Reactive cellulose-based thin films – a concept for multifunctional polysaccharide surfaces

Abstract

Reactive coatings of hydroxyethyl cellulose furoate in the form of thin films, suitable for the covalent immobilization of functional molecules, were developed and characterized in this work. The cellulose furoate derivatives were synthesized under homogeneous conditions by esterification of hydroxyethyl cellulose with 2-furoic acid. Reactive platform layers of these furoates were obtained by chemical surface modification of spin coated thin films with N,N′-carbonyldiimidazole. This chemistry allowed the covalent immobilization of functional molecules bearing primary and secondary amines on the films. The degree of substitution of the furoate thin films and their amino functionalized counterparts was determined gravimetrically by a quartz crystal microbalance (QCM-D) and correlated with infrared and X-ray photoelectron spectroscopy and zeta-potential measurements. Scanning electron- and atomic force microscopy showed changes in the morphologies that were influenced by the chemical reactions on the surface. The concept presented can be seen as a versatile method for immobilizing amine-containing (bio-)molecules to polysaccharide surfaces with the furoates having the potential for further reversible cross-linking in Diels–Alder reactions.

---

1 Introduction

Surface functionalization methods are important and common tools in manifold applications ranging from antifouling coatings to biosensors,¹ supports and models for living cells,² light-emitting or photovoltaic devices, batteries, membranes, corrosion protection, and colloid stability.^3–5 In particular polysaccharide (PS) materials and their surface modifications gained importance during previous years, owing to their biocompatible and renewable character, combined with a large number of possible derivatization reactions.⁶ Surface modifications can, besides others, be carried out by chemical reactions on bulk materials or by thin film depositions. They possess the advantage that less chemicals are needed to achieve a significant impact on the functionalities and properties of materials.

For PS, thin film coatings are often obtained by the synthesis of a soluble derivative and subsequent coating on a support using spin coating or layer-by-layer approaches.¹ PS thin film coatings in general can either serve as a model for bulk materials or can subsequently be surface modified by a large number of chemical reactions depending on the targeted application. Surface reactions give accessibility to derivatives that are difficult to produce under homogeneous conditions in solution due to cross-linking, which hampers subsequent processing into thin film coatings. Reactive carbonic acid esters of cellulose for instance form intra- and intermolecular carbonates⁷ in solution limiting the applicability of these interesting derivatives.⁸

Cellulose and its derivatives are also appropriate for post-modification of the hydroxyl groups at positions 2, 3, and 6 of the repeating unit. Tailor-made functional surfaces based on cellulose derivatives and solutions were reviewed recently emphasizing the great potential of these materials.⁹ Examples are regenerated cellulose from trimethylsilyl cellulose^10,11 or solutions of cellulose in N,N-dimethylacetamide/lithium chloride (DMAc/LiCl)¹² or N-methylmorpholine N-oxide/dimethyl sulfoxide (NMMO/DMSO).¹³ Moreover, spin coating of cellulose from aqueous colloidal suspensions was demonstrated successfully.⁹ Applications cover basic surface science, biosensors or low-protein adhesion coatings.¹¹

A step further in these developments is the design of reactive thin film based platforms allowing post-modification after coating the PS derivative on a substrate. This includes reversible cross-linking reactions or immobilization of e.g. biomolecules and especially the covalent binding of primary amines if biomedical, biotechnological, or biosensoric applications are targeted.^14–16 This work therefore aims at combining the advantages of PS derivatisation in solution with the post-modification of thin films. It develops a new approach for joining two functions in one PS film. An organo-soluble cellulose derivative bearing furoate moieties that could be cross-linked reversibly with maleimides leading to self-healing materials^17,18 is synthesized under homogeneous conditions from hydroxyethyl cellulose. The polymer is characterized by NMR- and IR spectroscopy and spin coated in the form of nano-metric thin films on solid substrates. Subsequent reaction of the remaining hydroxyl groups with N,N′-carbonyldiimidazole (CDI), avoiding in this way the negative consequences of cross-linking in solution, then enables the immobilization of primary amines as model substances for biomolecules (peptides, proteins etc.). To monitor the reactions on the thin films, surface sensitive techniques such as a quartz crystal microbalance with dissipation monitoring (QCM-D)^19–21 is used and compared with surface analytical results of attenuated total reflectance infrared spectroscopy (ATR-IR) and X-ray photoelectron spectroscopy. Surface charges are investigated by zeta-potential measurements and changes in morphology are analysed by atomic force microscopy. The whole concept would allow the development of reactive PS based platform layers that serve for the immobilization of primary amines and simultaneously provide a (Retro-) Diels–Alder reactive furoate substituent for further modification.

