Preprint
This is the submitted version of a paper published in Macromolecular materials and engineering (Print).
Citation for the original published paper (version of record): Jedvert, K., Elschner, T., Heinze, T. (2017)
Adsorption Studies of Amino Cellulose on Cellulosics
Macromolecular materials and engineering (Print), 302(7): 1700022 https://doi.org/10.1002/mame.201700022
Access to the published version may require subscription. N.B. When citing this work, cite the original published paper.
Permanent link to this version:
DOI: 10.1002/mame.((insert number))
Communication
Adsorption studies of amino cellulose on cellulosics
aKerstin Jedvert, Thomas Elschner, Thomas Heinze*
–––––––––
Dr. K. Jedvert, Prof. T. Heinze
Centre of Excellence for Polysaccharide Research, Institute of Organic Chemistry and Macromolecular Chemistry, Friedrich Schiller University of Jena, Humboldtstraße 10, D-07743 Jena, Germany
E-mail: thomas.heinze@uni-jena.de
Permanent address for K. Jedvert: Bio-based Fibres, Swerea IVF, P.O. Box 104, SE-431 22 Mölndal, Sweden
Dr. T. Elschner
Laboratory for Characterization and Processing of Polymers, Faculty of Mechanical Engineering, University of Maribor, Smetanova 17, SI-2000 Maribor, Slovenia –––––––––
Adsorption of a typical example of a new class of amino cellulose, namely 6-deoxy-6-(2-aminoethyl)amino cellulose at different pH-values and in the presence of electrolytes, onto cellulose model substrates is studied with surface plasmon resonance (SPR) and quartz crystal microbalance with dissipation (QCM-D). Unexpectedly, adsorption is consistently higher at a higher pH-value of 10, indicating that solubility and interactions between amine moieties and cellulose are more important than electrostatic interactions. The findings are highly relevant for the process to modify material surfaces with amino cellulose in water-based systems as a universal tool for changing the surface properties and chemistry. Potential applications for an antimicrobial all bio-based material could be found, e.g., as medical textiles or in the
biotechnology sector.
1. Introduction
A simple conversion of p-toluenesulfonyl (tosyl) cellulose with di- and oligoamines yields a novel class of cellulose derivatives containing amino moieties.[1] Such biopolymers are of
huge interest in fields of functional coatings and design of advanced materials, owing to the biocompatible environment and the usefulness of the amino groups. Water-soluble 6-deoxy-6-amino cellulose derivatives have shown to possess interesting properties and can, e.g., form thin, transparent films and monolayers that may be used for immobilization of enzymes and antibodies, useful as biosensors.[2, 3] The synthesis allows possibilities to design the molecule, thus, basicity, reactivity, and solubility can be tuned by varying the di- and oligoamines introduced into the cellulose backbone. Furthermore, these amino celluloses have shown to possess significant antimicrobial properties, e.g., against bacteria such as Staphylococcus aureus and Klebsiella pneumonia.[4] Furthermore, nanoparticle formulation of 6-deoxy-6-(ω-aminoalkyl) aminocellulose carbamates have shown strong antimicrobial activity and enhanced biocompatibility.[5] Antimicrobial and bacteriostatic properties of materials with amino-containing polymers grafted onto cellulose fibers or starch have also been described elsewhere.[6]
The aim of this work is to investigate the interaction between new
6-deoxy-6-(2-aminoethyl)amino cellulose derivatives and different cellulose substrates in water-based systems. Surface plasmon resonance (SPR) and quartz crystal microbalance with dissipation (QCM-D) were utilized for cellulose model surfaces, and the influence of varying pH-values and ionic strength were investigated. The long-term goal is to study possibilities for
functionalized, eco-friendly, cellulosic materials with antimicrobial properties, which could find potential applications, e.g., as medical textiles or in the biotechnology sector. The findings may also provide additional insight into the mechanisms behind the adsorption.
2.
Experimental Section
2.1. Amino cellulose
6-deoxy-6-(2-aminoethyl)amino cellulose (DAEAC) with a degree of substitution (DS) of 0.60 was synthesized according to Heinze et al.[1] An aqueous stock solution of 2 wt% of
DAEAC was used.
