A second life for low‐grade wool through formation of all‐keratin composites in cystine reducing calcium chloride–water–ethanol solution

BACKGROUND: Coarse low grade wool holds a share of more than 40% of the worldwide production of 1.2 million tons per year. Wool hair with a diameter above 32.5 μm and recycled wool waste represent an important source of high quality keratin. An efficient and simple shaping procedure to form all‐keratin composites could open a new approach to utilise wool keratin for production of sustainable and biodegradable all‐keratin composite. RESULTS: In this work the dissolution and regeneration of wool keratin was studied using a concentrated solution of calcium chloride–water–ethanol as solvent and thioglycolate as reducing agent to open disulphide bonds. Up to 70% of the wool keratin dissolved in the solvent at pH 7, 60 °C. After dilution with water a share of 80% of the total keratin could be obtained as regenerated composite structure while 20% of the protein remains in solution. Based on the model studies, all‐keratin composites were prepared by impregnation of wool with solvent followed by thermal consolidation in a heated press at 60 °C and 2.2–3.3 bar pressure. CONCLUSIONS: The new method for production of all‐keratin composites permits production of a protein‐based bio‐composite, which opens new applications for low value coarse wool and recycled wool waste. By use of cheap chemicals and thermal consolidation in standard equipment scale‐up of the technology is expected to be straightforward and commercially feasible, leading to a bio‐based and biodegradable composite material. © 2019 The Authors. Journal of Chemical Technology & Biotechnology published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.


INTRODUCTION
The worldwide annual production of wool is in the region of 1.2 million tons. 1,2 Dependent on the breed, the diameter of the wool varies from highly valuable superfine merino wool (< 16 μm) to low value coarse hair (> 32.5 μm). Only limited applications exist for coarse wool, which for example can be hydrolysed and used as a nitrogen-fertiliser. 3 Coarse wool and wool waste however could serve as a keratin source for bio-composite production. 4,5 Besides medical applications, e.g. for tissue engineering, the shaping of keratin-based resources also is of interest to produce non-petrol-based goods from sustainable and biodegradable sources. 6 To achieve wider use of keratin as biobased and biodegradable material appropriate processes for keratin shaping are required. 7 Due to the absence of thermoplasticity the formation of wool composites has to be achieved either by combination with thermoplastic matrix polymers, e.g. polyethylene, polypropylene, celluloseacetate or through dissolution of the protein and following regeneration. [8][9][10] A keratin solvent has to open the hydrogen bonding and the ionic bonding present between adjacent protein chains forming the -keratin structure of wool. 11 In addition reductive cleavage of disulphide bridges through action of suited reducing agents is a condition to achieve dissolution. 12 Representative examples for reducing agents are thioglycolate, mercaptoethanol and sulphites. [13][14][15] In aqueous media dissolution of wool can be achieved at pH 12 in the presence of thioglycolate as reducing agent, precipitation of a regenerate can be initiated by addition of acetic acid. 11,16 Recently highly concentrated salt solutions or ionic liquids in combination with reducing agents have been applied to dissolve keratin for example zinc chloride (ZnCl 2 ), calcium chloride (CaCl 2 ), sodium thiocyanate (NaSCN) and lithium bromide (LiBr) solutions, as well as ionic liquids, e.g. 1-butyl-3-methylimidazolium chloride or concentrated urea solutions. [17][18][19][20][21] During dissolution of keratin by ionic liquids the cystine bridges are opened and partly removed which bears the risk of degradation of the keratin microstructure. 22 Regeneration of ionic liquids previously used for keratin dissolution can be achieved by distillation. 23 A solution of L-cysteine in 8 mol L -1 urea at pH of 10.5 has been demonstrated to dissolve up to 72% of wool keratin, which however is far from being economic due to the costs for L-cysteine. 24,25 In another approach to prepare water soluble keratin, a mixture of 8 mol L -1 urea, 3%wt sodium dodecylsulphate and 3%wt sodium pyrosulphite has been applied for 30 min at 100 ∘ C. 26,27 Also mixtures of urea with mercaptoethanol have been used to dissolve keratin from feathers for preparation of composites with nanoclay-and styrene-based copolymers. 28 The concentrated molten urea solution however has been reported to react with the keratin, thus leading to an increased nitrogen content of the dissolved keratin fractions. 29 Also oxidative sulphitolysis has been proposed as a process to modify keratin during dissolution. 30 A weakness of the keratin solvents mentioned earlier is the undesired chemical modification of keratin during dissolution, e.g. through hydrolysis, formation of Bunte-salts. Keratin degradation occurs as a result of the rather harsh conditions prevalent during the process of dissolution, such as high alkalinity, elevated temperature and solvent reactivity.
