Food Chemistry 306 (2020) 125635

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Converting industrial organic waste from the cold-pressed avocado oil T

production line into a potential food preservative
Rahul Permala, Wee Leong Changb, Brent Sealea, Nazimah Hamida, Rothman Kama,
⁎

a
Food Science Department, Auckland University of Technology, New Zealand
b
Biomedical Science Department, Auckland University of Technology, New Zealand

ARTICLE INFO ABSTRACT

Keywords: The production of commercial cold pressed avocado oil (CPAO) generates large quantity of organic wastes such
Food preservation as pomace, seeds, peels and wastewater. During the early harvest season, for every 1000 kg of avocado fruits
Avocado oil processing processed, roughly 80 kg of oil is produced and wastewater accounted for the highest proportion (500 kg).
Spray drying Therefore, it is important to find an alternative application for this wastewater rather than its direct disposal into
SEM
landfills. Proximate analysis, total phenolic content (TPC) and antioxidant assays were conducted on the avo-
Total phenolic content
TBARS
cado wastes. Avocado wastewater (AWW) was spray dried into powder at different temperatures from 110 °C to
CUPRAC 160 °C, which concomitantly increased the TPC and antioxidant capacities of the AWW powder. The powder was
FRAP further applied as a preservative in pork sausages and was found to be effective in preventing lipid oxidation.
Folin-Ciocalteu assay
Waste conversion
Mass balance
Food processing

1. Introduction healthy blood lipid profiles (Perdomo et al., 2015). Avocado oil can also
enhance the bioavailability of fat-soluble vitamins and phytochemicals
One-third of the world’s food production for human consumption is from other fruits and vegetables that are naturally low in fat (Dreher &
either lost or wasted from farm to fork. In many cases, 39% of food Davenport, 2013).
losses occur during the manufacturing and processing of food products. Research indicates that avocado oil is abundant in phytochemicals,
These nutritious suboptimal foods are considered undesirable by the mainly chlorophylls, carotenoids and α-tocopherol (Wong et al., 2010).
consumer, based on sensory and visual deviations. As a result, the by- The antioxidative effect of these phytochemicals can help scavenge free
products are rarely used after disposal (Raak, Symmank, Zahn, radicals, as well as reduce the incidence of various diseases, including
Aschemann-Witzel, & Rohm, 2017). In recent years, production of cold- age-related macular degeneration (Woolf et al., 2009). A study by Lu
pressed avocado oil (CPAO) is on the increase accompanied by rapid et al. (2005) revealed that bioactive carotenoids from avocados, in
market expansion. Increased use of this oil has been due to its beneficial combination with diet-derived phytochemicals, may contribute to the
oxidative stability at high temperatures, similar to olive oil (Costagli & significant reduction of risk to cancer. Aside from health implications,
Betti, 2015). CPAO is also utilised for culinary purposes as it remains stable at high
The avocado (Persea americana Mill.), ‘Hass’ cultivar is mostly used temperatures with a smoke point of 250 °C, which makes it suitable for
as raw material for CPAO, as opposed to other varieties due to its su- shallow pan frying (Woolf et al., 2009). With this notion, the consumer
perior yields, fruiting characteristics, and thick skin that protects the perception of natural products as being less ‘polluted’, more nutritious
fruit during transportation (Wong et al., 2008). CPAO is made up of and better quality has contributed to its commercialisation.
approximately 10% polyunsaturated fats, 15–20% saturated fats and The process of CPAO (Fig. 1) first begins with washing the avocado
60–70% monounsaturates. According to Wong, Requejo-Jackman, and fruit and then draining it before de-stoning by separating avocado pulp
Woolf (2010), 60–80% of the monosaturated fatty acids in avocado oil from the seed and skin. Oil is extracted from the flesh as it goes through
is oleic acid. Research suggests that oleic acid can protect against in- a grinder where it is simultaneously crushed and sliced, breaking down
sulin resistance, decrease inflammation in the body and regulate cell walls and releasing the oil droplets. Subsequently, the most crucial

⁎
Corresponding author.
E-mail address: rothman.kam@aut.ac.nz (R. Kam).

https://doi.org/10.1016/j.foodchem.2019.125635
Received 13 February 2019; Received in revised form 29 August 2019; Accepted 30 September 2019
Available online 03 October 2019
0308-8146/ © 2019 Elsevier Ltd. All rights reserved.
R. Permal, et al. Food Chemistry 306 (2020) 125635

Fig. 1. Process flow diagram showing the extraction of cold pressed avocado oil in New Zealand’s Kerikeri production line with a three-phase decanter system during
early harvest season. Output of avocado wastewater, skin, seed, pomace and oil were reported as average of samples collected on three separate production days
within a week.

malaxing stage is where the avocado paste is continuously stirred at a (Domínguez et al., 2014; Wang, Bostic, & Gu, 2010). As an industry
controlled temperature between 40 °C to 50 °C. This technique allows practice, Olivado Ltd, a CPAO processing company based in Kerikeri,
oil droplets to agglomerate, making the downstream separation more New Zealand, uses the pomace as animal feed for bovines. A study by
efficient. A three-phase decanter then spins this malaxed paste in a Saavedra et al. (2017) found that the avocado seed and skin can be
horizontal drum separating the pomace (fibrous material), liquid phase turned into powders using a convective drying process, creating pow-
(wastewater) and the oil (CPAO). In some cases where a two-phase dered storable commodities with high antioxidant activity. Moreover,
decanter is used, the liquid phase can be further processed in a cen- avocado seed has shown potential use as biofuel due to its high com-
trifuge to remove the remaining CPAO from the wastewater (Costagli & bustible energy content of 19.145 MJ kg−1 (Perea-Moreno, Aguilera-
Betti, 2015; Wong et al., 2008). Ureña, & Manzano-Agugliaro, 2016).
Utilisation of the cold-press mechanical extraction method of avo- The disposal of wastewater is problematic for the manufacturer
cado oil has led to a significant accumulation of by-products, including because it cannot be directed into the drains due to its high volume
seed, pomace, peel, and the most abundant being wastewater. These (about 0.45 L wastewater per kg of avocado fruit) and level of organic
suboptimal parts, especially the skin and seed, contain essential anti- material. External contractors are usually required to collect and dis-
oxidants and vitamins that are commonly discarded into landfills pose of the wastewater, which is very costly. In this study, the spray

