Introduction

Cellulose fibers are one of the most abundant and widespread natural biopolymers on earth (Ramamoorthy et al. 2015). Due to their biodegradability, non-toxicity and sustainable character, cellulose fibers have been widely used in the paper industry (Wu and Farnood 2015), packaging industry (Khan et al. 2014), textile industry (Costa et al. 2013) and medical field (Tian et al. 2012). Compared to petroleum-based plastic products, one of the major drawbacks of cellulose fiber-based materials is their inherent highly hydrophilic nature, which restricts the practical applications of cellulose fibers where hydrophobicity is demanded (e.g., the packaging industry). With the increased awareness of environmental protection and decreasing petroleum resources, it is increasingly desirable to modify cellulose fibers to make them hydrophobic.

In the traditional papermaking industry, improvement of the hydrophobicity of cellulose fiber-based paper products is achieved via surface sizing or internal sizing (Hubbe 2014; Garcia-Ubasart et al. 2012). The hydrophobicity of cellulose fibers can be significantly improved (water contact angle >100º) by treating fibers with a sizing agent such as alkyl ketene dimer (AKD) or alkenyl succinic anhydrate (ASA) (Sun et al. 2014) or by enzymatic grafting ofa hydrophobic compound (Dong et al. 2015; Garcia-Ubasart et al. 2011). However, the obtained paper products still cannot meet the high water resistance requirements, for example, when paper products are used as food and drink packages. Therefore, the development of new methods to make hydrophilic cellulose fiber superhydrophobic (which is defined by a water contact angle >150º) is of utmost interest. Due to their relatively low cost and biodegradability, modified superhydrophobic cellulose fiber-based products would have numerous potential applications, such as beverage containers (Mourad et al. 2012), self-cleaning paper labels (Wang et al. 2015), bioactive paper (Sousa and Mano 2013), water-oil separation (Gao et al. 2015) and microfludic devices (Songok et al. 2014). Although various methods are used to prepare superhydrophobic material surfaces—for example, the sol-gel method (Pan et al. 2012), chemical vapor deposition (Crick et al. 2012), ink-jet printing (Zhang et al. 2015), etc., have been vigorously studied—research on superhydrophobic cellulose fiber-based products is limited because most methods for fabrication of a superhydrophobic surface cannot be easily applied to cellulose fibers because of the low efficiency. Recently, a number of approaches, including chemical grafting modification (Roy et al. 2009), spray coating (Ogihara et al. 2012), rapid expansion of supercritical CO2 (Werner et al. 2010) and plasma treatment (Balu et al. 2008), have been reported to prepare superhydrophobic cellulose fiber-based materials, but these techniques still involve several limitations, such as specialized and costly instrumentation, tedious fabrication procedures and low grafting efficiency (Huang et al. 2011).

The electrostatic layer-by-layer (LBL) self-assembly, introduced first by Decher and co-workers in the early 1990s (Decher 1997; Decher and Hong 1991), has gained considerable attentions in material surface modification because of its simplicity, low-cost and versatility. The LBL technique is based on the alternating deposition of oppositely charged polyelectrolytes on a variety of substrates and can be used to construct multilayers with tailored chemical composition and architecture on the micro- and nanometer size scales (Zhang et al. 2004). The construction process is easy, and the procedure can be adapted to almost any type of substrate surface as long as surface charges are present (Fu et al. 2005). So far, there are numerous reports regarding modifying the cellulose fiber surface using the LBL technique. Cellulose fiber-based products will obtain some new functional properties, such as flame resistance (Köklükaya et al. 2015), biological catalytic activity (Xing et al. 2007) and electroconductivity (Agarwal et al. 2009), and improved physical properties, such as high strength (Wu and Farnood 2014; Marais and Wågberg 2012), porosity (Lu et al. 2007) and stretch (Marais et al. 2014), by constructing LBL multilayers on the cellulose fiber surface. To the best of our knowledge, few attempts have been made to report the preparation of superhydrophobic paper products using the LBL technique: both Yang and Deng (2008) and Zhang et al. (2012) respectively constructed polydiallyldimethylammonium chloride/silica nanoparticle multilayers on cellulose fiber surfaces, followed by chemical modification with fluoroalkylsilane; the resulting fiber-based products exhibited water contact angles above 150º. Zhao et al. (2013) alternately deposited poly(acrylic acid) (PAA) and 2-hydroxy-4-methoxybenzophenone-5-sulfonic acid (HMBS)-intercalated layered double hydroxides (LDH) on cotton fiber surfaces followed by a fluoroalkylsilane treatment, and they rendered cotton fabric with dual functions of superhydorphobicity and UV-blocking. All the above studies involved fluoroalkylsilanes, which have extremely low surface free energy. However, such fluorochemicals are expensive and carry potential risks for human health and the environment (Li et al. 2008). Hence, the development of the non-fluorinated LBL building block is very important for the fabrication of environmentally friendly, superhydrophobic fiber-based products.

