscope (TEM, FEI Talos F200S). The thermal conductivities of CNT@PAN fabrics at dry and wet state were investigated through 5

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

sandwiching the fabric between two glass slides. The details are provided in the Supporting Information. 2.3. Solar-enabled evaporation and desalination of seawater. Indoor desalination: CNT@PAN nonwoven fabric was cut to have the size of 4 × 4 cm2. Its four corners with the identical squares (2 × 2 cm2) were cut off, forming a cross-like fabric. Subsequently, the cross-like fabric was wrapped on PS foam (size: 2 × 2 cm2, thickness: ~ 5 mm). The fabric-coated PS foam was implanted in a hole (size: ~ 2 × 2 cm2) at the center of a big rounded PS foam (diameter: 15 cm, thickness: 5 mm). To construct the evaporation device, the entire PS foam was floated on the simulated seawater (100 mL) contained in a beaker (100 mL), and the simulated seawater surface was fully covered by this PS foam. Under the illumination of simulated sunlight (Beijing Perfect Light Co. Ltd, Beijing) with different light intensities (such as 1, 3, or 5 kW m-2), the mass change of the whole device was recorded every 10 min by an electronic mass balance with an accuracy of 0.001g. At the same time, the surface temperature of the non-woven fabric was recorded by an IR camera (A300; FLIR Systems Inc.). The optical intensity of the light simulator was investigated by a Hand-Held Optical Meter Model (Newport 1918-C). The concentrations of ions in the simulated seawater and the condensed water collected during the evaporation process were determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES, Prodigy; USA). The condensed water was collected by putting the evaporation device on an inclined glass container. The steam condensed into fresh water on the inside surface of the container, then the fresh water was gradually accumulated and gathered at the bottom of container. Notably, all the indoor evaporation experiments were conducted at an ambient temperature of ~ 25 °C 6

ACS Paragon Plus Environment

Page 6 of 32

Page 7 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

and humidity of ~ 50%. Outdoor desalination: The outdoor experiment was carried out on the roof in Donghua University at daytime (8:00-18:00). A transparent inclined glass device was designed, and its cross section was a right-angled trapezoidal, where the length was 42 cm, the width was 45 cm, the left height was 15 cm, and the right height was 60 cm. In the interior bottom of the glass device, there was an electronic mass balance (1g in accuracy). A rectangular tank (28 × 35 × 10 cm3) with simulated seawater (3 kg, 3.5 wt% NaCl) was put on the mass balance. Subsequently, three parallel PS foams (9 × 30 × 1 cm3) were floated on the seawater with a gap of 0.7 cm. Then, the surface of PS foams was coated by the fabric (45 × 30 cm2). To ensure the efficient capillary absorption of seawater, a part of the fabric at the margin or gap of PS foams were soaked in the simulated seawater. Under natural sunlight illumination, the mass change of the whole polypropylene sink was recorded, and the internal and external temperature of the device was recorded by a thermometer. The light intensity of the natural sunlight was measured by a Hand-Held Optical Meter Model.

3. RESULTS AND DISCUSSION It is well known that when the clothes with adjustable texture and pores are polluted, they can easily be washed to remove any contaminations, without obvious change in the color and shape (Figure 1a). Inspired by these features, in the present study, we designed and prepared the flexible CNT-embedded PAN (CNT@PAN) non-woven fabrics for solar-driven seawater desalination. After the seawater evaporation, the accumulated solid salts on the fabric can easily be washed away through a plain hand-washing process due to the presence of 7

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

large micro-sized pores as channels (Figure 1b), and meanwhile, CNTs will not be exfoliated from the fabric due to the encapsulation by PAN. For the washing process, the desalted CNT@PAN non-woven fabric is immersed in simulated seawater (3.5 wt % NaCl solution) or real seawater (East China Sea) in a beaker. Then, the fabric is shaken in water to dissolve the solid salt (Movie S1 in supporting information).

Figure 1. Schematic illustration of washing clothes (a) and the process of removing the salt in non-woven fabric (b).