2
R. Permal, et al. Food Chemistry 306 (2020) 125635

drying method will be used to convert the wastewater into avocado Samples were then freeze-dried separately, ground into powder using a
wastewater (AWW) powder, which could potentially be used as a Sunbeam AutoGrinder II EM0420, and stored at −18 °C before further
functional food ingredient. Spray drying is ideal for processing the analysis.
wastewater because of its efficient single-step evaporation of the fluid
material into powder with a reproducible particle size (Gharsallaoui, 2.1.4. Chemicals
Roudaut, Chambin, Voilley, & Saurel, 2007). This is the first time that Neocuproine, ascorbic acid, ammonium acetate, 2,4,6-Tris(2-pyr-
such work has been investigated. idyl)-s-triazine (TPTZ), iron (III) chloride hexahydrate, propyl gallate,
The aims of this study were to investigate the amount of waste or sodium carbonate (Na2CO3), TBA (2-thiobarbituric acid), TEP (1,1,3,3-
by-products generated as a result of CPAO production, and to produce tetraethoxypropane), and gallic acid were purchased from Sigma-
AWW spray dried powder that was further examined in terms of its Aldrich. Acetic acid, hydrochloric acid (37%), and petroleum ether
suitability for use as a natural preservative. Proximate analysis on all (boiling point: 40-60 °C) were obtained from Fisher Chemicals. Sodium
four by-products from the CPAO process (seed, skin, pomace and acetate trihydrate was purchased from LabServ, UK. TBA (tri-
wastewater) were carried out. The spray dried AWW powder was chloroacetic acid), Na2-EDTA (ethylenediaminetetraacetic acid dis-
analysed in terms of particle size, yield, total phenolic content, anti- odium salt dehydrate), and Folin-Ciocalteu phenol reagent were
oxidant content, and colour. Finally, the feasibility of incorporating this sourced from Scharlau, Spain. Copper (II) chloride dihydrate was pur-
powder into pork sausages as a natural preservative to prevent lipid chased from VWR International, US. Monopotassium phosphate from
oxidation was explored. Interchem. Sulphuric acid (98%) was purchased from Labchem, US.
Ammonium molybdate was from Univar, US. Potassium sulphate was
2. Materials and method provided by ECP Labchem, NZ, and copper (II) sulphate was purchased
from Merck, US.
2.1. Materials
2.2. Methods
2.1.1. Fresh avocado fruits and percentage dry matter
The Orangewood orchard in Northland, New Zealand supplied fresh 2.2.1. Cold-press extraction of avocado oil and measuring avocado waste
Hass avocado fruits to Olivado Ltd for cold-pressed oil extraction in late outputs
October 2018 (early season). The fruits were held at 20 °C in wooden Yang et al. (2018) reported the processing steps and operating
crates to ripen. Two experienced assessors employed by Olivado Ltd conditions for CPAO production. Ripened avocado fruit (1000 kg) were
used industry standard tactile assessments to determine fruit ripeness supplied to the Olivado Ltd processing plant, Kerikeri, New Zealand.
(White, Woolf, Harker, & Davy, 1999). The avocado fruit was gently The avocado skin and seeds were collected from the destoner (Alfal
squeezed by the assessor’s hand to determine the firmness and hence, Laval, Sweden) in 200 L plastic bins and weighed using an industrial
ripeness. Twenty-five (n = 25) ripened avocado fruits were selected at platform scale ( ± 5 kg) (WS-701, Wedderburn). The avocado pomace
random from the wooden crates for proximate analysis. To determine and wastewater outputs from the decanter were determined using a 30
the percentage dry matter, thirty unripe avocado fruits were selected at L bucket and a timer. Mass flow rates of these components were mea-
random from multiple crates. The avocado skin and seed were removed, sured by weighing the total mass (kg) accumulated in a bucket after
and the remaining flesh was dried at 65 °C until constant weight 1 min. Likewise, the amount of pure avocado oil and residual water
(Gamble et al., 2010). The percentage dry matter of these early season were determined using the bucket and timer method. All measurements
avocados were found to be 24%. were replicated three times on three separate production days during
the week and the average was taken as the final value. Once the mass
2.1.2. Collection of waste samples flow rates were known, it could be proportioned to work out the total
Avocado wastewater, seed, skin and pomace from the Hass variety mass output.
were collected from the Olivado Ltd oil processing plant in triplicates
on three separate production days during the week. The seeds and skins 2.2.2. Proximate analysis of avocado by-products
were collected directly from the destoner and placed in 26 cm × 38 cm Standard AOAC (1997) methods were used to determine fat
snap lock plastic bags (Glad, Australia). The pomace was obtained from (method 920.39), ash (method 925.09), moisture (method 925) and
the decanter. Samples were immediately stored at 4 °C in an ice bath protein (method 954.01), in the avocado by-products. Fat content was
and transported within 3.5 h to the laboratory at Auckland University of determined using petroleum ether extraction according to the Soxhlet
Technology, New Zealand. Wastewater was stored in 5 L PET plastic principle (Thiex, Anderson, & Gildemeister, 2003) using a Gerhardt
containers and refrigerated before further use. A portion of the fresh SOXTHERM unit. Ash content was calculated from the difference be-
samples was used for proximate analysis, while the remaining portion tween initial dried and final sample weight after combustion at 550 °C
was freeze-dried using the Alpha 1–2 LDplus Laboratory Freeze Dryer for 5 h using a retort furnace (Perfect Fire III, HDTP-56–55). Moisture
for 48 h at −75 °C and 1 × 10-3 mbar, and then stored at −18 °C until content was measured based on the mass difference of sample after
needed for chemical analysis. oven drying at 105 °C for 24 h. Protein content was measured using the
Kjeldahl method. Approximately 0.5 g of powder was mixed with con-
2.1.3. Preparation of pork sausages centrated sulphuric acid for digestion on a heating block at 420 °C along
Pork belly was purchased from Countdown, a local supermarket in with potassium sulphate and copper (II) sulphate as catalysts. The
New Zealand. Pork belly was the meat of choice for sausages because of sample was then distilled using NaOH and H3B3O in the Vapodest 450
its high fat content. The pork belly was minced using a Kenwood Pro distillation unit. This unit then automatically titrated the sample using
1400 mincer and equally divided into three batches (450 g). Batch 1 0.1 M of HCl. Total protein was determined according to the AOAC
was used as a control with 4% (w/w) sodium chloride (NaCl) added to standard (AOAC, 1997). A nitrogen conversion factor of 6.25 was used,
increase the water holding capacity of meat (Bernthal, Booren, & Gray, as no specific conversion factor for avocado by-products exists.
1989) to make the sausage juicier. Batch 2 contained 4% NaCl and Available carbohydrate (ACH) and dietary fibre (DF) were assessed
0.04% sodium erythorbate (E316), a common synthetic antioxidant using a kit purchased from Megazyme (K-ACHDF) (Megazyme
(preservative) for processed meat products. Batch 3 comprised of 4% International Ireland Limited, Bray, Ireland). Due to high fat content of
NaCl and 0.2% AWW powder, spray dried at 160 °C. Each batch was the dried samples (> 10%), they were first defatted using the Soxhlet
separately mixed, cased in commercial hog casing (Dunninghams, NZ), principle as described previously, to ensure both ACH and DF assays
and baked at 180 °C for 20 min using a Piron PF7005D oven (Italy). presented accurate results.