2 Experimental section

2.1 Materials

2-Hydroxyethyl cellulose (2-HEC) with a viscosity average molar mass of 90 000 g mol⁻¹ was purchased from Sigma-Aldrich and dried at 60 °C in vacuum. Toluene (99.9%, Sigma) was dried over sodium hydroxide before use. Other chemicals and solvents were purchased from Sigma-Aldrich or Acros Organics and were used without further treatment.

Silicon wafers supplied by Silchem (Germany) were cut into 15 mm × 15 mm pieces. Microscope glass slides (25 mm × 75 mm) were obtained from Sigma-Aldrich. Quartz crystal microbalance (QCM-D) sensors with a gold layer (QSX303) were purchased from LOT-Oriel (Germany). Milli-Q water from a Millipore water purification system (MA, USA, resistivity 18.2 MΩ cm, pH 6.8) was used for the experiments.

All polysaccharide derivatives and the respective thin film are numbered according to the notation of the chemical structures in Fig. 1.

Fig. 1 Reaction scheme of the synthesis of hydroxyethyl cellulose furoate in bulk and subsequent post-modification of the thin film obtained by spin coating.

2.2 Syntheses

2.2.1 Synthesis of 2-hydroxyethyl cellulose 2-furoate 2. 2-Furoic acid (6.18 g, 55.1 mmol) was dissolved in 30 mL N-methyl-2-pyrrolidone (NMP) and CDI (8.93 g, 55.1 mmol) was added stepwise. The mixture was allowed to react over night at room temperature under stirring.

2-HEC (1, MS 2.5, 5 g, 18.4 mmol) was dissolved in 75 mL NMP at 90 °C for 2 h under stirring. After cooling to room temperature, the activated carboxylic acid was added and the combined solutions were stirred for 24 h at 60 °C. The reaction mixture was allowed to cool to room temperature and poured into 1.5 L deionized water. Subsequently, the precipitate was filtered off, washed three times with 500 mL water, and dried in vacuum at room temperature.

Yield: 93%, DS_furoate 1.72 (determined by ¹H NMR after peracetylation²²), FT-IR (ATR): 1730 cm⁻¹ (ν_C=O), soluble in DMSO, DMA, DMF, CHCl₃, THF, 1,4-dioxane, insoluble in toluene, heptane, acetone. (Solubility was determined by suspending 20 mg product in 0.5 mL solvent at room temperature or 60 °C over night. The product was considered to be soluble, when a clear solution was obtained.) ¹H NMR (250 MHz, DMSO-d₆): δ [ppm] = 7.95 (CH), 7.29 (CH), 6.67 (CH), 4.34, 3.70, 3.49, 3.35. ¹³C NMR (63 MHz, DMSO-d₆): δ [ppm] = 158.3 (C=O), 148.0 (CH), 144.2 (C), 119.0 (CH), 112.7 (CH), 70.2, 68.7, 64.2, 60.7.

2.2.2 Peracetylation. 250 mg of 2-hydroxyethyl cellulose 2-furoate in 5 mL pyridine was allowed to react with 5 mL of acetic anhydride over night at 60 °C. The polymer was precipitated in 250 mL of an aqueous solution of 0.5% sodium hydrogen carbonate and washed four times with deionized water (150 mL). Subsequently, the product was dried at room temperature in vacuum. FT-IR (KBr): no ν_OH.

