Improving the quality of pea yogurt: a study of sprouting and enzymatic hydrolysis using multidimensional instrumental analysis and sensory evaluation

Abstract

BACKGROUND

While pea yogurt represents an emerging plant‐based alternative with notable market potential and favorable nutritional attributes, its industrial development is limited by issues such as low acidity and beany flavor. This study employed multiple analytical techniques to systematically assess the influence of sprouting and enzymatic hydrolysis throughout the manufacturing process.

RESULTS

The contents of reducing sugar and trichloroacetic acid‐soluble nitrogen in both pea milk with 1 cm sprout length (Y1PM) and flavored protease‐treated pea milk (FPPM) were significantly increased (P < 0.05), thereby providing carbon and nitrogen sources for lactic acid bacteria fermentation in the pea milk matrix. Additionally, the phytic acid and tannin contents in Y1PM were significantly reduced (P < 0.05), whereas in FPPM antinutritional factors shifted from bound to free forms. Meanwhile, the beany flavor intensity of both pea milks was reduced, leading to an improvement in their flavor quality. Sprouted (SPY) and enzymatically treated pea yogurts (EPY) produced with Y1PM and FPPM substrates were evaluated against an untreated control (UPY). SPY and EPY exhibited lower pH and higher acidity compared with UPY, effectively overcoming the limited acid production typical of pea yogurt. Moreover, the reduction of off‐flavors and increase in aromatic compounds resulted in significantly higher sensory scores for flavor and taste.

CONCLUSION

This study establishes that both sprouting and enzymatic treatments are viable and effective biotechnological strategies for enhancing the overall quality of pea yogurt. The findings provide a theoretical foundation for developing high‐quality pea yogurt and improving the flavor profiles of pea‐based products. © 2026 Society of Chemical Industry.

INTRODUCTION

In response to global health and sustainable development concerns, reduced consumption of animal‐derived foods is being promoted worldwide, concurrently with a steady growth in the plant‐based food market.^{1, 2} Plant‐based milk alternatives (PBMAs) are recognized as an important component within the plant‐based food sector. The global PBMA market was valued at USD 32.38 billion in 2024 and is projected to reach USD 91.15 billion by 2032. ³ PBMAs are defined as emulsions manufactured from one or more plant‐based materials, including cereals, legumes, nuts, and other ingredients, resulting in a wide variety of products. ⁴ Legumes are particularly valued as they are abundant in balanced micronutrients and macronutrients, and are sources of essential B vitamins, minerals, and protein.^{5, 6, 7} Among legumes, peas (Pisum sativum) have gained significant attention due to their high protein and fiber content, richness in minerals, low fat content, low allergenicity, and well‐balanced amino acid profile. ⁸ Furthermore, peas are considered to possess a nutritional advantage over soybeans, primarily due to their lower allergenic potential. ⁹ Consequently, the development of yogurt derived from peas is considered to have substantial market prospects.

However, peas contain strong and persistent off‐flavors described as ‘beany’, ‘earthy’, and ‘green’, which are difficult to eliminate or mask, thereby limiting the development of pea yogurt. ¹⁰ These off‐flavors are primarily derived from volatile compounds generated through the oxidation of unsaturated fatty acids in peas.^{11, 12, 13, 14} Given the limited research on flavor improvement in pea products, strategies for enhancing the flavor of pea yogurt may be informed by pretreatment methods applied to soybean products. Accordingly, the flavor quality of pea yogurt can be modulated through physical methods (e.g., heat treatment), chemical methods (e.g., addition of organic solvents), or biological methods (e.g., fermentation).^{15, 16, 17} Among these, sprouting and enzymatic treatments represent biological strategies that better align with healthy consumption trends compared to physical and chemical approaches. Furthermore, sprouting and enzymatic treatments have been demonstrated to reduce the content of antinutritional factors in pea‐based products.^{18, 19} Nevertheless, research on the use of sprouting and enzymatic treatments to modulate the flavor quality of PBMAs remains limited. The effect of sprouting on off‐flavors such as ‘beany’ notes varies significantly among different legume types. For instance, sprouting has been reported to improve the flavor profile of fava bean flour, whereas it enhances woody and mushroom off‐flavor in lupin flour. ²⁰ Moreover, the impact of sprouting is influenced by its duration. Extended sprouting time in lentil and yellow pea flours increases the content of beany flavor‐related volatile compounds, whereas in chickpea flour, although beany notes are reduced, other off‐flavors may emerge. ²¹ Although beany flavor intensifies during the sprouting of yellow peas, the associated flavor carriers may become destabilized, potentially facilitating their removal in subsequent processing steps. ²¹ While sprouting time and substrate properties are known to influence flavor development during sprouting, their combined effect on the flavor profile of peas remains unclear.

The structure of proteins is modified by proteases, which alters their binding patterns with flavor compounds and consequently modulates the release or retention of flavors during hydrolysis.^{22, 23} Studies have shown that the hydrolysis of pea protein with alkaline protease promotes the release of esters and ketones, while aldehydes and disulfides are retained. Similarly, the hydrolysis of soybean flour reduces the content of alcohols such as 1‐octen‐3‐ol and 1‐hexanol. ²⁴ Protease hydrolysis may reduce hydrophobic binding with aromatic compounds, leading to their release, or, alternatively, expose more active sites that adsorb additional flavor molecules. ²⁵ However, the specific effects of proteolytic hydrolysis on the flavor of pea milk and the resulting pea yogurt remain unclear.

Therefore, this study aims to: (1) systematically evaluate the impact of sprouting on pea milk quality using a combination of physicochemical analysis and molecular sensory science; (2) determine the effects of enzymatic hydrolysis on pea milk quality through the aforementioned suite of analytical techniques; (3) compare the quality of the three pea yogurts by assessing their fermentation properties, volatile profiles, and sensory attributes to elucidate the mechanism by which sprouting and enzymatic pre‐treatments enhance the final product quality. The findings are expected to provide novel strategies for flavor regulation of PBMAs and enable targeted quality enhancement of pea yogurt.

MATERIALS AND METHODS

Samples and chemicals

Pea (Pisum sativum L.), named Zhongwan No. 6, was used in this study (Weifang City, Shandong Province, China), and the samples were stored at 4 °C after removing impurities.

Chromatographic‐grade n‐hexane (purity ≥97.0%) was purchased from MREDA (Beijing, China). The chromatographic‐grade n‐alkanes (C₇–C₄₀) were supplied by Sigma‐Aldrich (St Louis, MO, USA). Chromatographic‐grade 2‐methyl‐3‐heptanone (95%) was sourced from TCI (Shanghai, China). Liquid nitrogen was sourced from Xian Heyu Trading Co., Ltd (Beijing, China). He and N₂ (99.9992%) were produced by AP BAIF Gases Industry Co., Ltd (Beijing, China). The analytical‐grade sodium hydroxide, concentrated sulfuric acid, methyl red, bromothymol blue, anhydrous ethanol, hydrochloric acid, glucose, and boric acid are all sourced from Sinopharm Chemical Reagent Co., Ltd (Beijing, China). The analytical‐grade trichloroacetic acid (TCA) was provided by the Macklin Co., Ltd (Shanghai, China). Commercial direct injection starter was from Bioglan (Shanxi, China). Commercial proteases including alkaline protease, papain and flavor protease were purchased from Novozymes (China).

Preparation of sprouted peas

Sprouted peas were prepared based on the method described by Akkad et al., ²⁰ with minor modifications. The dried peas were weighed, washed, and soaked in distilled water (1:3, w/w) at 25 °C for 10 h. After discarding the soaking water, the peas were incubated in darkness at 25 °C. During incubation, the distilled water was replaced every 4 h. Sprouted peas with sprout lengths of 0, 1, 3, 5, and 7 cm were collected and designated as Y0, Y1, Y3, Y5, and Y7, respectively, for subsequent experiments.

Preparation of sprouted pea milk

Sprouted pea yogurt (SPY) was prepared based on the method of Yang et al., ²⁶ with slight modifications. Briefly, after washing and draining, the sprouted peas were weighed and mixed with seven times their mass of distilled water. The mixture was then ground at high speed for 3 min using a blender (model P165, Joyoung, Jinan, China). The slurry was filtered through double‐layered gauze and centrifuged at 1290 × g for 15 min at 4 °C to remove insoluble material. The supernatant was heated at 95 °C for 10 min to produce the sprouted pea milk (SPM), under the designations Y0 pea milk (Y0PM), Y1 pea milk (Y1PM), Y3 pea milk (Y3PM), Y5 pea milk (Y5PM), and Y7 pea milk (Y7PM). The prepared samples were immediately frozen by immersion in liquid nitrogen. These frozen samples were then stored in an ultrafreezer at −80 °C to preserve their integrity for subsequent analysis.

