Vapor Barrier Membranes Based on Polylactic acid and Cellulose Microfibers for the Building Envelope Application

This study aims to develop fully bio-based barrier membranes from polylactic acid (PLA) and cellulose microfibers (CMF) to control the transmission of water vapor through the building envelope. To improve the dispersibility of CMF in the PLA matrix, lactic acid (LA) monomer or oligomer was grafted onto the surface of CMF by a simple esterification reaction. Based on the morphological analysis, the LA-modified CMF (CMF-LA) showed excellent dispersion in the PLA matrix due to the improved interfacial adhesion between modified fibers and PLA. The results of thermal analyses indicated that the glass transmission temperature and thermal decomposition temperatures of materials enhanced after the addition of bio-fillers. According to the tensile test results, the PLA-based composites incorporated with the different content of CMF-LA (1–20 wt%) and dicumyl peroxide (PLALCD composites) showed higher values of tensile strength and Young’s modulus compared to the neat PLA. Moreover, the PLALCD composites displayed superior vapor barrier properties than the PLA/untreated CMF composites, and there was no significant decrement in the barrier performance of PLALCD composites when the CMF-LA content was up to 15 wt%. Additionally, the environmental impact of the prepared composite was studied by the life cycle assessment tool and results demonstrated that the incorporation of cellulose fibers into PLA reduced the global warming potential of materials. The obtained results suggest that the modification of CMF with LA can be considered a simple, cost-effective, and sustainable method to fabricate a PLA-based membrane for the building envelope application.


Introduction
The building envelope involves a series of materials to promote the indoor quality of buildings and fulfill the functions necessary for durability [1].Moreover, it is the main part of buildings responsible for protecting the interior atmosphere of buildings from external environmental impacts such as rain/snow, wind, sun, humidity, air, and acoustic pollution [2].Some of the most significant functions of building envelopes include air leakage, heat transfer resistance, thermal and mechanical stability, water permeability, and moisture diffusion [3].In a cold climate like Canada, the water vapor concentration inside of buildings is higher than in the exterior environment, leading to moisture flow through the building wall [4].The diffusion and accumulation of moisture in building envelopes cause mold growth, structural damage, and deterioration of building materials [5].The migration of moisture through walls, roof assembly, floors, and ceilings can be controlled by the installation of vapor barrier/ retarder membranes on the warm side of buildings, thereby preventing interstitial moisture condensation [6].
Currently, there are several types of barrier membranes used in the building industry.The plastic sheet of polyethylene (PE) is one of the most common vapor barriers in the market due to its interesting properties such as lightweight, low cost, durability, and good barrier properties [7].However, PE is a fossil-based synthetic polymer and its increasing production trend can lead to a significant negative impact on the environment [8].Consequently, new biobased materials derived from natural resources should be considered in the production of sustainable building envelope systems to alleviate problems associated with fossilbased plastics.Nowadays, biopolymers have received great attention among scientists due to their potential of replacing fossil-based plastics [9].These materials are synthesized from bio-based monomers or extracted from biomass and renewable sources and have similar physicochemical characteristics to conventional synthetic polymers [10].Polylactic acid (PLA), polycaprolactone, and polyhydroxyalkanoates are some of the well-known biopolymers currently used in several applications [11].Among all biopolymers, PLA has attracted considerable interest because of its notable properties such as excellent processability, thermal plasticity, and proper mechanical properties [12].PLA is derived from renewable resources such as corn, sugarcane, starch, and other sugar crops [13].It has comparable mechanical properties to fossil-based plastics such as polyethylene terephthalate and polystyrene [14].However, some disadvantages of PLA including high cost, poor thermal stability, and low melting strength can limit its applications [12].One of the strategies to overcome these drawbacks is the incorporation of renewable fillers (e.g., wood flour or natural fibers) into plastics [15].In this new class of materials, named woodplastic composite (WPC), the bio-fillers not only have a reinforcement effect but also improve sustainability and reduce the weight of final products [16].
Cellulose is one of the most abundant biomass materials in nature that can be extracted from wood pulp in the form of long micro/nanofibers [17].Over the past few decades, several studies investigated the possibility of replacing synthetic fibers like glass, carbon, and Kevlar with natural fibers [18].Cellulose microfibers (CMF) are ideal reinforcement for the fabrication of fully bio-based WPC due to their important advantages such as renewability, abundance, low cost, and excellent mechanical and rheological properties [19].However, the main challenges to producing CMF-reinforced polymer composites are: (i) their high moisture sensitivity and (ii) controlling the dispersion of hydrophilic cellulose fibers when blending them in a non-polar polymer [20].CMF has a large content of hydroxyl groups, and they can absorb moisture from the surrounding environment, thereby leading to fungal growth and swelling of fibers [21].Moreover, CMF tends to aggregate in a hydrophobic polymer matrix due to the strong intermolecular hydrogen bonding between the fibers [22].These undesirable phenomena could decrease the reinforcing effects of CMF, and consequently, lead to the deterioration of the mechanical and barrier properties of composites.These difficulties could be addressed by the surface modification of CMF to improve the compatibility and interfacial adhesion between fibers and the polymer matrix [23].Recently, numerous studies have investigated the modification effect of cellulose fibers with esterification [24], oxidation/amidation [25], silanization [26], alkalization [27], sol-gel process [28], surfactant [29], plasticizer [30], and polymer grafting [31], etc., and the obtained results indicated the improved dispersion of modified fibers in the polymer matrix.For instance, Sethi et al. reported a rapid and eco-friendly sustainable method for the functionalization of cellulose nanofibers (CNFs) with lactic acid (LA).The nano-papers prepared from the LAmodified CNFs were water-resistant and hydrophobic in nature and had superior mechanical properties and thermal stability under moist conditions [32].Moreover, Wu et al. modified cellulose nanocrystals (CNC) by LA to produce PLA/CNC nanocomposites.The results demonstrated that the crystallization rate of PLA was remarkably enhanced after the addition of modified CNC into PLA.Additionally, the mechanical properties of nanocomposites were improved due to the excellent compatibility of LA-modified CNC with PLA [33].
The main objective of this study is the development of fully bio-based vapor barrier membranes from PLA and CMF that can accomplish more sustainable products.The work also involves the surface modification of CMF with LA using a simple esterification reaction.The PLA-based membranes were fabricated by the solution casting method and then their properties were characterized in terms of morphology, chemical structure, thermal stability, mechanical properties, and barrier performance.In addition, the environmental impact (global warming potential) of developed membranes was investigated by the life cycle assessment tool.