2.3 Surface modification

2.3.1 Cleaning and coating of support materials. The procedures for cleaning of the substrates were already described in detail by Mohan et al.²³ In brief, silicon wafers, glass slides, and QCM-D crystals were treated with piranha solution and rinsed with Milli-Q water extensively. Subsequently, the support materials were spin coated with 2-hydroxyethyl cellulose 2-furoate from solutions in chloroform applying 4000 rpm (acceleration 2500 rpm s⁻¹) for 60 s.

2.3.2 Activation of hydroxyl groups applying N,N′-carbonyldiimidazole (CDI). The coated support materials were immersed in a solution of 1% CDI in toluene at 60 °C for 4 h and subsequently rinsed with 2-propanol and Milli-Q water. The films (3) were dried with a stream of nitrogen before measurements. FT-IR (ATR): 1730 cm⁻¹ (ν_C=O, furoate), 1773 cm⁻¹ (ν_C=O, 1H-imidazole-1-carboxylate), elemental composition (at% obtained by XPS) for DS_Im 0.91: calculated: C, 63.0; O, 32.1; N, 4.9 found: C, 65.3; O, 29.8; N, 4.9.

2.3.3 Aminolysis of 1H-imidazole-1-carboxylate. The activated films 3 (on solid support) were immersed in triethylene tetramine (TETA, 1% in toluene) at 60 °C for 4 h. Subsequently, the surface 4 was rinsed with 2-propanol and Milli-Q water and dried with a stream of nitrogen. For the blank sample an unmodified film of 2-hydroxyethyl cellulose 2-furoate (2) was treated under equal conditions. FT-IR (ATR): 1730 cm⁻¹ (ν_C=O, furoate, carbamate), elemental composition (at% obtained by XPS) for DS_TETA 0.65: calculated: C, 63.0; O, 30.2; N, 6.8 found: C, 66.0; O, 27.0; N, 6.8.

2.4 Measurements

2.4.1 NMR spectroscopy. NMR spectra were acquired on a Bruker Avance 250 MHz with 32 scans for ¹H NMR spectroscopy and 10 000 scans for ¹³C NMR spectroscopy (27 °C) applying 100 mg sample per mL solvent.

2.4.2 FT-IR spectroscopy. For supporting the complete functionalization of peracetylated 2 the FT-IR spectrum was recorded on a Nicolet AVATAR 370 DTGS spectrometer with the KBr technique. Thin films on QCM-D crystals were measured using a PerkinElmer Spectrum GX Series-73565 FTIR-spectrometer by the ATR technique applying 100 scans.

2.4.3 QCM-D experiments. The gold coated sensor crystals were measured in air with a QCM-D E4 from Q-Sense AB (Gothenburg, Sweden) at 21 °C. The relative resonant frequency (Δf) of the crystals as well as the relative dissipation factor (ΔD) were determined in comparison to the zero values at the beginning of the experiment (third overtone). All adsorption experiments were repeated at least three-times. The quartz crystals were AT-cut quartz with gold plated electrodes and with gold on the active surface. The fundamental frequency of quartz crystals is f₀ ≈ 5 MHz and the sensitivity constant C = 17.7 ng Hz⁻¹ cm⁻². Previous to measurements the sensors we blow-dried in a stream of nitrogen.

2.4.4 Atomic force microscopy. Atomic force microscopy (AFM) images were recorded in tapping mode (non-contact mode) on a Veeco Multimode Quadrax MM AFM (Bruker; Billerica, MA, USA). Scanning silicon cantilevers (NCH-VS1-W from NanoWorld AG, Neuchatel, Switzerland) possessed an average spring constant of 42 N m⁻¹ (force constant) and a resonance frequency of 270 to 320 kHz (coating: none). All measurements were performed at room temperature and under ambient atmosphere. The calculation of the root mean square roughness was done on 5 μm × 5 μm images. The image processing was carried out with the Nanoscope software (V7.30r1sr3; Veeco).

2.4.5 Stylus profilometry. The film thickness was measured using a DEKTAK 150 Stylus Profiler (Veeco). The scan length was set to 1000 μm for a period of 3 s. The diamond stylus possesses a radius of 12.5 μm. The force was adjusted to 3 mg with a resolution of 0.333 μm per sample and a measurement range of 6.5 μm. The profile was set to Hills and Valleys. For the determination of the film thickness each silicon wafer sample was scratched five times with tweezers. The profile was used to calculate the thickness from step height measurement.