3
R. Permal, et al. Food Chemistry 306 (2020) 125635

2.2.3. Spray drying was transferred to a glass vial, mixed with 500 µl of F-C phenol reagent,
Avocado wastewater was spray dried using a small scale Buchi mini and after 5 min of reaction, 1.5 mL of 20% Na2CO3 was was added. The
spray dryer B-290, Switzerland, equipped with a 0.70 mm spray nozzle. mixture was vortexed for ten seconds, covered with aluminium foil and
From preliminary tests, it was found that setting the wastewater feed left in the dark at room temperature (22 ± 0.5 °C) for 2 h. Solutions
flow rate to a constant 5.8 g per minute, aspiration rate at 37 m3/h, and were transferred into cuvettes and read at an absorbance of 765 nm
the atomising air at 49 m3/h, gave the highest yield of AWW powder. against a water blank. Gallic acid solutions at various concentrations
The inlet temperatures were altered from 110 °C to 160 °C at 10 °C in- (0–200 mg/L) were used for calibration. The TPC of samples were ex-
crements to investigate its effects on yield using Equation (1). Outlet pressed as gallic acid equivalents (mg GAE/g) by means of a standard
temperatures varied between 62 °C to 102 °C. The dried powders from curve (R2 = 0.9995).
the cyclone and collection vessel were further freeze dried to remove
residual moisture in the sample and then put into ziplock bags and 2.2.6. Scanning electron microscopy imaging and colour analysis
stored in an airtight plastic container at −18 °C until further analysis. Particle morphology for AWW powders was evaluated using a
All spray drying experiments were replicated three times. Hitachi SU-70scanning electron microscopy (SEM). The detectors
worked at a distance of 15.5 mm (WD = 15.5 mm) from the samples
Powder collected (g )
% yield = × 100 with an accelerating voltage of 5 kV applied for each sample. Sample
Mass of total solids in sample before spray drying (g )
micrographs were represented by a 20 μm scale. A platinum coating
(1) was applied before scanning using the Hitachi E-1045 Ion Sputter.
Particle size (μm) of the SEM images were measured using the ImageJ
software (version 2.0.0-rc-43/1.52i). Further details regarding colour
2.2.4. Antioxidant analysis
analysis on the AWW powders are provided in Supplementary in-
Extraction of antioxidants was carried out based on a protocol de-
formation SI.1 & SI.2.
scribed by Santo, Nunez, and Moya (2013) with slight adjustments.
Samples were homogenised using a T25 digital Ultra-Turrax in 4 mL of
2.2.7. Lipid oxidation measurement using TBARS
50% methanol solution at 10,000 rpm for 2 min, and left to stand for
Thiobarbituric acid reactive substances (TBARS) content was de-
1 h. Tubes were centrifuged using the Vortex-Genie II at 2500 rpm for
termined colorimetrically by the method of Vyncke (1970). When lipid
15 min at 20 °C, and the resulting supernatant transferred into a 10 mL
oxidation occurs in meat, it results in several undesirable products,
volumetric flask. This step was repeated using 70% acetone solution,
including malondialdehyde (MDA), which is associated with off-fla-
after which the volumetric flask was filled to the 10 mL mark using
vours and unpleasant smell (Ghani, Barril, Bedgood, & Prenzler, 2017).
distilled water. This solution was further diluted 1:10 using distilled
High levels of MDA indicate a higher degree of oxidation. About 1 g of
water.
the meat sample was homogenised with a 10 mL solution containing
Cupric ion reducing antioxidant capacity (CUPRAC) assay was carried
7.5% trichloroacetic acid solution (TCA), 0.1% propyl gallate and 0.1%
out according to the methods and principles detailed by Özyürek et al.
EDTA-Na2 for 30 s, then filtered using a Büchner funnel and centrifuged
(2011). Sample volume of 1 mL was added into 1 mL of CuCl2·2H2O
at 2000 rpm for 15 min. TBA reagent (0.02 M thiobarbituric acid in
(0.01 M), NH4AC (1 M, pH 7), neocuproine (0.075 M), and 0.1 mL dis-
distilled water) with a volume of 5 mL was added to 5 mL of filtrate in
tilled water to yield a total volume of 4.1 mL. The solution was left to
15 mL falcon tubes. The mixture was vortexed for 1 min, incubated in a
react for 5 min at room temperature and its absorbance was measured
water bath for 40 min at 100 °C and then cooled to ambient tempera-
against a reagent blank (1 mL of neocuproine, CuCl2·2H2O, NH2Ac so-
ture. The absorbance was measured at 532 nm and 600 nm (A532 nm -
lution and 1.1 mL water) using the GE Ultrospec 7000 spectro-
A600 nm) using a spectrophotometer against a blank containing 5 mL
photometer at 450 nm. A standard curve was plotted with Trolox (5 to
TCA and 5 mL TBA. This was done to account for turbidity and in-
170 mg/L-1; R2 = 0.998). The final antioxidant activity of the sample
strumental error. TBARS was expressed as mg MDA/kg of minced
was measured as mg Trolox equivalent (TE)/100 g powder.
sausages against a standard curve (R2 = 0.9996). The percent inhibi-
Ferric reducing antioxidant power (FRAP) assay was carried out as
tion of spray dried powder and E316 against TBARS was calculated as:
detailed by Benzie and Strain (1996). The final FRAP reagent was
composed of 1 mL TPTZ (0.01 M), 1 mL of FeCl3·6H2O (0.02 M), and % inhibition =
C T
× 100
10 mL of acetate buffer (0.3 M). A sample volume of 0.1 mL was added T (2)
to 0.9 mL of distilled H2O and 2 mL of the FRAP reagent. The samples where C is the control and T is the antioxidant treatment where either
were left to react for 5 min at room temperature and were spectro- spray dried powder or E316 was added to the sausage fillings.
photometrically measured against a reagent blank (2 mL FRAP reagent
1 mL H2O), at 593 nm. Trolox solutions with concentration varying 2.2.8. Statistical analysis
from 5 to 170 mg L-1 were used to generate a standard curve with Samples were analysed in triplicates where data were expressed as
R2 = 0.997. Results were expressed as mg TE/100 g powder. mean ± standard deviation. One-way analysis of variance (ANOVA)
Phosphomolybdenum assay was performed as outlined by Ivanišová, with Tukey pairwise comparison of means was performed using the
Kačániová, Petrová, Frančáková, and Tokár (2016). From each extract, XLSTAT software (version 2018.7). A difference of p ≤ 0.05 was con-
1 mL of sample was mixed with 2.8 mL KH2PO4 (0.1 M), 6 mL H2SO4 sidered significant.
(1 M), 0.4 mL (NH4)6Mo7O24 (0.1 M) and 0.8 mL of distilled water in
glass vials. Once mixed, the samples were incubated for 120 min at 3. Results and discussion
90 °C. They were then rapidly cooled and measured against a blank for
absorbance spectrophotometrically at 700 nm. Trolox (3 to 390 mg L-1; 3.1. Mass balance on the cold-press avocado oil production line
R2 = 0.996) was used as the standard. Results were expressed as mg
TE/ 100 g powder. Fig. 1 illustrates the CPAO production line employed by Olivado Ltd
with mass balance. Although a similar method was employed by
2.2.5. Total phenolic content (TPC) by Folin–Ciocalteu’s assay Costagli and Betti (2015), Olivado Ltd used a three-phase decanter for
The Folin-Ciocalteu’s (F-C) assay was carried out as described by separating the pomace, wastewater and CPAO without further treat-
Singleton, Orthofer, and Lamuela-Raventós (1999), with some altera- ment of wastewater after the decanting stage. In the Costagli and Betti
tions. Sample extraction was carried out the same as for antioxidant (2015) study, wastewater was centrifuged to recover a small portion of
analysis (section 2.2.4). After extraction, 1 mL of sample or standard oil. This step was omitted in the Olivado Ltd’s processing line. Olivado