In this study we investigated the potential of a concentrated CaCl 2 -water-ethanol solution (1:8:2, CWE) for dissolution of keratin. Due to its low toxicity Ajisawa's reagent has been extensively studied as a solvent for silk fibroin. [31][32][33][34][35][36] For controlled dissolution of wool a modification of the solvent through addition of a reducing chemical will be necessary to achieve reductive cleavage of the cystine bonds in keratin. In this article we report to our knowledge for the first time the use of a synergistic mixture of CaCl 2 -water-ethanol and thioglycolate (CWET) for dissolution of wool keratin and direct formation of an all-keratin composite through one-step thermal consolidation. 37 The dissolution of wool in the solvent was studied as a function of pH, temperature and time. Solvent stability as a function of time was studied by redox potential measurement. Through partial dissolution of wool fibres and regeneration of keratin in the presence of un-dissolved wool fibres a new all-keratin composite was obtained. The composites were characterised by laser scanning microscopy, Fourier-transform infrared (FTIR), moisture sorption, water uptake, ultimate tensile strength and elongation.

Composite preparation
Solvent (CWET) preparation: 0.25 mol, (37.3 g) CaCl 2 ·2H 2 O, 1.52 mol (27.4 g) water, 0.51 mol (23.4 g) ethanol, 2 mL colour indicator (Bromothymol blue solution, 0.1 g per 100 mL ethanol), and 0.06 mol (4 mL) of TGC were mixed at room temperature to form a homogenous solution, then pH was adjusted to pH 7 ± 0.3 by addition of NH 3 (25% wt). Wool sliver (2.0-4.3 g) was placed in a polyethylene plastic bag and a volume of 5 to 20 mL of solvent was added. The sample bag then was placed in another bag to avoid any spillover of solvent and reduce access of air oxygen. To support penetration of the solvent into the wool sliver the sample was squeezed three times by means of a laboratory padder (0.5 m min −1 , 3 bar). Consolidation of the impregnated wool then was achieved in a heated laboratory press (Servitec, Polystat 200 T) for 10 to 20 min at 60 ∘ C at a pressure of 2.2 to 3.3 bar (5 kN force on sample size 150-225 cm 2 ). After pressing the samples were removed from the bag and rinsed in excess water to coagulate dissolved keratin and to remove the solvent. The wet samples then were fixed between an absorbent soft paper and dried at ambient temperature overnight.

Moisture regain, water retention value and tensile properties
To determine moisture regain (MR), samples were milled and stored under normal climate conditions [20 ∘ C, 65% relative humidity (r.h.)] and weighed (m moist ). Then samples were dried at 105 ∘ C for 4 h, cooled down in a desiccator and weighed again (m dry ). The moisture content was calculated according to Eqn (1). Experiments were performed with two repetitions.
