Emulsion-polymerized flexible semi-conducting CNCs–PANI–DBSA nanocomposite films

Abstract

We have developed an essentially green, bottom-up approach for synthesising flexible, organic, semi-conducting nanocomposite films based on cellulose nanocrystals (CNCs) and polyaniline (PANI) through aqueous emulsion polymerization. Dodecylbenzene sulfonic acid (DBSA) was used as surfactant and dopant for the emulsion. CNCs and DBSA micelle concentrations, their structural organization and alignment with the monomer in the emulsion have a significant effect on the physical and mechanical properties of the resulting nanocomposite films with electrical conductivity reaching as high as 5.29 × 10⁻¹ S cm⁻¹ which falls in the electrical conductivity range of germanium (2.24 × 10⁻² S cm⁻¹) and silicon (0.43 × 10⁻⁵ S cm⁻¹) [S. L. Kakani, Electronics Theory and Applications, New Age International, 23–24, 2005]. The CNCs–PANI–DBSA nanocomposite films show a maximum tensile stress and strain values of 22 MPa and 0.89%, respectively and are significantly stronger and more flexible films than those obtained with, for instance, graphene/polyaniline composite paper or graphene paper, where it has been reported in the literature that the tensile strengths were 12.6 and 8.8 MPa, and maximum strains, 0.11 and 0.08%, respectively [Wang et al., ACS Nano, 2009, 3, 1745].

---

Introduction

Polyaniline (PANI), an inherently conducting polymer (ICP), has received extensive attention owing to its ease of synthesis, both chemically and electrochemically, and for its excellent chemical stability and relatively high electrical conductivity. Polyaniline has been used in many applications including anticorrosion coatings, batteries, sensors, separation membranes, and antistatic coatings.^1,2 However, despite its unique chemical and electronic properties, polyaniline has major processability limitations and inferior mechanical properties compared to conventional polymers. To date, great research efforts have been expended to overcome these limitations and to produce conductive nanocomposites and films based on ICPs including polyaniline, polypyrrole, polythiophene, polyacetylene and poly(p-phenylene vinylene) combined with a nanomaterial such as carbon nanotubes,^3,4 nanoclay^5,6 and graphene oxide⁷ for application in supercapacitors, light emitting composites or as low-cost disposable sensors.

Burgeoning interest in developing sustainable organic electronics has given impetus to exploring different forms of cellulosic materials. Cellulose nanocrystals, CNCs,‡ typically produced using strong sulfuric acid hydrolysis of bleached chemical pulp fibres or from tunicate or cotton fibres, possess excellent mechanical properties, as well as unique optical properties, self-assembly characteristics and a high degree of crystallinity.^8,9 Hydrogen bonding between the cellulose chains can stabilize the local structure in CNCs and play a key role in the formation of crystalline domains. CNCs can be regarded as anionic polyelectrolytes owing to the negatively charged sulfate ester groups on their surface. The surface charge is critical for the electrostatic repulsions between colloidal CNCs particles, which are, in turn, responsible for creating stable aqueous suspensions.¹⁰ A blended mixture of polyaniline or poly(p-phenylene ethylene), PPE, camphorsulfonic acid and a sulfate-functionalized tunicate CNCs in formic acid has been reported to produce a conductive film with electrical conductivities in the range of: 1.3 × 10⁻² to 5 × 10⁻² S cm⁻¹. This film also has good mechanical stiffness with elastic modulus of 9.5 GPa at 99.4% w/w of tunicate CNCs for PANI–CNCs and of ∼2.5 GPa for PPE–CNCs.¹¹ Moreover, CNCs have been used with poly(3,4-ethylenedioxythiophene)/poly(styrene sulfonate) (PEDOT:PSS)¹² or PEDOT alone¹³ to produce conductive films, where it was found that the electrical percolation threshold of PEDOT:PSS could be reduced by CNCs addition owing to their templating effect.¹²

Other forms of cellulose, primarily fibrous webs like bacterial cellulose (BC), microfibrillated cellulose (MFC) and cellulose nanofibrils (CNF), have been examined as flexible substrates for electronic materials. For instance, bacterial cellulose has been used to prepare (i) flexible PANI–BC membranes via in situ oxidative polymerization,¹⁴ (ii) electroactive cellulose-based polythiophene composites via FeCl₃-initiated oxidation co-polymerization with thiophene co-monomers,¹⁵ or (iii) core–sheath hybrid composites with polypyrrole (PPy) via in situ oxidative polymerization in a miscible two-phase system of dimethyl formamide (DMF) and water using ferric chloride (FeCl₃) as oxidant catalyst.¹⁶ Different protonic acids could be used in such constructs, e.g. p-toluenesulfonic acid (PTSA)^17,21 and dodecylbenzene sulfonic acid (DBSA),^18,20 as well as different oxidizing agents, e.g., (NH₄)₂S₂O₈, FeCl₃·6H₂O.^14,19 Similar approaches have been adopted to prepare conductive films using MFC and PPy via in situ chemical polymerization²⁰ or CNF and PANI via in situ oxidative polymerization.²¹

To the best of our knowledge, all methods reported in literature to prepare CNCs-based conductive nanocomposites primarily involve physical blending, layer-by-layer (LbL) deposition, electrochemical-co-deposition, latex technology or in situ oxidative polymerization.^{11–13,22,23} However, emulsion polymerization and the mechanism involved to promote the formation of the nanocomposite with desired electrical and mechanical properties have not yet been thoroughly examined. The principal objectives of the present study are two-fold: (1) to optimize a micellar aqueous system using dodecylbenzenesulfonic acid (DBSA) as dopant and surfactant to polymerize aniline in the presence of CNCs, and (2) to investigate the mechanistic role of CNCs and DBSA in the alignment of aniline monomer through electrostatic and hydrophobic interactions. NMR analysis was carried out to understand this mechanism. The structural, morphological, thermal and electrical properties of the nanostructured CNCs–PANI–DBSA composite were investigated to establish the structure–property interrelations in these flexible, sustainable, organic semiconductors. Lastly, we discuss, on the basis of this understanding, a mechanistic model for interactions within CNCs–DBSA–PANI nanocomposite films.