2.2. Preparation of cellulose model surfaces
The model surfaces (A) were cellulose-modified QCM-crystals (QSX 334) supplied by LOT-Quantum Design GmbH. The cellulose type was microfibrillated cellulose (MFC) with crystalline cellulose I and amorphous regions, according to the data sheet given by the manufacturer and Pääkkö et al.[7] The fibril diameters were 5-6 nm, and fibril aggregates with diameters of 10-20 nm existed. The cellulose was adhered on a SiO2-sensor base, the cellulose
thickness was ca 6 nm and the surface roughness was 3-4 nm (root mean square average). The second type (B) was made by spin-coting of trimethylsilyl cellulose (TMSC) onto plain gold QCM- or SPR crystals and regenerating TMSC into pure cellulose by an acid treatment.[8, 9]
2.3. QCM-D
A QCM-D E4 instrument (Q-Sense AB, Sweden) was used; the measurements were carried out at 21°C in duplicates. The relative resonant frequency (ΔF), and relative dissipation factor (ΔD), were recorded in comparison to the zero values at the beginning of the measurements (during rinsing). The third overtone was used for evaluation of results. Rinsing was performed with MilliQ water, MilliQ water with adjusted pH and/or with addition of 50 mM NaCl, corresponding to the conditions of the respective amino cellulose samples. The concentration of DAEAC was 0.1 wt% in all samples. Adjustment of the pH-level was done with 0.1 M HCl for pH-value 5, and with 0.1 M aqu. NaOH for pH-value 10. The flow rate was 0.1 mL min-1.
A Biacore X100 (GE Healthcare, Sweden) was used. Samples contained 0.1 wt% DAEAC and pH-value and electrolyte addition was done as described in 2.3. Pure MilliQ-water and MilliQ-water with addition of 50 mM NaCl were used as running buffers; and prior to each injection the system was run with MilliQ-water with conditions corresponding to the DAEAC-sample. The samples were injected for 180 sec. at a flow rate of 10 L min-1.
Since the calculated mass from SPR does not include incorporated water, it is possible to combine the QCM- and SPR-data to receive information about dynamic effects such as swelling and flexibility.[10] An approximation of the amount of water in the adsorbed layers was made by using Equation (1) [11]:
H2O(%) = (1 − ΔmSPR
ΔmQCM−d) 100 (1)
2.5. Dynamic Light Scattering (DLS)
DLS-measurements were performed at an ALV Laser CGS3 Goniometer equipped with ALV Avalanche correlator and He-Ne laser (λ=633 nm) at 25°C using the samples for the SPR-measurements. To receive the particle sizes, 10 measurements (each 10 sec.) were performed and CONTIN fit was used.
3. Results and Discussion
3.1. QCM-D
The QCM-system was run with rinsing solutions until stabilization of the baseline and then measurements were started. After five minutes, the samples were introduced and run for 90 min. Rinsing solutions were run again and the crystals were finally rinsed with pure MilliQ-water. It can be seen that the pH-value has a clear influence on the amount of 6-deoxy-6-(2-aminoethyl)amino cellulose (DAEAC) adsorbed for both model surfaces (Figure 1). The
highest adsorption was found at value 10 and the lowest for value 5, (original pH-value was determined to be 9.2). Although some DAEAC was removed during rinsing; it appeared that for all samples, except at pH-value 5 and in the presence of electrolyte, the majority of DAEAC was still attached to the crystals, indicating a mostly irreversible adsorption.
Figure 1. QCM-D results: change in frequency and dissipation after introducing 0.1 wt% DAEAC for 90 min and subsequently rinsing with MilliQ-water or 50 mM NaCl buffer; a) Crystal A, no salt b) Crystal A, with 50 mM NaCl c) Crystal B, no salt d) Crystal B, with 50 mM NaCl.
The amounts of DAEAC adsorbed were generally higher for all samples in the presence of electrolyte (frequency levels between -15 to -35 Hz) compared to the samples without salt (-10 to -20 Hz). The same trend is seen regarding the pH-value, i.e., the highest amount of DAEAC is adsorbed at a pH-value of 10.
The Sauerbrey equation[12] describes how the adsorbed mass is proportional to the change in frequency, and it was used to calculate the thickness of the adsorbed layers, see Table 1.
