Strawberry Nectar Colour Stability and Aroma: Influence of Cultivar, Harvest Time and Ripening Stage

Abstract

This study investigated the impact of cultivar, harvest time, and ripening stage of strawberries on their aroma concentration and profile, and colour stability of nectars produced from these strawberries. Purees from 12 different cultivars from two countries, collected at different ripening stages and harvest times, were analysed. Furaneol and mesifuran content was analysed using a gas chromatography–flame ionisation detector (GC-FID), and gas chromatography–mass spectrometry (GC-MS) was used to determine the content of 12 aroma compounds, including esters, C6 compounds, and lactones. Nectars produced from these purees had their colour stability measured over 12 weeks. Both the colour and aroma were greatly influenced by strawberry cultivar. Within cultivars, nectars produced from strawberries that had been harvested overripe showed higher colour stability and higher concentrations of aroma compounds than those harvested ripe from an earlier harvest, although some cultivars were more affected by harvest time than ripening stage. Aroma compounds that correlated significantly (p < 0.05) with a good colour after storage included furaneol, ethyl butanoate, hexanal, γ-decalactone and γ-dodecalactone, as well as the total concentration of aroma compounds. Only γ-decalactone concentrations correlated significantly with overall nectar colour stability, although this could be due to cultivar effects.

1. Introduction

Strawberry nectar is a popular beverage made from strawberry puree. In the European Union it is mandatory that it only contains puree, water, sugar, and acid. This regulation makes strawberry nectar a challenging product in relation to colour and flavour, because these two factors are important for consumer acceptance [1,2], but no colourants or flavour enhancers can be added to improve the nectar. These rules cause particular difficulties for strawberry nectars, as they have very poor colour stability. Nectars rapidly degrade from bright red, which is appealing to consumers, to an unacceptable brown colour [2,3].

The acceptability of a nectar’s colour can be quantified using the Acceptance Factor (AF) (calculated from CIELAB colour components). An AF value below 0.4 is considered as unacceptable to consumers, while a value above 0.7 indicates excellent colour, which is consistently well received [2]. This gives a simple method for tracking consumer acceptance—i.e., good colour—over time, and so can give an estimation of colour shelf-life by tracking how long strawberry nectars remain acceptable to consumers. Colour stability can subsequently be assessed by calculating a nectar’s AF before and after storage [4] to give a numerical value to how much the acceptability of nectars change over storage. While the initial acceptability of strawberry nectars is an important factor for the colour stability of nectars, it does not reliably predict colour stability over time [4]. Many studies have shown that factors such as cultivar, ripening stage, country of origin, and harvest time can significantly influence colour stability [5,6,7,8,9].

When analysing samples of the same cultivars, overripe strawberries harvested late in the harvest season produced nectars with the best colour stability [5,6]. Different cultivars produced nectars with very different inherent colour stabilities, with less ripe strawberries from some cultivars producing more stable nectars than those produced from more ripe strawberries from other cultivars [6,7]. Different cultivars respond differently to environmental factors, and so thoroughly investigating the influence of different parameters on different cultivars is of paramount importance. Alongside colour, flavour is the most important factor for consumer’s acceptance of strawberry products [10,11].

The flavour of strawberry nectar is determined by volatile organic aroma compounds, which account for less than 0.01% of the fresh weight of strawberries [12]. These aroma compounds are both numerous and diverse, with over 900 different identified substances including esters, terpenoids, furans, ketones, lactones, alcohols, and aldehydes [13]. The concentration and profile of aroma compounds vary significantly between different cultivars [14,15,16]. During ripening, the total concentration of aroma compounds can increase up to seven-fold [16]. Furthermore, these concentrations vary throughout the harvest season, depending on the cultivar [17].

Esters are responsible for the fresh and fruity odour of strawberries and are considered as the most important class of aroma compounds in these fruits. They account for the majority of strawberry aroma (comprising 25–90% of the total concentration), with over 130 different esters identified [16]. The most important esters include methyl butanoate, ethyl butanoate, and methyl hexanoate, although the exact concentrations and profiles vary among different cultivars [16]. Much of the increase in strawberry aroma during ripening is due to the increase in ester concentrations [14]. Consequently, esters including ethyl butanoate, methyl butanoate, and methyl hexanoate have been proposed as chemical markers for fully ripe strawberries [18,19,20].

Alcohols and aldehydes, especially the C6 compounds hexanal, hexenal, and hexenol, are responsible for the ‘green’ aroma notes in strawberries [16]. Their concentrations decrease during ripening and so have been suggested as markers for the identification of unripe strawberries [14,18,21]. The furanones 2,5-dimethyl-4-hydroxy-3(2H)-furanone (DMHF or furaneol) and 2,5-dimethyl-4-methoxy-3(2H)-furanone (DMMF or mesifurane), particularly furaneol, are considered key signature aromas of strawberries [22,23].

Despite its importance to strawberry aroma, many studies struggle to identify furaneol [14,21,24] due to its heat instability, which makes it difficult to detect by solid phase microextraction (SPME) methods [6]. Furaneol concentration is highest in overripe fruit [25] and so could serve as a marker for identifying overripe fruit. In strawberry juice, furaneol, (Z)-3-hexenal, and methyl butanoate have been identified as compounds essential for good flavour [26]. The profile and concentration of aroma compounds in strawberries are therefore very important for the flavour of strawberry nectars produced from them.

