Cellulose Nanocrystal/Polydimethylsiloxane hybrid membranes for air dehydration at elevated temperatures

doi:10.21203/rs.3.rs-4716356/v1

Download PDF

Research Article

Cellulose Nanocrystal/Polydimethylsiloxane hybrid membranes for air dehydration at elevated temperatures

https://doi.org/10.21203/rs.3.rs-4716356/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

This research study developed a dense composite membrane made of cellulose nanocrystal (CNC) and polydimethylsiloxane (PDMS) to efficiently separate water vapor from air at elevated temperatures up to 80°C. In this study a casting method was used to fabricate CNC/PDMS membranes. The water vapor permeability of the membrane samples was measured with a Payne diffusion cell (dry cup method) coupled with a Dynamic Vapor Sorption (DVS) instrument, while the nitrogen gas permeability was measured with a gas permeation cell. The results showed that the optimal CNC concentration of 2%, enhanced water vapor permeability at all temperatures up to 24.8% while increasing the selectivity slightly up to 3.1%. The membranes were characterized using AFM, FTIR, SEM, and TMA. measured the CTE of the prepared samples to study the dimensional stability as a function of temperature change. The optimized membranes showed an 8.9% lower value for CTE which results in higher thermal dimensional stability of the sample. The results have demonstrated that CNC-reinforced PDMS has potential to be used as selective membranes to remove water vapor from exhaust warm air such that the air recovers its drying capability and can be recirculated as the working medium in drying systems.

CNC

Membrane

Moisture Separation

PDMS

Permeability

Temperature

Recently, membrane separation technology for air dehumidification has become an interesting topic for researchers and industry. Water vapor is separated from the air through a dense membrane without any phase change or temperature change, so this process is also known as an isothermal membrane-based air dehumidification (IMAD) process [1]. IMAD has been applied in many industries, in which require moisture control during the manufacturing process, such as a drying step in food industries, chemical industries, pharmaceuticals and more [2], [3]. Moreover, when it is combined with other units, such as heating, ventilation, and air conditioning (HVAC) units in building systems, pre-air dehumidification could achieve cooling energy saving [4]. Similarly, it could be a potential energy-saving technology in wood drying processes [5]. Wood drying is an energy-demanding process that requires heat to warm up the air which circulates through a stack of lumber to remove its moisture [6]. After absorbing the moisture, the air in dry kilns is saturated and should be vented as exhaust. Fresh and cold air is introduced to the kiln; however, it once again needs extra energy to be heated [7], [8]. In contrast, if applying an air dehumidification membrane system to the dry kiln, it could remove the moisture from saturated air without any phase change, thereby allowing the warm air to be recycled into the system, retaining heat. The dehumidified air, which is still at a high temperature, can return to the drying system to run another cycle of drying [5].

Many polymer-based membrane materials, such as polydimethylsiloxane (PDMS), Polyether-block-amide (PEBAX), and Sulfonated poly(ether ether ketone) (SPEEK), can be used to make a dense membrane for air dehumidification[4] Among these materials, PDMS is commercially used in IMAD systems because of its low cost, chemical stability, nontoxicity, and good processability. When it is fabricated as a hollow-fiber form, PDMS tubular membranes have a high ratio of surface area to volume and are ease of scale-up[5], [9] However, most polymeric materials, including PDMS, are avoided in the working environment with a temperature greater than 50°C. For instance, in the study of flue gas dehydration using PEBAX®1074 and sulfonated SPEEK membrane materials, field tests were carried out when the temperature of flue gas was cooled below 50°C [10]. In another study, feed gas streams are cooled solely to accommodate a membrane gas separation process, and then they are heated back up [11]. Reducing the gas temperature and then heating it up add extra cost and energy to the process [11], [12].

In most softwood drying processes, the temperature of the air increases up to about 82°C [6]. Therefore, when testing the suitability of membranes for dehumidifying the exhaust air, it is important to address the influence of temperature on membrane’s performance (permeability and selectivity) instead of solely focusing on thermal stability/degradation analysis [13]. The performance of polymeric membrane materials can be improved by adding nanomaterials [12], ZnO [14], TiO₂ [15], SiO₂ [16], titanate nanotubes (TNTs) [17], Zeolite [18], [19] and more. However, most of these nanoparticles are non-sustainable and in some cases are petroleum based.

Cellulose nanocrystal (CNC), a class of nanomaterials with dimensions of 100–300 nm long and 5–70 nm in diameter derived from sustainable and renewable lignocellulosic biomass, can solve environmental problems related to other inorganic nanoparticles by serving as nature's storage for carbon dioxide. It has many attractive features, such as large specific surface area, high tensile strength and stiffness, abundance of surface hydroxyl groups, extremely low coefficient of thermal expansion (CTE), and more [20]–[23]. So, it has been used to substitute for inorganic nanoparticles (such as TiO₂, Al₂O₃) in film-type composites as reinforcement fillers to improve the strength, thermal expansion stability, optical property, etc. of the composites [24].

