**2. Materials and Methods**

Cellulose fibers commercially referred as cellulose filaments were received from Kruger Biomaterials Inc. (Montreal, QC, Canada), a privately held company that transforms renewable resources into sustainable essentials. The cellulose fibers used in this study were obtained in the form of suspension (2.5 wt%). The fibers were obtained by freeze-drying the suspension. PLA grade 4043D was used in this study and obtained from NatureWorks (Minnetonka, MN, USA). This particular grade of PLA is targeted for the preparation of membranes and films. Chloroform was used as the solvent for dissolving PLA and obtained from Sigma-Aldrich (Oakville, ON, Canada).

#### *2.1. Fiber Treatment*

For the treatment of cellulose fibers, an organic phosphoric acid ester-based surfactant was added to the CF suspension. The surfactant was obtained from EMCO-Inortech (Shwego-wett 6267), Montreal, QC, Canada. It is a biodegradable, VOC-free, anionic universal wetting, and dispersive additive. The concentration of surfactant used for treatment of the fibers was 20 wt% of dry fiber weight. The fiber suspension along with the surfactant was mixed using a high shear mixer at 2000 rpm for 15 min and then freeze-dried.

### *2.2. Film Preparation*

PLA membranes were prepared using solvent casting method. Briefly, 5 g of PLA was dissolved in 100 mL of chloroform under constant stirring at 600 rpm for 2 h at room temperature and atmospheric pressure. The solution was poured into a Petri dish and left in a fume hood at room temperature for evaporating the solvent. For preparation of the CF composite membranes, varying concentrations of untreated and surfactant treated CF were the added to dissolved PLA and stirred vigorously for another 30 min using a high shear mixer at 1200 rpm. At the end of the mixing process, the PLA-CF solution was homogenous, and no bubbles were observed. The prepared PLA membranes with varying concentrations of untreated cellulose fibers (UCF) and treated cellulose fibers (TCF) used for this study are mentioned in Table 1. Micrographs showing UCF and TCF are presented in Figure 1. Photographs of all prepared films used in this experiment are shown in Figure 2. All membrane samples were vacuum dried at 40 ◦C for 96 h to release any remaining solvent and then sealed and stored in zipped airtight bags for characterization.

**Figure 1.** Micrographs of cellulose fibers (**a**) untreated, (**b**) after surfactant treatment.


**Table 1.** Description of film samples prepared for this study.

(**b**)

**Figure 2.** Photographs of (**a**) PLA-UCF films with 1–20% UCF content; (**b**) PLA-TCF films with 1–20% TCF content.

#### *2.3. Methods*

#### 2.3.1. Vapor Permeability and Water Vapor Transmission Rate (WVTR)

The ability of a porous material to transfer moisture due to a vapour pressure gradient can be expressed by the vapour permeability of the material. The transfer of moisture during this process can take place due to three factors: diffusion (self-collision of water molecules), effusion (collision of water molecules with the pore walls), and liquid transfer (associated with capillary condensation) [42].

The vapor permeability and the WVTR of the composite membranes were determined according to ASTM standard E 96 using the wet cup method. The data used in this analysis were obtained a dynamic vapour sorption equipment DVS Advantage, SMS, London, UK. A SMS Payne Type Diffusion cell was used in this test that is specially designed to measure the permeability and moisture transmission rate of thin films in the DVS instrument. The Payne cell has two main components: cell lid that holds the test sample and a cell cup with a small reservoir. The test samples were cut to a diameter of 15.5–16.0 mm to perfectly fit in the Payne cell lid. The Payne cell cup was filled with 200 μL of water and the sample was placed between two O-rings. The bottom cell cup was screwed into the upper cell lid and placed on a metal sample pan for measurement in the DVS instrument. The samples were initially conditioned for 30 min at 95% RH and 25 ◦C. The test was run under 50% RH and 25 ◦C for 24 h and the cell was auto-weighed every minute by a DVS high mass ultra balance. The cell opening provided an active area of 113 mm2 for moisture transport. The sample thickness was measured using a micrometer and varied between the specimens (100–500 μm). The tests were performed in triplicate and the average reading was reported.

