Reducing waste by adding value to collagenic materials

Summary Research is aimed at the minimisation of industrial waste in the leather industry. This study is concerned with the processing of the chromed collagen residues produced at each of the different stages of leather production. They are as follows: Group I: chromed residuals - tanning and drying stage - crust state Group II: chromed residuals + fatty matter - fatliquoring stage Group III: chromed residuals + fatty matter + retanning agents - retanning stage Group IV: chromed residuals + fatty matter + retanning agents + dyestuff - dyeing stage Group V: chromed residuals + fatty matter + retanning agents + dyestuff + acrylic and/or polyurethanes - finish stage Group V residues represent the highest percentage of the total volume of residues and they are also the most complex. The group includes the trimmings from industrial manufacturers of leather products such as footwear, clothing and upholstery. Chromed collagen residues are those stabilised or crosslinked with chromium (III) salts which block any later treatment to obtain gelatine using conventional methods. A large part of this type of residue is used for the production of agglomerates known as regenerated leather from which diverse articles are manufactured. Introduction In the literature there are descriptions of several processes based on soft alkaline hydrolysis with magnesium and/or calcium oxides giving a dark green precipitate of chromium (III) strongly masked by amino acids and/or short chain peptides. This precipitate can be redissolved with acids and neutralised for re-use in the tanning phase. However, the grain surface of the leather retains dark stains, preventing the production of light coloured articles. The liquid phase is composed of a mixture of amino acids and/or short chain peptides with very low levels of chromium (III), but the high level of hydrolysis limits later use. In contrast, our process is based on the oxidation of chromium (III) to chromium (VI) using peroxides, generally hydrogen peroxide, in an alkaline medium. The collagen fibres of the leather are left with very little residual chromium and, more importantly, with the physical and chemical characteristics of the original fibre practically unaffected. By means of this process, the chromed collagen fibres are left at a stage previous to the tanning phase, remaining in the raw or pickle stage, not stabilised or crosslinked^1,2,3,4. Excess peroxide is completely eliminated from the chromium (VI), then acidified to the dichromate anion. The dichromate is then tightly bound to a column of ion exchange resin. Once the resin is completely saturated, the chromium is reduced to level (III) using an acid medium in the presence of the same peroxide. (This has been patented by our research team.) When the chromium has the same positive charge as that of the matrix of the resin it can be easily eluted giving a highly concentrated solution. By adjusting the pH of this solution it can be subsequently re-used in the tanning process. With this method, the degree of masking of the original collagen chromium (III) complex residue is not important as the complexing capacity is lost in the chromium (VI) oxidation, liberating all the ligands. Contrary to other published methods or processes, tanning with this recovered chromium (III) gives an extremely soft, clear grain usable for a whole range of articles of any colour. This patented technique can also be applied to other fields such as engraving and textile stamping and paper treatments. The treated dechromed collagen residues undergo a second process of highly controlled topochemical hydrolysis. The method of control depends on the degree of interaction of the following variables: * Macerators - alkaline, acid and liotropic materials. * Activators - peroxides (ie hydrogen peroxide), sodium percarbonate and sodium perborate, etc. Figure 1, presents an outline of the whole process, using solid waste in the raw state (crosslink - free), and chromium-containing wastes, as well as all the other solid crosslinked collagenic residues produced in the sequential production steps - fatliquoring, retanning, dyeing, finishing, etc⁵. Factorial design covers the optimum zone for each different type of residue. It is important to stress that during the process, certain amounts of peroxychromates are generated in situ, which are very active and with high oxidation capacity. These compounds contain 2.5 peroxy groups for each chromium atom, and are strongly linked to the ionic groups of the collagen structure. The presence of such peroxychromates during the second hydrolytic process, carried out at a temperature of 70ºC, accelerates the kinetics of the reaction, allowing the production of the byproduct, 'gelatine', with excellent characteristics within a minimum time frame. Within two hours there is a yield of 98% or higher^6,7,8. Depending on the application of the byproduct, eg, photography, cosmetics, pharmacology, veterinary use or medicine. It is necessary to purify the byproduct using ultrafiltration techniques to eliminate the remainder of chromium residues, amino acids and short chain peptides, and also to eliminate salts, increasing the quality of the final product. A pilot plant has been designed for processing all groups of residues; it is able to process quantities within a range of 25-50kg. Enough quantities of the byproduct, of low to medium quality, was obtained for use in the following industrial companies involved with the research: Pielcolor: for partial and/or total substitution of casein to be used in glossy finishes. Tabercolor: for partial and/or total substitution of casein in the production of inks and printing of decorative paper for furniture Incusa: as a pretanning or retanning resin, as a filling agent or as an intensifier for dyeing. In this last application, the byproduct is co-polymerised with acrylic or polyurethane monomers. 