When the solution-diffusion mechanism governs gas diffusion in the dense polymeric membranes, the gas molecules dissolve into a membrane, diffuse across the membrane thickness, and then desorb from the other side of the membrane [5]. This study aimed to develop a high-performance hybrid composite membrane for air dehumidification at elevated temperatures for wood drying applications. CNC was chosen as an additive to PDMS because of two hypotheses: 1) the surface hydroxyl groups of CNC could increase the moisture adsorption sites in the PDMS membrane to improve the solubility of water vapor on the membrane surface and create a large water vapor concentration difference for a fast diffusion and 2) CNC's low thermal expansion attribute could restrain the PDMS thermal expansion at elevated temperatures if CNCs disperse in PDMS uniformly.

To achieve the goal, four objectives of this study were to 1) synthesize CNC/PDMS membranes using CNC suspension with four concentrations (0%, 2%, 4%, and 6%); 2) measure the water vapor permeability, nitrogen gas permeability, their selectivity, and coefficient of thermal expansion of the membrane samples at three different temperatures (25°C, 50°C, and 80°C); 3) characterize the morphology of the membrane samples, and 4) determine the optimal CNC concentration for making a high-performance CNC/PDMS composite membrane.

2.1 Materials

PDMS prepolymer and curing agent kits (Dow SYLGARDTM 184 Silicone Elastomer kit) were purchased from Dow Inc. (MI, USA). A CNC suspension at 11.8% solids was purchased from the USDA Forest Products Laboratory, distributed by the Process Development Center of the University of Maine (ME, USA). CNCs were subjected to comprehensive characterization in a previous study [25]. The results of the study revealed that the CNC particles had an average length of 134 ± 52 nm and a width of 7 ± 2 nm. The aspect ratio of the particles was determined to be 19. Additionally, the study reported a sulfate half-ester (-OSO₃⁻) content of 335 mmol kg^− 1 for the CNCs.

2.2 Methods

2.2.1 Synthesis of membranes

The PDMS prepolymer and curing agent at the ratio of 10:1 by weight (e.g., 10 grams of PDMS prepolymer and 1 gram of curing agent) was mixed mechanically for 5 minutes at room temperature. The solution was degassed for 45 seconds with a planetary centrifugal bubble free mixer (THINKY ARE-310, Tokyo, Japan). Then the solution was cast on a Teflon plate using a casting knife. The wet film had a nominal thickness of 150 µm. The wet film was cured in an oven at 50°C for 24 h. The pure PDMS was used as control group (i.e., 0% CNC).

For making CNC/PDMS membranes, ten (10) grams of the PDMS prepolymer and 1 gram of the curing agent were mixed mechanically for 5 minutes at room temperature. Appropriate amount of CNC suspension was added to the 11-gram solutions to achieve the CNC solid concentrations of 2%, 4%, and 6% in total solids, respectively. The mixture was then mechanically mixed for 1 hour, degassed for 45 seconds, and cast on a Teflon plate using a casting knife to obtain a wet membrane with a nominal thickness of 130 and 150 µm for Pure PDMS and CNC/PDMS films, respectively [26], [27]. The wet membranes were first dried at room temperature in a vacuum desiccator with a vacuum pressure of 88 kPa for 1 hour, intended to remove any remaining air bubbles in the wet membranes. After that, the semi-cured membranes were transferred to an oven and cured at 50°C for 24 h [26].

The thickness of dry film samples was measured using a digital micrometer (Mitutoyo, IL, USA) with an accuracy of 0.001 mm. Multiple measurements were performed to determine the average thickness of the film samples.

2.2.2 Scanning electron microscopic (SEM) analysis

The morphology of the CNC/PDMS membrane samples was observed using a scanning electron microscope (SEM) (NVision 40, Zeiss, Germany). The sample was placed on a specimen mount using carbon tape and then was coated with conductive silver. The regions of interest of the sample were sputter coated with gold (23 nm) using a Cressington 108 auto sputter coater (Ted Pella Inc., Redding, CA, USA). The images were taken at an accelerating voltage of 3 kV.

2.2.3 Atomic Force Microscopy (AFM) analysis

The PDMS and CNC/PDMS samples were cut into 5 mm x 5 mm areas for observation on an AFM (MFP-3D, Oxford Instruments Oxon, UK). Images were collected with an AC200TS-R3 probes with 9 ± 6 (nN/nm) spring constants. Alternating current mode produced images with a set point of (0.4 V) and constant free air amplitude of (1 V).

2.2.4 ATR-FTIR analysis

The ATR-FTIR analysis was performed using a PerkinElmer Spectrum Two™ FTIR spectrometer (Shelton, CT, USA) to evaluate the nature of the interaction between the CNCs and PDMS in the membranes. Each sample was measured at a resolution of 4 cm^− 1. The data were obtained under an accumulation of 16 scans in the range of 450–4000 cm^− 1.