Modification of CMF with LA
Firstly, 5 g of CMF (weight of dried fibers) was added to 500 mL deionized (DI) water, and the suspension was mixed in a high-shear mixer at 2,500 rpm for 20 min.Then, 5 g of LA (CMF:LA (1:1)) and 50 mg of zinc acetate dihydrate (catalyst) were sequentially added to the prepared CMF suspension and the mixture was sonicated for 5 min at room temperature.To promote the esterification reaction, the mixture was transferred to an oil bath with a temperature of 170 °C and mixed in an Ultra-Turrax disperser at 18,000 rpm.The reaction was continued for 10 min and then stopped by adding 20 mL DI water to the mixture.Subsequently, the mixture was centrifuged at 14,000 rpm for 2 min, vacuum filtered, followed by washing three times with DI water to completely remove the unreacted LA.Finally, the LA-modified CMF (coded as CMF-LA) was freeze-dried for two days and then ground in a grinder (Retsch ZM100, Germany) to obtain the CMF-LA powder.Figure 1 shows the schematic image of the modification procedure of CMF with LA and the esterification reaction.

Preparation of PLA-based Composites
The bio-based composites were fabricated by a solutioncasting method.Priorly, PLA pellets and bio-fillers were vacuum-dried overnight at 70 °C to remove moisture.The desired amount of PLA was dissolved into 100 mL CHCl 3 under stirring at 600 rpm for 2 h at room temperature.Then, different content of untreated CMF or CMF-LA (1, 5, 10, 15, and 20 wt%) were added to the PLA solution and the mixture was stirred at 1,000 rpm for 1 h at room temperature.Finally, the prepared solutions were cast onto a glass petri dish and dried by solvent evaporation for two days in a fume hood at room temperature.The PLA-based composites incorporated with untreated CMF and CMF-LA were coded as PLACx and PLALCx, respectively (x indicates the content of bio-fillers).Additionally, to effectively graft PLA on the surface of CMF-LA, 1 wt% of DCP (radical initiator) was added to the PLA/CMF-LA solution.Then, the mixture was heated under reflux at 80 °C to initiate the decomposition of DCP while stirring at 1,000 rpm for 2 h.This series of composites were coded as PLAL-CDx.The formulation of all samples is given in Table 1.In each formulation, the total weight of all components of composites was 5 g.Before the characterization, the prepared composites were vacuum-dried at 40 °C for 4 days to remove the remaining solvent/moisture and then stored in a zipped airtight bag.

Scanning Electron Microscope (SEM)
The morphology of PLA-based composites and CMF before and after the modification was examined using a scanning electron microscope (SEM), JEOL 6360LV (Hitachi, Japan).Before the observation, the surface of the samples was coated with a thin layer of gold by an EMS 950x sputter coater (Hatfield, PA, USA).The analysis was operated under vacuum conditions and an accelerating voltage of 15 kV.

