Cellulose (2009) 16:467–479 DOI 10.1007/s10570-009-9281-5
Adsorption of chlorhexidine onto cellulosic fibers E. Gime´nez-Martı´n Æ M. Lo´pez-Andrade Æ A. Ontiveros-Ortega Æ M. Espinosa-Jime´nez
Received: 8 July 2008 / Accepted: 30 January 2009 / Published online: 20 February 2009 Ó Springer Science+Business Media B.V. 2009
Abstract Comparative investigations of adsorption properties of chlorhexidine (CHX) on two cellulose fibers, bleached cotton and viscose, were studied in order to obtain dry gauzes covered with known amount of this antiseptic. Adsorption isotherm results carried out at 293 and 323 K can be described by Langmuir isotherm model, nevertheless, at high concentration correlation is better to Freundlich isotherm. Electrokinetic potential evolution with CHX concentration, shows that initial negative zeta potential of cotton and viscose diminish its absolute value as the concentration of the treatment increases; both fibers present an isoelectric point at high concentration of CHX that is 0.3 mM for viscose and 0.8 mM for cotton. Electrostatic interactions between cationic groups of CHX and carboxylic acid groups of the fibers could explain adsorption at low concentration, but when it is higher than these values, possible hydrogen bonding between the amine groups of CHX and hydroxyl groups of cellulose could
explain increasing adsorption when it is hindered by electrostatic repulsion as it is predicted by Freundlich model, that describes heterogeneous surface and multilayer adsorption. Adsorption kinetics isotherms reveal that the process is quick with t1/2 values of 5.4 min for cotton and 2.8 min for viscose. Differences in adsorption behaviour between the two fibers could be attributed to structural differences as we have observed from estimation of CI index based on FTIR spectra. Values obtained 1.6 for viscose and 2.2 for cotton could explain that the amount of CHX adsorbed on viscose is higher than it is on cotton. Finally desorption experiments performed with 0.01 M of NaCl solution at room temperature and pH 6 reveals the possibility of therapeutical application of these fibers although further investigations must be done to optimize the process.
E. Gime´nez-Martı´n (&) A. Ontiveros-Ortega M. Espinosa-Jime´nez Department of Physics, University of Jae´n, Paraje de las Lagunillas s/n, Edificio A-3, Dependencia 412, 23071 Jaen, Spain e-mail:
[email protected]
Introduction
M. Lo´pez-Andrade Department of Health Science, University of Jae´n, Paraje de las Lagunillas s/n, Edificio A-3, Dependencia 034, 23071 Jaen, Spain
Keywords Chlorhexidine Cellulosic fibers Zeta potential Adsorption Desorption
Chlorhexidine salts, available commercially as chlorhexidine digluconate, are widely used for their antimicrobial properties. Although it is mainly used as antiseptic, CHX is also used as an antimicrobial agent, being one of the most important applications as an oral hygienic agent. Mouth rinses containing Chlorhexidine appear to be the most effective
123
468
chemical agents in plaque control. The high efficiency of CHX is mainly due to its cationic nature. The mechanism of action of CHX is that two symmetrically positioned p-chlorophenyl guanide groups can penetrated through the cellular wall of bacteria and irreversibly disrupt the bacterial membrane, thus killing the microorganism. CHX also, binds to oral mucosa by interacting with the carboxyl groups in the mucin layer of the surface thus avoiding the adsorption of the bacteria (Morton 1996). As well as the possibilities of topical application, it is well known for its effectiveness when it is used as a mouth rinse. Unfortunately, this is not recommended after oral surgery because of bleeding for the operated area. Another problem is that at usual concentrations, CHX produces side-effects such as extrinsic tooth staining. In such situations, the use of sterile, dry gauze covered with controlled concentration of CHX, which could easily be desorbed in oral physiological conditions, would be interesting to control topical applications so as to protect the surgery area from contaminations. Therefore, the aim of this work is to study the adsorption of CHX onto fibrous material in order to obtain nonwovens and gauzes covered with known amounts of the antiseptic compound. Cellulose is still the most commonly used type of fiber in pharmaceutical applications for nonwovens manufacture. Cotton nonwovens are applied as absorbent material for medical or personal care. Important improvements have been made in fiber selection. Mechanical cleaning, bleaching and fiber finishing make cotton nonwovens preferable to synthetic fibers. In the present work, initial experiments carried out using different types of synthetic fibers show that the amount of CHX uptake was very little compared to the amount adsorbed by cellulose fibers. Natural cellulose fibers are always cleaned in order to remove the non-cellulose compounds and to improve the reactivity of the polymer. Bleaching and mercerizing operations cause oxidation of glucose hydroxyl groups to carboxylic acid groups, thus providing cellulose fibrils of anionic reactive groups. Cellulose fibers, natural as well as regenerated, have a crystalline/amorphous microfibrillar structure. Elementary fibrils consist of a succession of crystallites and less ordered intermediate amorphous regions with lateral tie molecules connect laterally adjacent amorphous regions (Klemm et al. 1998). The differences