2.2.5 Measurement of water vapor permeability of membranes

The water vapor transmission rate (WVTR) was measured using a dynamic vapor sorption (DVS) instrument (Model: DVS Advance; Surface Measurement Systems, London, UK) as shown in Fig. 1. The sample was placed to seal a Payne cell. Silicone gel beads were placed in the cell to absorb water vapor that entered the cell to maintain a 0% RH in the headspace. The surrounding environment of the Payne cell was controlled by DVS to desired testing conditions. For each sample type, nine (9) replicates cut from three same membranes were tested. The thickness of each sample was measured using a digital micrometer. Each sample was tested at 25°C, 50°C, and 80°C and a constant RH of 60%. The increase of mass of the Payne cell assembly with time was recorded simultaneously. The tests followed the ASTM E96 Standard test method for water vapor transmission of materials (ASTM 2016). The H₂O permeability (P_H2O) was calculated using Eq. (1):

$$\:{P}_{{H}_{2}O}=\frac{WVTR\times\:{T}_{mem}}{\:{\varDelta\:}_{Vapor\:pressure}}=\frac{\left(\varDelta\:m/\varDelta\:t\right)\times\:{T}_{m}}{A\times\:\varDelta\:p\times\:18}\:\:\left(\frac{mol}{m\bullet\:s\bullet\:Pa}\right)\:$$

Where, Δm/Δt is the slope of a linear section of the mass versus time, g/s; T_m is the thickness of the sample membrane, m; A is the open area of the Payne cell, 1.18^10^− 4 m²; Δp is the partial pressure difference of water vapor, $\:\varDelta\:p=\frac{\varDelta\:RH}{100}\times\:Saturated\:water\:pressure$ at 25°C,50°C, and 80°C, Pa (Saturated water vapor pressure at 25°C = 3171 Pa, 50°C = 12351 Pa, 80°C = 47415 Pa);18 g/mol is the molecular weight of H₂O.

2.2.6 Measurement of nitrogen gas permeability of membranes

The pure nitrogen gas transmission rate (NGTR) was measured using a single gas transmission cell (Custom Scientific Instruments Inc, PA, USA) as shown in Fig. 2, following the ASTM D1423-82(2015) Standard test method for determining gas permeability characteristics of plastic film and sheeting (ASTM 2015). Samples were cut into circles with a diameter of 0.12 meter thickness was recorded. One sample was mounted to separate the cell into the upper and lower chambers. Compressed nitrogen gas flowed into the lower chamber, diffused through the sample, and entered the upper chamber. The volume expansion of the upper chamber was measured by recording the movement of a liquid color slug in the capillary tube. Nine (9) replicates for each sample type were tested. The transmission cell was placed in a water bath to maintain the desired temperatures of 25°C, 50°C, and 80°C. The pressure difference between the lower chamber and upper chamber was measured by a pressure gauge. The N₂ gas permeability was calculated using Eq. (2):

$$\:{P}_{{N}_{2}}=\frac{NGTR\times\:{T}_{m}}{\varDelta\:p}=\:\frac{{P}_{0}\:\times\:\:S\times\:\pi\:{D}^{2}\times\:{T}_{m}}{\varDelta\:p\times\:4\:A\:R\:T}\:\:\:\:\:\:\:\:\:\:\:\left(\frac{mol}{m\bullet\:s\bullet\:Pa}\right)\:$$

Where P₀ is the ambient pressure, 101325 Pa, S is the rate of rising of the capillary slug (m/s); D is the capillary tube’s inner diameter, 0.4957^10^− 3 m; A is the open area of the cell, 0.6655^10^− 2 m²; T_m is the thickness of the sample membrane, m; Δp is the pressure difference between the lower and upper chamber, i.e., gauge pressure, Pa; T is the temperature of gas, K; R is the gas constant (8.3143 m³.Pa/(mol.K)).

2.2.7 Calculation of Selectivity of water vapor and nitrogen

The selectivity (α) was obtained from the ratio of H₂O to N₂ permeability, Eq. (3).

$$\:\alpha\:=\:\frac{{P}_{{H}_{2}O}}{{P}_{{N}_{2}}}$$

2.2.8 Measurement of coefficient of thermal expansion (CTE)

The in-plane coefficient of thermal expansion (CTE) of the PDMS and CNC/PDMS membrane samples were measured using a thermomechanical instrument (Q400 TMA, TA Instrument). Samples were cut into 5 mm × 5 mm areas and one sample at a time was sandwiched between two crystal wafers to ensure force dispersion. An expansion probe rested on the surface of the samples with a 0.02 N preload. As the temperature was raised from 25°C to 100°C a heating rate of 5°C/min, the change in thickness of the sample was recorded, allowing the calculation of CTE from the slope of the resulting expansion temperature plots. CTE is calculated using Eq. 4 [23].

$$\:CTE=\:\frac{\varDelta\:L}{{L}_{0}\:\times\:\:\varDelta\:t}$$

Where L₀ is the initial length of the sample (µm); Δt is the temperature change (° C), and ΔL is the thermal expansion (or contraction) of the sample after the temperature change.

For each concentration, 3 samples were prepared and tested, where performed in triplicate. Therefore, the average of Nine (9) replicates for each sample type were measured and reported.

2.2.9 Statistical analysis

Two-way analysis of variance (ANOVA) was performed for two independent variables (CNC concentration and temperature) to understand the main effects and interaction effects of them on the water vapor and nitrogen gas permeability, and selectivity of the membrane samples. One-way ANOVA was conducted for the effect of the independent variable (CNC concentration) on CTE. All statistical analyses were run in OriginPro software version 2022b, with a level of significance of 0.05 [28].