Figure 2. Electrospinning preparation process (a) and the typical photo of CNT@PAN-2% fabric (b). Herein, CNT@PAN non-woven fabrics were prepared by electrospinning the PAN/CNT precursor solution (Figure 2a) with different CNT concentrations (1, 2 and 5 wt%) and volumes (5, 10, 15 mL). The thickness of each fabric was determined by a micrometer. With the increase of solution volume from 5 to 10 and 15 mL, the fabric prepared with 2 wt% CNT 8

ACS Paragon Plus Environment

Page 8 of 32

Page 9 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

solution have the increased thickness from 207 ± 18 μm to 423 ± 28 μm and 605 ± 25 μm, and can be abbreviated as CNT@PAN-2%-207, CNT@PAN-2%-423 and CNT@PAN-2%-605 fabrics. These fabrics with different thickness appear a similar color (Figure S2). In addition, when CNT concentration increases from 1 to 5 wt% with the same volume (5 mL), these CNT@PAN fabrics have a similar thickness (~207±18 μm) and X-ray diffraction peaks from 10 to 80 (Figure S3), but their color changes from gray-black to deep-black (Figure S4 and Figure 2b) It should be noted that the nonwoven cloths are able to be fabricated in a large-scale, and their area can be well adjusted in a wide range (10-4 to 100 m2) by controlling the collection area and needle numbers. In the present work, CNT@PAN fabrics with area of ~200 cm2 or ~900 cm2 were respectively prepared (Figure S5). The morphology of CNT@PAN non-woven fabrics were investigated by SEM. For comparison, pure PAN non-woven fabric was also investigated. SEM image (Figure 3a) shows that pure PAN fabric consists of nanofibers with diameters of ~ 300 ± 50 nm (Figure S6a) and micro-sized pores (size: 1-5 μm). PAN-CNT fabrics have similar microcosmic morphologies when changing CNT concentration (Figure S7), the spinning distance or the feeding rate (Figure S8). With CNT@PAN-2%-207 nonwoven fabric as a model, SEM image (Figure 3b) reveals that it is also composed of plenty of single filament nanofibers with smooth surfaces. But the diameters of CNT@PAN nanofibers exhibit a slight decrease to ~ 250 ± 50 nm (Figure S6b-d), resulting from the strong electrostatic repulsion among jet sprays due to the increased conductivity by CNT in the precursor solution

32.

Simultaneously, these CNT@PAN

nanofibers are interwoven and form a three-dimensional network structure in the non-woven fabric, resulting in the natural formation of many hierarchical micro-sized pores with a 9

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

diameter of 1-5 μm. Such plenty of micro-sized pores will result in the high porosity of the fabric (Table S2), which serves as the transport path of seawater in the seawater desalination process, and also facilitate the removal of solid salts in the washing process.

Figure 3. SEM image of pure PAN fabric (a). SEM (b) and TEM (c,d) images of typical CNT@PAN-2% fabric. To evaluate the presence and distribution of CNTs, CNT@PAN-2% fabric was investigated by TEM. Low-magnification TEM image (Figure 3c) indicates that the fabric consists of nanofibers with diameters of 200 ~ 300 nm, which agrees well with SEM results (Figure 3b). Typical high-magnification TEM image (Figure 3d) of single nanofiber reveals that there is a single CNT with a diameter of ~ 15 nm inside the fiber, suggesting the well encapsulation of CNT by PAN. Undoubtedly, this encapsulation can prevent the exfoliation of CNT from the fabric during the seawater desalination and washing process, ensuring good performance stability. For the continuous and efficient evaporation of seawater, photothermal membranes 10

ACS Paragon Plus Environment

Page 10 of 32

Page 11 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

should have good hydrophilicity. To measure the hydrophilicity, the water contact angle of the non-woven fabric was studied. When a tiny water droplet (2 μL) is placed on the surface of CNT@PAN fabrics with different CNT contents (1, 2 and 5wt%), the contact angle goes down from 107-110° at 0 s to nearly 0° at 7s (Figure 4a, Figure S9). This fact reveals that the water droplet can rapidly spreads out and infiltrates into the fabric, suggesting a high affinity of the fabrics to water. Such hydrophilicity ensures the continuous seawater supply for solar-driven evaporation. Interestingly, the wetting process by water confers an obvious change from grey-black or black color of dry fabrics to dark black of wet fabrics (Figure S2, Figure S4 and Figure 4b).