123
Cellulose (2009) 16:467–479
between natural and different types of regenerated cellulose fibers lie primarily in the size and orientation of crystalline and amorphous phase, size and shape of voids, number of interfibrillar bonds and the nature of impurities (Kra¨ssig et al. 1984, 1992). All these differences are responsible for the distinction in the adsorption character and reactivity in cellulose material. A significant influence on the adsorption properties of fibers is the amount of accessible hydroxyl and carboxyl groups, so as the proportion of amorphous regions where the adsorption processes take place (Stana-Kleinschek et al. 2001), because neither the water nor the aqueous solution of electrolyte, dyes or surfactant can penetrate into the crystalline regions of the fibers. In order to determine the influence of structural differences between cotton and viscose in adsorption of CHX, we have estimated crystallinity index based on FTIR spectroscopy as described by Sang Youn Oh et al. (2005). In the present work, a combined investigation of adsorption and electrokinetic determinations of CHX over natural and regenerated cellulose fibers has been done, with the aim of studying the mechanism of adsorption of CHX molecules onto cellulose. Results of adsorption isotherms have been analyzed considering Langmuir and Freundlich models. Langmuir isotherm Langmuir isotherm describes sorption onto specific homogeneous sites within an adsorbent where all sorption sites are identical and energetically equivalent and there is no interaction between molecules adsorbed on neighbouring sites. Langmuir models predict that adsorption occurs uniformly on the specific sites of the adsorbent, and once an adsorbate occupies a site, no further adsorption can take place at this site. Therefore a saturation value is reached. Intermolecular forces decrease rapidly with distance and so the model predicts monolayer coverage where the adsorbent has a finite capacity for the adsorbate (Lyklema 1995). The Langmuir model equation is: qe ¼
KL Ce 1 þ aL Ce
ð1Þ
where qe is the equilibrium concentration of adsorbate (mg g-1); Ce is the equilibrium adsorbate in the solution (mg dm-3), KL (dm3 g) and aL (dm3 mg-1) are Langmuir constant, being KL adsorption
Cellulose (2009) 16:467–479
469
equilibrium constant and KL/aL is the theoretical monolayer capacity q0. The constants are determined from lineal fit of the equation: Ce 1 aL ¼ þ Ce qe KL KL
ð2Þ
The linearity of Eq. 1 is only respected at low concentrations, where the model follows Henry’s law. Freundlich isotherm The Freundlich isotherm suggests that the ratio of solute adsorbed to the solute concentration is function of the solution. The empirical model was shown to be consistent with exponential distribution of active centres, characteristic of heterogeneous surface (Lyklema 1995). The empirical equation is: qe ¼ KF Ce1=nF
ð3Þ
where KF is the Freundlich constant (dm3 g-1) and 1/nF is the heterogeneity factor. If n [ 1 then the adsorption is favourable. From a linear fit of Eq. 4 these empirical values were obtained. ln qe ¼ ln KF þ
1 ln Ce nF
ð4Þ
The standard free energy of adsorption, DG0 (kJ mol-1), can be obtained from the Langmuir and Freundlich equations using Eq. 5, where K is constant in terms of dm3 mol-1 DG0 ¼ RT ln K
ð5Þ
Zeta potential evolution as function of the concentration of the treatment used provides information about electrostatic interactions between fiber surface and CHX molecules. It can be determined from streaming potential DU determinations, using the Helmholtz–Smoluchowski relationship (Jacobasch et al. 1985): DU:L:g n¼ Dp:Q:e:e0 :R L ¼ Rs vs Q
ð6Þ
where Dp is the pressure difference, g the dynamic viscosity of the electrolyte, (e.e0) the permittivity of the electrolyte, L and Q the length and cross-section
of the capillary, and R the electrical resistance across the plug. Equation 6 is perfectly adequate for most practical systems. The terms L/Q consist of two parameters neither of which can be easily measured. In the Fairbrother y Mastin approach the term L/Q is replaced by Rsvs where Rs is the electrical resistance of the plug when the measurement cell is filled with an electrolyte whose specific conductance vs, is determined with 0.1 N KCl. Finally, we have carried out desorption experiments at room temperature with the aim of elucidating the possibility of transferring the antiseptic compound from the tissue sample to the oral cavity. Optimization of this process will be held in posterior studies.
Experimental methods Materials Two kinds of nonwovens cellulosic fibers were used in the present work: natural cellulose, 100% pure hydrophilic bleached cotton of pharmaceutical applications from INDAS S.A, and regenerated cellulose, rayon viscose commercially named Fibrana supplied by SNIACE S.A, (Spain). Table 1 summarizes the technical characteristics of the viscose fibers used in the present work, which were provided from SNIACE S.A, and the characteristics of cotton fibers obtained from bibliography (Martinez de las Marı´as 1976). For all experiments, fibrous samples were rinsed repeatedly with deionized water, until the conductivity of washing water remained constant and then they were dried in an oven at 323 K.