3.1 Appearance and morphology of CNC/PDMS membrane samples

The appearance of four types of membrane samples is provided in Fig. 3. The Pure PDMS has the highest transparency, and the transparency of the film is decreased with increasing the CNC concentration. The SEM images (Fig. 4) show the top surface morphology of four membrane samples. The dispersion of CNC in PDMS is dependent on the concentration of CNC added. CNC particles are relatively small and distributed randomly in PDMS when 2% of CNC was added. In contrast, when the concentration of CNC was increased to 4% and 6%, the CNC aggregation became more significantly distinctive. In addition, these SEM images show the PDMS and CNC/PDMS membranes are dense membranes, and no pores or defects are observed at this level of magnification. AFM images (Fig. 5) were used to estimate the 3D shape and dimensions of CNC nanoparticles agglomerations. Most of the CNC particles have a height in the range of 10 nm to 100 nm. The maximum height of agglomerated CNC particles could reach over 100 nm. More high peaks are observed with the increase in CNC concentration.

3.2 ATR - FTIR analysis

Figure 6 shows the FTIR spectra of an air-dried CNC film (Control-1), pure PDMS sample (Control-2), and CNC/PDMS samples with CNC concentrations of 2%, 4%, and 6%. For Control-1 CNC film, a strong O-H stretching vibration at 3332 cm^− 1 and bending band of H-O-H bond at 1638cm^− 1 are observed [29], [30]. For Control-2 PDMS membrane, the featured peaks of PDMS include 2960 cm^− 1 for CH₃ group vibration, 1257 cm^− 1 for the symmetric deformation of the CH₃ group, 887 cm^− 1 for Si-C vibration, and 790 cm^− 1 for Si-O vibration [31]. When low concentrations of CNCs were blended in PDMS, the spectra of CNC/PDMS samples show that peaks at 3332 cm^− 1 and 1638cm^− 1 almost disappeared but the peak at 1057 cm^− 1 representing the stretching vibrations of C-O in CNC became slightly distinctive [30]. The results reveal that the addition of CNC from 2–6% did not change the general spectra pattern of CNC/PDMS samples. Therefore, the performance of CNC/PDMS membrane is mainly governed by the PDMS.

3.3 Permeability of water vapor

The values of water vapor permeability (Fig. 7) increased with the CNC concentration from 0% (pure PDMS) to 2% and then decreased. The mean ± SD values of water vapor permeability at 25°C for Pure PDM, 2% CNC/PDMS, 4% CNC/PDMS, and 6% CNC/PDMS were 30,683.6 ± 615.9, 38,295.6 ± 1,070.9, 35,883.6 ± 1,248.8, and 35,846.1 ± 1,338.3 Barrer, respectively. The addition of 2% CNC resulted in an increase of 24.8%, 30.9%, and 11.2% at 25°C, 50°C, and 80°C, respectively. Higher concentrations of CNC (4% and 6%) show a slightly decreasing trend for water vapor permeability, while the permeability values remained higher than those for pure PDMS membranes. This trend is the same in all three temperatures except for 80°C. When the temperature increased from 25°C to 80°C, the permeability of water vapor decreased dramatically.

Two-way ANOVA results reveal that the effects of CNC concentration, temperature and their interaction on water permeability were statistically significant (Table A.1 Supplementary document). Furthermore, the comparison between different CNC concentrations shows that the 2% CNC/PDMS is significantly different from other samples, but there is not any significant difference between pure PDMS (0% CNC), 4% CNC/PDMS, and 6% CNC/PDMS samples (Table A.2). Therefore, 2% CNC is the optimal concentration in terms of water vapor permeability.

3.4 Permeability of nitrogen gas

The values of Nitrogen gas permeability (Fig. 8) increased with the CNC concentration from 0% (pure PDMS) to 2% and then slightly decreased. The mean ± SD values of Nitrogen gas permeability at 25°C for Pure PDM, 2% CNC/PDMS, 4% CNC/PDMS, and 6% CNC/PDMS were 263.7 ± 4.6, 324.9 ± 18.7, 310.0 ± 8.4, and 310.3 ± 7.3 Barrer, respectively.

The nitrogen gas permeability (Fig. 8) was increased by 23.2%, 27.8%, and 16.0% at three temperatures when adding 2% CNC to the PDMS membrane. Similarly, higher concentrations of CNC (4% and 6%) show a slightly decreasing trend for nitrogen permeability compared with 2% CNC. The nitrogen permeability increased as the temperature was elevated from 25°C to 80°C.

Two-way ANOVA results show the effect of CNC concentration on nitrogen permeability was not statistically significant but the effect of temperature on permeability was statistically significant (Table A.3). The interaction between temperature and concentration was not statistically significant either. Furthermore, the comparison between different CNC concentrations reveals that there was not a statistically significant difference between 2%, 4%, and 6% CNC/PDMS membranes but all the CNC/PDMS were significantly different from the Pure PDMS (Table A.4). Therefore, 2% CNC is the optimal concentration in terms of nitrogen gas permeability.

3.5 Selectivity of water vapor and nitrogen

The selectivity of water vapor over nitrogen (Fig. 9) at 25°C of pure PDMS, 2%CNC/PDMS, 4%CNC/PDMS, and 6%CNC/PDMS were 116.3, 117.9, 115.7, and 115.5, respectively. A slight increase of about 3.1% was observed at 2% CNC addition, followed by a slight decrease when increasing the concentration of CNC at three temperatures. Two-way ANOVA results (Table A.5) show that the effect of CNC concentration on selectivity was not statistically significant but the effect of temperature on selectivity was statistically significant. The interaction between temperature and concentration was not statistically significant for selectivity.

