- Email: [email protected]
Hydrothermolysis of rapeseed cake in subcritical water. Effect of reaction temperature and holding time on product composition
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Available online at www.sciencedirect.com
ScienceDirect http://www.elsevier.com/locate/biombioe
Hydrothermolysis of rapeseed cake in subcritical water. Effect of reaction temperature and holding time on product composition kowska a,*, Paweł Wolak a, Esther Oliveros b Hanna Pin a
Department of Industrial Chemistry, Wrocław University of Economics, ul. Komandorska 118/120, 53-345 Wrocław, Poland b Laboratoire des IMRCP, UMR CNRS 5623, Universite´ Paul Sabatier (Toulouse III), 31062 Toulouse Ce´dex 9, France
article info
abstract
Article history:
We have investigated the hydrothermolysis of rapeseed cake, which is a residual biomass
Received 29 September 2011
with high protein and fat content. The effects of the process parameters, reaction tem-
Received in revised form
perature and holding time, on the composition of the products (amino acids and fatty
7 November 2013
acids) contained in the separated liquid fractions (water and ether, respectively) were
Accepted 14 March 2014
studied. To model the hydrothermal process the experimental design methodology was
Available online 14 April 2014
used. Based on this, optimized operational parameters were determined yielding highest amounts of amino acids and high conversion of triacylglycerols to fatty acids. A maximum
Keywords:
estimated yield of amino acids of 135.9 g kg-1 of rapeseed cake was obtained at 215 C after
Rapeseed cake
36 min. A further increase of both reaction temperature and holding time would lead to the
Subcritical water
Maillard reaction and the decomposition (deamination and decarboxylation) of the amino
Hydrolysis
acids. Maximum conversion of triacylglycerols to fatty acids of 0.91 was predited at 246 C
Amino acids
during 65 min. In this case, a further increase of both reaction parameters would lead to
Fatty acids
enhanced cisetrans isomerization of the unsaturated fatty acids produced.
Experimental design methodology
1.
Introduction
Rapeseed cake is a residual biomass from the pressing of oil from rapeseed. From the processing of rapeseed large amounts of rapeseed cake are released as by-product (about 650 kg from 1 t of rapeseed). Poland is one of the major rapeseed (Brassica napus L.) producers in Europe and the rapeseed growing area is forecast to further increase in the years to come. Rapeseed cake has a high protein and fat content. Traditionally, rapeseed cake has been applied as a livestock
* Corresponding author. Tel./fax: þ48 71 368 0275. kowska). E-mail address: [email protected] (H. Pin http://dx.doi.org/10.1016/j.biombioe.2014.03.028 0961-9534/ª 2014 Elsevier Ltd. All rights reserved.
ª 2014 Elsevier Ltd. All rights reserved.
feed. However, the forecasted increase in rapeseed oil production may cause difficulties in the agricultural application of rapeseed cake. For several years, research into rapeseed cake through thermochemical processing, e.g. pyrolysis yielding bio-oil and biocarbon [1,2], catalytic conversion producing bio-oil [3,4], and co-combustion with hard coal [5], has been conducted worldwide. However, these utilization methods do not allow recovery of the valuable components of the residual biomass, which is possible by using hydrothermal processes conducted in subcritical water. Due to its special properties [6e8], most
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
biomass raw materials may be easily hydrolysed to many valuable feedstocks, among them to amino acids and fatty acids. Depending on the biomass composition the hydrothermal processes of biomass treatment may lead to only amino acids, only fatty acids or mixtures which after separation give both products: aqueous phase containing amino acids and oil phase, rich in fatty acids [9e12]. Recently there was efficiently performed the hydrothermolysis of protein wastes derived from meal processing industry [13e15], seafood industry [16e21], brewery industry [22], silk industry [23] and also other raw materials rich in proteins such as bovine serum albumin [24] and whey protein isolate [25], sewage sludge [19], baby food [26], rice bran [27e30], soybean meal [28], bean dregs [31], microalgae [32,33], raw grass clippings [34] and water lettuce [35]. Several attempts also have been made at the hydrothermolysis of a fat-rich resources: vegetable oils [36e43], squid wastes [20,44], fish meal (horse mackerel) [16] and microalgae [45], which resulted in saturated, monoenic, dienic and polyenic fatty acids, including long chain polyunsaturated acids. In turn the conversion of scallop viscera wastes and squid entrails in a two-step process allowed production of separated fractions: rich in amino acids and fatty acids [43,46]. The hydrothermal methods (under subcritical conditions) have many advantages in relation to traditional methods of amino acids and fatty acids production based on chemical and microbiological syntheses, which are complicated, expensive, require toxic reagents and cause many purification problems. In contrast hydrothermal technologies do not require the addition and recovery of chemicals except for water in consequence allowing valuable materials to be produced with zero emissions [46]. The drying of raw material is not required which makes the process energy saving. Due to hightemperature, high-pressure conditions and the catalytic properties of water arising from its high ionic product under hydrothermal conditions, reactions are faster and may be easily controlled by temperature and pressure adjustment [13,34]. Also lower dielectric constant, viscosity and surface tension of subcritical water enhance the solubilization of treated raw materials and transport of the reagents which accelerate the reaction rates [6,7,13]. Other advantages of hydrothermal processes are limited equipment corrosion problems and simple operation [47]. Considering the above, hydrothermal processes are regarded as environmental friendly, cost effective and having a great potential for practical applications. Since investigations of rapeseed cake hydrothermolysis are rare (to the best of our knowledge no report has been published) herein we propose a hydrothermal path for rapeseed cake treatment yielding a liquid fraction rich in amino acids and fatty acids as an alternative method of its utilization. This study focuses on the effects of the reaction temperature and time of hydrothermolysis of rapeseed cake, as a first step towards a mechanistic interpretation and an evaluation of the feasibility of its technical development. The optimal experimental design (OED) methodology was used to establish a statistically significant reaction model as well as for optimizing the conditions for amino acids and fatty acids production.
2.
Materials and methods
2.1.
Materials and chemicals
51
Rapeseed cake was obtained from the Wilmar Oil Press Plant _ ´ rawina (the Lower Silesian Province, Poland). The moist in Zo material was dried to constant mass at a temperature of 103 C and ground to a grain size of below 1 mm. All solvents and reagents were purchased from SigmaeAldrich and Fluka (amino acids). Depending on the requirements of the particular analytical or investigative method, analytically or HPLCpure reagents were used.
2.2.
Reactor and experimental procedure
The hydrothermolysis of rapeseed cake was performed in a batch reactor (4576A, Parr Instrument Company, USA). A detailed description of the reactor equipment, as well a schematic diagram of experimental set-up (Fig. 1S), is given in Supplementary material. The investigation of hydrothermolysis of rapeseed cake was conduced in two stages. The aim of the first preliminary stage was to establish the range of temperature and holding time of hydrothermolysis of the raw material conversion to amino acids and fatty acids with significant yield. On the basis of these results the optimal experimental design was applied in two experimental series for optimization of amino acids (series 1) and fatty acids (series 2) production, which was the second stage of the study. In preliminary study the hydrothermolysis of the rapeseed cake was conducted at a reaction temperature (T) of 180, 200, 220, 240, 260 and 280 C. The reaction would be stopped once the intended temperature was reached (zero holding time), and after a holding time (t) of 5, 10, 20, 30, 40 and 60 min. In all experiments the pressure corresponded to the vapour pressure curve at the given temperature, or slightly exceeded it (at 180 Ce1.23 MPa, 200 Ce1.85 MPa, 220 Ce2.88 MPa, 240 Ce4.14 MPa, 260 Ce5.52 MPa, 280 Ce7.45 MPa). In the second stage of the study two series of designed experiments were performed in which reaction temperature and holding time were varied simultaneously: in experimental series 1, rapeseed cake hydrothermolysis was performed at temperature ranging from 180 to 240 C and holding time from 0 to 60 min, whereas in series 2, temperature varied from 200 to 260 C and holding time from 0 to 80 min. The experimental conditions are given in Supplementary Material (Table 1S). For each hydrothermolysis experiment, HPLC-pure water was subjected to degasification in an ultrasonic bath and purged with nitrogen. Rapeseed cake (10 g) in water (90 g) suspension was introduced into the reaction vessel, preheated to about 80 C. The reactor was closed and the reaction vessel with its content was purged with nitrogen under a pressure of 2 MPa and heated up at a rate of 10e15 C min1 to a predetermined temperature for 10e15 min, and kept at this temperature with an accuracy of 1 C. The time during which the reaction mixture components were kept at the prescribed temperature was considered as the reaction time (holding time). When the reaction was over, the reaction vessel was
52
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
cooled down to about 90 C for about 5 mine10 min and after the system was decompressed the vessel was emptied and washed with water, to reach the final volume of the aqueous fraction of 150 mL.
2.3. Separation of rapeseed cake hydrothermolysis products The hydrothermolysis of the rapeseed cake yielded a liquid product and a post-reaction solid residue. The water fraction was separated by filtering it through a PTFE membrane filter (Sartorius, SRP 15, 0.45 mm). For the sake of the simplicity and the effectiveness of separation [36,37,43] and requirements of chromatographic analysis of oil fraction components for its separation diethyl ether was used. Diethyl ether (100 mL) was added to the permeate and after one-step extraction the top layer was separated from the bottom layer. The diethyl ether from the top layer was removed by evaporation in a vacuum drier at a temperature of 45 C, whereby the fraction containing ether-soluble components (ES fraction) was obtained and analysed for free fatty acid (FFA) content. The bottom layer containing water-soluble substances (WS fraction) was diluted with water to a volume of 250 mL and amino acids and volatile fatty acids (VFA) content was determined. The yields (Yi) of amino acids or volatile fatty acids contained in the WS fractions were calculated using the following equation (Eq. (1)): Yi g kg
1
¼ mi =mrc
(1)
where mi is the mass (g) of the i reaction product, and mrc is the mass (kg) of rapeseed cake subjected to hydrothermolysis. In the ES fractions the degree of conversion (YFFA) of the triacylglycerols contained in the rapeseed cake, leading to FFA, was calculated from Eq. (2): YFFA ð Þ ¼ AVi =SVrc
(2)
where AVi is the acid value determined in the ES fractions, and SVrc is the saponification value of the rapeseed cake.
2.4.
Analyses and analytical methods
Rapeseed cake dried to constant mass was investigated, except for moisture content. The elementary composition of the rapeseed cake was determined using the Vario EL III apparatus provided by Elementar Analysensysteme GmbH. The content of dry matter and diethyl ether extractable substances in rapeseed cake was determined using The National Forage Testing Association (NFTA) procedures [48], while the content of ash, protein, sulphuric acid soluble lignin and acetyl groups was determined using the National Renewable Energy Laboratory (NREL) protocols [49e51]. Lignocellulosic content (without calcination step) was assayed in the rapeseed cake using the detergent method [48,52,53]. The rapeseed cake was subjected to chromatographic analysis to determine the content and composition of bound [54,55] amino acids. The free amino acids content and composition in the WS fractions obtained in the first preliminary stage of the study was also determined by chromatographic analysis, while the concentration of the free
amino acids in the WS fractions achieved in the second stage of the study, was determined by the ninhydrin method using a procedure according to Watchararuji et al. [28]. The proportion of individual amino acids contained in the WS fractions was depicted as their mass fraction in relation to 100 g of all detected amino acids expressed as: QA (g mA 1001 g1 mSA) where mA stands for individual amino acid, and mSA stands for the sum of all detected amino acids. The contents of fatty acids in the rapeseed cake and in the ES fractions, as well as the contents of formic, acetic, propionic, isobutyric, butyric, isopentanoic and pentanoic acids in the WS fractions were determined by gas chromatography. The proportion of individual fatty acids was depicted as their mass fraction in relation to 100 g of all detected fatty acids (QFFA) expressed as. QFFA (g mFFA 1001 g1 mSFFA) where mFFA stands for individual fatty acid and mSFFA stands for the sum of all detected fatty acids. The operating conditions of all applied chromatographic analyses and ninhydrin method are described in Supplementary material. The acid value of the rapeseed cake (AVrc) and that of the ES fractions (AVi) [56] and the saponification value of the rapeseed cake (SVrc) [57] were determined. All the analytical determinations were performed in triplicate and the mean values were calculated.
2.5.
Modelling and optimization method
The hydrothermolysis of rapeseed cake implies complex reaction manifolds, and the product distribution changes strongly with the variation of the process parameters. The OED methodology [58e60] was used to facilitate the interpretation of experimental results and to determine the process parameters for an optimal yield of amino acids and degree of conversion of triacylglycerols to fatty acids in the WS and ES fractions, respectively. The OED methodology makes use of statistical tools to select a minimum set of experiments adequately located in a chosen experimental region (experimental matrix) and to determine the coefficients of a mathematical model that represents the variations of the experimental response of interest with the best possible precision. In this methodology, the effective values of the process parameters to be varied are coded or normalized, thus enabling the quantitative analysis of the results for different units and ranges of variation of the variables. The relation between normalized and effective variables is given by Eq. (3): xi ¼ ui ui;0
Dui
(3)
where xi is an independent normalized variable and values of xi range from 1 to þ1, ui is an effective variable, ui,0 is the value of the effective variable at the centre of the experimental region (corresponding to xi ¼ 0) and Dui is the step given by Eq. (4): Dui ¼ ui;max ui;min
2
(4)
In this work, the effects of two variables were investigated: reaction temperature (T or u1) and holding time (t or u2).