The WVTR was determined by dividing the slope of the linear portion of the weight gain versus time curve by the tested surface area (interior cell area = 113 mm2) of the sample using the following equation:

$$\text{WVTR} \left(\frac{\text{g}}{\text{h} \cdot \text{m}^2}\right) = \frac{\text{slope} \left(\text{g/h}\right)}{\text{cell area} \left(\text{m}^2\right)}\tag{1}$$

WVTR describes the rate of water permeating through a test specimen into the headspace volume of a container, which differs in relative humidity (ΔRH).

Vapor permeability, P hence is represented as:

$$P\left(\frac{\text{g}}{\text{h}\cdot\text{m}\cdot\text{Pa}}\right) = \frac{\text{WVTR}\left(\text{g}/\text{h}\cdot\text{m}^2\right)\cdot\text{d}\left(\text{m}\right)}{\Delta\text{P}\left(\text{Pa}\right)\left(\text{m}^2\right)}\tag{2}$$

where d is the test specimen thickness (m)

It is simpler to compare WVTR between samples with varying thicknesses, by eliminating the thickness factor and reporting it as:

$$\text{Normalized WVT} \left( \frac{\text{g}}{\text{h} \cdot \text{m}} \right) = \text{WVTR} \left( \frac{\text{g}}{\text{h} \cdot \text{m}^2} \right) \cdot \text{d} \left( \text{m} \right) \tag{3}$$

#### 2.3.2. Water Retention

The water retention test was performed according to ASTM D-570-95 using the 24 h immersion test method. The specimen was placed in a container of water at room temperature, rested at its edge and entirely immersed. At the end of 24 h, the specimen was removed from water, wiped free of surface moisture with a dry cloth, and weighed to the nearest 0.001 g immediately. The test was performed in triplicates and the average reading was reported. The test specimens were thin films of 150 mm in diameter.

The percentage of water absorption (WA) was calculated according to the following equation:

$$\text{WA} = \frac{\text{m}\_{24} - \text{m}\_0}{\text{m}\_0} \times 100 \tag{4}$$

where m0 is the initial mass of the test specimen (g), m24 is the mass of the test specimen after partial immersion for 24 h (g).

#### 2.3.3. Scanning Electron Microscopy

Photomicrographs of the samples were captured using a Thermo Fisher Scientific Scanning Electron Microscope (SEM), model FEI Quanta 250 (Hillsboro, OR, USA) operating at 15 kV. The samples were gold coated to obtain high magnification of morphology and texture.

#### 2.3.4. Thermogravimetric Analysis

The thermal degradation behaviour of samples was studied using a thermogravimetric analyser TGA/DTA 851e Mettler Toledo instrument (Columbus, OH, USA). Approximately 8–10 mg of sample was placed in an uncovered 70 μL alumina crucible to determine mass loss during heating. The specimens were heated at a rate of 10 ◦C/min from 25 to 950 ◦C under nitrogen atmosphere purged at 50 mL/min. The test was performed in triplicate.

#### 2.3.5. Differential Scanning Calorimetry

The melting point and enthalpy of the samples were determined by a differential scanning calorimeter DSC Mettler Toledo 822/e (Columbus, OH, USA). Approximately 7 mg of sample was placed in a sealed aluminium cell to determine melting and glass transition temperatures and enthalpies during heating. The samples were heated from 25 ◦C to 250 ◦C at 10 ◦C/min under nitrogen. The test was performed in triplicate.

#### 2.3.6. Dynamic Mechanical Analysis

Glass transition temperature (Tg) was determined by dynamic mechanical analysis based on the maximum of the tan delta peak, which represents the ratio of storage modulus to loss modulus. The tests were conducted using a Q800 dynamic mechanical analyzer (DMA) from TA instruments, New Castle, DE, USA. The tests were conducted in tension mode at a frequency of 1 Hz and an amplitude of 10 μm. The test samples were cut from the films using a laser machine having a dimension of 20 mm (length) by 4.5 mm (wide) and approximate thickness of 0.2 mm. All specimens were initially conditioned at 25 ◦C in the DMA chamber, and then dynamic heating scans were performed from 25 to 150 ◦C at 3 ◦C/min. The test was performed in triplicate and the average reading was reported.

