Adsorption Characteristics of Lignosulfonates in Salt-Free and Salt

Jul 20, 2011 - The driving forces of self-assembly of SL and PDAC are discussed based on the solution behaviors and adsorption characteristics of SL i...
0 downloads 11 Views 3MB Size
ARTICLE pubs.acs.org/Biomac

Adsorption Characteristics of Lignosulfonates in Salt-Free and Salt-Added Aqueous Solutions Xinping Ouyang,† Yonghong Deng,† Yong Qian,† Pan Zhang,† and Xueqing Qiu*,†,‡ †

School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou, People's Republic of China, 510640 ‡ State Key Lab of Pulp and Paper Engineering, Guangzhou, People's Republic of China, 510640

bS Supporting Information ABSTRACT: Five sodium lignosulfonate (SL) fractions with narrow molecular weight distribution and known salt content were used as the polyanion to build up layer-by-layer self-assembly multilayers with poly(diallyldimethylammonium chloride) (PDAC) as polycation. It is interesting to find that the salt-free SL is hardly adsorbed on the PDAC surface, but the SL in salt-added solutions can be self-assembled well with PDAC to form SL/PDAC multilayers. When the five SL fractions dissolved in saline solutions are adsorbed on the PDAC surface by a self-assembly technique, SL with higher Mw shows a higher adsorption amount than does SL with lower Mw. The driving forces of self-assembly of SL and PDAC are discussed based on the solution behaviors and adsorption characteristics of SL in salt-free and saltadded aqueous solutions. A possible self-assembled mechanism of SL and PDAC is electrostatic or cation-π interactions, but the SL cannot be adsorbed onto the PDAC surface without a hydrophobic interaction. With the addition of enough salt, the Coulomb interaction of SL becomes negligible, but the adsorption amount increases, indicating that the electrostatic interaction is not the main driving force of SL/PDAC self-assembly. For adsorption of SL in saline solution onto the PDAC surface, the cation π interaction is the main driving force, and the hydrophobic interaction plays an important role in the adsorbed amount.

’ INTRODUCTION Sodium lignosulfonate (SL) is generally present as a byproduct in spent sulfite pulping liquor from the paper and pulping industry. In view of favorable wettability, adsorptivity, and dispersive ability, SL can be used as a polymer dispersant in various application areas.1a e The performance of SL in the field of polymer dispersants is dependent on the adsorption characteristics of SL on solid particles, which is significantly influenced by the SL solution behaviors. A better understanding of the solution behaviors and adsorption characteristics of SL is necessary for facilitating its value-added application. The molecular conformation of lignosulfonates in solutions has been widely reported.2 5 Goring introduced a microgel model,2 but Myrvold considered that the structure of lignosulphonate was not a microgel but a randomly branched polyelectrolyte.3 Kontturi et al. found that the effective charge number of lignosulfonates dropped suddenly to zero at a critical temperature and suggested that the charge discharge transition might have something to do with the spherical conformation.4a d Based on a curve fitting result of small-angle X-ray scattering experiments, Vainio et al. reported that an oblate spheroid shape might describe the average shape of the lignosulfonate particles in saline solutions.5 The molecular conformation of lignosulfonates in solutions is still in debate. A further investigation is necessary to clarify the molecular conformation of SL. The adsorption behavior of SL on solid particles is reported to be affected by its molecular weight (Mw)6 8 and ionic strength of dispersion system.9 Among these earlier studies there exist two r 2011 American Chemical Society

difficulties in understanding the adsorption characteristics of a SL layer at a molecular level. On the one side, the structure of SL is very complex, and the raw SL material recovered from the spent pulping liquor has a wide molecular weight distribution and unknown salt content. On the other side, the dispersed solid particles are of all kinds of shapes, and the adsorbed thickness of lignosulfonate is very thin,10 which makes it hard to detect the adsorption characteristics of SL. To overcome the first difficulty, salt-free SL samples with a low polydispersity index of molecular weight are required. To overcome the second difficulty, we can select flat substrates instead of irregular solid particles and use a well-known layer-by-layer (LBL) selfassembly technique11,12 to prepare SL multilayer films instead of a thin SL monolayer. In this way, the adsorption behaviors of SL can be well characterized by a wide variety of physical techniques. Until now, there have been a few reports about the selfassembled multilayers of lignin and cations.13 16 For the driving force of self-assembly, Notley and Norgren considered that the adsorption of PDAC to lignin is of a pure electrosorption nature.15 In contrast, Pillai and Renneckar concluded that the self-assembly of lignin biopolymers on PDAC is driven by a strong cation π interaction.16 Further investigation of the adsorption mechanism behind the self-assembly process is necessary for Received: June 14, 2011 Revised: July 18, 2011 Published: July 20, 2011 3313

dx.doi.org/10.1021/bm200808p | Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules

ARTICLE

understanding the basic structure and intermolecular interaction of the complex lignin biopolymers. In the present work, five SL fractions with narrow Mw distribution isolated by preparative column chromatography are used as polyanions to fabricate LBL self-assembled multilayers with PDAC as the polycation, and the effect of salt on the solution behaviors and adsorption characteristics of SL were investigated for the purpose of giving insight into the adsorption mechanism of SL on the PDAC surface.