Table 1 Technical characteristics of viscose and cotton fibers Viscose
Cotton
Title (dtex)
1.7
1.52
Length (mm)
38.0
27.0
Tenacity (wet) (cN/tex)
9.5
24.5
Elongation (wet) (%)
30
7–7.4
Time absorption (s)
B6
B10
Absorption capacity (%)
18
7–11
Invisible waste (%)
0.13
0.5
Ash residue (%)
1.10
1.5
123
470
Chlorhexidine used in the present work is presented as a salt solution, chlorhexidine digluconate with initial concentration of 20% w/v aqueous solution, and was kindly supplied by DENTAID S.A, (Spain), and the identification of the product is Ph. Eur. Chlorhexidine digluconate is designated as 1,10 -hexamethylenebis [5-(p-chlorophenyl)biguanide]di-D-gluconate, and has a molecular formula is C22H30Cl2N102(C6H12O7). The molecular weight is 897.8 g mol-1 and its chemical structure is shown in Scheme 1. Deionized water with conductivity of ca. 1 lS cm-1 was used to prepare the different solutions.
Cellulose (2009) 16:467–479
time to achieve equilibrium as we can assume because absorbance of the sample remains constant. The pH of the experiments is that of CHX aqueous solution and is ranged from 6 to 8. The amount of the antiseptic adsorbed on the fibers at equilibrium, qe, (mg g-1 dried fiber) was estimated from the difference between the initial and final concentration of CHX in solution after sorption equilibrium had been attained. Fiber samples were removed and dried in an oven at 323 K and then concentration of the residual CHX solution was determined with calibration equation from the absorbance of the solutions at a wavelength kmax 234 nm using a Hitachi U-2000 spectrophotometer.
Methods Characterization of fibers samples Pellets of ca. 2 mg of cellulosic samples were mixed with 200 mg of spectroscopic grade KBr. IR spectra (4,000–400 cm-1) were recorded using a FTIR spectrometer, Bruker 27, with a spectral range of 10,000–400 cm-1. Adsorption equilibrium experiments Pellets of 1 g of fibrous samples of cotton and viscose were used in the adsorption and electrokinetic experiments. For this purpose, fibrous samples were first conditioned with 100 mL of chlorhexidine digluconate aqueous solution, in the range of 10-5 M (0.001%)–5.10-3 M (0.5%) at two different temperature, 293 and 323 K in a thermostated waterbath (WNB Memmmert) with constant stirring and a temperature accuracy of ±0.1 K during 48 h, enough
Kinetic of adsorption experiments Pellets of 1 g of fiber were immersed in 250 mL of 0.05 mM, clorhexidine digluconate solution, in a Pyrex conical flask and placed in a thermostated bath at 293 K with constant stirring. Samples were removed each minute during the first hour and after each 5 min during the next 2 h. Determination of the amount of CHX adsorbed was obtained as before. Data from kinetic of adsorption isotherms obtained were fitted to empirical first order model following the equation qt ¼ qe ð1 ekt Þ
ð7Þ
where qt is the amount of CHX adsorbed on the fiber surface at time t, qe is the amount of CHX adsorbed at equilibrium, and k the empirical rate constant. The experimental data in the form ln½1=ð1 qt =qe Þ
ð8Þ
versus t were fitted to obtain k. The half-time of adsorption, t1/2, time required for the fiber to uptake half as much CHX as it will uptake at equilibrium is determined from equation t1=2 ¼ ln 2=k
ð9Þ
Fick’s equation, Eq. 10, may be applied to estimate the approximated diffusion coefficient, D, at the cylindrical walls of the fiber (Peters 1975). qt 4 Dt 1=2 ¼ 1=2 ð10Þ a2 qe p
Scheme 1 Chlorhexidine digluconate molecule
123
where a is the fiber radius that has been taken as 30 nm as proposed by Wa¨gberg and Ha¨gglund in accordance with the data of van de Ven and Alince
Cellulose (2009) 16:467–479
471
(Wa¨gberg and Ha¨gglund 2001; van de Ven and Alince 1997). The apparent diffusion coefficient can 2 be obtained by plotting the values qqet versus t under the assumption that the surface concentration is constant, although this condition is impossible to maintain.
adsorption experiments at 293 K were used for zeta potential measurements. The samples were dried and after f was determined with an EKA, Electrokinetic Analyzer, (Anton Paar, KG), based on streaming potential technique. For f determinations we used 0.001 N of KCl as streaming solution.
Desorption experiments
Results and discussion
Desorption experiments were carried out using the fibrous samples obtained in the adsorption experiments. These samples present known amounts of CHX adsorbed that was estimated in adsorption experiments. The samples were dried, and then immersed in 100 mL of 0.01 M of NaCl solution at a constant temperature of 293 K, for 2 h in a thermostated bath with constant stirring. The amount of CHX desorbed was estimated by measuring the absorbance of the residual solution once equilibrium is attained, (constant absorbance of the liquid), at kmax of 234 nm wavelength.