4
R. Permal, et al. Food Chemistry 306 (2020) 125635

Ltd sends its CPAO into a primary storage tank after extraction where proximate composition.
inert nitrogen is sparged into the oil. Sparging removes oxygen that AWW as a by-product of CPAO production is a component, which
may have dissolved into the oil and delays rancidity of the oil. has not been studied in any previous literature. Table 1 shows that this
The mass balance (Fig. 1) shows that for every 1000 kg of avocado component contained significantly higher lipid content compared to
fruit processed, around 78 kg or 85 L of CPAO is produced. The most skin, seed and pomace. Research by Wong et al. (2014) also found that
significant mass of waste was wastewater (448 kg), followed by skin the wastewater component contained the highest amount of lipid
and seeds with a combined weight of 274 kg, pomace (150 kg) and compared to the other by-products. This high level of lipid content can
residual water (50 kg). Wong, Eyres, and Ravetti (2014) also reported a be attributed to the processing stage when wastewater was first sepa-
general mass balance summary but based it on a two-phase decanter rated in the decanter (Fig. 1), where some of the oil droplets were too
system for an avocado oil processing plant. They showed that proces- small to be separated. Yang et al. (2018) monitored avocado oil droplet
sing of 796 kg of avocado fruit yielded wastewater (408 kg), skin and size with light microscopy over time during the malaxing process. They
seed (195 kg), and pomace (120 kg). In both studies the largest by- reported that even after 120 min of malaxing at 45 ± 5 °C, oil dro-
product by mass was wastewater. plets < 5 µm that were observed using a microscope, were difficult to
The total oil yield was calculated to be around 67% in this study separate in the decanter.
(refer to Supplementary information SI.3 for calculation), which is Skin and pomace had a substantial dietary fibre content of 20.1%
reasonable as Costagli and Betti (2015) reported similar yields of be- and 16.4% respectively with very low amounts of available carbohy-
tween 57 and 68%. Wong et al. (2014); Wong et al. (2010) explained drates (ACH), similar to other reported studies (Domínguez et al.,
that up to 75% of the oil could be extracted from early-season avocado 2014); Saavedra et al. (2017). On the other hand, the seed contained
(samples with 15% oil by fresh weight). However, up to 85% of the 31.7% of ACH making it an unconventional, yet viable starch source.
maximum oil (22.5% oil by fresh weight) can be extracted from late- Domínguez et al. (2014) explained that despite their high starch con-
season avocados. The majority of unextracted oil remained in the tent, the seeds have a high polyphenol concentration, which imparts a
wastewater, followed by seed, skin, and lastly pomace. bitter taste and could be toxic at higher levels.

3.2. Proximate composition in the avocado by-products 3.3. Spray drying and powder morphology

Proximate composition of avocado oil by-products is shown in Fig. 1 shows that during the early avocado harvest season, with
Table 1. The results demonstrated that avocado wastewater was the every 1000 kg of avocados used to manufacture CPAO, approximately
highest in moisture (88.3%) content followed by the pomace (82.8%), 448 kg of wastewater was discarded as waste. This makes up to almost
skin (75.3%) and seed (57.0%). The fat and ash quantified in waste- half of the total avocado fruit input. So far, no studies have char-
water were significantly higher but lowest in protein when compared to acterised this by-product and investigated its application in food. This
the other components. All proximate values were reasonably similar to study is the first to carry out physicochemical analysis of wastewater,
what has been reported by Morais et al. (2017). In retrospect, Vinha, and to convert the wastewater using spray drying into AWW powder.
Moreira, and Barreira (2013) reported that skin and seeds possessed a Inlet drying temperatures of 110 °C through to 160 °C produced pow-
lower moisture content of 69% and 54% respectively, and a higher fat ders that did not exceed above 32% yield (Table 3). Garofulić, Zorić,
content of 2.20% and 14.7% respectively. This variability in the prox- Pedisić, and Dragović-Uzelac (2016) explained that a suitable spray
imate composition of the avocado constituents could be due to various drying yield of 50% or higher is deemed commercially viable. Samples
reasons. Araújo, Rodriguez-Jasso, Ruiz, Pintado, and Aguilar (2018) from this study were solely spray dried without incorporating any mi-
stated that the constituent and proximate value of avocado is mainly croencapsulation agent leading to powder stickiness and consequently
dependent on the variety, ripeness, climate, and composition of soil and exhibited low yields with high variability.
fertiliser used, all of which would have been slightly different for each Stickiness affects yield as most of the spray dried product is attached
sampled batch in the two studies. Avocado samples from the Morais to the drying chamber and cyclone of the spray dryer. Garofulić et al.
et al. (2017) study were sourced from Brazil, whereas samples from (2016) explained that fruit juices usually exhibit stickiness and caking,
Vinha et al. (2013) were from Portugal, meaning that both fruits would due to a high amount of low molecular mass sugars such as glucose and
have been exposed to different environmental conditions. Neither of the fructose, which have low glass transition temperatures. Consequently,
studies reported that harvesting season had a significant impact on this encourages the tendency of the powder to stick the dryer walls

Table 1
Proximate composition of avocado by-products and avocado wastewater (AWW) powder.
Proximate composition of avocado flesh and by-products (dry basis % w/w)
2
Samples Dietary Fibre Available Carbohydrate Protein Ash Lipid
Skin 81.4 ± 2.0 1.2 ± 2.0 8.1 ± 0.4 3.6 ± 0.4 6.9 ± 2.0
Seed 19.5 ± 1.3 69.1 ± 1.1 4.9 ± 0.2 3.7 ± 0.2 3.7 ± 0.7
Wastewater 22.2 ± 3.4 0.9 ± 3.4 10.3 ± 7.7 17.9 ± 2.6 53.8 ± 9.4
Pomace 72.1 ± 1.2 0.6 ± 1.2 12.8 ± 1.3 7.0 ± 0.6 9.3 ± 6.4
Flesh ND 26.4 ± 2.1 6.0 ± 0.4 6.4 ± 1.6 61.3 ± 3.4
Proximate composition of avocado flesh and by-products (wet basis % w/w)
1 2
Samples Quantity Generated (kg) Moisture Content Dietary Fibre Available Carbohydrate Protein Ash Lipid
Skin 153 ± 5 75.3 ± 0.5 20.1 ± 0.5 0.3 ± 0.5 2.0 ± 0.1 0.9 ± 0.1 1.7 ± 0.5
Seed 121 ± 5 57.0 ± 0.9 8.4 ± 0.6 29.7 ± 0.5 2.1 ± 0.1 1.6 ± 0.1 1.6 ± 0.3
Wastewater 448 ± 18 88.3 ± 0.1 2.6 ± 0.4 0.1 ± 0.4 1.2 ± 0.9 2.1 ± 0.3 6.3 ± 1.1
Pomace 150 ± 9 82.8 ± 0.3 12.4 ± 0.2 0.1 ± 0.2 2.2 ± 0.2 1.2 ± 0.1 1.6 ± 1.1
Flesh 726 ± 10 76.5 ± 0.5 ND 6.2 ± 0.5 1.4 ± 0.1 1.5 ± 0.4 14.4 ± 0.8

Values represent the means of triplicates expressed in percentages (dry and wet basis, % w/w), ± depicts standard deviation of the means.
ND = no data.
1
Based on 1000 kg of avocado fruit input from early harvest season into a three-phase decanter cold-press avocado oil extraction process. Values are reported as
means from triplicate runs from three separate production days in the early season and ± depicts standard deviation of the means.
2
Samples were analysed in duplicates.