Figure 4. (a) Water contact angle of CNT@PAN-2% fabric at 0 s (left) and 7 s (right). (b) Photos of dry (left) and wet (right) CNT@PAN fabric. (c) Absorption spectra of pure PAN, dry typical CNT@PAN fabric with different CNT content (1, 2, 5wt%). (d) Absorption spectra of typical wet CNT@PAN fabric with different CNT contents (1, 2, 5wt%). (e) Surface temperature changes of dry PAN and CNT@PAN fabrics with different CNT contents (1, 2, 5wt%) under simulated sunlight irradiation (1kW m-2). 11

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

A strong and broad photoabsorption is the prerequisite for efficient seawater evaporation. Furthermore, the optical properties of different dry CNT@PAN fabrics were measured by a UV–vis-NIR spectrophotometer. For comparison, dry pure PAN fabric was also analyzed. PAN fabric has very low absorbance (< 20%) in the entire UV-vis-NIR region (350-2500 nm). While all dry CNT@PAN fabrics have the enhanced photoabsorption in a wide range (350-2500 nm) (Figure 4c), due to the presence of CNTs with excellent photoabsorption

33.

The thickness of CNT@PAN-2% nonwoven fabrics in the range of

200-600 μm has no apparent effects on photoabsorption (Figure S10), probably resulting from that 200 μm thickness is adequate. In addition, the electrospinning parameters also have a negligible effect on the photoabsorption of CNT@PAN-2% (Figure S11). When CNT concentration in the precursor solution increases from 1 to 2 and 5 wt%, the absorbance of dry CNT@PAN fabric goes up rapidly in the entire UV-vis-NIR region, and the total solar absorbance efficiency of the dry fabrics are calculated to be 65.2, 72.6, 81.7%. Since wet fabrics become deeper black, we further investigated the photoabsorption of wet fabrics (Figure 4d). Interestingly, all these wet fabrics exhibit the enhanced absorbance compared to dry ones (Figure 4c, d). For example, the average absorbance goes up from 72.6 % for dry CNT@PAN-2% to 90.8% for wet CNT@PAN-2%, and wet CNT@PAN-2% fabric has similar photoabsorption compared with wet CNT@PAN-5% fabric. The darker color and increase in photoabsorption of wet fabric should be attributed to the decrease in the reflectivity from the air-fabric interface to air-water-fabric interface, which is well demonstrated in the previous report 34. Therefore, the present CNT@PAN non-woven fabric has excellent hydrophilicity, and the wet fabric exhibits strong photoabsorption in broad 12