Contact Angle Measurement
The contact angle test was used to study the effect of modification on the hydrophilicity nature of CMF.Moreover, to examine the contact angle of CMF-LA after removing the self-polymerizing LA on the surface of fibers, the CMF-LA was dispersed in CHCl 3 for 2 h at 600 rpm.Then, the mixture was vacuum filtered, followed by washing several times with solvent, and then vacuum-dried in the oven overnight at 70 °C.Thereafter, 0.1 g of dried bio-fillers (untreated CMF and CMF-LA before and after dispersing in solvent) were compressed for 5 min at 30,000 KN into a disc of 13 mm in diameter and 0.5 mm in thickness, using the MTS Alliance RT/5 (Eden Prairie, Minnesota, USA).Three discs were prepared for each bio-filler.Then, the contact angle of a DI water droplet was measured by a contact angle goniometer, FTA200 (Portsmouth, VA, USA).The contact angle values were determined from the instrument's software (FTA32) and the average values of contact angle versus time were reported over 10 s.

Fourier-transform Infrared Spectroscopy (FT-IR)
A Fourier transform infrared spectroscopy (FT-IR), Bruker INVENIO® R (Billerica, MA, USA) was used to investigate the chemical structure of bio-fillers (untreated CMF and CMF-LA before and after dispersing in solvent) and PLAbased composites.The scans were collected in absorbance mode from 4000 to 400 cm −1 at 4 cm −1 resolutions.

Thermogravimetric Analysis (TGA)
The thermal stability of bio-fillers and PLA-based composites was studied using a thermogravimetric analysis (TGA), TGA/DSC 3 + Mettler Toledo instrument (Mississauga, ON, Canada).The samples were heated from 30 to 930 °C (heating rate of 10 °C/min) under nitrogen purging of 50 mL/min.

Differential Scanning Calorimetry (DSC)
The differential scanning calorimetry (DSC) analysis was used to characterize the thermal properties of bio-fillers and PLA-based composites.The DSC analysis was performed using a DSC Mettler Toledo 822/e (Columbus, OH, USA).The materials were heated from 25 to 400 °C at a heating rate of 10 °C/min under nitrogen purging of 50 mL/min.

Dynamic Mechanical Analysis (DMA)
The thermomechanical behavior of PLA-based composites was studied by the dynamic mechanical analysis (DMA) with a TA Instruments Q800 (New Castle, DE, USA).In this regard, samples (thickness of ~ 150 μm) were cut into a rectangular shape (2 cm × 0.45 cm) using a laser machine.The measurement was performed in tension mode at an amplitude and frequency of 10 μm and 1 Hz, respectively.The composites were heated from 30 to 100 °C at a heating rate of 3 °C/min.

Tensile Test
The mechanical performance of PLA-based membranes was determined by means of the tensile test according to the ASTM D882 standard [34] using the MTS QTest/5 (Eden Prairie, Minnesota, USA).Each specimen (thickness of ~ 150 μm) was cut into a rectangular shape (10 cm × 1 cm) using a laser machine.The analysis was performed at a crosshead speed of 12.5 mm/min with a load cell of 500 N.The tensile test was performed in triplicates (n = 3) and the average values of tensile strength (MPa), Young's modulus (MPa), and elongation at break (%) were reported.

Water Vapor Transmission Rate (WVTR)
The barrier performance of developed PLA-based membranes was studied using the water vapor transmission rate (WVTR) test according to ASTM standard E96 (methods A and B) [35].All membranes (thickness of ~ 150 μm) were mounted in the mouth of a permeability cup with an internal diameter of ~ 6 cm.The surrounding specimen's area was sealed with silicone sealant to avoid the migration of water vapor through boundaries.In method A (desiccant method), the cup was filled with ~ 102 g of CaCl 2 desiccant within 6 mm of the specimen.In method B (water method), the cup was filled with distilled water to a level of 19 ± 6 mm below the specimen.The relative humidity inside of the cup in methods A and B were 0% and 100%, respectively.All sample cups were weighed (W 0 ) and then placed in a controlled temperature/humidity chamber (21 °C/40% RH).The weight of cups was measured every day for three weeks (W n ) and then the graphs of gained water (W n −W 0 ) and evaporated water (W 0 −W n ) versus time were plotted for methods A and B, respectively.The WVTR value was calculated using the equation below: where G is the steady-state weight change (g), t is time (h), and A is the cup mouth area (0.00255 m 2 ).To compare the WVTR values of membranes with different thicknesses, the normalized WVTR value was described by eliminating the thickness factor as follows: where d is the thickness of each membrane (m).Three replications were considered for each composite (n = 3) and the average values of normalized WVTR were reported.