2.4.6 X-ray photoelectron spectroscopy. XPS spectra were recorded using a Thermo Scientific instrument equipped with a monochromatic Al K_α X-ray source (1486.6 eV). High resolution scans were acquired at a pass energy of 50 eV and a step size (resolution) of 0.1 eV. Wide scans were acquired with pass energy of 100 eV and a step size of 1.0 eV. All spectra were normalized to the Au 4f7/2 peak. Charge compensation was performed with an argon flood gun. The average chemical composition was calculated from wide scan spectra in two different locations on each surface. The peaks were fitted using a Gaussian/Lorenzian mixed function employing Shirley background correction (Software Thermo Avantage v5.906). All analyses were performed at room temperature.

2.4.7 Contact angle measurements. The surface energies of the films were measured by using Dataphysics contact angle measurement system OCA35 (Dataphysics, Germany) with the sessile drop method and a drop volume of 3 μL. Contact angles of four different solvents, namely water (Erbil), ethylene glycol (van Oss et al.), formamide (van Oss et al.), and diiodomethane (Gonzales-Martin) were evaluated according to the published surface tension parameters applying the Young–Laplace method. The references written in brackets refer to the used software of the contact angle device. All measurements were carried out at room temperature and were performed at least five times and an average value was calculated (Table 1).

Table 1 Contact angles (CA) of water, ethylene glycol, formamide, and diiodomethane on thin films 2–4

Sample	CAwater [°]	CAglycol [°]	CAformamide [°]	CAdiiodomethane [°]
2 (furoate)	76.6 ± 0.2	52.7 ± 1.3	50.4 ± 2.8	41.9 ± 1.5
3 (Im)	68.2 ± 1.0	44.2 ± 1.6	53.1 ± 2.2	34.0 ± 2.3
4 (TETA)	64.7 ± 0.6	35.5 ± 1.3	36.5 ± 1.5	35.2 ± 0.8

---

2.4.8 Zeta potential measurements. The zeta potential was determined on microscope glass slides in a self-made apparatus “Zeta-Oszi” using the principle of oscillating streaming potential.²⁴ The electrolyte was composed of 1 mM KCl solution in water and the pH was adjusted by automatic titration from pH 9 to pH 2. Each measurement was carried out for 60 s recording 600 points and repeated three times at the same pH value.

3 Results and discussion

3.1 Synthesis of hydroxyethyl cellulose furoate

The derivatization of polysaccharides (PS) with furoate moieties is an innovative approach to obtain materials, which can be cross-linked reversibly. Up to now, there are only a few examples for PS-based polymer films¹⁷ or hydrogels¹⁸ possessing furoate groups that undergo (Retro)-Diels–Alder reaction with e.g. bismaleinimides as cross-linker. First synthetic attempts to synthesize hydroxyethyl cellulose furoate were based on the acid chloride, but a low DS value (1.2) was obtained at more than 6-fold excess of reagent.¹⁷ Moreover, the product had to be peracetylated to increase solubility.

Within this work, the mild and chlorine-free esterification applying N,N′-carbonyldiimidazole (CDI) leads to a highly functionalized product (DS 1.72) at low molar ratio (1 : 3, mol repeating unit : mol reagent). As shown in Fig. 1 (1, 2) the reaction was performed homogeneously in NMP at 60 °C over night. The furoic acid was activated with CDI separately in order to avoid side products or cross-linking. The biopolymer derivative obtained is not colored and possesses good solubility in several organic media, e.g. DMSO, chloroform, THF, and dioxane. Thus, this biopolymer derivative may be easily shaped into films, particles, and fibers.