Experimental section

Materials

Aniline, ammonium persulfate (APS), 4-dodecylbenzenesulfonic acid (DBSA) and hydrochloric acid were purchased from Sigma-Aldrich, and used without further purification. Aqueous suspension of CNCs was prepared by controlled sulfuric acid hydrolysis of commercial bleached kraft pulp. Hydrolysis was carried out with 64% (w/w) sulfuric acid at 45 °C for 25 min with constant stirring (8.75 mL of a sulphuric acid solution per g pulp). Immediately following the acid hydrolysis, the suspension was diluted 10-fold with cold deionized (DI) water to quench the reaction and the mixture was allowed to settle overnight. The clear top layer was decanted and the remaining white cloudy layer was centrifuged at 4000 rpm for 10 min. The resultant precipitate was rinsed 2 times and dialyzed against DI until constant pH. After dialysis, the suspension was dispersed by subjecting it to ultrasound treatment using a Fisher Sonic Dismembrator for 30 min at 60% output. After sonication, the suspension was filtered with 42 Whatman filter paper and the resultant aqueous suspension was approximately 5 wt%.

CNCs–aniline–DBSA emulsion preparation (step 1)

In a 500 mL Erlenmeyer flask, 40 mL of a protonated form (H-form) CNCs suspension (4.7 wt%) was mixed with 110 mL of DI water and sonicated for 10 min at 60% maximum power. Then, 0.92 mL of aniline was added to the aqueous CNCs suspension under constant stirring followed by the addition of 3.08 mL of 4-dodecylbenzenesulfonic acid (DBSA) while mixing vigorously using an overhead stirrer (model IKA RW 20) at room temperature for approximately 1 h (refer to Table 1 for all emulsion ratios used in this work).

Table 1 Emulsion polymerization conditions for preparation of PANI–DBSA and CNCs–PANI–DBSA nanocomposites. Molar ratio APS : aniline = 1 (0.05 M) was used

Sample code		Mass ratio (CNCs : aniline)	Molar ratio (DBSA : aniline)	DBSA concentration (M)
CNCs–PANI–DBSA	A	2	0.25	0.0125
	B	2	0.5	0.025
	C	2	1	0.05
	D	2	1.5	0.075
	E	1	2	0.1
	F	2	2	0.1
	G	5	2	0.1
PANI–DBSA		0	1	0.05

---

Flexible CNCs–PANI–DBSA nanocomposite film preparation (step 2)

Ammonium persulfate (APS) (2.28 g) dissolved in 50 mL of deionized water was added to the CNCs–aniline–DBSA emulsion prepared in step 1. The APS-to-aniline molar ratio of 1 was used; their concentration in the solution was 0.05 M. The mass ratio of CNCs-to-aniline was 2 and the molar ratio of DBSA-to-aniline covered the range 0.25 to 2. In order to assess the effect of CNCs concentration, experiments were performed at a constant molar ratio of DBSA-to-aniline equal to 2 but using a mass ratio of CNCs-to-aniline of: 1, 2 and 5 (see Table 1). The polymerization was performed under vigorous stirring using an overhead stirrer (model IKA RW 20) at room temperature for 24 h, and a dark green-coloured CNCs–PANI–DBSA suspension was obtained. The suspension was centrifuged in order to remove by-products and the remaining thick dark green suspension was dialyzed against DI water until pH of the dialysis water stabilized at 6. The suspension was then mixed for 15 min and sonicated at 60% of maximum power for 10 minutes to produce a homogeneous and well dispersed suspension of CNCs–PANI–DBSA. A desired amount of the suspension was doped with HCl (1 N) for 24 h, then air dried at ambient conditions.

Preparation of PANI–DBSA (control sample)

Polyaniline was prepared in the same conditions following step 1 and step 2 procedures described above but in the absence of CNCs. The molar ratio DBSA : aniline was set to 1. At the end of the reaction, the suspension was dialyzed against DI water until pH of dialysis water stabilized at 6, to obtain a dark green powder. The powder was then pressed using a Carver press (model 4122) at 5 tonnes for 1 minute at room temperature to make pellets of polyaniline.

Characterization

For ¹H NMR study, 0.67 M solution of aniline in D₂O was prepared. To dissolve aniline in D₂O, the pH was adjusted to ∼4.7–4.9 using 1 M HCl. DBSA stock solution (∼0.2 M) was prepared by dissolving 130 mg of DBSA in 2 mL of D₂O. Another CNCs stock suspension was also prepared by dispersing 250 mg of freeze dried sodium form CNCs in 5 mL of D₂O. From these three stock solutions, small amounts were withdrawn so as to have 0.6 mL total volume in each NMR tube. Incremental amounts of CNCs stock suspension were added to aniline or aniline–DBSA complex to have mass ratio CNCs to aniline of 1, 2 and 5. The molar ratio of DBSA to aniline was fixed at 2 (0.1 M). Another NMR tube of aniline–DBSA complex solution without CNCs was prepared for comparison.

UV-Vis spectra of CNCs–PANI–DBSA aqueous suspensions were performed in the range between 260 and 1000 nm using a Cary 5000 spectrometer.

Apparent particle size was determined by photon correlation spectroscopy (Zetasizer 3000, Malvern Instruments, UK) which uses dynamic laser light scattering. Samples of CNCs–aniline–DBSA complex emulsions and CNCs–PANI–DBSA suspensions were prepared in 1 mM NaCl solution. The final CNCs concentration was adjusted to 0.05% (w/w). All runs were at least duplicated, with the reported values being the mean particle diameter for each material.

