Studies of process intensification for the recovery of high molecular weight beta-glucan from cereals

Resumen

In the recent years there has been an increase in the concern for the control of diseases such as cholesterol, diabetes or obesity by means of natural products. Those natural products that claim to have positive effects on human health are known as nutraceuticals. (1-3),(1-4)-b-D-glucan, a sort of soluble fiber or dietary fiber, is an example of nutraceutical according to the claims issued by the FDA (Federal Drug Administration, USA; FDA, 2005) and the EFSA (European Food Safety Authority). More specifically b-glucans are a kind of non-starchy long and linear polysaccharides composed of b-D-glucose units. Glucose molecules are linked by means of the O-glucosidic bond: a ß-(1-4) in the 70% of the cases that is separated by a single ß-(1-3) linkage. The structure of the b-glucan in shown in figure 1. This polysaccharide can be found in the cell walls of different cereals, such as oat, barley, rye and, in lesser extent, wheat. The range of concentrations of b-glucans in these cereals is 0.5-11% (Cui et al, 2011) nevertheless the presence in cereals such as wheat can be considered as residual (0.5-2%). The aim of this PhD Thesis is to make studies in order to develop an efficient process to produce concentrates rich in controlled molecular weight b-glucans, using cereals as raw material. It is expected to obtain high molecular weight b-glucans that can be used e. g. as food additive and to obtain low molecular weight b-glucans to be used e. g. in active compounds formulation processes. This work was started with the selection of the raw material. It was used barley developed by the Instituto Tecnológico Agrario de Castilla y León (ITACYL; Valladolid, Spain). Two different waxy cultivars from 2006 were used, differing in the composition besides other morphological differences: H13 (hulless, 4.64±0.18% in b-glucan) and D24 (hulled, 6.02±0.37% in b-glucan). In chapter 1, a literature review was done and it could be concluded that in the literature there was not any work in which the different process parameters affecting the b-glucan extraction process were systematically studied. It was selected 55°C as the optimal temperature to extract b-glucan, since a compromise between the b-glucan solubility and the co-extraction of other species present in the barley (mainly starch) was required, and water was selected as solvent. To study the effect that process parameters had on the extraction yield and molecular weight of b-glucan and to find of combination of them that maximize the extraction yield were the aims of this chapter. Particle size (100-250-500um), solvent to solid ratio (5-8-15), stirring rate (200-600-1000rpm) and extraction time (1-3-5hours) were the process parameters studied. After this work it was concluded that particle size had a significantly higher influence on the extraction process than the other parameters studied (stirring rate, extraction time and barley variety), indicating that mass transfer steps are limiting the overall extraction process. The conventional extraction process in the optimal conditions lasted 3 hours (1000rpm stirring rate, solvent to solid ratio of 8 and particle size 100um) had an extraction yield of 62.8% for barley H13 and 73.4% for D24. It was also seen that the molecular weight of the extracted b-glucan was very low, in the range of 45-55kDa, so in order to increase the MW of the b-glucan barley should be pretreated in order to deactivate the endogenous b-glucanases and preserve the native high MW of the b-glucan. Barley was boiled in aqueous ethanol for two hours and used as raw material. The b-glucan extracted from this barley had a significantly higher MW (>400kDa), but the extraction yield decreased dramatically, due to the difficulty to keep the water-barley suspension as a consequence of its high viscosity. Results are shown in annex I. In the annex II, a different pretreatment was tried: barley was milled in order to get different milling fractions, with different compositions and particle size. It was seen the importance of the fraction used as raw material, as different extraction yields and product compositions were obtained. In order to overcome the mass transfer limitations, which could not be solved in the convectional stirred tank process, new extraction processes were proposed. In chapter 2 b-glucans were extracted from barley by means of the ultrasound technology (Ultrasound Assisted Extraction ¿ UAE). Ultrasounds are related to mass transfer intensification processes, increasing the extraction yields and reducing the operation times compared to the conventional processes. In this work pretreated barley was used as raw material. Amplitude of oscillation (60-80-100%), cycle (pulsed or continuous sonication) and extraction time (3-10-17 minutes) were the process parameters studied. Results revealed that UAE was an efficient process to extract b-glucans, providing higher extractions yields than the conventional CSTR process in much shorter extraction times, up to 10 minutes. Moreover, the specific energy required in the extraction was substantially lower than in the conventional process (here it was 1460kJ/L), which allowed to talk about b-glucan extraction intensification. On the other hand, high energy outputs (>500kJ/L) involve an important depolymerization of the b-glucan, not providing increases in the extraction yield. As a consequence of these results, the extraction conditions must be carefully selected in order to maximize extraction yield, molecular weight or both simultaneously. Chapters 3 to 5 are