Preparation of enzymatically treated pea milk

The processing conditions for enzymatically treated pea milk (EPM) were optimized and the final product was prepared based on the method of García Arteaga et al., ²⁷ with slight modifications. After washing and soaking 50 g of dry pea seeds in 150 g of distilled water at 25 °C for 12 h, the total mass was adjusted to 200 g with additional distilled water. The mixture was then ground at high speed for 5 min using a blender. Following the filtration and centrifugation procedures described for SPM, non‐enzymatically treated pea milk (NEPM) was obtained. 100 mL of pea milk was enzymatically treated using three different enzymes (flavor protease, FP; alkaline protease, AP; papain, PP) at five concentration levels each. The specific activity concentrations employed were as follows: FP: 100, 150, 200, 250, and 300 U L⁻¹; AP: 500, 1500, 2500, 3500, and 4500 U L⁻¹; PP: 15 000, 30 000, 45 000, 60 000, and 75 000 U L⁻¹. The mixture was adjusted to the enzyme‐specific optimal pH and heated at 50 °C for 1 h to facilitate enzymatic hydrolysis. Subsequently, a sensory evaluation of the EPM was performed, assessing color, texture, and flavor. Based on the sensory scores, the optimal concentrations were selected as 150 U L⁻¹ for FP, 1500 U L⁻¹ for AP, and 30 000 U L⁻¹ for PP. The enzymatic reactions were terminated by heating at 95 °C for 10 min to obtain EPM. The resulting EPM samples were designated as flavor protease‐treated pea milk (FPPM), alkaline protease‐treated pea milk (APPM), and papain‐treated pea milk (PPPM) for subsequent experiments.

Determination of protein, fat, reducing sugar, and TCA‐soluble nitrogen content

The protein content was determined using the Kjeldahl method with a conversion factor of 6.25. The fat content was measured using the chloroform–methanol method. The reducing sugar content was quantified by direct titration. The TCA‐soluble nitrogen content was analyzed by adding an equal volume of 30% (w/w) TCA to the pea milk, followed by thorough mixing, centrifugation for 15 min, and subsequent nitrogen measurement in the supernatant. ²⁸

Determination of antinutritional factor content

Phytic acid and tannin concentrations were determined using commercial assay kits from Boxbio (Beijing, China) and Solarbio (Beijing, China), respectively, following the manufacturers’ instructions. Briefly, standard curves were constructed by measuring the absorbance of sodium phytate at 500 nm and tannic acid at 275 nm. The regression equations for phytic acid and tannin in SPM were $y=0.68x+0.099\left(R^2=0.999\right)$ and $y=31.7x+0.811\left(R^2=0.998\right)$, respectively, while those in EPM were $y=0.66x+0.108\left(R^2=0.998\right)$ and $y=31.72x+0.825\left(R^2=0.998\right)$, respectively. The analyte contents in the samples were then calculated based on their measured absorbances.

E‐nose analysis

E‐nose (PEN3, Airsense, Germany) was used to analyze the flavor profile of SPM and EPM. Briefly, the samples (5.0 g) were added to a 20 mL headspace vial and allowed to equilibrate for 30 min at 25°C. The testing time for each sample was 180 s, and the instrument cleaning time between samples was 80 s. ²⁹

Preparation of pea yogurt

The preparation of pea yogurt was carried out according to the method described by Ma et al., ²⁸ with minor modifications. Untreated pea milk (UPM, prepared as described for EPM), Y1PM, and FPPM were used as substrates. Each substrate was supplemented with 1.5% glucose, 7% sucrose, 0.25% soybean peptide, and 0.15% disodium phosphate. Following pH adjustment to 7.3, the mixtures were heated to 95 °C for 10 min and cooled to 43°C. A direct‐set starter culture containing Streptococcus thermophilus and Lactobacillus bulgaricus (1:1 ratio), each with a viable cell count of 1.0 × 10¹⁰ CFU g⁻¹, was inoculated at 6% (w/w). Fermentation was carried out at 43 °C until the pH reached 4.7. After fermentation, the products were stored at 4 °C for 12 h and were then designated as untreated pea yogurt (UPY), sprouted pea yogurt (SPY), and enzymatically treated pea yogurt (EPY).

Determination of acidity and pH

Yogurt was fermented until the desired endpoint was reached, during which samples were taken at 1 h intervals to measure pH and acidity. Briefly, the pH was measured in triplicate using a pH meter (FE28‐Standard, Mettler Toledo, Switzerland) on 10.0 g samples of pea yogurt. For acidity determination, 10.0 g of pea yogurt was mixed with 20 mL of CO₂‐free distilled water and titrated with 0.1 mol L⁻¹ NaOH to an endpoint of pH 8.3. The volume of NaOH consumed was recorded, and the acidity was calculated using Eqn 1: ²⁸

$$X=\frac{C\times \left(V_1-V_0\right)\times 100}{m\times 0.1}$$ (1)

where X represents the acidity of the sample; C represents the molar concentration of the NaOH standard solution; V ₁ represents the volume of the NaOH standard solution consumed by the sample titration; V ₀ represents the volume of the NaOH standard solution consumed by the blank titration; and m represents the mass of the sample.

Sensory evaluation

A sensory evaluation of SPM, EPM, and pea yogurt was conducted by a trained panel of 12 members (7 females and 5 males, aged 23–30 years) at the Molecular Sensory Science Laboratory of Beijing Technology and Business University (BTBU). Sensory evaluation of SPM and EPM was adapted from Xu et al. ³⁰ Prior to the experiment, all panelists were trained in quantitative descriptive analysis for 2 weeks in a clean, odor‐free sensory evaluation laboratory. This study was conducted with the informed consent of all participants and was approved by the University Scientific Research Ethics Committee of BTBU (2023‐2081). The evaluation was performed at 25 °C using 10 mL samples presented in 20 mL headspace vials, each labeled with a random three‐digit code. Based on a preliminary list of sensory descriptors generated by the panel for SPM and EPM, the five most frequently cited terms were selected for subsequent profiling. The final descriptors for SPM were ‘beany’, ‘grassy/green’, ‘earthy’, ‘oil‐oxidation’, and ‘mushroom’. For EPM, the descriptors were ‘beany’, ‘grassy/green’, ‘earthy’, ‘milk’, and ‘sweet’. The intensity of each descriptor was scored on a scale from 0 (none) to 5 (high intensity) with 0.5‐point increments. Each sample was evaluated in triplicate with a 1‐min interval between assessments.

Sensory evaluation of the three pea yogurt types was adapted from Ma et al. ²⁸ Samples served in coded plastic cups were assessed for color, texture, aroma, and taste. Panelists scored each attribute on a scale of 1 (least desirable) to 10 (most desirable) in 0.5‐point increments. All evaluations were performed in triplicate, and the data were summarized as radar charts.

Volatile compounds in pea milk and pea yoghurt

Headspace solid‐phase microextraction

A mixture of 5.0 g of sample, 0.5 g of NaCl, and 1 μL of 2‐methyl‐3‐heptanone (internal standard, 163.2 μg kg⁻¹ in n‐hexane) was prepared in a headspace vial. The vial was then equilibrated at 50 °C for 20 min. Volatile compounds were subsequently extracted from the headspace for 40 min at 50 °C using a divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS, 50/30 μm) solid‐phase microextraction fiber (Supelco, USA).

Gas chromatography–mass spectrometry analysis

Volatile compounds in SPM, EPM, and pea yogurt were detected using a 8890‐5977C GC–MS (Agilent, USA) device with DB‐WAX Polar Column (60 m × 0.25 mm, 0.25 μm). The temperature program for the chromatographic column was set as follows: the initial temperature of 40 °C was maintained for 2 min, then increased to 75 °C at a rate of 7°C min⁻¹, and finally to 230 °C at 5 °C min⁻¹ for 2 min. The temperature of the MS ion source was set to 230 °C. The MS detector was operated in electron impact mode at 70 eV, with a mass scan range of 35–350 m/z. ¹⁰

Qualitative analysis

A 1 μL aliquot of the n‐alkane standard mixture (C₇–C₄₀) was injected under the same conditions used for the GC–MS analysis to determine the retention times (RTs). The retention indices (RIs) of the volatile compounds were calculated based on the RTs of the n‐alkanes. The MS, RIs, and flavor characteristics of the target compounds were then compared with data obtained from the NIST 17 MS database and standards to identify volatile compounds.

Semi‐quantitative analysis

The relative concentration of each volatile compound was determined using the internal standard quantitative method. The semi‐quantitative content of each target compound was calculated based on the ratio of its chromatographic peak area to that of the internal standard (2‐methyl‐3‐heptanone), as shown in Eqn 2:

$$\begin{array}{c}{C}_x=\frac{A_n}{A_x}\times C_n\end{array}$$ (2)

where $C_x$ represents the relative concentration of the target compound; $C_n$ represents the concentration of the internal standard compound; $A_x$ represents the peak area of the compound of interest; and $A_n$ represents the peak area of the internal standard compound.

Relative odor activity value calculation

Equation 3 was used to calculate the relative odor activity values (r‐OAVs) of volatile compounds.

$$\begin{array}{c}\end{array}\mathrm{r}-\mathrm{OAV}=\frac{C}{T}$$ (3)

where C represents the relative concentration of volatile compounds; and T represents the odor threshold (OT) of this compound reported in the literature.