53
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
From among the possible experimental designs based on the quadratic model, the Doehlert array was selected [58,60]. The values of the experimental response y were fitted to the empirical quadratic polynomial model: y ¼ b0 þ b1 x1 þ b2 x2 þ b11 x21 þ b22 x22 þ b12 x1 x2
(5)
where b0, b1, b2, b11, b22 and b12 are the coefficients of the quadratic model. A Doehlert matrix for 2 variables (T and t) contains 7 uniformly distributed experiments, which, using normalized independent variables xi (Table 1), may be represented by the vertices and the centre of a hexagon. The degree of conversion of the rapeseed cake protein fraction and fatty fraction, defined respectively as the yield of amino acids (y ¼ YA) in the WS fractions, and the degree of conversion of triacylglycerols, leading to fatty acids (y ¼ YFFA) in the ES fractions, was modelled and optimized. Standard deviations sb of experimental responses y were evaluated by repeating three times the experiments at the centres of the experimental domains (xi ¼ 0). The values of coefficients bi of the quadratic model (Eq. (5)) were calculated for dependent variables YA and YFFA. The NEMROD software (Version 2002, LPRAI, France) was used for calculating (by the least-square regression method) the model coefficients and evaluating the significance of the regression and for validity tests.
3.
Results and discussion
3.1.
Composition of rapeseed cake
The rapeseed cake has the molecular formula [C3.9H6.3O2.5N0.4S0.02]n, and the following elementary composition: carbon (46.9%), hydrogen (6.3%), oxygen (40.7%), nitrogen (5.6%) and sulphur (0.5%). The rapeseed cake dry matter content amounted to (mean values and standard deviations, respectively) 905.2 g kg1 3.7 g kg1, ash to 68.8 g kg1 0.6 g kg1, total protein to 349.8 g kg1 2.6 g kg1 (calculated from the total Kjeldahl Nx6.25 [48]), diethyl ether extractable substances represented 106.6 g kg1 7.3 g kg1, hemicellulose 42.2 g kg1 1.4 g kg1, acetyl groups 28.1 g kg1 1.5 g kg1, cellulose 135.3 g kg1 2.1 g kg1, sulphuric acid insoluble lignin 112.9 g kg1 1.4 g kg1 and sulphuric acid soluble lignin 70.0 g kg1 1.9 g kg1. The density of the rapeseed cake is
Table 1 e Doehlert experimental design with normalized (coded) variables (xi). Experiment 1 2 3 4 5 6 7 70 a 700 a a
x1
x2
þ1 1 þ0.5 0.5 þ0.5 0.5 0 0 0
0 0 þ0.866 0.866 0.866 þ0.866 0 0 0
Experiments repeated at the centre of the experimental region to calculate the standard deviation on the response (sY).
0.62 g mL1, and the pH of the 100 g L1 water slurry has a value of 6.21. Table 2 shows the composition of the bound amino acids present in the raw material subjected to hydrothermolysis, whereas the composition of the fatty acids is shown in Table 3. The total bound amino acid content in the rapeseed cake amounts to 317.2 g kg1. Glutamic acid and aspartic acid predominate, followed by leucine, proline, valine, arginine and alanine. In the case of fatty acids, unsaturated acids, particularly oleic acid (C18:1n9) and linolenic acid (C18:2n6) predominate. No trans isomers were found to be present in the rapeseed cake. Its acid and saponification values are 15.3 and 192.5, respectively.
3.2. Effect of reaction temperature and holding time on the yield of amino acids in WS fractions The hydrothermolysis of rapeseed cake was investigated as a batch process. Under the chosen experimental conditions, subcritical water acted as solvent, reagent and catalyst. The protein content in the rapeseed cake was partly hydrolysed and decomposed in subcritical water and converted to watersoluble amino acids. Fig. 1 shows the effect of reaction temperature and time on the yield of amino acids in the WS fractions. At reaction time 0 min, the amino acid content increased with temperature from 180 C to 240 C, due to the increase of the ionic product of water in this temperature range. In the presence of hydronium and hydroxide ions, hydrolysis of the peptide bonds results in the breakdown of the proteins into smaller molecules of soluble proteins or amino acids [25]. The amino acid content also grows with increasing holding time from 0 to 10 min, before reaching a plateau. Similar to our results, Rogalinski et al. [24], Watchararuji et al. [28], Zhu et al. [31], Luo et al. [34,35] confirmed a great influence of reaction temperature and holding time on the hydrothermolysis of used raw materials, but the holding time
Table 2 e Bound amino acid content in rapeseed cake. Amino acids
Bound amino acid content (g kg1)
Bound amino acid mass fraction in relation to 31.7 g (%)
Total Alanine Arginine Cysteine Phenylalanine Glycine Histidine Isoleucine Aspartic acid Glutamic acid Leucine Lysine Methionine Proline Serine Threonine Tryptophan Tyrosine Valine
317.2 17.3 17.9 7.2 16.1 19.7 9.2 13.9 28.3 60.4 26.6 10.0 6.8 20.0 15.7 17.4 2.8 9.4 18.5
100.0 5.5 5.6 2.3 5.1 6.2 2.9 4.4 8.9 19.0 8.4 3.2 2.1 6.3 4.9 5.5 0.9 3.0 5.8
54
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Table 3 e Free fatty acid composition and content in rapeseed cake. Fatty acids Saturated acids (C8:0-C24:0) Monoenic acids C16:1n7 C18:1n7 Monoenic acids C16:1n9 C18:1n9 C20:1n9 Dienic acids C18:2n6 Polyenic acids C18:3n3
QFFA (g mFFA 1001 g1 mSFFA) 13.1 9.8 0.6 9.2 58.9 0.2 57.8 0.9 14.8 14.8 3.4 3.4
had less of an effect on amino acid yield when the temperature was higher than 220 C [28]. However, at higher temperatures (260 C and 280 C), the content of amino acids in the WS fractions decreased with increasing holding time. The decline of the yield of amino acids with increasing holding time at these temperatures is probably due to their degradation, i.e. deamination (resulting in the formation of ammonia) and decarboxylation (resulting in the formation of carbonic acids and amines, e.g., ethylamine and methylamine) [16,18,24,61]. This supposition may be confirmed by results presented in the literature. Rogalinski et al. [24] observed, that above 250 C, the amino acid decomposition rate exceeded the hydrolysis rate of protein and at 330 C there was almost complete degradation of all amino acids to ammonia, ethanoloamine, ornithine and carboxylic acids, such as acetic, propanoic, n-butyric and isobutyric. In other studies [61,62] it was also concluded that the hydrothermal decomposition of alanine and glycine leads to formation of ammonia, methylamine, ethylamine, acetaldehyde, diketopiperazine (dimerization product of glycine) and carboxylic acids, such as acetic, formic, glycolic, lactic, propionic and pyruvic. In our work, the gas phase was not analysed for products (e.g. volatile amines); however the content of some selected VFAs (formic, acetic, propionic, butyric, isobutyric, pentanoic and isopentanoic) as products of
Fig. 1 e Effect of reaction temperature and holding time on the yield of amino acids in WS fractions obtained from hydrothermolysis of rapeseed cake.
deamination of the amino acids [16,19,24,61,63] was determined in the WS fractions. From among the identified VFAs, formic and acetic acids were produced in the largest amounts of 3.3 g kg1 and 1.0 g kg1, respectively. However, these acids may result not only from the deamination of the amino acids resulting from the hydrolysis of the rapeseed cake protein fraction, but also from the decomposition of the cellulose fraction components [19,22,29,64,65]. Formic acid could also be produced by the degradation of the furfurals, whereas acetic acid might be the product of the acetyl cleavage from acetylated sugar monomers of the hemicellulose polymer [66]. In the investigated WS fractions, formic acid was present in variable proportions, and its yield did not show a clear correlation with the primary reaction parameters. The amount of acetic acid did not significantly in the whole investigated range, except that its yield decreased as the holding time was increased at a constant temperature. The yield of the other volatile acids in the WS fractions was very low (0.4e0.9 g kg1), but increased slightly with temperature and holding time. Another reason for the decrease of the amino acid yield in the WS fractions might be the Maillard condensation reaction, by which the amino groups of the amino acids, peptides or proteins, and the carbonyl groups (or the hemiacetal groups) present in oligomer and polymeric carbohydrates would produce melanoidins [67] (dark brown nitrogenous polymers with a nutty odour), whose main thermal degradation products at temperatures ranging from 100 to 300 C are pyrazines, pyridines, pyrroles and furans [68].
3.3.
Profiles of amino acid contents
Fig. 2 shows the profile and proportion of amino acids (QA) in selected WS fractions obtained from the hydrothermolysis of the rapeseed cake. In all WS fractions the predominant amino acid was glutamic acid whose content increased with reaction temperature and holding time. In the WS fractions obtained at 280 C and holding times of 30 and 60 min, the mass fraction of glutamic acid was higher than 50 g 1001 g1. The mass
Fig. 2 e Composition and content of amino acids in WS fractions obtained from hydrothermolysis of rapeseed cake: 1: serine, 2: glutamic acid, 3: glycine, 4: alanine, 5: valine, 6: leucine, 7: other amino acids (aspartic acid, treonine, proline, tyrosine, arginine, isoleucine, phenylalanine, histidine, lysine, cysteine, methionine, tryptophan).
55
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Fig. 3 e Effect of reaction temperature and holding time on the conversion of triacylglycerols to fatty acids (YFFA [ AVi/ SVrc, Eq. (2)) in ES fractions obtained from hydrothermolysis of rapeseed cake.
fraction of the other amino acids contained in the WS fractions would not exceed 10 g 1001 g1. The proportion of the amino acids in the WS fractions was found to be different from that of the bound amino acids in the rapeseed cake. As a result of the changes taking place in the course of the decomposition of the rapeseed cake protein fraction in subcritical water, the structure of the amino acids and their proportion in the WS fractions would change. The changes might be due to the decomposition of the less stable amino acids present in the reaction mixture, depending on the adopted reaction temperature and holding time [46], and to the thermal stability of the other amino acids under these conditions [18,69]. For example, alanine could be formed by transamination of serine with pyruvic acid, and glycine could result from the retroaldol condensation of serine with formaldehyde [61]. The concentrations of these two amino acids in the WS fractions obtained at temperatures from 180 C to 260 C and holding times from 0 to 60 min were in fact higher
than those determined in rapeseed cake, which was in agreement with previous studies presented in the literature [61,62]. A significant decrease of alanine and glycine contents, due to their thermal degradation, was however only observed in the WS fractions obtained at 280 C. Ammonia is the primary thermal degradation product regardless of the amino acid involved. Polar amino acids such as aspartic acid undergo decomposition releasing large amounts of ammonia, in contrast to nonpolar amino acids such as alanine, leucine and valine that remain stable. Amino acids containing more than one nitrogen atom (arginine, histidine, tryptophan) generated higher amounts of ammonia than others. The high content of negatively charged glutamic acid in the WS fractions can be explained by its high thermal stability due the additional methylene group in its side chains [70].
3.4. Effect of reaction temperature and holding time on the yield of fatty acids in ES fractions Fig. 3 shows the effects of the temperature and holding time on the yield of fatty acids in ES fractions formed as a result of the hydrolysis of triacylglycerols contained in the rapeseed cake. At temperatures from 180 to 240 C, the conversion of triacylglycerols to fatty acids that are relatively thermally stable in such conditions would increase with reaction temperature and holding time, reaching a value of 0.85e0.9 in 10e20 min. These results, describing the influence of reaction temperature and holding time on the yield of fatty acids, confirmed the previous results obtained by King et al. [36], Holliday et al. [37], Alenezi et al. [42] and Milliren et al. [43], who successfully hydrolysed soybean [36,37,43], linseed [36], coconut [36] and sunflower [42] oils in subcritical conditions. The increasing ionic product in the temperature range investigated promoted the hydrolysis of triacylglycerols. Also fatty acids may have acted as catalysts promoting the hydrolysis of triacylglycerols as well as of intermediate reaction products, such as di- and monoglycerols [43]. Such a
Table 4 e Profile and content of free fatty acids in selected ES fractions obtained from hydrothermolysis of rapeseed cake. QFFA (g mFFA 1001 g1 mSFFA) Saturated acids (C12:0eC22:0) C16:0 C18:0 Trans isomers C18:1 C18:2 Monoenic acids C16:1n7 C18:1n7 Monoenic acids C16:1n9 C18:1n9 C20:1n9 Dienic acids C18:2n6 Polyenic acids C18:3n3
Reaction temperature ( C)/holding time (min) 180/0
180/30
180/60
240/0
240/30
240/60
280/0
280/30
280/60
13.2 9.6 2.4 1.0 0.8 0.2 6.9 0.6 6.3 65.3 0.0 64.4 0.9 11.8 11.8 1.8 1.8
12.4 9.6 1.9 1.0 0.8 0.2 6.6 0.6 6.0 57.7 0.0 56.8 0.9 17.5 17.5 4.8 4.8
12.6 9.8 1.9 1.1 0.8 0.3 6.6 0.6 6.0 57.5 0.0 56.7 0.8 17.4 17.4 4.8 4.8
12.9 10.0 2.0 1.1 0.8 0.3 7.0 0.6 6.4 57.2 0.0 56.5 0.7 17.1 17.1 4.7 4.7
13.0 9.9 2.0 1.5 1.1 0.4 6.6 0.6 6.0 62.4 0.0 61.6 0.8 13.5 13.5 3.0 3.0
12.7 10.0 1.8 1.7 1.3 0.4 6.1 0.6 5.5 61.3 0.0 60.6 0.7 14.8 14.8 3.4 3.4
12.9 10.1 2.0 2.3 1.6 0.7 5.9 0.5 5.4 60.8 0.0 60.1 0.7 15.1 15.1 3.0 3.0
13.4 10.3 2.3 5.5 4.2 1.3 5.7 0.7 5.0 59.1 0.0 58.5 0.6 14.0 14.0 2.3 2.3
17.0 10.8 2.3 6.6 5.3 1.3 5.4 0.5 4.9 55.8 0.1 55.1 0.6 13.4 13.4 1.8 1.8
56
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
affected by their molecular structure. With the increase of reaction temperature the yield of fatty acids which contained double bonds decreased, polyenic acids demonstrated a slight or quick decrease, and saturated fatty acids reached a plateau.