Life Cycle Assessment (LCA)
The environmental impact of fully bio-based membranes was investigated using the life cycle assessment (LCA) tool, openLCA software (version 1.11.0).The LCA utilized in this study is in accordance with the ISO 14,040 and ISO 14,044 standards [36].In this study, the developed PLA-based membranes were selected for the case study.To reduce the complexity of LCA analysis and create insights faster about the internal process, only the extraction and manufacturing/processing steps of materials were considered in this study (Cradle-to-gate analysis).
The background information on the production of each component (including PLA, CMF, and LA) was obtained by referring to the Ecoinvent 3.8 database (Cut-off system model).The name of each component of membranes in the Ecoinvent database is provided in Table 2.The (2) NormalizedWVTR = WVTR × d functional unit of 2.5 g of membranes was considered for the LCA analysis.For instance, the PLALC10 membrane consists of 2.2, 0.25, and 0.05 g of PLA, CMF, and LA, respectively.

Characterization of CMF
The morphology of cellulose fibers before and after the modification was characterized by SEM and the results are presented in Fig. 2. As can be seen, the untreated CMF showed large aggregates and a network of entangled fibers (Fig. 2a-c).The presence of aggregates is due to the van der Waals force and hydrogen bonding between hydroxyl groups of CMF [31].Figure 2d-f indicates that the morphology of cellulose fibers became more defined and uniform after the modification with LA.This result might be attributed to the replacement of OH groups with LA oligomers on the surface of CMF that reduced the formation of hydrogen bonding and entanglements among fibers.
Figure 3 shows the contact angle test results of bio-fillers including untreated CMF and CMF-LA before and after dispersing in solvent (CHCl 3 ).As can be observed, the contact angle of untreated CMF was 16.33 ± 2.1° (Fig. 3a) at 5 s, whereas the value enhanced to 46.13 ± 1.1° (Fig. 3b) after the modification of CMF with LA.According to Fig. 3d, CMF-LA showed higher values of contact angle than untreated CMF in the range of 2-10 s.The hydrophilic nature of CMF is attributed to the presence of abundant hydroxyl (OH) groups on the surface of cellulose fibers.After the LA-treatment, OH groups of CMF were replaced with LA oligomers that could reduce the hydrophilicity of cellulose fibers [32,33].Moreover, it can be seen in Fig. 3c that the contact angle value slightly decreased to 41.01 ± 0.8° (at 5 s) after the dispersion of CMF-LA in CHCl 3 .The result might be due to the removing the hydrophobic polymerized PLA with low molecular weight on the surface of CMF [33].
The FT-IR results of untreated CMF and CMF-LA before and after dispersing in CHCl 3 are provided in Fig. 4. In the spectra of untreated CMF, the absorption bands at 3320, 2888, 1090, and 1028 cm −1 are assigned to the O-H and C-H, and C-OH stretching of secondary and primary alcohols of cellulose, respectively [37].In the spectra of CMF-LA, the intensity of the O-H band decreased after the LA-treatment.Moreover, new bands at 1747 and 1750 cm −1 appeared in the spectra of CMF before and after dispersing in CHCl 3 , respectively, which corresponded to the stretching vibration of the ester group.The appearance of the ester band around 1750 − 1735 cm −1 confirms the successful esterification reaction between CMF and LA [32].Similarly, Wu et al. reported that a band at 1740 cm −1 (the carbonyl stretching area) appeared after the modification of cellulose nanocrystals (CNC) with LA, indicating a successful grafting of LA oligomers on the surface of CNC [33].The esterification reaction between CMF and LA was illustrated in Fig. 1b.During the modification of CMF, the esterification reaction began when the suspension of CMF and LA was heated at 170 °C.Then, DI water was gradually removed from the mixture and LA oligomers were grafted onto the surface of CMF [32,33].
Figure 5a shows the TGA results of CMF before and after the modification with LA.In the temperature range below 100 °C, the mass of fibers was slightly decreased, which is due to the moisture removal.The untreated CMF was thermally decomposed in a single step from 280 to The first heating DSC thermograms of CMF before and after the LA-treatment are provided in Fig. 5b.The untreated CMF showed a decomposition temperature (T d ) of 340.1 °C, which is attributed to the breakage of glycosidic bonds in cellulose [31].After the modification, the value of T d increased to the value of 345.3 °C.These observations are in line with TGA results, indicating the improved thermal stability of CMF after the modification with LA.Wu et al. have also reported that LA-modified CNC showed higher thermal stability in comparison with untreated CNC [33].