FTIR analysis
Determination of carboxylic group The COOH content was determined using the following literature method (Davidson 1948). 0.2 g of dry samples fibers were placed in 100 cm3 with 25 cm3 of 0.03% methylene blue (C.I. Basic Blue 9) solution and 25 cm3 of borate buffer solution (pH 8.5) and the samples were stirred during 20 h at room temperature. After, 2.5 cm3 of the residual solution and 5 cm3 of 0.1 M HCl were diluted in a volumetric flask to 50 cm3 with distilled water. Absorbance was measured at a kmax of 667 nm and concentration was calculated from calibration graphs. The value of COOH (mmol g-1) content was determined using Eq. 8 where A is the total amount of free methylene blue (mg), and E is the weight of the dry sample. CCOOH ¼
0:00313ð7:5 AÞ E
ð11Þ
Zeta potential measurements The determination of the evolution of zeta potential of the fibers surface as function of the concentration of chlorhexidine digluconate solution used in the treatment was studied. Fibrous samples used for
The results of FTIR analysis of the two cellulose fibers are shown in Fig. 1 where the spectra of cotton (red), viscose (blue) and the characteristic bands are represented. We can observe similar chemical composition, but with some differences between the two spectra. Comparing viscose with cotton spectra we can observe that the maximum absorbance of hydrogen-bonded OH stretching, at around 4,000–2,995, is shifted to a higher wave number from 3,344 to 3,450 and the absorbance decreases from 0.94 to 0.73 showing that the intermolecular hydrogen bonding is decreased in the regenerated cellulose fibers. The CH stretching mode at 2,900 cm-1 (0.58) is also reduced to lower wave number 2,891 (0.46). The band at 1,647 could be due to carboxylic group although this is quite difficult to identify because the OH-bending mode due to bound water is also observed in this region. This band is more intensive in viscose samples and shift to lower value 1,641 cm-1(0.31) than for cotton 1,647 cm-1(0.14). The absorbance at 1,430 cm-1 decreases whereas that at 898 cm-1 increases. The first is known as a ‘‘crystalline’’ and the second as an ‘‘amorphous’’ adsorption band. The absorbance ratio A1430/A900 is defined as an empirical crystallinity index CI and the decrease in this factor reflects lower crystallinity as suggested by Tyrone (1997). We have estimated the value of CI for the two fibers based on the results presented. Correlation between the structural characteristics of cellulose fibers and the amount of acid groups present in the natural and regenerated fibers has been extensively studied (Fras-Zemljic et al. 2004, 2008). Therefore, we have estimated the amount of acid groups in the fiber with Eq. 11. Results obtained are shown in Table 2. We can observe that CI for cotton is higher than for viscose fibers and that the amount of carboxylic groups is higher in the regenerated
123
560.00
435.67
519.59
666.60 616.44 898.30
1647.86
0.0
2134.56
706.70
1281.90 1236.01
1372.43
1430.97
2900.78
1318.10
1337.18
1164.09
3344.59
1.0 0.8 0.6 0.2
0.4
Absorbance Units
1059.07
Cellulose (2009) 16:467–479
1113.87
472
3500
3000
2500
2000
1500
1000
500
Wavenumber cm-1
Fig. 1 FTIR spectra of cotton (red) and viscose (blue) with bands wave number
fibers. As was described (Fras-Zemljic et al. 2008), the amount of accessible carboxyl groups is lowered by an increase in the degree of crystallinity. Adsorption The results of adsorption equilibrium of CHX on viscose and cotton are shown in the isotherms presented in Figs. 2 and 3, where it is plotted qe, mg of CHX adsorbed per g of dry fiber versus Ce final equilibrium concentration of CHX in solution expressed in mg dm-3. We can observe that temperature favours the adsorption process as the amount of CHX adsorbed increased with temperature in both cases. When comparing results from both figures we observe that adsorption onto viscose is higher than onto cotton fibers, and that the shape of the isotherms is compatible with Langmuir isotherm model. Data of adsorption were fitted to Langmuir and Freundlich isotherm in the form expressed in Eqs. 2 and 4.
