Effects of convection-, vacuum-and freeze-drying on antioxidant, physicochemical properties, functional properties and storage stability of stink bean (Parkia speciosa) powder

Abstract

This study aims to investigate the antioxidant, physicochemical and functional properties of convection-, vacuum-and freeze-dried stink bean (Parkia speciosa) powder upon storage at various relative humidity (RH) at room temperature, 25 °C. Both convection- and vacuum-dried samples exhibited stronger DPPH free radical scavenging activity (7.62 ± 1.77 and 10.38 ± 0.63 mg AA/g·db respectively) and ferric ion-chelating (FIC) ability (16.55 ± 1.29 and 18.88 ± 2.36 mg/mL·db respectively) compared to the fresh and freeze-dried samples. Stink bean powder had low water solubility index, water holding capacity and oil holding capacity but it had the potential as emulsifier in food systems. Apparent formation of clumps were observed after 25-day storage of powder at RH of 43%, 54% or 75% due to absorption of moisture from surroundings. Generally, the antioxidant properties and colour of the powders deteriorated after storage especially at higher RH. Vacuum drying would be the most suitable drying method to produce stink bean powder because of its enhanced antioxidant capacities, light colour and relatively more stable after storage.

Introduction

Parkia speciosa is known as stink bean but it is also commonly known as “petai” in Malaysia. P. speciosa bears green long and twisted pods which hang in small bundles. Each pod contains flat, broad and oval-shaped green seeds that are encapsulated with seed coats (Kamisah et al. 2013). As the name suggested, stink beans are rather unique because they have strong and distinctive bitter taste and pungent odour contributed by the presence of cyclic polysulfide compounds such as hexathionine, tetrathiane, triothiolane, pentathiopane, pentathiocane and tetrathiepane. (Kamisah et al. 2013). Nevertheless, stink beans remain popular as a form of delicacy among Southeast Asians whether they are eaten cooked or raw. Stink beans are considered underutilized because they are grown and consumed traditionally for their medicinal properties (Lim 2012). Among 25 selected tropical plants studied, Wong et al. (2006) had reported the highest total phenolic content and cupric ion chelating activity in aqueous extract of stink beans, indicating high antioxidant activity. They might have the ability to reduce oxidative stress as a result of the damage by free radicals and hence, reducing the risks of inflammatory diseases, cardiovascular diseases, cancer and neurological disorders. Other research has also shown that P. speciosa seeds exhibited hypoglycaemic activity in which blood glucose level was effectively reduced in alloxan-induced diabetic rats (Jamaluddin et al. 1994).

Drying is one of the most traditional forms of food preservation. With the advancement of technology, there are various types of drying techniques available. Convection drying is one of the most practiced drying methods due to its low cost. It utilizes hot air as a drying medium by circulating in the oven and passing over the food material continuously to carry moisture away from it (Rahman and Perera 2007). However, convection drying may affect the bioactive, nutritional and sensory qualities of the food products negatively. Losses in phytochemical content and antioxidant properties have been associated with the initial enzymatic degradation of antioxidant compounds and thermal degradation of phytochemical compounds (Chan et al. 2009; Chong and Lim 2012). Gumusay et al. (2015) found that the TPC and ascorbic acid content of convection oven-dried tomatoes and gingers were significantly lower than vacuum-dried, freeze-dried and fresh ones.

Hence, other alternatives of drying techniques like vacuum drying and freeze drying may be used for better food preservation. Vacuum drying allows drying of food product under a reduced pressure condition so that water can be evaporated below boiling temperature while freeze drying involves the freezing of food material, also under reduced pressure condition so that ice crystals can be sublimed into vapour (Rahman and Perera 2007). Both vacuum drying and freeze drying allow the drying of food materials without exposure to high temperature and high oxygen level, so they are very suitable for products that are heat-labile and oxidized easily. Hence, degradation of bioactive compounds and nutraceutical components due to heat and oxidation can be minimized (Strumillo and Adamiec 1996). In fact, a study had shown that vacuum oven-dried rosehip has higher retention level of antioxidant capacity (57.5%) and phenolic compounds (25.1%) (Quintero et al. 2014). Studies had also shown that there was no significant difference in the TPC and ascorbic acid content between fresh and freeze-dried tomatoes (Gümüşay et al. 2015; Chang et al. 2006). In addition to that, the retention level of ascorbic acid was reported to be more than 60% most of the time in freeze-dried products (Karam et al. 2016).

With the increasing health consciousness in today’s era, new natural ingredients, particularly, fruits and vegetables extracts and powders become more demanding in the food industry. For example, fruit and vegetable powders have been used in the beverage industry as functional food additives to improve the nutritional value of product, as flavouring agent or natural colouring agent (Karam et al. 2016). In the case of stink beans, since their medicinal properties like antioxidant activity and hypoglycaemic activity had been reported in previous works (Kamisah et al. 2013; Jamaluddin et al. 1994), they can probably be converted into powder to be used as ingredients in food. Stink beans may be bitter and have a peculiar smell similar to bitter melon powders but the latter had been reported to be incorporated into muffin due to their functional properties (An 2014). Hence, stink beans are potentially made into powder for future food applications. Gan and Latiff (2011) had shown that stink bean pod powders might have the potential as functional powder in food application due to its antioxidant activities. However, to date, no research has been conducted on producing stink bean powders.