The molecular structure of hydroxyethyl cellulose furoate (2) was shown by NMR spectroscopy. In the ¹H NMR spectrum the signals arising from the furan ring appear at 7.95, 7.29, and 6.67 ppm, which are well separated from the remaining resonances of the repeating unit (Fig. 2, top). Therefore, the DS could be calculated from the integral intensities of furoate- and methyl protons after peracetylation of the sample.²² The structure was evidenced by ¹³C NMR spectroscopy assigning the signals to the corresponding position and the linkage (ester moiety) to the polymer backbone that can be clearly observed due to the carbonyl resonance at 158 ppm (Fig. 2, bottom).

Fig. 2 ¹H NMR spectrum of peracetylated hydroxyethyl cellulose furoate recorded in CDCl₃ (top) and ¹³C NMR spectrum of hydroxyethyl cellulose furoate (2) recorded in DMSO-d₆ (bottom).

3.2 Film formation of hydroxyethyl cellulose furoate

In order to obtain thin films of the cellulose derivative, which are uniform and suitable for post-modification, chloroform solutions were applied for spin coating on silicon wafers. Using 0.5% polymer solution, a smooth and evenly distributed film is obtained. According to profilometry the film thickness is 19.7 ± 0.5 nm.

3.3 Post-modification of films

Considering features of thin films obtained from cellulose derivatives, multifunctionality can be achieved by modification of the remaining hydroxyl groups. Thus, next to furoate moieties that could be cross-linked reversibly by Diels–Alder reaction, 1H-imidazole-1-carboxylate groups could be integrated into the film, which provides a reactive platform for various materials by aminolysis. In a first step, the hydroxyethyl cellulose furoate film (2) was allowed to react with a solution of CDI in toluene (Fig. 1). Toluene was found to be an aprotic organic solvent that does not dissolve the film but CDI, which is a prerequisite for this reaction. In the second reaction, the 1H-imidazole-1-carboxylate (3) was allowed to react with triethylene tetramine (TETA) to yield the corresponding hydroxyethyl cellulose furoate carbamate (4). It should be pointed out that this model amine is an example for an inexhaustible pool of amino compounds that may be covalently bound to the surface. TETA provides a high density of amino groups and a relatively high molecular mass, which is advantageous for monitoring the reaction with QCM-D.

3.3.1 Qualitative and quantitative evaluation. For a quantitative analysis of the surface modification, QCM-D experiments were performed in air. To show the applicability of toluene as reaction medium spin coated sensor crystals were immersed twice in pure solvent under reaction conditions (4 h, 60 °C, Fig. 3, top). The frequency shift (Δf) of about −240 Hz (after spin coating) changes to about −190 Hz. Thus, loosely bound molecules were removed from the surface. However, the mass of the films was constant after the first treatment.

Fig. 3 Stitched QCM-D measurements in air, Δf change in frequency (black), ΔD change in dissipation (gray); top: stability of the film in toluene, center: reaction with TETA after CDI activation, bottom: treatment with TETA without CDI activation.

The mass increase during the first reaction of films with CDI, Δf changes to −221 Hz (Fig. 3, center). Due to the low dissipation (ΔD ≈ 0) the film is rigid and Sauerbrey equation²⁵ is valid. Thus, the DS of 1H-imidazole-1-carboxylate (DS_Im) could be calculated according to the following equation. M₂ is the molecular mass of one repeating unit of 2 (g mol⁻¹) and represents the net molecular mass increase by 1H-imidazole-1-carboxylate (94.07 g mol⁻¹):

The change in frequency from −186 to −221 Hz leads to a DS_Im of 0.87 assuming a homogeneous esterification in the whole film. The activation is efficient and provides a high density of reactive groups.

Moreover, the formation of covalent bonds on the films could be shown by FT-IR ATR spectroscopy of the QCM-D crystals (Fig. 4). In addition to the typical C=O stretching vibration at 1730 cm⁻¹ arising from furoate, a second C=O vibration for the 1H-imidazole-1-carboxylate²⁶ at 1773 cm⁻¹ appears. The integral areas of these peaks allow the prediction of the DS_Im, which was found to be 0.96. This value is in accordance to QCM-D measurements.

