foods Article Effect of Ultra-High Pressure Homogenization (UHPH) and Conventional Thermal Pasteurization on the Volatile Composition of Tiger Nut Beverage Idoia Codina-Torrella 1,2, * , Joan Josep Gallardo-Chacón 1 , Bibiana Juan 1 , Buenaventura Guamis 1 and Antonio José Trujillo 1, * 1 Centre d’Innovació, Recerca i Transferència en Tecnologia dels Aliments (CIRTTA-UAB), TECNIO, XIA, MALTA-Consolider, Department of Animal and Food Science, Facultat de Veterinària (Edifici V), Universitat Autònoma de Barcelona (UAB), Bellaterra, 08193 Cerdanyola del Vallès, Spain 2 Department of Agri-Food Engineering and Biotechnology, EEABB, Universitat Politècnica de Catalunya (UPC), Parc Mediterrani de la Tecnologia, Campus del Baix Llobregat (Edifici D4), c/Esteve Terradas, 8, 08860 Castelldefels, Spain * Correspondence: idoia.codina@upc.edu (I.C.-T.); toni.trujillo@uab.cat (A.J.T.) Abstract: Tiger nut beverages are non-alcoholic products that are characterized by their pale color and soft flavor. Conventional heat treatments are widely used in the food industry, although heated products are often damaging to their overall quality. Ultra-high pressure homogenization UHPH) is an emerging technology that extends the shelf-life of foods while maintaining most of their fresh characteristics. The present work deals with the comparison of the effect of conventional thermal homogenization-pasteurization (H-P, 18 + 4 MPa at 65 ◦ C, 80 ◦ C for 15 s.) and UHPH (at 200 and 300 MPa, and inlet temperature of 40 ◦ C), on the volatile composition of tiger nut beverage. Headspace-solid phase microextraction (HS-SPME) was used for detecting volatile compounds of beverages, which were then identified by gas chromatography-mass spectrometry (GC-MS). A total of 37 different volatile substances were identified in tiger nut beverages, which were primarily grouped into the aromatic hydrocarbons, alcohols, aldehydes and terpenes chemical families. Stabilizing Citation: Codina-Torrella, I.; treatments increased the total amount of volatile compounds (H-P > UHPH > R-P). H-P was the Gallardo-Chacón, J.J.; Juan, B.; treatment that produced the most changes in the volatile composition of RP, while treatment at Guamis, B.; Trujillo, A.J. Effect of 200 MPa had a minor impact. At the end of their storage, these products were also characterized by Ultra-High Pressure Homogenization the same chemical families. This study evidenced the UHPH technology as an alternative processing (UHPH) and Conventional Thermal of tiger nut beverages production that minimally modifies their volatile composition. Pasteurization on the Volatile Composition of Tiger Nut Beverage. Keywords: tiger nut beverage; ultra-high pressure homogenization (UHPH); pasteurization; volatile Foods 2023, 12, 683. https://doi.org/ profile; headspace-solid phase microextraction; gas chromatography-mass spectrometry 10.3390/foods12040683 Academic Editor: Osman Erkmen Received: 29 December 2022 Revised: 1 February 2023 1. Introduction Accepted: 2 February 2023 Tiger nut beverages are non-alcoholic products obtained from the aqueous extract Published: 4 February 2023 of tiger nut tubers (Cyperus esculentus L.), which are produced and consumed worldwide. One of the most consumed is “horchata de chufa”, a Spanish traditional beverage that is characterized by its pale color and soft flavor [1]. These complex dispersions are char- acterized by a high percentage of carbohydrates (>50%), fat (~2%), fiber (~1%), and a Copyright: © 2023 by the authors. limited content of protein (~1%) [1,2]. Nowadays, the consumption of vegetal beverages is Licensee MDPI, Basel, Switzerland. increasing at the global level, so the beverage industry is focused on improving the safety This article is an open access article and shelf-life of these highly perishable products. Conventional thermal pasteurization distributed under the terms and and ultra-high temperature sterilization (UHT) are the most applied treatments for the conditions of the Creative Commons microbiological and physicochemical stabilization of marketable tiger nut beverages with Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ an extended shelf-life. However, according to the severity of these treatments, the fresh 4.0/). product results in an undesirable loss of its most appreciated sensory characteristics, such Foods 2023, 12, 683. https://doi.org/10.3390/foods12040683 https://www.mdpi.com/journal/foods Foods 2023, 12, 683 2 of 12 as its pale color and almond-like flavor and taste. After heating, different unpleasant flavors appear in these beverages because of the biochemical reactions activated by temperature (i.e., Maillard reaction, caramelization processes, or fat-oxidation reactions, among others), which in some cases, cause significant inconveniences to the consumers. Some authors have characterized commercial thermal-pasteurized and UHT tiger nut beverages and reported that the volatile profiles of these beverages were mainly composed of aldehydes, alcohols, terpenes, and aromatic compounds [3–5]. Off-flavors related to some of these components are pungent, oxidized, burnt, or bitter, among others, the reason why the industry tends to mask them through the addition of different flavorings [6] and look for novel stabilizing processes that preserve the overall characteristics of the untreated product, as well. Ultra-high pressure homogenization (UHPH) is a novel technology that has been demonstrated to allow the microbial and physicochemical stability of different fluid food- stuffs while maintaining the most nutritional and sensory characteristics of the fresh product [7]). Codina-Torrella et al. [8,9] demonstrated the industrial relevance of