5
R. Permal, et al. Food Chemistry 306 (2020) 125635

Fig. 2. SEM images of AWW powders, spray dried at different temperatures: (A) 110 °C; (D) 140 °C; (F) 160 °C. Samples were randomised images of what was seen
under the SEM for each spray drying condition. Note that the whole scale bar is 20.0 um and each increment is 2.0 um.

leading to low product yield. Although avocado samples were higher in reported that Hass avocado peel contained a higher TPC of 25.3 CE/g
fat than sugar, high-fat products can also result in stickiness due to the (DM) compared to seed (9.5 mg CE/g). In contrast, Wang et al. (2010)
presence of low melting point triglycerides as reported by Bhandari, reported that the highest TPC was observed in avocado seeds, 5.6 g
Datta, and Howes (1997). As seen in Table 1, it can be seen that the GAE/100 g DM, and was the least in peels, 1.2 g GAE/100 g DM. This
most abundant component after moisture in wastewater were lipids. could be explained by the fact that different extraction solutions were
Table 1 also showed that wastewater was naturally low in carbo- used. Wang et al. (2010) used a mixture of acetone, water and acetic
hydrates, so it cannot efficiently encapsulate the oils. As carbohydrates acid at a ratio of 70:29.7:0.3 respectively, while Kosińska et al. (2012)
have low interfacial properties that are required for increased micro- extracted TPC using 80% methanol, at a solid to solvent ratio of 1:8,
encapsulation, gums or proteins are usually incorporated (Gharsallaoui with 2 additional methanol rinses.
et al., 2007). Chasekioglou, Goula, Adamopoulos, and Lazarides (2017) Spray dried AWW powder did not differ significantly (p < 0.05) in
similarly reported extensively lower yields of about 3% as a result of TPC amongst the different drying temperatures. However, there was a
stickiness while spray drying olive mill wastewater (OMWW). To slight increase in TPC with increasing inlet temperature. Interestingly,
overcome this, maltodextrin was added as a carbohydrate drying agent the TPC of spray dried wastewater samples were significantly higher
at a ratio of 1 (OMWW solids)/(drying aid solids) that increased the than freeze dried samples as shown in Table 2. This was similar to a
yield to 23.7%. study by Soong and Barlow (2004) where a higher phenolic content was
SEM micrographs of the spray dried powder are presented in Fig. 2. found in heated (140–180 °C) mango seed kernel powder (MSKP)
All spray dried powders appeared as agglomerates of smaller particles compared to freeze dried MSKP.
rather than separate individual particles. At lower inlet temperature
(110 °C), the powders appeared to be tightly gelled together, and the
3.5. Antioxidant properties
particle morphology was less defined. However, at 160 °C, the ag-
glomerates appeared more morphologically distinguished and sepa-
As one protocol is not precise enough to establish the complete
rated. Table 3 shows that the particle size distribution was about the
antioxidant potential of a natural extract, this study utilised three as-
same for Sample A (2.7 ± 0.57 μm) and D (2.6 ± 0.21 μm) with the
says at different pH values to evaluate the by-products of avocado oil
most significant difference seen in Sample F (3.5 ± 0.8 μm). However,
production including seed, pomace, skin and wastewater (spray dried).
it was evident that Sample F, with the highest heat application (160 °C),
Reducing power of extracts of each by-product was determined using
displayed a lower degree of coalescence, resulting in larger, more dis-
CUPRAC and FRAP assays. CUPRAC involves reduction of Cu (II) to Cu
tinguished particles. This is similar to findings reported by Dantas,
(I), whereas the FRAP method involve reduction of Fe(III) to Fe(II) by
Pasquali, Cavalcanti-Mata, Duarte, and Lisboa (2018) who spray dried
antioxidants. The third protocol looked at total antioxidant activity
avocado pulp mixed with milk sugar and maltodextrin.
using the phosphomolybdenum assay, based on the reduction of Mo(VI)
When air inlet temperature is low, the evaporation rate produces
to Mo(V).
microcapsules with increased water content, membrane density and
The skin of avocado displayed the highest antioxidant activity in
poor fluidity. Hence Sample A (110 °C) exhibited very densely packed
two (FRAP and phosphomolybdenum) out of the three assays carried
agglomerates that slowly start to coalesce together. On the other hand,
out. This was expected as several articles similarly concluded that the
a high inlet temperature causes rapid moisture evaporation conse-
skin of the avocado to be high in antioxidant activity and reducing
quently resulting in cracks in the membrane (Gharsallaoui et al., 2007),
ability (Kosińska et al., 2012; Rodríguez-Carpena, Morcuende, Andrade,
which was evident in Samples D and F. Fig. 2 shows that the mor-
Kylli, & Estévez, 2011; Wang et al., 2010). Nonetheless, the results
phology of Sample A was mostly irregular and very rough. With the
differed from a study by Vinha et al. (2013) who found that avocado
increase in spray drying temperature, Samples D and F although ag-
seeds exhibited significantly higher antioxidant activity compared to
glomerated together, showed internal spherical shape with an irregular
the skin of the avocado. Vinha et al. (2013) concluded that both skin
surface and smoother surfaces.
and seed had high antioxidant activities similar to our results sum-
marised in Table 2.
3.4. Determination of total phenolic content (TPC) Rodríguez-Carpena et al. (2011) reported that TPC and CUPRAC
assays resulted in antioxidant activity of peels and seed extracts that
The avocado skin contained the highest TPC compared to the other were considerably higher than avocado flesh extracts. This was also
CPAO by-products, and the lowest TPC was observed in pomace evident in our study as seen in Table 2, where CUPRAC and TPC values
(Table 2). Fruit peels are known to be rich in polyphenols and anti- of flesh (0.5 g TE/100 g and 0.6 g GAE/100 g powder respectively),
oxidants, as they protect the fruit from oxidative stress caused by were considerably lower compared to avocado skin, seed and spray
sunlight and high temperatures (Mokbel & Fumio, 2005). Total phe- dried powders. Wang et al. (2010) reported that the antioxidant capa-
nolic content in skin, seed and pomace were 13.7, 8.1 and 3.6 g GAE/ city of avocado seeds and peels were several folds higher than those
100 g dried mass (DM) respectively. Kosińska et al. (2012) similarly reported for raw blueberry (5.3 mg GAE/g), a fruit which is popularly