#### **3. Results and Discussion**

Moisture transfer in porous materials takes place due to the presence of a vapour pressure gradient between their top and bottom surfaces. Figure 3 represents the water vapour transmission rate (WVTR) of the prepared membranes. It was seen that, overall, the vapour transmission increased with increasing CF content in the samples. The detailed data analysis of vapour permeability of the composites is reported in Table 2.

The normalized WVTR is an important factor to consider when measuring permeability of composites with different thicknesses. The WVTR of the PLA-TCF was similar to PLA-UCF composites at lower CF concentrations. As the CF concentration increased beyond 10%, the composites with treated fibers showed significantly lower vapor transmission rates. The WVTR of PLA-TCF20 was 73% lower than that of PLA-UCF20. The increase in vapour permeability and WVTR at higher CF concentration is linked to the pore network and structure of the composite. The surfactant treatment resulted in good dispersion of fibers in the PLA matrix, thereby reducing formation of voids and limiting capillary movement of water molecules. The treatment, however, did not completely block microscopic pores in the samples that are needed for vapour diffusion.

**Figure 3.** WVTR of the films with varying CF concentration.


The increase in the water vapour permeability at higher humidity level is related to the enhanced transport of moisture. This phenomenon is induced by the transfer of liquid in the microscopic pores of the material that are filled with water due to capillary condensation [42]. For materials that show hysteresis in their sorption isotherm, it has been reported earlier that their water vapour permeability is dependent on the moisture content [43].

The water absorption of the composites was calculated as percentage of absorption with respect to initial mass. As seen in Figure 4, the water absorption behaviour of the composites was significantly affected by the treatment of CF keeping the WA values to a minimum. PLA-UCF composites showed higher water absorption due to the absence of treatment on CF. PLA-UCF20 showed the highest absorption having a WA value of over 7% for 24 h immersion.

**Figure 4.** Water absorption behaviour of the films with varying CF concentration.

When compared to the treated fiber composite having the same CF concentration (PLA-TCF20), PLA-UCF20 showed a four-fold increase in water absorption. In addition to enhanced dispersion due to the CF treatment, it is evident that the chemical structure of the CF has also been altered, reducing available surface hydroxyl groups that are responsible for the higher water uptake in the samples.

A close examination of the composite morphology revealed that samples without CF treatment showed poor dispersion in the PLA matrix whereas the treatment was found to be effective for preparing homogenous PLA-CF samples. In Figure 5b,c, it can be seen that the untreated fibers and PLA have poor interfacial adhesion and the fibers show agglomeration at the surface. On the other hand, Figure 5d,e show that samples with surfactant treated CF are well dispersed in the PLA matrix.

The presence of the acid phosphate ester surfactant on the surface of nanocellulose facilitates their dispersion in the polymer matrix and improves the nucleation effect [40,44,45]. The surfactant acts as a stabilizing agent for obtaining a stable dispersion of cellulose fibers in the matrix [46,47]. The surfactant treatment has also been found to be effective in preventing the agglomeration of CF in solvents such as chloroform, thereby enhancing dispersibility in the polymer matrix [48].

The thermal properties of the PLA-CF films were investigated by TGA, DSC, and DMA analysis to determine the effect of surfactant treatment on the thermal stability of the polymer composites. The thermal characteristics of the composites are mentioned in Table 3.


**Table 3.** Thermal properties of films and fibers.

**Figure 5.** *Cont*.

**Figure 5.** Micrograph of prepared films (**a**) PLA; (**b**) PLA-UCF10; (**c**) PLA-UCF20; (**d**) PLA-TCF10 and; (**e**) PLA-TCF20.

Thermogravimetric analysis of PLA, untreated and treated CF and the composite films shows a multistep degradation as seen in Figure 6. The initial step (50–100 ◦C) can be attributed to a loss of moisture. The treated CF shows lower thermal stability and weight loss between 250–300 ◦C corresponding to its respective DTG peaks in Figure 7 which are mainly associated with the effect of the surfactant on the fiber [46,48]. The PLA-TCF20 shows a smaller degradation peak corresponding to the surfactant coated fibers in the matrix.