UHPH to improve the overall quality and shelf-life of tiger nut beverages. These authors also observed that UHPH caused lesser changes in the sensory profile of the raw beverage (color, flavor, taste, etc.) if compared with the conventional thermal-pasteurization processing. To the best of our knowledge, no study exists to determine the effect of UHPH on the volatile profile of tiger nut beverages. The limited existing data about UHPH-treated vegetal bev- erages reported that, in general, UHPH processing causes fewer changes in the volatile composition of these products than the observed in beverages treated with conventional thermal technologies [10,11] To complement the studies conducted to date, this study aimed to determine the effect of conventional thermal-pasteurization and UHPH treatments on the volatile profile of tiger nut beverages. This study would allow the industry to have a better understanding of which aromatic compounds are generated during the storage of these beverages, to evaluate which could be the most suitable treatment of pasteurization to preserve the original volatile profile of this product. At the same time, this work has generated more knowledge about the impact of UHPH technology on the volatile profile of vegetal beverages. 2. Materials and Methods 2.1. Tiger Nut Beverages Production and Processing Tiger nuts beverages were produced and processed at the Pilot Plant of Universitat Autònoma de Barcelona (SPTA-UAB), as described by Codina-Torrella et al. [7]. The overall experiment was performed in triplicate. 2.1.1. Tiger Nut Beverages Production The proportion of tubers:water corresponded to 1:8 (w:w). After pressing and filter- ing the ground product, 8% of sucrose (w/w) was mixed with the liquid extract. This mixture was considered the raw product (RP), which general composition (%, w/w) corre- sponded to 12.99 ± 0.18 total solids, 10.30 ± 0.60 nitrogen-free materials, 2.01 ± 0.02 fat, 0.54 ± 0.02 protein, and 0.13 ± 0.01 ash. The composition of RP was not affected by the treatment applied. Before the application of all stabilizing treatments, 0.05% of α-amylase enzyme (Bialfa, Biocon Española, S.A., Franqueses del Vallès, Spain) was added to the RP (holding time of 10 min, at room temperature), to hydrolyze the starch granules. Qualitative determination of starch (Total Starch Assay Procedure kit, Amyloglucosidase/α-amylase method, K-TSTA 404-2009, Megazyme International Ireland Ltd., Wicklow, Ireland) demon- strated that this component was hydrolyzed. 2.1.2. Beverage Treatments: UHPH, Homogenization-Pasteurization Two different pasteurizing UHPH treatments were performed by using an ultra-high pressure homogenizer, at a flow rate of 120 L/h (Model: DRG No. FPG11300:400 Hygienic Homogenizer, Stansted Fluid Power Ltd., Harlow, UK) at two different pressures, 200 and 300 MPa, and the same inlet temperature (Ti) of 40 ◦ C. The temperature of UHPH- Foods 2023, 12, 683 3 of 12 treated beverages increased by 24.2 ◦ C between pressures ranging from 200 to 300 MPa [8]. Temperature after the UHPH valve corresponded to 92.1 ± 1.7 and 116.3 ± 4.3 ◦ C for the 200 and 300 MPa treatments, respectively, and the residence time of the product at these temperatures was estimated to be <0.7 s. The outlet temperature of products corresponded to 15.3 ± 1.1 and 17.1 ± 1.6 ◦ C in 200 and 300 MPa treatments, respectively. Conventional treatment of Homogenization-Pasteurization (H-P) was also applied to the RP using an indirect heat system composed of a double-stage homogenizer positioned upstream (Model X68, Soavi B. and Figli, S.P.A., Parma, Italy) and a multitube tubular heat exchanger at a flow rate of 1000 L/h (laminar flow) (6500/010, GEA Finnah GmbH, Ahaus, Germany). Beverages were homogenized at pressures of 18 MPa (first stage-valve) and 4 MPa (second stage-valve) at 65 ◦ C, and subsequently pasteurized at 80 ◦ C for a holding time of 15 s. Samples (RP, H-P, 200 MPa, and 300 MPa) were collected in sterile glass bottles of 1 L of capacity with twist-off caps (Apiglass Envases y Material Apícola, S.L., Barcelona, Spain) inside a laminar flow cabin (Mini-V cabin, Telstar Technologies, S.L., Terrassa, Spain) and were stored at refrigeration temperature (4 ◦ C) until their analyses Microbiological shelf-life of stored beverages corresponded to 3, 5, 30 and 57 days for the RP, H-P and UHPH processed beverages at 200 and 300 MPa, respectively, according to Codina-Torrella et al. [9]. 2.2. Procedure of HP-SPME and GC-MS Tiger nut’s beverage samples were evaluated after production and during their pre- viously established shelf-life. Volatile compounds were analyzed following the method optimized by Klein et al. [4], with some modifications. All analyses were performed in triplicate. 2.2.1. HP-SPME Extractions An aliquot of 2 mL of each beverage was placed in different vials with 3 µL of an internal standard (4-methyl-2-pentanol in methanol, 5 ppm). Vials were incubated for 10 min at 40 ◦ C for their stabilization, at continuous homogenization (stirring with a mag- netic stirrer at 700 rpm). After sample stabilization, SPME fiber of 85 µm (DVB/CAR/PDMS, Supelco, Bellefonte, PA, USA) was exposed to the vial headspace, for 30 min at 40 ◦ C, in which volatile compounds were adsorbed. 