Therefore, the objectives of this research were to investigate the effects of convection-, vacuum- and freeze-drying on the antioxidant properties of stink bean powders and to characterize their physicochemical and functional properties. The effects of storage conditions on the physical appearances and antioxidant properties of the stink bean powders processed by three different drying methods were also determined.

Materials and methods

Raw materials

Bundles of stink bean pods of 2–3 weeks upon maturity of plant were bought from a local wet market in Pudu, Kuala Lumpur to obtain a total of about 1.8 kg of stink bean samples for the whole study.

Chemicals

Methanol (Fisher Scientific UK, AR grade) was used for the DPPH assay and extraction of seeds; Folin–Ciocalteau’s phenol reagent (Merck), anhydrous sodium carbonate (Friendmann Schmidt, AR grade), gallic acid (R&M Chemicals, 99%) for TPC assay; potassium dihydrogen phosphate and disodium hydrogen phosphate (R&M Chemicals, AR grade), potassium ferricyanide (Acros Organics, 99%), trichloroacetic acid (Fisher Scientific UK, 99%), iron(III) chloride (R&M Chemicals, AR grade) for FRAP assay; 2,2-diphenyl-1-picrylhydrazyl (Sigma-Aldrich, 90%) and L-ascorbic acid (R&M chemicals, AR grade) for DPPH assay; iron (II) sulfate (R&M Chemicals, AR grade) and ferrozine (Acros Organics, 98%) for FIC assay; sodium chloride (R&M Chemicals, AR grade), magnesium chloride hexahydrate (Friendmann Schimdt, AR grade), potassium carbonate (Friendmann Schimdt, AR grade) and magnesium nitrate hexahydrate (Friendmann Schimdt, AR grade) for storage stability at different relative humidities test; methanol HPLC grade (Friendemann Schimdt) and formic acid (R&M Chemicals, 98–100%) for HPLC analysis.

Sample preparation

Stink beans that were encapsulated with seed coats were removed from their pods. The seed coats were then peeled and removed from the seeds. The stink beans were split into halves to facilitate drying. The stink beans were then kept in a tightly closed plastic container and stored at − 20 °C before further use. The initial moisture content of the stink bean samples is 74%.

Different drying treatments on stink beans

Approximately 200 g of stink bean samples are prepared for each drying treatment. Stink beans were dried in a convection oven (Memmert, UF 110, Germany) at 60 °C for 48 h, operated at a maximum fan speed while for vacuum drying, stink beans were dried for 72 h in a vacuum oven (Binder, VD 115, Germany) connected to a vacuum pump (KNF, Laboport, USA), operated at a minimum pressure of 0.12 inch Hg and at 60 °C. Stink beans were also lyophilised for 48 h in a vacuum flask at about 0.100 mBar and − 85 °C in a benchtop freeze-dryer (LabConco, FreeZone 4.5 Liter, USA). The drying time was determined based on the time taken to dry the samples to constant weight, that is around 5% of moisture content. The dried stink beans were ground into powder using a mixer grinder (Panasonic, MX-AC210S, Malaysia) and then sieved (0.6 mm mesh size). The powder samples were kept in amber bottles and stored in desiccators containing silica gels at room temperature, 25 °C before further analysis. Fresh stink beans that were not dried were used as control.

Extraction of stink beans for antioxidant activity analysis

Extraction of stink beans was carried out using methanol based on 1 g of dried powder samples in 50 mL of 50% (v/v) methanol. For control or fresh samples, about 3.5 g of fresh stink beans (equivalent to 1 g of dry basis) were crushed into homogenate using a pestle in a mortar and mixed with 50% (v/v) methanol. All the samples in 50 mL of 50% (v/v) methanol were subjected to continuous shaking at 25 °C and 200 rpm in an incubator shaker (Labwit, ZHWY-200B, Shanghai, China) for1 h. Extracts were filtered under suction using a Whatman filter paper No. 3 and stored at – 80 °C for further analyses.

Total phenolic content (TPC)

The TPC was determined using the Folin-Ciocalteu method as described by Chong and Lim (2012). Three hundred microliters of samples was added into test tubes, followed by 1.5 mL of 10% (v/v) Folin-Ciocalteu’s reagent and 1.2 mL of 7.5% (w/v) sodium carbonate. Test tubes were subsequently left to stand for 30 min at ambient temperature before measuring the absorbance at 765 nm using a UV–Vis spectrophotometer (Perkin Elmer, Lambda 25, USA). A calibration curve was obtained with gallic acid in which the equation was y = 0.0115x−0.005 (R² = 0.9998), where y represents absorbance and x, concentration of gallic acid in mg/L. The TPC was expressed in milligrams of gallic acid equivalent (GAE) per 1 g dry basis (mg GAE/g·db) of sample.