Fig. 4 FT-IR ATR spectra of thin films of hydroxyethyl cellulose furoate (2), hydroxyethyl cellulose furoate 1H-imidazole-1-carboxylate (3), and hydroxyethyl cellulose furoate carbamate (4) recorded on QCM-D crystals.

XPS results allow the qualitative and quantitative evaluation of the activated film considering C1s- and N1s scans (Fig. 5). Next to the three peak components arising from hydroxyethyl cellulose furoate at 284.4 eV (C–H), 286.0 eV (C–O), and 288.6 eV (O=C–O) an additional peak at 285.2 eV (C–N) occurs. Furthermore, the imidazolide is visible in the N1s scan; two peaks representing C=N and (C=O)N at 398.9 and 401.0 eV. These values are in accordance with literature data.^27,28 The elemental composition of the film indicates a DS_Im of 0.91 (nitrogen content). Thus, the DS_Im value is 0.9 as proved by three independent techniques (QCM-D, IR, XPS, Table 2).

Fig. 5 Selected XPS results of thin films: C1s scan of hydroxyethyl cellulose furoate (2, top left), C1s scan of hydroxyethyl cellulose furoate 1H-imidazole-1-carboxylate (3, top right), N1s scan of 3 (bottom left), and N1s scan of hydroxyethyl cellulose furoate carbamate (4, bottom right); scans (black), fits (gray), residuals (offset, gray), background (light gray).

Table 2 Showing degree of substitution (DS) by three independent techniques

Sample	DS (QCM)	DS (IR)	DS (XPS)
3 (Im)	0.87 ± 0.06	0.96 ± 0.10	0.91 ± 0.01
4 (TETA)	0.68 ± 0.10	—	0.65 ± 0.03

---

The aminolysis of the activated polymer thin films (3) with TETA was also quantified with QCM-D and XPS. The increase of mass results in a frequency shift to −236 Hz (Fig. 3, center). Using analogous calculations to the previous example, the DS_TETA was determined. is the net molecular mass increase by triethylene tetramine carbamate (172.23 g mol⁻¹):

The change in frequency from −186 to −236 Hz could be interpreted as DS_TETA 0.68. Thus, the conversion of the 1H-imidazole-1-carboxylate into the functional carbamate is about 80%. IR spectroscopy does not show C=O vibration of the imidazolide and thus, hydroxyl groups are regenerated. The formation of covalent bonds could not be proven by FT-IR spectroscopy unambiguously due to the overlapping of signals arising from the ester- (1730 cm⁻¹) and the carbamate moieties. The C=O stretching vibration for cellulose carbamates is usually observed at 1710 cm⁻¹.⁷ However, in the spectrum of the N1s scan (XPS) of sample 4 one peak at 399.5 eV, arising from amine groups is visible. Considering the elemental composition the DS_TETA was found to be 0.65 calculated from the nitrogen content of the XPS measurements.

The treatment of polymer films (2), not CDI-activated, with TETA results in no significant frequency shift (Fig. 3, bottom), i.e. the blank tests show that no significant amounts of amine were adsorbed on the surface. Moreover, no nitrogen could be detected by XPS in the samples.

3.3.2 Change of surface properties. Furthermore, the post-modification of the thin films was detected by zeta-potential measurements using the principle of oscillating streaming potential.²⁴ A blank glass surface and the film of the cellulose derivatives (2, 3) show negative zeta potentials under alkaline and neutral conditions (Fig. 6). The negative zeta potential of these three materials can be caused by three effects. Adsorbed anions from the electrolyte solution, negatively charged carboxyl moieties from oxidized cellulose and dissociated silanol groups in the case of blank glass. The results are in accordance with literature.^29,30 The isoelectric point appears in acid medium due to the protonation of functional groups. However, zeta-potential measurements are appropriate to evidence the aminolysis with TETA. By this reaction amino groups are introduced in the thin film and a covalent carbamate linkage is formed. The obtained film (4) possesses a high density of amino groups resulting in a positive zeta potential, which increases with decreasing pH value that is related to the pronounced presence of ammonium groups in acidic medium. The zeta potentials of film 4 are significantly higher than the values of the blank sample (TETA adsorbed to non-activated film 2). However, a very small amount could be adsorbed to the blank sample due to electrostatic interactions of a few negatively charged carboxylic groups and the amino groups of TETA.