It is a long-standing aim to find markers that can be used to identify strawberries that can produce colour-stable nectars [4,8,27] in order to produce nectars with the longest possible shelf-life and in order to reduce food waste and maximise profits. As the aroma of strawberries changes during ripening—C6 compounds decrease as the fruit ripens, while compounds such as esters and furanones increase during the overripe stages—and overripe strawberries produce nectars with good colour stability, this study aimed to determine if any aroma compounds could be used as markers for colour stability of nectars. To this end, strawberries were characterised according to their cultivar, ripening stage, and harvest time, and their aroma compounds were quantified, to investigate the impact that these parameters have on aroma. Nectars were produced from these strawberries, and the colour stability of these nectars were determined and the impact of cultivar, ripeness, and harvest quantified. The colour stability was then compared to the aroma compounds. In total, this study compared the colour stability of nectars to the aroma profiles and concentrations of 42 samples of strawberries from two different countries varying in ripening and harvest time.

2. Materials and Methods

2.1. Strawberry Fruits

Strawberry fruits were harvested in June and July 2021. Sixteen samples were harvested from open fields in Poland (region of Tomaszów Lubelski, Geo: 50.436 23.421; minimum 5 kg per variant) and included ripe and overripe strawberries from a single harvest of the cultivars ‘Marmolada’, ‘Rumba’, and ‘Roxana’, as well as ripe and overripe strawberries from two harvests along with overripe strawberries from a third harvest of cultivars ‘Elsanta’ and ‘Honeoye’. These strawberries were received by an industrial processor of strawberries, and the cultivars used were those grown for industrial production. Thirty samples of strawberries (minimum 2 kg per variant) were harvested in Austria from an open field in Zeiselmauer, Lower Austria (Geo: 48.326, 16.181). The cultivars ‘Allegro’, ‘Malling Centenary’, ‘Clery’, ‘Sibilla’, ‘Rendezvous’, ‘Limalexia’, and ‘Faith’ were collected ripe and overripe twice during the season. The cultivar ‘Malling Allure’ was also collected twice; however, these fruits remained partially white and did not ripen fully as defined by surface colour, so no overripe strawberries were collected. These cultivars were selected by farmers for their suitability for ‘pick-your-own’ strawberry harvesting. ‘Allegro’, ‘Honeoye’, ‘Rendezvous’, ‘Rumba’, and ‘Clery’ were considered early varieties. ‘Sibilla’ and ‘Malling Centenary’ are early to middle season. ‘Marmolada’ is middle season. ‘Limalexia’ and ‘Elsanta’ are middle to late season. ‘Faith’, ‘Roxana’, and ‘Malling Allure’ are late-season cultivars. Ripe and overripe designations were determined by external examination of the surface colour, where ripe fruits are fully red, and overripe fruits were dark red to purple in colour. All fruit samples were picked in the morning, frozen in 5 L plastic bags within two hours, and stored at −18 °C.

2.2. Preparation of Nectars

Frozen strawberries were defrosted for 24 h at 20 °C and processed into puree by a rotor mill with a 1 mm sieve (Feuma, Gößnitz, Germany). Purees were mixed with water, citric acid, and sugar to produce nectar (40% puree, 15 °Brix, 7.0 g/L titratable acidity) and homogenised with a hand blender (Philips, Drachten, The Netherlands). Nectar (140 g ± 1 g) was weighed into glass jars (212 mL), sealed with a screw lid, and pasteurised (80 °C, 10 min) in a water bath (Westfalia, Hagen, Germany). A minimum of 24 jars were produced for each variant. Austrian strawberries were pureed into one bulk (minimum 2 kg), whereas Polish samples were processed into three bulks (3 × 5 kg), with the exception of three ‘Elsanta’ variants (second harvest, ripe and overripe; third harvest) and one ‘Honeoye’ variant (second harvest, ripe), which were processed into one bulk (5 kg) due to fewer strawberries being available from these variants due to defrosting issues. Nectars were stored at 20 ± 3 °C in the dark.

2.3. Physical and Chemical Analysis of Puree

The bulked puree of all variants had their soluble solids, pH, and titratable acidity determined in order to produce nectars with the same fruit content, total soluble solids, and titratable acidity as mentioned in the previous chapter. The total soluble solids (°Brix) were measured using a refractometer (N-20, Brix 0~20%, ATAGO, Tokyo, Japan) [28]. pH values were determined with a pH meter (MultilineP4, WTW, Weilheim, Germany) and pH electrode (SenTix 41-3, WTW, Weilheim, Germany). Titratable acidity of puree was determined by titration to an endpoint of pH-value 7.0 (0.1 N NaOH (Titrisol ^®), Merck KGaA, Darmstadt, Germany) and multiplied by the acidity factor of tartaric acid (0.75) to express acidity as g/L [29].

2.4. Colour Measurements

CIELAB colour space components L* (lightness), a* (red to green), and b* (yellow to blue) were measured using a Minolta CM-5 spectrophotometer (spectrophotometric method, D65, 30 mm 10°, reflection measurement, gloss excluded, Minolta, Osaka, Japan). C* (Chroma) and h° (hue angle) were calculated as previously reported [30]. AF was calculated (AF = a*/h°) as previously reported [2]. Colour measurements were undertaken on nectars. All nectars were measured after pasteurisation on the day of production. Two samples from each nectar were measured in duplicate. The colour stability was quantified with different indicators for colour stability: AF4, AF8, and AF12 represent the AF at 4, 8, and 12 weeks. D4, D8, and D12 represent the difference in AF between the day of production and 4, 8, and 12 weeks, respectively [4].