3.5.
Fig. 4 e Graphical representation in effective variables (u1 [ T and u2 [ t) of experiments performed in series 1 and 2 (numbering of experiments as in Tables 5 and 6, respectively, in parentheses for series 1 and in brackets for series 2).
hypothesis would correspond to a mechanism where, in the first reaction step, fatty acids would dissociate, inducing protonation of the carboxyl oxygen of acyl glycerols [39]. Another advantage of hydrothermolysis in subcritical water is the reduced induction period compared to the traditional fat splitting processes. However, with increased reaction temperature and holding time (260, 280 C and after 20 min) the yield of fatty acids significantly decreased. Even though, the fatty acids are stable in subcritical water [36], a higher reaction temperature [44] and longer holding time of the reaction mixture enhanced degradation and pyrolysis of the triacylglycerols and fatty acids [36,44] and partial depolymerization of fatty acids [40]. Kocsisova´ et al. [40] observed structural changes in the chains of the unsaturated fatty acids and their dimerization during hydrothermolysis of rapeseed oil. Simultaneously, in the work of Tavakoli et al. [44], it was determined that the thermal stability of fatty acids was
Profiles of fatty acid contents
Table 4 shows the profile and proportion of the fatty acids (QFFA) contained in selected ES fractions obtained from hydrothermolysis of rapeseed cake. The contents of fatty acids in the ES fractions did not show a clear dependence on reaction temperature or on holding time. The majority of saturated acids, such as palmitic and stearic acids, were stable under the chosen conditions and the unsaturated fatty acids were only weakly affected. The main unsaturated fatty acids contained in rapeseed cake, oleic and linoleic acids, did not degrade but underwent cisetrans isomerization (do not occur for acids produced under conventional industrial hydrolysis procedures) that depended on the reaction temperature and holding time. For instance, the total content of the trans isomers of oleic acid and linoleic acids increased with both reaction temperature and holding time, reaching 6.6 g 1001 g1 within 60 min at 280 C. Similar results presented King and Holliday [36,37], who detected the presence of trans isomers of oleic, linoleic and linolenic acids and observed that isomerization occurred for incomplete hydrolysis conducted between 270 and 320 C.
3.6. Modelling of the hydrothermolysis of rapeseed cake, using optimal experimental design Reaction temperature and holding time were the reaction parameters to affect the rate of hydrolysis of rapeseed cake and the production of free amino acids and fatty acids in subcritical water. On the basis of the results obtained in preliminary studies, in the second stage of this work, two experimental regions (Fig. 4) were explored using Doehlert matrices with reaction temperature and holding time as
Table 5 e Doehlert experimental design with effective (ui) variables and values of corresponding experimental and calculated responses (YA in WS fractions) for hydrothermolysis of rapeseed cake.
Table 6 e Doehlert experimental design with effective (ui) variables and values of corresponding experimental and calculated responses (YFFA in ES fractions) for hydrothermolysis of rapeseed cake.
Experimental series 1
Experimental series 2
Experiment 1 2 3 4 5 6 7 70 c 700 c
u1 T (oC)
u2 t (min)
YA-expa (g kg1)
YA-cal.b (g kg1)
240 180 225 195 225 195 210 210 210
30 30 60 0 0 60 30 30 30
125.8 112.9 111.0 55.6 102.2 118.9 128.3 136.1 135.1
130.1 108.9 106.7 59.9 97.9 123.2 133.2 133.2 133.2
Experiment 1 2 3 4 5 6 7 70 c 700 c
a
a
b
b
Experimental values. Calculated values. c Experiments repeated at the centre of the experimental region to calculate the standard deviation on the response: sYA ¼ 7.0.
u1 T (oC)
u2 t (min)
YFFA-exp.a (g kg1)
YFFA-cal.b (g kg1)
260 200 245 215 245 215 230 230 230
40 40 80 0 0 80 40 40 40
0.80 0.65 0.92 0.60 0.70 0.71 0.85 0.85 0.86
0.83 0.62 0.90 0.63 0.67 0.73 0.86 0.86 0.86
Experimental values. Calculated values. c Experiments repeated at the centre of the experimental region to calculate the standard deviation on the response: sYFFA ¼ 0.006.
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
57
Fig. 5 e Two- (a, c) and three-dimensional (b, d) representations of the yield of amino acids YA (a, b) in WS fractions and of the conversion of triacylglycerols to fatty acids YFFA (c, d) in ES fractions, as a function of reaction temperature and holding time during the hydrothermolysis of rapeseed cake.
effective variables (ui). In experimental series 1, reaction temperature was limited to the range from 180 to 240 C, with holding times up to 60 min. These experiments aimed at optimizing the amino acid content in the WS fractions. Experimental series 2 was aimed at optimizing the conversion of triacylglycerols leading to free fatty acids in the ES fractions at reaction temperatures from 200 to 260 C and holding times of 0e80 min. The conditions of the experiments carried out in the defined experimental regions and the obtained values of the experimental responses: yield of amino acids in series 1 and the conversion of triacylglycerols to fatty acids in series 2, are listed in Tables 5 and 6, respectively. The results of experimental series 1 show that experiments performed at relatively low reaction temperature (195 C) and shortest holding time (0 min) led to the lowest value of YA. Yields higher than 100 g kg1 may be reached also at lower temperatures (180 and 195 C) but choosing longer holding times (30 and 60 min respectively, Exps. 2 and 6). However,
performing the reaction simultaneously at high reaction temperature and long holding time resulted in a drop of YA (Exp. 3). Experimental series 2 was focused on higher reaction temperatures (up to 260 C) and longer holding times (up to 80 min). The conversion of triacylglycerols to fatty acids (YFFA) increased with both reaction temperature and holding time (Exps. 2e7); however, in the investigated experimental region responding to simultaneously highest reaction temperature and long holding time, a drop of fatty acid yield was observed (Exp. 1). Attempts to fit the experimentally determined amino acid yield and the conversion of triacylglycerols to fatty acids to a quadratic model (Eq. (5)) led to satisfactory results (predicted values are consistent with experimental ones, within experimental errors). The following quadratic models were calculated by multi-linear least-square regression from the results obtained from experimental series 1 (Eq. (6)) and series 2 (Eq. (7)):
58
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Table 7 e Overview of hydrothermolysis of selected biomass origin substrates leading to amino acids and fatty acids. Raw material
Experimental conditions for the best yield of desired product
Yield of main product
20 g kg1 of dry mass 9.14 g kg1 of raw material 19.65 g kg1 of raw material 325 g kg1 protein
[15]
0.86 g kg1 of dry mass 755 g kg1 of raw silk 203 g kg1 of raw silk 4.35 g kg1 of dry mass 83 g kg1 of raw material
[34] [38] [38] [35] [25]
Rapeseed cake
330 C, 30 min, with addition of CO2 220 C, 30 min 200 C, 30 min 250 C, 60 min, substrate concentration of 10 g L1 230 C, 30 min, water:feed ratio ¼ 9 220 C, 60 min, water:feed ratio ¼ 50 160 C, 60 min, water:feed ratio ¼ 20 200 C, 30 min 264 C, 29 min, with addition of 0.83 M sodium bicarbonate 215 C, 36 min
135.9 g kg1 of dry mass
Present study
Fatty acids Coconut oil Linseed oil Soybean oil Squid wastes Sunflower oil Rapeseed cake
270 280 270 240 350 246
>97% conversion >97% conversion >97% conversion 440 g kg1 oil 92.8% conversion 91% conversion
[36] [36] [36] [44] [42] Present study
Amino acids Bean dregs Deoiled rice bran Deoiled soybean Hog hair Lawn grass clipping Silk waste-fibroin Silk waste-sericin Water lettuce Whey protein isolate
YA ¼ 133:2 þ 10:8x1 þ 20:8x2
13:8x21
C, C, C, C, C, C,
15 15 20 40 15 65
min min min min min min
43:7x22
31:5x1 x2
(6)
In experimental series 1 the standard error for coefficients b0, b1, b2 was 4.0, for b11, b22 6.4, and for b12 8.1. YFFA ¼ 0:9 þ 0:1x1 þ 0:1x2 0:1x21 0:1x22 þ 0:1x1 x2
(7)
In experimental series 2 the standard error for coefficients b0, b1, b2 was 0.003, for b11, b22 0.005 and for b12 0.006. Two- (a, c) and three-dimensional (b, d) representations of YA (a, b) in the WS fractions and YFFA in the ES fractions (c, d) as a function of reaction temperature and holding time, calculated using the above models are shown in Fig. 5. The contour plots in Fig. 5a and c represent the curves of constant value of the estimated responses YA and YFFA using the quadratic models. In the investigated experimental region, the optimal range of protein hydrolysis leading to amino acids was predicted at reaction temperatures of 200e230 C and holding times from 46 to 25 min, respectively. The maximum value of amino acids was 135.9 g kg1 (3.94) predicted at 215 C and 36 min. The obtained results are in relatively good agreement with results published in the work of Luo et al. [34], they employed surface methodology to analyse the influence of the water: feed ratio, reaction temperature and holding time on hydrothermolysis of lawn grass clippings leading to obtain amino acids. The highest yield of amino acids was obtained with water:feed ration of 5.5 at 230 C and 30 min. The maximum region for the conversion of triacylglycerols to fatty acids was estimated for reaction temperatures and holding times from 240 to 252 C and 74 to 55 min, respectively, with a maximum conversion of 0.91 (0.005) at 246 C and 65 min. In contrast to our results, Johnson et al. [45] studying microalgae hydrothermolysis, determined that short holding time (shorter as 30 min) with high temperatures
Reference
[31] [28]
(300e350 C) maximized the yield of fatty acids before degradation occurred. As shown, the hydrothermolysis of rapeseed cake in the optimal region proceeded with high yields of both amino acids and fatty acids in comparison with literature data (Table 7). Since the hydrothermal biomass conversion to valuable chemicals significantly depends on composition of applied raw material, higher yields were obtained only for the processes in which the raw materials were pure protein or oil substances. Owing to the fact that rapeseed cake has been found as an attractive raw material for production of both amino and fatty acids in the separate processes, it would be advantageous to perform the hydrothermolysis in a two-step process, as proposed by Tavakoli and Yoshida [20,21] who hydrolysed respectively squid entrails and scallop viscera to simultaneously amino acids and oil fraction rich in free fatty acids. The aim of the first step of rapeseed cake hydrothermolysis could be the production of amino acids as the primary product. On the basis of the present study we assess that the optimal conditions for amino acid production are appropriate for the first step of the two-step process, since the hydrolysis of triacylglycerols under these conditions occurred only with the 0.77 conversion. As the post-reaction residue still contained unreacted fat components it might be used as the raw material for the production of fatty acids in the second step.
4.
Conclusions
It was found that rapeseed cake due to its high content of protein and fat components is useful raw material for production of amino acids and fatty acids in the hydrothermolysis process under subcritical conditions. Given the
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
complexity of the process, the optimal experimental design methodology was chosen to elucidate the dependence of the yield of amino and fatty acids on the two reaction parameters (temperature and holding time). On the basis of the OED experiments it was found that the yield of amino acids reached highest values equal to approximately 134.1 g kg1 at ranges of reaction temperatures from 200 to 232 C and holding times of 46e25 min respectively, and the predicted maximum value of amino acids was 135.9 g kg1 at 215 C and 36 min. The region of highest conversion degree of triacylgycerols to fatty acids was found at ranges of reaction temperatures from 240 to 252 C and holding times of 74e55 min respectively, with the maximum conversion degree of 0.9, and estimated maximum conversion degree was 0.91 at 246 C and 65 min. Since the optimal parameters of amino acid production are significantly lower than for fatty acids (their conversion in the region optimal for amino acids reached only 0.77 conversion), it is worth considering carrying out the hydrothermolysis in a twostep process: the first step is hydrolysis of proteins to amino acids, and the second step is hydrolysis of the remaining solid post-reaction residue containing the fat component to fatty acids. After isolation and proper purification, the bioproducts, amino acids and fatty acids, obtained from the hydrothermolysis of the rapeseed cake, may be used for various aims of industrial interest.
Acknowledgements The authors acknowledge the financial support of this work (project No. N N523 494134) provided by the Ministry of Science and Higher Education of Poland during the years 2008e2010. The authors also thank Prof. Dr. Andre´ M. Braun (Karlsruhe Institute of Technology, Germany) and Dr. Irena JacukowiczSobala (University of Economics in Wrocław, Poland) for their valuable critical comments and suggestions.
[5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.biombioe.2014.03.028.