ACS Paragon Plus Environment

Page 12 of 32

Page 13 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

UV-vis-NIR region (350-2500 nm). Wide and strong photoabsorption usually induces the excellent photothermal conversion performance. To evaluate the photothermal performance, the surface temperatures of dry/wet PAN-CNT fabric as well as pure PAN fabric were respectively recorded by an IR camera, under simulated sunlight irradiation (1.0 kW m-2) (Figure 4e, Figure S12). Obviously, pure PAN fabric shows a negligible increase in the surface temperature, such as by ~ 3.3 °C at 200 s of irradiation (Figure 4e). With the presence of CNT (1-5 wt%), the surface temperature of dry PAN-CNT fabric goes up rapidly from 0 to 30 s under irradiation and then exhibits a relative plateau with small fluctuations (± 2 °C). Importantly, with the increase of CNT content from 1 wt% to 2 wt% and 5 wt%, the temperature elevation (ΔT), which is calculated from Figure 4e, goes up from ~ 26 °C to 32 °C and ~ 42 °C at 200s. This large temperature elevation should result from the enhanced photoabsorption of the CNT content and also confirms that CNT in fabrics can rapidly and efficiently convert light to thermal energy. In addition, the wetting of PAN-CNT fabric can confer the decrease of temperature elevation. For example, the wet CNT@PAN-2% fabric and CNT@PAN-5% fabric exhibit a similar temperature elevation of ~16 °C at 200s (Figure S12), due to similar photoabsorption, high thermal conductivity (~ 0.6 W m-1 K-1) of water and water-evaporation-induced heat loss 35. For practical applications in seawater desalination, all CNT@PAN fabric will be wet. Since the wet CNT@PAN-2% and CNT@PAN-5% fabrics have similar photothermal properties, CNT@PAN-2% fabric is used as a model for the future characterization, taking cost into account. It is well known that the thermal conductivity of photothermal membranes has serious 13

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

effects on the thermal transfer19,

25.

Herein, we investigated the thermal conductivity of

CNT@PAN-2% fabric at a dry and wet state by sandwiching fabrics between two glass slides with an IR camera. The dry fabric shows an obvious temperature gradient along the thickness of the fabric layer (inset of Figure 5a). Its thermal conductivity is determined to be 0.136 W m-1 K-1 (Figure 5a) which is lower than that of pure PAN film (~ 0.26 W m-1 K-1) 36, since the dry fabric is porous (Table S2) and has plenty of air with low thermal conductivity of 0.024 W m-1 K-1. Once CNT@PAN-2% fabric is soaked with water, the wet fabric shows an inconspicuous temperature gradient (inset of Figure 5b). The thermal conductivity of the wet fabric is 0.418 W m-1 K-1 (Figure 5b) which is lower than that of pure water (0.6 W m-1 K-1) 35 and pure CNTs (~ 3000 W m-1 K-1)

37,

due to the presence of PAN matrix as the thermal

insulator. It should be noted that the thermal conductivity (0.418 W m-1 K-1) of wet fabric is nearly 3 times that (0.136 W m-1 K-1) of dry fabric, due to the high thermal conductivity (0.6 W m-1 K-1) of water in the pores of wet fabric.

Figure 5. Thermal conductivity of CNT@PAN-2% fabric at dry (a) and wet (b) state. The insert in the picture is the representative picture taken by the IR microscope. Obviously, large thermal conductivity will induce the rapid heat dissipation from the photothermal membranes to bulk water, which will reduce the evaporation rate of seawater. It 14

ACS Paragon Plus Environment

Page 14 of 32

Page 15 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

has been demonstrated that PS foam has a very low thermal conductivity (~ 0.04 W m-1 K-1 ) 38

and can serve as the heat-insulation layer for the photothermal membrane (such as GO film

18,

graphene sheets membrane 23, carbon/nonwoven paper 39) to enhance the evaporation rate

of seawater. To reduce the heat loss, herein we also used a cross-like CNT@PAN-2% fabric to coat a PS foam (2 × 2 cm2), and then the fabric-coated PS foam was floated on the surface of seawater, as vividly shown in the photos of indoor evaporation device (Figure 6a). It should be pointed out that the contact area between the fabric and water is only at the edge of PS foam, which will ensure a continuous supply of seawater for evaporation and simultaneously suppresses the heat dissipation from the bulk water. Under the simulated sunlight irradiation (1 kW m-2), the surface temperature of the fabric with or without PS foam substrate was captured by an IR camera (Figure 6b, c). In the absence of PS foam substrate, the fabric temperature goes up slowly from 25.0 °C at 0 s to 32.0 °C at 100 s, and then maintains nearly a constant value (32.0 ± 1 °C) during 100-200 s. Importantly, with PS foam substrate, the fabric temperature increases rapidly from 25.0 °C at 0 s to ~ 43.0 °C at 100 s, and then maintians a constant temperature at ~ 43.0 ± 1.5 °C with the further increase of the irradiation time to 200 s. The temperature elevation (ΔT: ~ 18.0 °C) in the presence of PS foam is ~ 2.5 times that (ΔT: ~ 7.0 °C) without PS foam. To further investigate the temperature distribution, COMSOL Multiphysics software was used to simulate the above evaporation system through a basic mass transfer model (the details are shown in supporting information). The simulated results show that the maximum temperature at fabric surface is about 32.4 °C for the fabric without PS foam and 42.6 °C for that with PS foam (Figure 6d), which matches well with the experimental results (Figure 6b,c). Thus, both experimental and 15