6
R. Permal, et al. Food Chemistry 306 (2020) 125635

Table 2
Antioxidant properties and total phenolic content of freeze-dried avocado by-products and AWW powder.
Samples CUPRAC FRAP Phosphomolybdenum TPC
g TE/100 g powder g TE/100 g powder g TE/100 g powder g GAE/100 g powder

Freeze-dried avocado by-products

Seed 2.1 ± 0.09d,e 7.7 ± 0.35b 8.1 ± 1.00b 8.1 ± 0.28b
Pomace 7.1 ± 0.10c 4.0 ± 0.01d 6.8 ± 1.90b,c 3.6 ± 0.13d
Skin 7.0 ± 0.12c 13.7 ± 0.76a 12.0 ± 1.19a 13.7 ± 0.40a
Flesh 0.5 ± 0.01e 0.3 ± 0.02f 0.8 ± 0.01e 0.6 ± 0.02e
Wastewater 3.3 ± 0.87d 2.0 ± 0.19e 3.2 ± 0.4d,e 1.6 ± 0.12e
AWW powder
A (110 °C) 8.8 ± 0.87b,c 3.1 ± 0.13d,e 5.0 ± 1.18c,d 3.9 ± 0.24c,d
B (120 °C) 8.0 ± 0.84c 3.6 ± 3.60d 6.5 ± 0.74b,c 4.4 ± 0.38c,d
C (130 °C) 11.3 ± 1.36a,b 4.1 ± 4.07d 7.3 ± 0.31b,c 5.1 ± 0.58c
D (140 °C) 11.1 ± 0.70a,b 4.0 ± 0.33d 7.2 ± 0.41b,c 4.7 ± 0.86c,d
E (150 °C) 12.6 ± 0.52a 4.2 ± 4.18d 7.4 ± 0.38b,c 5.0 ± 0.44c
F (160 °C) 11.1 ± 0.96a,b 5.4 ± 0.56c 7.6 ± 0.47b,c 5.1 ± 0.30c

Data are mean ± standard deviation on dry weight basis. Subscript letters in the column on each parameter do not differ statistically by Tukey’s test (p < 0.05)
where samples were analysed in triplicates. Spray dried powder parameters: see Table 2. Trolox equivalent (TE). Gallic acid equivalent (GAE).

known to have high antioxidant capacity. In this study, CUPRAC

showed the lowest antioxidant activity for seed (2.1 TE/100 g DM)
extracts, and the highest in spray-dried powder extracts (between 8.8
and 12.6 g TE/100 g powders) compared to FRAP and phosphomo-
lybdenum assays. The reducing ability of the dried powder increased
with increasing spray drying inlet temperature similar to TPC results. A
study on the effects of spray drying temperatures on lemongrass leaf
extracts similarly showed that a gradual increase in spray drying tem-
perature from 110 °C through to 150 °C led to a significant (P < 0.05)
increase in antioxidant activity (Tran & Nyugen, 2018). This pattern of
increased antioxidant activity alongside an incremental increase in inlet
temperature was seen in the CUPRAC, FRAP and phosphomolybdenum
assays employed in this study as depicted in Table 2.
Fig. 3. TBARS values (mg MDA/kg) of pork sausages that contained spray dried
avocado waste water (AWW) powder or synthetic antioxidant (E316) as pre-
3.6. TBARS content in pork sausages mixed with AWW powder servative. Subscript letters that are the same do not differ statistically using the
Tukey’s test (p > 0.05). Measurements of each sample was carried out in tri-
Lipids readily undergo oxidation reactions, driven by the degree of plicates. Error bars represent the standard deviation of means. Sample F
(160 °C) was used in this assay.
its unsaturated fatty acids and the presence of catalysts and molecular
oxygen in the environment. Several oxygenated products such as per-
oxyl and hydroperoxides are generated during lipid oxidation, which oxidation because the MDA content was significantly lower than the
potentially leads to more undesirable characteristics in meat products control sample that had the highest level of MDA (0.85 mg MDA/kg of
during storage. The thiobarbituric acid-reactive substances (TBARS) dried sausage meat). This was expected as there were no antioxidants in
assay indicates the level of secondary products produced through lipid the sausage to prevent lipid oxidation in the control. On the other hand,
oxidation (Alirezalu et al., 2018). the sausage containing E316 and AWW powder yielded 0.67 and
Preliminary analysis using the CUPRAC assay showed that the an- 0.68 mg MDA/kg of dried sausages respectively. There were no sig-
tioxidant capacity of E316 was five times that of the AWW powder nificant differences between using AWW powder at 0.20% w/w and
spray dried at 160 °C. Fig. 3 presents the degree of lipid peroxidation of synthetic E316 at 0.04% w/w (p > 0.05). The ability of AWW powder
free fatty acids using the TBARS test in three different batches of sau- to prevent lipid oxidation in the sausages would infer that there are
sages containing the following additives, 4% NaCl (control), 4% NaCl lipid soluble antioxidants present in the powder.
with 0.04% E316, and 4% NaCl with 0.20% AWW powder. The results A study by Rodríguez-Carpena et al. (2011) utilised the TBARS
indicated that both E316 and AWW powder were able to inhibit lipid assay to study the effects of seed and peel extracts from both Hass and

Table 3
Summary of spray drying parameters and colour properties of AWW powder.
Sample A Sample B Sample C Sample D Sample E Sample F

1
Inlet temperature (˚C) 110 120 130 140 150 160
1
Outlet temperature (˚C) 60–79 65–85 69–82 84–86 90–96 90–120
1
Yield (%) 21.1 ± 6.73 19.4 ± 2.96 22.6 ± 6.27 18.56 ± 2.83 18.3 ± 3.07 32 ± 7.86
Particle Size (μm) 2.7 ± 0.57 N.D N.D 2.6 ± 0.21 N.D 3.5 ± 0.8
2
Powder colour properties
L* 69.4 ± 5.11 N.D N.D 69.4 ± 1.66 N.D 64.2 ± 1.37
a* 0.2 ± 0.35 N.D N.D 0.5 ± 0.10 N.D 2.0 ± 0.70
b* 25.0 ± 0.72 N.D N.D 27.6 ± 0.70 N.D 30.9 ± 0.51

Data are mean ± standard deviation on dry weight basis.