[19]
[20]
references [21] ¨ zc¸imen D, Karaosmanog lu F. Production and [1] O characterization of bio-oil and biochar from rapeseed cake. Renew Energy 2004;29(5):779e87. [2] Ucar S, Ozkan AR. Characterization of products from the pyrolysis of rapeseed oil cake. Bioresour Technol 2008;99(18):8771e6. [3] Giannakopoulou K, Lukas M, Vasiliev A, Brunner C, Schnitzer H. Low pressure catalytic co-conversion of biogenic waste (rapeseed cake) and vegetable oil. Bioresour Technol 2010;101(9):3209e19. [4] Giannakopoulou K, Lukas M, Vasiliev A, Brunner C, Schnitzer H. Conversion of rapeseed cake into bio-fuel in a
[22]
[23]
[24]
59
batch reactor: effect of catalytic vapour upgrading. Micropor Mesopor Mater 2010;128(1e3):126e35. Kandefer S, Baron J, Olek M. Co-burning of rapecake and hard coal in a grate furnace. Ochr Pow Pr Odp 2006;42(3):69e76 [in Polish]. Akiya N, Savage PE. Roles of water for chemical reactions in high-temperature water. Chem Rev 2002;102(8):2725e50. Kruse A, Dinjus E. Hot compressed water as reaction medium and reactant. Properties and synthesis reactions. J Supercrit Fluids 2007;39(3):362e80. Kruse A, Dinjus E. Hot compressed water as reaction medium and reactant. 2. Degradation reactions. J Supercrit Fluids 2007;41(3):361e79. Peterson AA, Vogel F, Lachance RP, Froling M, Antal MJ, Tester JW. Thermochemical biofuel production in hydrothermal media: a review of sub- and supercritical water technologies. Energy Environ Sci 2008;1(1):32e65. Zhu G, Zhu X, Fan Q, Wan X. Recovery of biomass wastes by hydrolysis in sub-critical water. Resour Conserv Recycle 2011;55(4):409e16. Toor SS, Rosendahl L, Rudolf A. Hydrothermal liquefaction of biomass: a review of subcritical water technologies. Energy 2011;36(5):2328e42. Pavlovic I, Knez Z, Skerget M. Hydrothermal reactions of agricultural and food processing wastes in sub- and supercritical water: a review of fundamentals, mechanisms, and state of research. J Agric Food Chem 2013;61(34):8003e25. Cheng H, Zhu X, Zhu C, Qian J, Zhu N, Zhao L, et al. Hydrolysis technology of biomass waste to produce amino acids in sub-critical water. Bioresour Technol 2008;99(9):3337e41. Esteban MB, Garcı´a AJ, Ramos P, Ma´rquez MC. Kinetics of amino acid production from hog hog hair by hydrolysis in sub-critical water. J Supercrit Fluids 2008;46(2):137e41. Esteban MB, Garcı´a AJ, Ramos P, Ma´rquez MC. Sub-critical water hydrolysis of hog hair for amino acid production. Bioresour Technol 2010;101(7):2472e6. Yoshida H, Terashima M, Takahashi Y. Production of organic acids and amino acids from fish meat by sub-critical water hydrolysis. Biotechnol Progr 1999;15(6):1090e4. Quitain AT, Sato N, Daimon H, Fujie K. Production of valuable materials by hydrothermal treatment of shrimp shells. Ind Eng Chem Res 2001;40(25):5885e8. Daimon H, Kang K, Sato KN, Fujie K. Development of marine waste recycling technologies using sub- and supercritical water. J Chem Eng Jpn 2001;34(9):1091e6. Quitain AT, Faisal M, Kang K, Daimon H, Fujie K. Lowmolecular-weight carboxylic acids produced from hydrothermal treatment of organic wastes. J Hazard Mater 2002;B93(2):209e20. Yoshida H, Tavakoli O. Sub-critical water hydrolysis treatment for waste squid entrails and production of amino acids, organic acids, and fatty acids. J Chem Eng Jpn 2004;37(2):253e60. Tavakoli O, Yoshida H. Conversion of scallop viscera wastes to valuable compounds using sub-critical water. Green Chem 2006;8(1):100e6. Lamoolphak W, Goto M, Sasaki M, Suphantharika M, Muangnapoh Ch, Prommuag Ch, et al. Hydrothermal decomposition of yeast cells for production of proteins and amino acids. J Hazard Mater 2006;B137(3):1643e8. Lamoolphak W, De-Eknamkul W, Shotipruk A. Hydrothermal production and characterization of protein an amino acids from silk waste. Bioresour Technol 2008;99(16):7678e85. Rogalinski T, Herrmann S, Brunner G. Production of amino acids from bovine serum albumin by continuous sub-critical water hydrolysis. J Supercrit Fluids 2005;36(1):49e58.
60
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
[25] Espinoza AD, Morawicki RO. Effect of additives on subcritical water hydrolysis of whey protein isolate. J Agric Food Chem 2012;60(20):5250e6. A, Gu¨lbay S, Uskan B, Canel M. Biomass decomposition [26] Sinag in near critical water. Energ Convers Manage 2010;51(3):612e20. [27] Sereeewatthanawut I, Prapintip S, Watchiraruji K, Goto M, Sasaki M, Shotipruk A. Extraction of protein and amino acids from deoiled rice bran by subcritical water. Bioresour Technol 2008;99(3):555e61. [28] Watchararuji K, Goto M, Sasaki M, Shotipruk A. Value-added subcritical water hydrolysate from rice bran and soybean meal. Bioresour Technol 2008;99(14):6207e13. [29] Pourali O, Asghari FS, Yoshida H. Sub-critical water treatment of rice bran to produce of valuable materials. Food Chem 2009;115(1):1e7. [30] Sunphorka A, Chavasiri W, Oshima Y, Ngamprasertsith S. Kinetic studies on rice bran protein hydrolysis in subcritical water. J Supercrit Fluids 2012;65:54e60. [31] Zhu G, Zhu X, Fan Q, Liu X, Shen Y, Jiang J. Study of amino acids from bean dregs by hydrolysis in sub-critical water. Chin J Chem 2010;28(10):2033e8. [32] Barreiro DL, Prins W, Ronsse F, Brilman W. Hydrothermal liquefaction (HTL) of microalgae for biofuel production: state of the art review and future prospects. Biomass Bioenergy 2013;53:113e27. [33] Garcia-Moscoso JL, Obeid W, Kumar S, Hatcher PG. Flash hydrolysis of microalgae (Scenedesmus sp.) for protein extraction and production of biofuels intermediates. J Supercrit Fluids 2013;82:183e90. [34] Luo G, Cheng X, Shi W, Strong PJ, Wang H, Ni W. Response surface analysis of the water:feed ratio influences on hydrothermal recovery from biomass. Waste Manage 2011;31(3):438e44. [35] Luo G, Shi W, Chen X, Ni W, Strong PJ, Jia Y, et al. Hydrothermal conversion of water lettuce biomass at 473 or 523 K. Biomass Bioenergy 2011;35(11):4855e61. [36] Holliday RL, King JW, List GR. Hydrolysis of vegetable oils in sub- and supercritical water. Ind Eng Chem Res 1997;36(3):932e5. [37] King JW, Holliday RL, List GR. Hydrolysis of soybean oil in a subcritical water flow reactor. Green Chem 1999;1(6):261e4. [38] Pinto JSS, Lanc¸as FM. Hydrolysis of corn oil using subcritical water. J Brazil Chem Soc 2006;17(1):85e9. [39] Minami E, Saka S. Kinetics of hydrolysis and methyl esterification for biodiesel production in two-step supercritical methanol process. Fuel 2006;85(17e18):2479e83. [40] Kocsisova T, Juhasz J, Cvengros J. Hydrolysis of fatty acids esters in subcritical water. Eur J Lipid Sci Tech 2006;108(8):652e68. [41] Moquin PHL, Temelli F. Kinetic modeling of hydrolysis of canola oil in supercritical media. J Supercrit Fluids 2008;45(1):94e105. [42] Alenezi R, Leeke GA, Santos RCD, Khan AR. Hydrolysis of sunflower oil under subcritical water conditions. Chem Eng Res Des 2009;87(6):867e73. [43] Milliren AL, Wissinger JC, Gottumukala V, Schall CA. Kinetics of soybean oil hydrolysis in subcritical water. Fuel 2013;108:277e81. [44] Tavakoli O, Yoshida H. Squid oil and fat production from squid wastes using subcritical water hydrolysis: free fatty acids and transesterification. Ind Eng Chem Res 2006;45(16):5675e8. [45] Johnson MC, Tester JW. Lipid transformation in hydrothermal processing of whole algal cells. Ind Eng Chem Res 2013;52(32):10988e95. [46] Tavakoli O, Yoshida H. Conversion to valuable compounds using sub-critical water. Green Chem 2006;8(1):100e6.
[47] Ruiz HA, Rodrı´guez-Jasso RM, Fernandes BD, Vecente AA, Teixeira JA. Hydrothermal processing, as an alternative for upgrading agriculture residues and marine biomass according to the biorefinery concept: a review. Renew Sust Energ Rev 2013;21:35e51. [48] Undersander D, Mertens DR, Thiex N. Forage analyses procedures. Omaha, Nebrasca: The National Forage Testing Association; 1993 Jul. p. 147. [49] Hames B, Scarlata C, Sluiter A. Determination of protein content in biomass. revised 2008, May. Golden, Colorado: National Renewable Energy Laboratory; 2008. p. 8. Report No. TP-510-42625. [50] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D, et al. Determination of structural carbohydrates and lignin in biomass. Golden, Colorado: National Renewable Energy Laboratory; 2008 Apr. p. 18. revised 2011, Jul; Report No. TP510-42618. [51] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D. Determination of ash in biomass. Golden, Colorado: National Renewable Energy Laboratory; 2005 Jul. p. 8. Report No. TP510-42622. [52] Van Soest PJ, Robertson JB, Lewis BA. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci 1991;74(10):3583e97. [53] Carrier M, Loppinet-Serani A, Denux D, Lasnier JM, HamPichavant F, Cansell F, et al. Thermogravometric analysis as a new method to determine the lignocellulosic composition of biomass. Biomass Bioenerg 2011;35(1):298e307. [54] Lisiewska Z, Kmiecik W, Korus A. The amino acid composition of kale (Brassica oleracea L. var. acephala), fresh and after culinary and technological processing. Food Chem 2008;108(2):642e8. ski P, Sobczyn ska L. Amino [55] Lisiewska Z, Kmiecik W, Ge˛bczyn acid profile of raw and as-eaten products of spinach (Spinacia oleracea L.). Food Chem 2011;126(2):460e5. [56] Standard test method for acid value of fatty acids and polymerized fatty acids. West Conshohocken, Pennsylvania: American Society for Testing and Materials; 1998. p. 2. ASTM D1980-87. [57] Standard test method for determination of the saponification value of fats and oils. West Conshohocken, Pennsylvania: American Society for Testing and Materials; 2006. p. 2. ASTM D5558-95. [58] Atkinson AC, Donev AN. Optimum experimental design. Oxford: Oxford University Press; 1992. [59] Rasch D, Verdooren LR, Gowers JI. Fundamentals in the design and analysis of experiments and surveys- Grundlagen der Plannung und Auswertung von Versuchen und Erhebungen. Mu¨nchen: Oldenburg Wissenschaftsverlag; 1999. [60] Ferreira SLC, Dos Santos WNL, Quintella CM, Neto BB, Bosque-Sebdra JM. Doehlert matrix. A chemometric tool for analytical chemistry e review. Talanta 2004;63(4):1061e7. [61] Sato N, Quitain AT, Kang K, Gaimon H, Fujie K. Reaction kinetics of amino acid decomposition in high-temperature and high-pressure water. Ind Eng Chem Res 2004;43(13):3217e22. [62] Klingler D, Berg J, Vogel H. Hydrothermal reactions of alanine and glycine in sub- and supercritical water. J Supercrit Fluids 2007;43(1):112e9. [63] Goto M, Obuchi R, Hirose T, Sakaki T, Shibata M. Hydrothermal conversion of municipal organic waste into resources. Bioresour Technol 2004;93(3):279e84. [64] Dı´az MJ, Cara C, Ruiz E, Romero I, Moy M, Castro E. Hydrothermal pre-treatment of rapeseed straw. Bioresour Technol 2010;101(7):2428e35.
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
A, Gu¨lbay S, Uskan B, Gu¨llu¨ M. Comparative studies of [65] Sinag intermediates produced from hydrothermal treatments of sawdust and cellulose. J Supercrit Fluids 2009;50(2):121e7. [66] Carvalheiro F, Esteves MP, Parajo JC, Pereira H, Girio FM. Production of oligosaccharides by autohydrolysis of brewery’s spent grain. Bioresour Technol 2004;91(1):93e100. [67] Peterson AA, Lachance RP, Tester JW. Kinetic evidence of the Maillard reaction in hydrothermal biomass processing: glucoseeglycine interactions in high-temperature, highpressure water. Ind Eng Chem Res 2010;49(5):2107e17.
61
[68] Tehrani KA, Kersiene M, Adams A, Venskutonis R, De Kimpe N. Thermal degradation studies of glucose/glycine melanoidins. J Agric Food Chem 2002;50(14):4062e8. [69] Abdelmoez W, Nakahasi T, Yoshida H. Amino acids transformation and decomposition in saturated subcritical water conditions. Ind Eng Chem Res 2007;46(16):5286e94. [70] Sohn M, Ho CT. Ammonia generation during thermal degradation of amino acids. J Agric Food Chem 1995;43(12):3001e3.