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

theoretical results demonstrate that the addition of PS foam substrate confers the great enhancement of temperature elevation of CNT@PAN-2% fabric under simulated sunlight irradiation. This enhancement should be attributed to that PS foam with low thermal conductivity can effectively suppress the heat dissipation from the fabric to the bulk water, which result in the efficient heating at the wet fabric–air interface. Undoubtedly, this interfacial heating facilitates the effective evaporation of seawater absorbed in the wet fabric.

Figure 6. (a) Fabrication process of the indoor solar-driven evaporation device. The temperature change (b) and the balanced temperature (c) of CNT@PAN-2% fabric without or with PS foam after 200 s irradiation under simulated sunlight (1 kW m-2). (d) Simulated temperature distribution of CNT@PAN-2% fabric without or with PS foam at steady state. To investigate the evaporation performance, the mass change of simulated seawater (3.5 wt% NaCl solution) was measured by using CNT@PAN fabric-coated PS foam under simulated sunlight irradiation with different light intensities (1, 3, or 5 kW m-2) (Figure 7a). For comparison, the natural evaporation of simulated seawater was also studied in the dark or under light irradiation. The weight loss of simulated seawater in the dark is determined to be 0.225 kg m-2 at ~ 25 °C in 60 min, corresponding to a low natural evaporation rate of 0.225 kg 16

ACS Paragon Plus Environment

Page 16 of 32

Page 17 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

m-2 h-1 in dark (Figure S13). Under one sun illumination (1 kW m-2), the weight loss rate due to natural evaporation increases to 0.64 kg m-2 h-1 (Figure 7a), which is similar to the previous reports

23-24.

Importantly, when CNT@PAN-2% fabric-coated PS foam is used to evaporate

seawater, the evaporation rate goes up to 1.44 kg m-2 h-1, which is comparable to previous results (such as 1.50 kg m-2 h-1 for Cu MoF 40, 1.38 kg m-2 h-1 for PPy/ airlaid paper 41; 1.46 kg m-2 h-1 for carbonization wood

42).

In addition, both the electrospinning parameters (feeding

rate, spinning distance, Figure S14) and the thickness (~207, ~423, ~605 μm, Figure S15) of CNT@PAN-2% fabric have no obvious effect on the weight loss of simulated seawater, due to their similar photoabsorption in the entire range (350-2500 nm, Figure S10 and Figure S11). It should be noted that with the increase of CNT content from 1 wt% to 5 wt%, the CNT@PAN fabric has the enhanced weight loss of seawater in 60 min and thus improved evaporation rate (from 1.28 to 1.44 kg m-2 h-1, Figure S16), resulting from higher photoabsorption with increasing CNT content (Figure 4c,d). These facts confirm that CNT@PAN-2% fabric have good evaporation rate due to the superior photoabsorption and photothermal effect of CNT@PAN fabric as well as high heat-insulation by PS foam.