3.6 Coefficient of thermal expansion (CTE) of membranes

The dimensional stability of CNC/PDMS membranes influences how well the membrane operates at elevated temperatures. It was evaluated in terms of the coefficients of thermal expansion (CTE) of pure PDMS and CNC/PDMS membranes (Fig. 10). The mean ± SD of CTE for pure PDMS, from room temperature up to 100°C, was 301.1 ± 12.6 µm/m.°C. This measured value is in good agreement with the published values for CTE of PDMS which is 309 ppm/°C [32]. After adding 2% CNC, the mean ± SD of CTE was reduced to 274.4 ± 8.5 µm/m.°C, which is about 12% lower than that of pure PDMS. This decrease in CTE was expected because CNC has a very low CTE of 9 µm/m.°C [33]and it helps to restrain the expansion of PDMS at elevated temperatures when CNCs are dispersed in PDMS [23]. However, the mean ± SD of CTE of 4% CNC/PDMS and 6% CNC/PDMS was 269.5 ± 13.2 and 264.3 ± 8.6, respectively. Increasing the CNC concentration to 4% and 6% did not further decrease the CTE greatly. One-way ANOVA results (Table A.6) show that the decrease in the CTE between pure PDMS sample and 2% CNC/PDMS sample was statistically significant. Also, the comparison between different CNC concentrations (Table A.7) shows that there is not a significant difference between 2%, 4% and 6% CNC/PDMS samples and all these samples are significantly different from Pure PDMS. Once again, 2% CNC is the optimal concentration in terms of CTE.

4.1 Influence of CNC agglomeration on membrane appearance and properties

When CNC suspension was mechanically mixed with the PDMS prepolymer and curing agent solution and then dried until fully cured, CNC agglomeration occurred due to the poor compatibility of hydrophilic CNC and hydrophobic PDMS. When more agglomerated CNC particles were formed as the increase of CNC concentration, the membrane transparency decreased ascribed to the mismatch of refractive indexes between CNC and PDMS. Combined SEM and AFM images, the CNC particles are present on the membrane surface with needle-like shapes.

The water vapor and nitrogen gas permeability, and CTE results of CNC/PDMS membrane samples show that 2% CNC was the optimal concentration. Increasing the CNC concentration were not further enhancing the membrane performance. The reason is also explained by the poor compatibility. The compatibility of CNC and PDMS might be improved by modifying the CNC, such as the silylation of CNC [26], [27]

4.2 Effects of temperature on permeability of water vapor and nitrogen gas

Gas transport through dense polymeric membranes is governed by the solution-diffusion mechanism. Therefore, the effects of temperature on both solution and diffusion processes should be considered. The influence of temperature on solubility has been well addressed in terms of the van't Hoff relationship. Gas solubility correlates with its condensability. For less condensable gases (e.g., N₂ and O₂) that often have lower critical temperatures (Table 1), gas solubility increases with an increase in the temperature. However, for condensable gases and vapors (e.g., water vapor) solubility decreases with increasing temperature [11], [34]. The diffusion of gas molecules in a dense membrane is a thermally activated process and increases as the temperature increases, which is well described by the Arrhenius equation [34] However, for water vapor permeability, sorption is the preponderant factor [35], Which means the magnitude of solubility change is greater than that of diffusivity change (because of high condensability which correlates with the high critical temperature as shown in Table. 1), thereby resulting in a decrease in permeability. This explains why water vapor permeability decreases, as illustrated in Fig. 7, when the temperature increases from 25°C to 80°C.

However, for the N₂ and O₂ molecules which are larger in kinetic diameter and less condensable, while increasing the temperature, there is increase in both the diffusivity and the solubility. This leads to an increase in permeability [11] This phenomenon is evident in the increased nitrogen gas permeability when the temperature rises from 25°C to 80°C, as shown in Fig. 8.

4.3 Effectiveness of CNC as a nanofiller on water vapor and nitrogen gas permeability

In this study, a bio-based material CNC was used as a nano filler in the PDMS matrix to improve the PDMS membrane performance for air dehydration, to the best of our knowledge, which is the first time to report this type of application of CNC. CNC’s renewable and sustainable features make it possible as alternatives to non-renewable nanofillers, like zeolites (ZIF), silicon dioxide (SiO₂), and titanate nanotubes (TNTs), reported in the previous studies [16]–[19]. While in most of the other studies regarding mixed matrix membranes, the nano filler they used in PDMS is not bio based and not renewable. The CNC used in this study is renewable and can be produced from sustainable sources, two features that make it superior to other nano particles such as zeolites (ZIF). On the other hand, to the best of our knowledge, this is the first time to make a mixed matrix membrane with PDMS for improving water vapor permeability of PDMS. Our study revealed that the water vapor permeability and nitrogen permeability of PDMS membrane at room temperature were found to be 3,683 Barrer and 263 Barrer, respectively, consistent with previous studies [37], [38]. After adding the optimal concentration of 2% CNC, the water vapor and nitrogen gas permeability at room temperature were increased by 24.8% and 23.5 %, respectively. I the study conducted by Mao et al. (2012), the addition of approximately 30 wt.% of ZIF-L to PDMS led to a significant 8.0% increase in water vapor permeability at 40°C [18]. Similarly, Sahin et al. (2020) conducted another study where they added 20 wt.% of ZIF-71 to PDMS and observed a significant enhancement of around 34% in the permeability of nitrogen gas at 35°C [19]. CNCs have demonstrated several benefits in comparison to ZIF nanoparticles. For instance, they require a significantly lower weight ratio to achieve a similar improvement in water vapor and nitrogen gas permeability.