However, the correlation is only valid for thin, rigid, and evenly distributed monolayers and the change in dissipation has to be small.[13] Thus, by taking the changes in dissipation into account, and using Voigt modeling, it is possible to study the viscoelastic properties by QCM-D.[10] Calculated values from such modeling are also presented in Table 1.
Adsorption of positively charged polyelectrolytes onto negatively charged surfaces is normally governed by electrostatic interactions and is influenced by ionic strength, concentration of adsorbent, pH-value, and the properties of the solid substrate.[14] For
polysaccharides, there are additional factors that may affect the adsorption, such as degree of polymerization (DP), DS, and solubility. The driving force of polymer adsorption is always enhanced if the solubility is decreased. A typical example is the adsorption of cationic starch onto cellulose that is increased in the presence of an electrolyte.[15] Higher adsorption at higher ionic strength has also been shown for cellulose-4-[N-methylamino]butyrate
hydrochloride (CMABC) on thin films of cellulose.[16] Similar to our results, the adsorption of CMABC was lower at a lower pH-value. These unexpected results were discussed on the protonation/deprotonation of the amine group of CMABC due to changes in pH-value, which influenced the solubility to a large extent. Hence, the changes in conformation at higher pH-values lead to a reduced solubility and thus an increased adsorption, which could also be the explanation for the behavior found in the present study. For DAEAC, the primary amines protonate around pH-value 9 while the secondary amines protonate around pH-value 4.5[17]. This means that at pH-value 5; the polymer persistence length is high due to electrostatic repulsion, while at pH-value 10; the polymer is coiled and more polymer can be adsorbed. Additionally, the secondary amino groups could be responsible for binding with their electron pairs, and the NH-groups cannot bind when they are protonated. Furthermore, it can be expected that the DS and the substitution pattern of the amino cellulose may also have an influence on the adsorption behavior.
The values for density and viscosity of the films adsorbed are close to water, which is realistic considering that DAEAC forms a soft film, which contains incorporated water. The
thicknesses calculated from the Voigt model are naturally higher compared to values based on the Sauerbrey equation. They are in the range of mono/bi-layer in the case of unadjusted pH-value, no addition of salt, while a higher pH-value results in somewhat thicker films.
Furthermore, it may be concluded that the two types of model cellulose surfaces produce similar results, which was expected due to the similarities in molecular structure for both types of cellulose. However, from the viscosity- and shear results, it can be seen that the films are more rigid and compact for crystal A without salt. This can be explained by lower thickness and higher crystallinity of the model cellulose surface in this case (ca 6 nm
compared to 25 nm for crystal B[18]). Another difference between the crystals was the amount of swelling prior to the measurements, which was higher for the crystal A, most likely due to a more fibrillar type of cellulose. Thus, the equilibrating time prior to the QCM-measurements was longer for these crystals, but could be reduced by pre-swelling.
3.2. SPR
The adsorbed amount in mass per surface area (ΔmSPR) is linearly proportional to the relative
response (ΔRU), according to:
𝛥𝑚𝑆𝑃𝑅 = 𝐶𝑆𝑃𝑅𝛥𝑅𝑈 (2)
CSPR is a proportionality constant and has been calibrated (using a large amount of different
proteins) to 0.065 ng cm-2 for adsorption on flat surfaces.[10] The constant has also been determined for ethyl(hydroxyethyl)cellulose to 0.080 ng cm-2.[11] The relative SPR-responses were recorded and adsorbed mass was calculated using Equation (2) and assuming a constant of 0.07 ng cm-2 (Table 1).
The water contents in the layers of DAEAC adsorbed were reasonably high for all samples; however, the amounts for samples containing NaCl were generally higher (70-80% compared to ca. 60-65%). According to the SPR data, addition of electrolyte resulted in a slight decrease in adsorbed mass, except at pH-value 5, whereas the QCM-D data show an increase in
adsorbed mass for all samples. Thus, the addition of salt results in different compactness of the adsorbed layer. Generally, addition of electrolytes reduces the persistence length so that more of the polymer fits on the surface, which consequently leads to increased adsorption. The charges of the electrolyte screens intra- and intermolecular repulsions, hence, the
conformation of the adsorbed polymer becomes less extended and the polymer forms a thicker layer. Furthermore, addition of incorporated ions in the adsorbed films could increase
adsorbed mass to some extent. Adsorbed films of cellulose derivatives have also previously been shown to contain high amounts of water. For carboxymethyl celluloses on regenerated cellulose surfaces; the adsorbed layers contained 90-95% water.[19] Similar amounts were found for ethyl(hydroxyethyl) cellulose at both hydrophilic and hydrophobic surfaces.[11] The reason for the difference in results for the samples at pH-value 5 with electrolyte could be due to a high amount of desorbed polymer in this case.