Figure 7. The evaporation rate (a) and thermal conversion efficiency (b) at different optical densities for CNT@PAN-2% fabric. 17

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

Furthermore, we further investigated the effects of light intentisy on the evaporation rate and efficiency by CNT@PAN-2% fabric-coated PS foam. When the light intensity increases from 1 to 3 and 5 sun, the evaporation rate increase to 4.12 and 6.61 kg m-2 h-1 (Figure 7a), respectively. Meanwhile, the solar energy conversion efficiency is calculated by the following equation: 𝑚ℎ𝐿𝑉

𝜂 = 𝐶𝑜𝑝𝑡𝑃0

(1)

where 𝑚 represents the evaporation rate, ℎ𝐿𝑉 is the total latent enthalpy of the liquid-vapor phase change of water, 𝐶𝑜𝑝𝑡 denotes the optical concentration, 𝑃0 is the normal solar radiation intensity (1 kW m-2). It should be noted that ℎ𝐿𝑉 changes under different temperature. For example, the ℎ𝐿𝑉 should be 2438.5 KJ kg-1 at 26.4 °C (Figure S17) for 3.5 wt% NaCl solution under 1sun, 2400.1 KJ kg-1 at 43.5 °C (Figure 6b) for CNT@PAN-2% fabric under 1 sun (1 kW m-2), 2357.9 KJ kg-1 at 60.0 °C (Figure S18) for CNT@PAN-2% fabric under 3 sun (3 kW m-2), and 2332.8 KJ kg-1 at 72.0 °C (Figure S18) for CNT@PAN-2% fabric under 5 sun (5 kW m-2). The calculation details are shown in the Spporting Information. The solar energy efficiency is calculated to be 28.1% for simulated seawater only (3.5 wt% NaCl) under the irradiation of 1 sun. With the presence of the CNT@PAN-2% fabric-coated PS foam, the solar energy efficiency goes up to 81.0% at 1 sun, 85.0% at 3 sun and 82.8% at 5 sun. (Figure 7b). It should be noted that solar irradiation with high intensity induces relatively high efficiency and evaporation rate, but it is difficult to achieve such strong solar irradiation in a natural environment. Thus, the efficiency at 1 sun is more significant for the practical applications. The evaporation efficiency (81.0%) at 1 sun is much higher or comparable to previous reports (Table S3), such as bilayer structure carbon 18

ACS Paragon Plus Environment

Page 19 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

foam with exfoliated graphite (64%)25, bilayer wood (57.3%)43, and plasmonic wood (70 %)44, graphene aerogel (83%)24. Thus, the above findings demonstrate that CNT@PAN fabric is a superior photothermal material for solar-driven seawater desalination. Normally, the dissolved salt in the seawater will be crystallized and separated out in the case of the salt concentration beyond the saturation point. In fact, the solid salt (NaCl, KCl, etc) accumulation on the photothermal membrane is inevitable during the process of solar-driven desalination of real seawater after a long time21, 39. To observe the accumulation of solid salt, the simulated seawater with a high concentration (21 wt% NaCl solution) was evaporated by CNT@PAN-2% fabric-coated PS foam under the simulated light irradiation (1kW m-2). Before the irradiation, the fabric exhibits a black surface without solid particles (Figure 8a). After just 1 h of irradiation, there is a thin-layer salt crystal on the fabric surface. With further increase of the illumination time to 2-5 h, the accumulation of solid salt appears evidently from the edge to the center of fabric (Figure 8a). Correspondingly, the effects of accumulation of solid salt on the evaporation rate were investigated. Obviously, the evaporation rate declines from 1.44 kg m-2 h-1 at 1 h to 0.65 kg m-2 h-1 after 5 h (Figure 8b). Such a great decrease (55%) should be attributed to the formation of solid salt, which can not only blocks the incident light but also impedes the vapor transport from the internal membrane to the external. Based on these facts, the severe accumulation of solid salt can be confirmed and it seriously deteriorates the evaporation rate from CNT@PAN fabric-coated PS foam.

19

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8. Photographs (a) and evaporation rates (b) of CNT@PAN-2% fabric during 5-hour evaporation of 21 wt% NaCl solution under simulated sunlight (1 kW m-2).