Available online at www.sciencedirect.com
ScienceDirect http://www.elsevier.com/locate/biombioe
Hydrothermolysis of rapeseed cake in subcritical water. Effect of reaction temperature and holding time on product composition kowska a,*, Paweł Wolak a, Esther Oliveros b Hanna Pin a
Department of Industrial Chemistry, Wrocław University of Economics, ul. Komandorska 118/120, 53-345 Wrocław, Poland b Laboratoire des IMRCP, UMR CNRS 5623, Universite´ Paul Sabatier (Toulouse III), 31062 Toulouse Ce´dex 9, France
article info
abstract
Article history:
We have investigated the hydrothermolysis of rapeseed cake, which is a residual biomass
Received 29 September 2011
with high protein and fat content. The effects of the process parameters, reaction tem-
Received in revised form
perature and holding time, on the composition of the products (amino acids and fatty
7 November 2013
acids) contained in the separated liquid fractions (water and ether, respectively) were
Accepted 14 March 2014
studied. To model the hydrothermal process the experimental design methodology was
Available online 14 April 2014
used. Based on this, optimized operational parameters were determined yielding highest amounts of amino acids and high conversion of triacylglycerols to fatty acids. A maximum
Keywords:
estimated yield of amino acids of 135.9 g kg-1 of rapeseed cake was obtained at 215 C after
Rapeseed cake
36 min. A further increase of both reaction temperature and holding time would lead to the
Subcritical water
Maillard reaction and the decomposition (deamination and decarboxylation) of the amino
Hydrolysis
acids. Maximum conversion of triacylglycerols to fatty acids of 0.91 was predited at 246 C
Amino acids
during 65 min. In this case, a further increase of both reaction parameters would lead to
Fatty acids
enhanced cisetrans isomerization of the unsaturated fatty acids produced.
Experimental design methodology
1.
Introduction
Rapeseed cake is a residual biomass from the pressing of oil from rapeseed. From the processing of rapeseed large amounts of rapeseed cake are released as by-product (about 650 kg from 1 t of rapeseed). Poland is one of the major rapeseed (Brassica napus L.) producers in Europe and the rapeseed growing area is forecast to further increase in the years to come. Rapeseed cake has a high protein and fat content. Traditionally, rapeseed cake has been applied as a livestock
* Corresponding author. Tel./fax: þ48 71 368 0275. kowska). E-mail address: [email protected] (H. Pin http://dx.doi.org/10.1016/j.biombioe.2014.03.028 0961-9534/ª 2014 Elsevier Ltd. All rights reserved.
ª 2014 Elsevier Ltd. All rights reserved.
feed. However, the forecasted increase in rapeseed oil production may cause difficulties in the agricultural application of rapeseed cake. For several years, research into rapeseed cake through thermochemical processing, e.g. pyrolysis yielding bio-oil and biocarbon [1,2], catalytic conversion producing bio-oil [3,4], and co-combustion with hard coal [5], has been conducted worldwide. However, these utilization methods do not allow recovery of the valuable components of the residual biomass, which is possible by using hydrothermal processes conducted in subcritical water. Due to its special properties [6e8], most
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
biomass raw materials may be easily hydrolysed to many valuable feedstocks, among them to amino acids and fatty acids. Depending on the biomass composition the hydrothermal processes of biomass treatment may lead to only amino acids, only fatty acids or mixtures which after separation give both products: aqueous phase containing amino acids and oil phase, rich in fatty acids [9e12]. Recently there was efficiently performed the hydrothermolysis of protein wastes derived from meal processing industry [13e15], seafood industry [16e21], brewery industry [22], silk industry [23] and also other raw materials rich in proteins such as bovine serum albumin [24] and whey protein isolate [25], sewage sludge [19], baby food [26], rice bran [27e30], soybean meal [28], bean dregs [31], microalgae [32,33], raw grass clippings [34] and water lettuce [35]. Several attempts also have been made at the hydrothermolysis of a fat-rich resources: vegetable oils [36e43], squid wastes [20,44], fish meal (horse mackerel) [16] and microalgae [45], which resulted in saturated, monoenic, dienic and polyenic fatty acids, including long chain polyunsaturated acids. In turn the conversion of scallop viscera wastes and squid entrails in a two-step process allowed production of separated fractions: rich in amino acids and fatty acids [43,46]. The hydrothermal methods (under subcritical conditions) have many advantages in relation to traditional methods of amino acids and fatty acids production based on chemical and microbiological syntheses, which are complicated, expensive, require toxic reagents and cause many purification problems. In contrast hydrothermal technologies do not require the addition and recovery of chemicals except for water in consequence allowing valuable materials to be produced with zero emissions [46]. The drying of raw material is not required which makes the process energy saving. Due to hightemperature, high-pressure conditions and the catalytic properties of water arising from its high ionic product under hydrothermal conditions, reactions are faster and may be easily controlled by temperature and pressure adjustment [13,34]. Also lower dielectric constant, viscosity and surface tension of subcritical water enhance the solubilization of treated raw materials and transport of the reagents which accelerate the reaction rates [6,7,13]. Other advantages of hydrothermal processes are limited equipment corrosion problems and simple operation [47]. Considering the above, hydrothermal processes are regarded as environmental friendly, cost effective and having a great potential for practical applications. Since investigations of rapeseed cake hydrothermolysis are rare (to the best of our knowledge no report has been published) herein we propose a hydrothermal path for rapeseed cake treatment yielding a liquid fraction rich in amino acids and fatty acids as an alternative method of its utilization. This study focuses on the effects of the reaction temperature and time of hydrothermolysis of rapeseed cake, as a first step towards a mechanistic interpretation and an evaluation of the feasibility of its technical development. The optimal experimental design (OED) methodology was used to establish a statistically significant reaction model as well as for optimizing the conditions for amino acids and fatty acids production.
2.
Materials and methods
2.1.
Materials and chemicals
51
Rapeseed cake was obtained from the Wilmar Oil Press Plant _ ´ rawina (the Lower Silesian Province, Poland). The moist in Zo material was dried to constant mass at a temperature of 103 C and ground to a grain size of below 1 mm. All solvents and reagents were purchased from SigmaeAldrich and Fluka (amino acids). Depending on the requirements of the particular analytical or investigative method, analytically or HPLCpure reagents were used.
2.2.
Reactor and experimental procedure
The hydrothermolysis of rapeseed cake was performed in a batch reactor (4576A, Parr Instrument Company, USA). A detailed description of the reactor equipment, as well a schematic diagram of experimental set-up (Fig. 1S), is given in Supplementary material. The investigation of hydrothermolysis of rapeseed cake was conduced in two stages. The aim of the first preliminary stage was to establish the range of temperature and holding time of hydrothermolysis of the raw material conversion to amino acids and fatty acids with significant yield. On the basis of these results the optimal experimental design was applied in two experimental series for optimization of amino acids (series 1) and fatty acids (series 2) production, which was the second stage of the study. In preliminary study the hydrothermolysis of the rapeseed cake was conducted at a reaction temperature (T) of 180, 200, 220, 240, 260 and 280 C. The reaction would be stopped once the intended temperature was reached (zero holding time), and after a holding time (t) of 5, 10, 20, 30, 40 and 60 min. In all experiments the pressure corresponded to the vapour pressure curve at the given temperature, or slightly exceeded it (at 180 Ce1.23 MPa, 200 Ce1.85 MPa, 220 Ce2.88 MPa, 240 Ce4.14 MPa, 260 Ce5.52 MPa, 280 Ce7.45 MPa). In the second stage of the study two series of designed experiments were performed in which reaction temperature and holding time were varied simultaneously: in experimental series 1, rapeseed cake hydrothermolysis was performed at temperature ranging from 180 to 240 C and holding time from 0 to 60 min, whereas in series 2, temperature varied from 200 to 260 C and holding time from 0 to 80 min. The experimental conditions are given in Supplementary Material (Table 1S). For each hydrothermolysis experiment, HPLC-pure water was subjected to degasification in an ultrasonic bath and purged with nitrogen. Rapeseed cake (10 g) in water (90 g) suspension was introduced into the reaction vessel, preheated to about 80 C. The reactor was closed and the reaction vessel with its content was purged with nitrogen under a pressure of 2 MPa and heated up at a rate of 10e15 C min1 to a predetermined temperature for 10e15 min, and kept at this temperature with an accuracy of 1 C. The time during which the reaction mixture components were kept at the prescribed temperature was considered as the reaction time (holding time). When the reaction was over, the reaction vessel was
52
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
cooled down to about 90 C for about 5 mine10 min and after the system was decompressed the vessel was emptied and washed with water, to reach the final volume of the aqueous fraction of 150 mL.
2.3. Separation of rapeseed cake hydrothermolysis products The hydrothermolysis of the rapeseed cake yielded a liquid product and a post-reaction solid residue. The water fraction was separated by filtering it through a PTFE membrane filter (Sartorius, SRP 15, 0.45 mm). For the sake of the simplicity and the effectiveness of separation [36,37,43] and requirements of chromatographic analysis of oil fraction components for its separation diethyl ether was used. Diethyl ether (100 mL) was added to the permeate and after one-step extraction the top layer was separated from the bottom layer. The diethyl ether from the top layer was removed by evaporation in a vacuum drier at a temperature of 45 C, whereby the fraction containing ether-soluble components (ES fraction) was obtained and analysed for free fatty acid (FFA) content. The bottom layer containing water-soluble substances (WS fraction) was diluted with water to a volume of 250 mL and amino acids and volatile fatty acids (VFA) content was determined. The yields (Yi) of amino acids or volatile fatty acids contained in the WS fractions were calculated using the following equation (Eq. (1)): Yi g kg
1
¼ mi =mrc
(1)
where mi is the mass (g) of the i reaction product, and mrc is the mass (kg) of rapeseed cake subjected to hydrothermolysis. In the ES fractions the degree of conversion (YFFA) of the triacylglycerols contained in the rapeseed cake, leading to FFA, was calculated from Eq. (2): YFFA ð Þ ¼ AVi =SVrc
(2)
where AVi is the acid value determined in the ES fractions, and SVrc is the saponification value of the rapeseed cake.
2.4.
Analyses and analytical methods
Rapeseed cake dried to constant mass was investigated, except for moisture content. The elementary composition of the rapeseed cake was determined using the Vario EL III apparatus provided by Elementar Analysensysteme GmbH. The content of dry matter and diethyl ether extractable substances in rapeseed cake was determined using The National Forage Testing Association (NFTA) procedures [48], while the content of ash, protein, sulphuric acid soluble lignin and acetyl groups was determined using the National Renewable Energy Laboratory (NREL) protocols [49e51]. Lignocellulosic content (without calcination step) was assayed in the rapeseed cake using the detergent method [48,52,53]. The rapeseed cake was subjected to chromatographic analysis to determine the content and composition of bound [54,55] amino acids. The free amino acids content and composition in the WS fractions obtained in the first preliminary stage of the study was also determined by chromatographic analysis, while the concentration of the free
amino acids in the WS fractions achieved in the second stage of the study, was determined by the ninhydrin method using a procedure according to Watchararuji et al. [28]. The proportion of individual amino acids contained in the WS fractions was depicted as their mass fraction in relation to 100 g of all detected amino acids expressed as: QA (g mA 1001 g1 mSA) where mA stands for individual amino acid, and mSA stands for the sum of all detected amino acids. The contents of fatty acids in the rapeseed cake and in the ES fractions, as well as the contents of formic, acetic, propionic, isobutyric, butyric, isopentanoic and pentanoic acids in the WS fractions were determined by gas chromatography. The proportion of individual fatty acids was depicted as their mass fraction in relation to 100 g of all detected fatty acids (QFFA) expressed as. QFFA (g mFFA 1001 g1 mSFFA) where mFFA stands for individual fatty acid and mSFFA stands for the sum of all detected fatty acids. The operating conditions of all applied chromatographic analyses and ninhydrin method are described in Supplementary material. The acid value of the rapeseed cake (AVrc) and that of the ES fractions (AVi) [56] and the saponification value of the rapeseed cake (SVrc) [57] were determined. All the analytical determinations were performed in triplicate and the mean values were calculated.
2.5.
Modelling and optimization method
The hydrothermolysis of rapeseed cake implies complex reaction manifolds, and the product distribution changes strongly with the variation of the process parameters. The OED methodology [58e60] was used to facilitate the interpretation of experimental results and to determine the process parameters for an optimal yield of amino acids and degree of conversion of triacylglycerols to fatty acids in the WS and ES fractions, respectively. The OED methodology makes use of statistical tools to select a minimum set of experiments adequately located in a chosen experimental region (experimental matrix) and to determine the coefficients of a mathematical model that represents the variations of the experimental response of interest with the best possible precision. In this methodology, the effective values of the process parameters to be varied are coded or normalized, thus enabling the quantitative analysis of the results for different units and ranges of variation of the variables. The relation between normalized and effective variables is given by Eq. (3): xi ¼ ui ui;0
Dui
(3)
where xi is an independent normalized variable and values of xi range from 1 to þ1, ui is an effective variable, ui,0 is the value of the effective variable at the centre of the experimental region (corresponding to xi ¼ 0) and Dui is the step given by Eq. (4): Dui ¼ ui;max ui;min
2
(4)
In this work, the effects of two variables were investigated: reaction temperature (T or u1) and holding time (t or u2).