2.5. Determination of Furanone

Furanones were calculated on the bulked puree of each variant. Determination of furanones (mesifuran and furaneol) was undertaken as described in Brandes et al. [6]. Frozen puree was defrosted in a water bath at 20 °C. A 10 g quantity of defrosted puree and 10 mL citrate buffer were mixed in a 50 mL centrifuge tube for 5 min and centrifuged (1132× g, 10 min). A 3 mL quantity of the supernatant was combined with 90 µL internal standard (maltol, 250 mg/L in 20% ethanol) and 1 mL tert-butyl methyl ether in a 15 mL centrifuge tube. A 4 g quantity of sodium dihydrogen phosphate monohydrate was added, mixed (10 min), and centrifuged (2010× g, 7 min). The supernatant was then analysed with a gas chromatography–flame ionisation detector (GC-FID) from Agilent Technologies 7820A (Agilent Technologies, Santa Clara, CA, USA), under the following conditions: 1 µL injection volume, splitless, injection temperature 200 °C, column: DB-Wax 60 m length, 0.32 mm ID, 0.25 µm film thickness (Agilent Technologies, Santa Clara, CA, USA), 3.5 mL/min helium gas, constant flow, temperature programme: initial 45 °C for 3 min, heating to 80 °C at 30 °C/min, holding time 0 min, heating to 167 °C at 2 °C/min, holding time 0 min, heating to 250 °C at 40 °C/min, holding time 12 min, FID detector 250 °C. The quantification was based on peak areas with maltol as internal standard. Extraction and analysis of Austrian samples were carried out in technical duplicates, except for ‘Allegro’ (2nd harvest, overripe) due to small sample size (lack of availability of this variant due to environmental factors during harvesting) and large quantities of puree required for nectar production. Polish samples were extracted and analysed from two purees of each variant where possible, or two technical repeats when not possible.

2.6. Determination of Other Aroma Compounds

Aroma compounds were calculated on the bulked puree of each variant. Determination of concentration of methyl butanoate, ethyl butanoate, butyl ethanoate, isoamyl acetate, methyl hexanoate, hexyl acetate, trans-2-hexenol, hexanal, linalool, trans-2-hexenal, γ-decalactone, and γ-dodecalactone was undertaken by SPME followed by gas chromatography–mass spectrometry (GC-MS). Extraction of strawberry puree was performed as described for mesifuran and furaneol determination. Supernatant from the first centrifugation (5 mL) was combined in a 20 mL SPME vial with 2 g sodium chloride, a magnetic stirrer bar, and 50 µL internal standard (3,4-dimethylanisole, 5 mg/L in 10% ethanol) for SPME. Strawberry extracts were analysed by GC-MS (gas chromatograph: Agilent Technologies 6890N (Agilent Technologies, Santa Clara, CA, USA). MS-Detector: Agilent Technologies 5975 (Agilent Technologies, St. Clara, CA, USA), inert mass selective detector (MSD) with triple-axis detector (Agilent Technologies, St. Clara, CA, USA), autosampler (Combi PAL, CTC Analytics, Zwingen, Switzerland), and data analysis software (MSD ChemStation E.02.00.493). Column: OPTIMA FFAP, 60 m, 0.25 mm ID, 0.25 µm film thickness (Avantor Inc., Radnor, PA, USA), 19.9 mL helium/min, splitless. Initial temperature: 250 °C. Temperature programme: 40 °C for 8 min, 5 °C/min to 250 °C for 10 min, run time 60 min. Extraction and analysis of Austrian samples were performed in technical duplicate except for ‘Allegro’ (2nd harvest, ripe) and ‘Faith’ (2nd harvest, ripe and overripe) due to small sample size due to a lack of availability of this variant due to environmental factors during harvesting and large quantities of puree required for nectar production. Polish samples were extracted and analysed from two purees of each variant where possible, or two technical repeats when not possible.

2.7. Statistical Analysis

Statistical analysis (ANOVA, Tukey HSD, curve fitting, regression analysis, and Pearson correlations) was carried out using IBM SPSS 26 version: 29.0.0.0 (241) (Statistical Package for the Social Sciences).

3. Results

3.1. Effects on Colour Stability

The colour and colour stability of the nectars varied greatly by cultivar, harvest time, and ripening stage, whereas individual cultivars respond differently to these parameters. The decrease in AF of each cultivar, separated by harvest time and ripening stage, is shown in Figure 1 and Table S1. Differences in mean AFs after 12 weeks of storage (D12) between AF0 (time of nectar production) and AF12 (12 weeks of nectar storage) are shown in Figure 2 and Table S1 and provide an index for colour stability.

3.1.1. Cultivar

Cultivar was a significant determining factor of the colour and colour stability of strawberry nectar. Considering the mean AF0 value of all ripening stages and harvest times within each cultivar, the cultivar with the best initial colour (highest mean AF0) was ‘Marmolada’ (0.94). The cultivars ‘Elsanta’ (0.92), ‘Rumba’ (0.90), and ‘Faith’ (0.90) also exhibited high mean AF0 values, with no significant difference between these cultivars and ‘Marmolada’. The lowest mean AF0 was found in ‘Rendezvous’ (0.77), although there was no significant difference (p < 0.05) between the cultivars ‘Rendezvous’, ‘Malling Centenary’ (0.78), ‘Malling Allure’ (0.79), ‘Limalexia’ (0.80), ‘Sibilla’ (0.81), ‘Clery’ (0.82), and ‘Allegro’ (0.81).