SEM Test
Figure 6 presents the photographic images of all prepared composites containing up to 20 wt% of bio-fillers.It can be observed from this figure that the dispersion of cellulose fibers in PLA remarkably improved after the modification of CMF with LA.SEM analysis was also conducted to study the morphology of composites incorporated with 10 wt% of untreated CMF or CMF-LA.Figure 7 presents the SEM images of the top surface and tensile fracture surface of PLAC10, PLALC10, and PLALCD10 composites.The SEM images of PLAC10 (Fig. 7a and b) indicate a poor dispersion of untreated CMF where fibers were agglomerated in the PLA matrix.Moreover, a clear formation of voids can be observed around the fibers on the tensile fracture surface of the PLAC10 composite (Fig. 7c), confirming a poor interfacial adhesion between fibers and PLA.Our previous investigation also revealed that untreated CMF showed poor dispersibility in the PLA matrix [29,31].SEM images corresponding to the PLALC10 composite (Fig. 7d and e) demonstrate that the agglomeration of CMF was reduced after the modification of CMF with LA.It can be concluded that the presence of grafted LA oligomers on the surface of CMF led to improved interfacial compatibility with the hydrophobic PLA matrix.Likewise, SEM images of the PLALCD10 composite indicate that CMF-LA were homogeneously dispersed in PLA, and the  7g and h).Moreover, the SEM image of the fracture surface of PLALCD10 (Fig. 7i) exhibits that there was no obvious void around the fibers.These results might be due to the presence of DCP (radical initiator) that led to the formation of a cross-linked/branched structure between PLA and grafted LA oligomers on the surface of CMF and resulted in the enhancement of matrix-CMF interfacial adhesion.Similarly, Zheng et al. reported that the addition of DCP into poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV)/CNC composites significantly improved the dispersion of CNC in the PHBV matrix [39].
It is worth noting that the presence of void in PLA/CMF composites can negatively affect the barrier properties since it creates a preferential pathway for the migration of water vapor molecules [40].The above observations suggest that the modification of CMF with LA presents a simple and effective method for the fabrication of PLAbased membranes with an improved dispersibility of fibers, thereby reducing the void formation and the deterioration of the barrier performance of materials.

FT-IR Test
The chemical structure of the neat PLA and its composites incorporated with 10 wt% of bio-fillers was studied by the FT-IR analysis and the results are presented in Fig. 8.The neat PLA showed absorption bands at 2999, 2950, and 1753 cm −1 that are assigned to the ν asC-H , ν sC-H , and ν C=O vibrations, respectively.Moreover, the peaks at 1259, 1087, and 869 cm −1 are corresponding to the ν COC , ν sCOC , and ν C−COO , respectively, which is in agreement with previous studies [12].No major differences can be observed between the spectra of the neat PLA and its composites.However, in the spectra of the PLAC10 composite, a new band appeared at 3012 cm −1 that is attributed to the O-H stretching band of cellulose.This observation might be because of the formation of intermolecular hydrogen bonds between the terminal hydroxyl and carbonyl groups of PLA and OH groups of CMF [41].According to the spectra of PLALC10 and PLA-LCD10 composites, the O-H stretching band of CMF was disappeared in these composites.It could be concluded that OH groups of CMF were mostly replaced by the LA oligomers after the modification, and as a result, more interfacial interaction can be formed between PLA and LA-modified CMF.
Improving the chemical interactions in fiber-reinforced polymer composites is a very important consideration when designing a vapor barrier membrane.As mentioned earlier, poor interfacial adhesion can lead to the formation of voids in composites, which has a significant negative impact on the thermomechanical and barrier performance of materials.Dhar et al. reported that the presence of DCP in PLA/CNC composites could induce PLA grafting on the CNC surface, thereby leading to the improvement of the interfacial interaction between PLA and CNC [38].To explain the effect of DCP in the PLA/CMF-LA composite, the possible chemical reaction between PLA and CMF-LA is provided in Fig. 9.As shown in this figure, DCP decomposes into free radicals at high temperatures.Then, the generated peroxide radicals can facilitate the abstraction of hydrogen from PLA and LA oligomers on the surface of CMF that forms free radicals on the backbone of components.As a result, the propagation of radical reactions causes the grafting of PLA onto the CMF-LA surface while forming a cross-linked/branched structure between PLA and LA oligomers [42].Therefore, the presence of DCP in the composite can improve the interfacial compatibility between PLA and CMF-LA, and consequently, lead to a uniform dispersion of modified fibers in the PLA matrix.