For determination of the water retention value (WRV) the samples (0.3 g) were placed in deionised water and shaken overnight. The samples then were collected by filtration and capillary water was removed by centrifugation for 10 min at 4000 × g (5580 U min −1 ; wool fibres at 2000 g, 4000 U min −1 ). The wet samples were weighed (m wet ) and dried at 105 ∘ C for 4 h. WRV was calculated according to Eqn (2). 38 After cooling down in the desiccator the weight of the dried samples was determined (m dry ). Experiments were performed with two repetitions.
www.soci.org C Fitz-Binder, T Pham, T Bechtold with dimension of 4 cm length and 1 cm width. Tensile strength experiments were performed without pretension, with speed of 1 cm min −1 (tests were made at least as double determination). Thickness of samples was measured by means of a sliding calliper. The solution was heated to 40 ∘ C or 60 ∘ C, pH was adjusted to pH 7 ± 0.3 by addition of NH 3 (25% wt) and a mass of 0.5 g of wool was added. The pH value and redox potential were measured in regular intervals for a total duration of 90 min. At the end of the experiment the solution was filtered through a paper filter and regeneration of the dissolved protein was initiated by addition of 100 mL deionised water. The insoluble matter in the filter and the regenerate were collected, rinsed with deionised water, dried at 80 ∘ C and weighed. Experiments were performed with two (40 ∘ C) and three (60 ∘ C) repetitions.

Solubility of keratin in CWET
To study maximum solubility at a temperature of 60 ∘ C, experiments were performed overnight with 0.2 to 0.8 g wool per 10 g of solvent. The effect of dissolution time was studied by treatment of 0.5 g wool in 20 g CWET at pH 7 at a temperature of 60 ∘ C for 1 to 6 h and overnight. Insoluble parts then were removed by filtration of the viscous solution through a PES-screen (23 mesh, open width ca 0.6 mm). The dissolved keratin was precipitated in 100 mL water. The regenerates then were dried in the oven at 80 ∘ C for 1 h.

Determination of protein content in solution
Soluble protein was determined by photometry using Coomassie Blue G-250 staining. To analyse the dissolved protein remaining in solution after regeneration, 0.5 g of keratin solution was diluted with 50 mL of deionised water and filtered to remove any precipitated keratin. A volume of 10 mL filtrate then was diluted with water to 50 mL and 1 mL of this solution was mixed with 1 mL of the Coomassie reagent. Absorbance was measured at 595 nm. Freshly prepared solvent was used to determine the blank value. The Coomassie Blue G-250 reagent solution was prepared by dissolution of 100 mg dye in 50 mL ethanol (95%wt). After addition of 100 mL phosphoric acid (H 3 PO 4 ) (85% w/v) the solution was filtered through a Whatman #1 paper filter. To study time dependence of the dissolution process, a mass of 0.5 g wool was treated in 20 mL of solvent (2.4%wt wool) and samples were analysed for soluble protein content after regular times up to a maximum duration of 3 h. For comparison experiments with addition of TGC after a period of 3 h pre-swelling in CWE solvent were performed. In another series the amount of TGC was increased. Results are given as mean of double determination.

Stability of CWET solvent
Reducing conditions in CWET will initiate the reductive cleavage of the disulphide groups in keratin. An analysis of reducing properties of the solvent in terms of stability of pH and redox potential was required to confirm time stable conditions during the period of thermal consolidation. Potentiometric titrations were undertaken to study the interrelation between pH and reduction potential of TGC in the solvent, and to assess the stability of the solution against air oxygen. Bromothymol blue was added to the solvent as a colour indicator to avoid presence of alkaline conditions and possible hydrolysis of the keratin. The colour of Bromothymol blue changes from yellow to blue in the pH interval of 5.8 to 7.6. The green colour of the indicator thus indicates a solvent pH of 6 to 7.
Titration curves demonstrate the interrelation between redox potential and pH value (Fig. 1). The titration curves shown in Fig. 1(a) indicate the equivalent point of TGC (pK a 3.55) in the pH range of 2.0 to 6.0. 39,40 Thus in the solvent used for the experiments TGC will be present in dissociated form as thioglycolate. Under these experimental conditions the thiol group will be present in un-dissociated form as the pK a of this group has been reported with pK a = 10.40 for TGC. 41 With increasing pH the redox potential in solution decreases ( Fig. 1(b)). At pH 7 a redox potential of −350 mV is observed which is in agreement with results measured in diluted aqueous solution. 42 The slope of the graphs in Fig. 1(b) can be determined from linear approximation with 57.9-58.7 mV pH −1 (r = 0.986-0.994). This corresponds to a reduction reaction according to Eqn (3) (pH < 3) and 4 (pH > 6).