Considering individual ripening stages and harvest times, the cultivars with the highest initial AFs were ‘Elsanta’ (1.00, overripe, 7 July 2021) and ‘Faith’ (0.99, overripe, 5 July 2021) and the cultivars with the lowest initial AFs were ‘Malling Centenary’ (0.68, ripe, 14 June 2021) and ‘Rendezvous’ (0.71, ripe, 16 June 2021).

After 12 weeks of storage, the cultivar ‘Malling Allure’ (0.12) had the lowest mean AF12 value, which was significantly lower than for all other cultivars (p < 0.05). On average, nectars from this cultivar became unacceptable to consumers (AF < 0.4) within just three weeks of storage. This poor colour stability is likely linked to the fact that these fruits never turned fully red, with a surface colour that was a mix of red and white, even when left on the plant for extended periods. Consequently, no overripe strawberries were available for this cultivar. The highest mean AF12 value was observed for ‘Marmolada’ (0.69), indicating that nectars produced from this cultivar had close to excellent colour (AF > 0.7) even after 12 weeks of storage. This value was significantly higher (p < 0.05) than for all other cultivars. In addition, nectars produced from the cultivars ‘Rendezvous’ (0.59), ‘Honeoye’ (0.58), ‘Faith’ (0.56), ‘Elsanta’ (0.51), and ‘Roxana’ (0.50) maintained a mean AF value above 0.4 after 12 weeks storage. Considering individual ripening stages and harvest times, the cultivars with the highest AF12 values were ‘Marmolada’ (0.73, overripe, 30 June 2021) and ‘Elsanta’, ‘Honeoye’, and ‘Faith’ (all three 0.68, overripe, different harvest times), and the cultivars with the lowest AF12 values were ‘Malling Allure’ (0.11, ripe, 1 July 2021; 0.14, ripe, 5 July 2021) and ‘Sibilla’ (0.20, ripe, 16 June 2021).

The cultivar ‘Rendezvous’ had the lowest mean AF0 value but one of the highest AF12 values and therefore was the most colour-stable, which is represented by the lowest mean D12 value (difference between AF0 and AF12, 0.18) (Figure 2). The cultivars ‘Marmolada’ (0.25) and ‘Honeoye’ (0.30) also had low D12 values, indicating good colour stability in addition to a very good initial colour. The highest D12 value was found for ‘Malling Allure’ (0.67), as was expected given its good initial colour but poor colour acceptance 12 weeks after storage. The cultivars ‘Limalexia’ (0.52) and ‘Sibilla’ (0.49) also had high mean D12 values.

3.1.2. Country of Origin

Nectars produced from Polish strawberries had significantly higher AF0 (t(330) = 13.10 p < 0.001) and AF12 (t(330) = 9.48 p < 0.001) values, as well as significantly lower D12 (t(330) = −4.16 p < 0.001) values, when compared with nectars from Austrian strawberries. Despite this study comparing different cultivars, this indicates that nectars produced from Polish strawberries had a more acceptable initial colour, a more acceptable colour after 12 weeks, and therefore a better colour stability. This is likely because strawberries in Poland are specifically cultivated for processing, with cultivars selected for their suitability for processing. Furthermore, strawberries harvested at the correct ripening stage for processing (i.e., the ripe strawberries) tend to be more mature than those harvested for the fresh market. As a result, there was less variation between the ripe and overripe Polish strawberries used in this study.

3.1.3. Ripening Stage and Harvest Time

From the samples from Poland, the cultivars ‘Marmolada’, ‘Roxana’, and ‘Rumba’ were only harvested once per season, at ripe and overripe stages. For ‘Roxana’ and ‘Rumba’, there were no significant differences in AF0, AF12 and D12 values between the ripe or overripe samples. However, for ‘Marmolada’ the overripe strawberries are more suitable for nectars with significantly higher AF12 and lower D12 values compared to the ripe samples, despite no significant differences in AF0 values between the ripening stages being observed. Therefore, for ‘Marmolada’, the overripe strawberries led to more colour stable nectars. The cultivars ‘Elsanta’ and ‘Honeoye’ were harvested three times, where the third harvest only comprised overripe strawberries. For ‘Honeoye’, AF12 values increased significantly and D12 values decreased significantly with each successive harvest time. However, there were no significant differences in colour stability (AF12) between ripe and overripe strawberries within the same harvest time. In contrast, for ‘Elsanta’, overripe strawberries from the finial harvest showed the highest AF12 and lowest D12 values. However, of nectars of the first two harvests, those from ripe strawberries showed significantly higher colour stability than nectars from overripe strawberries from the same harvest.

For all cultivars collected in Austria (except for ‘Malling Allure’ which had no overripe samples), AF0 and AF12 values in general were higher for nectar produced from strawberries harvested at overripe stage than in ripe stage and later in the season than earlier in the season. This was most obvious in the case of ‘Allegro’, ‘Malling Centenary’, ‘Clery’ ‘Sibilla’, and ‘Limalexia’ (Figure 1). Cultivars ‘Allegro’ and ‘Clery’ showed no significant differences in D12 values between any of the nectar samples, indicating that colour stability is not influenced by ripening stage or harvest time (Figure 2). Nectars from ripe strawberries of ‘Malling Centenary’ from the first harvest led to significantly less colour stable nectars (high D12 value) compared to the samples from other ripening stages or harvest times. Nectars from overripe strawberries of the second harvest of ‘Sibilla’ showed a significant higher colour stability (low D12 value) than those from the other nectars. ‘Rendezvous’ was unique among the cultivars, with the most stable nectar produced from ripe strawberries from the first harvest. ‘Limalexia’ strawberries from the second harvest exhibited greater stability, with nectars from overripe strawberries being more stable than those from ripe strawberries at both harvest points. ‘Faith’ showed no significant difference between the ripe samples of the two harvests, but the nectars from the overripe strawberries from the first harvest were more stable than those from the second.