Beltran et al. reported an increase in the gas permeability of PDMS after adding modified SiO₂, due to the increase in the free volume of the membrane [16]. Such an increase in free volume led to an increase in the diffusion coefficient. These findings suggested that PDMS, a rubbery polymer, may exhibit an increase in free volume upon the incorporation of nonporous fillers of SiO₂ [16]. Li et al. (2010) found that adding TNTs into the PDMS matrix resulted in a significant enlargement of the fractional free volume (FFV)[17]. The enlargement of the FFV created more diffusion paths for small penetrants, and consequently, increased the gas permeability of the nanocomposite membranes. In this study, the effectiveness of CNC on the enhancement of water vapor and nitrogen gas permeability could be also explained by the modified FFV.

In this study, CNC/PDMS membranes with the CNC weight concentrations of 0%, 2%, 4%, and 6% were successfully synthesized. The samples were defect-free and contained no observable pores under microscopic imaging because of the degassing step during preparation. The SEM images and AFM results revealed the CNC nanoparticles were dispersed in the PDMS matrix either randomly or in small agglomerations depending on CNC concentrations. The FTIR spectra confirmed the existence of CNCs in the CNC/PDMS samples. The addition of CNC alters the permeability and selectivity of the membranes. The optimal CNC concentration was 2% with enhanced water vapor permeability at all temperatures up to 24.8% while increasing the selectivity slightly up to 3.1%. In addition, the 2% CNC/PDMS samples showed an 8.9% lower value for CTE which results in higher thermal dimensional stability of the sample. The future work would be the modification of CNCs, e.g., silylation, to increase the compatibility between CNCs and PDMS polymer and increase the dispersion of CNCs in the PDMS matrix.

Acknowledgement: This project was supported by the USDA National Institute of Food and Agriculture, McIntire-Stennis (Project number ME0-42205) through the Maine Agricultural & Forest Experiment Station; the U.S. Department of Agriculture’s Agricultural Research Service (USDA ARS Agreement No. 58-0204-6-003 & No. 58-0204-9-166); the US Forest Service and US Endowment for Forestry and Communities-P3Nano Advancing Commercialization of Cellulose Nanomaterials (Agreement 21-00166).

In addition, authors would like to thank Dr. Elliot Sanders for help with the AFM testing and iLab UMaine for help to prepare SEM images.

Funding

This project was supported by the USDA National Institute of Food and Agriculture, McIntire-Stennis (Project number ME0-42205) through the Maine Agricultural & Forest Experiment Station; the U.S. Department of Agriculture’s Agricultural Research Service (USDA ARS Agreement No. 58-0204-6-003 & No. 58-0204-9-166); the US Forest Service and US Endowment for Forestry and Communities-P3Nano Advancing Commercialization of Cellulose Nanomaterials (Agreement 21-00166).

Conflict of interest/Competing interests

This work has no conflicts of interest to disclose.

Ethics approval and consent to participate

Not applicable. This study did not involve human or animal subjects.

Consent for publication

All the images and tables presented in this work are original results. Content adopted from other works has been cited carefully.

Data availability

All research data is included in the article.

Authorship Contributions

Nasim Alikhani: Methodology, Running tests, Visualization, Writing original draft.

Ling Li: resources, writing – review and editing, and funding acquisition.

Jinwu Wang: resources, writing – review and editing, and funding acquisition.