53
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
From among the possible experimental designs based on the quadratic model, the Doehlert array was selected [58,60]. The values of the experimental response y were fitted to the empirical quadratic polynomial model: y ¼ b0 þ b1 x1 þ b2 x2 þ b11 x21 þ b22 x22 þ b12 x1 x2
(5)
where b0, b1, b2, b11, b22 and b12 are the coefficients of the quadratic model. A Doehlert matrix for 2 variables (T and t) contains 7 uniformly distributed experiments, which, using normalized independent variables xi (Table 1), may be represented by the vertices and the centre of a hexagon. The degree of conversion of the rapeseed cake protein fraction and fatty fraction, defined respectively as the yield of amino acids (y ¼ YA) in the WS fractions, and the degree of conversion of triacylglycerols, leading to fatty acids (y ¼ YFFA) in the ES fractions, was modelled and optimized. Standard deviations sb of experimental responses y were evaluated by repeating three times the experiments at the centres of the experimental domains (xi ¼ 0). The values of coefficients bi of the quadratic model (Eq. (5)) were calculated for dependent variables YA and YFFA. The NEMROD software (Version 2002, LPRAI, France) was used for calculating (by the least-square regression method) the model coefficients and evaluating the significance of the regression and for validity tests.
3.
Results and discussion
3.1.
Composition of rapeseed cake
The rapeseed cake has the molecular formula [C3.9H6.3O2.5N0.4S0.02]n, and the following elementary composition: carbon (46.9%), hydrogen (6.3%), oxygen (40.7%), nitrogen (5.6%) and sulphur (0.5%). The rapeseed cake dry matter content amounted to (mean values and standard deviations, respectively) 905.2 g kg1 3.7 g kg1, ash to 68.8 g kg1 0.6 g kg1, total protein to 349.8 g kg1 2.6 g kg1 (calculated from the total Kjeldahl Nx6.25 [48]), diethyl ether extractable substances represented 106.6 g kg1 7.3 g kg1, hemicellulose 42.2 g kg1 1.4 g kg1, acetyl groups 28.1 g kg1 1.5 g kg1, cellulose 135.3 g kg1 2.1 g kg1, sulphuric acid insoluble lignin 112.9 g kg1 1.4 g kg1 and sulphuric acid soluble lignin 70.0 g kg1 1.9 g kg1. The density of the rapeseed cake is
Table 1 e Doehlert experimental design with normalized (coded) variables (xi). Experiment 1 2 3 4 5 6 7 70 a 700 a a
x1
x2
þ1 1 þ0.5 0.5 þ0.5 0.5 0 0 0
0 0 þ0.866 0.866 0.866 þ0.866 0 0 0
Experiments repeated at the centre of the experimental region to calculate the standard deviation on the response (sY).
0.62 g mL1, and the pH of the 100 g L1 water slurry has a value of 6.21. Table 2 shows the composition of the bound amino acids present in the raw material subjected to hydrothermolysis, whereas the composition of the fatty acids is shown in Table 3. The total bound amino acid content in the rapeseed cake amounts to 317.2 g kg1. Glutamic acid and aspartic acid predominate, followed by leucine, proline, valine, arginine and alanine. In the case of fatty acids, unsaturated acids, particularly oleic acid (C18:1n9) and linolenic acid (C18:2n6) predominate. No trans isomers were found to be present in the rapeseed cake. Its acid and saponification values are 15.3 and 192.5, respectively.
3.2. Effect of reaction temperature and holding time on the yield of amino acids in WS fractions The hydrothermolysis of rapeseed cake was investigated as a batch process. Under the chosen experimental conditions, subcritical water acted as solvent, reagent and catalyst. The protein content in the rapeseed cake was partly hydrolysed and decomposed in subcritical water and converted to watersoluble amino acids. Fig. 1 shows the effect of reaction temperature and time on the yield of amino acids in the WS fractions. At reaction time 0 min, the amino acid content increased with temperature from 180 C to 240 C, due to the increase of the ionic product of water in this temperature range. In the presence of hydronium and hydroxide ions, hydrolysis of the peptide bonds results in the breakdown of the proteins into smaller molecules of soluble proteins or amino acids [25]. The amino acid content also grows with increasing holding time from 0 to 10 min, before reaching a plateau. Similar to our results, Rogalinski et al. [24], Watchararuji et al. [28], Zhu et al. [31], Luo et al. [34,35] confirmed a great influence of reaction temperature and holding time on the hydrothermolysis of used raw materials, but the holding time
Table 2 e Bound amino acid content in rapeseed cake. Amino acids
Bound amino acid content (g kg1)
Bound amino acid mass fraction in relation to 31.7 g (%)
Total Alanine Arginine Cysteine Phenylalanine Glycine Histidine Isoleucine Aspartic acid Glutamic acid Leucine Lysine Methionine Proline Serine Threonine Tryptophan Tyrosine Valine
317.2 17.3 17.9 7.2 16.1 19.7 9.2 13.9 28.3 60.4 26.6 10.0 6.8 20.0 15.7 17.4 2.8 9.4 18.5
100.0 5.5 5.6 2.3 5.1 6.2 2.9 4.4 8.9 19.0 8.4 3.2 2.1 6.3 4.9 5.5 0.9 3.0 5.8
54
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Table 3 e Free fatty acid composition and content in rapeseed cake. Fatty acids Saturated acids (C8:0-C24:0) Monoenic acids C16:1n7 C18:1n7 Monoenic acids C16:1n9 C18:1n9 C20:1n9 Dienic acids C18:2n6 Polyenic acids C18:3n3
QFFA (g mFFA 1001 g1 mSFFA) 13.1 9.8 0.6 9.2 58.9 0.2 57.8 0.9 14.8 14.8 3.4 3.4
had less of an effect on amino acid yield when the temperature was higher than 220 C [28]. However, at higher temperatures (260 C and 280 C), the content of amino acids in the WS fractions decreased with increasing holding time. The decline of the yield of amino acids with increasing holding time at these temperatures is probably due to their degradation, i.e. deamination (resulting in the formation of ammonia) and decarboxylation (resulting in the formation of carbonic acids and amines, e.g., ethylamine and methylamine) [16,18,24,61]. This supposition may be confirmed by results presented in the literature. Rogalinski et al. [24] observed, that above 250 C, the amino acid decomposition rate exceeded the hydrolysis rate of protein and at 330 C there was almost complete degradation of all amino acids to ammonia, ethanoloamine, ornithine and carboxylic acids, such as acetic, propanoic, n-butyric and isobutyric. In other studies [61,62] it was also concluded that the hydrothermal decomposition of alanine and glycine leads to formation of ammonia, methylamine, ethylamine, acetaldehyde, diketopiperazine (dimerization product of glycine) and carboxylic acids, such as acetic, formic, glycolic, lactic, propionic and pyruvic. In our work, the gas phase was not analysed for products (e.g. volatile amines); however the content of some selected VFAs (formic, acetic, propionic, butyric, isobutyric, pentanoic and isopentanoic) as products of
Fig. 1 e Effect of reaction temperature and holding time on the yield of amino acids in WS fractions obtained from hydrothermolysis of rapeseed cake.
deamination of the amino acids [16,19,24,61,63] was determined in the WS fractions. From among the identified VFAs, formic and acetic acids were produced in the largest amounts of 3.3 g kg1 and 1.0 g kg1, respectively. However, these acids may result not only from the deamination of the amino acids resulting from the hydrolysis of the rapeseed cake protein fraction, but also from the decomposition of the cellulose fraction components [19,22,29,64,65]. Formic acid could also be produced by the degradation of the furfurals, whereas acetic acid might be the product of the acetyl cleavage from acetylated sugar monomers of the hemicellulose polymer [66]. In the investigated WS fractions, formic acid was present in variable proportions, and its yield did not show a clear correlation with the primary reaction parameters. The amount of acetic acid did not significantly in the whole investigated range, except that its yield decreased as the holding time was increased at a constant temperature. The yield of the other volatile acids in the WS fractions was very low (0.4e0.9 g kg1), but increased slightly with temperature and holding time. Another reason for the decrease of the amino acid yield in the WS fractions might be the Maillard condensation reaction, by which the amino groups of the amino acids, peptides or proteins, and the carbonyl groups (or the hemiacetal groups) present in oligomer and polymeric carbohydrates would produce melanoidins [67] (dark brown nitrogenous polymers with a nutty odour), whose main thermal degradation products at temperatures ranging from 100 to 300 C are pyrazines, pyridines, pyrroles and furans [68].
3.3.
Profiles of amino acid contents
Fig. 2 shows the profile and proportion of amino acids (QA) in selected WS fractions obtained from the hydrothermolysis of the rapeseed cake. In all WS fractions the predominant amino acid was glutamic acid whose content increased with reaction temperature and holding time. In the WS fractions obtained at 280 C and holding times of 30 and 60 min, the mass fraction of glutamic acid was higher than 50 g 1001 g1. The mass
Fig. 2 e Composition and content of amino acids in WS fractions obtained from hydrothermolysis of rapeseed cake: 1: serine, 2: glutamic acid, 3: glycine, 4: alanine, 5: valine, 6: leucine, 7: other amino acids (aspartic acid, treonine, proline, tyrosine, arginine, isoleucine, phenylalanine, histidine, lysine, cysteine, methionine, tryptophan).
55
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Fig. 3 e Effect of reaction temperature and holding time on the conversion of triacylglycerols to fatty acids (YFFA [ AVi/ SVrc, Eq. (2)) in ES fractions obtained from hydrothermolysis of rapeseed cake.
fraction of the other amino acids contained in the WS fractions would not exceed 10 g 1001 g1. The proportion of the amino acids in the WS fractions was found to be different from that of the bound amino acids in the rapeseed cake. As a result of the changes taking place in the course of the decomposition of the rapeseed cake protein fraction in subcritical water, the structure of the amino acids and their proportion in the WS fractions would change. The changes might be due to the decomposition of the less stable amino acids present in the reaction mixture, depending on the adopted reaction temperature and holding time [46], and to the thermal stability of the other amino acids under these conditions [18,69]. For example, alanine could be formed by transamination of serine with pyruvic acid, and glycine could result from the retroaldol condensation of serine with formaldehyde [61]. The concentrations of these two amino acids in the WS fractions obtained at temperatures from 180 C to 260 C and holding times from 0 to 60 min were in fact higher
than those determined in rapeseed cake, which was in agreement with previous studies presented in the literature [61,62]. A significant decrease of alanine and glycine contents, due to their thermal degradation, was however only observed in the WS fractions obtained at 280 C. Ammonia is the primary thermal degradation product regardless of the amino acid involved. Polar amino acids such as aspartic acid undergo decomposition releasing large amounts of ammonia, in contrast to nonpolar amino acids such as alanine, leucine and valine that remain stable. Amino acids containing more than one nitrogen atom (arginine, histidine, tryptophan) generated higher amounts of ammonia than others. The high content of negatively charged glutamic acid in the WS fractions can be explained by its high thermal stability due the additional methylene group in its side chains [70].
3.4. Effect of reaction temperature and holding time on the yield of fatty acids in ES fractions Fig. 3 shows the effects of the temperature and holding time on the yield of fatty acids in ES fractions formed as a result of the hydrolysis of triacylglycerols contained in the rapeseed cake. At temperatures from 180 to 240 C, the conversion of triacylglycerols to fatty acids that are relatively thermally stable in such conditions would increase with reaction temperature and holding time, reaching a value of 0.85e0.9 in 10e20 min. These results, describing the influence of reaction temperature and holding time on the yield of fatty acids, confirmed the previous results obtained by King et al. [36], Holliday et al. [37], Alenezi et al. [42] and Milliren et al. [43], who successfully hydrolysed soybean [36,37,43], linseed [36], coconut [36] and sunflower [42] oils in subcritical conditions. The increasing ionic product in the temperature range investigated promoted the hydrolysis of triacylglycerols. Also fatty acids may have acted as catalysts promoting the hydrolysis of triacylglycerols as well as of intermediate reaction products, such as di- and monoglycerols [43]. Such a
Table 4 e Profile and content of free fatty acids in selected ES fractions obtained from hydrothermolysis of rapeseed cake. QFFA (g mFFA 1001 g1 mSFFA) Saturated acids (C12:0eC22:0) C16:0 C18:0 Trans isomers C18:1 C18:2 Monoenic acids C16:1n7 C18:1n7 Monoenic acids C16:1n9 C18:1n9 C20:1n9 Dienic acids C18:2n6 Polyenic acids C18:3n3
Reaction temperature ( C)/holding time (min) 180/0
180/30
180/60
240/0
240/30
240/60
280/0
280/30
280/60
13.2 9.6 2.4 1.0 0.8 0.2 6.9 0.6 6.3 65.3 0.0 64.4 0.9 11.8 11.8 1.8 1.8
12.4 9.6 1.9 1.0 0.8 0.2 6.6 0.6 6.0 57.7 0.0 56.8 0.9 17.5 17.5 4.8 4.8
12.6 9.8 1.9 1.1 0.8 0.3 6.6 0.6 6.0 57.5 0.0 56.7 0.8 17.4 17.4 4.8 4.8
12.9 10.0 2.0 1.1 0.8 0.3 7.0 0.6 6.4 57.2 0.0 56.5 0.7 17.1 17.1 4.7 4.7
13.0 9.9 2.0 1.5 1.1 0.4 6.6 0.6 6.0 62.4 0.0 61.6 0.8 13.5 13.5 3.0 3.0
12.7 10.0 1.8 1.7 1.3 0.4 6.1 0.6 5.5 61.3 0.0 60.6 0.7 14.8 14.8 3.4 3.4
12.9 10.1 2.0 2.3 1.6 0.7 5.9 0.5 5.4 60.8 0.0 60.1 0.7 15.1 15.1 3.0 3.0
13.4 10.3 2.3 5.5 4.2 1.3 5.7 0.7 5.0 59.1 0.0 58.5 0.6 14.0 14.0 2.3 2.3
17.0 10.8 2.3 6.6 5.3 1.3 5.4 0.5 4.9 55.8 0.1 55.1 0.6 13.4 13.4 1.8 1.8
56
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
affected by their molecular structure. With the increase of reaction temperature the yield of fatty acids which contained double bonds decreased, polyenic acids demonstrated a slight or quick decrease, and saturated fatty acids reached a plateau.