In our previous work (Li et al. 2012), we improved the hydrophobic character of cellulose fiber surfaces using LBL self-assembly of cationic polyacrylamide (CPAM) and lignosulfonates (LS). The initial water contact angle of cellulose fiber modified with a (CPAM/LS)5 multilayer reached 109.2º, but the long-term stability of hydrophobicity was poor, and a superhydrophobic fiber surface was not obtained. However, it is worth mentioning that most of the hydrophobic C–C groups in LS macromolecules were probably oriented to the multilayer-air interface of the system to minimize the surface energy of the cellulose fibers. A similar mechanism has been suggested by Feng et al. (2003) for nanofibers of polyvinyl alcohol (PVA), which are highly hydrophobic compared to a smooth film of the same material. This is because of enrichment of hydrophobic –CH2– at the solid-air interface. Moreover, Zhai et al. (2004) and Gustafsson et al. (2012) have concluded that heat treatment can significantly improve the surface hydrophobicity of micro-structured poly(allylamine hydrochloride) (PAH)/PAA multilayers through reorientation of hydrophobic groups in polymers. Inspired by these finding, our present work focuses on introducing typical hydrophobic long chains into LS macromolecules through the Mannich reaction. The modified LS (LS-amine, LSA) was alternately deposited with PAH on cellulose fiber surfaces. The formation of PAH/LSA multilayers on cellulose fibers was characterized, and the effects of PAH/LSA multilayers on the wetting and physical strength properties of cellulose fibers were also investigated.

Experimental

Materials

Fully bleached eucalyptus kraft pulp fibers were kindly provided by Yunnan Yunjing Forestry & Pulp Co., Ltd. (Yunnan, China). Before use, the fibers were washed, and the carboxyl groups of the fibers were converted to their sodium forms according to Marais et al. (2014). Poly(allylamine hydrochloride) with an average molecular weight of 17,500 was purchased from Sigma-Aldrich. Lignosulfonates (LSs) were kindly provided by Jiangmen Sugarcane Chemical Factory Co., Ltd. (Jiangmen, China). Before use, the LSs were purified as described in our previous work (Li et al. 2012). The dodecylamine (Aladdin Co., Ltd., Shanghai, China) was used as received. The Milli-Q water (Millipore, MA, USA) with a resistivity of 18 MΩ/cm was used in all the experiments. All other chemical reagents (formaldehyde, hydrochloric acid and sodium hydroxide) were of analytical grade and used without any further purification.

Methods

Synthesis of lignosulfonates-amine

Lignosulfonates-amine (LSA) was prepared by referring to the reported by Yue et al. (2011) with a few modifications. Briefly, 10.0 g of LS and 7.3 g of dodecylamine were added to 100 ml Milli-Q water maintained at 40 °C, and then the pH of the solution was adjusted to 10.5 using NaOH. The solution was added to a three-necked flask. Subsequently, 5.3 g of formaldehyde solution (37 wt%) was added by portions over a 30-min period with continuous stirring. The reaction was carried out at 90 °C for 6 h. Thereafter, the reaction solution was added to about three times its volume of ethanol to precipitate LSA. Finally, the LSA was recovered by filtration, washed in ethanol several times and then dried under vacuum for further use. The Mannich reaction general LS formula is shown in Fig. 1.