**Figure 6.** TGA curves of films and fibers with and without treatment.

**Figure 7.** DTG curves of films and fibers with and without treatment.

The next main degradation step for all samples occurs between 350 and 400 ◦C, associated with the cellulose, hemicellulose, and thermally stable compounds in the polymer matrix. The main DTG peak corresponding to the peak degradation temperature (Tmax) for PLA was 368 ◦C and for the PLA-CF samples it was between 354 and 372 ◦C, showing that there was no degradation taking place in the composites at lower temperatures. It was observed that the addition of TCF in PLA slightly increases the thermal stability of the matrix due to better reinforcement as seen in Table 3.

Figure 8 shows the endothermic melting peaks for neat PLA, PLA-UCF20, and PLA-TCF20 composites. The three materials displayed similar curves and minor differences in their melting temperatures were observed, indicating that CF dispersion does not significantly affect the thermal properties of the polymer matrix.

**Figure 8.** DSC curves of films with and without CF treatment.

From the DMA analysis, the glass transition temperatures (Tg) of the composites were evaluated using the tan δ peak temperature. It was found that both the fiber content as well as fiber treatment had an effect on the Tg, as seen in Figure 9.

**Figure 9.** Glass transition temperature of films with varying CF concentration.

PLA composites prepared with surfactant treated fibers showed relatively higher Tg when compared to PLA composites containing untreated CF. The improvement in Tg (up to 10%) could be associated with the enhanced CF dispersion and interface compatibility due to surface modification. On the other hand, the PLA-UCF composites showed a lower Tg compared to neat PLA. This could be due to the presence of residual water in the system, which causes a degradation of PLA. Moreover, it was observed that the Tg increased with higher CF content in the PLA composites. The addition of acid phosphate ester-based surfactant has been reported to enhance dispersibility, mechanical, and stress-transfer properties of cellulose reinforced composites [49].

#### **4. Conclusions**

The preparation of bio-based PLA membranes incorporating micro-fibrillated cellulose fiber (CF) using a solvent casting method have been reported in this paper. The effects of CF modification, using an anionic organic phosphoric acid ester-based surfactant, on composite structure, morphology, and properties were investigated. The physical properties of the composite were significantly enhanced when prepared with treated CF. The composites containing treated CF absorbed very low amounts of water, even at 20 wt% CF loading, yet these composites were permeable to water vapour. The CF treatment resulted in a good dispersion of fibers in the PLA matrix, as seen using scanning electron microscopy, thereby reducing void formation and limiting capillary movement. However, the incorporation of treated CF did not completely block the microscopic pores in the composite samples that are needed for vapour diffusion, as seen in the vapour permeability results. The CF treatment also improved the thermal properties of the composite, increasing their glass transition temperature and peak degradation temperature. The melting point of the composites did not change before and after CF treatment. The prepared composites were breathable and showed better resistance to water. Using a relatively inexpensive treatment, smart biobased membranes have been developed that have great potential in the building industry by increasing the hygroscopic performance as well as reducing the embodied energy of buildings. Further work is recommended considering the durability and weathering of the PLA-CF composites as well as the preparation of larger sized films and assessment of their hygroscopic performance in test cells.

**Author Contributions:** Conceptualization, A.H. and P.B.; methodology, A.H. and P.B.; formal analysis, A.H.; investigation, A.H.; resources, P.B.; data curation, A.H.; writing—original draft preparation, A.H.; writing—review and editing, A.H. and P.B.; visualization, A.H.; supervision, P.B.; project administration, P.B.; funding acquisition, A.H. and P.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors are grateful to Natural Sciences and Engineering Research Council of Canada for the financial support through its IRC and CRD programs (IRCPJ 461745-18 and RDCPJ 524504-18) as well as the industrial partner, Kruger Inc., of the NSERC industrial chair on eco-responsible wood construction (CIRCERB). This work was also supported by the funds received from Quebec Research Funds – Nature and Technology (FRQNT) under the Merit Scholarship Program for Foreign Students, Quebec-India Postdoctoral scholarships (PBEEE 2I, 2019-2020).

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**