2.2.2. GC-MS Analysis Adsorbed volatiles were desorbed in the gas chromatograph (GC) injector port, in splitless mode, at 250 ◦ C for 3 min. The split valve was opened, and the fiber was kept in the injector for 15 min for a cleaning step. The headspace of the volatile compounds was analyzed using an automated GC (model: 6890 Series II, Agilent, Santa Clara, CA, USA). The analysis was carried out on a 60 m × 0.25 mm internal diameter capillary column, with a film thickness of 0.25 µm (TRB-Wax, Agilent technologies). The mass spectrometry (MS) selective detector (model: 5972 Agilent, Santa Clara, CA, USA) was used in electron impact ionization mode with a mass range of 30–250 m/z. Before each analysis, the fiber was preconditioned for 1 h at 250 ◦ C. The temperature was programmed in two stages. The initial temperature was kept at 40 ◦ C for 5 min, and then, it increased at the rate of 10 ◦ C/min to the temperature of 250 ◦ C and held for 10 min. Tentative identification of volatile compounds was achieved by comparing their mass spectra with those of the mass spectra libraries Willey 1n.l and NIST 0.5 (National Institute of Standards and Technology). Relative retention times of detected compounds were also determined by injecting 1 µL of alkane standard solutions (Alkane standard solution C8–C20 from Sigma–Aldrich and Connecticut ETPH calibration mixture C9–C36, with purity greater than 95%) in triplicate with a split ratio of 1:200. Signals were processed using Agilent MSD Productivity ChemStation Enhanced Data Analysis software (Agilent, Santa Clara, CA, USA). Confirmation of the identification of hexanal, pentanal, 1-octen- 3-one, 2,3-pentanodione, 1-hexanol, 1-pentanol, 1-octen-3-ol and 2-penthyl furan (Sigma- Foods 2023, 12, 683 4 of 12 Aldrich, St. Louis, MO, USA) was performed by comparing GC retention times and mass spectra of individual components with those authentic reference compounds injected under the same conditions. The results of volatile compounds were expressed as microgram (µg) equivalents of 4-methyl-2-pentanol internal standard per milliliter (mL) of tiger nuts’ milk beverage. The limit of quantification (LOQ) of metabolites was also determined by measuring the average noise and the standard deviation values of 10 blanks (calibration matrix). Average noise plus 10 times standard deviation was used for LOQ. Relative standard deviation (RSD) was calculated for five replicate measurements of each compound solution. Only compounds with RSD <10% were finally considered. 2.3. Statistical Analysis One-way analysis of variance (ANOVA) was performed on volatile compounds, by using the GLM procedure of Statgraphics (Statgraphics Inc., Chicago, IL, USA). Tukey test was used for the data comparison and significant differences were determined at the 5% level of probability. Principal component analysis (PCA) was performed to reduce the data in two dimensions and identify patterns of variation in the results. R software (R software, Auckland, New Zealand) was used for this purpose. In this paper, data showed corresponds to the mean ± standard error. 3. Results and Discussion 3.1. Effect of Treatments on the Volatile Composition of Beverages Analysis of GC-MS of all beverages revealed about 37 different volatile substances in tiger nut beverages (Table 1). RP was characterized by aromatic hydrocarbons and alcohols, which represented ~52.3 and ~36.2% of the total, respectively, followed in importance by aldehydes (~6.4% of the total) and terpenes (~3.7% of the total). The main volatile compounds found in this sample were toluene, ethanol, 1-octanol, 1-nonanol, nonanal, and limonene. The significant presence of aromatic hydrocarbons in tiger nut beverages could be attributed to different origins, such as the breakdown of plant carotenoids, the uptake by plants of chemical substances from the environment, or due to the traditional practice of burning the aerial part of the plant before harvesting of tubers [12–14]. It is also reported in the literature that alcohols are one of the most representative aromatic groups of raw tiger nuts (~68%), followed by aldehydes (~6.4%), pirazines (~5.6 %), and terpenes (~4.2%) [15]. As shown in Table 1, the application of stabilizing treatments to the RP increased the total amount of volatile compounds in this sample, according to H-P > UHPH > R-P (p < 0.05). H-P beverage was characterized by aromatic hydrocarbons (~48.7%) and alcohols (~37.5%), followed by aldehydes (~6%) and another minor groups of components (~7.3%) represented by terpenes, furans, ketones, esters, phenolic compounds, and acids. Some of these compounds were probably formed during the sample exposure to the temperature, which enhances lipid oxidation and browning reactions [16]. UHPH-treated beverages at 200 and 300 MPa were also characterized by aromatic hydrocarbons (50.3 and 57.5%, respectively) and alcohols (41.7 and 32.2%, respectively), but in this case, terpenes took the third place in importance (5.9 and 5.7%, respectively). Comparing both UHPH-treated beverages, higher amounts of ketones and aldehydes were observed in the 300 MPa sample, which was mainly attributed to the synergic effect of the higher temperature reached after the high-pressure valve (~92 and ~116 ◦ C, respectively in 200 and 300 MPa treatments) and the homogenization pressure onto the food matrix’s components. 