TGA and DSC Tests
The thermal stability of PLA-based composites was investigated by TGA analysis, and the thermogravimetric curves and characteristic temperatures of materials are provided in Fig. 10a-c; Table 3, respectively.All composites showed thermal degradation from 310 to 380 °C.The initial mass loss at temperatures below 100 °C is because of the evaporation of water in samples.As can be seen, the neat PLA displayed a single thermal degradation step with a T max located at 365.9 °C, which is due to the hydrolysis and oxidative chain scission of PLA.The degradation temperature of the neat PLA at 10% (T 10 ) and 90% weight loss (T 90 ) are 149.4 and 374.1 °C, respectively.In the TGA thermograms of composites, it is evident that the incorporation of either untreated CMF or CMF-LA into PLA enhanced the onset (T 10 ) and end-set (T 90 ) decomposition temperatures of materials.A similar result was reported by Botta et al. in PLA/graphene composites, where the thermal degradation temperatures of materials were shifted towards higher values in the presence of graphene nanoparticles [43].It can be concluded that the addition of highly thermally stable reinforcement (CMF) into PLA protected polymer chains from thermal decomposition, thereby improving the thermal stability of materials [31].
A closer look at Table 3 reveals that the PLALCD composite showed higher values of thermal decomposition temperatures compared to the PLAC and PLALC composites in low-and high-temperature ranges.This might be due to the presence of DCP in the PLALCD composite, which led to the grafting of PLA chains onto the surface of CMF-LA.Moreover, Dhar et al. demonstrated that grafting PLA onto CNC by the addition of a small amount of DCP remarkably improved the thermal stability of materials [38].Therefore, Fig. 9 Schematic of the chemical reaction between PLA and CMF-LA the presence of DCP can form a network-like structure in fiber-reinforced PLA composites that protect materials from temperature rise in the event of a fire by increasing the required activation energy level for the thermal decomposition of materials [44].
The effect of bio-fillers on the thermal properties of PLAbased composites was investigated by the DSC analysis and the first heating DSC curves and thermal characteristics of materials are given in Fig. 10d-f; Table 3, respectively.The neat PLA exhibited a glass transition (T g ) and melting peak (T m ) at 45.6 and 148.7 °C, respectively.It can be observed in Table 3 that the melting temperature of PLA changed insignificantly after the incorporation of either untreated CMF or CMF-LA, while the T g of PLA-based composites shifted to higher temperatures.These results could be because of the chemical interaction between bio-fillers and PLA, thereby It should be noted that the increase of T g was more significant in the case of PLALCD composites so that, with increasing CMF-LA loading from 5 to 20 wt%, the value of T g enhanced from 50.3 to 53.4 °C.It might be explained that the formation of cross-linked/branched structures and grafting of PLA onto the CMF-LA surface could restrict the segmental movements of PLA, and consequently, shift the T g to the higher values.

