Development of an integrated and green biorefinery from winery waste. Application to wine lees and grape stems

  1. Romero Díez, Rut
Supervised by:
  1. María José Cocero Alonso Director
  2. Ana Alexandra Figueiredo Matias Director
  3. Soraya Rodríguez Rojo Director

Defence university: Universidad de Valladolid

Fecha de defensa: 13 December 2018

  1. María Luisa González San José Chair
  2. Alexandre de Almeida Paiva Babo de Almeida Paiva Secretary
  3. Alexander Navarrete Muñoz Committee member

Type: Thesis


Nowadays, the valorisation of residues generated in the agro-food industry for the recovery of high added values compounds, is one of the most important fields of study. A clear example of this, is the vinification grape-derived sub-products. Due to the fact that the wine sector is one of the most important activities all around the world, grapes are one of the most cultivates fruit all over the world. However, the vinification process generates huge amount of waste. Only in Spain, around 2-3 million tons of residues are generated annually. In other words, 25kg of sub-products are produced per 100L of red wine. Depending on the vinification step in which they are generated, different residues can be distinguished. The most abundant waste is the grape pomace since it represents the 62% of the total residues generated in a winery. Grape pomace is constituted by grape pulps, skins and seeds. It is obtained after the crushing of the grapes performed to get the must. The second most abundant residue are wine lees (14%). Lees are the dregs that settled at the bottom of the vessels after the wine fermentation or aging processes. Finally, grape stems and dewatered sludge are the least abundant sub-products, each of them represent a 12%. The former, are the woody skeleton of the grape bunches and they are obtained after the destemming process, just before the grape crushing. Grape stems, as grape pomace, has a lignocellulosic composition. This means that, apart from their high content of phenolic compounds, cellulose, hemicellulose and lignin are present in relative significant concentrations (up to 25%). Consequently, grape stems are also a rich source of these polymers that can be also hydrolysed into their monomeric sugars components and, posteriorly, converted in high added value products such as fuels, bulk chemicals and materials. Finally, the dewatered sludge comes from the washing and disinfection steps. Although these wastes are considered as “non-hazardous” residues, they present a high organic load. Therefore, if they are not treated properly they can cause several environmental problems that are partially phytotoxic effects and/or soil acidification. This elevated content of organic matter, is to a large extent due to their high concentration of polyphenols. Polyphenols present more than one aromatic ring with several hydroxyl groups as substituents. They present recognized health-promoting effects due to their antioxidant, antimicrobial, anti-inflammatory and cardio protective properties. Thanks to this, they are compounds with a strong potential application in food, cosmetics, and pharmaceutical industries. Thus, several and different processes have been developed for the recovery of polyphenols from grape sub-products, especially grape pomace. Though, few literature is available regarding the extraction of high-added-value compounds from either wine lees or grape stems. Additionally, as it was previously mentioned grape stems are a rich source of biopolymers and sugars which can be further converted into bulk materials. For this reason, this thesis is focused on the valorisation in terms of high added value compounds of both wine lees and grape stems by alternative greener methodologies. The general idea was to maximize and intensify the extraction of high added value compound from these two residues in terms of polyphenols, and also sugars and oligomers for the case of grape stems. To do so, a total of five chapters were developed in this thesis. In Chapter I the identification of the main phenolic compounds present in aging wine lees extracts, mainly anthocyanins, was carried out. Moreover, correlations between them and different antioxidant activities were developed. In Chapter II and III, the kinetic extraction curves of anthocyanins from different types of wine lees were achieved in order to obtain the parameter values (solid-liquid ratio, solvent composition and temperature) that maximized the extraction of anthocyanins. Once these parameters were obtained, several pre-treatments were applied before the conventional solid-liquid extraction so as to intensify the extraction process. These pre-treatments involved the use of microwaves (MW), ultrasounds (US) and enzymes. Like in the previous chapter, the antioxidant activities of the different extracts was determined and main compounds were identified. The valorisation of grape stems was done in Chapters IV and V. In Chapter IV, the extraction kinetics of total polyphenols and flavonoids from grape stems were studied in the same way that for wine lees. Similarly MW were applied as a pre-treatment in order to enhance the polyphenols recovery. Regarding Chapter V, it was focus on the obtention of extracts rich in sugars and oligomers using a hydrothermal hydrolysis. Temperature and operating time were the variables studied in this chapter. In Chapter I aging wine lees were characterized in terms of total phenolic (TPC, as gallic acid equivalents GAE) and total flavonoid contents (TFC, as catechin equivalents, CAE). Aging wine lees come from the step in which wine is aged inside wood barrels. Once the solid phase of the wine lees were freeze-dried, they were subjected to several solid-liquid extractions. These extractions were carried out using a solid-liquid ratio of 0.025 (0.25g of dry lees in 10mL of solvent), stirring for 5min at room temperature followed by 10 min of sonication in a ultrasonic bath. Solvents with different polarities (water, methanol, ethanol, two hydroalcoholic mixtures and acetone) were used. Total phenolic (TPC) and total flavonoid contents (TFC) were determined and expressed per gram of dried extract (DE). The mixture of 75% vol. ethanol showed the highest values with 254mgGAE/gDE and 146mgCATE/gDE, respectively. To determine the antioxidant activities (AA) of the extracts, different assays were studied. The highest HORAC (as catechin equivalents, CAE), HOSC and FRAP values (both as Trolox equivalents, TE) were obtained for the hydroalcoholic mixture with a 75% vol. ethanol (4,690µmolCAE/gDE, 4,527µmolTE/gDE and 2,197µmolTE/gDE, respectively). In contrast, for ORAC method (as Trolox equivalents, TE), methanol extract showed the best value with 2,771µmolTE /gDE. Furthermore, correlations between TPC, TFC, individual phenolic compounds and the different AA were determined. Most relevant compounds contributing to AA were identified using data from mass spectrometry. It could be asserted that anthocyanins were the major compounds present in the wine lees extracts. Those anthocyanins which presented the 6-p-coumaroyl moiety strongly contributed (p<0.10) to FRAP, as well as, gallic acid and the two flavan-3-ols detected. Besides, anthocynins significantly (p<0.10) contributed to ORAC in a negative way. Anthocyanins also contributed negatively and positively (but not significantly) to HORAC and HOSC, respectively. Depending on the solvent used different amounts of the individual compounds were extracted which could have higher or lower activity against oxygen radicals (ROO•) or (HO•) affecting the antioxidant capacity of the extracts. Although, there are some few works in literature regarding the recovery of phenolic compounds from wine lees, none of them have studied the extraction kinetics and the parameters that maximize their extraction through conventional solid-liquid process. Furthermore, the use of pre-treatments such as microwaves (MW) or ultrasound (US) seemed to be suitable alternatives to intensify the process and to increase the polyphenols recovery. Thus, in Chapter II, firstly the extraction kinetics of anthocyanins (AC, as malvidin equivalents, MALE…) from different wine lees in conventional solid-liquid extraction were studied. AC were chosen as target compounds, since they are the main sub-family of polyphenols found in grapes. The influence of parameters such as temperature, solid-liquid ratio (RS-L) and type of solvent (hydro-alcoholic mixtures) was also considered. Best parameter values chosen to get most AC out from Port win lees (2.78mgMALE/gDRY-LEES) were: a temperature of 25ºC, with a RS-L of 0.10 (g/mL) and with a 50%vol. ethanol mixture. After 15min of extraction, a steady AC content was achieved. Same conditions were also applied to first fermentation and second fermentation wine lees. Final AC yields of 3.04mgMALE/gDRY-LEES and 2.09mgMALE/gDRY-LEES were obtained, respectively. Once the extraction kinetics were studied, the application of MW and US as pre-treatments to the conventional extraction in order to increase AC yield were assessed. With the help of a statistical surface response design, the optimum conditions which maximize the final AC content of the extracts were obtained. Evaluated parameters in this design were the RS-L (g/mL), the solvent composition (% vol. ethanol) and the pre-treatment time (s). The optimal values for each parameter were: a RS-L of 0.140 (g/mL), a hydro-alcoholic mixture of 