3.1.1. Main Groups of Volatile Patterns: Aromatic Hydrocarbons, Alcohols and Aldehydes Aromatic hydrocarbons were the most detected compounds of beverages, according to H-P ≥ 200 MPa = 300 MPa ≥ RP (Table 1). In all samples, toluene was the most repre- sentative compound, followed by the isomers of xylene, ethylbenzene and styrene. On the contrary, m-cymene was only identified in UHPH-treated beverages. These components, which have been associated with earthy and musty off-flavors, are described as an impor- Foods 2023, 12, 683 5 of 12 tant contaminant group of processed food, and their formation is related to heating and several processing techniques [17,18]. Table 1. Concentration of volatile compounds (µg of 4 methyl-2-pentanol/mL) in tiger nut beverages, grouped by their chemical family. Treatment 4 Family Group Compound ID 1 KI 2 KI lit. 3 RP H-P 200 MPa 300 MPa Aromatic Toluene MS, IR, S 1053.0 1042 3.419 ± 0.961 b 9.835 ± 1.567 ab 6.182 ± 1.949 ab 7.424 ± 0.738 ab Hydrocarbons Ethylbenzene MS, IR, S 1132.2 1124 0.028 ± 0.002 b 0.048 ± 0.009 ab 0.086 ± 0.018 a 0.089 ± 0.007 a p-xylene MS, IR, S 1141.4 1150 0.013 ± 0.001 c 0.199 ± 0.050 a 0.036 ± 0.003 b 0.068 ± 0.003 b m-xylene MS, IR 1146.3 1150 0.071 ± 0.006 b 0.216 ± 0.039 a 0.247 ± 0.048 a 0.289 ± 0.018 a o-xylene MS, IR 1191.9 1182 0.036 ± 0.002 c 0.167 ± 0.030 b 0.114 ± 0.020 b 0.213 ± 0.017 a Styrene MS, IR 1256.5 1273 0.024 ± 0.012 c 0.061 ± 0.011 ab 0.076 ± 0.024 a 0.030 ± 0.003 b m-cymene MS, IR, S 1288.4 1267 200 MPa = 300 MPa > RP. These compounds are broadly used as quality markers of oils because their formation is associated with fatty acid oxidation, among others [19]. The most representative alcohols of tiger nut beverages were 1-nonanol, 1-octanol and ethanol (Table 1). UHPH samples also presented a high content of 1-hexanol, which might originate from the linoleic acid degradation, and, as reported in the literature, their presence is associated with bitter and floral aromas [20]. Homogenized sample at 200 MPa was characterized by the highest content of 1-octen-3-ol, which is associated with mushroom aromas [16]. Concerning the aldehydes, three different compounds were detected in the bev- erages evaluated in this study, which corresponded to hexanal, octanal and nonanal. Cantalejo (1997) [15] had previously reported that total aldehydes represented the ~6.4 % of total volatile compounds in tiger nuts, of which benzaldehyde was the most abundant (5%), followed by hexanal, nonanal, and octanal. Nonanal was the most abundant compound of RP, H-P, and 200 MPa beverages (H-P > RP > 200 MPa), with values around 89, 89, and 97.8% of total aldehydes, respectively. On the contrary, the most representative aldehyde in the 300 MPa sample was hexanal (~48% of the total), followed by octanal (~31% of the total). As observed, the H-P treatment caused the most significant increase in the total amount of aldehydes, which could be explained by the highest lipid oxidation reactivity due to heat processing (Table 1). The 200 MPa-treated beverages presented the lowest values, which suggested that this treatment could improve the oxidative stability of the RP. The presence of these compounds is commonly related to grassy, green, and beany flavors [21–23]. Although UHPH-treated products might be expected to be oxidized faster (due to the increase in the number of fat particles and the subsequent increase of total fat surface exposed to oxidation [24], new interactions created during the UHPH process between denatured proteins and the other components of the dispersion might result in a significant protective effect against fat droplet’s oxidation [25]. Differences observed between both UHPH-treated samples (200 and 300 MPa) were probably attributed to the higher temperature reached during the treatment [8]. Technological treatments increased substantially (p < 0.05) the presence of total ter- penes in samples, not showing significant differences between treated beverages. Terpenes are ubiquitous compounds of vegetables, which are produced during their metabolism to fight pests and other diseases [26]. Limonene was the most relevant terpene in all samples, and which content increased significantly (p < 0.05) after processing. In line with this, other authors also detected limonene as the predominant terpene in tiger nut beverages [4,6]. Badui (2006) [27] suggested that heat treatments could affect the compounds that keep emulsified limonene, which therefore let it release. The increase of terpenes in UHPH bev- erages could be related to the effect of pressure on limonene liberation. To a lesser extent, other terpenes were detected in these beverages, such as β-pinene in both UHPH-treated beverages, α-pinene and γ-terpinene in RP and UHPH beverages, and 1-α-terpineol in H-P and UHPH samples. The presence of these compounds is associated with pine, citric, or musty aromas [27,28]. 3.1.2. Secondary Compounds of Beverages: Ketones, Acids, Esters, Phenolic Compounds, and Furans Five different ketones were detected in treated tiger nut beverages (Table 1). Their highest presence was detected in the 300 MPa sample, whereas no differences were observed between H-P and 200 MPa samples (p > 0.05). Contents of 6-methyl-5-hepten-2-one and 1-phenylethanone were not affected by the treatment (p > 0.05), while 2-nonanone and 1-etanone contents increased after 300 MPa and H-P processes. Ketones are described to be derived from Maillard, Strecker, and lipid oxidative reactions [29,30], and their presence had been described in tiger nuts by-products after roasting [15,31]. In this study, ketones were not detected in the RP, probably due to their under-representation in raw tubers Foods 2023, 12, 683 7 of 12 (~0.68 of total volatile compounds). Klein et al. (2014) [4] also identified eight different ketones in conventional pasteurized and sterilized tiger nut beverages, the amount of which also increased according to the severity of the heat treatment. In the literature, these volatile compounds had been related to kindly aromas (caramel, sweet, fruit-like, or buttery) but also with mushroom or green-beany unpleasant flavors [16,30]. Acids, esters, phenolic compounds, and furans corresponded to the minor groups of volatile compounds in all samples, and their prevalence increased (p < 0.05) after H-P, if compared with beverages treated by UHPH (Table 1). Three different acids (hexanoic, benzoic, and butanoic) were