Figure 9. Photograph (a) and SEM image (b) of CNT@PAN-2% fabric after washing. (c) Photoabsorption spectra of wet CNT@PAN-2% fabric before and after different washing times (1st, 5th, 10th, 15th). (d) Evaporation rates of CNT@PAN-2% fabric in the cycling test. Each cycle is 1 h. Usually, the salt accumulation has an adverse effect not only on the evaporation rate but also for photothermal membranes. Particularly, separated solid salt (such as NaCl and KCl) may fill the pores and damage the nanostructure of photothermal membranes. Thus, it is necessary to remove the solid salt by washing the photothermal membranes. However, most 20

ACS Paragon Plus Environment

Page 20 of 32

Page 21 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

nanostructured photothermal membranes (Au film aerogel

21)

15,

GO-based membrane

18

and graphene

cannot be easily washed, since the washing process can destroy these films.

Through the desired design, the present CNT@PAN nonwoven fabric can easily be detached from the PS foam and then be washed through a plain hand-washing process to remove the solid salt, as vividly demonstrated in Figure 1and Movie S1. After the washing, the fabric can be effortlessly re-coated on PS foam for the next evaporation of seawater. Then, the influence of the washing process on the morphology, photoabsorption, and evaporation performance of CNT@PAN nonwoven fabric was investigated. When the CNT@PAN-2% fabric is re-used and washed for 15 times, the washed fabric retains its black color with a smooth surface (Figure 9a), which is similar to the initial fabric (Figure 4b, 6a). SEM image (Figure 9b) shows that the washed fabric is still composed of nanofibers with a diameter of ~ 250 ± 50 nm, which is also similar to the SEM image (Figure 3b) of the initial fabric. In addition, after being washed for different times (1-15), the wet fabric was investigated by UV-vis-NIR spectrometer. The wet fabrics washed for 1-15 times still maintain a wide and strong absorbance (90.3~ 92.5%) in the UV-vis-NIR region (350-2500 nm) (Figure 9c). And this photoabsorption is analogous to that (Figure 4d) of the unwashed wet fabric. Furthermore, the leachate, which is used to wash fabrics for different times (1st, 5th, 10th and 15th), was collected and then measured by UV-vis-NIR spectrophotometer. All these leachates show no absorption in entire UV-vis-NIR region (300-1100 nm) (Figure S19), which indicates that CNT cannot leach out from the fabric during the washing process due to the well encapsulation of CNT by PAN (Figure 3d). In addition, the PS foam was re-coated with the washed CNT@PAN fabric to re-evaporate the simulated seawater (3.5 wt% NaCl solution) 21

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 32

under 1sun for 1 h in each cycle, and the evaporation rate was also calculated for each cycle (Figure 9d). Obviously, the evaporation rate remains to be 1.4 ± 0.04 kg m-2 h-1 during 15 cycle test times, without any decay in the rate. Based on the above results, CNT@PAN nonwoven fabric can easily be washed to remove solid salts, and there is no obvious adverse effects from washing step for the morphology and photoabsorption/evapor- ation performance of fabrics. To systematically investigate the desalination effect, the actual seawater (from East China Sea) and the simulated seawater (3.5 wt% NaCl solution) were evaporated by CNT@PAN-2% fabric-coated PS foam under the irradiation of simulated sunlight (1 kW m-2). The freshwater was collected by condensing the watersteam, and the ion concentrations in the freshwater and simulated seawater were measured by inductively coupled plasma atomic emission spectroscopy (ICP-AES). The original simulated seawater has a high Na+ concentration of ~ 1.407×104 mg L-1. Importantly, Na+ concentration in the fresh water drops greatly to ~ 1.08 mg L-1, which is about four orders of magnitude decline (Figure 10a). This Na+ concentration (~ 1.08 mg L-1) is far below the salinity level (such as Na+< 200 mg L-1) of drinking water as defined by the World Health Organization (WHO)

45.