Statistical analysis

All experiments were conducted in triplicate, and data are presented as mean ± standard deviation. Statistical significance (P < 0.05) was determined by analysis of variance (ANOVA) followed by Duncan's multiple range test, using SPSS Statistics software (version 26.0; IBM Corp., Armonk, NY, USA). Heatmaps were generated with TBtools‐II. ³¹ Principal component analysis (PCA) of the E‐nose data was performed using WinMuster software (v1.6.2; Airsense Analytics GmbH, Schwerin, Germany). All other figures were plotted using Origin 2022 (OriginLab, Northampton, MA, USA).

RESULTS AND DISCUSSION

Contents of protein, fat, reducing sugar, TCA‐soluble nitrogen, and antinutritional factors in SPM and EPM

The protein, fat, reducing sugar, and TCA‐soluble nitrogen contents of SPM and EPM are presented in Fig. 1(A,B). The results indicated that sprouting led to a significant decrease in the contents of protein and fat in SPM. Conversely, the contents of reducing sugar and TCA‐soluble nitrogen increased significantly over the same period. As pea sprouts elongate, their protein and fat contents decrease, while the contents of reducing sugars and TCA‐soluble nitrogen increase. This metabolic shift is initiated by the activation of endogenous enzymes during sprouting, which break down macromolecular substances (proteins, fats, and starches) into small molecules to provide energy for tissue growth. ³² Furthermore, some of these degradation products, such as amino acids and peptides, are likely translocated to the growing parts of the embryo to serve as substrates for the synthesis of new proteins. Consequently, the dynamic balance between the rates of protein synthesis and degradation governs the net protein content observed in sprouting pea seeds. ³³ The contents of fat, TCA‐soluble nitrogen, and reducing sugars in the EPM increased significantly (P < 0.05). This is because exogenous proteases broke down the proteins into small peptides and amino acids, which simultaneously disrupted the protein matrix that had encapsulated starch and fat, thereby releasing these components. The liberated starch was then hydrolyzed into reducing sugars by endogenous starch enzymes.^{34, 35} In summary, the contents of reducing sugars and TCA‐soluble nitrogen increased significantly (P < 0.05) during both sprouting and enzymatic treatment, thereby providing readily available nitrogen and carbon sources, which are essential for the subsequent fermentation of pea milk. ³⁶

Figure 1.

Contents of protein, fat, reducing sugar, and trichloroacetic acid (TCA)‐soluble nitrogen in sprouted pea milk (A) and enzymatically treated pea milk (B); contents of antinutritional factor in sprouted pea milk (C) and enzymatically treated pea milk (D). Y0PM/Y1PM/Y3PM/Y5PM/Y7PM: pea milk prepared with 0/1/3/5/7 cm sprouts; APPM, alkaline protease‐treated pea milk; FPPM, flavor protease‐treated pea milk; NEPM, non‐enzymatically treated pea milk; PPPM, papain‐treated pea milk. Different letters above the bars indicate significant differences (p < 0.05), bars sharing the same letter are not significantly different.

Figure 1C,D show the changes in antinutritional factor content in SPM and EPM, respectively. The sprouting process enhanced the activities of phytase and polyphenol oxidase in peas, which effectively degraded antinutritional factors such as phytic acid and tannins in SPM. This degradation improved the nutritional quality of the product and increased the bioavailability of minerals. ³⁷ The content of free phytic acid and tannins in EPM increased paradoxically. This is because the hydrolysis of proteins by exogenous protease disrupted the phytic acid/tannin–protein complexes, which promoted the conversion of bound antinutritional factors into their free forms. However, this increase is beneficial, as it facilitates the subsequent effective removal of these antinutritional factors during fermentation processes. ³⁸

Sensory evaluation and E‐nose analysis of SPM and EPM

The sensory evaluation results for SPM and EPM are presented in Fig. 2(A,B), respectively. For SPM, all treatments shared a strong beany flavor, the intensity of which increased with sprout length. Notably, Y1PM showed a distinct profile compared to Y0PM, characterized by a reduction in grassy and beany flavors and a slight increase in earthy and oil‐oxidation notes. For EPM, the overall odor of APPM was relatively mild. In contrast, PPPM exhibited intense beany and grassy notes, while its pleasant flavors (e.g., milk, sweet) were less pronounced, correlating with its lower acceptability. Conversely, FPPM demonstrated a more desirable profile, with a marked reduction in beany odor and a concurrent enhancement of milk and sweet flavors. This may be attributed to perceptual interaction, and it is precisely this integration mechanism at the perceptual level that enables the sweet and milky flavors in pea milk to partially mask its beany flavor. ³⁹

Figure 2.

Sensory evaluation score of sprouted pea milk (A) and enzymatically treated pea milk (B); PCA score plot for E‐nose analysis of sprouted pea milk (C) and enzymatically treated pea milk (D). Y0PM/Y1PM/Y3PM/Y5PM/Y7PM: pea milk prepared with 0/1/3/5/7 cm sprouts; APPM, alkaline protease‐treated pea milk; FPPM, flavor protease‐treated pea milk; NEPM, non‐enzymatically treated pea milk; PPPM, papain‐treated pea milk.

The odor characteristics of SPM and EPM were further analyzed using an E‐nose, and the results are presented in Fig. 2(C,D). PCA showed that the cumulative variance explained by the first two principal components was 99.91% (with PC1 = 97.06%) for SPM and 99.72% (with PC1 = 95.56%) for EPM, establishing PC1 as the primary factor distinguishing the samples in each case. The E‐nose findings were consistent with the sensory evaluation. Specifically, E‐nose analysis revealed significant differences in the odor profiles of both SPM and EPM. For SPM, the distinctions intensified with increasing sprout length, while in EPM the PPPM sample exhibited the most distinct profile.

Volatile compounds of SPM and EPM

Analysis of volatile compounds in SPM

The volatile compounds in SPM were analyzed by GC–MS, with their r‐OAVs calculated and presented in Table 1. A total of 73 volatile compounds were detected in SPM, belonging to five chemical categories. Volatile compounds with high OAVs significantly influenced the overall odor characteristics of the samples. ⁴⁰ The volatile compounds contributing most significantly to the overall odor of SPM were identified as key aldehydes, alcohols, and heterocyclic compounds. These included nonanal (r‐OAV = 72.5; fatty), hexanal (r‐OAV = 5.91; grassy, green), and (E)‐2‐hexenal (r‐OAV = 1.32; grassy, green) from the aldehyde group; 1‐hexanol (r‐OAV = 48.13; grassy) and 1‐octen‐3‐ol (r‐OAV = 39.07; mushroom) from the alcohol group; and 2‐methoxy‐3‐secbutylpyrazine (r‐OAV = 7278; beany, earthy), 2‐isopropyl‐3‐methoxypyrazine (r‐OAV = 188.21; beany, earthy), and 2‐amylfuran (r‐OAV = 36.94; beany) from the heterocyclic compounds. Methoxypyrazines, which are crucial for the characteristic flavor of peas, were not detected in Y0PM or Y1PM but emerged in pea milk from later sprouting stages. This emergence can be explained by the proteolysis during sprouting, where storage proteins are broken down into amino acids. As sprouting progresses, the accumulating amino acids can subsequently undergo Strecker degradation to form methoxypyrazines. ⁴¹ Similarly, compounds such as 2‐butyloct‐2‐enal, likely released from the decomposition of macromolecules such as proteins and starch, were also detected in all SPM except for Y0PM and Y1PM. ²¹

Table 1.