identified in the RP, of which benzoic and butanoic increased significantly (p < 0.05) after H-P treatment. Hexanoic acid was not detected in the H-P bev- erage. On the contrary, UHPH treatment at 200 MPa did not cause significant differences (p < 0.05) in the percentage of the total acids, in comparison to the RP (only butanoic acid was detected in this sample), and no acids were detected in the beverage homogenized at 300 MPa. Among others, the contribution of acids in food aroma has been related to cheese and acidic odors [16]. Concerning the esters, ethylcaprylate was only detected in the sample treated by heat pasteurization. It had been previously reported the low preva- lence of esters in raw tiger nuts (1.45% of the total), of which ethylcaprylate represented ~0.1% of the total [15]. In the current study, only one furan (2-pentyl furan) was detected in RP and H-P samples, according to RP < H-P (p < 0.05) (Table 2). Lower amounts of phenolic compounds were identified in all treated beverages. H-P beverages showed the highest content, probably due to the heat degradation of phenolic acids from tiger nuts (such as tannins). 4-vinil-2-metoxyphenol was identified in all beverages, according to RP > UHPH ≥ RP. On the contrary, phenol-2-metoxy and phenol were only detected in H-P and UHPH samples, respectively. According to the literature, the presence of phenolic acids in food is related to green and harsh grassy odors [32,33]. Furans are widely associated with heating, which origin is reported to be related to Maillard reactions and the oxidation processes of unsaturated fatty acids [23]. Cantalejo (1997) [15] isolated different furans in raw tiger nuts, which increased in number and concentration after the tuber’s roasting, and Klein et al. (2014) [4] also identified four types of furans in different commercial UHT Tiger nut beverages. No furans were detected in both UHPH-treated beverages, probably due to the lesser effect of temperature during this process, in comparison to the H-P. Among others, these compounds had been related to buttery, almond-like, sweet, and green bean-like odors [34,35]. Table 2. Percentage variance and loading accounted by the first two principal components (PC1 and PC2) of tiger nut beverages’ volatile profile. Principal Components 1 Name PC1 PC2 Aromatic Hydrocarbons Toluene 0.045 0.248 Ethylbenzene 0.189 0.148 p-xylene −0.062 0.236 m-xylene 0.178 0.182 o-xylene 0.152 0.204 Styrene −0.081 0.070 m-cymene −0.165 0.135 1,2,4-trimethylbenzene 0.130 0.071 Naftalene 0.065 0.196 Aldehydes Hexanal 0.205 0.122 Octanal 0.123 0.196 Nonanal −0.207 0.117 Foods 2023, 12, 683 8 of 12 Table 2. Cont. Principal Components 1 Name PC1 PC2 Alcohols Ethanol −0.169 0.127 1-hexanol 0.122 0.216 2-octanol −0.184 0.100 1-octen-3-ol 0.184 0.180 1-heptanol 0.052 −0.036 1-octanol −0.165 0.085 1-nonanol −0.016 0.271 Phenolic Compounds Phenol-2-metoxy −0.164 0.154 Phenol 0.196 0.063 4-vinil-2-metoxyphenol −0.221 0.106 Ketones 3-octanone 0.205 0.127 6-methyl-5-hepten-2-one −0.108 0.070 2-nonanone 0.221 0.096 1-phenylethanone 0.014 0.244 1-etanone −0.157 0.162 Terpenes α- pinene 0.232 −0.056 β- pinene 0.208 0.067 Limonene 0.080 0.243 γ-terpiene 0.046 −0.215 l-α–terpineol −0.023 0.219 Acids Hexanoic −0.061 −0.209 Benzoic −0.219 0.076 Butanoic −0.161 0.171 Furans 2-pentil furan −0.206 0.121 Esters Etilcaprilate −0.148 0.178 Variance explained (%) 39 32 1 PC1: First Principal Component, PC2: Second Principal Component. 3.2. Principal Component Analysis (PCA) Figure 1 shows the distribution of samples in the principal components 1 (PC1) and 2 (PC2). This clustering method reduces the dimensionality of multivariate data and preserves the variance therein [36]. In this study, PC1 and PC2 explained 71% of the global variability of the dataset (Table 2). As observed in Figure 1, samples were perfectly distributed along these two axes in three main groups differentiated by the stabilizing process. PC1 and PC2 scores separate the different samples of beverages (Table 2). As shown in Figure 1, UHPH samples could be grouped in the same group, although these two samples also presented some differences. PC1 could be entirely related to the treat- ment. The highest negative scores (which represent strong influence) in PC1 are detected in some aromatic hydrocarbons, two alcohols (ethanol and 2-octanol), 4-viny-2-metoxyphenol, 2-pentil-furan and nonanal, which are related to heat-damage effects and separate treated Foods 2023, 12, x FOR PEER REVIEW 9 of 13 Foods 2023, 12, 683 9 of 12 Variance explained (%) 39 32 1 PC1: First Principal Component, PC2: Second Principal Component. 3.2. Principal beverages Component throughAnalysis (PCA) the axis. In this component, large positive loadings in two ketones (3-octanone andthe Figure 1 shows 2-nonanone) distribution and 1-octen-3-ol of samples were principally in the principal influenced components by 300 and 1 (PC1) and 2 (PC2). This clustering method reduces the dimensionality of multivariate data and pre- (α and 200 MPa treatments, respectively, while high loadings for hexanal and two terpenes serves the variance β-pinene) weretherein linked[36]. withInthe this study, effect of PC1 UHPH. andConcerning PC2 explained 71% PC2, of the large globalscores in positive variability of the dataset (Table 2). As observed in Figure 1, samples were perfectly dis- aromatic hydrocarbons, aldehydes, two alcohols (1-hexanol, 1-nonanol), 1-phenylethanone, tributed and along thesewere limonene two identified axes in three main groups as markers differentiated of treated samples by theand (H-P stabilizing UHPH).pro- Large nega- cess.tive scores of γ-terpiene and hexanoic acid were related to the