Figures 4 and 5 show linear plots of viscose to Langmuir and Freundlich model and Figs. 6 and 7 are similar for cotton fibers. Table 3 summarizes the values of Langmuir and Freundlich parameters for the two fibers. We have considered only linear fits to Freundlich model at high concentration. We can observe that in both cases there is good correlation (R2) to Langmuir model and although throughout the
Table 2 Empirical crystallinity index, CI and amount of acid groups, COOH, of viscose and cotton fibers CI
Acid groups (mmol kg-1)
Viscose
1.6
50
Cotton
2.2
30
123
Fig. 2 Amount of CHX adsorbed on viscose fibers as function equilibrium concentration of chlorhexidine digluconate in the liquid phase at 293 and 323 K
Cellulose (2009) 16:467–479
Fig. 3 Amount of CHX adsorbed on cotton fibers as function equilibrium concentration of chlorhexidine digluconate in the liquid phase at 293 and 323 K
Fig. 4 Linear fit of adsorption data of CHX on viscose presented in Fig. 2 to Langmuir model
range of concentration tested Freundlich fits exhibit very low correlation, in the range of high concentration (initial concentration of 0.5 mM) experimental data fits well to Freundlich isotherm model. Also standard free energy of adsorption, DG0 has been estimated from experimental data. Langmuir data present good correlation in the four cases tested. The value of the parameters KL, aL and q0 are higher for viscose fibers than for cotton. Also -DG0 values, (indicative of the spontaneity of the process) are slightly higher for viscose fibers. From this data we can observe than the process is more favourable when the adsorbent is regenerated cellulose and that a rise in temperature increases adsorption. The first is probably
473
due to higher reactivity and adsorption capacity of regenerated cellulose fibers as was observe by many investigators, (Kra¨ssig 1992; Ribitsch et al. 2001; Stana-Kleinschek et al. 2002) due to the presence of bigger amounts of reactivity groups provided by oxidation treatment. The influence of temperature is positive because, as has been pointed out by other investigators (Perineau and Gaset 1981) temperature favours the dilatation of the fibers pores and then causing the hydrogen bonding between the polymer chains to get weaker so there are more reactive groups of the polymer accessible to interact with the cationic molecule of CHX. Data of Freundlich model correlation at high concentration must be considered. The values of KF are high in the four situations tested but keep higher for viscose fiber; also nF [ 1 as is required for favourable process and DG0 is quite similar for the two fibers, but its value rises with temperature. Considering Langmuir correlation, specific interaction between the cationic groups of CHX molecule and the carboxyl acid groups of cellulose surface could be responsible for adsorption process at low concentration. However, monolayer capacity obtained from the linear fits is lower than the maximum amount of CHX adsorbed deduced from experimental data of Figs. 1 and 2. Also, the shape of the isotherms show that a plateau is not reached in the conditions tested. In our opinion, another kind of non specific interactions must be responsible for adsorption at high concentration when the interaction between CHX and cellulose is not so specific. In our opinion, as is predicted by Freundlich model, multilayer adsorption could take place, possibly by hydrophobic interactions as hydrogen bonding between biguanide group, and p-chlorophenol of CHX molecule with hydroxyl groups of cellulose polymer. However, to confirm this last hypothesis, determination of evolution of free energy components of the fibers, polar electronacceptor c? and electron-donor c- and the apolar Lifshitz-van der Waals cLW, with the concentration of CHX in the treatment should be done in order to obtain information about other interfacial interactions. Such determinations have been done in our laboratory with acrylic fibers and results are presented in previous works (Chibowski et al. 1998; Espinosa-Jime´nez et al. 2002; Gime´nezMartı´n et al. 2007).
123
474
Cellulose (2009) 16:467–479
Fig. 5 Linear fit of adsorption data of CHX on viscose presented in Fig. 2 to Freundlich model
Fig. 6 Linear fit of adsorption data of CHX on cotton presented in Fig. 3 to Langmuir model
Zeta potential results f-potential evolution of the two cellulose fibers, cotton and viscose, with the concentration of chlorhexidine digluconate used in the treatment of the fibers, were determined and the results plotted in Fig. 8. Experimental data were recorded using 1 mM KCl solution as streaming potential solution at a pH 6–7. In these conditions cellulose fibers are negatively charged due
123
to ionization of carboxylic groups, and as Jacobash et al. (1985) found, IEP of different types of cellulose fibers is very similar, approximately 2.7. Besides, previous works show a plateau in f-potential values for pH between 6 and 10, and that in this pH-range, negative zeta potential of cotton and viscose present their maximum value (Stana-Kleinschek et al. 2001; Reischl et al. 2006). f-potential values for untreated viscose and cotton fibers used in this work were approximately -20 and -12 mV, respectively. Similar results were found working with cellulose fibers by Espinosa-Jimenez et al. (2002). In Fig. 8 we can observe that qualitative electrokinetic behaviour of the two fibers tested is similar, although absolute value of f-potential of cotton fibers is higher than it is for viscose ones. As suggested by Ribitsch et al. (StanaKleinschek et al. 2002), the adsorption of water and electrolyte solution causes interfibrillar swelling of the surface layers, but although the size of the active surface is increased, the nature of the dissociable groups should not change. Swelling itself causes a reduction of f-potential because of the shift of the shear plane into the liquid phase and, in our opinion, as viscose shows more hydrophilic character than cotton, zeta potential should be smaller as can be observed in Fig. 8. As the concentration of CHX used in the