Fig. 1
figure 1

Mannich reaction general formula of lignosulfonates

Full size image

Preparation of superhydrophobic cellulose fiber-based paper

The process of preparing PAH/LSA multilayers on cellulose fiber surfaces is simply illustrated in Scheme 1. Step 1: the negatively charged cellulose fibers were first immersed in PAH solution (0.1 g/l, pH 7.5) for 15 min to render the cellulose fibers positively charged, followed by rinsing thoroughly three times with Mill-Q water to remove excess PAH. Step 2: the positively charged cellulose fibers were then immersed in LSA solution (0.2 g/l, pH 12) for 15 min, followed by rinsing thoroughly three times with Mill-Q water to remove the excess LSA. (PAH/LSA) n multilayers with a desirable bilayer number (n) can be prepared on cellulose fibers surfaces by simply repeating steps 1–2 for n times. The PAH/LSA multilayer-modified cellulose fibers were made into handsheets with an average grammage of 80 g/m2 using a semiautomatic sheet former equipped with a water circulation system. The handsheets were cut in half, and one half of each sheet was further heat-treated at 160 °C for 30 min. Here, we labeled non-extra-heat-treated handsheets as NHTs and heat-treated handsheet as HTs.

Scheme 1
scheme 1

Schematic diagram illustrating the preparation process of superhydrophobic cellulose fiber surfaces via LBL multilayers modification

Full size image

Characterization

X-ray photoelectron spectroscopy (XPS) was performed using an Axis Ultra DLD spectrometer (Kratos Analytical Ltd., UK) with monochromatic Al Kα radiation as the X-ray source. Fourier transform infrared spectroscopy (FTIR) was performed on a Bruker VERTEX-70 spectrometer. A zetasizer-nano ZS (Malvern Instruments, Malvern, UK) was used to measure the zeta potential of LSA as a function of pH. A Mütek SZP-10 zeta potential tester (BTG Group, Germany) based on the streaming potential method was used to measure the zeta potential of modified cellulose fibers. The surface morphologies of the modified cellulose fibers were characterized by tapping mode atomic force microscopy (AFM, Nanoscope IIIa Multimode, Veeco Co., USA). At least five different fibers from the same sample were scanned several times, and the representative images (1.5 × 1.5 μm) were chosen for presentation. Measurement and processing of the AFM images were performed with Version 5.12r3 software (Veeco Co., USA), and no image processing except flattening was done. Contact angles were measured using a SL200B optical contact angle meter (Kino USA Industry Co., Ltd., USA) at ambient temperature, and the values were averages from measurements made at at least five different positions for each sample. The tensile strength of the handsheet was tested using a Lorentzen and Wettre’s Tensile Tester (AB Lorentzen and Wettre, Sweden), and the values presented were the average from at least five measurements for each sample. The zero-span tensile strength was determined using a Pulmac zero-span tensile tester (Pulmac International Inc., Middlesex, CT, USA) according to the ISO 15361:2000 standard.

Results and discussion

Synthetic lignosulfonates-amine (LSA) characterization

Figure 2 shows the FIIR spectra of the original LS and LSA. The absorption peak intensities at 2930, 2850, 1380 and 1468 cm−1, attributed to the C–H stretching vibration and C–H bending vibration of the –CH3 and –CH2–, respectively, were significantly increased in the FTIR spectrum of LSA in comparison with those of LS. The absorption peak at 840 cm−1 is due to the stretching vibrations of C–H for C2, C5 and C6 in the aromatic ring. This peak intensity decreased in the FTIR spectrum of LSA in comparison with that of LS, indicating the –H of C5 in the aromatic ring was substituted after the reaction. Moreover, the obvious absorption peak at 721 cm−1, attributed to the –(CH2) n – (n ≥ 4) bending vibration, was present in the FTIR spectrum of LSA. The above result is probably due to the Mannich reaction introducing long-chain aliphatic amine to the LS macromolecules. However, it is worth noting that the absorption peaks of N–H and C–N at 3300 and 1040 cm−1, respectively, could not be clearly detected, which was attributed to the overlapping with the strong peak of –OH and C–O in the same frequency region. Thus, according to the FTIR results it can be preliminarily deduced that dodecylamine was introduced into the LS macromolecules.