RP, which contributed to differentiating this untreated beverage from the others. Figure 1. Loadings plot after principal component analysis of the individuals in the plane defined by Figurethe1.two Loadings plot aftercomponents first principal principal component (PC1 and analysis of the(39%): PC2). PC1 individuals in the plane First Principal defined which Component, by the two first principal components (PC1 and PC2). PC1 (39%): First Principal Component, explains 39% of total variability; PC2 (32%): Second Principal Component, which explains which 32% of explains 39% of total variability; PC2 (32%): Second Principal Component, which explains 32% of total variability; RP: raw product; H-P: homogenization-pasteurization at 18 + 4 MPa (pressure of total variability; RP: raw product; H-P: homogenization-pasteurization at 18 + 4 MPa (pressure of ◦ C for 15 s; 200 MPa: ultra-high pressure first stage first stage valve valve + second + second stageand stage valve) valve) and80 65, and 65,°Cand 80 s; for 15 200 MPa: ultra-high pressure ho- mogenization at 200 MPa homogenization at and 200 Ti of 40and MPa °C; Ti 40 ◦ultra-high 300ofMPa: C; 300 MPa:pressure homogenization ultra-high pressure at 300 MPa homogenization at and Ti300ofMPa 40 °C.and Ti of 40 ◦ C. PC1 According and PC2 scores separate to their the different volatile profile, samples 200 MPaofand beverages (Table 2).would RP beverages As shown be the most in Figure 1, UHPH similar samples products, couldby followed bethe grouped 300 MPa in the andsame the group, althoughCodina-Torrella H-P samples. these two sam- et al. [8] ples reported also presented some that previously differences. 200 MPa PC1 could becaused treatment entirelylesser related to the in changes treatment. The the physicochemical highest negative scores characteristics (color,(which represent viscosity, oxidationstrong influence) reactions) in PC1 of RP, are detected compared in someobserved with changes aromatic in thehydrocarbons, raw base aftertwo 300alcohols MPa and (ethanol and 2-octanol), 4-viny-2-metoxyphenol, 2- H-P treatments. pentil-furan and nonanal, which are related to heat-damage effects and separate treated 3.3. Changes beverages through inthe the axis. Volatile Profile In this of Beverages component, during large Theirloadings positive Storage in two ketones (3- octanone Figure and 2-nonanone) 2 shows the and 1-octen-3-ol evolution werefamily of each principally groupinfluenced of volatilebycompounds 300 and 200of stored MPabeverages. treatments,Therespectively, whileof volatile profile high RPloadings for hexanalsince was not evaluated and its twoshelf-life terpenesonly (α and β- corresponded pinene) to 3 were days. linked Duringwith theirthe effect the storage, of UHPH. Concerning total amount PC2,compounds of volatile large positive scoresa in showed significant aromatic hydrocarbons, decrease (p < 0.05) inaldehydes, two alcohols all pasteurized beverages. (1-hexanol, Secondary 1-nonanol), products 1-phenyleth- generated from lipid anone, and limonene oxidation reactions, were identified as browning non-enzymatic markers ofreactions treated samples (Maillard(H-Pandand UHPH). caramelization), and Largethenegative scores of γ-terpiene microbiological evolution of and hexanoic each sample acid were related probably to thethe explained RP,differences which con-observed tributed to differentiating between these beverages this untreated and their beverage from the corresponding others. after production. At the end homologs According to their volatile profile, 200 MPa and of the storage time, the aromatic profiles of tiger nut beverages RP beverages would werebe the most by the represented similar products, followed by the 300 MPa and the H-P samples. aromatic hydrocarbons, alcohols, aldehydes, and terpenes families. In contrastCodina-Torrella et al. [8] to H-P and reported previously that 200 MPa treatment caused lesser changes in the physicochemical 200 MPa samples, the alcohols increased in 300 MPa beverage at the end of the storage time. These changes were probably attributed to the reaction of aldehydes and ketones derived from lipid oxidative reactions that occurred in the sample over time. Aldehydes decreased in importance in H-P beverages (Figure 2), while in both UHPH beverages, their content increased significantly (p < 0.05). Homogenized beverages at 300 MPa showed the highest content of aldehydes. Our previous research [9] suggested faster oxidation reactions in 300 MPa samples, probably due to the increase of total fat surface by the increase in the Foods 2023, 12, 683 10 of 12 number of oil droplets after the homogenization process. Terpenes only increased in the 300 MPa sample (Figure 2). Concerning the minority groups of volatile compounds, total furans and phenolic compounds increased in all beverages, although their content in the H-P sample was significantly higher than in their homologs. On the contrary, ketones, acids, Foods 2023, 12, x FOR PEER REVIEW 11 of 13 and esters decreased in all beverages over time (Figure 2), which demonstrated that these aromatic family groups only were involucrate in the sensory profile of these beverages at the beginning of their shelf-lives. Figureof2. volatile Figure 2. Evolution Evolutioncompounds of volatile compounds in tiger in tiger nut nut beverages beverages during during theirtheir storage storage atat44◦°C. C. a–d For each a–d For each group ofgroup familyofcompounds, family compounds, valuesvalues with different with different letters letters differed differed significantly significantly (p (p < 0.05). H-P: < 0.05). homogenization-pasteurization H-P: homogenization-pasteurization at 18 +at4 18 + 4 MPa MPa (pressure (pressure of first of first stagevalve stage valve ++ second secondstage stagevalve) at 65, and ◦ 80 °C for 15 s (-); 200 MPa: ultra-high pressure homogenization at 200 MPa and Ti = 40 °C valve) at 65, and 80 C for 15 s (-); 200 MPa: ultra-high pressure homogenization at 200 MPa (….); 300 MPa: ultra-high pressure homogenization at 300 MPa and Ti = 40 °C (----); Concentration: and Ti = 40 ◦ C (μg. . equivalents . ); 300 MPa:of 4-methyl-2-pentanol ultra-high pressure internal standard per mL homogenization of tiger at 300 MPa nuts’ andmilk Ti = 40 ◦ C beverage. (—-); Concentration: µg equivalents of 4-methyl-2-pentanol internal standard per mL of tiger nuts’ milk beverage. 