Fig. 6 Zeta-potential (oscillating streaming potential) in dependence on pH value of thin films measured on glass slides: □ piranha-cleaned glass, ▲ hydroxyethyl cellulose furoate (2), ★ hydroxyethyl cellulose furoate 1H-imidazole-1-carboxylate (3), ○ blank sample (2 was treated with TETA), ∇ hydroxyethyl cellulose furoate carbamate (4).

Considering the surface energies of the thin films there are slight changes after performing the surface reactions. All films are hydrophobic cellulose derivatives. The total surface energy of the film is 37 mN m⁻¹ for hydroxyethyl cellulose furoate (2), 39 mN m⁻¹ for the 1H-imidazole-1-carboxylate (3), and 44 mN m⁻¹ for the carbamate (4).

The changes in surface morphology could be evaluated by atomic force microscopy (Fig. 7). The thin film of cellulose ester (2) is smooth and uniform but some small particles stick to the surface. The surface reactions in toluene lead to a porous structure and the particles are removed from the surface (Fig. 7 center, right). In accordance with QCM-D measurements leaching of loosely bound polymer chains take place. Nevertheless, the film thickness increases from 19.7 ± 0.5 nm via 28.8 ± 1.3 nm to 31.9 ± 2.1 nm during the reactions, determined by profilometry. Thus, the film is probably expanded by changing the morphology and bounding of substituents to the polymer chains (increase of mass).

Fig. 7 5 × 5 μm AFM images (height) of thin films of hydroxyethyl cellulose furoate 2 (left, rms 1.28 nm), hydroxyethyl cellulose furoate 1H-imidazole-1-carboxylate 3 (center, rms 1.49 nm), and hydroxyethyl cellulose furoate carbamate 4 (right, rms 1.67 nm).

4 Conclusions

In the present work, hydroxyethyl cellulose furoate with high DS was synthesized by a mild and chlorine-free esterification applying CDI. Thin films of the biopolymer derivative were obtained by spin coating. The furoate moieties may undergo reversible cross-linking by (Retro)-Diels–Alder reaction with bismaleinimides and thus, these films are promising in field of self-healing materials. However, this topic will be a subject of further studies. In order to design advanced materials, multifunctionality was achieved by post-modification of the remaining hydroxyl groups. Beyond cross-linkable furoate moieties 1H-imidazole-1-carboxylate groups could be integrated into the film applying CDI, which provides a reactive platform coating for various biological species by aminolysis. The model amine TETA was studied representing an example for an inexhaustible pool of amino compounds that may be covalently bound to the surface. The degree of substitution (DS) of the thin films was determined by means of QCM-D and proven with IR-spectroscopy as well as XPS. Thus, the surface sensitive technique QCM-D was applied for the determination of the DS of cellulose-based thin films for the first time. The change of the surface properties was measured by means of oscillating streaming potential. SEM and AFM imaging showed changed surface morphologies. In principle, hydrophobic, hydrophilic, charged, or responsive surfaces may be designed according to this concept, which will be investigated in subsequent studies in context with application as functional biomaterials. For example, self-healing films with antifouling properties could be used for coating of medical instruments. Therefore, we aim at developing a method that avoids harmful solvents in order to reduce the toxicity of the material.

Acknowledgements

The financial support of the German Research Foundation (DFG, Research Fellowship EL843/1-1) is gratefully acknowledged. We thank the coworkers of the NMR platform of Friedrich Schiller University Jena.