’ EXPERIMENTAL SECTION Materials. The raw SL, supplied by Shixian Papermaking Co. Ltd. (China), was recovered from spent sulfite pulping liquor. A total of 10 wt % of SL aqueous solutions was first filtered to remove the insoluble solid matter, and then the low Mw impurities were removed by using an ultrafiltration apparatus (Wuxi Membrane Science and Technology Co., China) with a 1000 Da (Da) cutoff membrane. Five SL fractions with narrow Mw distribution were obtained through the method of gel column chromatographic separation, as described in ref 17. To remove salt, SL separated by column chromatography was further treated by column chromatography with water as an eluant and then followed by dialysis against water. Poly(diallyldimethylammonium chloride) (PDAC, Mw of 200000 350000, 20% solution, Aldrich) was used as a polycation and was diluted to a concentration of 0.1 mmol/L (repeated unit). Water used in this work was ultrapure water obtained from a Millipore water purification system, and the water resistivity is larger than 18 MΩ 3 cm. The other reagents were purchased commercially as analytical grade products and used directly without further purification, unless otherwise indicated. Characterization. The determination of electrical conductivity was conducted with 1 g/L of SL at 25 C with a DDSJ-308A conductivity meter (Shanghai Shengci Co., China). The Mw and distributions of SL were determined by gel permeation chromatography (GPC) with Ultrahydragel 120 and Ultrahydragel 250 columns. The 0.10 mol/L NaNO3 aqueous solution with pH 8 was used as the eluent at a flow rate of 0.5 mL/min. The effluent was monitored at 280 nm with a Waters 2487 UV Absorbance Detector (Waters Corp., U.S.A.) at a flow rate of 1.0 mL/min. Poly(styrene sulfonate) (PSS) was employed as the standard substance. Elemental analysis of SL was performed by Elemental Analyzer (PE2400II, Perkin-Elmer). The UV vis absorption measurements were performed with a UV vis spectrophotometer (UV-2450, Shimadzu Corp., Japan). For the sample solutions with water as the solvent, water was scanned at the same wavelength as a baseline. The fluorescence measurements were performed on a Fluorosens System (Gilden Photonics Ltd., England) equipped with a 150 W Xenon arc Lamp at 298 K. The excitation spectra of samples at different concentrations were recorded in a range of 200 520 nm using an emission wavelength of 530 nm, the slit width for both emission and excitation was 2.5 nm, and integration time was 100 ms. Before measurement, the solution was ultrasonicated for 10 min and then kept undisturbed for 10 min. Static light scattering (SLS) and dynamic light scattering (DLS) were performed on a commercial light scattering instrument (ALV/CGS-3, Germany) equipped with a multi-τ digital time correlator (ALV-7004) and a solid-state He Ne laser (JDS-Uniphase, output power = 22 mW, at = 632.8 nm). Reduced Rayleigh ratios were obtained for the angular range 30 150. The effect of fluorescent emission on the light-scattering of SL was eliminated by inserting a high performance laser-line bandpass filter (NT47-494) between the sample solution and the detector unit.22 24 All the experiments were performed at 25 C over

Figure 1. Conductivity of 1 g/L sodium lignosulfonates obtained from different separation methods: sample 1, ultrafiltration; sample 2, ion exchange after ultrafiltration; sample 3, column chromatographic separation after ultrafiltration; sample 4, dialysis after chromatographic separation. a range of concentrations and detection angles. The derivative refractive index increment (dn/dC) was measured by DnDc-2010 differential refractometer from WGE Dr Buresr (Germany). The wavelength at which dn/dC is determined is 620 nm. The dn/dC values for all SL samples were determined reproducibly. Film thickness was determined by the spectroscopic ellipsometer (UVISEL-NIR-FGMS, Horiba). Surface roughness was measured by AFM (DI Multimode SPM nanoscope V, Veeco). Self-Assembled Film. The SL/PDAC LBL self-assembled films of desired bilayers were deposited on the Quartz slide substrates (50  14  8 mm). Prior to deposition, the Quartz slide was sonicated in a 98% H2SO4/30%H2O2 solution (piranha solution) for 1 h and in a H2O/ H2O2/NH4OH (5:1:1) solution for 1 h, followed with a thorough rinse and dried with an air stream. The film deposition process involved the repeated sequential dipping of the substrate into the polycation and polyanion solutions at pH 6.7 for 10 min, with rinsing between each of the deposition steps.