40% vol. ethanol and a time pre-treatment of 90s. When MW were used at these conditions, AC extraction yield was doubled for Port wine lees (6.20mgMALE/gDRY-LEES) and the required time to achieve a constant yield was reduced from 15min to 90s, since no further increase in AC yield was observed in the subsequent conventional extraction at previously selected conditions. MW pre-treatment applied to first and second fermentation wine lees, increases the anthocyanin yield 1.50 and 1.40 times, respectively. Meanwhile, US only shortened extraction time in less proportion (from 15 to 5min). Putative identification of main extract compounds was performed by LC/MS-MS. It was interesting to notice the identification of one pyranoanthocyanin (Vitisin A) in all the different types of wine lees. Apart from the pre-treatments studied in Chapter II, it was thought that the application of an enzymatic hydrolysis pre-treatment to wine lees will increase the release of polyphenols that may be linked or absorbed in the cell wall of yeast, one of the major components of wine lees. Therefore, in Chapter III, two different enzymes, Glucanex and Mannaway were tested for this purpose and, additionally, a blend of both enzymes. Incubation times for the enzymatic hydrolysis to take place, ranged from 5 to 60min. No significant differences were found in the final AC extraction yield between them. The extract obtained when the enzymatic blend was used with an incubation time of 5min, was the one which showed the highest increment (50%) in the AC extraction yield respect the conventional extraction. On the other hand, antioxidant activity of the enzymatic extracts was evaluated via ORAC assay. The highest ORAC value was achieved for extract from first fermentation wine lees treated with Mannaway. Anthocyanins were identified via mass spectrometry and no differences in composition were detected respect to the conventional extract. The sole difference was found after pre-treatment with Mannaway with an un-identified compound detected at 280nm. Hitherto it was the work developed for wine lees. Concerning grape stems, a similar procedure was followed. In Chapter IV, an intensification of the polyphenol extraction process was also performed. Firstly, the extraction kinetics of total polyphenols (TPC) and flavonoids (TFC) content were studied following the same procedure exposed in Chapter II for wine lees. A RS-L of 0.10g/mL, a solvent with a 50% vol. of ethanol and 75ºC were selected as the best conditions for the recovery of bioactives. At these conditions, a TPC of 38 ± 1mgGAE/gDS and a TFC of 38 ± 1mgCATE/gDS were achieved. Since the use of MW as pre-treatment significantly increase the recovery of phenolic compounds from wine lees, this procedure was also proposed to grape stems. In this case, MW pre-treatment increased the extraction yield of TPC and TFC in a 19% and 24% respect to the conventional solid-liquid extraction, respectively. Additionally, identification and quantification of the major stilbenes (resveratrol and ɛ-viniferin) and flavonoids (catechin and epicatechin) was carried out with HPLC-DAD-MS/MS. According to literature, grape stems extracts have interesting anti-fungal properties. Thus, activity against Botrytis cinarea was determined together with the antioxidant activity, measured by ORAC assay. Putting aside the recovery of phenolics compounds, Chapter V was focused in valorization of grape stems as source of cellulose, hemicellulose and lignin due to their lignocellulosic composition, as it was aforementioned. A hydrothermal process for grape stems conversion into sugars and oligomers was employed for this purpose. Studied temperatures ranged from 100ºC to 180ºC during 20-minute experiments in order to evaluate its influence on sugars recovery. Finally, a temperature of 140ºC seemed to be the most suitable to maximize the sugars yield (264mg/gDS). Besides, the effect of the operating time was also determined at this temperature. Operational times from 10 to 30min were tested. Results brought out that the higher the time, the higher the yield in terms of sugars and total carbon content. In addition, within the biorefinery concept, the effect of the MW pre-treatment carried out in Chapter IV for the extraction of polyphenols was evaluated on the conversion of grape stems into oligomers and sugars. Extractives were removed, as well as the free sugars of the raw material. On the other hand, the effect of the MW which could have an effect on the oligomers, disrupting them and improvement their extraction (10%). Finally, a kinetic model was used to fit the experimental data with an average absolute deviation around 15% for sugars and oligomers and 30% for degradation products.