1
Spray dried samples were performed as triplicates (n = 3)
2
Samples measured in triplicates (n = 3)

7
R. Permal, et al. Food Chemistry 306 (2020) 125635

Fuerte avocado cultivars in pork patties. Results from the study con- AOAC (1997). Official Methods of Analysis, Association of Official Analytical Chemists (16th
cluded that percentage inhibition of avocado extracts ranged from 72.4 Edition). USA: Arlington.
Araújo, R. G., Rodriguez-Jasso, R. M., Ruiz, H. A., Pintado, M. M. E., & Aguilar, C. N.
to 91.5%. These values were higher than the current study where (2018). Avocado by-products: Nutritional and functional properties. Trends in Food
percentage inhibition was calculated to be 21.7% (NaCl + E316) and Science & Technology, 80, 51–60. https://doi.org/10.1016/j.tifs.2018.07.027.
19.9% (NaCl + wastewater powder). The significant difference in oxi- Benzie, I. F. F., & Strain, J. J. (1996). The Ferric Reducing Ability of Plasma (FRAP) as a
Measure of “Antioxidant Power”: The FRAP Assay. Analytical Biochemistry, 239(1),
dation inhibition values in Rodríguez-Carpena’s study could be due to 70–76. https://doi.org/10.1006/abio.1996.0292.
the fact that TBARS values of the patties were measured after being Bernthal, P. H., Booren, A. M., & Gray, J. I. (1989). Effect of sodium chloride con-
chilled for 15 days at 5 °C. In this study TBARS was assessed after centration on pH, water-holding capacity and extractable protein of prerigor and
postrigor ground beef. Meat Science, 25(2), 143–154. https://doi.org/10.1016/0309-
samples were cooked. Additionally, the patties from Rodríguez-Carpena 1740(89)90029-6.
et al. (2011) study used seed and skin waste components, and according Bhandari, B. R., Datta, N., & Howes, T. (1997). Problems Associated With Spray Drying Of
to our results these by-products possessed a higher antioxidant activity Sugar-Rich Foods. Drying Technology, 15(2), 671–684. https://doi.org/10.1080/
07373939708917253.
and reducing ability (Table 2), which can explain the higher degree of
Chasekioglou, A. N., Goula, A. M., Adamopoulos, K. G., & Lazarides, H. N. (2017). An
oxidation inhibition in the pork patties. approach to turn olive mill wastewater into a valuable food by-product based on
spray drying in dehumidified air using drying aids. Powder Technology, 311, 376–389.
4. Conclusion https://doi.org/10.1016/j.powtec.2017.02.008.
Costagli, G., & Betti, M. (2015). Avocado oil extraction processes: Method for cold-pressed
high-quality edible oil production versus traditional production. Journal of
The present work was carried out to evaluate the proximate com- Agricultural Engineering, 46(3), 8. https://doi.org/10.4081/jae.2015.467.
position, antioxidant capacities and total phenolic content of avocado Dantas, D., Pasquali, M. A., Cavalcanti-Mata, M., Duarte, M. E., & Lisboa, H. M. (2018).
Influence of spray drying conditions on the properties of avocado powder drink. Food
seeds, peels, pomace and wastewater, which are all by-products from Chemistry, 266, 284–291. https://doi.org/10.1016/j.foodchem.2018.06.016.
the avocado oil production process. The total mass of by-products Domínguez, M. P., Araus, K., Bonert, P., Sánchez, F., San Miguel, G., & Toledo, M. (2014).
constituted approximately 92% of the whole avocado fruit. Avocado The Avocado and Its Waste: An Approach of Fuel Potential/Application
[Domínguez2016]. In G. Lefebvre, E. Jiménez, & B. Cabañas (Eds.). Environment,
seeds were the only by-product that contained a large portion of di- Energy and Climate Change II: Energies from New Resources and the Climate Change (pp.
gestible carbohydrates 29.7% (wet basis) of the whole seed. 199–223). Cham: Springer International Publishing. https://doi.org/10.1007/698_
Furthermore, it was found that wastewater contained the highest lipid 2014_291.
Dreher, M. L., & Davenport, A. J. (2013). Hass Avocado Composition and Potential Health
content (6.3%) and could be successfully spray dried without the use of Effects. Critical Reviews in Food Science and Nutrition, 53(7), 738–750. https://doi.org/
carriers. Although the powder yield was low mainly due to the high 10.1080/10408398.2011.556759.
lipid content, spray drying was found to increase antioxidant activity, Gamble, J., Harker, F. R., Jaeger, S. R., White, A., Bava, C., Beresford, M., ... Woolf, A.
(2010). The impact of dry matter, ripeness and internal defects on consumer per-
and total phenolic content by 20- and 8-fold respectively compared to
ceptions of avocado quality and intentions to purchase. Postharvest Biology and
avocado flesh. Colour was not influenced much by increasing spray Technology, 57(1), 35–43. https://doi.org/10.1016/j.postharvbio.2010.01.001.
drying temperatures. However, increasing spray drying inlet tempera- Garofulić, I. E., Zorić, Z., Pedisić, S., & Dragović-Uzelac, V. (2016). Optimization of sour
tures resulted in less agglomeration of AWW powders. Antioxidant cherry juice spray drying as affected by carrier material and temperature. Food
Technology and Biotechnology, 54(4), 441.
activities of seed and peel were the highest across the CUPRAC, FRAP, Ghani, M. A., Barril, C., Bedgood, D. R., & Prenzler, P. D. (2017). Measurement of anti-
phosphomolybdenum and TPC assays, with pomace having the lowest oxidant activity with the thiobarbituric acid reactive substances assay. Food
values. The spray dried wastewater powder had the highest antioxidant Chemistry, 230, 195–207. https://doi.org/10.1016/j.foodchem.2017.02.127.
Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A., & Saurel, R. (2007). Applications
activity using the CUPRAC assay but had significantly lower FRAP, of spray-drying in microencapsulation of food ingredients: An overview. Food
phosphomolybdenum and TPC antioxidant values compared to seed Research International, 40(9), 1107–1121. https://doi.org/10.1016/j.foodres.2007.
and skin by-products. Application of the spray dried AWW powder into 07.004.
Ivanišová, E., Kačániová, M., Petrová, J., Frančáková, H., & Tokár, M. (2016). The eva-
cooked pork sausages was found to be as effective as using a synthetic luation of antioxidant and antimicrobial effect of Tussilago farfara L. and Cetraria
antioxidant such as sodium erythorbate (E316) in inhibiting lipid oxi- islandica L. Scientific Papers: Animal Science & Biotechnologies/Lucrari Stiintifice:
dation. Zootehnie si Biotehnologii, 49(2).
Kosińska, A., Karamać, M., Estrella, I., Hernández, T., Bartolomé, B., & Dykes, G. A.
(2012). Phenolic Compound Profiles and Antioxidant Capacity of Persea americana
Declaration of Competing Interest Mill. Peels and Seeds of Two Varieties. Journal of Agricultural and Food Chemistry,
60(18), 4613–4619. https://doi.org/10.1021/jf300090p.
Lu, Q.-Y., Arteaga, J. R., Zhang, Q., Huerta, S., Go, V. L. W., & Heber, D. (2005). Inhibition
The authors declare that they have no known competing financial
of prostate cancer cell growth by an avocado extract: Role of lipid-soluble bioactive
interests or personal relationships that could have appeared to influ- substances. The Journal of Nutritional Biochemistry, 16(1), 23–30. https://doi.org/10.
ence the work reported in this paper 1016/j.jnutbio.2004.08.003.
Mokbel, M., Saif, & Fumio, H. (2005). Antibacterial and Antioxidant Activities of Banana
(Musa, AAA cv. Cavendish) Fruits Peel. American Journal of Biochemistry and
Acknowledgement Biotechnology, 1, 125–131. https://doi.org/10.3844/ajbbsp.2005.125.131.
Morais, D. R., Rotta, E. M., Sargi, S. C., Bonafe, E. G., Suzuki, R. M., Souza, N. E., ...
The authors gratefully acknowledge the technical support of Visentainer, J. V. (2017). Proximate composition, mineral contents and fatty acid
composition of the different parts and dried peels of tropical fruits cultivated in
Olivado Ltd and financial support from Callaghan Innovation New Brazil. Journal of the Brazilian Chemical Society, 28(2), 308–318.
Zealand under contract number OLIVA1701/PROP-57274-FELLOW- Özyürek, M., Güçlü, K., Tütem, E., Sözgen Başkan, K., Erçağ, E., Karademir Çelik, S., ...
OLIVADO for this research project. Without their ongoing support, this Apak, R. (2011). A comprehensive review of CUPRAC methodology, 3. https://doi.org/
10.1039/C1AY05320E.
study would not be possible. Perdomo, L., Beneit, N., Otero, Y. F., Escribano, Ó., Díaz-Castroverde, S., Gómez-
Hernández, A., & Benito, M. (2015). Protective role of oleic acid against cardiovas-
Appendix A. Supplementary data cular insulin resistance and in the early and late cellular atherosclerotic process. 75
75 Cardiovascular diabetology, 14. https://doi.org/10.1186/s12933-015-0237-9.
Perea-Moreno, A.-J., Aguilera-Ureña, M.-J., & Manzano-Agugliaro, F. (2016). Fuel
Supplementary data to this article can be found online at https:// properties of avocado stone. Fuel, 186, 358–364. https://doi.org/10.1016/j.fuel.
doi.org/10.1016/j.foodchem.2019.125635. 2016.08.101.
Raak, N., Symmank, C., Zahn, S., Aschemann-Witzel, J., & Rohm, H. (2017). Processing-
and product-related causes for food waste and implications for the food supply chain.
References Waste Management, 61, 461–472. https://doi.org/10.1016/j.wasman.2016.12.027.
Rodríguez-Carpena, J.-G., Morcuende, D., Andrade, M.-J., Kylli, P., & Estévez, M. (2011).
Alirezalu, K., Hesari, J., Nemati, Z., Munekata, P. E. S., Barba, F. J., & Lorenzo, J. M. Avocado (Persea americana Mill.) Phenolics, In Vitro Antioxidant and Antimicrobial
(2018). Combined effect of natural antioxidants and antimicrobial compounds during Activities, and Inhibition of Lipid and Protein Oxidation in Porcine Patties. Journal of
refrigerated storage of nitrite-free frankfurter-type sausage. Food Research Agricultural and Food Chemistry, 59(10), 5625–5635. https://doi.org/10.1021/
International. https://doi.org/10.1016/j.foodres.2018.11.048. jf1048832.
Saavedra, J., Córdova, A., Navarro, R., Díaz-Calderón, P., Fuentealba, C., Astudillo-