’ RESULTS AND DISCUSSION SL Fractions. Derived from byproduct in spent pulping liquor, raw SL usually contains various salts that affect the properties of SL significantly. Conductivity measurement is the most convenient method for testing the salt content of SL samples. Figure 1 shows the conductivity of SL obtained from different isolation methods. The conductivity of SL sample is 10372 μS/cm after ultrafiltration, and then becomes 358 μS/cm followed by ion exchange. During gel column chromatographic separation, the eluant used in the process is 0.20 mol/L NaCl aqueous solution.17 After column chromatographic separation, the salt content in SL solution was high. When the sample is further treated by column chromatography with Milli Q water as an eluant, the conductivity of SL sample is 57 μS/cm. After being treated by dialysis against the Milli Q water, SL has the lowest conductivity of 15 μS/cm. There has not been any reports on the electrical conductivity of SL so far because it is very difficult to obtain the pure SL. Therefore, a low of 15 μS/cm of the conductivity can be attributed to SL itself. In view of the difference in the properties of SL with different Mw, SL fractions with narrow Mw distribution were prepared by gel column chromatographic separation for the following fundamental investigation. The Mw distributions of SL and its five fractions fractionated by column choromatography are presented in Figure 2. Before chromatographic separation, SL had a 3314

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules

Figure 2. Weight-average Mw distributions (by GPC) of SL and its five fractions.

Figure 3. UV vis absorption spectra of the five SL fractions, where the absorption intensity at 280 nm is kept the same by adjusting the concentration: SL-1, C = 0.1288 g/L; SL-2, C = 0.1104 g/L; SL-3, C = 0.0920 g/L, SL-4: C = 0.0920 g/L; SL-5, C = 0.0920 g/L.

weight-averaged Mw of 10500 Da with polydispersity index of 4.55. After chromatographic separation, each of the five SL fractions has a separated Mw with a polydispersity index less than 1.5. Besides Mw distribution, SL also has a polydispersity in the loading density of the functional sulfonic groups. When SL is isolate into five fractions, the content of the sulfonic groups vary with different SL fractions. By elemental analysis, it is estimated that SL includes elemental sulfur ranging from 8.60 to 5.65 wt % with increasing Mw (Table S1), indicating that the contents of sulfonic groups decrease with the increase of Mw. Spectral Analysis. Based on its aromatic structure, SL strongly absorbs UV light and exhibits the main characteristic absorption of the π π* electron transition. Figure 3 shows the UV vis absorption spectra of the five SL fractions. The absorption peak at about 280 nm are designated B band of the π π* transition of the C9 unit. The broad shoulder band of SL ranging from 290 to 500 nm is attributed to the overlapped bands of the C9 unit with the other conjugated groups formed in the pulping process.18 Here, the absorption band at 280 nm is considered as the characteristic maximum of SL, indicating that the guaiacyl structural units are predominant in this SL fractions.18 The π π interaction of the aromatic groups in lignin derivatives has been extensively investigated.19 21 SL is an amphiphilic biopolymer consisting of hydrophobic aromatic groups and hydrophilic sulfonic groups. When SL is dispersed into water,

ARTICLE

Figure 4. Fluorescent emission spectra of SL-3, PDAC, SL-3/PDAC mixture, SL-3/NaCl mixture, and SL-3/PDAC/NaCl mixture: SL-3, 0.8 g/L; PDAC, 0.4 mmol/L; NaCl, 0.1 mol/L.