Electrical conductivity measurements were carried out using the four-probe method²⁴ directly on the surface of air-dried CNCs–PANI nanocomposite films, while for polyaniline alone, the powder was compressed to 20 mm diameter and ∼1 mm thickness pellets using a Carver press (model 4122).

Thermogravimetric analyses (TGA) were carried out using TA Instruments Q50 Thermogravimetric Analyzer. The heating rate was set at 20 °C min⁻¹ over a temperature range of 30–600 °C. Measurements were carried out in nitrogen atmosphere, with a flow rate of 20 cm³ min⁻¹.

Fourier transform infra-red (FTIR) spectra were made on air-dried CNCs–PANI nanocomposite films and on PANI–DBSA powder using ATR sampling technique with a Smart Orbit Nicolet 6700 FTIR spectrometer (diamond 30 000–200 cm⁻¹) in the range 500–4000 cm⁻¹. Each FTIR spectrum was obtained using 4 cm⁻¹ scanning resolution and 64 scans, and the spectra were processed using OMNIC (version 7.3 Thermo Electron Corporation) software.

Tensile testing was carried out on CNCs–PANI–DBSA nanocomposite films cut into rectangular strips of approximately 6 mm × 25 mm with a known thickness using a Deben micro-tensile testing rig equipped with a 200 N load cell. All tensile tests were carried out at ambient laboratory conditions (∼23 °C). The gauge length was 10 mm and special PTFE-coated grips with textured surface were used to clamp the strip's extremities.

The morphological properties were observed using an FEI Quanta 400F field-emission scanning electron microscope, operated in high vacuum mode at 5 kV accelerating voltage. The air dried CNCs–PANI–DBSA nanocomposite films were mounted on aluminium stubs using adhesive carbon tabs, and stubs were sputter-coated with AuPd.

Results and discussion

Interaction mechanisms in CNCs–aniline–DBSA emulsions

Effect of DBSA in CNCs–aniline–DBSA emulsions. In addition to its role as surfactant, DBSA was used as dopant because it provides the low-pH environment necessary for the chain growth of conducting polymers. Fig. 1 (inset) shows that increasing DBSA concentration in the medium at constant mass ratio (CNCs : aniline = 2) leads to different aggregation behaviour of surfactant molecules in the CNCs–aniline–DBSA complex. At molar ratios (DBSA : aniline) lower than 2, we observed milky colloidal suspensions because of the formation of an insoluble anilinium–DBSA complex. Whereas at DBSA : aniline = 2 molar ratio, the complex formed a clear, stable and less viscous suspension. Fig. 1 shows that at molar ratio DBSA : aniline less than 2, the particles are in the micron range. However, the size significantly drops to reach 71 nm when the molar ratio is increased (DBSA : aniline = 2). On the other hand, pH of the CNCs–aniline–DBSA emulsion decreases from 6 to 3 with increasing molar ratio of DBSA : aniline from 0.25 to 1, then stabilizes at 2 when the DBSA : aniline molar ratio reaches 1.5 and beyond.

Fig. 1 Apparent particle size (solid circles) and pH (open circles) of CNCs–aniline–DBSA emulsions before polymerization at different molar ratios of DBSA : aniline and at constant mass ratio, CNCs : aniline = 2.

It is well known that surfactants aggregate spontaneously in aqueous media to form micelles when the concentration is over their critical micellar concentrations (CMC).²⁵ Indeed, few or no micelles will be formed in the solution when surfactant concentration is lower than its CMC. Moreover, the size and structure of the micelles are greatly dependent on the concentration of the surfactant and also influenced by the solutes solubilized in the micelle solution.²⁵ For instance at a DBSA : aniline molar ratio less than 2, DBSA concentration is most likely below its CMC leading to no or few micelles formation. In addition, the local concentration of aniline will be higher at this concentration causing an increase in the pH of the medium which is likely too high for protonation and solubilisation of aniline. This phenomenon will lead to coalescence of monomer droplets which explains the higher particle size observed. At a DBSA : aniline = 2 molar ratio however, more micelles are formed with smaller size. Upon reaching the critical micellar concentration, the hydrophobic micellar core formed by the long hydrophobic hydrocarbon side chain and the hydrophilic interface formed by the sulfonate ester groups of DBSA are rearranged in a way where the hydrophobic tails are oriented towards the interior of the micelles leaving the hydrophilic sulfonate groups in contact with the aqueous medium containing the hydrophilic CNCs particles. Since CNCs bear a significant number of anionic sulfonate groups on their surfaces, these anionic moieties will result in an electrostatic repulsion between CNCs and DBSA micelles leading to micelles stabilization in the emulsion. This micellar rearrangement could induce the aniline monomer to be organized in a fashion where its benzene ring and amine moieties interact with both DBSA micelles and CNCs particles.

Role of CNCs, aniline and DBSA in the emulsion. To investigate the mechanistic roles of CNCs, aniline and DBSA micelles involved in the CNCs–aniline–DBSA emulsion, their interactions were examined using ¹H NMR spectroscopy of aniline. Fig. 2a and b show, respectively, the ¹H NMR spectra of CNCs–aniline and CNCs–aniline–DBSA complexes at DBSA concentration above CMC (DBSA : aniline = 2) but using different mass ratios of CNCs : aniline. These figures indicate the effect of CNCs and DBSA micelles on aniline protons.

Fig. 2 ¹H NMR spectra of (a) CNCs–aniline and (b) CNCs–aniline–DBSA emulsions with different molar ratios of CNCs : aniline at constant molar ratio (DBSA : aniline = 2) compared to pure aniline in D₂O. (*) peaks are assigned to DBSA protons.