DMA and Tensile Tests
The dynamic mechanical properties of PLA-based composites were investigated by DMA analysis and the storage modulus (E′) versus temperature curves are presented in Fig. 11.Moreover, Table 4 shows the T g and of E′ (at 30 °C) of materials.In all composites, the significant drop of E′ at temperatures around 50-85 °C is due to the glass transition effect of PLA.According to Table 4, the E′ value for the neat PLA was 1345.9MPa, and this value significantly decreased after the addition of untreated CMF or CMF-LA into PLA.For instance, the E′ value of PLAC composites incorporated with 5, 10, and 20 wt% of untreated CMF were 1163.5, 896.3, and 669.4 MPa, respectively.The results might be because of the agglomeration of untreated CMF in the PLA matrix that negatively affects the storage modulus of materials [31].According to the DMA test results, the T g of PLA decreased to ~ 61 °C after the addition of 5 wt% of untreated CMF or CMF-LA; however, the T g enhanced to nearly 63-65 °C at higher loading of bio-fillers (10 and 20 wt%).The increase in T g by the incorporation of bio-fillers was previously confirmed using the DSC analysis.It should be mentioned that the PLALCD20 composite exhibited the highest T g value (65.9 °C) among all composites.As mentioned earlier, this result could be explained by the improvement of the interfacial adhesion between CMF-LA and PLA in the presence of DCP which can restrict the mobility of polymer chains, thereby increasing the glass transition temperature of PLA [45].
To estimate the mechanical properties of developed PLAbased membranes, the tensile test was carried out and the results including stress-strain curves and mechanical characteristics are given in Fig. 12 and Table 4, respectively.As can be observed in Fig. 12a, the neat PLA showed Young's modulus, tensile strength, and elongation at break (%) values of 2.5 GPa, 23.4 MPa, and 37.5%, respectively.In the PLAC composites, the mechanical properties dramatically decreased when more than 5 wt% of untreated CMF was added into PLA.This is due to the poor compatibility and agglomeration of fibers in the PLA matrix that form high local stress around fibers in composites [46].The SEM image of the tensile fracture surface of the PLAC10 composite (Fig. 7c) also confirmed the presence of fiber agglomerates in the PLA matrix.As can be seen in Table 4, the PLALC composites showed superior mechanical properties than the PLAC composites.Interestingly, further improvements in the mechanical properties were achieved in the PLALCD composites.For instance, the PLALCD20 composite (incorporated with 20 wt% CMF-LA and 1 wt% DCP) exhibited higher values of tensile strength and modulus by ~ 34% and ~ 29%, respectively, in comparison with the neat PLA.The results can be explained by the following reasons: (i) the grafting of PLA onto the CMF-LA surface in the presence of DCP led to enhanced interfacial adhesion between the modified fibers and the PLA matrix and (ii) the DCP-induced chain-extension and formation of polymer cross-linking can contribute to the increase in the mechanical properties of materials.Zheng et al. also reported that the addition of 1% DCP into PHBV/CNC composites significantly enhanced the values of tensile strength and modulus because of the efficient polymer chain entanglement and improved interfacial adhesion at the interface PHBV and CNC.
The mechanical characteristics of barrier membranes, such as modulus and tensile strength should be taken into account when specifying materials [47].The commercial polyethylene (PE) vapor barrier 6 mil (0.1524 mm of thickness) currently used in the building industry has a tensile strength value of 19.1 ± 1.4 MPa.According to Table 4, the PLALCD composites incorporated with the different content of CMF-LA displayed higher values of the tensile strength (24-45 MPa) compared to the neat PLA or the commercial PE vapor barrier.Moreover, the vapor barrier membranes should present proper flexibility to cover building walls and move with the building [48].It has been reported that the Fig. 11 DMA storage modulus curves of PLA-based membranes required range of elongation for these membranes is from 4 to 200% based on their scope of use in a building [49].It was observed from the results that the incorporation of biofillers into PLA reduced the flexibility of materials.This might be due to the formed cross-linked/branched structures in composites that restricted the mobility of PLA segments to dissipate energy under the tensile force, thereby leading to the reduction of the flexibility of composites [39].Similarly, Wu et al. reported that the addition of high-modulus CNC into PLA reduced the flexibility and toughness of nanocomposites [33].However, the PLALCD composite having reasonable flexibility (~ 1.5−2%) offered superior mechanical strength and modulus compared to the neat PLA and other composites (PLAC and PLALC composites) and it could be considered a potential candidate for the fabrication of vapor barrier membranes in the building envelope application.

WVTR Test
The barrier performance is one of the key characteristics of polymeric membranes.In a building wall, moisture transfer takes place due to the water vapor concentration differences existing between the two sides of building envelopes.The installation of vapor barrier membranes on the interior side of building walls can control and retard the transmission rate of moisture diffusion and consequently avoid mold growth, excessive moisture accumulation, and deterioration of building envelope materials [50].There are several parameters that affect the barrier properties of plastic sheeting membranes, including hydrophobic/hydrophilic nature, crystallinity, porosity, fiber orientation, fiber/matrix adhesion, and the volume fraction of fibers [51].
The vapor barrier properties of developed PLA-based membranes were studied using the water vapor transmission rate (WVTR) test (water and desiccant methods) and the obtained results are presented in Fig. 13.In the water method (wet cup) (Fig. 13a), the normalized WVTR values displayed a continuous increase with untreated CMF loading.For instance, the WVTR value of neat PLA increased by 547% in PLAC20 composite.This is because of the hydrophilic characteristic of cellulose fibers and the formation of voids between PLA and fibers, resulting in a capillary movement of water vapor molecules.Previous studies also reported a higher transmission rate of water vapor in Fig. 12 Stress-strain curves of the neat PLA and PLA-based membranes polymer composites when incorporated with cellulose-based reinforcement [29].However, the PLALC and PLALCD composites showed superior barrier performance compared to the PLAC composites.These observations could be explained by the following factors: firstly, the OH groups of CMF were replaced by LA oligomers, which decreased the hydrophilicity of CMF (as shown in the contact angle test results).Secondly, the improved dispersion of modified fibers in the PLA matrix reduced the formation of voids and defects in composites that provides migration pathways for water vapor molecules.Interestingly, in the PLALCD composites, there was no significant increase in the value of normalized WVTR when the CMF-LA content was up to 15 wt%.This might be attributed to the better interfacial interaction and more effective chemical reaction between the modified cellulose fibers and PLA in the presence of DCP.According to the national building code of Canada, the value of water vapor permeance of barrier membranes should not be greater than 15 ng/Pa.s.m 2 , which is equal to the normalized WVTR value of ~ 3.16 g/h.m for a membrane with a thickness of 150 μm [52].Figure 13 demonstrates that the normalized WVTR values of all prepared composites were less than 0.0015 g/h.m.Therefore, PLA-based composites could be considered appropriate candidates for the development of barrier membranes for building envelope applications.
It is worth mentioning that the migration of moisture through plastic materials is based on the absorption-diffusion-desorption mechanism.The membranes with the ability to release diffused moisture in their structures can avoid moisture buildup, interstitial condensation, and the deterioration of building envelope materials [53].In this regard, the WVTR test desiccant method (dry cup) was carried out to measure the drive moisture into a heated dry building from the exterior (for instance during a pouring rain).As can be seen in Fig. 13b, the addition of bio-fillers increased the transmission rate of moisture through the membranes.However, PLALCD composites showed lower values of normalized WVTR, suggesting improved barrier performance.