Above pH 7.5 precipitation of Ca-thioglycolate begins, which thus defines the maximum pH for application of the concentrated solution. Due to its high pK a value the thiol group does not dissociate and a reaction according to Eqn (5) can be neglected.
The stability of the solvent was studied by observation of redox potential and pH as function of time during storage at 40 ∘ C and 60 ∘ C and in presence of wool (Fig. S1 in Supporting Information). At a solution temperature of 60 ∘ C a redox potential of less than −550 mV was achieved within the first 10 min of the experiments, which stabilised between −520 mV to −550 mV. The results indicate stable redox potential conditions in solution for at least 90 min at 60 ∘ C. Solution pH remained constant within ±0.1 pH units. Thus under the experimental conditions applied oxidative effects through presence of air oxygen or pH change will be neglectable. wileyonlinelibrary.com/jctb Regeneration Dissolution
By reduction of the cystine groups in keratin (1) through thioglycolate (2) the corresponding cysteine groups (4) are formed as side chains. As a by product dithioglycolate (5) is formed. Formation of mixed disulphides (3) through reaction of thioglycolate with keratin has to be considered as possible reaction leading to soluble products which remain in dissolved state during regeneration.
To determine conditions of maximum solubility of keratin the influence of the amount of wool used and the effect of the dissolution time both were investigated. In the concentration range of 2.0 to 7.4%wt keratin (0.2-0.8 g wool per 10 g solvent) the share of regenerated wool decreases with mass of wool used and accordingly the mass of un-dissolved residue increases (Fig. 3). The decrease in share of regenerated keratin with higher mass of wool also is due to the formation of jellylike components which are difficult to filter before regeneration and thus contribute to the mass of non-dissolved parts.
The reasons for the observed incomplete dissolution of wool are still unclear. The amount of non-regenerable dissolved protein remains almost constant. www.soci.org C Fitz-Binder, T Pham, T Bechtold   Possible hydrolysis of the protein was studied in solution experiments with prolonged dissolution time. An increase in soluble, non-regenerable protein should indicate hydrolytic degradation (Fig. 4).
The mass fraction of non-dissolved wool decreased with dissolution time below 20%wt, however the amount of regenerated keratin did not increase to the same extent. The maximum of regenerated keratin was observed after 3-4 h of dissolution, with a mass fraction of 50 to 55%wt of total wool used. The amount of non-regenerable protein increased with time from initially 16%wt (1 h) to 47%wt during 22 h of treatment, thus indicating possible hydrolytic degradation of keratin.
An alternative procedure to dissolve keratin, bases on the use of concentrated 8 mol L -1 urea solution at pH values above 7. 12,25 In this process thiol-based reducing agents such as mercaptoethanol or L-cystein are required to open the disulphide bonds and to achieve solubility of the keratin. Solutions with a protein content of up to 30%wt could be prepared with these solvents, which is comparable to the findings in this study. However, the treatment in solutions with pH value above 9 also led to some hydrolytic degradation. As a result, a reduction in the molecular weight of the proteins in solution is observed. 25 The use of ionic liquids for keratin dissolution has been studied in preparation of keratin solution of up to 10%wt. As a disadvantage disulphide cleavage of sulphide bondings and losses in sulphur content were observed in dissolution-regeneration experiments using ionic liquids. Thus, chemical changes in the structure of keratin appear during dissolution in ionic liquids. 21,22 The use of an ionic liquid system which could be regenerated by distillation for dissolution of up to 15%wt feather keratin has been reported by Idris et al. 23 Nearly 60% of the dissolved keratin could be regenerated also from ionic liquid solutions by addition of methanol, which is at comparable level to the results achieved in this study. 23 Thus, besides the high dissolution capacity of the CWET system also the low pH value of the solvent is advantageous to prevent hydrolytic degradation of the keratin during dissolution.