Different cultivars were harvested on different dates, since each cultivar has its own harvest period. For example, the first harvest of the cultivar ‘Faith’ was collected on the same day as the second harvests of ‘Malling Centenary’ and ‘Allegro’. There were significant positive correlations between harvest time and both AF0 (r = 0.58 p < 0.001) and AF12 (r = 0.31 p = 0.034), but no significant correlation with D12 (r = −0.04, p = 0.81) (Table S2). These results support the finding that a later harvest time leads to strawberry nectars with a more acceptable colour at time of production, though it has less impact on the colour stability of these nectars.

3.2. Aroma Compounds

All puree samples from Poland were analysed for aroma compounds. From the Austrian samples, due to a small sample size due to environmental factors limiting the quantity of strawberries available from these cultivars, as well as the large quantities necessary to produce sufficient nectars for colour analysis, aroma analysis was not conducted on the second harvest of ‘Clery’ or the overripe strawberries from the first harvest of ‘Limalexia’ and ‘Sibilla’. As a result, 42 samples of strawberries were analysed for their aroma compounds.

The average concentrations and percentages of the different aroma compounds for each strawberry puree sample are shown in Table S3. The total concentration of mesifuran and furaneol detected by GC-FID are shown in Figure 3, and the concentration of the aroma compounds detected by GC-MS are shown in Figure 4. All the compounds were detected in every strawberry puree sample, although their concentration and profile in some cases varied greatly depending on the cultivar, as well as the ripening stage and harvest time, whereas for other cultivars, variation between samples were low. Overall, mesifuran, furaneol, methyl butanoate, and ethyl butanoate were the major aroma compounds measured in the samples (Figure 3 and Figure 4). In individual samples, the highest cumulative amounts of aroma compounds were found in puree of ‘Sibilla’ (21.556 mg/L, overripe, 29 June 2021; 18.981 mg/L, ripe, 29 June 2021) and ‘Faith’ (14.429 mg/L, overripe, 5 July 2021) and the lowest amounts were found for ‘Rumba’ (1.645 mg/L, overripe, 24 June 2021; 1.712 mg/L, ripe, 24 June 2021) and ‘Malling Centenary’ (1.944 mg/L, ripe, 14 June 2021).

3.2.1. Furanones (Mesifuran and Furaneol)

Cultivar

Both the concentration and proportion of mesifuran and furaneol varied between the cultivars (Figure 3, Table S3). However, because concentrations of mesifuran and furaneol for most cultivars also varied greatly between individual ripening stages and harvest times, comparisons of mean values are not always meaningful. Therefore, if looking at individual samples, the cultivars with the highest mesifuran concentrations were ‘Malling Centenary’ (12.301 mg/L, overripe, 29 June 2021) and ‘Malling Allure’ (8.028 mg/L, ripe, 5 July 2021) and the cultivars with the lowest mesifuran concentrations were ‘Clery’ (0.064 mg/L, ripe, 14 June 2021) and ‘Sibilla’ (0.134 mg/L, ripe, 16 June 2021). For furaneol, the highest concentrations were found in ‘Faith’ (27.286 mg/L, overripe, 5 July 2021; 24.874 mg/L, overripe, 29 June 2021 mg/L) and ‘Malling Centenary’ (22.614 mg/L, overripe, 29 June 2021) and the lowest concentrations were found in ‘Rumba’ (1.346 mg/L, overripe, 24 June 2021; 1.678 mg/L, ripe, 24 June 2021) and ‘Sibilla’ (1.721, ripe, 16 June 2021). The highest amounts of furanones were found in ‘Malling Centenary’ (34.915 mg/L, overripe, 29 June 2021) and ‘Faith’ (32.190 mg/L, overripe, 5 July 2021; 29.093 mg/L, overripe, 29 June 2021) and the lowest amounts were found in ‘Rumba’ (1.627 mg/L, overripe, 24 June 2021; 2.034 mg/L, ripe, 24 June 2021) and ‘Sibilla’ (1.855 mg/L, ripe, 16 June 2021).

Ripening Stage and Harvest Time

The Polish puree samples from ‘Marmolada’, ‘Roxana’ and ‘Rumba’ showed no significant differences in furanones between ripe and overripe strawberries (Figure 3, Table S3). ‘Honeoye’ showed an increase in concentration between the first and second harvests, but no significant difference between ripe and overripe strawberries, with the proportions of each of the furanones remaining unchanged. ‘Elsanta’ was an outlier, showing the highest concentration of aroma compounds in puree from ripe strawberries from the first harvest, which was significantly higher than the concentration in both ripe and overripe strawberries from the second harvest.

For the Austrian samples ‘Malling Centenary’, ‘Allegro’, ‘Sibilla’, ‘Rendezvous’, and ‘Faith’, the concentration of furanones was significantly higher in strawberry puree from later harvest times. In particular, puree from overripe strawberries from ‘Faith’ showed a remarkably higher (approximately nine-fold) concentration of furanones compared to those from ripe strawberries.