Fig. 2
figure 2

FT-IR spectra of original lignosulfonates (a) and lignosulfonates-amine (b)

Full size image

Since the LSA could not be easily confirmed by FTIR analysis, the elemental compositions of LS and LSA were determined using a PE 2400II element analysis apparatus (PerkinElmer Co., USA) for further analysis, and the results are displayed in Table 1. The nitrogen content increased to 4.59 % in LSA from 1.27 % in LS. These data further indicated that the long-chain aliphatic amine was introduced into LS macromolecules through the Mannich reaction.

Table 1 Elemental composition of LS and LSA
Full size table

Figure 3 shows the zeta potential-pH curve of original LS and LSA. The original LS was negatively charged under all pH conditions, owing to the electrolysis of sulfonic, carboxyl and phenolic hydroxyl groups in LS macromolecules (Liu et al. 2009). After modification, the zwitterionic character of LSA was obtained, and the isoelectric point (IEP) of the LSA was about 3.8, indicating the colloidal properties of LS changed drastically after Mannich reaction modification. The observed changes of LSA could be caused by the protonation of secondary amine groups on the LSA macromolecules, which further verified that the dodecylamine has been grafted onto the LS macromolecules. Therefore, to keep LSA negatively charged for the electrostatic interaction with PAH, LSA solutions made performed at pH 12 in all the experiments.

Fig. 3
figure 3

Zeta potential of LS and LSA in aqueous solution (0.2 g/l) with different pHs

Full size image

Formation of PAH/LSA multilayers on cellulose fiber surfaces

The surface chemical compositions of original and modified cellulose fibers were investigated by XPS, and the survey scan XPS spectra of original and PAH/LSA multilayer-modified cellulose fibers are illustrated in Fig. 4. Distinctive peaks at 284.6 and 530.6 eV demonstrated the appearance of carbon and oxygen, respectively (Fig. 4a). When PAH/LSA multilayers were formed on cellulose fibers, distinctive peaks at 167.9, 197.8, 284.6, 401.6 and 530.6 eV indicated the presence of sulfur, chlorine, carbon, nitrogen and oxygen, respectively (Fig. 4b–d). Herein, the nitrogen element derived from both the amine group in the PHA macromolecule and the secondary amine group in the LSA macromolecule, whereas the sulfur element only originated from the sulfonic group in LSA, and the chlorine element only belonged to the PHA macromolecules. Therefore, the sulfur and chlorine elements could be considered as characteristic elements of PHA/LSA multilayers, and the growth of PAH/LSA multilayers on cellulose fibers can be represented by the increase in elemental contents of sulfur and chlorine. The insets are the high-resolution XPS spectra of the S 2p and Cl 2p regions for the modified cellulose fibers. With the LBL deposition, the peak intensities of S 2p and Cl 2p increased with increasing bilayer number, confirming the growth of PAH/LSA multilayers on cellulose fiber surfaces.

Fig. 4
figure 4

XPS spectra of original and PAH/LSA multilayer-modified cellulose fibers. a Original cellulose fiber; b–d (PAH/LSA)1, (PAH/LSA)3 and (PAH/LSA)5 multilayer-modified cellulose fibers, respectively. Insets show the high-resolution spectra of S 2p and Cl 2p

Full size image

Because the surface charge of cellulose fibers depends on the polyelectrolytes in the outermost layer, the zeta potentials of the cellulose fibers deposited with different layers were measured to demonstrate the formation of PAH/LSA multilayers. Figure 5 shows the zeta potential of modified fibers as a function of layer numbers. The original cellulose fibers had a negative potential of −63.1 mV. When PAH was deposited in the outermost layer, the zeta potential of the cellulose fiber surface reversed to a positive value of 34.2 mV. After that, the deposition of LSA reversed the surface potential to −27.6 mV. The alternating reversal trend in the zeta potential was observed along with the further increasing layer number, demonstrating the stepwise growth of PAH/LSA multilayers on the cellulose fiber surface and reproducibility of deposition process.