The ¹H NMR spectrum of pure aniline in D₂O shows three peaks assigned as: two triplets for the meta (m) and para (p) protons at 7.26 and 7.07 ppm, respectively, and one doublet at 7.01 ppm for the ortho (o) proton. Addition of CNCs to aniline causes the resonance peaks of aniline protons in ortho, meta and para positions of the aniline benzene ring to shift downfield suggesting that the interaction between CNCs and aniline causes de-shielding of these protons. The change of chemical shift (Δδ) indicated for each peak in Fig. 2a shows that the downfield shift of the meta proton of aniline benzene ring is slightly higher than that observed for the ortho and para protons. Knowing that CNCs are considered as a polyanion bearing sulfate ester groups on their surfaces, and that aniline could be protonated in the presence of high concentrations of this strong acid polyelectrolyte, the positively-charged aniline acts as a counterion that complexes with the anionic CNCs.

Therefore, electrostatic or acid–base interactions may contribute to the observed downfield chemical shift of aniline resonance peaks. The line broadening of aniline protons slightly increase with increasing CNCs concentration, as well. This behaviour was already observed with poly(acrylic acid) (PAA),²⁶ where the aniline proton peak broadening was attributed to the behaviour of aniline being part of the polyelectrolyte, PAA, which results in shortening the relaxation time and leads to the broadening of the resonance peaks.

On the other hand, the addition of DBSA above the CMC with a molar ratio DBSA : aniline of 2 to aniline in the absence of CNCs causes the opposite effect (Fig. 2b). Indeed, the resonance peaks of aniline benzene ring protons shift upfield with merging of the para and ortho protons peaks. This indicates that DBSA causes shielding of the aniline aromatic protons in the micelles formed by DBSA. Similarly, the line widths of the aniline protons peaks also increase at this high concentration of DBSA owing to the inhomogeneous nature of the medium caused by the presence of micelles.²⁷ Upon increasing CNCs concentration from mass ratio CNCs : aniline = 1 to CNCs : aniline = 5, a slight gradual downfield shift of the resonance peaks was observed leading to a merging of the meta proton peak with the ortho and para peaks. The shielding of the aromatic protons of aniline in the presence of DBSA above the CMC is believed to be due to an increase in the hydrophobic micellar environment as proposed by Liu et al.²⁷

On the basis of these ¹H NMR results, and acknowledging that the solubilization of organic molecules in micelles is a dynamic process involving hydrophobic and electrostatic interactions,²⁶ we can explain the upfield shift of aniline protons in the presence of DBSA micelles by the fact that the aromatic ring of aniline is most likely exposed to the hydrophobic hydrocarbon core of the micelle. Whereas the downfield shift of aniline protons caused by the presence of CNCs suggest that the benzene ring of the aniline molecule is believed to be intercalated between the benzenesulfonate head groups of DBSA micelles while the quaternary amine groups are oriented towards the sulfonate esters of the surrounding CNCs suspension to electrostatically complex with them as illustrated in Fig. 3. This alignment of aniline suggests that CNCs are acting as a template for the DBSA micelles, which, in turn, can be regarded as a suitable template for the aniline monomer prior to polymerization. In fact, previous studies have clearly demonstrated using a similar surfactant, dodecylbenzenesulfonic acid sodium salt (SDBS), that the micelles may serve as nanoreactors or templates for aniline monomer prior to enzymatic synthesis of conducting linear polyaniline.^28,29

Fig. 3 Schematic representation of the mechanism of interactions between CNCs, aniline and DBSA in CNCs–aniline–DBSA emulsion.

Semiconducting nanocomposite films via emulsion polymerization

Particle size characterization of CNCs–PANI–DBSA suspensions. When polymerization proceeds by addition of ammonium persulfate, anilinium cations are polymerized within the DBSA micelle and cellulose nanocrystals resulting in CNCs–PANI–DBSA nanocomposite suspensions that differ greatly in terms of stability. More stable suspensions were obtained at higher DBSA concentration. Fig. 4 shows the relationship between particle size of the synthesized CNCs–PANI–DBSA nanocomposite suspensions, the molar ratio of DBSA : aniline, and the mass ratio of CNCs : aniline. When the DBSA concentration increases in the reaction medium, the particle size of the resulting CNCs–PANI–DBSA nanocomposite suspension decreases 7 fold while the particle size hardly changes with increasing CNCs concentration at a constant molar ratio (DBSA : aniline = 2). This confirms the necessity of the presence of micelles for the formation of a structurally ordered complex between CNCs, aniline monomer and DBSA micelles to produce a stable nanocomposite suspension. CNCs and DBSA have a synergistic effect on the alignment of aniline monomer during the complexation step before initiation of polymerization, which dictates the final properties of the CNCs–PANI–DBSA nanocomposite. A schematic representation of the proposed emulsion polymerization mechanism to obtain CNCs–PANI–DBSA nanocomposites is shown in Fig. 5.

Fig. 4 Apparent particle size (PCS) of CNCs–PANI–DBSA nanocomposite suspensions as function of molar ratio, DBSA : aniline, and mass ratio, CNCs : aniline.

Fig. 5 Reaction mechanism (a) and proposed schematic representation (b) of the emulsion polymerization process for CNCs–PANI–DBSA nanocomposites.