3.2.2. Esters

The esters measured in this study (methyl butanoate, ethyl butanoate, butyl ethanoate, isoamyl acetate, methyl hexanoate, and hexyl acetate) comprised between 22% and 97% of the total concentration of aroma compounds measured by GC-MS (Figure 4, Table S3). Of these, methyl butanoate and ethyl butanoate accounted for the majority of esters in all the tested samples, although the exact proportions varied across different samples.

Cultivar

Looking at individual samples, the cultivars with the highest concentrations of esters were ‘Sibilla’ (20.943 mg/L, overripe, 29 June 2021; 18.183 mg/L, ripe, 29 June 2021), and ‘Malling Centenary’ (12.836 mg/L, overripe, 29 June 2021). In contrast, the concentrations were lowest in the Polish samples ‘Honeoye’ (0.573 mg/L, overripe, 21 June 2021; 0.593 mg/L, ripe, 21 June 2021) and ‘Rumba’ (0.808 mg/L, ripe, 24 June 2021).

Ripening Stage and Harvest Time

The Polish puree samples from ‘Rumba’, ‘Roxana’, and ‘Marmolada’ had no significant differences in either concentration or proportion of any esters between ripe and overripe strawberries. For ‘Honeoye’ and ‘Elsanta’, there were no significant differences in ester concentrations between the different harvest times and ripening stages. However, the proportion of ethyl butanoate was higher in the second harvest for ‘Honeoye’ and highest in the third harvest for ‘Elsanta’.

For the Austrian puree samples ‘Allegro’, ‘Sibilla’, ‘Malling Centenary’, ‘Rendezvous’, and ‘Faith’, methyl butanoate concentrations were significantly higher in overripe strawberries from the second harvest compared with ripe strawberries from the first harvest. ‘Allegro’ and ‘Clery’ showed higher levels of methyl butanoate in overripe strawberries, while ‘Rendezvous’ had higher concentrations in fruit from the second harvest. Ethyl butanoate concentrations were higher in the second harvest for ‘Allegro’, ‘Malling Centenary’, and ‘Sibilla’. The proportion of methyl and ethyl butanoate varied across harvests, with a higher proportion of ethyl butanoate in the second harvest for ‘Malling Centenary’ and ‘Allegro’, while ‘Rendezvous’ had a lower proportion in the second harvest. In contrast, ‘Clery’ and ‘Sibilla’ had higher proportions of ethyl butanoate in overripe strawberries. Butyl ethanoate concentrations were significantly lower in first-harvest ripe fruit for ‘Malling Centenary’ and ‘Faith’. Isoamyl acetate levels were significantly higher in the second harvest for ‘Allegro’ and ‘Sibilla’, while methyl hexanoate was significantly higher in overripe strawberries from the second harvest for ‘Faith’. Hexyl acetate did not have significant differences between ripening stage or harvest time for any cultivars.

3.2.3. C6 Compounds and Linalool

C6 compounds trans-2-hexenol, hexanal, and trans-2-hexenal were detected in much lower concentrations, contributing between 0.6% and 10% of total compounds detected by GC-MS (Figure 4, Table S3). Linalool levels were also very variable, making up between 0.3% and 20%, depending on the sample.

Cultivar

Looking at individual samples, the cultivars with the highest combined concentrations of C6 compounds and linalool were ‘Honeoye’ (0.736 mg/L, overripe, 30 June 2021) and ‘Rendezvous’ (0.630 mg/L, overripe, 16 June 2021). In contrast, the concentrations were lowest in the samples of ‘Malling Centenary’ (0.078 mg/L, ripe, 14 June 2021) and ‘Allegro’ (0.112 mg/L, ripe, 14 June 2021).

Ripening Stage and Harvest Time

There were very few differences between harvest times and ripening stages for C6 compounds and linalool. Hexanal showed no significant differences for any of the cultivars, despite reports suggesting that these compounds are higher in less ripe strawberries [14,18,21]. Trans-2-hexenol even had a significantly higher concentration in overripe strawberries from the second harvest compared with ripe strawberries from the first harvest and showed a higher proportion in ‘Allegro’ ripe strawberries from the first harvest. Trans-2-hexenal was higher in both concentration and proportion in overripe strawberries from the first harvest. Linalool was more abundant in the second harvest for ‘Allegro’ and had a higher proportion in ripe ‘Clery’ strawberries and ripe first harvest ‘Allegro’ samples compared with samples from overripe strawberries.

3.2.4. Lactones

The lactones γ-decalactone and γ-dodecalactone accounted for between 1.6% and 65.2% of the total aromas measured by GC-MS (Figure 4, Table S3). The concentration and proportion of these compounds in the puree varied greatly depending on the cultivar, as well as the ripening stage and harvest time.

Cultivar

Looking at individual samples, the cultivars with the highest concentrations of lactones were Honeoye (4.280 mg/L, overripe, 30 June 2021) and Faith (3.500 mg/L, overripe, 5 July 2021) and the cultivars with the lowest concentrations were ‘Malling Allure’ (0.070 mg/L, ripe, 5 July 2021; 0.093 mg/L, ripe, 29 June 2021) and ‘Sibilla’ (0.139 mg/L, ripe, 16 June 2021).

Ripening Stage and Harvest Time

The concentration of γ-decalactone in ‘Allegro’, ‘Rendezvous’, and ‘Faith’ increased with ripeness and later harvest time, resulting in a significant difference between puree from ripe strawberries of the first harvest and puree from overripe strawberries of the second harvest. This pattern was also observed for γ-dodecalactone in cultivars ‘Allegro’ and ‘Faith’. The concentration of γ-decalactone was significantly higher in the overripe strawberries of the first harvest of ‘Malling Centenary’, but this was not the case for the second harvest. The proportion of γ-decalactone was higher in the first harvest for ‘Honeoye’ and ‘Malling Centenary’ compared to subsequent harvests, while ‘Limalexia’ had a higher proportion in the second harvest. Additionally, ‘Faith’ had a higher proportion in overripe than ripe strawberries.