these hydrophobic aromatic groups tend to come together for minimizing entropic loss. When the aromatic groups approach each other at a distance that is small enough to produce coupling effect and energy splitting, they form π π aggregates. The effect of salt on the π π aggregation of the aromatic groups can be investigated by UV vis spectroscopy and fluorescent excitation spectroscopy. With the addition of NaCl ranging from 1  10 5 to 2 mol/L, both UV vis spectroscopy and fluorescent excitation spectroscopy show that the SL in salt-added solutions has no spectral change, indicating that salt has no influence on π π aggregation of the aromatic groups in SL. Usually, salt causes polyelectrolytes to shrink because it screens the charges and inhibits the Coulomb interactions in polyelectrolytes. It is possible that the aromatic groups in π π aggregates are charge-free, so salt does not cause these aromatic groups to shrink. Based on SL “microgel model” noted by Goring et al.,2 SL is assumed to have free charge only on the surface, and its interior is likely built by the charge-free aromatic groups in π π aggregates. Salt can still make SL polyelectrolytes to shrink because of charge-screening, but this shrinkage only occurs at the charged surface of the SL microgel. It is reported that there exists cation π interaction between SL and PDAC.16 This cation π interaction can be detected by fluorescent excitation spectroscopy. Figure 4 shows the fluorescent emission spectra of SL, PDAC, SL/PDAC mixture, SL/ NaCl mixture, and SL/PDAC/NaCl mixture. Both SL and PDAC solutions have a maximum emission wavelength of 530 nm. With the addition of PDAC, the maximum emission peak of SL has a spectral shift from 530 to 500 nm, indicating that there exists cation π interaction between PDAC and the aromatic groups of SL. As shown in Figure 4, NaCl does not cause any energy shift of SL in the absence of PDAC. However, when NaCl is added into SL/PDAC solutions, it causes an increasing intensity in the emission wavelength at 500 nm, demonstrating that the cation π interaction between PDAC and SL is enhanced with addition of salt. Light Scattering. Figure 5 is a typical Zimm plot of SL-3 in aqueous solutions with a concentration ranging from 0.2040 to 0.6105 g/L. The absolute Mw, the Z-averaged radius of gyration ÆRgæ, and the second virial coefficient A2 of SL are estimated to be 1.05  105 g/mol, 57 nm, and 1.51  10 6 mol 3 dm3/g2, respectively. A2 describes the strength of interaction between 3315

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules

Figure 5. Typical Zimm plot of SL (SL-3) in aqueous solution at 25 C, where C ranges from 0.2040 to 0.6105 g/L. K is a constant for a given solution and temperature and k is a constant to spread the plot.

the molecule and the solvent, and A2 > 0 indicates the molecules tend to stay in solution. The Mw measured by SLS is larger than that obtained by GPC. SLS is an absolute method but GPC is a relative method for Mw determination. The Mw standards used in GPC measurement are PSS with a linear structure other than lignosulfonates with a branched structure, which causes difference in volume due to different shapes and extension. SL had more compact structure than PSS standards, and thus, the absolute Mw of SL was larger than that of PSS standard at the same elution volume.24,25 The polyelectrolytes effect in light scattering measurement can be inhibited by addition of salt, but salt leads to a slight increase in molecular weight and a significant decrease in Rg and dn/dC of SL (Table S2). A slight increase in molecular weight indicates a slight molecular aggregation. The effective charge of SL-3 decreases with increasing salt content, resulting in molecular shrinkage and hence the Rg decrease. The decrease of dn/dC may be caused by molecular shrinkage.25,26 The dn/dC values and the molecular parameters of SL obtained from the SLS measurement were different for the five SL fractions (Table S3). For salt-free SL solutions, there was an increase in the dn/dC with increasing Mw, which is in agreement with ref 2a. However, when 0.1 mol/L NaCl is added into the five SL fractions, the dn/dC values have an apparent decrease, especially for SL with higher Mw. SL with higher Mw has a lower percentage of sulfonic groups (Table S1). When salt at the same concentration is added into the five SL fractions, SL with higher Mw has a higher molar ratio of salt to the sulfonic groups and, thus, has a larger molecular shrinkage. Molecular shrinkage causes dn/dC to decrease. The Mw of SL measured by SLS is larger than that obtained by GPC, and the Mw of the SL with higher Mw has a more significant enlarging effect. This phenomenon was also reported in ref 25, but the reason is unclear until now. The SL microgel model may be useful to explain it. The SL microgel with higher Mw has more compact structure than does the SL microgel with lower Mw, and thus, its ratio of the absolute Mw (by SLS) to relative Mw (by GPC) is larger. DLS was used to determine the hydrodynamic radius (Rh) of SL in saline aqueous solutions. Here, Rh is the radius of an equivalent sphere that has a friction f identical to that of the SL particle during its Brownian motion in solution. Figure 6 shows the CONTIN analysis of the DLS measurement for SL in saltfree and salt-added solutions in terms of Mw. For both salt-free and salt-added SL solutions, there exists a bimodal Rh distribution