LCA Study
The information regarding the environmental impact of materials helps decision-makers in choosing sustainable materials at the early stage of designing a product.Nowadays, the life cycle assessment (LCA) became an accepted and powerful management tool to analyze the environmental performance of a product, including the material extraction, production, use phase as well as end-of-life stage [54].In this regard, the LCA tool was used to investigate the environmental impact of developed PLA-based membranes.Figure 14 shows the global warming potential (GWP) values Fig. 13 The normalized WVTR values of PLA-based membranes: a water method (wet cup) and b desiccant method (dry cup) Fig. 14 The global warming potential from LCA study of PLA-based membranes of materials.Overall, it is evident that the incorporation of either untreated CMF or CMF-LA into PLA mitigated the GWP value of composites.Indeed, the GWP values are generally reduced by increasing the content of bio-fillers and replacing the portion of PLA with cellulose fibers in composites.For instance, the PLA-based membranes incorporated with 15 wt% of CMF showed a decrease of 15.2% in the GWP value.It is commonly accepted in LCA investigation that a difference higher than 10% can be considered significant in the global warming scores [55].Therefore, the development of membranes from PLA and cellulose fibers has shown significant potential for producing a sustainable vapor barrier membrane with lower embodied energy.

Conclusion
The object of this study was to develop fully bio-based barrier membranes for building envelope applications.In a cold climate like Canada, the water vapor concentration inside the building is higher than outside; therefore, a vapor barrier membrane should be installed on the interior side of the building to control the migration of moisture through the building walls.In this regard, plastic barrier membranes were fabricated from a biopolymer (PLA) and a renewable reinforcement (CMF) via the solvent casting method.To improve the compatibility between CMF and PLA, the bio-filler was modified by LA using a simple esterification reaction.Moreover, the effect of LA-treatment on the morphology, thermomechanical, and barrier properties of membranes was investigated.The morphological studies demonstrated that the modified CMF homogenously dispersed in the PLA matrix, thereby reducing the formation of voids in composites.According to the TGA results, the thermal stability of materials improved thanks to the addition of biofillers into PLA.The tensile test results indicated that the PLALCD composites incorporated with different content of CMF-LA (1-20 wt%) had higher values of the tensile strength (24-45 MPa) compared to the neat PLA (23 MPa), indicating the improved mechanical performance of materials.Regarding the WVTR test, PLALCD composites displayed superior barrier properties than PLAC and PLALC composites, and no significant decrement in the barrier performance of PLALCD composites containing up to 15 wt% of CMF-LA was observed when it is compared with the neat PLA.Additionally, the provided data regarding the life cycle assessment tool revealed that the addition of bio-fillers into PLA significantly reduced the global warming potential of materials.Therefore, it could be concluded that the developed membranes from PLA and LA-modified CMF have potential in barrier membrane applications as an alternative to fossil-based materials.Further work is recommended to fabricate membranes by a solvent-free process such as the extrusion method.Moreover, it would be interesting to study the durability as well as mechanical and barrier properties of membranes after aging since they may be in service for the whole building's service life.

Fig. 1 a
Fig. 1 a Schematic illustration of the modification of CMF with LA and b the esterification reaction of CMF and LA

Fig. 2 Fig. 3
Fig. 2 SEM images of: a, b, and c untreated CMF and d, e, and f CMF-LA

Fig. 5 aFig. 6
Fig. 5 a TGA curves and b DSC first heating scans of untreated CMF and CMF-LA

Fig. 7 Fig. 8
Fig. 7 SEM images of top surface and tensile fracture surface of: a, b, and c PLAC10, d, e, and f PLALC10, and g, h, and i PLALCD10

Fig. 10 a
Fig. 10 a, b, and c Thermal degradation TGA curves and d, e, and f first heating DSC scans of PLA-based membranes

Table 1
The formulation of PLA-based composites incorporated with untreated CMF or CMF-LA

Table 2
The Inventory data of materials

Table 3
Thermal parameters of CMF, LA-CMF, and PLA-based membranes from TGA and DSC analyses