References

1. C. Zhao, L. Y. Li, M. M. Guo and J. Zheng, Chem. Pap., 2012, 66, 323–339.
2. M. Tanaka, A. P. Wong, F. Rehfeldt, M. Tutus and S. Kaufmann, J. Am. Chem. Soc., 2004, 126, 3257–3260.
3. G. Decher, Science, 1997, 277, 1232–1237.
4. C. S. Peyratout and L. Dahne, Angew. Chem., Int. Ed., 2004, 43, 3762–3783.
5. S. V. Orski, K. H. Fries, S. K. Sontag and J. Locklin, J. Mater. Chem., 2011, 21, 14135–14149.
6. V. Thakur and A. Singha, Surface Modification of Biopolymers, Wiley, 2015.
7. T. Elschner, K. Ganske and T. Heinze, Cellulose, 2013, 20, 339–353.
8. T. Elschner and T. Heinze, Macromol. Biosci., 2015, 15, 735–746.
9. C. Wang, R. A. Venditti and K. Zhang, Appl. Microbiol. Biotechnol., 2015, 99, 5791–5799.
10. M. Schaub, G. Wenz, G. Wegner, A. Stein and D. Klemm, Adv. Mater., 1993, 5, 919–922.
11. T. Mohan, R. Kargl, A. Doliška, A. Vesel, S. Köstler, V. Ribitsch and K. Stana-Kleinschek, J. Colloid Interface Sci., 2011, 358, 604–610.
12. J. Eriksson, M. Malmsten, F. Tiberg, T. H. Callisen, T. Damhus and K. S. Johansen, J. Colloid Interface Sci., 2005, 284, 99–106.
13. S. M. Notley and L. Wagberg, Biomacromolecules, 2005, 6, 1586–1591.
14. T. Elschner, F. Scholz, P. Miethe and T. Heinze, Macromol. Biosci., 2014, 14, 1539–1546.
15. R. Kargl, V. Vorraber, V. Ribitsch, S. Köstler, K. Stana-Kleinschek and T. Mohan, Biosens. Bioelectron., 2015, 68, 437–441.
16. M. I. Montanez, Y. Hed, S. Utsel, J. Ropponen, E. Malmstrom, L. Wagberg, A. Hult and M. Malkoch, Biomacromolecules, 2011, 12, 2114–2125.
17. J. Ax and G. Wenz, Macromol. Chem. Phys., 2012, 213, 182–186.
18. C. M. Nimmo, S. C. Owen and M. S. Shoichet, Biomacromolecules, 2011, 12, 824–830.
19. M. Andersson, J. Andersson, A. Sellborn, M. Berglin, B. Nilsson and H. Elwing, Biosens. Bioelectron., 2005, 21, 79–86.
20. T. Elschner, A. Doliška, M. Bračič, K. Stana-Kleinschek and T. Heinze, Carbohydr. Polym., 2015, 116, 111–116.
21. F. Hook, M. Rodahl, P. Brzezinski and B. Kasemo, Langmuir, 1998, 14, 729–734.
22. S. Hornig, T. Liebert and T. Heinze, Macromol. Biosci., 2007, 7, 297–306.
23. T. Mohan, R. Kargl, A. Doliška, H. M. A. Ehmann, V. Ribitsch and K. Stana-Kleinschek, Carbohydr. Polym., 2013, 93, 191–198.
24. M. Reischl, S. Köstler, G. Kellner, K. Stana-Kleinschek and V. Ribitsch, Rev. Sci. Instrum., 2008, 79, 113902.
25. G. Sauerbrey, Z. Phys., 1959, 155, 206–222.
26. S. Alila, A. M. Ferraria, A. M. B. do Rego and S. Boufi, Carbohydr. Polym., 2009, 77, 553–562.
27. Y. Liu, T. T. Y. Tan, S. J. Yuan and C. Choong, J. Mater. Chem. B, 2013, 1, 157–167.
28. Z. X. Xu, T. Li, Z. M. Zhong, D. S. Zha, S. H. Wu, Q. Liu, W. D. Xiao, X. R. Jiang, X. X. Zhang and J. T. Chen, Biopolymers, 2011, 95, 682–694.
29. T. Ristić, T. Mohan, R. Kargl, S. Hribernik, A. Doliška, K. Stana-Kleinschek and L. Fras, Cellulose, 2014, 21, 2315–2325.
30. C. Werner, U. König, A. Augsburg, C. Arnhold, H. Körber, R. Zimmermann and H. J. Jacobasch, Colloids Surf., A, 1999, 159, 519–529.

---