3.5.
Fig. 4 e Graphical representation in effective variables (u1 [ T and u2 [ t) of experiments performed in series 1 and 2 (numbering of experiments as in Tables 5 and 6, respectively, in parentheses for series 1 and in brackets for series 2).
hypothesis would correspond to a mechanism where, in the first reaction step, fatty acids would dissociate, inducing protonation of the carboxyl oxygen of acyl glycerols [39]. Another advantage of hydrothermolysis in subcritical water is the reduced induction period compared to the traditional fat splitting processes. However, with increased reaction temperature and holding time (260, 280 C and after 20 min) the yield of fatty acids significantly decreased. Even though, the fatty acids are stable in subcritical water [36], a higher reaction temperature [44] and longer holding time of the reaction mixture enhanced degradation and pyrolysis of the triacylglycerols and fatty acids [36,44] and partial depolymerization of fatty acids [40]. Kocsisova´ et al. [40] observed structural changes in the chains of the unsaturated fatty acids and their dimerization during hydrothermolysis of rapeseed oil. Simultaneously, in the work of Tavakoli et al. [44], it was determined that the thermal stability of fatty acids was
Profiles of fatty acid contents
Table 4 shows the profile and proportion of the fatty acids (QFFA) contained in selected ES fractions obtained from hydrothermolysis of rapeseed cake. The contents of fatty acids in the ES fractions did not show a clear dependence on reaction temperature or on holding time. The majority of saturated acids, such as palmitic and stearic acids, were stable under the chosen conditions and the unsaturated fatty acids were only weakly affected. The main unsaturated fatty acids contained in rapeseed cake, oleic and linoleic acids, did not degrade but underwent cisetrans isomerization (do not occur for acids produced under conventional industrial hydrolysis procedures) that depended on the reaction temperature and holding time. For instance, the total content of the trans isomers of oleic acid and linoleic acids increased with both reaction temperature and holding time, reaching 6.6 g 1001 g1 within 60 min at 280 C. Similar results presented King and Holliday [36,37], who detected the presence of trans isomers of oleic, linoleic and linolenic acids and observed that isomerization occurred for incomplete hydrolysis conducted between 270 and 320 C.
3.6. Modelling of the hydrothermolysis of rapeseed cake, using optimal experimental design Reaction temperature and holding time were the reaction parameters to affect the rate of hydrolysis of rapeseed cake and the production of free amino acids and fatty acids in subcritical water. On the basis of the results obtained in preliminary studies, in the second stage of this work, two experimental regions (Fig. 4) were explored using Doehlert matrices with reaction temperature and holding time as
Table 5 e Doehlert experimental design with effective (ui) variables and values of corresponding experimental and calculated responses (YA in WS fractions) for hydrothermolysis of rapeseed cake.
Table 6 e Doehlert experimental design with effective (ui) variables and values of corresponding experimental and calculated responses (YFFA in ES fractions) for hydrothermolysis of rapeseed cake.
Experimental series 1
Experimental series 2
Experiment 1 2 3 4 5 6 7 70 c 700 c
u1 T (oC)
u2 t (min)
YA-expa (g kg1)
YA-cal.b (g kg1)
240 180 225 195 225 195 210 210 210
30 30 60 0 0 60 30 30 30
125.8 112.9 111.0 55.6 102.2 118.9 128.3 136.1 135.1
130.1 108.9 106.7 59.9 97.9 123.2 133.2 133.2 133.2
Experiment 1 2 3 4 5 6 7 70 c 700 c
a
a
b
b
Experimental values. Calculated values. c Experiments repeated at the centre of the experimental region to calculate the standard deviation on the response: sYA ¼ 7.0.
u1 T (oC)
u2 t (min)
YFFA-exp.a (g kg1)
YFFA-cal.b (g kg1)
260 200 245 215 245 215 230 230 230
40 40 80 0 0 80 40 40 40
0.80 0.65 0.92 0.60 0.70 0.71 0.85 0.85 0.86
0.83 0.62 0.90 0.63 0.67 0.73 0.86 0.86 0.86
Experimental values. Calculated values. c Experiments repeated at the centre of the experimental region to calculate the standard deviation on the response: sYFFA ¼ 0.006.
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
57
Fig. 5 e Two- (a, c) and three-dimensional (b, d) representations of the yield of amino acids YA (a, b) in WS fractions and of the conversion of triacylglycerols to fatty acids YFFA (c, d) in ES fractions, as a function of reaction temperature and holding time during the hydrothermolysis of rapeseed cake.
effective variables (ui). In experimental series 1, reaction temperature was limited to the range from 180 to 240 C, with holding times up to 60 min. These experiments aimed at optimizing the amino acid content in the WS fractions. Experimental series 2 was aimed at optimizing the conversion of triacylglycerols leading to free fatty acids in the ES fractions at reaction temperatures from 200 to 260 C and holding times of 0e80 min. The conditions of the experiments carried out in the defined experimental regions and the obtained values of the experimental responses: yield of amino acids in series 1 and the conversion of triacylglycerols to fatty acids in series 2, are listed in Tables 5 and 6, respectively. The results of experimental series 1 show that experiments performed at relatively low reaction temperature (195 C) and shortest holding time (0 min) led to the lowest value of YA. Yields higher than 100 g kg1 may be reached also at lower temperatures (180 and 195 C) but choosing longer holding times (30 and 60 min respectively, Exps. 2 and 6). However,
performing the reaction simultaneously at high reaction temperature and long holding time resulted in a drop of YA (Exp. 3). Experimental series 2 was focused on higher reaction temperatures (up to 260 C) and longer holding times (up to 80 min). The conversion of triacylglycerols to fatty acids (YFFA) increased with both reaction temperature and holding time (Exps. 2e7); however, in the investigated experimental region responding to simultaneously highest reaction temperature and long holding time, a drop of fatty acid yield was observed (Exp. 1). Attempts to fit the experimentally determined amino acid yield and the conversion of triacylglycerols to fatty acids to a quadratic model (Eq. (5)) led to satisfactory results (predicted values are consistent with experimental ones, within experimental errors). The following quadratic models were calculated by multi-linear least-square regression from the results obtained from experimental series 1 (Eq. (6)) and series 2 (Eq. (7)):
58
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
Table 7 e Overview of hydrothermolysis of selected biomass origin substrates leading to amino acids and fatty acids. Raw material
Experimental conditions for the best yield of desired product
Yield of main product
20 g kg1 of dry mass 9.14 g kg1 of raw material 19.65 g kg1 of raw material 325 g kg1 protein
[15]
0.86 g kg1 of dry mass 755 g kg1 of raw silk 203 g kg1 of raw silk 4.35 g kg1 of dry mass 83 g kg1 of raw material
[34] [38] [38] [35] [25]
Rapeseed cake
330 C, 30 min, with addition of CO2 220 C, 30 min 200 C, 30 min 250 C, 60 min, substrate concentration of 10 g L1 230 C, 30 min, water:feed ratio ¼ 9 220 C, 60 min, water:feed ratio ¼ 50 160 C, 60 min, water:feed ratio ¼ 20 200 C, 30 min 264 C, 29 min, with addition of 0.83 M sodium bicarbonate 215 C, 36 min
135.9 g kg1 of dry mass
Present study
Fatty acids Coconut oil Linseed oil Soybean oil Squid wastes Sunflower oil Rapeseed cake
270 280 270 240 350 246
>97% conversion >97% conversion >97% conversion 440 g kg1 oil 92.8% conversion 91% conversion
[36] [36] [36] [44] [42] Present study
Amino acids Bean dregs Deoiled rice bran Deoiled soybean Hog hair Lawn grass clipping Silk waste-fibroin Silk waste-sericin Water lettuce Whey protein isolate
YA ¼ 133:2 þ 10:8x1 þ 20:8x2
13:8x21
C, C, C, C, C, C,
15 15 20 40 15 65
min min min min min min
43:7x22
31:5x1 x2
(6)
In experimental series 1 the standard error for coefficients b0, b1, b2 was 4.0, for b11, b22 6.4, and for b12 8.1. YFFA ¼ 0:9 þ 0:1x1 þ 0:1x2 0:1x21 0:1x22 þ 0:1x1 x2
(7)
In experimental series 2 the standard error for coefficients b0, b1, b2 was 0.003, for b11, b22 0.005 and for b12 0.006. Two- (a, c) and three-dimensional (b, d) representations of YA (a, b) in the WS fractions and YFFA in the ES fractions (c, d) as a function of reaction temperature and holding time, calculated using the above models are shown in Fig. 5. The contour plots in Fig. 5a and c represent the curves of constant value of the estimated responses YA and YFFA using the quadratic models. In the investigated experimental region, the optimal range of protein hydrolysis leading to amino acids was predicted at reaction temperatures of 200e230 C and holding times from 46 to 25 min, respectively. The maximum value of amino acids was 135.9 g kg1 (3.94) predicted at 215 C and 36 min. The obtained results are in relatively good agreement with results published in the work of Luo et al. [34], they employed surface methodology to analyse the influence of the water: feed ratio, reaction temperature and holding time on hydrothermolysis of lawn grass clippings leading to obtain amino acids. The highest yield of amino acids was obtained with water:feed ration of 5.5 at 230 C and 30 min. The maximum region for the conversion of triacylglycerols to fatty acids was estimated for reaction temperatures and holding times from 240 to 252 C and 74 to 55 min, respectively, with a maximum conversion of 0.91 (0.005) at 246 C and 65 min. In contrast to our results, Johnson et al. [45] studying microalgae hydrothermolysis, determined that short holding time (shorter as 30 min) with high temperatures
Reference
[31] [28]
(300e350 C) maximized the yield of fatty acids before degradation occurred. As shown, the hydrothermolysis of rapeseed cake in the optimal region proceeded with high yields of both amino acids and fatty acids in comparison with literature data (Table 7). Since the hydrothermal biomass conversion to valuable chemicals significantly depends on composition of applied raw material, higher yields were obtained only for the processes in which the raw materials were pure protein or oil substances. Owing to the fact that rapeseed cake has been found as an attractive raw material for production of both amino and fatty acids in the separate processes, it would be advantageous to perform the hydrothermolysis in a two-step process, as proposed by Tavakoli and Yoshida [20,21] who hydrolysed respectively squid entrails and scallop viscera to simultaneously amino acids and oil fraction rich in free fatty acids. The aim of the first step of rapeseed cake hydrothermolysis could be the production of amino acids as the primary product. On the basis of the present study we assess that the optimal conditions for amino acid production are appropriate for the first step of the two-step process, since the hydrolysis of triacylglycerols under these conditions occurred only with the 0.77 conversion. As the post-reaction residue still contained unreacted fat components it might be used as the raw material for the production of fatty acids in the second step.
4.
Conclusions
It was found that rapeseed cake due to its high content of protein and fat components is useful raw material for production of amino acids and fatty acids in the hydrothermolysis process under subcritical conditions. Given the
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
complexity of the process, the optimal experimental design methodology was chosen to elucidate the dependence of the yield of amino and fatty acids on the two reaction parameters (temperature and holding time). On the basis of the OED experiments it was found that the yield of amino acids reached highest values equal to approximately 134.1 g kg1 at ranges of reaction temperatures from 200 to 232 C and holding times of 46e25 min respectively, and the predicted maximum value of amino acids was 135.9 g kg1 at 215 C and 36 min. The region of highest conversion degree of triacylgycerols to fatty acids was found at ranges of reaction temperatures from 240 to 252 C and holding times of 74e55 min respectively, with the maximum conversion degree of 0.9, and estimated maximum conversion degree was 0.91 at 246 C and 65 min. Since the optimal parameters of amino acid production are significantly lower than for fatty acids (their conversion in the region optimal for amino acids reached only 0.77 conversion), it is worth considering carrying out the hydrothermolysis in a twostep process: the first step is hydrolysis of proteins to amino acids, and the second step is hydrolysis of the remaining solid post-reaction residue containing the fat component to fatty acids. After isolation and proper purification, the bioproducts, amino acids and fatty acids, obtained from the hydrothermolysis of the rapeseed cake, may be used for various aims of industrial interest.
Acknowledgements The authors acknowledge the financial support of this work (project No. N N523 494134) provided by the Ministry of Science and Higher Education of Poland during the years 2008e2010. The authors also thank Prof. Dr. Andre´ M. Braun (Karlsruhe Institute of Technology, Germany) and Dr. Irena JacukowiczSobala (University of Economics in Wrocław, Poland) for their valuable critical comments and suggestions.
[5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.biombioe.2014.03.028.