Fig. 5
figure 5

Zeta potential of the modified cellulose fiber with different layer numbers

Full size image

Furthermore, AFM was employed to reveal the surface morphologies of PAH/LSA multilayer-modified cellulose fibers, and the representative AFM phase images and 3D morphological images are presented in Fig. 6. For the original cellulose fiber surface, it was obvious that it completely consisted of regularly ordered microfibrils and exhibited a relatively flat topography with a root-mean-square (RMS) roughness of about 6.3 nm (Fig. 6a). The surface morphology of th PAH monolayer-modified cellulose fiber with an RMS roughness of about 6.3 nm was quite similar to that of the original cellulose fiber, and the PHA could not be distinguished from the image (Fig. 6b). When the PAH/LSA multilayers were deposited on the cellulose fiber surface, a great deal of granular substances gradually appeared (Fig. 6c–e), thus causing an increase in the surface roughness of the modified cellulose fibers. The surface RMS roughness of cellulose fiber surfaces modified with (PAH/LSA)1, (PAH/LSA)3 and (PAH/LSA)5 multilayers was measured as 28.5, 30.4 and 51.8 nm, respectively. These granular substances having a mean diameter of 90–136 nm were regarded as LSA granules. This can be explained in that the amount of granules increased along with the increasing bilayer number, which was consistent with the increased sulfur element content as described by previous XPS analysis.

Fig. 6
figure 6

AFM images of original and modified cellulose fiber surfaces: a original cellulose fiber; b cellulose fiber modified with a PAH monolayer; c–e cellulose fiber modified with a (PAH/LSA)1, (PAH/LSA)3 and (PAH/LSA)5 multilayer, respectively. The left pictures are phase images, and the right pictures are 3D images from height phase. The scanning scale is 1.5 × 1.5 μm

Full size image

Wetting properties of handsheets prepared from modified fibers

Figure 7 shows the water contact angles of handsheets prepared from PAH/LSA multilayer-modified cellulose fibers with and without extra heat treatment, respectively. The handsheet prepared from original cellulose fibers is highly hydrophilic because of the natural hydrophilic character of cellulose, resulting in a rapid absorption of water droplets into the handsheet. Zigzag growth of the water contact angles was observed for both non-heat-treated and heat-treated handsheets, depending on whether PAH or LSA was deposited in the outermost layer. The water contact angle of the handsheet with LSA in the outermost layer was higher than that of the handsheet with PAH in the outermost layer, indicating the LSA deposited in the outermost layer plays an important role in cellulose fiber's hydrophobic modification. This may be due to more hydrophobic groups of LSA than PAH being exposed to the outside of multilayers. Moreover, it was clearly shown that extra heat-treatment further increased the hydrophobicity of the same handsheets. When a 10-layer (5-bilayer) PAH/LSA multilayer was formed on cellulose fiber surfaces, the handsheet surface turned hydrophobic with a water contact angle of 120.3º. After heat treatment, the water contact angles reached up to 151.7º, indicating a superhydrophobic surface was obtained.