S. Zhao et al., “Simultaneous heat and water recovery from flue gas by membrane condensation: Experimental investigation,” Appl Therm Eng, vol. 113, pp. 843–850, Feb. 2017, doi: 10.1016/j.applthermaleng.2016.11.101.
M. Qu, O. Abdelaziz, Z. Gao, and H. Yin, “Isothermal membrane-based air dehumidification: A comprehensive review,” Renewable and Sustainable Energy Reviews, vol. 82, pp. 4060–4069, Feb. 2018, doi: 10.1016/J.RSER.2017.10.067.
G. P. Thiel and C. Atkinson, “The potential for energy savings in U.S. houses by using isothermal dehumidification,” Sci Technol Built Environ, vol. 27, no. 6, pp. 776–787, 2021, doi: 10.1080/23744731.2021.1879564.
D. Bergmair, S. J. Metz, H. C. De Lange, and A. A. Van Steenhoven, “Modeling of a water vapor selective membrane unit to increase the energy efficiency of humidity harvesting,” in Journal of Physics: Conference Series, Institute of Physics Publishing, Nov. 2012, p. 012161. doi: 10.1088/1742-6596/395/1/012161.
N. Alikhani, L. Li, J. Wang, D. Dewar, and M. Tajvidi, “Exploration of membrane-based dehumidification system to improve the energy efficiency of kiln drying processes: Part I Factors that affect moisture removal efficiency,” Wood and Fiber Science, 2020, Accessed: Feb. 05, 2021. [Online]. Available: http://wfs.swst.org/index.php/wfs/article/view/2960
W. T. Simpson, Dry kiln operator’s manual. U.S. Dept. of Agriculture, Forest Service, Forest Products Laboratory, 1991.
H. L. Field and J. M. Long, “Heating, Ventilation, and Air Conditioning,” Introduction to Agricultural Engineering Technology, pp. 333–358, 2018, doi: 10.1007/978-3-319-69679-9_23.
D. V. (David V. Chadderton, Air conditioning: a practical introduction. E & FN Spon, 1997.
C. Z. Liang and T. S. Chung, “Robust thin film composite PDMS/PAN hollow fiber membranes for water vapor removal from humid air and gases,” Sep Purif Technol, vol. 202, pp. 345–356, Aug. 2018, doi: 10.1016/J.SEPPUR.2018.03.005.
H. Sijbesma, K. Nymeijer, R. van Marwijk, R. Heijboer, J. Potreck, and M. Wessling, “Flue gas dehydration using polymer membranes,” J Memb Sci, vol. 313, no. 1–2, pp. 263–276, 2008.
M. Rezakazemi, M. Sadrzadeh, and T. Matsuura, “Thermally stable polymers for advanced high-performance gas separation membranes,” Prog Energy Combust Sci, 2018.
Z. Jahan, M. B. K. Niazi, M. B. Hägg, and yvind W. Gregersen, “Decoupling the effect of membrane thickness and CNC concentration in PVA based nanocomposite membranes for CO2/CH4 separation,” Sep Purif Technol, vol. 204, pp. 220–225, Oct. 2018, doi: 10.1016/J.SEPPUR.2018.04.076.
J. D. Jovanovic, M. N. Govedarica, P. R. Dvornic, and I. G. Popovic, “The thermogravimetric analysis of some polysiloxanes,” Polym Degrad Stab, vol. 61, no. 1, pp. 87–93, Jan. 1998, doi: 10.1016/S0141-3910(97)00135-3.
S. Ammar, K. Ramesh, B. Vengadaesvaran, S. Ramesh, and A. K. Arof, “Amelioration of anticorrosion and hydrophobic properties of epoxy/PDMS composite coatings containing nano ZnO particles,” Prog Org Coat, vol. 92, pp. 54–65, Mar. 2016, doi: 10.1016/J.PORGCOAT.2015.12.007.
D. T. Vaimakis-Tsogkas, D. G. Bekas, T. Giannakopoulou, N. Todorova, A. S. Paipetis, and N. M. Barkoula, “Effect of TiO2 addition/coating on the performance of polydimethylsiloxane-based silicone elastomers for outdoor applications,” Mater Chem Phys, vol. 223, pp. 366–373, Feb. 2019, doi: 10.1016/J.MATCHEMPHYS.2018.11.011.
A. Beltran, G. Nisola, E. Cho, and E. Lee, “Organosilane modified silica/polydimethylsiloxane mixed matrix membranes for enhanced propylene/nitrogen separation,” Appl Surf Sci, 2011, Accessed: Apr. 25, 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0169433211013067
B. Li et al., “Rubbery polymer-inorganic nanocomposite membranes: Free volume characteristics on separation property,” Ind Eng Chem Res, vol. 49, no. 24, pp. 12444–12451, Dec. 2010, doi: 10.1021/IE101142B.
H. Mao et al., “Highly selective and robust PDMS mixed matrix membranes by embedding two-dimensional ZIF-L for alcohol permselective pervaporation,” J Memb Sci, vol. 582, pp. 307–321, Jul. 2019, doi: 10.1016/J.MEMSCI.2019.04.022.
F. Şahin, B. Topuz, and H. Kalıpçılar, “ZIF filled PDMS mixed matrix membranes for separation of solvent vapors from nitrogen,” J Memb Sci, 2020, Accessed: Apr. 25, 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0376738819325840
C. Gomri, M. Cretin, and M. Semsarilar, “Recent progress on chemical modification of cellulose nanocrystal (CNC) and its application in nanocomposite films and membranes-A comprehensive review,” Carbohydr Polym, vol. 294, p. 119790, Oct. 2022, doi: 10.1016/J.CARBPOL.2022.119790.
P. Cruz-Tato et al., “Metalized Nanocellulose Composites as a Feasible Material for Membrane Supports: Design and Applications for Water Treatment,” Environ Sci Technol, vol. 51, no. 8, pp. 4585–4595, Apr. 2017, doi: 10.1021/ACS.EST.6B05955/SUPPL_FILE/ES6B05955_SI_001.PDF.
O. Nechyporchuk, M. N. Belgacem, and J. Bras, “Production of cellulose nanofibrils: A review of recent advances,” Ind Crops Prod, vol. 93, pp. 2–25, Dec. 2016, doi: 10.1016/J.INDCROP.2016.02.016.
N. Saba and M. Jawaid, “A review on thermomechanical properties of polymers and fibers reinforced polymer composites,” Journal of Industrial and Engineering Chemistry, vol. 67, pp. 1–11, Nov. 2018, doi: 10.1016/J.JIEC.2018.06.018.
N. Azizi, T. Mohammadi, and R. B. Research, “Comparison of permeability performance of PEBAX-1074/TiO2, PEBAX-1074/SiO2 and PEBAX-1074/Al2O3 nanocomposite membranes for CO2/CH4,” Chemical Engineering Research, 2017.
M. S. Reid, M. Villalobos, and E. D. Cranston, “Benchmarking Cellulose Nanocrystals: From the Laboratory to Industrial Production,” Langmuir, vol. 33, no. 7, pp. 1583–1598, Feb. 2017, doi: 10.1021/ACS.LANGMUIR.6B03765.
M. Planes, J. Brand, and S. Lewandowski, “Improvement of the thermal and optical performances of protective polydimethylsiloxane space coatings with cellulose nanocrystal additives,” Applied Materials and Interfaces, 2016, Accessed: Feb. 26, 2022. [Online]. Available: https://pubs.acs.org/doi/abs/10.1021/acsami.6b09043
T. Xia, H. Yuwen, and N. Lin, “Self-bonding sandwiched membranes from PDMS and cellulose nanocrystals by engineering strategy of layer-by-layer curing,” Compos Sci Technol, 2018.
OriginLab Corporation, “Origin(Pro).” Northampton, MA, USA., 2022.
E. Inácio, D. Souza, and M. D. Applications, “Thermal and crystallization behavior of PLA/PLLA-grafting cellulose nanocrystal,” Materials Sciences and Applications, 2020, Accessed: Feb. 26, 2022. [Online]. Available: https://www.scirp.org/html/4-7702507_97682.htm
T. Jayaramudu, H. Ko, H. Kim, J. Kim, and M. RM, “Electroactive hydrogels made with polyvinyl alcohol/cellulose nanocrystals,” Materials, 2018, Accessed: Feb. 26, 2022. [Online]. Available: https://www.mdpi.com/335546
X. Cui, G. Zhu, Y. Pan, Q. Shao, M. Dong, and Y. Zhang, “Polydimethylsiloxane-titania nanocomposite coating: fabrication and corrosion resistance,” Polymer (Guildf), vol. 138, 2018.
A. Müller, M. C. Wapler, and U. Wallrabe, “A quick and accurate method to determine the Poisson’s ratio and the coefficient of thermal expansion of PDMS,” Soft Matter, vol. 15, no. 4, pp. 779–784, Jan. 2019, doi: 10.1039/C8SM02105H.
J. A. Diaz, X. Wu, A. Martini, J. P. Youngblood, and R. J. Moon, “Thermal Expansion of Self-Organized and Shear-Oriented Cellulose Nanocrystal Films,” 2013, doi: 10.1021/bm400794e.
B. W. Rowe, L. M. Robeson, B. D. Freeman, and D. R. Paul, “Influence of temperature on the upper bound: Theoretical considerations and comparison with experimental results,” J Memb Sci, vol. 360, no. 1–2, pp. 58–69, Sep. 2010, doi: 10.1016/J.MEMSCI.2010.04.047.
A. S. Spatafora Salazar, P. A. Sáenz Cavazos, H. Mújica Paz, and A. Valdez Fragoso, “External factors and nanoparticles effect on water vapor permeability of pectin-based films,” J Food Eng, vol. 245, pp. 73–79, Mar. 2019, doi: 10.1016/J.JFOODENG.2018.09.002.
S. J. Metz, “WATER VAPOR AND GAS TRANSPORT THROUGH POLYMERIC MEMBRANES,” Nov. 2003. Accessed: Feb. 05, 2021. [Online]. Available: https://research.utwente.nl/en/publications/water-vapor-and-gas-transport-through-polymeric-membranes
M. Leemann, G. Eigenberger, H. S.-J. of M. Science, and undefined 1996, “Vapour permeation for the recovery of organic solvents from waste air streams: separation capacities and process optimization,” Elsevier, Accessed: Apr. 25, 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/0376738895001301
S. Koester, F. Roghmans, M. W.-J. of M. Science, and undefined 2015, “Water vapor permeance: The interplay of feed and permeate activity,” Elsevier, Accessed: Apr. 25, 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0376738815001957