Volatile compound content of sprouted pea milk

No.	Compounds	Odor description f	RI g	Relative concentration (μg kg−1) h					r‐OAV i
				Y0PM	Y1PM	Y3PM	Y5PM	Y7PM
Aldehydes (31)
1	(E)‐2‐Butenal	Flower	1050	ND	ND	0.45 ± 0.34a	0.60 ± 0.085a	ND	0
2	Hexanal	Grassy; green	1090	203.21 ± 15.43d	175.78 ± 10.55d	278.31 ± 14.88c	459.68 ± 17.03b	591.15 ± 49.15a	5.91
3	(E)‐2‐Pentenal	Green; fruity	1135	6.70 ± 3.39b	ND	6.13 ± 0.94b	7.0 ± 0.916b	12.54 ± 4.60a	0.01
4	Heptanal	Green; fatty	1193	6.66 ± 0.28d	6.23 ± 0.10d	12.53 ± 0.87b	10.62 ± 0.57c	15.09 ± 0.31a	5.39
5	2‐Hexenal	Green	1226	6.37 ± 0.33b	10.92 ± 0.25a	1.09 ± 0.11c	1.39 ± 0.07c	1.58 ± 0.05c	0.36
6	(E)‐2‐Hexenal	Grassy; green	1227	ND	ND	80.8 ± 2.15c	89.65 ± 1.38b	117.05 ± 2.65a	1.32
7	Octanal	Herbal; fruity	1297	8.79 ± 0.60cd	7.87 ± 0.42d	9.95 ± 0.53c	13.01 ± 0.7b	15.49 ± 1.28a	26.39
8	(E)‐2‐Heptenal	Grassy; beany	1332	9.12 ± 0.28e	11.37 ± 0.22d	32.12 ± 1.66b	28.24 ± 0.69c	37.95 ± 0.96a	0.95
9	Nonanal	Grassy	1401	48.7 ± 1.72cd	52.32 ± 0.85c	44.21 ± 2.75d	79.75 ± 5.03a	73.7 ± 3.38b	72.5
10	(E)‐2‐Octenal	Cucumber; fatty	1437	13.77 ± 0.24e	20.56 ± 0.48d	75.06 ± 5.08b	52.72 ± 1.63c	81.49 ± 3.83a	27.16
11	(E,E)‐2,4‐Heptadienal	Fatty; green	1500	0.50 ± 0.11d	0.53 ± 0.06d	5.23 ± 0.52b	2.82 ± 0.67c	8.64 ± 1.49a	0.56
12	Decanal	Sweet	1505	2.47 ± 0.12c	2.72 ± 0.08c	5.29 ± 1.1b	6.29 ± 0.06ab	7.23 ± 0.73a	2.41
13	Benzyaldehyde	Almond; nutty	1529	7.84 ± 0.28e	14.34 ± 0.28d	26.1 ± 0.89c	60.59 ± 3.34a	45.76 ± 3.7b	0.08
14	(E)‐2‐Nonenal	Green; cucumber	1542	1.2 ± 0.08d	1.73 ± 0.01d	7.65 ± 0.92b	5.09 ± 0.35c	12.34 ± 2.02a	64.95
15	Undecanal	Waxy; soapy	1610	0.32 ± 0.01b	0.3 ± 0.01b	1.06 ± 0.15a	1.21 ± 0.04a	1.2 ± 0.16a	0.1
16	β‐Cyclocitral	Herbal	1629	ND	0.6 ± 0.018d	0.79 ± 0.033c	1.393 ± 0.043a	0.987 ± 0.04b	0.46
17	(E)‐2‐Decenal	Green; earthy	1650	2.79 ± 0.11d	5.49 ± 0.11c	7.89 ± 0.39b	12.78 ± 0.83a	12.74 ± 0.5a	0.75
18	2‐Butyloct‐2‐enal	Fatty; green	1674	ND	ND	1.56 ± 0.26c	5.25 ± 0.54a	2.98 ± 0.14b	ND
19	(E,E)‐2,4‐Nonadienal	Fatty	1704	0.47 ± 0.05c	0.7 ± 0.09c	7.61 ± 0.26a	3.81 ± 0.06b	7.7 ± 0.5a	77
20	4‐Ethylbenzaldehyde	Sweet	1708	0.33 ± 0.02a	0.27 ± 0.01ab	ND	0.22 ± 0.01bc	0.15 ± 0.11c	0.03
21	Dodecanal	Waxy; citrus	1709	0.33 ± 0c	0.35 ± 0.02c	1.7 ± 0.09b	1.74 ± 0.06ab	1.9 ± 0.17a	0.19
22	3,7‐Dimethyl‐2,6‐octadienal	Citrus	1722	ND	ND	0.85 ± 0.04a	ND	0.68 ± 0.01b	0.03
23	(E)‐2‐Undecenal	Fresh; waxy	1734	1.1 ± 0.06d	2.18 ± 0.12c	3.56 ± 0.09b	5.7 ± 0.23a	5.5 ± 0.28a	4.07
24	(E,Z)‐2,4‐Decadien‐1‐al	Fatty	1764	ND	0.47 ± 0.03c	2.36 ± 0.09b	3.90 ± 0.18a	3.62 ± 0.25a	97.53
25	2,4‐Decadienal	Green; fatty	1764	ND	ND	14.58 ± 0.64c	18.75 ± 0.84b	21.70 ± 0.74a	0.16
26	4‐Isopropylbenzaldehyde	Green; herbal	1771	ND	ND	4.41 ± 0.13	ND	ND	0.01
27	Tridecanal	Soapy; citrus	1830	2.4 ± 0.17e	4.38 ± 0.05d	24.10 ± 0.7a	9.88 ± 0.21c	15.26 ± 0.99b	2.41
28	2,4‐Undecadienal	Fatty; citrus	1912	ND	ND	1.74 ± 0.21	ND	ND	3.99
29	Tetradecanal	Fatty; waxy	1932	6.90 ± 0.53e	11.88 ± 0.68d	116.24 ± 4.63a	40.16 ± 2.08c	64.65 ± 1.73b	1.06
30	Pentadecanal	Fresh; waxy	2045	ND	ND	1.26 ± 0.13	ND	ND	0
31	1‐Hexadecanal	Cardboard	2153	ND	ND	2.57 ± 0.11	ND	ND	ND
Alcohols (21)
32	Ethanol	Alcohol	926	7.12 ± 0.19a	2.88 ± 0.51b	ND	ND	ND	0
33	1‐Penten‐3‐ol	Green	1163	2.53 ± 0.19a	1.68 ± 0.02b	ND	1.78 ± 0.05b	ND	0.01
34	1‐Pentanol	Oily	1255	11.23 ± 2.94a	6.18 ± 0.64b	6.10 ± 0.39b	6.26 ± 0.64b	8.29 ± 1.82b	0.07
35	(Z)‐2‐Penten‐1‐ol	Grassy; green	1325	3.80 ± 0.53b	6.28 ± 0.09a	ND	ND	ND	0.01
36	1‐Hexanol	Grassy	1359	121.98 ± 9.93e	141.97 ± 0.37d	167.42 ± 5.71c	269.55 ± 11.76a	223.44 ± 4.74b	48.13
37	(E)‐3‐Hexen‐1‐ol	Earthy	1369	0.36 ± 0.15bc	0.45 ± 0.04bc	0.69 ± 0.06a	0.50 ± 0.02b	0.35 ± 0.03c	0.01
38	(Z)‐3‐Hexen‐1‐ol	Grassy	1391	2.34 ± 0.22a	2.03 ± 0.01b	ND	2.31 ± 0.09a	1.48 ± 0.10c	0
39	(E)‐2‐Hexenol	Green	1412	ND	1.67 ± 0.02c	1.97 ± 0.15c	3.46 ± 0.40a	2.67 ± 0.09b	0.01
40	1‐Octen‐3‐ol	Mushroom; earthy	1455	2.76 ± 0.23d	2.11 ± 0.02d	58.60 ± 3.41a	8.88 ± 0.68c	25.85 ± 0.02b	39.07
41	1‐Heptanol	Musty	1462	8.96 ± 0.91a	5.88 ± 0.22b	4.04 ± 0.16c	5.57 ± 0.52b	5.38 ± 0.29b	1.66
42	2‐Ethylhexanol	Citrus	1496	ND	ND	2.29 ± 0.16a	1.08 ± 0.24b	2.50 ± 0.06a	0
43	(E)‐2‐Hepten‐1‐ol	Fatty; green	1518	4.52 ± 0.37b	4.08 ± 0.12c	1.42 ± 0.04e	4.93 ± 0.22a	2.14 ± 0.03d	0
44	Linalool	Citrus	1552	ND	0.80 ± 0.01a	0.43 ± 0.03b	0.10 ± 0.01c	0.45 ± 0b	3.62
45	1‐Octanol	Waxy; green	1565	16.99 ± 0.71a	13.78 ± 0.33bc	12.27 ± 0.93c	13.23 ± 1.24bc	14.69 ± 0.54b	0.14
46	(E)‐2‐Octen‐1‐ol	Green; citrus	1620	4.12 ± 0.27e	5.28 ± 0.10d	6.36 ± 0.44c	10.30 ± 0.24a	6.88 ± 0.28b	0.52
47	2‐Octen‐1‐ol	Green	1622	0.46 ± 0.04c	ND	25.00 ± 1.72a	ND	9.02 ± 0.25b	0.33
48	1‐Nonanol	Fresh; fatty	1667	16.45 ± 0.11a	13.82 ± 0.18b	10.31 ± 0.63c	15.94 ± 1.72a	14.94 ± 0.94ab	0.37
49	(E)‐2‐Nonen‐1‐ol	Green	1712	ND	ND	1.03 ± 0.09b	0.78 ± 0.07c	1.19 ± 0.06a	0.01
50	(E)‐2‐Decenol	Citrus	1768	ND	1.40 ± 0.03b	9.92 ± 0.83a	9.06 ± 0.40a	9.75 ± 0.76a	ND
51	Vernol	Citrus	1781	ND	ND	0.78 ± 0.03a	ND	0.35 ± 0.02b	0
52	2‐Decenol	Citrus	1795	ND	ND	1.59 ± 0.19	ND	ND	ND
Ketones (12)
53	2,3‐Pentanedione	Sweet	1063	2.47 ± 0.34a	0.29 ± 0.19b	ND	1.70 ± 1.06a	ND	0.12
54	3‐Octanone	Herbal; sweet	1262	0.59 ± 0.22d	0.37 ± 0.04d	9.88 ± 0.32a	2.11 ± 0.26c	6.25 ± 0.11b	0.46
55	2‐Octanone	Earthy; herbal	1292	0.94 ± 0.03a	0.51 ± 0.04c	0.45 ± 0.02c	0.71 ± 0.12b	0.48 ± 0.04c	0.02
56	2,3‐Octandione	Asparagus	1327	2.95 ± 0.453d	ND	12.79 ± 0.76a	8.78 ± 0.15c	11.58 ± 0.16b	ND
57	6‐Methylhept‐5‐en‐2‐one	Grassy; green	1344	1.7 ± 0.05d	1.77 ± 0.02cd	1.95 ± 0.02c	3.56 ± 0.16a	2.79 ± 0.18b	0
58	2‐Nonanone	Green; earthy	1397	0.22 ± 0.01d	0.22 ± 0.01d	0.44 ± 0.02b	0.29 ± 0.02c	0.52 ± 0.01a	0
59	3‐Octen‐2‐one	Herbal; sweet	1416	2.45 ± 0.11d	2.16 ± 0.03d	8.64 ± 0.43b	6.44 ± 0.30c	9.12 ± 0.15a	0.04
60	3‐Nonen‐2‐one	Fatty; cheese	1519	ND	ND	1.64 ± 0.14a	ND	1.96 ± 0.01a	0
61	3,5‐Octadien‐2‐one	Fruity	1525	ND	ND	1.87 ± 0.11ab	1.64 ± 0.15b	2.01 ± 0.12a	ND
62	(E,E)‐3,5‐Octadien‐2‐one	Fruity; grassy	1577	0.91 ± 0.07c	0.67 ± 0.01d	1.10 ± 0.01b	1.12 ± 0.11b	1.31 ± 0.09a	0.01
63	Acetophenone	Sweet	1656	ND	ND	1.56 ± 0.14b	ND	5.74 ± 2.11a	0.09
64	Neryl Acetone	Fatty	1784	0.67 ± 0.03d	0.63 ± 0.02d	1.25 ± 0.18c	2.23 ± 0.16a	1.56 ± 0.13b	ND
Acids (3)
65	3‐Methylvaleric acid	Sour; stinky	1799	ND	ND	0.52 ± 0.2	ND	ND	0.01
66	1‐Hexanoic acid	Sour; stinky	1851	ND	ND	0.23 ± 0.01a	0.26 ± 0.2a	0.49 ± 0.37a	0
67	Nonanoic acid	Waxy; cheese	2153	ND	ND	ND	ND	0.96 ± 0.09	0
Heterocyclic compounds (6)
68	2‐Ethylfuran	Beany	945	3.17 ± 0.29d	9.82 ± 1.62b	6.80 ± 0.62c	13.22 ± 1.44a	10.20 ± 1.38b	0
69	2‐Butylfuran	Fruity	1133	ND	ND	4.91 ± 0.46	ND	ND	0.98
70	2‐Amylfuran	Beany	1237	56.86 ± 2.19d	62.06 ± 0.97d	214.22 ± 13.08a	117.09 ± 9.19c	156.73 ± 2.41b	36.94
71	2‐Isopropyl‐3‐methoxypyrazine	Beany; earthy	1439	ND	ND	0.73 ± 0.26	ND	ND	188.21
72	3,4‐Lutidine	Green	1490	ND	1.27 ± 0.02c	3.65 ± 0.22a	2.94 ± 0.04b	3.13 ± 0.04b	0.73
73	2‐Methoxy‐3‐secbutylpyrazine	Musty; earthy	1503	ND	ND	7.28 ± 0.58a	0.15 ± 0.04c	3.00 ± 0.18b	7278