4. Conclusions In this study, a total of 37 different volatile substances were identified in tiger nut 4. Conclusions beverages. Raw product (RP) was characterized by aromatic hydrocarbons and alcohols, which In this study, a represented ~52.3 and ~36.2% total of 37 different volatileofsubstances the total, respectively, followed were identified inintiger importance nut by beverages. Raw aldehydes product(~6.4% (RP) wasof the total) and terpenes characterized (~3.7% of by aromatic the total). Theand hydrocarbons application alcohols, of stabi- which representedlizing~52.3 treatments and of homogenization-pasteurization ~36.2% of the total, respectively, (H-P and UHPH) followed increased the total in importance amount of volatile compounds of samples, in comparison to the RP, according to H-P > UHPH > R-P. H-P treatment induced the most important changes in the raw base, while Foods 2023, 12, 683 11 of 12 by aldehydes (~6.4% of the total) and terpenes (~3.7% of the total). The application of stabilizing treatments of homogenization-pasteurization (H-P and UHPH) increased the total amount of volatile compounds of samples, in comparison to the RP, according to H-P > UHPH > R-P. H-P treatment induced the most important changes in the raw base, while treatment at 200 MPa had a minor impact. H-P-treated beverage was characterized by aromatic hydrocarbons (~48.7%) and alcohols (~37.5%), followed by aldehydes (~6%). On the contrary, UHPH-treated beverages at 200 and 300 MPa were characterized by aromatic hydrocarbons (50.3 and 57.5%, respectively) and alcohols (41.7 and 32.2%, respectively), but in this case, terpenes took the third place in importance (5.9 and 5.7%, respectively). Differences observed in both UHPH samples were probably attributed to the synergic effect of pressure and temperature. During the storage time, the total amount of volatile compounds decreased in all beverages. The volatile profile of these products was also characterized by the high content of alcohols, aldehydes, aromatic hydrocarbons, and terpenes. Results obtained in this study evidenced that H-P treatment has a greater impact on the chemical volatile profile of tiger nut beverages if compared with the UHPH process. According to this, UHPH technology could be proposed as an alternative processing of pasteurization in tiger nut beverages production to obtain fresh like products. Author Contributions: Conceptualization, I.C.-T., B.G. and A.J.T.; methodology, I.C.-T., A.J.T., J.J.G.-C. and B.J.; data processing: I.C.-T. and J.J.G.-C.; writing—original draft preparation I.C.-T. and A.J.T.; review and editing: I.C.-T., A.J.T., J.J.G.-C. and B.J.; supervision, A.J.T. All au- thors have read and agreed to the published version of the manuscript. Funding: This research was funded by MALTA-CONSOLIDER project (CONSOLIDER-INGENIO 2010 Program, Ministry of Science and Innovation of Spain, funding number: CSD 2007-00045). Data Availability Statement: Data is contained within the article. Acknowledgments: The authors acknowledge SPTA-UAB staff for their unconditional support and technical advice during the production and processing of tiger nut beverages. Conflicts of Interest: The authors declare no conflict of interest. References 1. CRDO Consejo Regulador de la, D.O. Horchata de Chufa de Valencia. 2022. Available online: http//www.chufadevalencia.org/ (accessed on 3 November 2022). 2. Alegría-Torán, A.; Farré-Rovira, R. Horchata y Salud: Aspectosnutricionales y Dietéticos; Jornada Chufa y Horchata: Tradición y Salud; Fundación Valenciana de Estudios Avanzados, Ed.; Consellería de Agricultura, Pesca y Alimentación: Valencia, Spain, 2003; pp. 55–70. 3. Clemente-Villalba, J.; Cano-Lamadrid, M.; Issa-Issa, H.; Hurtado, P.; Hernández, F.; Carbonell-Barrachina, A.A.; López-Lluch, D. Comparison on sensory profile, volatile composition and consumer’s acceptance for PDO or non PDO tigernut (Cyperus esculentus L.) milk. LWT 2021, 140, 110606. [CrossRef] 4. Klein, B.; Gallardo-Chacón, J.J.; Codina-Torrella, I.; Trujillo, A.J.; Juan, B. Evaluation of volatile compounds of “Tiger nut beverage” (Orxata de xufla) headspace by optimized solid-phase micro-extraction. OALib J. 2014, 1, 1–15. [CrossRef] 5. Ezeh, O.; Gordon, M.H.; Niranjan, K. Tiger nut oil (Cyperus esculentus L.): A review of its composition and physico-chemical properties. Eur. J Lipid Sci Technol. 2014, 116, 783–794. [CrossRef] 6. Morell, J.; Barber, S. Chufa y Horchata: Características Físicas, Químicas y Nutritivas; Valencia: C.S.I.C. Instituto de Agroquímica y Tecnología de Alimentos: Valencia, Spain, 1983. 7. Dumay, E.; Chevalier-Lucia, D.; Picart-Palmade, L.; Benzaria, A.; Gràcia-Julià, A.; Blayo, C. Technological aspects and potential applications of (ultra) high-pressure homogenisation. Trends Food Sci. Technol. 2012, 31, 13–26. [CrossRef] 8. Codina-Torrella, I.; Guamis, B.; Ferragut, V.; Trujillo, A.J. Potential application of ultra-high pressure homogenization in the physico-chemical stabilization of tiger nuts’ milk beverage. Innov. Food Sci. Emerg. Technol. 