2. Statistical treatment of the results 2.1. Experimental design Experiments were carried out according to a Box and Hunter Central Rotatable and Orthogonal Composite Design [a] for four variables. The variables were the following: X₁, temperature [ºC], X₂, time [min], X₃, M of NaOH and X₄, mL of H₂O₂. Variable levels are given in table 1. The coded and uncoded levels of the variables for each experiment as well as the values for bloom degree of the gelatines are given in Table 2. All experiments gave residual chromium contents lower than 30 ppm and efficiencies higher than 99%. Therefore both responses were not analysed because they were considered optimum in the complete experimental range. The influence of temperature, time, concentration of NaOH and volume of H₂O₂ on the Bloom degrees was adjusted using the following second-order polynomial function where; Y = b₀ + S_ib_iX_i + S_ijb_ijX_iX_j for _i ≥ _j where b₀ is the independent term directly related to the mean value of the experimental plan, b_i the regression coefficient which explains the influence of the variables in their linear form, and b_ij the regression coefficients that account for the influence of the variables in their quadratic form, which could define optimum zones and possible interactions between them.^9,10,11 The regression equation relating the influence of the variables with the Bloom degree, the determination coefficient R_, the equation F-Snedecor coefficient and the significance level after removing from the model the non significant variables through the application of the Stepwise regression procedure were the following: Bloom= 3444.79-8405.46*NaOH + 12448*NaOH_ -64.18*Temp + 0.4937*Temp_ The two variables explaining the influence of the NaOH concentration on Bloom Degree have P-values lower than 0.01%, indicating that they are significantly different from zero at the 99.9% confidence level. The influence of temperature explained through the term Temp_ have P-values lower than 10%, indicating that they are significantly different from zero at 90% confidence level. Neither Time nor H₂O₂ showed significant differences from zero, that is to say that they do not influence the bloom if they vary through the experimental range tested. The determination coefficient R_ = 74.79%, indicates that the model as fitted explains the 74.79% of the variability in Bloom: the model is highly significant due to the influence of the NaOH concentration on bloom. Figure 2 shows the influence of NaOH concentration and temperature on bloom degrees. The highest Bloom results are obtained when NaOH is 0.1M and 0.175M. The influence of temperature on the bloom is rather small when compared with the influence of NaOH. When the NaOH concentration is 0.25 M, the Bloom degree is practically zero. The polynomial model used to explain the influence of the studied variables on bloom is not the most adequate. The best values of bloom were obtained at 0.175 M NaOH. Based on the best results some exploratory tests using the EVOP technique were planned. 2.2. EVOP design Experiment number 10 (Table 2) gave the best bloom result. Therefore, based on this experiment, a first cycle of an EVOP plan was designed. Taking into account that only NaOH concentrations and temperature had significant influence on bloom, the two-level factorial plan on which the experimental design is based was centred on this experiment with the range of different variables as follows: Central Point: X₁, Temperature = 57.5ºC X₂, time = 52.5 min X₃, NaOH concentration = 0.175M X₄, H₂O₂ amount = 1 mL It was decided to explore variations on bloom due to the NaOH concentration ranging from 0.15 to 0.20M and a temperatures range of 52.5 to 62.5ºC. Neither the H₂O₂ amount nor time significantlyinfluence the bloom results. Therefore, the H₂O₂ amount was held constant at 1mL and the time was modified from 40 to 50 minutes. The experimental conditions and the bloom results are shown in Table 3. Experiment number 37 gave the best results, better than those obtained in the first experimental plan. Therefore, a second EVOP cycle centred on this experiment was designed using the same methodology. The NaOH concentration ranged from 0.125 to 0.175 M, and the temperature from 47.5 and 57.5ºC. H₂O₂ amount was kept constant at 1mL and time varied between 47.5 to 52.5 minutes. Time effect is clouded by those of NaOH and temperature. By the effect of the interaction between NaOH concentration and temperature. The results are shown in Table 4. Experiment number 41 gave the best result and so can be considered the optimum conditions from the point of view of the Bloom degree. 