Note: All compounds were identified by mass spectra and retention indices.

Abbreviation: ND, not detected; Y0PM/Y1PM/Y3PM/Y5PM/Y7PM: pea milk prepared with 0/1/3/5/7 cm sprouts.

^a–eDifferent superscript letters indicate significant differences (p < 0.05).

^f Odor description was obtained from the website: http://www.thegoodscentscompany.com.

^g Retention indices on DB‐WAX capillaries.

^h Relative concentration of aroma‐active compounds calculated by internal standard semi‐quantitatively.

ⁱ The r‐OAV was obtained by calculating the maximum concentration of each compound in sprouted pea milk.

A heatmap visualization of volatile compound concentrations in SPM is provided in Fig. 3(A). With increasing sprout length, the concentrations of volatile compounds, notably hexanal, (E,E)‐2,4‐nonadienal, and hexanol, increased significantly (P < 0.05), which agrees with prior research on yellow pea powder. ²¹ This analysis also indicated distinct sample clustering, with stronger correlations observed between Y0PM and Y1PM and between Y5PM and Y7PM. As summarized in Fig. 3(B), SPM contained 45–66 volatile compounds with total concentrations of 606.53 ± 44.36 to 1709.74 ± 97.34 μg kg⁻¹. Aldehydes were the dominant class, followed by alcohols and heterocyclic compounds. The following biochemical pathways for key compounds are proposed: Aldehydes such as hexanal and (E)‐2‐nonenal are generated from unsaturated fatty acids (linoleic acid, α‐linolenic acid) by lipoxygenase (LOX) and hydroperoxide lyase (HPL).^{42, 43, 44} A subsequent conversion of these aldehydes by alcohol dehydrogenase (ADH) yields corresponding alcohols, including hexanol and (E)‐2‐nonenol.^{43, 45} Meanwhile, 2‐amylfuran, an important odor contributor, originates from oxidized polyunsaturated fatty acids or thermally degraded amino acids. ⁴⁵ The changes in volatile compound content across SPM are depicted in Fig. 3(C). Relative to Y0PM, the concentrations of aldehydes, alcohols, and heterocyclic compounds in Y1PM remained statistically unchanged; however, a significant increase was observed in aromatic compounds. In contrast, all other SPM samples exhibited significant increases (P < 0.05) in the concentrations of most volatile compound classes compared to Y0PM, a finding that aligns with the sensory evaluation results.

Figure 3.

Heatmap analysis of volatiles (A), accumulation chart of species and concentrations of volatiles (B), and changes in the content of volatile compounds (C) in sprouted pea milk. Y0PM/Y1PM/Y3PM/Y5PM/Y7PM: pea milk with 0/1/3/5/7 cm sprouts. Different letters above the bars indicate significant differences (p < 0.05), bars sharing the same letter are not significantly different.

Analysis of volatile compounds in EPM

As in the analysis of volatile compounds in SPM, the concentrations and r‐OAVs of volatile compounds in EPM are compiled in Table 2. A total of 72 volatile compounds were detected in EPM, belonging to six chemical categories. The overall aroma of EPM was primarily driven by aldehydes (e.g., nonanal, r‐OAV = 22.67, fatty; hexanal, r‐OAV = 1.15, green, grassy), alcohols (e.g., hexanol, r‐OAV = 39.73, green, grassy; 1‐octen‐3‐ol, r‐OAV = 7.36, mushroom), and heterocyclic compounds (e.g., 2‐amylfuran, r‐OAV = 10.60, beany). Based on r‐OAV values, the contributions of other volatile classes were negligible. Notably, hexanal is a key contributor to the off‐flavor in pea milk. ¹⁰ Its significant decrease in EPM compared to NEPM may be attributed to the enzymatic treatment, which potentially exposes more active sites for the precise adsorption of this aldehyde. ⁴⁶

Table 2.