8
R. Permal, et al. Food Chemistry 306 (2020) 125635

Castro, C., ... Galvez, L. (2017). Industrial avocado waste: Functional compounds Wang, W., Bostic, T. R., & Gu, L. (2010). Antioxidant capacities, procyanidins and pig-
preservation by convective drying process. Journal of Food Engineering, 198, 81–90. ments in avocados of different strains and cultivars. Food Chemistry, 122(4),
https://doi.org/10.1016/j.jfoodeng.2016.11.018. 1193–1198. https://doi.org/10.1016/j.foodchem.2010.03.114.
Santo, M. G.d., Nunez, C. V., & Moya, H. D. (2013). A new method for quantification of White, A., Woolf, A., Harker, R., & Davy, M. (1999). Measuring avocado firmness:
total polyphenol content in medicinal plants based on the reduction of Fe(III)/1,10- Assessment of various methods. Revista Chapingo Serie Horticultura, 5, 389–392.
phenanthroline complexes. Advances in Biological Chemistry, 03(06), 11. https://doi. Wong, M., Ashton, O., Requejo-Jackman, C., McGhie, T., White, A., Eyres, L., . . . Woolf,
org/10.4236/abc.2013.36059. A. (2008). Avocado Oil: The Color of Quality. In Color Quality of Fresh and Processed
Singleton, V. L., Orthofer, R., & Lamuela-Raventós, R. M. (1999). Analysis of total phenols Foods (Vol. 983, pp. 328-349): American Chemical Society. Retrieved from http://dx.
and other oxidation substrates and antioxidants by means of folin-ciocalteu reagent. doi.org/10.1021/bk-2008-0983.ch024. Retrieved 2017/12/03. 10.1021/bk-2008-
Methods in Enzymology: Vol. 299, (pp. 152–178). Academic Press. 0983.ch024.
Soong, Y.-Y., & Barlow, P. J. (2004). Antioxidant activity and phenolic content of selected Wong, M., Eyres, L., & Ravetti, L. (2014). 2 - Modern Aqueous Oil
fruit seeds. Food Chemistry, 88(3), 411–417. https://doi.org/10.1016/j.foodchem. Extraction—Centrifugation Systems for Olive and Avocado Oils. In W. E. Farr & A.
2004.02.003. Proctor (Eds.), Green Vegetable Oil Processing (pp. 19-51): AOCS Press. Retrieved
Thiex, N. J., Anderson, S., & Gildemeister, B. (2003). Crude fat, diethyl ether extraction, from http://www.sciencedirect.com/science/article/pii/B9780988856530500054.
in feed, cereal grain, and forage (Randall/Soxtec/submersion method): Collaborative https://doi.org/https://doi.org/10.1016/B978-0-9888565-3-0.50005-4.
study. Journal of AOAC International, 86(5), 888–898. Wong, M., Requejo-Jackman, C., & Woolf, A. (2010). What is unrefined, extra virgin cold-
Tran, T. T. A., & Nyugen, H. V. H. (2018). Effects of Spray-Drying Temperatures and pressed avocado oil. Inform AOCS, 21(4), 189–260.
Carriers on Physical and Antioxidant Properties of Lemongrass Leaf Extract Powder. Woolf, Allan, Wong, Marie, Eyres, Laurence, McGhie, Tony, Lund, Cynthia, Olsson, Shane,
Beverages, 4(4), 84. Wang, Yan, Bulley, Cherie, Wang, Mindy, Friel, Ellen, & Requejo-Jackman, Cecilia
Vinha, A. F., Moreira, J., & Barreira, S. V. (2013). Physicochemical Parameters, (2009). Gourmet and Health-Promoting Specialty Oils (pp. 73–125). Elsevier. https://
Phytochemical Composition and Antioxidant Activity of the Algarvian Avocado doi.org/10.1016/B978-1-893997-97-4.50008-5.
(Persea americana Mill.). Journal of Agricultural Science, 5(12). Yang, S., Hallett, I., Rebstock, R., Oh, H. E., Kam, R., Woolf, A. B., & Wong, M. (2018).
Vyncke, W. (1970). Direct Determination of the Thiobarbituric Acid Value in Cellular Changes in “Hass” Avocado Mesocarp During Cold-Pressed Oil Extraction.
Trichloracetic Acid Extracts of Fish as a Measure of Oxidative Rancidity. Fette, Seifen, Journal of the American Oil Chemists' Society, 95(2), 229–238. https://doi.org/10.
Anstrichmittel, 72(12), 1084–1087. https://doi.org/10.1002/lipi.19700721218. 1002/aocs.12019.