Fig. 7
figure 7

Water contact angles of handsheets at contact times of 60 s

Full size image

It is well known that there are two basic ways to improve the hydrophobicity of a surface: (1) by increasing the surface roughness; (2) by lowering the surface energy. On one hand, AFM results have shown that the PAH/LSA multilayers efficiently increased the surface RMS roughness of cellulose fibers by controlling the multilayer number. On the other hand, the surface energies of handsheets with and without extra heat treatment were tested respectively according to the method reported by our previous work (Li et al. 2012), and the results are presented in Tables 2 and 3, respectively. In this study, the water and diiodomethane were used as the test liquids, and the contact angles were recorded at contact times of 60 s. When PAH/LSA multilayers were constructed on cellulose fibers, the surface energy of the handsheet decreased with the increase in bilayer number. This can be explained in that parts of the hydrophobic long chain groups in LSA macromolecules were exposed at the outside of multilayers and parts of them toward the inside of multilayers in order to provide smaller steric hindrance, and the hydrophobic long chain groups were randomly arranged and crossed between layers (as shown in Scheme 1). The surface energy was further decreased when an extra heat treatment was made, which is suggested to be a result of reorientation where most of the hydrophobic long chain groups in LSA macromolecules were oriented to the multilayer-air interface after a high-temperature induction treatment (as shown in Scheme 1). This is in accordance with the reported results of Gustafsson et al. (2012), who found extra heat treatment significantly improved the hydrophobicity of the smooth substrate (e.g., SiO2) modified with PAH/PAA multilayers compared to that of non-heat-treated substrates, considering that an increase in temperature increased the mobility within the layers and thus resulted in reorientation of most hydrophobic groups.

Table 2 Surface energy of the handsheet prepared from modified cellulose fibers without any extra heat treatment
Full size table
Table 3 Surface energy of the handsheet prepared from modified cellulose fibers with heat treatment
Full size table

Moreover, high stability of superhydrophobic paper products is required in many packaging applications. Therefore, the stability of superhydrophobic paper was also investigated in this study. The experiments were carried out by placing water droplets onto modified handsheets at contact times from 1 to 72 h, and the corresponding water contact angles are presented in Fig. 8. It was found that the water contact angle was kept above 150º in the whole contact process, which demonstrated that high stability of the superhydrophobic paper surface was obtained, and this superhydrophobic paper will have potential application in the field of liquid packaging, such as paper cups, beverage containers, etc. However, it is worth noting that PAH is a chlorinated compound, and it is still not clear whether PAH in the multilayers will migrate into the liquid when the obtained superhydrophobic paper comes in contact with liquid during use or storage. In this study, migration of PAH in multilayers into liquid was investigated. We used hot pure water as a model liquid, and the experiments were carried out by immersing heat-treated handsheets prepared from (PAH/LSA)5 multilayer-modified fibers (0.5 g) in 90 °C hot pure water up to 24 h, followed by measuring the adsorbable organic halides (AOX) in pure water and the water contact angle of the handsheet. The AOX was determined according to Chinese standard GB/T15959 (1995) by using a Multi X® 2500 AOX analyzer (Analytik Jena AG, Eisfeld, Germany). The measured AOX value of pure water was 3.67 μg/l, whereas a handsheet immersed in pure water had an AOX value of 3.69 μg/l, only slightly higher than that of pure water. It was also found that the water contact angle of the handsheet was still kept above 150o after the immersing experiment, which demonstrated that high stability and high safety of PAH/LSA multilayers were developed.

Fig. 8
figure 8

Water contact angles of the handsheet prepared from (PAH/LSA)5 multilayer-modified cellulose fibers with heat treatment at different contact times

Full size image

Physical strength of handsheets prepared from modified fibers

The tensile strength tests were carried out for handsheets prepared from PAH/LSA multilayer-modified cellulose fibers with and without extra heat treatment to evaluate the impact of the LBL multilayers and heat treatment on the physical strength properties of cellulose fibers. The results are presented in Fig. 9. For the non-heat-treated handsheet, the tensile index continuously increased until the layer number reached up to five. Subsequently, the tensile index showed a decreasing trend with more layers deposited, and the tensile index depended on the composition of the outermost layer of the multilayers. The tensile index was significantly higher when PAH was deposited in the outermost layer as compared to when LSA was in the outermost layer. When a five-layer (2.5-bilayer) PAH/LSA multilayer was deposited on the cellulose fiber surfaces, the tensile strength index increased by 121 % compared to that of the handsheet made from original cellulose fibers (called the control handsheet), whereas when a ten-layer (5-bilayer) multilayer was deposited on the cellulose fiber surfaces, the tensile index improved by 33.9 % compared with that of the control handsheet. These results indicated that the PAH/LSA multilayer can efficiently improve the physical strength of handsheets. Moreover, it can be observed that subsequent extra heat treatment led to improvement in the tensile strength for all the handsheets. There was a 57.8 % increase in the tensile strength index when a ten-layer (5-bilayer) PAH/LSA multilayer was formed on the cellulose fiber surfaces compared to the control handsheet, suggesting a heat treatment has a significantly positive impact on handsheet strength.