Determination of antioxidant activity

DPPH radical-scavenging activity assay

The 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical-scavenging activity of the sample extracts was determined based on the method described by Chong and Lim (2012). Different dilutions of 1 mL extracts were added to 2 mL of DPPH (5.9 mg/100 mL of 100% methanol). The tubes were left to stand for 30 min at room temperature and the absorbance was then measured at 517 nm. The percentage of DPPH scavenged was obtained by using the formula;

$$\text{Percentage of DPPH scavenged}=\frac{{\text{A}}_{\text{control}}{\text{- A}}_{\text{sample}}}{{\text{A}}_{\text{control}}}\times 100;$$

where A_control was the absorbance of the control while A_sample was the absorbance of the sample extracts.

Based on the study by Chong and Lim (2012), the DPPH free radical-scavenging activity was expressed as ascorbic acid equivalent antioxidant capacity (AEAC) in mg ascorbic acid/g·db calculated using the formula as follows:

$$\text{AEAC}(\text{mg AA/g}\times \text{db})=\frac{{\text{IC}}_{50\left(\text{ascorbate}\right)}}{{\text{IC}}_{50\left(\text{sample}\right)}}\times {10}^3;$$

where the IC₅₀ of ascorbic acid was 0.00366 mg/mL, obtained from a calibration curve of ascorbic acid at 50% DPPH radical scavenging activity.

Ferric-reducing antioxidant power activity (FRAP)

In assessing the FRAP, different dilutions of 1 mL of sample were added to 2.5 mL of phosphate buffer (0.2 M; pH 6.6) and 2.5 mL of 1% (w/v) of potassium ferricyanide (Chan et al. 2009). The mixture was incubated in a water bath at 50 °C for 20 min. Upon incubation, 2.5 mL of 10% (w/v) trichloroacetic acid was added into the mixture to stop the reaction. Subsequently, the mixture was separated into aliquots of 2.5 mL and diluted with 2.5 mL of water. Five hundred microliters of 0.1% (w/v) ferric chloride solution was then added to each diluted aliquot. Upon leaving to stand for 30 min at room temperature, the absorbance was measured at 700 nm. FRAP activity was expressed as milligrams of gallic acid equivalent per 1 g of dry basis sample (mg GAE/g·db) based on the calibration equation with gallic acid, y = 16.606x (R² = 0.9921), where y is absorbance and x is concentration of gallic acid in mg/mL.

Ferrous ion-chelating (FIC) ability

FIC ability of the sample extracts was determined according to Chong and Lim (2012). Sample extracts were subjected to a rotary evaporator (EYELA, N-1000, Tokyo Rikakikai, Japan) to concentrate the extract to about 3–4 times. Different dilutions (0.2, 0.5 and 1 mL) of 1 mL of extracts were added to 1 mL of iron (II) sulfate (0.1 mM), followed by 1 mL of ferrozine (0.25 mM). After 10 min, the absorbance was measured at 562 nm. When measuring the absorbance, the samples were paired with their respective blanks. The ability of extracts to chelate ferrous ions was calculated as follows:

$$\text{Chelating effect}\%=\left(1-\frac{A_{\mathit{sample}}}{A_{\mathit{control}}}\right)\times 100$$

where A_sample and A_control are absorbance of the sample and negative control, respectively.

Results were being expressed as 50% chelating effective concentration, CEC₅₀ (mg/mL).

HPLC

Sample preparation for HPLC

Dried stink beans of different drying techniques were prepared and extracted as described earlier but the ratio of sample extraction was based on 10 grams of dried samples to 100 mL of 50% (v/v) methanol instead. The extraction procedure was repeated for a total of three times. The sample extracts were then subjected to the rotary evaporator maintained at 30 °C at low pressure before they were freeze-dried for 48 h. The freeze-dried extracts were then stored in − 20 °C freezer before further HPLC analyses.

For HPLC analysis, the freeze-dried (powder) samples were weighed and dissolved in 80% water/20% methanol to obtain a concentration of 5 mg/mL. The samples were sonicated and then filtered using 0.22 µm, Nylon syringe filters into HPLC vials.

Sample analysis

A High Performance Liquid Chromatography system (HPLC) [Agilent 1260 Infinity HPLC System, USA] equipped with a quaternary pump (G1311B, 1260 Quat Pump), a standard autosampler (G1329B, 1260 DL ALS), a thermostated column compartment (G1316A, 1260 TCC) and a diode array detector (G4212B, 1260 DAD) were used. The separation was performed at room temperature using the C-18 column (Agilent, Poroshell 120 EC-C18, 4.6 × 150 mm 2.7-Micron). At a flow rate of 1.0 mL/min and injection volume of 50 µL, using solvent A (water + 0.5% formic acid) and solvent B: (methanol + 0.5% formic acid), the gradient programme was performed as follows: 5% to 50% B in 60 min, it reached 80% B at 100 min then to 100% B at 115 min, it maintained at 100% B till 125 min and finally went back down to 5% B at 130 min.