Cellulose (2009) 16:467–479
475
Fig. 7 Linear fit of adsorption data of CHX on cotton presented in Fig. 3 to Freundlich model Table 3 Isotherms constants and correlation coefficients from Langmuir and Freundlich isotherms models Langmuir isotherm KL (dm3 mol-1)
Freundlich isotherm 2
0
aL (dm3 mg-1)
q0 (mg g-1)
R
DG (kJ mol-1)
KF (dm3 mol-1)
nF
R2
DG0 (kJ mol-1)
Cotton 293 K
94
0.008
13
0.993
-11.2
2,700
8.3
0.992
-19.5
Cotton 323 K
166
0.013
14.2
0.994
-13.9
2,200
4.5
0.986
-21.0
Viscose293 K
392
0.021
20.8
0.998
-14.7
3,100
4.2
0.960
-19.8
Viscose323 K
308
0.012
28.6
0.991
-15.5
3,400
3.8
0.994
-22.7
treatment of the fibers increases, negative value of f-potential decreases until an IEP is reached. For higher concentration than IEP, positive values of f-potential are reached but ageing values corresponding to viscose fibers are higher than for cotton ones. To explain this behaviour electrostatic interaction between carboxylic group of cellulose and the two cationic groups of CHX must be taken into account. As we observe in Fig. 8, the IEP for viscose fibers takes place at a concentration of CHX used in treatment of 1,014 mg L-1. For this initial concentration, the amount of CHX adsorbed by viscose deduced from data of Fig. 2 is 16 mg kg-1 (17.8 mmol kg-1). The same data obtained with cotton fibers give initial
concentration of 2,690 mg L-1 that corresponds to a concentration of CHX on the fiber of 12 mg kg-1 (13.4 mmol kg-1). The values obtained from determination of the contents of carboxylic groups shown in Table 2 are 50 mmol kg-1 for viscose fibers and 30 mmol kg-1 for cotton ones. From these results it is difficult to determine if at IEP saturation of the fiber is reached. On considering the values of theoretical monolayer capacity, q0, from Langmuir correlation data (at 293 K) which are 20 mmol kg-1 for viscose and 11 mmol kg-1 for cotton it is possible to assume that when IEP is reached monolayer saturation takes place. The relationship between these values and the contents of COOH must be carefully studied in
123
476
Fig. 8 Evolution of zeta potential of viscose and cotton fibers with initial concentration of chlorhexidine digluconate used in adsorption treatment. Streaming solution is 1 mM of KCl
accordance with stoichiometry of the reaction between CHX molecule and COOH terminal groups. It is possible that one molecule of CHX binds with two COOH groups as CHX possesses two cationic groups on its molecule. In this situation, interaction with viscose would be approximately 18 mm kg-1 of CHX to 25 mmol kg-1 of COOH groups which correspond to a 72% degree of interaction, and in the case of cotton fibers 12 mmol kg-1 of CHX to 15 mmol kg-1 of COOH groups that will correspond to 80% degree of interaction. In our opinion these values are too high because accessibility of COOH groups would be between 58 and 60%. When f-potential of the fibers changes its sign, the surface of the fibers becomes positive. The increase in the amount of the cationic molecule adsorbed when the process is hindered by electrostatic repulsion between the positive charged fibers surface and the cationic molecule of CHX, could be due to the hydrophobic interactions, possible hydrogen bonding between biguanide group and p-chlorophenol of CHX with hydroxyl groups of cellulose, and also between the molecules of CHX previously adsorbed in the monolayer of the fiber surface and those molecules in the liquid phase.
Cellulose (2009) 16:467–479
suggests a first-order process and good correlation (R2 = 0.998) is found when experimental data fits to this empirical model. The rate constant can be determined from the Eq. 7. Finally, Fick’s equation, Eq. 10 was applied to make an estimation of diffusion coefficient D, of CHX through the fibrous plug. In this equation a is the fiber radius that has been assumed as 30 nm as proposed by Wa¨gberg and Ha¨gglund (2001) in accordance with the data of van de Ven and Alince. The results obtained are presented in Table 4. It can be observed that the process is very fast. Values of half-time obtained from the linear fit of Eq. 10, and the values of empirical rate constant are quite high when compared with values found in our laboratory whilst working with other acrylic fibers and cationic surfactants and dyes (Espinosa-Jime´nez et al. 1997; Chibobowski et al. 1998; Gime´nezMartin and Espinosa-Jime´nez 2005). In order to consider the influence of diffusion process in this situation, that is, low concentration of CHX and room temperature, we have estimated the empirical diffusion coefficient. Values obtained are similar to those obtained by Roger and colleagues with lyocell fibers (Ibbett et al. 2008) which were D = 6–15. 10-12 m2 s-1. Also, Wa¨gberg and Ha¨gglund (2001) have investigated differences in saturation adsorption between different polymers of LMW and HMW. They have observed that LMW polymers with molecular ˚ mass of 8,750 g mol-1 and radius of gyration of 86A can penetrate the fiber wall completely. As CHX molecules present a molecular mass of 897.8 g mol-1
Kinetics of adsorption The time-adsorption isotherms of viscose and cotton treated with 0.5 mM of chlorhexidine digluconate at 293 K are plotted in Fig. 9. The shape of the curves
123
Fig. 9 Isotherms of kinetic of adsorption of 0.05 mM of chlorhexidine digluconate on viscose and cotton at 293 K
Cellulose (2009) 16:467–479
477
Table 4 Adsorption rate constant, half-time of adsorption, amount adsorbed at equilibrium and empirical diffusion coefficient for viscose and cotton fibers treated with 0.05 mM of chlorhexidine digluconate at 293 K k 9 103( min
-1
)
t1/2( min )
qe (mg/g)
D 9 10-10 (cm2 s-1)
Viscose
246
2.8
4.7
14.1
Cotton
127
5.4
3.3
10.0
it would be expected that its radius of gyration would ˚ , and then diffusion of CHX be little than 86A molecules into the porous plug could justify the rapid adsorption at least at low concentration.