2.3. The relationship between NaOH concentration, temperature and the bloom degree Collating the results of the two EVOP cycles and experiments from the Box and Hunter design conducted at NaOH concentration levels up to 0.175M, the relationship between X (NaOH concentration) and Y (bloom degree) seems to fit an exponential function as follows: θ being the NaOH concentration below which the efficiency of the reaction suddenly drops. a is the initial slope of the curve for values of NaOH concentration very close to θ. It can be considered as an initial conversion factor or initial rate of transformation. β is related to the maximum conversion capacity of the system. The lower the NaOH concentration the higher values of bloom reached. The maximum bloom degree attainable is α/(β.e) at X (NaOH concentration) equal to θ+(1/β). The experiments to explore the influence of the NaOH concentration on Bloom degree are shown in table 5. Through the application of the Non-linear regression procedure [c] equation (1) has been fitted to the experimental data of Table 5, Y being the bloom degree and X the NaOH concentration in mM. First estimators of α, β and θ were obtained by linear regression of the equation transformed by logarithms. The final fitted model was as follows: Y=100.73(X-96.18)*EXP (-0.036(X-96.18)) According to this equation the lowest NaOH concentration to reach high efficiency in the reaction is 96.18 mM, and the concentration at which the maximum theoretically attainable bloom degree of 1,030 will be 123.96 mM. The determination coefficient R_ is 95.53%, which indicates that the model explains the 95.53% variability in bloom degrees due to NaOH concentration. The remaining 4.47% can be attributed to the intrinsic variability of the raw material and the influence of the other variables. The standard deviation of the residuals is 63.03 bloom. This value can be used to construct prediction limits for new observations. 2.4. Inclusion of the effect of the temperature and time on bloom degree Residuals obtained after removing the influence of NaOH concentration were analysed in order to estimate the influence of temperature and time, and their interaction on the bloom degree by the application of the polynomial regression technique. Only the linear effect of temperature on the bloom degree was seen to be slightly significant and therefore it was included in the final model. For this purpose the variable temperature was transformed into the new variable Z as follows: Z = (Temp-65)/7.5, giving the final equation as follows: Y=95.61(X-95.96)*EXP(-0.036(X-95.96))-31.36*Z See figure 3. When the effect of the temperature is included, at 47.5ºC the lowest NaOH concentration to reach high efficiency in the reaction is 95.96 mM, and the concentration at which the maximum theoretically attainable bloom degree of 1,030 is reached will be 123.96 mM. The determination coefficient R_ is now 97.03%, which indicates that the model explains the 97.03% of the variability in bloom degrees due to NaOH concentration and temperature. This contributes to an increase in the explanation level of 1.5%, reducing the unexplained variation to 2.97%. This variation can be attributed to the intrinsic variability of the raw material and the influence of other non-controlled variables. The standard deviation of the residuals has been reduced down to 54.5 bloom degrees. This value can be used to construct prediction limits for new observations. 3. Industrial application Several types of byproducts obtained at the optimum zone according to the response surface given by the factorial experimental design (see figure 3). These were used in three different industries by Pielcolor (leather chemical maker), Tabercolor (paper printing industry); and Incusa (bovine tanner). For the first two, the main goal was to use the byproduct obtained by dechroming chromed-collagenic residues according to the methodology outlined in this work, as a substitute for casein. In both cases, two parameters should be taken into account, the viscosity and the low chromium content of the final solution where the concentration of dry weight byproduct has to be in the range of 20 to 30%. With respect to the viscosity, it has been established that such a solution stored at 4ºC would have to be viscous liquid or fluid, capable of being poured easily from the barrel or container - an average of 1.400-1.600cP. A similar solution of casein behaves in this manner, however, the gelatine byproduct is an entirely different protein - spaghetti-like rods - while the casein is a globular protein. In the majority of gelatines their physical state at this temperature is completely solid. Therefore, some changes in the sodium hydroxide, temperature and time variables were introduced into the factorial experimental design, to produce a major cleavage of the collagen molecule without strong hydrolysis, which would produce amino acids or small chain peptides. Pielcolor are a chemical supplier for the leather industry, dealing with polyurethane, acrylics, and corresponding copolymers for finishing leather. One variety is the glossy finish to produce a specific manufactured article very popular throughout the world. Traditionally, this type of glossy finish is based on casein. Several pilot-plant trials were carried out in this industry to compare the whole set of casein-based products with gelatine-based ones. This has allowed us to come to the following conclusions about the gelatine-based products: * there are no major differences between casein and gelatine application to leather * gelatine possesses excellent