Volatile compound content of enzymatically treated pea milk

No.	Compounds	Odor description e	RI f	Relative concentration (μg kg−1) g				r‐OAV h
				NEPM	FPPM	APPM	PPPM
Aldehydes (18)
1	Hexanal	Grassy; green	1091	115.42 ± 6.30a	103.87 ± 9.15b	22.45 ± 1.29c	93.88 ± 2.96b	1.15
2	(E)‐2‐Hexenal	Grassy; green	1223	1.62 ± 0.10c	2.71 ± 0.06b	0.45 ± 0.02d	4.89 ± 0.53a	0.06
3	Octanal	Herbal; fruity	1294	3.77 ± 0.71c	5.78 ± 0.25b	1.19 ± 0.05d	6.80 ± 0.25a	11.58
4	(E)‐2‐Heptenal	Grassy; beany	1328	1.16 ± 0.17c	3.31 ± 0.36b	0.81 ± 0.05c	6.81 ± 0.16a	0.17
5	Nonanal	Grassy	1397	24.93 ± 0.58a	21.34 ± 1.29b	9.90 ± 0.55d	14.89 ± 0.29c	22.67
6	(E)‐2‐Octenal	Cucumber; fatty	1432	1.03 ± 0.04c	2.13 ± 0.17b	0.59 ± 0.01d	11.05 ± 0.14a	3.68
7	Citronellal	Herbal; citrus	1467	ND	ND	ND	0.69 ± 0.20	0.01
8	(E,E)‐2,4‐Heptadienal	Fatty; green	1494	ND	ND	ND	1.11 ± 0.03	0.07
9	Decanal	Sweet	1501	1.74 ± 0.05b	1.06 ± 0.19c	ND	2.56 ± 0.50a	0.85
10	Benzyaldehyde	Almond; nutty	1512	2.25 ± 0.01c	3.49 ± 0.06b	1.20 ± 0.06d	66.57 ± 0.84a	0.09
11	(E)‐2‐Nonenal	Green; cucumber	1523	ND	0.51 ± 0.03b	ND	3.27 ± 0.13a	354.00
12	Undecanal	Waxy; soapy	1606	2.67 ± 0.43a	0.92 ± 0.19b	ND	ND	0.21
13	Β‐Cyclocitral	Herbal	1624	0.60 ± 0.03b	0.60 ± 0.05b	0.36 ± 0.02c	2.88 ± 0.05a	0.96
14	4‐Ethylbenzaldehyde	Sweet	1706	0.49 ± 0.02c	0.66 ± 0.03b	0.27 ± 0.03d	3.00 ± 0.07a	0.23
15	(E)‐2‐Undecenal	Fresh; waxy	1731	ND	ND	ND	0.42 ± 0.02	0.30
16	2,4‐Decadienal	Green; fatty	1762	ND	ND	ND	2.50 ± 0.11	0.02
17	Dodecanal	Waxy; citrus	1766	0.15 ± 0.02b	0.11 ± 0.01b	ND	1.45 ± 0.17a	0.14
18	Tetradecanal	Fatty; waxy	1933	ND	ND	ND	11.82 ± 2.67	0.11
Alcohols (24)
19	Ethanol	Alcohol	990	94.76 ± 3.74b	105.48 ± 2.64a	ND	85.71 ± 3.42c	0.00
20	1‐Penten‐3‐ol	Green	1162	3.37 ± 0.63c	5.56 ± 0.13b	2.49 ± 0.02d	7.83 ± 0.65a	0.29
21	3‐Hexanol	Alcohol	1200	0.77 ± 0.29a	0.53 ± 0.05a	ND	ND	0.00
22	Isopentanol	Alcohol; fruity	1210	1.17 ± 0.07b	1.86 ± 0.15a	0.59 ± 0.001c	1.17 ± 0.07b	0.00
23	1‐Pentanol	Oily	1250	6.38 ± 0.24c	11.87 ± 0.25a	3.46 ± 0.08d	8.78 ± 0.43b	0.08
24	(Z)‐2‐Penten‐1‐ol	Grassy; green	1320	1.78 ± 0.27c	3.68 ± 0.47b	1.59 ± 0.02c	9.19 ± 0.34a	0.01
25	1‐Hexanol	Grassy	1354	222.46 ± 37.54a	204.82 ± 2.16a	57.65 ± 0.35b	215.47 ± 8.17a	39.73
26	(E)‐3‐Hexen‐1‐ol	Leaf; earthy	1385	1.31 ± 0.14d	6.98 ± 0.07a	3.24 ± 0.05b	2.77 ± 0.17c	0.06
27	(Z)‐3‐Hexen‐1‐ol	Grassy	1387	10.52 ± 1.91a	ND	ND	12.32 ± 0.44a	0.01
28	(Z)‐2‐Hexenol	Grassy; green	1407	ND	ND	ND	0.17 ± 0.09	0.00
29	1‐Octen‐3‐ol	Mushroom; earthy	1450	3.45 ± 0.41c	7.79 ± 1.40b	4.33 ± 0.15c	11.03 ± 0.27a	7.36
30	1‐Heptanol	Musty; leaf	1456	6.84 ± 0.62c	16.84 ± 0.27a	5.98 ± 0.21d	11.96 ± 0.46b	3.12
31	6‐Methyl‐5‐Hepten‐2‐ol	Grassy; green	1463	0.16 ± 0.06a	0.25 ± 0.07a	ND	0.26 ± 0.10a	0.00
32	(E)‐2‐Hepten‐1‐ol	Fatty; green	1517	1.15 ± 0.16b	ND	ND	3.89 ± 0.11a	0.00
33	3,7‐Dimethyloct‐6‐en‐3‐ol	Woody; citrus	1535	ND	ND	ND	3.71 ± 0.21	ND
34	Linalool	Citrus; flower	1547	0.52 ± 0.01b	0.68 ± 0.04b	0.41 ± 0.02b	383.85 ± 14.00a	1744.77
35	1‐Octanol	Waxy; green	1559	7.00 ± 0.55d	19.32 ± 0.32a	8.19 ± 0.26c	11.92 ± 0.41b	0.15
36	(E)‐2‐Octen‐1‐ol	Green; citrus	1614	0.47 ± 0.09d	0.71 ± 0.06b	0.58 ± 0.02c	4.27 ± 0.01a	0.01
37	2‐Octen‐1‐ol	Green	1617	0.31 ± 0.01a	0.22 ± 0.02b	ND	ND	0.01
38	1‐Nonanol	Fresh; fatty	1662	ND	ND	9.85 ± 1.02b	17.70 ± 0.31a	0.40
39	(Z)‐3‐Nonenol	Mushroom	1685	0.58 ± 0.03	ND	ND	ND	ND
40	α‐Terpineol	Citrus; herbal	1699	ND	ND	ND	0.44 ± 0.03	0.00
41	Vernol	Flower; citrus	1779	ND	ND	ND	0.46 ± 0.01	0.00
42	Benzyl alcohol	Flower	1790	ND	ND	ND	2.45 ± 0.30	0.00
Ketones (14)
43	3‐Hexanone	Sweet; grape	1063	1.16 ± 0.04	ND	ND	ND	0.03
44	4‐Methyl‐3‐penten‐2‐one	Earthy	1137	ND	ND	ND	2.33 ± 0.46	0.01
45	2‐Heptanone	Sweet	1188	7.65 ± 0.83c	17.27 ± 0.22a	3.54 ± 0.20d	14.74 ± 0.88b	0.12
46	5‐Methyl‐3‐heptanon	Sweet	1259	1.56 ± 0.91bc	2.30 ± 0.3b	0.81 ± 0.10c	6.09 ± 0.79a	0.15
47	2‐Octanone	Earthy; herbal	1289	0.64 ± 0.06b	1.86 ± 0.04a	0.48 ± 0.06c	1.77 ± 0.13a	0.04
48	2,3‐Octandione	Asparagus	1323	0.57 ± 0.1b	4.65 ± 0.13a	0.45 ± 0.01b	ND	ND
49	6‐Methylhept‐5‐en‐2‐one	Grassy; green	1339	2.08 ± 0.34c	5.33 ± 0.16b	0.97 ± 0.05d	5.96 ± 0.12a	0.01
50	2‐Nonanone	Green; earthy	1392	0.21 ± 0.03c	0.67 ± 0.03b	0.13 ± 0.01d	1.78 ± 0.01a	0.01
51	(E,E)‐3,5‐Octadien‐2‐one	Fruity; grassy	1480	ND	ND	ND	0.57 ± 0.15	ND
52	2‐Decanone	Fruity	1496	ND	0.58 ± 0.11b	ND	1.36 ± 0.06a	0.16
53	3‐Nonen‐2‐one	Cheese	1514	ND	ND	ND	0.71 ± 0.05	0.00
54	Acetophenone	Almond; sweet	1649	1.03 ± 0.1	ND	ND	ND	0.02
55	Dihydro‐beta‐ionone	Woody	1773	0.15 ± 0.01a	0.08 ± 0.02b	ND	ND	ND
56	Neryl acetone	Fatty	1782	0.23 ± 0.02b	0.20 ± 0.01b	ND	1.40 ± 0.13a	ND
Acids (2)
57	1‐Hexanoic acid	Sour; stinky	1798	0.89 ± 0.08c	1.81 ± 0.04b	1.85 ± 0.10b	3.39 ± 0.59a	0.00
58	Nonanoic acid	Waxy; cheese	2121	ND	0.43 ± 0.53a	ND	0.43 ± 0.20a	0.00
Esters(9)
59	Isoamyl butyrate	Fruity	1267	ND	ND	ND	0.73 ± 0.05	0.05
60	Isoamyl isovalerate	Sweet; fruity	1299	ND	ND	ND	2.3 ± 0.22	0.12
61	Ethyl octanoate	Sweet; fruity	1437	0.30 ± 0.04c	0.50 ± 0.01b	0.17 ± 0.003d	3.61 ± 0.08a	0.05
62	Ethyl nonanoate	Fruity	1541	0.64 ± 0.05a	0.64 ± 0.09a	ND	ND	0.00
63	γ‐Caprolactone	Sweet	1700	ND	ND	0.34 ± 0.04	ND	0.00
64	Nerylacetate	Flower	1716	ND	ND	ND	0.43 ± 0.01	0.00
65	Ethyl phenylacetate	Sweet	1749	0.42 ± 0.14	ND	ND	ND	0.00
66	γ‐Heptalactone	Sweet	1817	ND	ND	ND	0.45 ± 0.01	0.00
67	γ‐Octanolactone	Sweet	1933	1.56 ± 0.14c	2.67 ± 0.06a	2.07 ± 0.03b	ND	0.12
Heterocyclic compounds (5)
68	2‐Amylfuran	Beany	1232	29.69 ± 1.15b	26.37 ± 3.41b	7.91 ± 0.17c	63.58 ± 3.77a	10.60
69	2‐Methoxy‐3‐methylpyrazine	Nutty	1339	0.12 ± 0.07	ND	ND	ND	0.00
70	(Z)‐Linalool oxide (furanoid)	Flower	1425	ND	0.18 ± 0.04b	0.07 ± 0.008b	75.01 ± 2.82a	0.75
71	(E)‐Linalool oxide (furanoid)	Flower	1475	ND	ND	ND	36.09 ± 0.27	0.39
72	Tetramethylpyrazine	Chocolate; nutty	1481	ND	ND	0.62 ± 0.02	ND	0.00

Note: All compounds were identified by mass spectra and retention indices.