FTIR of CNCs–PANI–DBSA nanocomposite films. Fig. 6 shows that emulsion polymerization yields CNCs–PANI–DBSA nanocomposites that share common peaks with CNCs and pristine PANI–DBSA. All spectra exhibit the characteristic broad band for the O–H group of CNCs appearing at around 3278 cm⁻¹ and the characteristic absorptions of polyaniline around 1553, 1483, 1290 and 800 cm⁻¹. The peaks around 1483 cm⁻¹ and 1553 cm⁻¹ result from C=C stretching vibrations of N–benzenoid–N (B) and N=quinoid=N (Q) moieties in the PANI chains. The peaks at 1290 and 800 cm⁻¹ correspond to the C–N stretching and to the out-of-bending vibration of the C–H bond of p-di-substituted benzene ring respectively. We can also observe apparition of two other peaks at around 2920 and 1026 cm⁻¹, especially at molar ratios, DBSA : aniline above 1. These bands are attributed to the alkyl chain C–H vibration band of DBSA molecule and to the asymmetric stretching of SO₃⁻, respectively.²⁸ The bands are much more attenuated at low DBSA concentrations especially for low DBSA : aniline molar ratio (sample A, Table 1), which probably suggests that less conducting polyaniline is formed at this concentration.

Fig. 6 FTIR spectra of CNCs, PANI–DBSA and CNCs–PANI–DBSA nanocomposite films at different molar ratios of DBSA : aniline = 0.25 (sample A), 0.5 (sample B), 1 (sample C), 1.5 (sample D), and 2 (sample F).

Conductivity of CNCs–PANI–DBSA nanocomposite films. The emulsion polymerization process results in conductive CNCs–PANI–DBSA films. Generally, the electrical conductivity measured at ambient conditions was in the same order of magnitude as the control sample (PANI–DBSA) except sample A which was prepared at the lowest DBSA concentration (Table 2). A trend of initial increase followed by a slow decrease in electrical conductivity was observed when increasing molar ratio DBSA : aniline from 0.25 to 2 at a constant mass ratio (CNCs : aniline = 2). The low conductivity observed at the lowest DBSA concentration may be due to the fact that pH of the medium is not low enough and micelles are not formed at this DBSA concentration to solubilize the monomer and to generate the emeraldine salt as the only conducting form of polyaniline. Acidic conditions (pH ≤ 3) are usually required to assist the solubilisation of the aniline in water and to avoid excessive formation of undesired branched products.²⁹ However, conductivity of the nanocomposite starts to decrease at molar ratio DBSA : aniline > 1. The pH at DBSA : aniline > 1 concentrations is less than 3 which is perfect for the formation of the conducting emeraldine salt polymer. Moreover, a steep decrease in conductivity of the nanocomposite was observed when increasing CNCs content at a constant molar ratio, DBSA : aniline = 2. This behaviour indicates that there might be different parameters affecting the polymer structure and the conductivity of the nanocomposite and will be discussed in more detail in the following sections.

Table 2 Emulsion polymerization conditions for preparation of PANI–DBSA and CNCs–PANI–DBSA nanocomposites. Molar ratio APS : aniline = 1 (0.05 M) was used

Sample code		Mass ratio (CNCs : aniline)	Molar ratio (DBSA : aniline)	Conductivity^a (S cm^−1)
a Electrical conductivity was measured for samples doped with HCl (1 N) for 24 hours and dried at RT.
CNCs–PANI–DBSA	A	2	0.25	1.36 × 10^−3 ± 1.47 × 10^−4
	B	2	0.5	1.70 × 10^−1 ± 3.51 × 10^−2
	C	2	1	5.29 × 10^−1 ± 3.10 × 10^−2
	D	2	1.5	1.81 × 10^−1 ± 7.02 × 10^−3
	E	1	2	1.79 × 10^−1 ± 9.53 × 10^−3
	F	2	2	1.45 × 10^−1 ± 3.13 × 10^−3
	G	5	2	6.94 × 10^−2 ± 5.56 × 10^−3
PANI–DBSA		0	1	9.26 × 10^−1 ± 2.20 × 10^−1

---

Thermal properties of CNCs–PANI–DBSA nanocomposite films. TGA and Derivative TGA (DTGA) response curves of a typical air-dried CNCs–PANI–DBSA nanocomposite film are shown in Fig. 7. Initial mass loss at temperatures between room temperature and approximately 125 °C can be attributed to bound water and volatiles, whereas protonated-form CNCs undergo two-stage degradation at 160 °C and 340 °C (see ESI, Fig. S1†). Primary degradation of H-CNCs dominates the temperature range 120–240 °C, with secondary degradation of H-CNCs continues beyond 240 °C. DBSA–PANI degradation, however, begins in the same range, 120–240 °C, but the loss of excess DBSA and the oxidation of the polyaniline structure become prominent at 290 °C.³⁰ Subsequently, the degradation of bound DBSA and the decomposition of the polyaniline backbone itself dominate the region starting at 420 °C.³⁰ It is worthwhile noting that the onset temperature corresponding to CNCs degradation increased with increasing DBSA concentration (Fig. S1†), and similarly, the onset degradation temperatures corresponding to DBSA and polyaniline degradation also increased (see ESI, Table S1†) compared to the original PANI–DBSA indicating the possible occurrence of strong interactions between CNCs and polyaniline. It is also possible that the difference in molecular weights of synthesized polyaniline (at different preparation conditions) contribute to the differences in thermal properties. Alas, a clear demarcation between the two possible causes responsible for thermal variations is difficult to conclusively discern.

Fig. 7 TGA and DTGA curves of air-dried CNCs–PANI–DBSA nanocomposite films. Mass loss at 125 °C can primarily be attributed to bound water and volatiles, whereas the region 120–240 °C is predominated by mass loss attributed to H-form CNCs. The decomposition of polyaniline and DBSA dominates the regions starting at 290 °C and 420 °C. The integrals of areas 290 °C and 420 °C (peaks B and C) were measured for each sample to calculate the derivative weight values of PANI–DBSA.