3.2.5. Harvest Time

Harvest time correlated significantly (p < 0.05) with the concentration of mesifuran (r = 0.31) and isoamyl acetate (r = 0.35) and the percentage of ethyl butanoate (r = 0.32) (Table S2). This suggests that towards the end of the harvest the levels of mesifuran and isoamyl acetate increase, as well as the proportion of ethyl butanoate in the aroma composition.

3.3. Correlation of Aroma Compounds with Colour and Colour Stability

Table 1. Correlation (r) between the mean concentration (mg/L) of aroma compounds and Acceptance Factor (AF). AF0: day of nectar production; AF12: after 12 weeks of nectar storage; D12: difference in AF between the day of nectar production (AF0) and 12 weeks of nectar storage (AF12); * p < 0.05, ** p < 0.01.

	Mesifuran	Furaneol	Methyl Butanoate	Ethyl Butanoate	Butyl Ethanoate	Isoamyl Acetate	Methyl Hexanoate	Hexyl Acetate	trans-2-Hexenol	Hexanal	Linalool	trans-2-Hexenal	γ-Deca-lactone	γ-Dodeca-lactone	Total by GC-MS
AF0	0.31 *	0.54 **	0.13	0.43 **	0.37 *	0.18	−0.07	0.14	0.05	0.33 *	−0.09	0.28	0.28	0.46 **	0.41 **
AF12	0.13	0.39 *	0.11	0.33 *	0.29	0.06	0	0.13	0.11	0.35 *	0.15	0.27	0.55 **	0.45 **	0.39 **
D12	0.03	−0.23	−0.06	−0.15	−0.14	0.04	−0.05	−0.08	−0.11	−0.24	−0.25	−0.14	−0.54 **	−0.29	−0.25

shows the correlation between the mean concentration of each aroma compound in strawberry puree (mg/L) and AF0, AF12, and D12, and

Table 2. Correlation (r) between the proportion (%) of aroma compounds and Acceptance Factor (AF). AF0: day of nectar production; AF12: after 12 weeks of nectar storage; D12: difference in AF between the day of nectar production (AF0) and 12 weeks of nectar storage (AF12); * p < 0.05, ** p < 0.01.

	Mesifuran	Furaneol	Methyl Butanoate	Ethyl Butanoate	Butyl Ethanoate	Isoamyl Acetate	Methyl Hexanoate	Hexyl Acetate	trans-2-Hexenol	Hexanal	Linalool	trans-2-Hexenal	γ-Deca-lactone	γ-Dodeca-lactone
AF0	0.03	−0.03	−0.45 **	0.23	0.28	−0.13	−0.35 *	−0.08	−0.17	0.05	−0.2	0.18	0.23	0.24
AF12	−0.06	0.06	−0.32 *	0.07	0.12	−0.41 **	−0.36 *	−0.24	−0.29	−0.2	−0.18	0.07	0.34 *	0.2
D12	0.09	−0.09	0.13	0.06	0.03	0.45 **	0.25	0.26	0.27	0.3	0.11	0.03	−0.29	−0.1

shows the correlation between the proportion (%) of aroma compounds, with these parameters. The initial colour (AF0) significantly correlated (p < 0.01) with the concentration of furaneol, ethyl butanoate, γ-dodecalactone, and the total concentration of aroma compounds detected by GC-MS, indicating strawberries high in these compounds produce nectars with the most acceptable initial colours. There were also weaker significant (p < 0.05) correlations with mesifuran, butyl ethanoate, and hexanal. The only significant correlations AF0 had with the proportions of the compounds was a negative correlation with methyl butanoate (p < 0.01) and methyl hexanoate (p < 0.05), suggesting nectars with high levels of these esters had poorer initial colour.

The AF12 significantly correlated with the concentration of γ-decalactone, γ-dodecalactone, and the total concentration of aroma compounds detected by GC-MS (p < 0.01) and furaneol, hexanal, and ethyl butanoate (p < 0.05). The correlations with furaneol, γ-decalactone, and γ-dodecalactone concentration are likely due to the fact that these compounds were highest in overripe strawberries from the late harvests, which also lead to nectars with the highest AF12 values. Although hexanal showed no significant difference between ripening stage and harvest time within cultivars, it was particularly high in the cultivar ‘Honeoye’, which also had very high AF12 values, potentially explaining the correlation.

The only compound whose concentration had a significant correlation with D12 (colour stability) was γ-decalactone. This was present in high concentrations in cultivars ‘Honeoye’, ‘Rendezvous’, and ‘Faith’, which were particularly colour-stable nectars. However, γ-decalactone was low in the cultivar ‘Marmolada’, which had the most stable colour (lowest D12 value). This suggests that cultivars with naturally high γ-decalactone will not necessarily lead to colour stable nectars.

As shown in Table 2, the proportion of methyl butanoate and methyl hexanoate (p < 0.05) as well as isoamyl acetate (p < 0.01) were negatively correlated with AF12, while the proportion of isoamyl acetate positively correlated with D12 (p < 0.01). This suggests that strawberries with a high proportion, but not concentration, of isoamyl acetate produce nectars with poor colour stability. γ-Decalactone was the only compound whose proportion had a positive correlation with AF12.