3.3. DLS
The results of DLS for the different pH-values indicate that the hydrodynamic radius of DAEAC was highest with ca. 2-9 nm for the samples at pH-value of 5 (without addition of salt), while in the presence of NaCl the hydrodynamic radius was ca. 1-5 nm (Figure 2).
Figure 2. DLS-results: hydrodynamic radii of DAEAC in aqueous solution; a) Unadjusted pH-value (ca 1 nm), b) pH-value 10 (ca 2-8 nm/2-5 nm in presence of salt), c) pH-value 5 (2-9 nm/1-5 nm in presence of salt).
Changes in hydrodynamic radius could be an effect of a change in conformation of the DAEAC, i.e. a reduction in size could be due to a decreased solubility (more coil-like structure).
4. Conclusions
There is a higher adsorption of amino cellulose onto cellulosic materials at a higher pH-value, shown for both types of model surfaces, indicating that electrostatic interactions do not play a
major role. On the contrary, an interaction of the amine moieties with the cellulose is predominating. Moreover, the presence of electrolytes leads to higher adsorption levels, probably due to changes in polymer conformation, which will be studied in detail by light scattering techniques. The adsorbed films are viscoelastic and contain significant amounts of water. The findings are important for the process to design material surfaces with amino cellulose as a universal tool for changing the surface properties and chemistry, e.g., in the field of biotechnology. Additional experiments are planned to study the interaction of the amino cellulose on different substrates in detail, e.g., with surface-enhanced Raman spectroscopy. In another paper, the influence of amino cellulose modification of cellulose fibers on dyeability will be discussed.
Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.
Acknowledgements: The Swedish Research Council Formas is gratefully acknowledged for financial support. Dr. Petra Prokop, (Forschungszentrum für Medizintechnik und
Biotechnologie, Bad Langensalza), is gratefully acknowledged for assistance with SPR-measurements. Mr. Oliver Eckardt (IOMC, FSU, Jena) is gratefully acknowledged for assistance with DLS-measurements.
Received: Month XX, XXXX; Revised: Month XX, XXXX; Published online: DOI: 10.1002/mame.((insert number))
Keywords: adsorption, 6-deoxy-6-(2-aminoethyl)amino cellulose, QCM-D, SPR, pH-dependence
[1] T. Heinze, A. Pfeifer, A. Koschella, J. Schaller, F. Meister, 2016, J. Appl. Polym. Sci., 133, DOI: 10.1002/APP.43987.
[2] T. Heinze, M. Siebert, P. Berlin, A. Koschella, 2016, Macromol. Biosci., 16, 10-42. [3] P. Berlin, D. Klemm, A. Jung, H. Liebegott, R. Rieseler, J. Tiller, 2003, Cellulose, 10, 343-367.
[4] K. Roemhild, C. Wiegand, U-C. Hipler, T. Heinze, 2013, Macromol. Rapid Commun., 34, 1767-1771.
[5] C. Wiegand, M. Nikolajski, U-C. Hipler, T. Heinze, 2015, Macromol. Biosci., 15, 1242-1251.
[6] H. Xiao, Y. Guan, US 2014/0303322 A1, 2014.
[7] M. Pääkkö, M. Ankerfors, H. Kosonen, A. Nykänen, S. Ahola, M. Österberg, J. Ruokolainen, J. Laine, P. T. Larsson, O. Ikkala, T. Lindström, 2007, Biomacromolecules, 8, 1934-1941.
[8] T. Mohan, R. Kargl, A. Doliška, H. M. A. Ehmann, V. Ribitsch, K. Stana-Kleinschek,
2013, Carbohyd. Polym., 93, 191-198.