The interdependence between CNCs concentration and conductivity of the nanocomposite films is explored in Fig. 8. The derivative weight is calculated from the integrals at 290 °C and 420 °C (see ESI, Fig. S2 and S3†), which are most likely dominated by the degradation of DBSA and polyaniline.§ The calculated derivative weight values were then plotted with the corresponding conductivities for each sample as functions of DBSA : aniline molar ratios and CNCs : aniline mass ratios (Fig. 8a and b). These results illustrate the close relationship between conductivity and the amount of polyaniline present in the nanocomposite. Above molar ratio DBSA : aniline = 1, the conductivity and the amount of polyaniline start to decrease (Fig. 8a). Similarly, increasing CNCs concentration in the system while keeping DBSA concentration constant (above CMC at a molar ratio DBSA : aniline = 2) leads to decreased conductivity and polyaniline content (Fig. 8b). In fact, it was observed after polymerization and centrifugation at high DBSA concentration, a homogeneous green supernatant on top of the formed nanocomposite precipitate. This observation reveals that at high DBSA concentrations—where well organized DBSA micelles are formed—a water soluble polyaniline stabilized by DBSA molecules may be present after polymerization. This can subsequently be discarded by centrifugation and washing, leading to a decrease in the polyaniline amount adsorbed onto CNCs. This behaviour can be explained as follows:

Fig. 8 Derivative weight and electrical conductivity of CNCs–PANI–DBSA nanocomposite films as function of: (a) molar ratio DBSA : aniline and (b) mass ratio CNCs : aniline.

(i) As stated in the previous section, there is a synergistic interaction mechanism between DBSA, aniline and CNCs in the emulsion before initiation of polymerization. Moreover, previous work reported that the polyaniline formed with the micellar system using SDBS is water soluble in its doped state.^27,29 Thus, the presence of excessive amount of DBSA in the emulsion will probably hinder the interaction between aniline and CNCs by solubilizing a significant amount of aniline monomers in the micelles formed by DBSA molecules leading to the formation of excess free water soluble polyaniline strongly interacted with DBSA, which is discarded during the washing step.

(ii) Another reason could be that excessive amount of DBSA and/or increasing CNCs concentration will lead to excess protonation of the amine groups with DBSA in polyaniline and/or hinder this protonation process by CNCs between polyaniline and DBSA. These effects could result in the retardation of charge transfer along the polymer and thereby a decrease in conductivity. This behaviour was also observed for DBSA-doped polyaniline nanoparticles prepared by reverse micelle polymerization at different DBSA concentrations,³⁰ and for other conducting polymeric systems such as polypyrrole,³¹ where the authors reported that further insertion of dopant molecules within PPy chains created a steric barrier for the charge carrier movement between PPy particles.

UV-Vis and elemental analysis of CNCs–PANI–DBSA nanocomposites. To gain further evidence of the relation between conductivity and doping level, UV-visible analysis was carried out for CNCs–PANI–DBSA dispersions in water at different DBSA concentrations. The UV-Vis spectra (Fig. 9) show the presence of three bands in the regions 350–370 nm, 780–820 nm, and a shoulder at around 420 nm. The presence of the two characteristic bands at 420 nm and 800 nm indicates that the polymer is in its emeraldine salt form.³² The band at ∼350 nm arises from π–π* (band gap) transition corresponding to benzenoid rings (B) while the two other bands at ∼420 nm and ∼800 nm may be assigned as polaron/bipolaron band transition.^3,33 The band at ∼800 nm is attributed to quinoid rings (Q).³ At low molar ratio of DBSA : aniline of 0.25, the band at ∼350 nm indicative of the extent of conjugation is not observed (Fig. 9). However, this band starts to prominently appear when DBSA content is increased. The ratio of absorbance (Q/B) of the low energy band (quinoid) to the high energy band (benzenoid) increases from 1.05 to 1.44 with increasing DBSA content indicating a higher concentration of polaronic species generated from higher doping, leading to better protonation level of polyaniline backbone. Indeed, the percentages of nitrogen (N) and sulfur (S) measured by elemental analysis were used to confirm this result (Table 3). S/N ratios of CNCs–PANI–DBSA nanocomposite samples increase with increasing the molar ratio DBSA : aniline, demonstrating that a strong complexation of DBSA with the polymer backbone lead to a high doping level of polyaniline.

Fig. 9 UV-Vis spectra of CNCs–PANI–DBSA nanocomposite films at different molar ratios (DBSA : aniline) compared to PANI–DBSA. The quinoid (Q) to benzenoid (B) ratios are indicated for each spectra.

Table 3 Elemental composition (wt%) and S/N ratio calculated for CNCs–PANI–DBSA nanocomposite films

Molar ratio	C (%)	H (%)	N (%)	S (%)	S/N
0.25	52.94	6.47	3.48	3.05	0.88
0.5	54.72	7.60	3.25	3.32	1.02
1	50.17	7.03	3.61	4.33	1.20
2	50.19	6.13	2.27	3.49	1.54

---

On the other hand, the polaron band at ∼800 nm undergoes a hypsochromic shift from 844 nm to 790 nm by increasing the DBSA concentration in the system. This hypsochromic shift is most likely due to the steric repulsion between the bulky DBSA molecules bound to polyaniline. The steric repulsions are believed to increase the torsion angle between adjacent rings on substitution, leading to increasing the average band gap in the ensemble of conjugated polymeric system.³⁴ The increase in the band gap consequently leads to a decrease in the conductivity of the polymer which could also explain the decrease in electrical conductivity for CNCs–PANI–DBSA nanocomposites at high DBSA concentration.