Abbreviations: APPM, alkaline protease‐treated pea milk; FPPM, flavor protease‐treated pea milk; ND, not detected; NEPM, non‐enzymatically treated pea milk; PPPM, papain‐treated pea milk.

^a–d Different superscript letters indicate significant differences (p < 0.05).

^e Odor description was obtained from the website: http://www.thegoodscentscompany.com.

^f Retention indices on DB‐WAX capillaries.

^g Relative concentration of aroma‐active compounds calculated by internal standard semi‐quantitatively.

^h The r‐OAV was obtained by calculating the maximum concentration of each compound in sprouted pea milk.

The concentration profile of volatile compounds in EPM is presented in the heatmap (Fig. 4(A)). Multivariate analysis revealed distinct clustering among the samples, with FPPM and NEPM exhibiting the strongest correlation. The diversity and abundance of volatile compounds in EPM are summarized in Fig. 4(B). A total of 34–59 volatiles were identified, with total concentrations ranging from 154.96 ± 5.07 to 1252.66 ± 49.90 μg kg⁻¹. Alcohols were the most abundant chemical class, followed by aldehydes and heterocyclic compounds. The variation in the content of volatile compounds in EPM is depicted in Fig. 4(C). Compared to NEPM, FPPM showed no significant changes in the concentrations of aldehydes, alcohols, or heterocyclic compounds, but exhibited significant increases (P < 0.05) in ketones, acids, and esters. In contrast, APPM demonstrated significant reductions (P < 0.05) in aldehydes, alcohols, ketones, esters, and heterocyclic compounds, coupled with a significant increase (P < 0.05) in acids. Conversely, PPPM showed significant increases (P < 0.05) in all categories of volatile compounds. The increased levels of ketones, acids, and esters in FPPM likely contributed to its sweet and milky flavors, which effectively masked beany and grassy notes, a conclusion consistent with the sensory evaluation results.

Figure 4.

Heatmap analysis of volatiles (A), accumulation chart of species and concentrations of volatiles (B), and changes in the content of volatile compounds (C) in enzymatically treated pea milk. APPM, alkaline protease‐treated pea milk; FPPM, flavor protease‐treated pea milk; NEPM, non‐enzymatically treated pea milk; PPPM, papain‐treated pea milk. Different letters above the bars indicate significant differences (p < 0.05), bars sharing the same letter are not significantly different.

Fermentation characteristics of SPM and EPM

Y1PM and FPPM were selected as matrices for fermentation from the sprouted and enzymatically treated pea milks, respectively. This selection was made owing to the capacity of Y1PM to reduce antinutritional factors without significantly increasing off‐flavors while providing more abundant carbon and nitrogen sources for fermentation, whereas FPPM not only enhanced the content of carbon and nitrogen sources but also effectively reduced or masked off‐flavor. As depicted in Fig. 5, the fermentation kinetics of UPM, Y1PM, and FPPM followed broadly similar trends in pH and acidity. A key distinction emerged after 1 h, with both Y1PM and FPPM exhibiting lower pH and higher acidity than UPM throughout the remainder of the process. This enhanced acidification is attributed to the higher concentrations of TCA‐soluble nitrogen and reducing sugars in the pre‐treated milks, which supported robust microbial activity. This enhancement effectively mitigates the typical challenge of weak acid production in pea yogurt fermentation. ²⁸

Figure 5.

Fermentation curves of pea milks: (A) pH of pea milk; (B) acidity of pea milk. FPPM, flavor protease‐treated pea milk; UPM, untreated pea milk; Y1PM, pea milk with 1 cm sprouts.

Volatile compounds in pea yogurt

Following the analysis of the pea milk substrates, the volatile compounds in the three pea yogurts were characterized (Table 3). A total of 54 volatile compounds were identified, falling into six distinct chemical categories. As indicated by the results in the table, the key volatile compounds contributing to the overall flavor profile of pea yogurt included aldehydes such as nonanal (r‐OAV = 4.19; fatty) and (E)‐2‐nonenal (r‐OAV = 2.53; green); alcohols such as hexanol (r‐OAV = 17.25; grassy) and 1‐octen‐3‐ol (r‐OAV = 8.93; mushroom); as well as 2‐amylfuran (r‐OAV = 2; beany) and 1‐hexanoic acid (r‐OAV = 3.07; sour). Notably, aldehydes, alcohols, and heterocyclic compounds, which are the primary sources of off‐flavors (beany, grassy), were already present in the fermentation substrates. In contrast, acids and esters, which impart sweet, fruity, and cheesy notes, were formed during fermentation. ⁴⁷ The combined effect of the reduction of key off‐flavor compounds and the generation of pleasant aroma compounds demonstrates that fermentation is an effective strategy for improving the flavor of pea‐based products. ⁴⁸ Compared to unfermented pea milk, the concentrations of aldehydes and alcohols were significantly reduced in the yogurt. This reduction can be attributed to the biotransformation of aldehydes by lactic acid bacteria and the competitive binding of newly formed alcohols to protein sites, which may displace other odorants.^{9, 49} A key and novel finding of this study was the complete degradation of 1‐octen‐3‐ol (mushroom odor) following the combined sprouting and fermentation treatment. This suggests that the sprouting pretreatment likely modifies the pea substrate composition, thereby creating more favorable conditions for the subsequent microbial fermentation to thoroughly convert or eliminate this specific compound.

Table 3.