Physicochemical and functional properties of stink beans powder

Moisture content

The moisture content of powder samples was determined according to AOAC official method 935.29 with slight modifications (AOAC International 1995). About 3.50 g of powder sample was weighed into a crucible before drying in a convection oven overnight at about 100 °C. The moisture content (wet basis) of the powder was calculated as follows:

$$\text{Moisture content},\text{wb}\left(\%\right)=\frac{m_b-m_a}{m_b}\times 100\%$$

where $m_b$ and $m_a$ are the mass of powder before drying and after drying at 100 °C, respectively.

Water activity

The water activity (a_w) of powder samples was measured using a water activity meter (Aqualab 4TE, Decagon Devices, USA).

Bulk density

The bulk density of powder samples was determined as described by Tuyen et al. (2010) with slight modifications. About 2 g of powder was weighed and transferred into an empty 25-mL measuring cylinder and it was held on a vortex vibrator for 1 min. The bulk density was obtained from the ratio of mass of the powder and the volume occupied in the measuring cylinder.

Water solubility index (WSI)

The WSI of powder samples was determined according to the method described by Tuyen et al. (2010). About 2.5 g of stink bean powder were vigorously mixed with 30 mL of distilled water in a 50-mL centrifuge tube, incubated in a 37 °C water bath for 30 min. Then, the tubes were centrifuged for 20 min at 11,000 g in a bench-top centrifuge (Eppendorf, Centrifuge 5810R, Germany). The supernatant was carefully collected in a pre-weighed glass petri dish and oven dried at a temperature of 100 °C. The WSI (%) was calculated as the percentage of dried supernatant with respect to the original mass of the stink bean powder which was 2.5 g.

Water-holding capacity (WHC) or oil-holding capacity (OHC)

WHC and OHC of the powder samples were determined using the method of Gan and Latiff (2011) with slight modifications. About 1 g of powder sample was weighed and added with 10 mL of distilled water or soybean oil (Soya Lite) and vigorously mixed and left for 30 min. The suspensions were then centrifuged at 2300 g for 30 min in a bench-top centrifuge (Thermoscientific, Sorvall Legend RT + , UK). WHC was expressed as gram of water held per gram of powder sample and OHC was expressed as gram of oil held per gram of powder sample. The density of soybean oil was 0.912 g/mL.

Emulsifying activity (EA) and emulsion stability (ES)

EA and ES were assessed using the method as described by Gan and Latiff (2011). About 5 g of stink bean powder was added with 100 mL of distilled water and vigorously mixed. The suspension was then homogenized with a homogenizer (IKA, T25 digital Ultra-Turrax, Germany) at 24,000 rpm for 2 min. One hundred mililiters of soybean oil (density of 0.912 g/mL) were then added to each homogenized sample and homogenized again for another 2 min using the same parameters. Ten mililiters of the emulsions were then transferred to 15-mL centrifuge tubes to be centrifuged at 1200 g for 5 min and the emulsion volume was measured. For the determination of the emulsion stability, the emulsions prepared earlier were heated at 80 °C in a water bath for 30 min. Then, the emulsion was cooled at room temperature and again, 10 mL of the emulsions were transferred to 15-mL centrifuge tubes to be centrifuged at 1200 g for 5 min. Emulsion activity and emulsifying stability was calculated as follows:

$$\text{EA or ES}=\frac{\text{Emulsified layer volume after centrifuge}}{\text{Original volume of the emulsion in the centrifuge tube}}\times 100$$

Storage stability of stink bean powder

Storage stability of powder samples was determined according to Kuan et al. (2016). About 3 g of powder samples was stored in desiccators containing saturated salt solutions equilibrated to relative humidities (RH) of 33% (magnesium chloride), 43% (potassium carbonate), 53% (magnesium nitrate) and 75% (sodium chloride) for 25 days at room temperature of about 25 °C. During the storage and handling of dried powders in food industries, dried powders may be exposed to different environmental conditions, which include the RHs that have been selected for this study (Carolina et al. 2007). After the storage period of 25 days, the powder samples were observed and their physical appearances were photographed. The antioxidant properties of the powder after 25-day storage were assessed based on their TPC, AEAC and FRAP, as described earlier. Fresh stink bean powder samples were used as controls (before storage).

Colour determination

Colour analysis of the powder was carried out using a Colorflex Spectrophotometer (HunterLab, ColorFlex EZ) using the CIE L*, a* and b* colour scale, in which L* shows the lightness of the sample ranging 0 (black) to 100 (white), +a* indicates the reddish colour while -a* indicates the greenish colour, +b* represents yellowish colours and -b* represents bluish colours. Chroma (C) indicates the colour intensity, in which it is calculated as C = $\sqrt{\left(a{\ast }^2+b{\ast }^2\right)}$; while the hue angle (Hº) = tan⁻¹ $\left(\frac{b\ast }{a\ast }\right)$. The hue angle values range from 0º or 360º (red hue), 90º (yellow hue), 180º (green hue) to 270º (blue hue) (Tuyen et al. 2010).

Statistical analyses

All analyses of the samples were carried out in three replications (n = 3). All data were presented as mean ± SD. The data were analysed using Statistical Package for Social Sciences (SPSS) version 23. Statistical analyses were evaluated by one-way analysis of variance (ANOVA) and post hoc Tukey HSD (Honestly Significant Difference) test in which values with P < 0.05 were considered statistically significant.