Mechanical properties of CNCs–PANI–DBSA nanocomposite films. Doped CNCs–PANI–DBSA nanocomposite films are mechanically distinct from the PANI–DBSA polymer. The former are flexible smooth films after doping with HCl and drying at room temperature, while PANI–DBSA is obtained as a powder or film with limited structural integrity. The tensile behaviour of CNCs–PANI–DBSA nanocomposite films as a function of increasing DBSA or CNCs concentrations is shown in Fig. 10a and b. The mechanical properties of CNCs–PANI–DBSA nanocomposite films changed significantly with increasing DBSA content. At molar ratio of DBSA : aniline = 0.5, the nanocomposite shows a maximum stress and strain values of 22 ± 2.16 MPa and 0.89 ± 0.20%, respectively, while at higher molar ratio DBSA : aniline = 2, the tensile strength and strain decrease to 8 MPa and 0.56%, respectively. However, increasing CNCs in the system at a constant DBSA concentration leads to enhancement of the nanocomposite strength properties. For instance, the tensile strength and strain reached 16 ± 2.89 MPa and 0.62 ± 0.10% at mass ratio of CNCs : aniline = 5, while at lower mass ratio of CNCs : aniline = 1, the strain stayed practically unchanged and the tensile strength decreased to 6 MPa. The decrease in tensile strength for CNCs–PANI–DBSA nanocomposites at high DBSA concentration could be a result of the steric effect of DBSA's long alkyl chains limiting CNCs from connecting with each other and leading to a weakened system. However, when CNCs concentration is increased, CNCs will act as an efficient scaffold for polyaniline owing to the intermolecular and hydrogen bonds existing between CNCs nanoparticles. A similar effect was observed with nanocellulose–polypyrrole (NC/PPy) composites,²³ where the post-synthetic inclusion of bare nanocellulose (NC) fibre to the composite enhanced the mechanical strength of NC–PPY which is due to H-bonding between adjacent pristine NC fibres.

Fig. 10 Stress vs. strain curves of CNCs–PANI–DBSA nanocomposites at (a) constant mass ratio (CNCs : aniline = 2) but different molar ratios of DBSA : aniline, and at (b) constant molar ratio (DBSA : aniline = 2) but different mass ratios of CNCs : aniline.

These are significantly stronger and more flexible films than those obtained with, for instance, graphene/polyaniline composite paper or graphene paper, where it has been reported that the tensile strengths were 12.6 and 8.8 MPa, and maximum strains, 0.11 and 0.08%, respectively.³⁵ It is worthwhile noting that researchers have recently attempted to develop conductive fabrics by in situ polymerization of woven cotton fabrics into a solution of pyrrole and cellulose nanocrystals,³⁶ or produce conductive nanopaper comprising multi-walled carbon nanotubes (MWCNTs) and cellulose nanocrystals via vacuum filtration using hydrophobic polyvinyl difluoride membrane filters.³⁷ In the former, the mechanical properties essentially relied on the nature of the woven fabric, resulting in flexible conductive fabrics (stretch ∼ 18%).³⁶ Whereas in the case of the hybrid nanopaper, the mechanical performance depended on the concentrations of MWCNTs and CNCs, and typically resulted in less extensible systems (stretch ∼ 0.02).³⁷

Morphological properties of CNCs–PANI–DBSA nanocomposite films. Fig. 11a and b depict SEM images (at low and high magnifications) of air-dried CNCs–PANI–DBSA nanocomposite film prepared at mass ratio of CNCs : aniline = 5 and molar ratio of DBSA : aniline = 2 before doping with HCl. The SEM images clearly indicate that at sufficiently high CNCs concentrations, we can still preserve the chiral nematic structure typical of self-assembled CNCs in the nanocomposite. However, doping with HCl disrupts the chiral structure of CNCs (Fig. 11c and d). This demonstrates that the present emulsion method does retain the structural organization of CNCs particles, as well as improves the mechanical properties of the nanocomposite films. If a different dopant to HCl is used, chiral nematic organization can be preserved in the semi-conductive films after doping at the appropriate ionic strength.

Fig. 11 SEM images of air-dried CNCs–PANI–DBSA nanocomposite film (sample G, Table 1) before doping (a and b) and after doping (c and d) with HCl.

Conclusions

Aqueous emulsion polymerization – a scalable and essentially green process – was successfully used to prepare flexible and semi-conducting CNCs–PANI–DBSA nanocomposite films, where the CNCs' original chiral, nematic organization is preserved. Our work has shown that the mechanical and conductive properties can be tailored to suit the desired end-use application. This development provides promising sustainable organic materials for use in electronic and opto-electronic applications. Our in-depth investigation helps provide strong insights into the structure–property interrelationships of the sustainable, organic semi-conducting films.

CNCs, which can be regarded as strong acid polyelectrolytes, and the micelles formed by DBSA have both shown to act as effective templates for aniline before initiation of the polymerization process. A stable suspension of CNCs–PANI–DBSA nanocomposite can be obtained depending on the micellar concentration of the surfactant used, and by changing CNCs and DBSA concentrations we could fine-tune the mechanical and electrical properties of the final nanocomposite films. Further doping with HCl has shown to alter the structural organization of CNCs in the nanocomposite film due to high ionic strength. However, enhancement in the mechanical properties attributed to CNCs as well as the electrical conductivity can be maintained. It is possible, of course, to use different dopants and produce flexible, organic semi-conducting films that retain chiral, nematic organization.

Acknowledgements

The Transformative Technology Program, Natural Resources Canada, and the NSERC-IPDF program are gratefully acknowledged for funding this work.