4. Discussion

Strawberry nectar is a popular beverage but has a very poor colour stability, leading to a rapid degradation from bright red to a brown colour during storage, which is unacceptable for consumers (Acceptance Factor AF < 0.4). For nectar production it is therefore of great importance to choose the optimal starting material, which results in a particularly colour-stable nectar. In this study we investigated the influence of cultivar, harvest time, and ripening stage on the colour stability and aroma profile to determine if any aroma compounds could be used as markers for colour stability of nectars.

Ideally, cultivars with a high AF value at the date of nectar production (AF0) that—even more importantly—decreases only slightly during storage (low D12 value; difference between AF0 and AF12 after 12 weeks of storage) are used for production of nectars with high colour stability. Promising cultivars were mainly ‘Marmolada’ and ‘Honeoye’, and to a lesser extent also ‘Rendezvous’ ‘Faith’, ‘Allegra’, ‘Roxana’, and ‘Elsanta’. On the other hand, ‘Malling Allure’, ‘Limalexia’, and ‘Sibilla’ showed a low colour stability (D12 value) and therefore seem to not be suitable for nectar production. In our study, nectars produced from strawberries from the same cultivar tended to have better colour stability when produced from overripe strawberries from the last harvest, which is in agreement with previous studies [5,7,8,31,32].

Aroma profile and concentrations varied with cultivar, with esters being highest in the cultivar ‘Sibilla’ and lactones more prominent in cultivars ‘Honeoye’ and ‘Faith’. Generally, the lowest concentrations of aroma compounds were found in the ripe strawberries from the first harvest, while the highest concentrations were found in overripe strawberries from a later harvest. However, this pattern was cultivar-dependent, with ‘Elsanta’ showing lower aroma concentrations in overripe strawberries. An increase in total aroma concentration during ripening has been widely reported [14,16]. In particular, furaneol is known to increase to its maximum concentration as strawberries become overripe [25], which was observed in this study for some cultivars, but not all. Furaneol was the most abundant aroma component, although comparison with compounds detected by GC-MS is complicated by differences in detection methods. For compounds detected by GC-MS, the cultivar determined whether esters or lactones were more abundant.

Some cultivars in this study showed greater variation by harvest time than by ripening stage, with later harvests giving strawberries with higher concentrations of aroma compounds. This contrasts with findings by Passa et al. [17], who reported that strawberries produced the highest aroma levels in the middle of the harvest period. Our study found no significant difference in the concentration of trans-2-hexenal, trans-2-hexenol, and hexanal between overripe and ripe strawberries. These compounds are typically known to decrease during ripening and are at their lowest levels in overripe strawberries [14,18,21], suggesting that they could be able to be used to differentiate between ripening stages; however, this was not observed in the current study. Hexanal was even significantly positively correlated with the nectar colour after storage, which is contrary to what would be expected from the colour results. This is likely due to a high concentration of hexanal in ‘Honeoye’, which was a particularly colour-stable cultivar.

The inconsistencies across different cultivars with regard to their aroma compound behaviour could be due to changes in the environment between different harvests, as different cultivars have different harvest windows, and so could be exposed to different climates. Additionally, differences between ripe and overripe fruit may be different during different harvest periods, as early in the harvest, strawberries are not allowed to become as overripe as strawberries later in the harvest. This could explain why different cultivars responded differently to ripeness and harvest.

The only compound whose concentration significantly correlated with colour stability was γ-decalactone, which is known to contribute an ‘overripe’ flavour to strawberries [33]. In some cultivars, the concentration of γ-decalactone increases in overripe strawberries, which may explain its correlation with colour stability. γ-Dodecalactone and furaneol were found to be markers for a good colour after storage, and therefore long shelf-life. This shows the combined importance of these aromas, such as γ-dodecalactone in particular, has been found to be of upmost importance in consumer flavour preferences of strawberries [34], and furaneol is considered one of the most important aromas for strawberries [10,11]. Therefore, it is important to produce nectars that are high in these compounds, as they both provide a desirable flavour and are linked to good colour. Nectars high in these compounds are therefore acceptable to consumers on two different fronts.

5. Conclusions

The concentration and proportion of aroma compounds varied with cultivar and, within cultivars, with harvest time and ripening stage. Some cultivars showed greater variation with ripening stage, while others varied more with harvest time. Some cultivars (especially those samples from Poland) did not show significant variation. In general, overripe strawberries from the latest harvest had a higher concentration of aroma compounds, although this pattern was cultivar-dependent. The colour stability of nectars also varied by cultivar, with nectars generally being more stable when produced from overripe strawberries from a later harvest. As both aroma concentration and colour stability were elevated in overripe, late-harvest strawberries, several aroma compounds were correlated with an acceptable colour after storage (AF12). This study aimed to identify compounds that could be used as markers for colour stability; the only compound that had a significant correlation with the change in colour acceptance (i.e., colour stability D12) was γ-decalactone. This suggests that strawberries high in γ-decalactone tend to be more conducive to producing nectars that are more colour stable. However, this correlation between higher γ-decalactone levels and colour stability could be coincidental and unique to the cultivars used in this study, and so analysis of a wider range of cultivars would be necessary to confirm this correlation. Many compounds, including furaneol, ethyl butanoate, hexanal, γ-decalactone, and γ-dodecalactone concentrations, and the total concentration of aroma correlated significantly with a good colour of nectar stored for 12 weeks, suggesting that good colour and good aroma are linked.