2017, 40, 42–51. [CrossRef] 9. Codina-Torrella, I.; Guamis, B.; Zamora, A.; Quevedo, J.M.; Trujillo, A.J. Microbiological stabilization of tiger nuts’ milk beverage using ultra-high pressure homogenization. A preliminary study on microbial shelf-life extension. Food Microbiol. 2018, 69, 143–150. [CrossRef] 10. Ferragut, V.; Valencia-Flores, D.C.; Pérez-González, M.; Gallardo, J.; Hernández-Herrero, M. Quality characteristics and shelf-life of ultra-high pressure homogenized (UHPH) almond beverage. Foods 2015, 4, 159–172. [CrossRef] 11. Poliseli-Scopel, F.H.; Gallardo-Chacón, J.J.; Juan, B.; Guamis, B.; Ferragut, V. Characterisation of volatile profile in soymilk treated by ultra high pressure homogenisation. Food Chem. 2013, 141, 2541–2548. [CrossRef] Foods 2023, 12, 683 12 of 12 12. Bansal, V.; Kim, K.-H. Review of PAH contamination in food products and their health hazards. Environ. Int. 2015, 84, 26–38. [CrossRef] 13. Johnson, A.E.; Nursten, H.E.; Self, R. Aromatic hydrocarbons in foodstuffs and related materials. Chemy Ind. 1969, 1, 10–12. 14. Mackiewicz-Walec, E.; Krzebietke, S.J.; Sienkiewicz, S. The Influence of Crops on the Content of Polycyclic Aromatic Hydrocarbons in Soil Fertilized with Manure and Mineral Fertilizers. Int. J. Environ. Res. Public Health 2022, 19, 13627. [CrossRef] 15. Cantalejo, J.M. Analysis of Volatile Components Derived from Raw and Roasted Earth-Almond (Cyperus esculentus L.). J Agric. Food Chem. 1997, 45, 1853–1860. [CrossRef] 16. Lozano, P.R.; Drake, M.; Benitez, D.; Cadwallader, K.R. Instrumental and sensory characterization of heat-induced odorants in aseptically packaged soy milk. J. Agric. Food Chem. 2007, 55, 3018–3026. [CrossRef] 17. Mazza, G.; Pietrzak, E.M. Headspace Volatiles and Sensory Characteristics of Earthy, Musty Flavored Potatoes. Food Chem. 1990, 36, 97–112. [CrossRef] 18. Singh, L.; Varshney, J.G.; Agarwal, T. Polycyclic aromatic hydrocarbons’ formation and occurrence in processed food. Food Chem. 2016, 199, 768–781. [CrossRef] 19. Zhou, Y.; Fan, W.; Chu, F.; Wang, C.; Pei, D. Identification of volatile oxidation compounds as potential markers of walnut oil quality. Food Chem. 2018, 83, 2745–2752. [CrossRef] 20. Yang, X.; Yang, F.; Liu, Y.; Li, J.; Song, H.-L. Identification of Key Off-Flavor Compounds in Thermally Treated Watermelon Juice via Gas Chromatography–Olfactometry–Mass Spectrometry, Aroma Recombination, and Omission Experiments. Foods 2020, 9, 227. [CrossRef] 21. Plutowska, B.; Wardencki, W. Aromagrams-Aromatic profiles in the appreciation of food quality. Food Chem. 2007, 101, 845–872. [CrossRef] 22. Vichi, S.; Pizzale, L.; Conte, L.; Buxaderas, S.; López-Tamames, E. Solid-phase microextraction in the analysis of virgin olive oil volatile fraction: Modifications induced by oxidation and suitable markers of oxidative status. J. Agric. Food Chem. 2003, 51, 6564–6571. [CrossRef] 23. Yuan, S.H.; Chang, K.C. Selected odor compounds in cooked soymik as affected by soybean materials and direct steam injection. J. Food Sci. 2007, 72, 481–486. [CrossRef] 24. Gohtani, S.; Sirendi, M.; Yamamoto, N.; Kajikawa, K.; Yamano, Y. Effect of droplet size on oxidation of docosahexanoic acid in emulsion system. J. Dispers. Sci. Technol. 1999, 20, 1319–1325. [CrossRef] 25. Hebishy, E.; Buffa, M.; Guamis, B.; Blasco-Moreno, A.; Trujillo, A.J. Physical and oxidative stability of whey protein oil-in-water emulsions produced by conventional and ultra high-pressure homogenization: Effects of pressure and protein concentration on emulsion characteristics. Innov. Food Sci. Emerg. Technol. 2015, 32, 79–90. [CrossRef] 26. Pontin, M.; Bottini, R.; Burba, J.L.; Piccoli, P. Allium sativum produces terpenes with fungistatic properties in response to infection with Sclerotium cepivorum. Phytochemistry 2007, 115, 152–160. [CrossRef] 27. Badui, S. Química de los Alimentos, 4th ed.; Pearson Educación: Mexico City, Mexico, 2006. 28. Ayala-Zavala, J.F.; González-Aguilar, G.A.; Del-Toro-Sánchez, L. Enhancing safety and aroma appealing of fresh-cut fruits and vegetables using the antimicrobial and aromatic power of essential oils. J. Food Sci. 2009, 74, 84–90. [CrossRef] 29. Ho, C.T.; Carlin, J.T. Formation and aroma characteristics of heterocyclic compounds in foods. Am. Chem. Soc. USA 1989, 388, 92–104. 30. Wilkens, W.F.; Lin, F.M. Gas chromatographic and mass spectral analyses of soybean milk volatiles. J. Agric. Food Chem. 1970, 18, 333–336. [CrossRef] 31. Lasekan, O. Volatile constituents of roasted tigernut oil (Cyperus esculentus L.). J. Sci. Food Agric. 2012, 93, 1055–1061. [CrossRef] [PubMed] 32. Achouri, A.; Boye, J.I.; Zamami, Y. Soybean variety and storage effects on soymilk flavor and quality. J. Food Qual. 2008, 30, 731–744. [CrossRef] 33. Wang, Z.H.; Dou, J.; Macura, D.; Durance, T.D.; Nakai, S. Solid phase extraction for GC analysis of beany flavors in soymilk. Food Res. Int. 1997, 30, 503–511. [CrossRef] 34. Bressanello, D.; Liberto, E.; Cordero, C.; Rubiolo, P.; Pellegrino, G.; Ruosi, M.R.; Bicchi, C. Coffee aroma: Chemometric comparison of the chemical information provided by three different samplings combined with GC-MS to describe the sensory properties in cup. Food Chem. 2017, 214, 218–226. [CrossRef] 35. Neffat, M.; Sriti, J.; Hamdaoui, G.; Kchouk, M.E.; Marzouk, B. Salinity Impact on Fruit Yield, Essential Oil Composition and Antioxidant Activities of Coriandrum sativum Fruit Extracts. Food Chem. 2009, 124, 221–225. [CrossRef] 36. Lee, B.M.; Shim, G.A. Dietary exposure estimation of benzo[a]pyrene and cancer risk assessment. J. Toxicol. Environ. Health 2007, 70, 1391–1394. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.