ARTICLE

corresponding to a “fast mode” and “slow mode”. The Rh,fast of the SL fractions increases with the increase of Mw, but Rh,slow of the five SL fractions does not have such a similar law. When 0.1 mol/L of NaCl is added into SL aqueous solution, the averaged size of Rh,fast decreases, and the relative intensity ratio of the Rh, slow to the Rh,fast decreases. The origin of the “fast mode” is attributed to the single molecules, but that of the “slow mode” is in debate and controversy.27 30 Figure 7 presents the DLS results of SL-3 aqueous solutions with different salt concentrations, which shows that the averaged size of Rh of the single SL molecules decreases with the increase of the concentration of salt, resulting from molecular shrinkage. When the salt concentration increases, the relative intensity ratio of the Rh,slow to the Rh,fast decreases gradually. As noted by F€orster et al., the “slow mode” of SL was related to the Coulomb interactions, and its disappearance means elimination of the Coulomb interactions.27,28 According to F€orster et al., Coulomb interactions become negligible if the molar concentration ratio cmonomer/csalt < 25 for branched charged polymers (or cmonomer/ csalt < 1 for linear charged polymers);27,28 therefore, the theoretical NaCl concentration to eliminate the Coulomb interactions was calculated to be about 5  10 5 mol/L for branched charged polymers (or 1.18  10 3 mol/L for linear charged polymers). However, the elimination of the Coulomb interactions for SL solutions occurs only when the NaCl concentration reaches 1 mol/L (Figure 9). It is obvious that the actual salt content required to eliminate the Coulomb interactions of SL is much larger than the calculated salt content based on F€orster’s theory, but the reason is unclear. One of the possible reasons is that the structure of SL is a compact microgel; the penetration of the hydrated counterion inside the polyelectrolyte body should be much easier in a branched or linear charged polymer than in a compact microgel. Another explanation can be referred to Kontturi et al.,4d who found that the effective charge of SL in 1 mol/L NaCl aqueous solution was lost at a critical temperature (35 C) in an external electric field and concluded that this charge discharge transition could occur only when the macromolecule is a compact sphere. Self-Assembly of SL and PDAC. Five SL fractions with narrow Mw distribution and known salt content were used as polyanion dipping solutions to build up LBL self-assembly multilayers with polycation PDAC. The λ280 nm is considered to be the characteristic absorption peak of SL, and thus, UV vis spectroscopy can be used to monitor the self-assembly process of SL/PDAC multilayer films. Figure 8 shows the UV vis spectra of the SL/PDAC multilayers varying with the number of bilayers. Here, 1 g/L SL-5 in the 0.1 mol/L NaCl aqueous solution is used for the polyanion dipping solution. Except for the first several bilayers having a substrate effect, a linear increase of the absorbance at λ280nm with the increase of the number of bilayers is observed during self-assembly process of SL and PDAC. It indicates that the adsorption process is reproducible from layer to layer and an equivalent amount of SL is adsorbed on the PDAC surfaces in each dipping cycle. Here, the absorbance at λ280nm in the UV vis spectra of the SL/PDAC multilayers represents the adsorption amount of SL. As shown in Figure 9a, the adsorbed amount of SL increases with increasing ionic strength, indicating that ionic strength has a significant influence on the adsorption behaviors of SL. Except for the first three bilayers having a substrate effect, the adsorption amount of SL had a linear increase with the number of bilayers. Figure 9b shows a plot of the absorbance at λ280nm for the 3316

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules

ARTICLE

Figure 6. DLS: CONTIN analysis of five SL fractions in (a) “salt-free” solution and (b) “salt-added” solution (0.1 mol/L NaCl solutions).

Figure 7. DLS: CONTIN analysis of SL-3 in saline solutions with different NaCl concentrations.

Figure 8. UV vis spectra of SL/PDAC multilayers with different bilayer numbers (0.1 wt % SL-5 with 0.1 mol/L NaCl).

10-bilayer SL/PDAC multilayers as function of CNaCl. With increasing ionic strength, the adsorption amount of SL has a sharp increase when CNaCl is 0 0.1 mol/L, and then has a slight increase when CNaCl is between 0.1 and 1.0 mol/L. The SL with different Mw has a different adsorption behavior. Figure 10a shows the relationship between the adsorption amount of five SL fractions in saline solutions and number of bilayers. Because SL with different Mw has a different content of C9 units, the concentration of five SL fractions used for selfassembly is adjusted to reach the same absorption intensity at 280 nm. During self-assembly of SL and PDAC, the adsorption