[9] M. Schaub, G. Wenz, G. Wegner, A. Stein, D. Klemm, 1993, Adv. Mater., 5, 919-922. [10] F. Höök, B. Kasemo, T. Nylander, C. Fant, K. Sott, H. Elwing, 2001, Anal. Chem., 73, 5796-5804.
[11] J. Hedin, J-E. Löfroth, M. Nydén, 2007, Langmuir, 23, 6148-6155. [12] G. Sauerbrey, 1959, Zeitschrift für Physik, 155, 206-222.
[13] B. D. Vogt, E. K. Lin, W-L. Wu, C. C. White, 2004, J. Phys. Chem. B, 108, 12685-12690.
[14] I. Szilagyi, G. Trefalt, A. Tiraferri, P. Maroni, M. Borkovec, 2014, Soft Matter, 10, 2479-2502.
[15] K. S. Kontturi, T. Tammelin, L-S. Johansson, P. Stenius, 2008, Langmuir, 24, 4743-4749.
[16] T. Mohan, C. S. P. Zarth, A. Doliška, R. Kargl, T. Grießer, S. Spirk, T. Heinze, K. Stana-Kleinschek, 2013, Carbohyd. Polym., 92, 1046-1053.
[17] L. Fras Zemljič, D. Čakara, N. Michaelis, T. Heinze, K. Stana Kleinschek, 2011, Cellulose 18, 33-43.
[18] T. Mohan, R. Kargl, A. Doliška, A. Vesel, V. Stefan Köstler Ribitsch, K. Stana-Kleinschek, 2011, J. Colloid Interface Sci., 358, 604–610.
[19] Z. Liu, H. Choi, P. Gatenholm, A. R. Esker, 2011, Langmuir, 27, 8718-8728.
Table 1. Results from QCM-D and SPR-measurements (sensor type A was specifically for QCM-D measurements). Thicknesses, density, viscosity and shear were calculated according to the Sauerbrey equation or Voigt model.
Exper iment no. pH Ty pe of cry stal Salt addi tion (50 mM NaC l) Adso rbed mass [ng cm -2], QC M-D, (Std. dev.) Thick ness [nm], Sauer brey, QCM -D Thic kness [nm] Voigt , QC M-D Den sity [kg m -3], Voi gt, QC M-D Visc osity [kg ms -1], Voig t, QC M-D She ar [Pa ], Voi gt, QC M-D Aver age rel. resp onse, SPR Adso rbed mass [ng cm -2], SPR, (Std. dev.) H 2 O [ % ] 1 Un adj. A No 222 (6.4) 2.22 2.58 112 8 0.00 37 201 000 2 pH 10 A No 250 (0.4) 2.49 2.81 996 0.00 53 422 700 3 pH 5 A No 171 (18) 1.70 - - - - 4 Un adj. A Yes 262 (24) 2.62 5.55 971 0.00 23 114 800 5 pH 10 A Yes 453 (46) 4.53 7.51 972 0.00 22 132 100 6 pH 5 A Yes 162 (0.9) 1.62 3.42 105 3 0.00 17 123 600 7 Un adj. B No 219 (23) 2.19 2.93 105 6 0.00 24 643 90 1298 91 (8.6) 58 .6 8 pH 10 B No 261 (36) 2.61 5.11 100 6 0.00 17 101 300 1459 102 (1.0) 60 .9 9 pH 5 B No 161 (5.6) 1.61 - - - - 823 58 (5.3) 64 .2 10 Un adj. B Yes 355 (37) 3.55 6.43 101 4 0.00 19 171 100 1107 77 (13) 78 .1 11 pH 10 B Yes 409 (24) 4.09 7.94 986 0.00 19 145 400 1253 88 (11) 78 .6 12 pH 5 B Yes 209 (0.2) 2.09 2.83 113 6 0.00 29 349 300 878 61 (0.7) 70 .5
Adsorption of amino cellulose onto cellulose surfaces is investigated and an unexpected pH-dependence is found. The adsorption is consistently higher at a higher pH-value of 10.
The findings indicate that solubility and conformation of the amino celluloses are potentially highly important. Novel amino cellulose derivatives possess amazing properties, e.g. antimicrobial. This work is another step towards advanced, functionalized cellulosic materials.
K. Jedvert, T. Elschner, T. Heinze*