related to the extraction of b-glucans from pressurized hot water (PHW), in different ways: batch or semi-continuous. PHW term refers to the water in liquid state in the range 100°C (boiling point) to 374°C (critical point) by the application of pressure. The transport properties of water (such as density, surface tension, viscosity and diffusivity) change dramatically when changing pressure and, specially, temperature. In chapter 3, a concentrate of b-glucans and starch (36.5% and 25.4%, respectively) was used in order to check its solubility and behavior in the PHW (range of temperature 120 to 170°C and range of times 10 to 300 minutes). It was seen that b-glucan could be dissolved in PHW, but once dissolved a rapid degradation of the molecule occurred: a rapid decrease of the MW was observed as well as the presence of sugar degradation products in the solution. In chapter IV it was done the extraction of b-glucan from not pretreated barley in a pressurized stirred vessel (batch operation mode). Temperature was changed in the range (135-180°C) and extraction times in the range (15-75 minutes). The experimental conditions were selected to maximize the extraction yield: results done at 155°C for 18 minutes at 50bar led to an extraction yield of 52.7% and MW equal to 200kDa. These results are remarkable, as in one single step high molecular weight b-glucan were extracted in a short extraction process: it was not necessary to perform the deactivation of b-glucanases pretreatment. In chapter 5, a flow extraction process was studied, and b-glucan was extracted in a 50ml semicontinuous extractor (fixed bed). b-glucan extraction yield was similar to that obtained in the batch process in a 4ml/min flow rate of water at 155°C for 105minutes. The major advantage of the fixed-bed extraction process was the preservation and the fractionation of the b-glucan in terms of MW during the extraction. The extracted b-glucan were in the range 500-100kDa, decreasing when increasing the extraction length. In this process no sugar degradation products, such as HMF, were detected. The extraction of b-glucan from other raw materials was also studied (wheat bran). It was observed similar results for wheat bran, in terms of temperature. The results obtained were used to calculate dimensionless numbers that would allow to do scale-up processes for the fixed-bed extraction of b-glucan. This research was done at the Technical University Hamburg-Harburg (Hamburg, Germany). In chapter 6 the purification of the extracted b-glucan was studied (downstream process). The use of water and its lack of selectivity makes that other species present in barley, specially starch, are coextracted unpurifying the liquid extract. Starch was hydrolyzed by means of a-amylase: it was studied the hydrolysis length, the temperature and the enzyme dose. The results allowed to do the hydrolysis in lower times than usual (only 9 minutes) keeping removal efficiencies higher than 90% with enzyme doses lower than those used in processes found in the literature: 100uL/g of starch at 55°C. Traditional processes do the starch hydrolysis at 95°C for 1 hour. In a second step, a significant intensification of the process was achieved by dosing the enzyme during the ultrasound assisted extraction step (7 minutes at 55°C), resulting in a starch removal of 90%. An intensification of the extraction and purification processes was successfully accomplished, as they were done simultaneously in one single step and at the same temperature. In chapter 6 was also studied the separation of the dextrins or sugars formed as a consequence on the starch hydrolysis. These molecules are low molecular weight compared to the b-glucan, so a membrane process was used to separate them: diafiltration (polysulfone membrane, MWCO 100kDa) done in a tangential flow cell. This operation way allowed to reduce the concentration of oligosaccharides in the solution by 45% in average (in some cases it was higher than 75%) in short times, with the subsequent increase in the concentration of b-glucan. Despite the fast fouling of the membrane observed (decrease of the permeability value, recovered after the cleaning process) it was concluded that the use of membranes to purify and concentrate b-glucans extracted from barley was technical viable. Finally, in chapter 6 it was concluded that the combination of ultrasounds, enzymatic hydrolysis and membrane technology allowed to obtain high purity b-glucans in simple and easy way, compared to the conventional process (time and energy consuming). In chapter 7 special attention was paid on the energy consumed during the different extraction processes proposed in this thesis. It was evaluated the energy required to pretreat the barley (deactivation of b-glucanases and milling) and to perform the extraction of the three main processes presented in this work: stirred tank extraction, UAE and bath PHW. The energy consumed was expressed in kWh per gram of b-glucan extracted. UAE was seen as the best process, as the lowest energy was consumed (0.14-0.17kWh/g b-glucan in 3 to 15.9 minutes processes) which together with the dramatic reduction of the extraction times lead to talk about a real process intensification compared to the stirred tank extraction (3h, 0.27kWh/g b-glucan). PHW process was seen as the highest energy consumer (6.0kWh/g b-glucan). In spite of the good results in extraction time reductions and the lack of pretreatments to obtain high molecular weight (MW) b-glucan, PHW resulted to be an unfavorable process form an energetic point of view, mainly due to the low barley to water ratio (1:50) used during the extraction.