Protein staining with Coomassie Blue G-250 was used to determine the concentration of non-regenerable protein remaining in solution after regeneration (Fig. 5). The Coomassie staining indicates a continuous increase in concentration of soluble proteins with time of dissolution, which is in agreement with the results shown in Fig. 4. For comparison also experiments with doubled amount of thioglycolate and with pre-swelling in the solvent before addition of TGC are shown.
The influence of solvent pH was studied by dissolution experiments at different pH values (pH 6.3, pH 7.3 and pH 7.9). The redox potential measured in solution decreased from −245 mV (pH 6.3), to −437 mV (pH 7.3) and −586 mV (pH 7.9). The mass of regenerated keratin decreased from 59.8% (pH 6.3) to 59.4% (pH 7.3) and 21.8% at pH 7.9. Coomassie staining also indicated substantially higher content of dissolved non-regenerable proteins at pH 7.9.
FTIR analysis of the regenerate and the insoluble part was performed to investigate for possible changes in protein structure during dissolution and the presence of mixed disulphides (structure 3 in Fig. 2). In the FTIR spectra however no significant differences between the untreated wool and the regenerates were observed ( Supporting Information, Fig. S2).
wileyonlinelibrary.com/jctb  These results indicate that the use of a solvent with pH 6-7 and a treatment time below 2 h will minimise the risk of hydrolytic degradation of keratin.

Preparation of all keratin composites
When a layer of wool fibres is treated with a limited amount of solvent only parts of the wool will be able to dissolve and will precipitate during regeneration. The major part of the wool fibres will remain un-dissolved and thus form the fibrous matrix. A general scheme of the process is given in Fig. 6.
For preparation of an all-keratin composite a plane layer of wool fibres formed from sliver was impregnated with a limited amount of CWET solvent and then consolidated in a heated press. After pressing, the solvent was washed out with water. During Table 1. Experimental conditions applied for keratin composite production: mass of solvent (calcium chloride-water-ethanol and thioglycolate (CWET)) used per gram of wool, duration of consolidation in press and direction of wool fibres in the composite, thickness and ultimate tensile strength (mean ± standard deviation) (solvent CWET 1.134 ± 0.006 g mL −1 ) CWET regeneration in water dissolved keratin coagulates and serves as a binder between un-dissolved wool fibres, thereby forming an all-keratin composite. In a series of experiments the mass of solvent used in relation to the amount of wool fibres and the conditions of consolidation were varied to elaborate their influence on the physical properties of the composite (Table 1). Composites were prepared with parallel orientation of the wool fibres and also with 90 ∘ crossed fibres. In the photomicrographs obtained by the laser scanning microscopy the composite structure formed by regenerated keratin and un-dissolved wool fibres can be seen. Dependent on the amount of CWET used, the composite structure becomes more or less compact. In Sample 1 the fibrous structure of the wool fibres still dominates. When higher amounts of CWET are used, the mass of regenerated keratin increases and a more compact composite structure is obtained (Sample 6, Fig. 7(c)).
The values for the ultimate tensile strength increase with amount of CWET used and reach the maximum value of 15 MPa at a ratio of 5 to 6 g CWET used per 1 g of wool (Fig. 8). The values for the ultimate tensile strength are considerably below the values of wool hair (150-250 MPa 43 ), which can be explained with the high porosity of the composite structure. A photograph of a representative sample (Sample 8) after tensile strength testing is given in the Supporting Information (Fig.S3).