amount has a linear increase with the number of bilayers, and the slope increases with increasing Mw. It indicates that the SL with higher Mw has a larger adsorbed amount on the PDAC surface. However, when PDAC-modified slides were dipped into the salt-free SL solutions, the SL was hard to deposit on the PDAC surface. Figure 10b shows the relationship between the absorption amount of the five salt-free SL fractions and the number of dipping cycles. Obviously, the adsorption amount is about zero for 10-layer SL/PDAC multilayers taking the experimental error into account, and it is difficult to detect the difference in the adsorption amount for the salt-free SL fractions with different molecular weights. For the five SL fractions in salt-added solutions, we have also investigated the film thickness and surface roughness of SL/ PDAC multilayers by ellipsometer and AFM and found that both of them increased with increasing molecular weight (Table S4). However, for the five SL fractions in salt-free solutions, the film thickness of SL/PDAC multilayers is about zero. The surface is rather flat, and it is hard to detect the difference in roughness among the five SL/PDAC multilayers obtained from SL with different molecular weights. Driving Force of Self-Assembly. It is interesting that the SL in salt-added solutions can be self-assembled well with PDAC to form SL/PDAC multilayer film, but the salt-free SL is hardly adsorbed on the PDAC surface. SL with different Mw has a different adsorpion amount, but this difference only appears when the SL dipping solution possess a certain ionic strength. Although the layer-by-layer self-assembly mechanism and the influence of ion strength on this self-assembly have been widely investigated for linear or branched charged polymers,31 34 the adsorption mechanism of SL on PDAC and its salt effect seem to have a different rule regarding the specific microgel structure of SL in aqueous solutions. The adsorption mechanism of SL on PDAC through layer-bylayer self-assembly technique is initially considered as a pure electrosorption nature.15 For SL in salt-free solutions, the structure of SL is assumed to be a swollen microgel with charged groups only at the surface, and each charged sulfonic group is surrounded by two hydration shells. The first hydration shell is a tightly bound inner shell where the water molecules are adjacent to the charged groups, and the second hydration shell is a loose bound outer shell that is critical for the groups to keep its charges.4d The adsorption of SL on PDAC is related to the ionic binding between a hydrated polyanion and a hydrated polycation taking place with sharing of the second hydration shells. Because the molecular distance between the mass center of SL microgel 3317

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules

ARTICLE

Figure 9. (a) The absorbance at λ280nm of SL/PDAC multilayers varying with bilayer numbers. The SL/PDAC multilayers were obtained from SL-5 in saline solutions with different NaCl concentrations. (b) The absorbance at λ280nm of 10-bilayer SL/PDAC multilayers as a function of CNaCl.

Figure 10. Relationship of the absorbance at 280 nm of SL/PDAC multilayers and number of bilayers. The five SL/PDAC multilayers were obtained from five SL fractions: (a) SL in salt-added solutions; (b) SL in salt-free solutions.

Figure 11. Schematic representation for self-assembled adsorption characteristics of SL in salt-free and salt-added solutions.

and PDAC is much larger than that between linear polyanion (i.e., PSS) and PDAC, the self-assembled driving force of PDAC and SL with cation anion interaction is not strong to bind SL and PDAC.31 Pillai and Renneckar suggested that the self-assembly of SL biopolymers on PDAC is driven by a cation π interaction.16 During self-assembly of salt-free SL and PDAC, the swollen SL “microgel” is surrounded by two hydration shells, so the distance between the aromatic rings in SL and PDAC is not small enough for charge transfer from the benzene rings in SL to PDAC. Therefore, the cation π interaction is not strong enough to bind SL and PDAC. Figure 11 illustrates the adsorption behaviors of SL on PDAC. For SL in salt-free solutions, the electrostatic and cation π interactions are the possible self-assembled mechanism of SL and PDAC, but the SL cannot be adsorbed onto the PDAC surface

without hydrophobic interaction. SL microgel may be adsorbed on PDAC driven by electrostatic and cation π interactions, but the SL microgel is easily desorbed from the SL/PDAC film during the rinse process, resulting in a net adsorption amount of approximate zero. Actually, the effect of salt on the solution and adsorption behaviors of polyelectrolyte is very complex.31 33 The main difficulties for understanding the solution behaviors are related to the long-range character of Coulomb interactions and to the nonlinear character of charge-screening of polyions.34 Here, NaCl is added into the SL polyanion to screen the Coulomb interactions. Considering the effect of ionic strength on the adsorption behaviors of polyanions, four aspects35 can be taken into account: (1) the anions of salt compete against the polyanions for binding sites; (2) the cations of salt shield the polyanions to reduce the affinity between polyanions and polycations; (3) the change of the molecular conformation results in variation of the hydrophobic interaction; (4) the shrinkage of polyelectrolytes causes an increase in adsorption amount on the substrates with the same binding sites. Among the four aspects, the hydrophobic interaction and the shrinkage of polymer chains are considered as the main factors for the effect of salt on the adsorption behaviors of SL. Moreover, the unique microgel structure of SL should also be taken into account because it may cause a different adsorption behavior from the other branched and linear polyelectrolytes. 3318