Results obtained from desorption experiments carried out in 0.01 M NaCl solution at room temperature are presented in Figs. 10 and 11. In these isotherms we have represented the amount of CHX desorbed as function of the concentration of the previous treatment realized on the adsorption experiments and also, as function of the temperature at which experiments have been done. We can notice that equilibrium is reached within 2 h, because concentration of CHX in liquid phase remains constant. We have observed that the amount of CHX desorbed is slightly higher when the adsorption has been done at 293 K onto cotton fibers, where the value of empirical adsorption energy DG0 obtained from Langmuir model (Table 3) is
lower. The fact that the amount of CHX molecules desorbed is low could be due to a release of molecules that were bonded to less energetic sites, possibly by hydrophobic interactions. In our opinion, an increase in salt concentration up to 10 mM of NaCl, could give rise to a decrease in lateral interactions between the adsorbed molecules, and also a decrease in the interaction between the cationic group of CHX and the anionic groups of the fiber surface, thus adsorbed molecules could be released from the fiber surface. It must be indicated that in the present work, desorption has been studied in conditions that could easily be held in oral cavity in order to obtain physiological application of nonwoven or gauzes with CHX. The results obtained, could be indicative of a possible mechanism for a controlled release of CHX in the mouth. However, new experiments in other conditions, such as lower pH and higher salt concentration, must be considered in order to optimize desorption/adsorption amount of CHX ratio.
Fig. 10 Amount of CHX desorbed from viscose fibers at 293 K in 0.01 M of NaCl, versus initial concentration of chlorhexidine digluconate solutions used in adsorption experiments carried out at 293, and 323 K
Fig. 11 Amount of CHX desorbed from cotton fibers at 293 K in 0.01 M of NaCl, versus initial concentration of chlorhexidine digluconate solutions used in adsorption experiments carried out at 293, and 323 K
Desorption experiments
123
478
Cellulose (2009) 16:467–479
Conclusions From experimental results presented in this work the following may be concluded: 1.
2.
3.
4.
5.
Adsorption experiments carried out with digluconate chlorhexidine on natural and regenerated cellulose fibers at different temperature show that these fibers could be used as adsorbent for CHX, the amount of CHX adsorbed on regenerated fibers being higher that on cotton ones. Data from isotherms present good correlation to Langmuir model adsorption but at higher concentration experimental data also fits well to Freundlich adsorption model. From analysis of the evolution of f-potential with the concentration of CHX in treatment together with adsorption isotherms results, we consider that uptake of CHX from cellulose fibers could be due to a combination of electrostatic interactions and same types of hydrophobic interactions. First, at low concentration, the mechanism of adsorption would be governed principally by electrostatic forces between cationic groups of CHX and carboxyl acid groups of cellulose fibers. When f-potential of the fiber is positive, hydrogen bonding between biguanide group and p-chlorophenol of CHX with hydroxyl groups of cellulose, and possibly with CHX molecules previously adsorbed, could be responsible for increasing adsorption and correlation to Freundlich isotherm. To confirm this last hypothesis, in future work, we will realize determination of evolution of free energy components of the fiber as function of the treatment. Desorption experiments show that CHX can be desorbed from these fibers in such physical conditions that can be easily created in oral cavity, that is, NaCl solution in a concentration similar to the concentration of physiological serum, and at room temperature. Optimization of this process is necessary and will be done in future experiments. Using the results presented in this study, we consider the possibility of obtaining cellulose gauzes covered with Chlorhexidine from which this antiseptic compound could be desorbed in physiological conditions.
123
Acknowledgments We express our gratitude to DENTAID S.A for providing us with chlorhexidine digluconate used in this project and all the necessary related information. We express our gratitude to DGICYT (Direccio´n General de Investigacio´n Cientı´fica y Te´cnica), Spain, Project No.FIS2005-06860-C02-02.