Results and discussion

Antioxidant properties

Total phenolic content (TPC), DPPH radical scavenging activity, ferric reducing power (FRAP) and ferrous ion-chelating (FIC) ability

All stink bean powders showed no significant differences (P $\ge$ 0.05) in their TPC and FRAP across their controls and different drying methods. This implies that the phenolic compounds in samples are relatively stable despite being subjected to heat treatment at 60 °C through convection- and vacuum-drying. The temperature, 60 °C was selected for the convection- and vacuum-drying simply because Mercer and Myhara (2012) recommended that the drying of fruits and vegetables should not exceed 60 °C to ensure that nutritional quality is retained. During convection-drying, the movement of air acts as a drying medium to help to facilitate and speed up the drying process and hence, only 48 h are required to completely remove about 74% of moisture content from the seeds compared to vacuum-drying which takes 72 h as it operates in the absence of air to create low pressure (Rahman and Perera 2007). However, both convection- and vacuum-dried stink bean powders showed significant increase (P < 0.05) in their AEAC, indicating stronger DPPH free radical-scavenging activity (Table 1). This result probably indicates that phenolic compounds in stink beans are related to their reducing power but not their free-radical scavenging activities. Besides phenolic compounds, other compounds like aromatic amines and sulfur-containing compounds may also react directly with free radicals through several mechanisms such as donating protons, addition of radicals, redox reaction and radical recombination (Yu et al. 2002). In a study, Leong and Shui (2002) had classified the strength of free radical-scavenging of fruits from low to extremely high and according to the study, dried stink beans are comparable to strawberries (4.72 mg AA/100 g) in which they are considered high in free radical-scavenging activity. Similarly, stronger FIC ability was observed in convection- and vacuum-dried stink beans as they had lower CEC₅₀ (Table 1). Since both oven-drying methods are capable of enhancing the AEAC and FIC ability, it was suggested that heat treatment might facilitate the release of bound antioxidants. Bound antioxidants refer to antioxidants that are chemically bound to food matrix or physically entrapped in the food matrix (Brewer 2011). Heat treatment may facilitate the release of these antioxidants causing increase in the antioxidant activity in stink beans, resulting in the release of antioxidant compounds that exhibit both free radical-scavenging activities and metal chelating ability (Brewer 2011). This finding was in accordance with Kang et al. (2006), which reported the increase in radical scavenging activities and capacities in ginseng after oven-drying. Chang et al. (2006) had shown that tomatoes exhibited stronger FIC ability of up to 16% increase after oven-drying. Antioxidant properties of the freeze-dried samples were not significantly affected as they are the same as that of the control samples in terms of their TPC and DPPH free radical-scavenging activities. This can be supported by Chang et al. (2006) and Gümüşay et al. (2015) who reported that there was no significant difference in the TPC and ascorbic acid content between fresh and freeze-dried tomatoes. In freeze drying, the food material is frozen and the pressure is lowered to below its triple point, causing the ice crystals to sublime to become vapour. As freeze-drying is operated at a very low processing temperature (absence of heat) under low pressure in the absence of oxygen, thus degradation of phenolic compounds may be minimized and preserved (Strumillo and Adamiec 1996).

Table 1.

Total phenolic content (TPC), DPPH radical scavenging activity (AEAC), ferric-reducing antioxidant power (FRAP) and ferrous ion-chelating ability (FIC) of stink beans before (control) and after convection-, vacuum- and freeze-dried (mean ± SD, n = 3)

Drying method	TPC (mg GAE/g·db)	AEAC (mg AA/g·db)	FRAP (mg GAE/g·db)	FIC: CEC50 (mg/mL·db)
Control	7.09 ± 0.76a	2.95 ± 0.10a	5.70 ± 0.95a	> 25.00a
Convection	6.59 ± 1.48a	7.62 ± 1.77b	6.50 ± 0.72a	16.55 ± 1.29b
Vacuum	5.94 ± 0.75a	10.38 ± 0.63c	6.40 ± 0.44a	18.88 ± 2.36b
Freeze drying	6.87 ± 1.08a	5.13 ± 0.11a	7.17 ± 0.72a	23.48 ± 0.93c

AA ascorbic acid; CEC₅₀, effective concentration at 50% chelating ability. Mean values (n = 3) with different superscript small letters (a–d) in each column indicate significant differences between the control samples and the convection-dried, vacuum-dried and freeze-dried samples (P < 0.05)

HPLC analysis

HPLC chromatogram (Fig. 1) showed that most compounds in the seeds of stink beans were relatively polar as they were detected at earlier retention time when the mobile phase was about 80% water based on its gradient programme. Interestingly, the chromatogram showed greater amounts of compounds in convection oven- and vacuum-dried samples compared to that of the control and freeze-dried seeds. Changes in amounts of compounds could be observed between retention times (RT) of 1.5 and 5 min, particularly, an unknown compound at RT = 1.57 min appeared to increase significantly in amount after both convection oven- and vacuum-drying. This result is comparable to the increase in AEAC of stink beans after convection and vacuum drying, thus the unknown compounds might possibly possess free radical scavenging activities. A further study has to be carried out to identify this compound.