Notes and references

1. T. G. Skotheim, R. L. Elsenbaumer and J. R. Reynolds, Handbook of Conducting Polymers, Marcel Dekker, New York, 2nd edn, 1998, pp. 883–1007.
2. P. Chandrasekhar, Conducting Polymers, Fundamentals and Applications: A practical Approach, Kluer Academic, Boston, 1999, pp. 434–435.
3. Y. Liao, C. Zhang, Y. Zhang, V. Strong, J. Tang, X. G. Li, K. K-Zadeh, E. M. V. Hoek, K. L. Wang and R. B. Kaner, Nano Lett., 2011, 11, 954–959.
4. B. C. Ku, D. K. Kim, J. S. Lee, A. Blumstein, J. Kumar and L. A. Samuelson, Polym. Compos., 2009, 30, 1817–1824.
5. J. Lin, Q. Tang, J. Wu and H. Sun, Sci. Technol. Adv. Mater., 2008, 9(2), 025010.
6. B. Winkler, L. Dai and A. W. H. Mau, J. Mater. Sci. Lett., 1999, 18, 1539–1541.
7. C. Zhu, J. Zhai, D. Wen and S. Dong, J. Mater. Chem., 2012, 22, 6300–6306.
8. Y. Habibi, L. A. Lucia and O. J. Rojas, Chem. Rev., 2010, 110, 3479–3500.
9. W. Y. Hamad and T. Q. Hu, Can. J. Chem. Eng., 2010, 88, 392–402.
10. J.-F. Revol, H. Bradford, J. Giasson, R. H. Marchessault and D. G. Gray, Int. J. Biol. Macromol., 1992, 14, 170–172.
11. O. van der Berg, M. Schroeter, J. R. Capadona and C. Weder, J. Mater. Chem., 2007, 17, 2746–2753.
12. E. Tkalya, M. Ghislandi, W. Thielemans, P. van der Schoot, G. de With and C. Koning, ACS Macro Lett., 2013, 2, 157–163.
13. J. Fan, W. Shao, G. Xu, X. T. Cui and X. Luo, RSC Adv., 2014, 4, 24328–24333.
14. W. Hu, S. Chen, Z. Yang, L. Liu and H. Wang, J. Phys. Chem. B, 2011, 115, 8453–8457.
15. K. Sakakibara and T. Rosenau, Holzforschung, 2012, 66, 9–19.
16. H. Wang, L. Bian, P. Zhou, T. Jian and W. Tang, J. Mater. Chem. A, 2013, 1, 578–584.
17. L. Byoung-Ho, K. Hyun-Joong and Y. Han-Seung, Curr. Appl. Phys., 2012, 12, 75–80.
18. J. A. Marins, B. G. Soares, K. Dahmouche, S. J. L. Ribeiro, H. Barud and D. Bonemer, Cellulose, 2011, 18, 1285–1294.
19. D. Müller, J. S. Mandelli, J. A. Marins, B. G. Soares, L. M. Porto, C. R. Rambo and G. M. O. Barra, Cellulose, 2012, 19, 1645–1654.
20. G. Nyström, A. Mihranyan, A. Razaq, T. Lindström, L. Neyholm and M. Strømme, J. Phys. Chem. B, 2010, 114, 4178–4182.
21. L. H. C. Mattoso, E. S. Medeiros, D. A. Baker, J. Avloni, D. F. Wood and W. Orts, J. Nanosci. Nanotechnol., 2009, 9, 2917–2922.
22. S. Y. Liew, S. Shariki, A. Vuorema, D. A. Walsh, F. Marken and W. Thielmans, in Functional Materials from Renewable Sources, ed. F. Liebner and T. Rosenau, ACS Symposium Series, 2012, vol. 1107, ch. 5, pp. 75–106.
23. A. Mihranyan, M. Esmaeili, A. Razaq, D. Alexeichlik and T. Lindström, J. Mater. Sci., 2012, 47, 4463–4472.
24. F. M. Smits, Bell Syst. Tech. J., 1958, 37, 711–718.
25. J. H. Fendler, Membrane Mimetic Chemistry: Characterizations and Applications of Micelles, Microemulsions, Monolayers, Bilayers, Vesicles, Host-Guest Systems, and polyions, John Willey & Sons, New York, 1982.
26. W. Li, P. A. McCarthy, D. Liu, J. Huang, S. C. Yang and H. L. Wang, Macromolecules, 2002, 35, 9975–9982.
27. W. Liu, A. L. Cholli, R. Nagarajan, J. Kumar, S. Tripathy, F. F. Bruno and L. Samuelson, J. Am. Chem. Soc., 1999, 121, 11345–11355.
28. W. Liu, J. Kumar and S. Tripathy, Langmuir, 2002, 18, 9696–9704.
29. Y. Cao, P. Smith and A. J. Heeger, Synth. Met., 1992, 48, 91–97.
30. Y.-G. Han, T. Kusunose and T. Sekino, Synth. Met., 2009, 159, 123–131.
31. C. Sasso, N. Bruyant, D. Beneventi, J. Faure-Vincent, E. Zeno, M. Petit-Conil, D. Chaussy and M. N. Belgacem, Cellulose, 2011, 18, 1455–1467.
32. N. Kuramoto and A. Tomita, Synth. Met., 1997, 88, 147–151.
33. H. K. Lin and S. A. Chen, Macromolecules, 2000, 33, 8117–8118.
34. S. Sharma and D. Kumar, Indian J. Eng. Mater. Sci., 2010, 17, 231–237.
35. D. W. Wang, F. Li, J. Zhao, W. Ren, Z. G. Chen, J. Tan, Z. S. Wu, I. Gentle, G. Q. Lu and H. M. Cheng, ACS Nano, 2009, 3, 1745–1752.
36. A. Hebeish, S. Farag and T. I. Shaheen, Carbohydr. Polym., 2016, 151, 96–102.
37. Y. Ren, T. Wang, Z. Chen, J. Li, Q. Tian, H. Yang and Q. Xu, Appl. Surf. Sci., 2016, 385, 521–528.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra13610a

‡ This is the terminology adopted by the International Standardization Committee (ISO). Cellulose nanocrystals are also referred to, in the literature, as nanocrystalline cellulose (NCC) or cellulose nanowhiskers (CNW).

§ As alluded to above, secondary H-CNCs degradation continues beyond 240 °C, however, it may be reasonable to approximate that 290 °C and 420 °C are dominated by DBSA and polyaniline degradation.

---