References Chibowski E, Espinosa-Jime´nez M, Ontiveros-Ortega A, Gime´nez-Martı´n E (1998) Surface free energy, adsorption and zeta potential in leacril/tannic acid system. Langmuir 14:5237–5244 Davidson GF (1948) The acidic properties of cotton cellulose and derived oxycelluloses. Part II. The absorption of methylene blue. J Text Inst 39:65–86 Espinosa-Jime´nez M, Gime´nez-Martı´n E, Ontiveros Ortega A (1997) Adsorption of N-cetylpyridinium chloride on leacril fibers: kinetics and thermodynamics. Textile res J 67(9):677–683 Espinosa-Jimenez M, Ontiveros-Ortega A, Perea-Carpio R, Gime´nez-Martı´n E (2002) Interfacial chemistry of fabric surfaces, encyclopedia of surface and colloid science. Marcel Dekker Inc, New York, pp 2770–2786 Fras-Zemljic L, Stenius P, Stana-Kleinschek K, Ribitsch V, Dolecek V (2004) Determination of dissociable groups in natural and regenerated cellulose fibers by different tritiation methods. J Appl Polym Sci 92:3186 Fras-Zemljic L, Persin Z, Steinius P, Stana-Kleinsschek K (2008) Carboxyl groups in pre-treated regenerated cellulose fibres. Cellulose 15:681–690 Gime´nez-Martı´n E, Espinosa-Jime´nez M (2005) Influence of tannic acid in Leacril/Rhodamina B system: thermodynamics aspects. Colloid Surfaces A 270–271:93–101 Gime´nez-Martı´n E, Ontiveros-Ortega A, Espinosa-Jime´nez M, Perea-Carpio R (2007) Electrokinetic effect and surface free energy behaviour in adsorption of a reactive dye onto Leacril pretreated with polyethyleneimide ion. J Colloid Interface Sci 311:394–399 Ibbett R, Taylor J, Christian Schuster K, Cox M (2008) Interpretation of relaxation and swelling phenomena in lyocell regenerated cellulosic fibers and textile associated with the uptake of solutions of sodium hydroxide. Cellulose 15:393–406 Jacobasch HJ, Bauo¨sk G, Schurz J (1985) Problems and results of zeta-potential measurements on fibers. Colloid Polym Sci 263:3–24 Klemm D, Philipp B, Heinze T (1998) Comprenhensive cellulose chemistry. Wiley_VCH Verlag, Winheim, pp 9–12 Kra¨ssig HA (1984) Struktur und reaktivita¨t von Cellulosafasesn. Das Papier 38:571 Kra¨ssig HA (1992) Cellulose, structure, accessibility and reactivity. Gordon & Breach, New York Lyklema J (1995) Fundamentals of interface and colloid science, vol I. Academic Press, London Martı´nez de las Marı´as P (1976) Quı´mica y Fı´sica de las fibras textiles, editorial Alhambra, Madrid Morton P (1996) Oral hygiene products and practice, products components: therapeutic agents. Marcel Dekker Inc, New York, pp 219–329
Cellulose (2009) 16:467–479 Perineau F, Gaset A (1981) Etude de l’adsorption de surfactants ioniques et non ioniques sur de la matie`re ve´ge´tale carbonise´e. Can J Chem 59(1):19–26 Peters RH (1975) Textile chemistry, vol III. Elsevier Scientific Publishing Company, New York, pp 149–172 Reishl M, Stana-Kleinsschek K, Ribitsch V (2006) Electrokinetic investigations of oriented cellulose polymers. Macromol Symp 244:31–47 Ribitsch V, Stana-Kleinschek K, Kreze T, Strnad S (2001) The significance of surface charge and structure on the accessibility of cellulose fibres. Macromol Mater Eng 286:648 Sang Youn Oh, Dong Il Yoo, Younsook Shin, Gon Seo (2005) FTIR analysis of cellulose treated with sodium hydroxide and carbon dioxide. Carbohydr Res 340:417–428 Stana-Kleinsschek K, Kreze T, Ribitsch V, Simona S (2001) Reactivity and electrokinetical properties of different
479 types of regenerated cellulose fibres. Colloid Surfaces 195:275–284 Stana-Kleinsschek K, Ribitsch V, Kreze T, Fras L (2002) Determination of the adsorption character of cellulose fibers using surface tension and surface charge. Mat Res Innovat 6:13 Tyrone LV (1997) Textile processing and properties. Elsevier, Amsterdam, pp 63–71 Van de Ven TGM, Alince B, (1997) Porosity of swollen pulp fibers evaluated by polymer adsorption. In: Baker CF (ed) Fundamentals of papermaking transations of the 11th reseacrch symposium held at cambridge, Pira International, Leatherhead, UK, p 771 Wa¨gberg L, Ha¨gglund R (2001) Kinetics and polyelectrolyte adsorption on cellulose fibers. Langmuir 17:1096–1103
123