Fig. 1.

Overlay chromatograms (280 nm) showing greater amount of compounds in b convection- and c vacuum-dried than a control and d freeze-dried stink beans

In terms of its antioxidant properties, vacuum-drying is the most appropriate drying technique for stink bean as not only that it helps to retain the total phenolic count but it also significantly increases its DPPH free radical scavenging activity and its FIC ability.

Physicochemical and functional properties

Moisture content and water activity

The moisture content of the stink bean powders of various drying methods was relatively moderate, falling within the range of about 4–5% (Table 2). The water activity of the powders obtained was low, ranging from 0.37 to 0.44. At water activity below 0.6, the powders can be considered as microbiologically stable. The properties of the stink bean powders are comparable to that of whole egg powder which has 5% moisture content and water activity of 0.40 (Fontana 2000).

Table 2.

The physicochemical and functional properties of convection oven-, vacuum oven- and freeze-dried stink bean powders (mean ± SD, n = 3)

Properties	Convection oven- dried powder	Vacuum oven- dried powder	Freeze-dried powder	Significance
Moisture content, wb (%)	5.01 ± 0.26a	4.55 ± 0.18a	4.45 ± 0.57a	ns
Water activity	0.44 ± 0.01a	0.40 ± 0.01b	0.37 ± 0.02b	P < 0.05
Bulk density (g/mL)	0.43 ± 0.01a	0.42 ± 0.01a	0.42 ± 0.01a	ns
Water solubility index (%)	31.13 ± 4.50a	31.83 ± 3.10a	32.28 ± 2.83a	ns
Water holding capacity (g/g)	2.67 ± 0.29a	2.41 ± 0.04a	2.28 ± 0.25a	ns
Oil holding capacity (g/g)	1.06 ± 0.26a	0.66 ± 0.20a	0.62 ± 0.17a	ns
Emulsifying activity (%)	50.79 ± 2.75a	60.71 ± 3.98b	58.97 ± 3.57ab	ns
Emulsifying stability (%)	56.06 ± 6.94a	57.58 ± 6.94a	61.93 ± 4.20a	ns

Mean values (n = 3) with different superscript small letters (a–c) in each row indicate significant differences between the samples of different drying methods (P < 0.05) with ‘ns’ indicating not significant

Bulk density

According to previous studies, most powders such as yam flour and Gac fruit powder commonly had bulk densities of about 0.50–0.90 g/mL (Roongruangsri and Bronlund 2016; Tuyen et al. 2010). The bulk density of stink bean powder (~ 0.42 g/mL) is considered low. This might be attributed to the particles sticking to each other due to moisture in the powder, resulting in more interspaces and a larger bulk volume (Goula and Adamopoulos 2005).

Water solubility index (WSI)

The WSI of stink bean powder was about 31–32%. However, in food application, it was suggested that it would be preferable for the powder to have a WSI of more than 50% (Roongruangsri and Bronlund 2016). The insolubility of stink bean powders in water might be attributed to the presence of water-insoluble materials such as cellulose, lignin and starch (Gan and Latiff 2011). Similar to many other bioactive compounds that are poorly water-soluble, stink bean powder warrants further study to improve its water solubility.

Water-holding capacity (WHC)

The WHC of the stink bean powders was about 2.28–2.67 g water/g sample, which was relatively low. WHC could probably be related to soluble dietary fibre content such as pectic-polysaccharides and gum. A study found that jack bean contained higher soluble dietary fibre (3.38%) than lima bean (0.77%) and at the same time, also had higher WHC than that of lima bean (Betancur-Ancona et al. 2004). This suggests that stink beans contain low soluble fibre content thereby resulting in low WHC.

Oil-holding capacity (OHC)

Table 2 shows that the OHC of stink bean powders was low, ranging from 0.62 to1.06 g oil/g sample. There are various factors that could potentially influence the OHC of dietary fibre including surface properties, charge density, thickness and hydrophobicity (Gan and Latiff 2011). Kuan and Liong (2008) suggested that OHC is associated with insoluble dietary fibre. In their study, okara and corn cob which were found to have the highest OHC and concurrently, also contain high insoluble dietary fibre content, which is mainly made up of cellulose, hemicellulose and lignin. These results imply that stink beans contain low insoluble dietary fibre and hence, low OHC.

Emulsifying activity (EA) and emulsifying stability (ES)

The stink bean powders were assessed for their potential as emulsifiers by quantifying their EA and ES. Table 2 shows that stink bean powders have moderate EA of 50.79–60.71% and moderate ES of 56.06–61.93%, which are comparable to the fibrous materials of okara (Kuan & Liong 2008). Kuan and Liong (2008) suggested that soluble polysaccharide fractions in the fibrous materials might contribute to the emulsifying properties. However, as stink bean powders have low WHC, they may contain low soluble fibre, which did not contribute to most of the emulsifying properties of stink beans. Having a protein content of about 8–9%, the emulsifying properties of stink beans are most likely contributed by their protein fractions. Proteins are known to be good emulsifiers because they are able to lower interfacial tension between hydrophobic and hydrophilic compounds in foods, thus they are able to participate in the formation of emulsions and stabilize them (Zayas 1997). The result suggests that stink bean powder could have the potential to be used as an emulsifier in food applications.