The ultimate tensile strength of selected samples is shown in Fig. 8. Due to the method of preparation the variation in physical properties of the composite samples still is considerably high, which also results from variations in sample thickness (Supporting Information, Table S1). A mass of 1.6 g of solvent per gram of wool was not sufficient to develop a uniform composite structure (Sample 1, Fig. 7(b)). When a mass of 2.6 to 3.  the composites is at the level of other protein-based composite structures. As an example values between 8 and 20 MPa have been reported for soy protein-based composites. 44 The ultimate tensile strength of keratin-based films from 100% regenerated keratin has been determined with 17 MPa. 18 For composites from silk fibroin and keratin the ultimate tensile strength has been reported to decrease with increasing keratin content to values of 9.1 to 17.7 MPa at a keratin content of 40 to 50%. 45 The relation between mass of solvent used per gram of wool and ultimate tensile strength of the composites is shown in Supporting Information Fig. S4. The level of uniformity in distribution of CWET in the fibre sample also influences the development of a compact composite structure. Higher variations in ultimate tensile strength and elongation were observed in particular at conditions where a high level of solvent penetration was achieved and a dense composite structure was obtained, e.g. Sample 6 ( Fig. 7(c)). For composites which exhibit a high level of ultimate tensile strength minor variations in composite consolidation already will lead to the observed higher variations in strength and elongation.
FTIR-ATR spectra of the keratin composites were performed to search for possible changes in chemical structure of the keratin. No differences in the spectra in terms of appearance of peaks or a significant shift in absorbance could be detected (spectra given in Supporting Information, Fig. S5).
A particular property of keratin-based material is their ability to absorb a high amount of moisture and to swell in water. Thus moisture content (MC) of the composites and WRV as a measure for swelling in water were determined (Fig. 9).
wileyonlinelibrary.com/jctb mass CWET per mass of wool (g/g) The higher accessibility of the porous regenerated keratin structure led to an increase in moisture content (20 ∘ C, 65% r.h.). The WRV mainly is an indicator for accessibility of liquid water, which is stored in highly swollen parts of the material and smaller pores. The experimental results indicate a distinct increase in WRV compared to the reference experiments with wool. For wool fibres a WRV of 37.9 ± 1.1%wt was measured, which increased to 54-70%wt in the keratin composites. This demonstrates the higher porosity of the composite thus leading to a higher capacity to retain water.
When WRV and moisture content are related to the amount of solvent used per gram of wool, a maximum appears when a mass of 4.6 g solvent is used per gram of wool. Use of low amounts of solvent, e.g. 2-3 g solvent per gram of wool led to incomplete composite formation. The more compact structure was obtained with high amounts of solvent used, e.g. when a high amount of 7.3 g CWET was used per 1 g of wool fibres, the WRV reduced to 54%wt.
The CWET solvent studied in this work as well as concentrated urea-based solvents reported in the literature use cheap chemicals systems which can be recycled with standard processes and can be disposed off without problems. 12,24,25 The application of the more expensive ionic liquids as keratin solvent requires careful regeneration and purification, which today still demands for substantial research and development in advance for possible commercialisation.

CONCLUSION
1 Dissolution of wool keratin was achieved in a CaCl 2 -water-ethanol solution after addition of thioglycolate as reducing agent which served as a representative for other thiol-based reducing agents. The combination of the concentrated salt solution with reducing agent permits dissolution of the wool keratin at neutral pH conditions and thus minimises risk of hydrolytic degradation. 2 An all-keratin composite consisting of wool fibres as reinforcement and regenerated keratin as matrix was prepared by direct thermal consolidation of wool fibres and solvent at 60 ∘ C and 2-3 bar pressure. The ultimate tensile strength of the composite depends on the amount of solvent used to consolidate the wool fibres. With the use of 5 g solvent per gram of wool the all-keratin composites exhibited an ultimate tensile strength of 15 MPa, which is comparable to other protein-based composites. A substantial increase in WRV and moisture sorption was observed when compared to wool fibres, which can be explained with the porous structure and higher accessibility of the composite structure. FTIR analysis both of regenerated protein and of composites formed did not indicate significant changes in the protein. 3 The formation of all-keratin composites opens new applications both for rather coarse types of wool from sheep breeding in the European Alps and for use of wool waste from textile production and recycling. The composites are of interest as a new sustainable and biodegradable material with high potential to substitute non-biodegradable petrol-based products.