Table 1 is available in the Supplementary Files section.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
11 Oct, 2024
Editor assigned by journal
11 Oct, 2024
Submission checks completed at journal
24 Sep, 2024
First submitted to journal
10 Jul, 2024

You are reading this latest preprint version

Cellulose Nanocrystal/Polydimethylsiloxane hybrid membranes for air dehydration at elevated temperatures

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and Methods

2.1 Materials

2.2 Methods

2.2.1 Synthesis of membranes

2.2.2 Scanning electron microscopic (SEM) analysis

2.2.3 Atomic Force Microscopy (AFM) analysis

2.2.4 ATR-FTIR analysis

2.2.5 Measurement of water vapor permeability of membranes

2.2.6 Measurement of nitrogen gas permeability of membranes

2.2.7 Calculation of Selectivity of water vapor and nitrogen

2.2.8 Measurement of coefficient of thermal expansion (CTE)

2.2.9 Statistical analysis

3. Results

3.1 Appearance and morphology of CNC/PDMS membrane samples

3.2 ATR - FTIR analysis

3.3 Permeability of water vapor

3.4 Permeability of nitrogen gas

3.5 Selectivity of water vapor and nitrogen

3.6 Coefficient of thermal expansion (CTE) of membranes

4. Discussion

4.1 Influence of CNC agglomeration on membrane appearance and properties

4.2 Effects of temperature on permeability of water vapor and nitrogen gas

4.3 Effectiveness of CNC as a nanofiller on water vapor and nitrogen gas permeability

5. Conclusions

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1