binding properties to pigments, dyes, waxes etc * good fastness to dry rub test although less when referring to wet rub test; probably this property can be modified by changing the crosslinking agent * good glossy and transparent effect * similar resistance to flexometer, lastometer * similar resistance to water absorption and to solvents Furthermore, gelatine does not stick during the ironing process such as -110ºC-120ºC at 80-100 Kg/cm₂ pressure and during a period of 20-60 seconds. Tabercolor are a company, which produces a special paper for decorating furniture. During the manufacturing process casein is used in different steps. These can be summarised as: a) a binder in the ink-making process b) during the impregnation process where paper is soaked in a set of various baths which contain the colours for a particular finish. Each bath contains a solution of casein, acting as a strong binder. As at Pielcolor, Tabercolor have run a series of pilot-plant applications using a gelatine-based product instead of casein. The final conclusions are similar to those obtained from Pielcolor. The gelatine has an excellent binding capacity, good glossiness and transparency. The effect is an increase in mechanical rubbing giving good fastness to dry rubbing, and a similar resistance to ink bleeding in relation to temperature, time and atmospheric conditions. Incusa produce a few tons of chrome-collagenic residues annually. They are capable of reusing chrome recovered from shavings and incorporating the byproduct obtained as a pretanning and/or retanning agent. The main target of this project is to substitute the synthetic resins such as acrylic and polyurethane for gelatine byproduct. The results from pilot-plant trials show that gelatine has an important filling effect on leather, similar to the other resins, and increases the leather surface shine after dyeing. However, when gelatine is applied alone, the leather obtained is rather hard. Mixing with a fatty acid or triglyceride can reduce this. Our research indicates the need to copolymerise the byproduct with a long chain alcohol or acid to give a permanent soft feel to the finished leather. Conclusions 1. Complete elimination of the physical volume occupied by the chromed collagenic residues, up to 98% if possible 2. To find a procedure for processing these types of byproduct residues. A dechromation process based on the oxidation with peroxides, in alkaline medium, followed by washing, reduces the chromium content from 50,000ppm in the original collagenic residues to 20-90ppm (based on the dry weight of the collagenic waste) 3. Recovery and reuse of the chromium (III) present in such residues. Separation and fixation of the ion chromate (CrO₄^2-) in a macroporous anion exchange resin, concentration of this ion in the resin, reduction 'in situ' back to chromium (III) with peroxides in acid conditions. Thus giving a straightforward elution to produce a concentrated solution of chromium (III). Finally, adjustment of its basicity level to 33% and addition of the necessary amount of sodium sulfate to give a reusable concentrated solution of chromium (III) to be incorporated into a conventional industrial tanning process 4. Isolation of industrially reusable byproducts. Once the majority of chromium has been separated from the original chromed-collagenic residues. Different types of byproducts such as gelatine can be obtained. Thus a range of the following quality byproducts were produced: Type A: low-medium quality Type B: medium-high quality Type C: high added-value byproduct with which application fields could include the veterinary, medicine, pharmacology, cosmetics fields etc 5. Both chromium (III) and dechromed collagenic residues can be industrially reused. Especially the latter, which can be transformed into raw material for the leather and paper industries. For example the resulting products may be used by Pielcolor and Tabercolor as a substitute for casein and at Incusa as pretanning or retanning agent 6. There is a substantial economical benefit 7. Based on the Box and Hunter design the influence of NaOH, temperature, time and H₂O₂ on the efficiency, cost and the bloom was explored. It was pointed out that only NaOH was a significant influence on the bloom degree and only temperature showed a scarce influence on the bloom degree. 8. All experimental conditions showed excellent results of efficiency and cost 9. Via the two EVOP cycles the best experimental conditions of the bloom degree were identified. By using the non-linear regression technique, the functional relationship between the bloom degree and, NaOH and temperature has been identified. Acknowledgements This research was sponsored by the project 'ppq-2.000-0213-p4-04' from the Spanish plan nacional de i + d + i (2.000-2.003). The programme of technological environmental processes. Awarded by the Spanish Science and Technology Ministry for the growth project gird-2002-00772 - Restorm. The author would like to express my sincerely acknowledgement to the Josefina Cartiel chemical support staff with the development of this work. *(1) - Ecotechnologies Department: IIQAB - C.S.I.C.-Barcelona - Spain (2) - Instituto Químico de Sarriá -I.Q.S.: Barcelona- Spain (3) - University Polytechnic of Lleida - Lleida - Spain (4) -Pielcolor - Barcelona - Spain (5) -Tabercolor - Cardedeu (Barcelona) - Spain (6) -Incusa - Valencia - Spain