[19]
[20]
references [21] ¨ zc¸imen D, Karaosmanog lu F. Production and [1] O characterization of bio-oil and biochar from rapeseed cake. Renew Energy 2004;29(5):779e87. [2] Ucar S, Ozkan AR. Characterization of products from the pyrolysis of rapeseed oil cake. Bioresour Technol 2008;99(18):8771e6. [3] Giannakopoulou K, Lukas M, Vasiliev A, Brunner C, Schnitzer H. Low pressure catalytic co-conversion of biogenic waste (rapeseed cake) and vegetable oil. Bioresour Technol 2010;101(9):3209e19. [4] Giannakopoulou K, Lukas M, Vasiliev A, Brunner C, Schnitzer H. Conversion of rapeseed cake into bio-fuel in a
[22]
[23]
[24]
59
batch reactor: effect of catalytic vapour upgrading. Micropor Mesopor Mater 2010;128(1e3):126e35. Kandefer S, Baron J, Olek M. Co-burning of rapecake and hard coal in a grate furnace. Ochr Pow Pr Odp 2006;42(3):69e76 [in Polish]. Akiya N, Savage PE. Roles of water for chemical reactions in high-temperature water. Chem Rev 2002;102(8):2725e50. Kruse A, Dinjus E. Hot compressed water as reaction medium and reactant. Properties and synthesis reactions. J Supercrit Fluids 2007;39(3):362e80. Kruse A, Dinjus E. Hot compressed water as reaction medium and reactant. 2. Degradation reactions. J Supercrit Fluids 2007;41(3):361e79. Peterson AA, Vogel F, Lachance RP, Froling M, Antal MJ, Tester JW. Thermochemical biofuel production in hydrothermal media: a review of sub- and supercritical water technologies. Energy Environ Sci 2008;1(1):32e65. Zhu G, Zhu X, Fan Q, Wan X. Recovery of biomass wastes by hydrolysis in sub-critical water. Resour Conserv Recycle 2011;55(4):409e16. Toor SS, Rosendahl L, Rudolf A. Hydrothermal liquefaction of biomass: a review of subcritical water technologies. Energy 2011;36(5):2328e42. Pavlovic I, Knez Z, Skerget M. Hydrothermal reactions of agricultural and food processing wastes in sub- and supercritical water: a review of fundamentals, mechanisms, and state of research. J Agric Food Chem 2013;61(34):8003e25. Cheng H, Zhu X, Zhu C, Qian J, Zhu N, Zhao L, et al. Hydrolysis technology of biomass waste to produce amino acids in sub-critical water. Bioresour Technol 2008;99(9):3337e41. Esteban MB, Garcı´a AJ, Ramos P, Ma´rquez MC. Kinetics of amino acid production from hog hog hair by hydrolysis in sub-critical water. J Supercrit Fluids 2008;46(2):137e41. Esteban MB, Garcı´a AJ, Ramos P, Ma´rquez MC. Sub-critical water hydrolysis of hog hair for amino acid production. Bioresour Technol 2010;101(7):2472e6. Yoshida H, Terashima M, Takahashi Y. Production of organic acids and amino acids from fish meat by sub-critical water hydrolysis. Biotechnol Progr 1999;15(6):1090e4. Quitain AT, Sato N, Daimon H, Fujie K. Production of valuable materials by hydrothermal treatment of shrimp shells. Ind Eng Chem Res 2001;40(25):5885e8. Daimon H, Kang K, Sato KN, Fujie K. Development of marine waste recycling technologies using sub- and supercritical water. J Chem Eng Jpn 2001;34(9):1091e6. Quitain AT, Faisal M, Kang K, Daimon H, Fujie K. Lowmolecular-weight carboxylic acids produced from hydrothermal treatment of organic wastes. J Hazard Mater 2002;B93(2):209e20. Yoshida H, Tavakoli O. Sub-critical water hydrolysis treatment for waste squid entrails and production of amino acids, organic acids, and fatty acids. J Chem Eng Jpn 2004;37(2):253e60. Tavakoli O, Yoshida H. Conversion of scallop viscera wastes to valuable compounds using sub-critical water. Green Chem 2006;8(1):100e6. Lamoolphak W, Goto M, Sasaki M, Suphantharika M, Muangnapoh Ch, Prommuag Ch, et al. Hydrothermal decomposition of yeast cells for production of proteins and amino acids. J Hazard Mater 2006;B137(3):1643e8. Lamoolphak W, De-Eknamkul W, Shotipruk A. Hydrothermal production and characterization of protein an amino acids from silk waste. Bioresour Technol 2008;99(16):7678e85. Rogalinski T, Herrmann S, Brunner G. Production of amino acids from bovine serum albumin by continuous sub-critical water hydrolysis. J Supercrit Fluids 2005;36(1):49e58.
60
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
[25] Espinoza AD, Morawicki RO. Effect of additives on subcritical water hydrolysis of whey protein isolate. J Agric Food Chem 2012;60(20):5250e6. A, Gu¨lbay S, Uskan B, Canel M. Biomass decomposition [26] Sinag in near critical water. Energ Convers Manage 2010;51(3):612e20. [27] Sereeewatthanawut I, Prapintip S, Watchiraruji K, Goto M, Sasaki M, Shotipruk A. Extraction of protein and amino acids from deoiled rice bran by subcritical water. Bioresour Technol 2008;99(3):555e61. [28] Watchararuji K, Goto M, Sasaki M, Shotipruk A. Value-added subcritical water hydrolysate from rice bran and soybean meal. Bioresour Technol 2008;99(14):6207e13. [29] Pourali O, Asghari FS, Yoshida H. Sub-critical water treatment of rice bran to produce of valuable materials. Food Chem 2009;115(1):1e7. [30] Sunphorka A, Chavasiri W, Oshima Y, Ngamprasertsith S. Kinetic studies on rice bran protein hydrolysis in subcritical water. J Supercrit Fluids 2012;65:54e60. [31] Zhu G, Zhu X, Fan Q, Liu X, Shen Y, Jiang J. Study of amino acids from bean dregs by hydrolysis in sub-critical water. Chin J Chem 2010;28(10):2033e8. [32] Barreiro DL, Prins W, Ronsse F, Brilman W. Hydrothermal liquefaction (HTL) of microalgae for biofuel production: state of the art review and future prospects. Biomass Bioenergy 2013;53:113e27. [33] Garcia-Moscoso JL, Obeid W, Kumar S, Hatcher PG. Flash hydrolysis of microalgae (Scenedesmus sp.) for protein extraction and production of biofuels intermediates. J Supercrit Fluids 2013;82:183e90. [34] Luo G, Cheng X, Shi W, Strong PJ, Wang H, Ni W. Response surface analysis of the water:feed ratio influences on hydrothermal recovery from biomass. Waste Manage 2011;31(3):438e44. [35] Luo G, Shi W, Chen X, Ni W, Strong PJ, Jia Y, et al. Hydrothermal conversion of water lettuce biomass at 473 or 523 K. Biomass Bioenergy 2011;35(11):4855e61. [36] Holliday RL, King JW, List GR. Hydrolysis of vegetable oils in sub- and supercritical water. Ind Eng Chem Res 1997;36(3):932e5. [37] King JW, Holliday RL, List GR. Hydrolysis of soybean oil in a subcritical water flow reactor. Green Chem 1999;1(6):261e4. [38] Pinto JSS, Lanc¸as FM. Hydrolysis of corn oil using subcritical water. J Brazil Chem Soc 2006;17(1):85e9. [39] Minami E, Saka S. Kinetics of hydrolysis and methyl esterification for biodiesel production in two-step supercritical methanol process. Fuel 2006;85(17e18):2479e83. [40] Kocsisova T, Juhasz J, Cvengros J. Hydrolysis of fatty acids esters in subcritical water. Eur J Lipid Sci Tech 2006;108(8):652e68. [41] Moquin PHL, Temelli F. Kinetic modeling of hydrolysis of canola oil in supercritical media. J Supercrit Fluids 2008;45(1):94e105. [42] Alenezi R, Leeke GA, Santos RCD, Khan AR. Hydrolysis of sunflower oil under subcritical water conditions. Chem Eng Res Des 2009;87(6):867e73. [43] Milliren AL, Wissinger JC, Gottumukala V, Schall CA. Kinetics of soybean oil hydrolysis in subcritical water. Fuel 2013;108:277e81. [44] Tavakoli O, Yoshida H. Squid oil and fat production from squid wastes using subcritical water hydrolysis: free fatty acids and transesterification. Ind Eng Chem Res 2006;45(16):5675e8. [45] Johnson MC, Tester JW. Lipid transformation in hydrothermal processing of whole algal cells. Ind Eng Chem Res 2013;52(32):10988e95. [46] Tavakoli O, Yoshida H. Conversion to valuable compounds using sub-critical water. Green Chem 2006;8(1):100e6.
[47] Ruiz HA, Rodrı´guez-Jasso RM, Fernandes BD, Vecente AA, Teixeira JA. Hydrothermal processing, as an alternative for upgrading agriculture residues and marine biomass according to the biorefinery concept: a review. Renew Sust Energ Rev 2013;21:35e51. [48] Undersander D, Mertens DR, Thiex N. Forage analyses procedures. Omaha, Nebrasca: The National Forage Testing Association; 1993 Jul. p. 147. [49] Hames B, Scarlata C, Sluiter A. Determination of protein content in biomass. revised 2008, May. Golden, Colorado: National Renewable Energy Laboratory; 2008. p. 8. Report No. TP-510-42625. [50] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D, et al. Determination of structural carbohydrates and lignin in biomass. Golden, Colorado: National Renewable Energy Laboratory; 2008 Apr. p. 18. revised 2011, Jul; Report No. TP510-42618. [51] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D. Determination of ash in biomass. Golden, Colorado: National Renewable Energy Laboratory; 2005 Jul. p. 8. Report No. TP510-42622. [52] Van Soest PJ, Robertson JB, Lewis BA. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci 1991;74(10):3583e97. [53] Carrier M, Loppinet-Serani A, Denux D, Lasnier JM, HamPichavant F, Cansell F, et al. Thermogravometric analysis as a new method to determine the lignocellulosic composition of biomass. Biomass Bioenerg 2011;35(1):298e307. [54] Lisiewska Z, Kmiecik W, Korus A. The amino acid composition of kale (Brassica oleracea L. var. acephala), fresh and after culinary and technological processing. Food Chem 2008;108(2):642e8. ski P, Sobczyn ska L. Amino [55] Lisiewska Z, Kmiecik W, Ge˛bczyn acid profile of raw and as-eaten products of spinach (Spinacia oleracea L.). Food Chem 2011;126(2):460e5. [56] Standard test method for acid value of fatty acids and polymerized fatty acids. West Conshohocken, Pennsylvania: American Society for Testing and Materials; 1998. p. 2. ASTM D1980-87. [57] Standard test method for determination of the saponification value of fats and oils. West Conshohocken, Pennsylvania: American Society for Testing and Materials; 2006. p. 2. ASTM D5558-95. [58] Atkinson AC, Donev AN. Optimum experimental design. Oxford: Oxford University Press; 1992. [59] Rasch D, Verdooren LR, Gowers JI. Fundamentals in the design and analysis of experiments and surveys- Grundlagen der Plannung und Auswertung von Versuchen und Erhebungen. Mu¨nchen: Oldenburg Wissenschaftsverlag; 1999. [60] Ferreira SLC, Dos Santos WNL, Quintella CM, Neto BB, Bosque-Sebdra JM. Doehlert matrix. A chemometric tool for analytical chemistry e review. Talanta 2004;63(4):1061e7. [61] Sato N, Quitain AT, Kang K, Gaimon H, Fujie K. Reaction kinetics of amino acid decomposition in high-temperature and high-pressure water. Ind Eng Chem Res 2004;43(13):3217e22. [62] Klingler D, Berg J, Vogel H. Hydrothermal reactions of alanine and glycine in sub- and supercritical water. J Supercrit Fluids 2007;43(1):112e9. [63] Goto M, Obuchi R, Hirose T, Sakaki T, Shibata M. Hydrothermal conversion of municipal organic waste into resources. Bioresour Technol 2004;93(3):279e84. [64] Dı´az MJ, Cara C, Ruiz E, Romero I, Moy M, Castro E. Hydrothermal pre-treatment of rapeseed straw. Bioresour Technol 2010;101(7):2428e35.
b i o m a s s a n d b i o e n e r g y 6 4 ( 2 0 1 4 ) 5 0 e6 1
A, Gu¨lbay S, Uskan B, Gu¨llu¨ M. Comparative studies of [65] Sinag intermediates produced from hydrothermal treatments of sawdust and cellulose. J Supercrit Fluids 2009;50(2):121e7. [66] Carvalheiro F, Esteves MP, Parajo JC, Pereira H, Girio FM. Production of oligosaccharides by autohydrolysis of brewery’s spent grain. Bioresour Technol 2004;91(1):93e100. [67] Peterson AA, Lachance RP, Tester JW. Kinetic evidence of the Maillard reaction in hydrothermal biomass processing: glucoseeglycine interactions in high-temperature, highpressure water. Ind Eng Chem Res 2010;49(5):2107e17.
61
[68] Tehrani KA, Kersiene M, Adams A, Venskutonis R, De Kimpe N. Thermal degradation studies of glucose/glycine melanoidins. J Agric Food Chem 2002;50(14):4062e8. [69] Abdelmoez W, Nakahasi T, Yoshida H. Amino acids transformation and decomposition in saturated subcritical water conditions. Ind Eng Chem Res 2007;46(16):5286e94. [70] Sohn M, Ho CT. Ammonia generation during thermal degradation of amino acids. J Agric Food Chem 1995;43(12):3001e3.