Storage stability of stink bean powders

Physical appearance

At different RH of 33%, 43%, 54% and 75%, all stink bean powders of different drying methods appeared to be physically similar (Table 3). The powder samples before storage (control) appeared to be free-flowing as they contained minimal moisture content. Changes on the physical appearance of the stink bean powders were observed after 25 days of storage at room temperature at different RH. Powder samples stored at 33% RH did not show any obvious physical changes and remained as free-flowing powder. However, at 43% RH, the powder samples began to show formation of clumps as caking occurred. At higher RH of 54% and 75%, larger clumps were observed (Table 3).

Table 3.

The physical appearance of stink bean powders before (control) and after storage of 25 days at different relative humidities (RH) at room temperature

Changes in the physical appearance of the powder samples across various RH were attributed to the migration of moisture from the surrounding into the powder samples (Lee et al. 2013). Water as a plasticizer, when absorbed by the powder, would lead to the increase in molecular mobility of soluble compounds in the food matrix, thereby allowing the formation of liquid bridges between neighbouring particles. As a result, caking and agglomeration of powder was observed (Mara Righetto and Maria Netto 2005).

Colour

Based on the powders before storage (control), freeze-dried powder showed to have the lightest colour (L* = 70.90 ± 1.19) followed by vacuum-dried (L* = 65.00 ± 0.32) and convection-dried powders (L* = 62.41 ± 0.65) (Table 4). Generally, after heat drying, samples became darker due to browning. Kamisah et al. (2013) reported that stink beans generally contain carbohydrates ranging from 13.2 to 52.9% and proteins ranging from 6.0 to 27.5%. Non-enzymatic browning may occur during drying as a result of reaction between reducing sugar (carbohydrates) and amino acids (protein) in stink beans. Concurrently, enzymatic browning could also occur under convection-drying adding to more browning. Enzyme like polyphenol oxidase (PPO) is an oxidative enzyme that would cause degradation of phenolic compounds and colour pigments (Terefe et al. 2010). The activity of enzymes varies with plant species and cultivars. In previous studies, PPO enzymes in two strawberry cultivars were found to be thermostable even after heating up to 100 °C while PPO enzymes in aubergines had shown to have minimal activity at 60 °C (Dogan & Dogan 2004; Terefe et al. 2010). These reports imply that drying at 60 °C might not be able to inactivate the oxidative enzymes completely in stink beans, leading to more browning. Therefore, in the presence of circulating air under convection-drying, enzymatic reaction might be able to take place in stink beans. Vacuum-dried powders were not as dark as convection-dried powders possibly because only non-enzymatic browning could occur during the drying at 60 °C. In the absence of air during vacuum-drying, enzymatic browning could not take place. Freeze-dried powders appeared to have the lightest colour and enhanced appearance because freeze-drying operates at extremely low temperature under vacuum conditions, so neither non-enzymatic nor enzymatic browning reactions could occur (Karam et al. 2016).

Table 4.

The colour of stink bean powders in terms of lightness (L*), a*, b*, hue angle, H and chroma, C before (control) and after storage of 25 days at different relative humidities (RH) at room temperature

Drying method	RH (%)	L*	A*	B*	Hue angle, H (°)	Chroma, C
CO	Control	62.41 ± 0.65aA	− 3.21 ± 0.05aA	40.25 ± 0.92aA	94.56 ± 0.16abA	40.38 ± 0.91aA
	33	62.73 ± 0.54aA	− 3.11 ± 0.08aA	37.74 ± 0.80bA	94.72 ± 0.20aA	37.87 ± 0.79bA
	43	60.58 ± 0.43bA	− 2.83 ± 0.06bA	38.65 ± 0.38bA	94.19 ± 0.13bA	38.76 ± 0.37aA
	54	59.79 ± 0.34bA	− 2.52 ± 0.09cA	38.60 ± 0.46bA	93.73 ± 0.11cA	38.68 ± 0.47aA
	75	57.20 ± 0.63cA	0.34 ± 0.06dA	38.62 ± 0.61bA	89.49 ± 0.08dA	38.62 ± 0.61aA
VO	Control	65.00 ± 0.32aB	− 4.45 ± 0.17aB	37.25 ± 0.54aB	96.81 ± 0.25acB	37.52 ± 0.54aB
	33	66.06 ± 0.60aB	− 4.79 ± 0.06bB	33.27 ± 0.48bB	98.19 ± 0.11bA	33.62 ± 0.48bB
	43	64.35 ± 0.25aB	− 4.75 ± 0.10bB	35.70 ± 1.25aB	97.59 ± 0.41bB	36.01 ± 1.23aB
	54	61.71 ± 0.72bB	− 4.26 ± 0.09aB	37.3 ± 0.24aB	96.51 ± 0.17cB	37.54 ± 0.23aB
	75	58.21 ± 0.70cB	− 1.77 ± 0.10cB	36.43 ± 0.90aB	92.78 ± 0.18 dB	36.47 ± 0.90aB
FD	Control	70.90 ± 1.19aC	− 9.46 ± 0.13aC	34.69 ± 0.74abC	105.26 ± 0.11aC	35.96 ± 0.75abB
	33	70.89 ± 0.66aC	− 8.95 ± 0.14bC	33.48 ± 0.55aC	104.96 ± 0.18aB	34.66 ± 0.56aB
	43	68.53 ± 1.05bC	− 8.28 ± 0.07cC	33.48 ± 0.55aC	103.89 ± 0.10bC	34.49 ± 0.55aB
	54	66.56 ± 0.43bC	− 7.68 ± 0.08dC	35.82 ± 0.32abC	102.10 ± 0.16cC	36.63 ± 0.31bB
	75	62.89 ± 0.49cC	− 6.29 ± 0.06eC	34.49 ± 0.52abC	100.33 ± 0.09dC	35.06 ± 0.52aB