In addition, after

evaporation of real seawater, we also investigated the concentration of five main ions (Na+, Mg2+, Ca2+, K+, B3+) in the original actual seawater and the resulting clean water. Before desalination, the ion concentration in the real seawater is very high, such as ~ 1.75 ×103 mg L-1 for Na+, ~ 4.63×102 mg L-1 for Mg2+, ~ 1.95×102 mg L-1 for Ca2+, ~ 1.71×102 mg L-1 for K+ and ~ 1.47 mg L-1for B3+ (Figure 10a). After the desalination, all concentrations of these ions in the resulting fresh water are below 1 mg L-1, with very high ion rejection, >99.6% for Na+, 22

ACS Paragon Plus Environment

Page 23 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Mg2+, Ca2+, K+, and 86.7% for B3+ (Figure 10b). Therefore, the present CNT@PAN nonwoven fabric has an excellent desalination effect for producing clean water from seawater.

Figure 10. (a) Salinities (the weight percentage of ion) of simulated seawater (3.5wt% NaCl solution) and primary ion Na+, K+, Mg2+, Ca2+ in actual sea water (East China sea) before (Original) and after (desalinated) solar thermal desalination. (b) Ion rejection of real seawater sample after desalinated.

Figure 11. (a) Typical photo of outdoor evaporation device. (b) Schematic diagram of large-area CNT@PAN-2% fabric for outdoor evaporation test with required number PS foams as insulators. Solar flux/external and the internal temperature of the evaporation device (c) and the evaporation rate/accumulated weight change (d) during the whole daytime.

23

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Besides stability and excellent desalination performances, the cost and scalability are also very significant for the practical applications. Undoubtedly, solar energy is a sustainable source and can irradiate CNT@PAN non-woven fabric to produce the water steam from seawater without much cost/pollution. Furthermore, CNT@PAN non-woven fabric can be prepared by a simple electrospinning technology in large-scale. Herein, we prepared a typical CNT@PAN-2% fabric with a relatively large area (45 × 30 cm2) for outdoor practical evaporation experiment. The practical application experiment was carried out in a tilted transparent condenser (Figure 11a). The large-area fabric was used to coat on the three parallel of PS foams on the surface of seawater (Figure 11b). The fabric at foam edge/gap were soaked in the simulated seawater to absorb and supply sufficient simulated seawater for evaporation. Under natural sunlight illumination, the solar intensity/temperature/evaporation rate/seawater mass-loss were recorded from 08:00 to 18:00. The solar flux exhibits a typical sunny day, and it increases from 0.44 kW m-2 at 8:00 to 0.79 kW m-2 at 14:00 and then decreases to 0.3 kW m-2 at 18:00 (Figure 11c). Similarly, the interior and exterior temperatures of the condenser reach the highest temperature of ~ 47 °C and ~ 35 °C at 14:00, respectively (Figure 11c). Meanwhile, the variation tendency of the evaporation rate is also similar to that of the light intensity. The evaporation rate increases from 0.619 kg m-2 h-1 at 9:00 to 1.04 kg m-2 h-1 at 14:00, and then decreases to 0.43 kg m-2 h-1 at 18:00 (Figure 11d). Furthermore, the total mass loss of seawater increases from 0 at 8:00 to 8.23 kg m-2 at 18:00 (Figure 11d). Therefore, ~8 kg per day freshwater can be produced by a 1 m-2 CNT@PAN-2% fabric, which is sufficient to meet the water demand of 3 individuals. It should be noted that by using a large-area fabric or the combination of fabric with numerous 24

ACS Paragon Plus Environment

Page 24 of 32

Page 25 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

parallel PS foams, such device can easily be enlarged, facilitating the practical applications in the large-scale for efficient outdoor solar-driven seawater desalination. Based on the above results, there are chiefly two advantages for solar-driven seawater desalination over traditional technologies (such as reverse osmosis, electrodialysis and membrane distillation). One is no or low energy consumption, since solar-driven desalination uses mainly solar energy that is abundant, green and renewable for evaporating seawater. On the contrary, traditional seawater desalination technologies have high electric energy consumption, such as ~3.57 kWh/m3 fresh water for reverse osmosis electrodialysis

46,

46,

~1.5 kWh/m3 for

and ~2.27 kWh/m3 for membrane distillation 4. The other is ultra-low ion

concentration (