Fig. 9
figure 9

Tensile strength index of non-heat-treated and heat-treated handsheets as a function of layer number, respectively

Full size image

Generally, the tensile strength mainly depends on the strength of individual fibers and the bonding strength between fibers. Therefore, the relative bonded area (RBA) between fibers is a main evaluation index for the bonding strength between fibers. The average strength of individual fibers was evaluated by zero-span tensile strength, and the results are shown in Fig. 10. The zero-span tensile strength index slightly increased with increasing layer numbers, which probably was because of the increase fiber wall thickness by depositing PAH/LSA multilayers on the cellulose fiber surface. The subsequent heat treatment further obviously improved the zero-span tensile strength of the handsheets. This may be a result of the formation of a covalent amide linkage in the PAH/LSA multilayers and between the multilayers and the fibers through induced thermal cross-linking, which is in agreement with the study results of Gustafsson et al. (2012).

Fig. 10
figure 10

Zero-span tensile strength index of non-heat-treated and heat-treated handsheets as a function of layer number

Full size image

The relative bonded area (RBA) between fibers was calculated according to the method of our previous work (Li et al. 2012), and the results are presented in Fig. 11. For the non-heat-treated handsheet, the RAB gradually increased until the layer number reached 5, and then decreased when more layers were deposited and depended on whether PAH or LSA was in the outermost layer. The RAB was higher when PAH was in the outermost layer than when LSA was in the outermost layer. This may be explained by the following plausible reasons: (1) the PAH and LSA may form a soft gel on the fiber surface when multilayers have the appropriate layer number, similar to the report of our preivous work and that of Maximova et al. (2001), increasing the RBA between fibers. With further growth of multilayers, the “gel” effect is attenuated or eliminated. (2) PAH is a strengthening additive agent used in the papermaking industry that can act as a glue to increase the RBA between fibers when deposited in outermost layer (Gimåker and Wågberg 2009; Rathi and Biermann 2000). (3) The hydrophobicity increased more when LSA was in the outermost layer, causing lower bonding between fibers. Moreover, the subsequent heat treatment resulted in a slight increase in RBA compared to the non heat-treated samples. This was possible because the increase in temperature increased both the mobility within the layers and the “gel” effect between PAH and LSA. A similar mechanism was suggested by Gustafsson et al. (2012).

Fig. 11
figure 11

Relative bonded area as a function of the layer number. The inset shows the light-scattering coefficient versus the tensile index of the handsheet

Full size image

Conclusions

A superhydrophobic paper with enhanced physical strength was prepared using a facile layer-by-layer (LBL) self-assembly of PAH and LSA on cellulose fiber surfaces, followed by a heat treatment at 160 °C for 30 min. With increasing layer numbers, two characteristic element contents (i.e., Cl and S) of PAH/LSA multilayers increased, the zeta potential of the modified fibers was regularly inversed, and the surface RMS roughness of modified cellulose increased. Wetting property testing results showed that the water contact angle was higher when LSA was in outermost layer than when PAH was in outermost layer, and the heat treatment improved the hydrophobicity of all the handsheets. PAH/LSA multilayers significantly decreased the surface energy of modified cellulose fibers after heat treatment, indicating that high temperature treatment induced the orientation of the most hydrophobic long chains of LSA to the multilayer-air interface. The heat-treated handsheet prepared from (PAH/LSA)5 multilayer-modified fibers can obtain a superhydrophobic surface with a water contact angle of 151.7º and high stability during contact with testing water droplets. In addition, the tensile strength properties of heat-treated handsheets made from multilayer-modified fibers were significantly enhanced. When a (PAH/LSA)5 multilayer was deposited on the cellulose fiber surface, the tensile strength increased by 57.8 % compared to that of handsheets made from original fibers.