CO convection oven-dried; VO vacuum oven-dried; FD freeze-dried. For the same drying method, mean values within each column with different superscript small letters (a–d) differ significantly with various RH (P < 0.05). On the other hand, for the same relative humidity, mean values within each column with different superscript capital letters (A–C) show significant differences between the different drying techniques

Upon storage of 25 days, all powder samples showed relatively stable chroma, a measure for colour saturation across different relative humidity. However, results (Table 4) showed that in increasing relative humidity, there is a significant decrease (P < 0.05) in the lightness values, L* and hue angles, H°, indicating the loss of greenness while approaching yellowness and redness. This shows that stink bean powders might undergo browning possibly due to both chemical oxidative reaction and enzymatic browning. As discussed earlier, enzymes might not be completely inactivated via drying. Hence, upon storage and exposure to air, enzyme reactions and chemical oxidations might occur resulting in further degradation of phenolic compounds and polymerization of brown pigments (Robards et al. 1999). The loss of greenness in stink bean powders during storage might be attributed to the degradation of chlorophyll by enzymes (Yamauchi and Watada 1991). It was observed that the colour of powders also became darker at higher RH because of high moisture and chemical reactions that were favoured due to the increase in substrate mobility (Liang et al. 2013). Generally, colour deterioration in all the powder samples was observed after storage of 25 days at various RH.

Antioxidant properties

In general, most stink bean powders appeared to decrease in their antioxidant properties after 25-day storage in terms of their TPC, AEAC and FRAP but minor fluctuations were observed at various relative humidity (RH) (Fig. 2). Some of these changes might be attributed to the phenolic compounds that could undergo hydrolysis, oxidation and complexation after storage (Zafrilla et al. 2003). As discussed earlier, enzyme activity and chemical oxidation may still be possible in stink bean powders during storage, thereby resulting in degradation of polyphenol and antioxidant compounds during the subsequent storage. Despite physical changes were observed in some of these powders, vacuum-dried powders seemed to be relatively stable in terms of TPC and FRAP as no drastic decrease was observed after storage as compared to other powders (Fig. 2).

Fig. 2.

The a Total phenolic content (TPC), b Ascorbic acid equivalent antioxidant capacity (AEAC), c Ferric-reducing power (FRAP) of convection oven-, vacuum-and freeze-dried stink bean powders before (control) and after storage at different RH of 33%, 43%, 54% and 75% at room temperature for 25 days (mean ± SD, n = 3). For the stink bean powders of the same drying method, mean values with different capital letters (A–C) differ significantly with various RH. For powders stored at the same RH, mean values with different superscript small letters (a–c) differ significantly with different drying methods (P < 0.05)

Conclusion

Convection- and vacuum-dried stink bean powders were found to have significantly higher AEAC and FIC ability compared to the fresh stink beans. The antioxidant properties of freeze-dried powders and the control samples seemed to have no significant differences. None of the drying methods causes loss in the antioxidant properties of the stink beans. Overall, vacuum-drying technique is the most appropriate in drying stink bean due to its increased DPPH radical-scavenging activity and FIC ability.

Regardless of drying methods, stink bean powders basically had moderate moisture content, low water activity and low bulk density. Stink bean powders were shown to be low in water solubility index, water holding capacity and oil holding capacity but they exhibited moderate emulsifying activity and emulsifying stability in which they could potentially be useful as emulsifiers for various food products applications.

After storage of 25 days at room temperature, powder samples stored at 33% RH remained free-flowing but starting at 43% RH, caking of powder was observed and at 75% RH, larger clumps were formed. Colour loss was observed after 25-day storage especially at higher RH whilst antioxidant properties were found to decrease after storage though the change of antioxidant properties varied with RH. Hence, stink bean powder would be best stored at RH below 43% in an airtight container to prevent deterioration in physical and nutritional qualities due to moisture and oxidation. Vacuum-drying would be the most suitable drying method to produce stink bean powder because of its enhanced antioxidant capacities, light colour and relatively stable in terms of its antioxidant properties after storage. The potential of stink bean powders in food applications can be further assessed and explored through application in various food matrices.