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules For SL in salt-added solutions, the adsorption amount of SL increases with salt contents. Especially, when enough salt (e.g., 1 mol/L NaCl) is added into SL solutions, the Coulomb interaction of SL becomes negligible (Figure 7), but the adsorption amount of SL continues to increase, indicating that the electrostatic interaction is not the main driving force of SL/PDAC self-assembly. Based on the solution behaviors and the adsorption characteristic of SL in saline solutions, we consider that the cation π interaction is the main driving force for adsorption of SL onto the PDAC surface, and the hydrophobic interaction plays an important role in the adsorbed amount (Figure 11). Based on the “microgel” model noted by Goring et al., where SL was assumed to have free charges only on the surface, a modified “microgel” model is put forward to assume that SL interior consists of the charge-free aromatic groups. These charge-free aromatic groups tend to form π π aggregates due to π π interactions. With the addition of salt, the SL microgel shrinks at the surface significantly, as detected by a decrease in Rg, because the surface charges are screened by a cloud of counterions. The SL microgel interior does not shrink, as detected by no spectral shift, because the component aromatic groups are charge-free, which guarantees that the coupling aromatic groups in π π aggregates is close to the PDAC-coated substrates. When the distance between the benzene rings of SL and polycation PDAC is small enough for experiencing charge transfer, the SL/PDAC charge transfer complex can be formed due to cation π interaction. Because there exists coupling aromatic groups in π π aggregates, cation π interaction is expected to be enhanced significantly when the π π aggregates of aromatic groups is close to the PDAC surface. Because of hydrophobic interaction, the distance between the benzene rings of SL and polycation PDAC decreases with increasing salt content, the cation π interaction is enhanced as detected by fluorescent emission spectroscopy (Figure 4). Hydrophobic interaction causes an increase in adsorption amount of SL at the solid liquid interface. SL with higher Mw has a stronger hydrophobic interaction. When these five SL fractions at the same C9 unit concentration were adsorbed on the PDAC surface by self-assembly technique, SL with higher Mw shows a higher adsorption amount than does SL with lower Mw (Figure 10a).

’ CONCLUSIONS This work studied the solution behaviors and the adsorption characteristics of SL in salt-free and salt-added aqueous solutions in term of molecular weight. Five SL fractions with narrow molecular weight distribution and known salt content were used as polyanion dipping solutions to build up layer-by-layer selfassembly multilayers with PDAC. Results show that hydrophobic interaction plays an important role in adsorption of SL at the solid liquid interface. Without hydrophobic interaction, SL cannot be adsorbed onto the PDAC surface. Because of hydrophobic interaction, SL with higher Mw shows a higher adsorption amount than does SL with a lower Mw. With the addition of enough salt, the Coulomb interaction of SL becomes negligible, but the adsorption amount increases, indicating that the electrostatic interaction is not the main driving force of SL/PDAC selfassembly. Cation π interaction is improved with increasing salt content, which is accompanied by an increase in adsorption amount of SL. It indicates that cation π interaction is the main driving force for SL/PDAC self-assembly.

ARTICLE

’ ASSOCIATED CONTENT

bS

Supporting Information. Weight-average Mw, polydispersity, and elemental composition of SL fractions, values of molecular parameters of SL fractions from light scattering measurements, and the film thickness and surface roughness of SL/ PDAC multilayers. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT The authors thank the financial support of the China Excellent Young Scientist Fund (20925622), the National Natural Science Foundation of China (20976064, 20876064), and the National Basic Research Program (973 Program; 2010CB732205). ’ REFERENCES (1) (a) Ouyang, X. P.; Ke, L. X.; Qiu, X. Q.; Guo, Y. X.; Pang, Y. X. J. Dispersion Sci. Technol. 2009, 30, 1. (b) Zhou, M. S.; Qiu, X. Q.; Yang, D. J.; Lou, H. M.; Ouyang, X. P. Fuel Process. Technol. 2007, 88, 375. (c) Li, Z. L.; Pang, Y. X.; Lou, H. M.; Qiu, X. Q. Bioresources 2009, 4, 589. (d) Ouyang, X. P.; Qiu, X. Q.; Lou, H. M.; Yang, D. J. Ind. Eng. Chem. Res. 2006, 45, 5716. (e) Liu, H.; Zhu, J. Y. Bioresour. Technol. 2010, 101, 9120. (2) (a) Rezanowich, A.; Goring, D. A. J. J. Colloid Sci. 1960, 15, 452. (b) Goring, D. A. I.; Vuong, R.; Gancet, C.; Chanzy, H. J. Appl. Polym. Sci. 1979, 24, 931. (3) Myrvold, B. O. Ind. Crops Prod. 2008, 27, 214. (4) (a) Kontturi, A. K. J. Chem. Soc., Faraday Trans. 1 1988, 84, 4033. (b) Kontturi, A. K.; Kontturi, K.; Niinikoski, P. J. Chem. Soc., Faraday Trans. 1991, 87, 1779. (c) Kontturi, A. K.; Ky€osti Kontturi, K.; Niinikoski, P.; Murtom€aki, L. Acta Chem. Scand. 1992, 46, 941. (d) Mafe, S.; Manzanares, J. A.; Kontturi, A. K.; Kontturi, K. Bioelectrochem. Bioenerg. 1995, 38, 367. (5) Vainio, U.; Lauten, R. A.; Serimaa, R. Langmuir 2008, 24, 7735. (6) Yang, D. J.; Qiu, X. Q.; Zhou, M. S.; Lou, H. M. Energy Convers. Manage. 2007, 48, 2433. (7) Anderson, P. J.; Roy, D. M.; Gaidis, J. M. Cem. Concr. Res. 1988, 18, 980. (8) Anita, A.; Marek, P. Miner. Eng. 2007, 20, 609. (9) Bai, B. J.; Wu, Y. F.; Grigg, R. B. J. Phys. Chem., C 2009, 113, 13772. (b) Grigg, R. B.; Bai, B. J. J. Colloid Interface Sci. 2004, 279, 36. (10) Palmqvist, L.; Holmberg, K. Langmuir 2008, 24, 9989. (11) Iler, R. K. J. Colloid Interface Sci. 1966, 21, 569. (12) Decher, G.; Hong, J. D. Makromol. Chem. Macromol. Symp. 1991, 46, 321. (13) Paterno, L. G.; Mattoso, L. H. C. Polymer 2001, 42, 5239. (14) Liu, H.; Fu, S. Y.; Li, H.; Zhan, H. Y. Ind. Crops Prod. 2009, 30, 287. (15) Notley, S. M; Norgren, M. Biomacromolecules 2008, 9, 2081. (16) Pillai, K. V.; Renneckar, S. Biomacromolecules 2009, 10, 798. (17) Ouyang, X. P.; Zhang, P.; Tan, C. M.; Deng, Y. H.; Yang, D. J.; Qiu, X. Q. Chin. Chem. Lett. 2010, 21, 1479. (18) Lin, S. Y.; Dence, C. W. Methods in Lignin Chemistry; SpringerVerlag: Berlin, 1992. (19) Deng, Y. H; Feng, X. J.; Zhou, M. S.; Qian, Y.; Qiu, X. Q. Biomacromolecules 2011, 12, 1116. (20) (a) Jelley, E. E. Nature 1936, 138, 1009. (b) Scheibe, G. Angew. Chem. 1936, 49, 563. 3319