Volatile compound content of pea yogurt

No.	Compounds	Odor description d	RI e	Relative concentration (μg kg−1) f			r‐OAV g
				UPY	SPY	EPY
Aldehydes (14)
1	Hexanal	Grassy; green	1083	4.08 ± 0.49b	3.54 ± 0.19b	7.45 ± 0.92a	0.07
2	3‐Methyl‐2‐butenal	Sweet; fruity	1202	ND	0.33 ± 0.02b	0.49 ± 0.05a	ND
3	(E)‐2‐Hexenal	Green	1220	1.04 ± 0.06b	0.75 ± 0.03c	1.30 ± 0.02a	0.01
4	Octanal	Citrus; fitty	1294	0.57 ± 0.24b	0.56 ± 0.27b	5.38 ± 1.25a	9.16
5	(E)‐2‐Heptenal	Green; fitty	1328	3.29 ± 0.35b	1.87 ± 0.33b	11.71 ± 2.23a	0.29
6	Nonanal	Fitty	1397	1.91 ± 0.21b	1.71 ± 0.07b	4.60 ± 0.21a	4.19
7	(E)‐2‐Octenal	Cucumber; fitty	1432	2.59 ± 0.93b	1.48 ± 0.14b	4.33 ± 1.09a	1.44
8	Furfural	Sweet	1458	0.25 ± 0.03a	0.34 ± 0.04a	0.37 ± 0.09a	0.00
9	Decanal	Sweet	1501	ND	ND	0.31 ± 0.05	0.10
10	Benzyaldehyde	Bitter almonds	1521	3.89 ± 0.08c	5.34 ± 0.21b	6.43 ± 0.26a	0.01
11	(E)‐2‐Nonenal	Cucumber; green	1523	ND	0.33 ± 0.03b	0.48 ± 0.04a	2.53
12	β‐Cyclocitral	Mint	1624	0.39 ± 0.04b	0.21 ± 0.01c	0.55 ± 0.02a	0.18
13	4‐Ethylbenzaldehyde	Sweet	1719	0.75 ± 0.05a	0.35 ± 0.02b	0.1 ± 0.01c	0.06
14	2,4‐Dimethylbenzaldehyde	Almond	1760	ND	2.13 ± 0.08a	1.22 ± 0.11b	ND
Alcohols (16)
15	1‐Penten‐3‐ol	Green	1160	2.02 ± 0.07a	1.72 ± 0.11b	ND	0.01
16	2‐Methylbutan‐1‐ol	Alcohol	1207	0.7 ± 0.01	ND	ND	0.00
17	Isopentanol	Alcohol; fruity	1210	ND	0.99 ± 0.10	ND	0.00
18	1‐Pentanol	Oily	1248	2.25 ± 0.17b	1.01 ± 0.13c	3.32 ± 0.13a	0.02
19	(Z)‐2‐Penten‐1‐ol	Grassy; green	1320	1.26 ± 0.04b	0.43 ± 0.05c	1.45 ± 0.04a	0.00
20	1‐Hexanol	Grassy; green	1354	86.84 ± 4.91a	53.78 ± 6.18b	96.57 ± 5.58a	17.25
21	(E)‐3‐Hexen‐1‐ol	Grassy; green	1384	2.97 ± 0.19b	0.40 ± 0.02c	4.26 ± 0.20a	0.00
22	1‐Octen‐3‐ol	Mushroom; earthy	1450	8.17 ± 0.21b	ND	13.4 ± 0.79a	8.93
23	1‐Heptanol	Musty; leaf	1456	4.19 ± 0.34b	2.83 ± 0.16c	5.39 ± 0.11a	1.00
24	2‐Ethylhexanol	Grassy; green	1489	ND	ND	0.68 ± 0.1	0.00
25	1‐Octanol	Waxy; green	1559	5.06 ± 0.26a	3.11 ± 0.22b	ND	0.04
26	(E)‐2‐Octen‐1‐ol	Green; citrus	1614	1.09 ± 0.17a	0.22 ± 0.08b	0.29 ± 0.03b	0.05
27	2‐Octen‐1‐ol	Green	1617	ND	ND	1.21 ± 0.21	0.02
28	1‐Nonanol	Fresh; fatty	1662	6.26 ± 0.13b	7.51 ± 0.09a	6.44 ± 0.44b	0.17
29	(Z)‐3‐Nonenol	Mushroom	1685	0.10 ± 0.02	ND	ND	ND
30	Phenethyl alcohol	Flower	1808	0.30 ± 0.02b	0.29 ± 0.06b	0.72 ± 0.05a	0.00
Ketones (11)
31	3‐Pentanone	Sweet; fruity	985	0.24 ± 0.14	ND	ND	0.01
32	2,3‐Pentanedione	Sweet	1054	3.74 ± 0.16a	2.37 ± 0.13b	2.31 ± 0.08b	0.19
33	2‐Heptanone	Sweet	1185	4.51 ± 0.30b	4.78 ± 0.34b	7.57 ± 1.05a	0.03
34	5‐Methyl‐3‐heptanon	Sweet	1257	0.53 ± 0.02c	0.95 ± 0.02a	0.86 ± 0.04b	0.02
35	3‐Hydroxy‐2‐butanone	Earthy	1286	2.10 ± 0.08b	1.45 ± 0.11c	2.44 ± 0.13a	0.17
36	1‐Octen‐3‐one	Mushroom; earthy	1304	0.74 ± 0.60b	ND	4.54 ± 0.04a	0.45
37	2,3‐Octandione	Asparagus	1321	ND	ND	0.40 ± 0.06	ND
38	6‐Methylhept‐5‐en‐2‐one	Green; citrus	1339	1.04 ± 0.05b	ND	1.97 ± 0.08a	0.00
39	2‐Nonanone	Green; earthy	1392	2.18 ± 0.02b	3.38 ± 0.16a	3.31 ± 0.06a	0.02
40	2‐Decanone	Fruity	1496	ND	ND	0.21 ± 0.01	0.02
41	2‐Hendecanone	Fruity; flower	1599	0.48 ± 0.02b	0.54 ± 0.03a	0.54 ± 0.01a	0.10
Acids (6)
42	Acetic acid	Sour	1444	ND	10.55 ± 0.5	ND	0.00
43	Butyric acid	Cheese	1626	ND	12.72 ± 1.73	ND	0.01
44	2‐Methylbutyric acid	Chesses	1690	ND	0.59 ± 0.09a	0.23 ± 0.14b	0.00
45	1‐Hexanoic acid	Sour	1777	54.52 ± 3.49b	49.58 ± 6.15b	110.42 ± 7.74a	3.07
46	n‐heptanoic acid	Sour; stinky	1895	ND	0.24 ± 0.14	ND	0.00
47	Octanoic acid	Cheese	1997	0.65 ± 0.12b	1.46 ± 0.26a	1.22 ± 0.51ab	0.00
Esters (4)
48	Hexyl butyrate	Fruity	1417	ND	0.21 ± 0.02	ND	0.00
49	Ethyl octanoate	Sweet; fruity	1437	0.25 ± 0.02c	0.58 ± 0.03a	0.32 ± 0.02b	0.03
50	Octyl formate	Fruity	1558	ND	ND	7.31 ± 0.88	ND
51	γ‐Caprolactone	Sweet; fitty	1926	1.29 ± 0.11b	1.54 ± 0.11b	2.80 ± 0.18a	0.16
Heterocyclic compounds (3)
52	2‐Amylfuran	Beany	1232	5.41 ± 0.27b	11.57 ± 1.16a	4.81 ± 0.52b	2.00
53	2,3,5‐Trimethylpyrazine	Chocolate; nutty	1409	ND	0.40 ± 0.03	ND	0.00
54	2‐Ethyl‐3,5‐dimethylpyrazine	Nutty	1498	ND	0.48 ± 0.05	ND	ND

Note: All compounds were identified by mass spectra and retention indices.

Abbreviations: EPY, enzymatically treated pea yogurt; ND, not detected; SPY, sprouted pea yogurt; UPY, untreated pea yogurt.

^a–c Different superscript letters indicate significant differences (p < 0.05).

^d Odor description was obtained from the website: http://www.thegoodscentscompany.com.

^e Retention indices on DB‐WAX capillaries.

^f Relative concentration of aroma‐active compounds calculated by internal standard semi‐quantitatively.

^g The r‐OAV was obtained by calculating the maximum concentration of each compound in sprouted pea milk.

The concentration profiles of volatile compounds in the three pea yogurts are presented in the heatmap (Fig. 6(A)), which revealed distinct clustering among the samples. Specifically, UPY and SPY exhibited a stronger correlation. As summarized in Fig. 6(B), the number of volatile compounds ranged from 37 to 42, with total concentrations between 186.96 ± 17.33 and 337.08 ± 16.88 μg kg⁻¹. Alcohols constituted the most abundant chemical class across all samples, followed by acids. EPY showed the highest total volatile concentration, followed by UPY, while SPY had the lowest. Changes in the important volatile compounds, based on r‐OAV values, are detailed in Fig. 6(C). A notable finding was the significant reduction (P < 0.05) of alcohols in SPY. This can be explained by a two‐stage process: during sprouting, alcohols are released from complexes with proteins or starch, and subsequent fermentation by lactic acid bacteria leads to their degradation or adsorption onto the protein matrix.^{9, 21} Conversely, EPY was characterized by a significant increase (P < 0.05) in acid content, contributing to its distinct flavor profile. ⁵⁰

Figure 6.

Heatmap analysis of volatiles (A), accumulation chart of species and concentrations of volatiles (B), and changes in the content of volatile compounds (C) in pea yogurt. EPY, enzymatically treated pea yogurt; SPY, sprouted pea yogurt; UPY, untreated pea yogurt. Different letters above the bars indicate significant differences (p < 0.05), bars sharing the same letter are not significantly different.

Sensory evaluation of pea yogurt

The sensory evaluation results of the pea yogurts are presented in Fig. 7. Compared to UPY, both SPY and EPY received higher scores for flavor and taste. The presence of acid compounds may help mask the beany off‐flavor in these products, contributing to their improved sensory profiles. Differences were also observed in texture. SPY received the lowest texture score, which can be attributed to its soft gel strength and observable whey separation. Regarding appearance, UPY achieved the highest score, likely due to its more desirable, milky color compared to the other samples.

Figure 7.

Sensory evaluation score of pea yogurt. EPY, enzymatically treated pea yogurt; SPY, sprouted pea yogurt; UPY, untreated pea yogurt.

CONCLUSIONS

This study systematically investigated the impact of sprouting and enzymatic pre‐treatments on pea milk for yogurt production. Both methods enhanced fermentation efficiency by mitigating antinutritional factors and providing a nutrient‐rich medium for lactic acid bacteria, thereby resolving the typical weak acidification problem. Y1PM and FPPM were identified as the optimal substrates through rigorous analysis. Flavor enhancement was achieved through distinct pathways: off‐flavor reduction in sprouted milk versus active aroma generation in EPM. The overall superior quality of EPY over SPY, particularly in texture, confirmed the greater efficacy of enzymatic hydrolysis. Our findings underscore the potential of employing combined pre‐treatment strategies for superior product quality and offer a strategic framework for flavor improvement in a wide range of plant‐based yogurts. It is worth noting that in the future industrialization of pea yogurt further attention should be paid to the feasibility of process scale‐up, the stability of the fermentation endpoint, and the retention of flavor and texture during shelf life, while also addressing consumer demand for clean‐label products.

FUNDING

This research was supported by the National Natural Science Foundation of China (No. 32202214), Beijing Natural Science Foundation (No. 6252002) and Cross‐Innovation Open Project of Food Flavor and Health, Beijing Technology & Business University (FFHCI‐2026004).

AUTHOR CONTRIBUTIONS

Wendong Wang: writing – original draft preparation. Liya Liu: investigation; methodology; formal analysis. Yixin Dai: investigation; methodology. Xiaoying Xiong: investigation; methodology. Shuang Bi: supervision; project administration. Ye Liu: visualization.

CONFLICT OF INTEREST STATEMENT

There are no conflicts of interest to declare.

DATA AVAILABILITY

Data will be made available on request.