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320

Biomacromolecules

ARTICLE

(21) (a) McRae, E. G.; Kasha, M. J. Chem. Phys. 1958, 28, 721. (b) Kasha, M. Radiat. Res. 1963, 20, 55. (c) Kasha, M.; Rawls, H. R.; El Bayoumi, M. A. Pure Appl. Chem. 1965, 11, 371. (22) Sarkanen, S.; Teller, D. C.; Stevens, C. R.; McCarthy, J. L. Macromolecules 1984, 17, 2588. (23) Guerra, A.; Gaspar, A. R.; Contreras, S.; Lucia, L. A.; Crestini, C.; Argyropoulos, D. S. Phytochemistry 2007, 68, 2570. (24) Fredheim, G. E.; Braaten, S. M.; Christensen, B. E. J. Chromatogr., A 2002, 942, 191. (25) Contreras, S.; Gaspar, A. R.; Guerra, A.; Lucia, L. A.; Argyropoulos, D. S. Biomacromolecules 2008, 9, 3362. (26) (a) Sorlier, P.; Rochas, C.; Morfin, I.; Viton, C.; Domard, A. Biomacromolecules 2003, 4, 1034. (b) Kulicke, W. M.; Horl, H. H. Angew. Makromol. Chem. 1983, 149. (c) Stejskal, J.; Horska, J. Makromol. Chem. 1982, 10, 2527. (d) Dupont, A.-L.; Harrison, G. Carbohydr. Polym. 2004, 58, 233. (27) F€orster, S.; Schmidt, M. Adv. Polym. Sci. 1995, 120, 51. (28) F€orster, S.; Hermsdorf, N.; B€ottcher, C.; Lindner, P. Macromolecules 2002, 35, 4096. (29) Borsali, R.; Nguyen, H.; Pecora, R. Macromolecules 1998, 31, 1548. (30) Sch€artl, W. Light Scattering from Polymer Solutions and Nanoparticle Dispersions; Springer: Berlin, 2007. (31) Durstock, M F; Rubner, M F Langmuir 2001, 17, 7865–72. (32) Farhat, T R; Schlenoff, J B Langmuir 2001, 17, 1184–94. (33) Milkova, V.; Radeva, Ts. J. Phys.: Condens. Matter 2010, 494107. (34) Borisov, O. V.; Zhulina, E. B. Eur. Phys. J. B 1998, 4, 205. (35) Lan, Q. D.; Bassi, A. S.; Zhu, J. X.; Margaritis, A. Chem. Eng. J. 2001, 81, 179.

3320

dx.doi.org/10.1021/bm200808p |Biomacromolecules 2011, 12, 3313–3320