EXPERIMENTAL PROCEDURE

Sampling of TOs

Freshly prepared CNO, PO, and PKO were purchased from selected locations of Kotokuraba (Cape Coast, Ghana) and Bantama (Kumasi, Ghana) local markets. All TOs were immediately stored in 1500 mL dark polypropylene bottles, away from light, to avoid premature photochemical reactions. The caps on each bottle containing the oils were well sealed and locked to limit oxygen penetration. Each set of TOs stored in the dark polypropylene bottle was divided into two subsets: one bottle was labelled as the control sample, and the other labelled as the test sample. The control sample was stored away from light, whereas the test samples were exposed to sunlight for seven consecutive weeks.

Irradiation Procedure

Seven samples of each oil test sample were exposed to sunlight for seven consecutive weeks (8 h of exposure per day) at Adum (Kumasi, Ghana), between May and June on sunny days. The mean maximum and minimum temperatures from May to June were 33–29 °C. After each day of exposure, the sample was analysed for the FFA, PV, IV, and colour content. All analyses and characterisations were performed at the Ghana Nut Company limited, Techiman (Ghana), and Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana.

Free Fatty Acids

The percentage (%) of FFA in the oil was evaluated according to the AOCS (Ca 5a-40) (AOCS, 2017) protocol. The oil samples were shaken thoroughly to ensure even mixing of the oil. The oil (2.0 g) was weighed with a Sartorius analytical balance in a conical flask. Neutralised ethanol (50 mL) in a flask was heated to 70 °C on a heating mantle, which dissolved all of the oils in the flask. Phenolphthalein indicator (2 to 3 drops) was added to the heated mixture, and the mixture was titrated against 0.1 N NaOH solution with constant swirling until the first appearance of the pink colour, which persisted for 15–20 s. The procedure was performed in triplicate, and the FFA level was calculated from equation 2.
\begin{equation} \%\ FFA=\frac{\text{T\ x\ N}\ }{\text{W\ }}\times 0.02\ mg,\text{\ \ \ \ }\text{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ }\text{\ \ \ \ \ \ \ \ \ \ }\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (1)\nonumber \\ \end{equation}
where T is the titer volume (mL), N is the normality of NaOH, and W is the weight of samples taken (g).

Peroxide Value

According to the AOCS (Cd 8b-90) protocol (AOCS Official Method Cd 8b-90, 2003), PV was evaluated by the titrimetric method, and the results were expressed in Meq O2 / kg. The control and test samples were shaken thoroughly to ensure homogeneity of the TOs. The TOs (5.0 g) were weighed with a Pasteur pipette into a 250 ml dark Erlenmeyer flask, and 50 mL of acetic acid-isooctane solution (3:2, v/v) was added to the oil. The mixture was swirled to ensure dissolution. Potassium iodide (0.5 mL) was pipetted into the flask, and the flask was stored away from light before the titration. The resulting solution was shaken for 1 min, after adding 30 mL of distilled water. SDS solution (0.5 mL, 10%) and approximately 10 mL of the starch indicator was added, and titration was performed. The solution was titrated with 0.1 N standardised sodium thiosulfate with constant and vigorous agitation to liberate the iodine from the solvent layer completely. Thiosulfate was added dropwise until the solution turned from blue-black to colourless. The procedure was performed in triplicate, and the PV was calculated using equation 2.
\begin{equation} \text{PV}\ \left(\frac{\text{meq}}{\text{kg}}\right)=\frac{\left(S-B\right)\text{\ x\ N}\ }{W}\times 1000\ \text{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ }\text{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ }\text{\ \ \ \ }\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (2)\ \nonumber \\ \end{equation}
where S is the volume of Na2SO3required by the samples (mL), B represents the volume of Na2SO3 for the blank sample (mL),N is the normality of sodium thiosulfate (0.1 N), and W is the weight of the sample (g).

Iodine Value

IV of oils was analysed according to the AOCS (Cd 1-25) protocol (AOCS, 2009), and the results were expressed in grams of iodine / 100 g of oil. The oils were shaken thoroughly to ensure even mixing, and 0.50 g of the oil was weighed with a Sartorius analytical balance in a 250 ml Erlenmeyer flask. Cyclohexane-acetic acid (1:1) solution (20 mL) was added to the flask, followed by 25 ml of Wijs solution; the mixture was swirled to ensure homogeneity. The flask was stored in a dark enclosed area for 1 h. Potassium iodide solution (20 mL) and distilled water (100 mL) were immediately added to the solution. The resulting solution was titrated with 0.1 N standardised sodium thiosulfate. At the endpoint, the yellow colour of the solution completely disappeared, at which point, 1–2 mL of the starch indicator was added to the solution. The titration was continued until the solution turned from blue to colourless. The procedure was performed in triplicate, and IV was calculated from equation 3.
\begin{equation} \text{IV}\ \left(\frac{\text{g\ of\ iodine}}{100g\ of\ Oil}\right)=\frac{\left(B-S\right)\text{\ x\ N}\ x\ 12.69}{W}\ \text{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ }\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (3)\ \nonumber \\ \end{equation}
where S is the volume of Na2SO3for all samples (mL), B is the volume of Na2SO3 for the blank sample (mL),N is the normality of sodium thiosulfate (0.1 N), and Wrepresents the weight of the sample (g).

Colour Content

According to the AOCS (Cc 13b-45) protocol (Associaton of Offical Analytical Chemists, 2000), the oil colour content was determined analytically. The instrument used was the Lovibond Tintometer PFX-i series. The results were expressed in the AOCS tintometer red (R) and yellow (Y) units.

FTIR Analysis

ATR-FTIR Analysis

A Bruker Alpha FTIR spectrometer equipped with platinum attenuated total reflectance (ATR-FTIR, Bruker, Karlsruhe, Germany) was used for the FTIR measurement. All measurements were performed at room temperature (25 °C) at the KNUST central laboratory (Kumasi, Ghana). The ATR-FTIR employed a diamond crystal, which was cleaned with isopropanol before any background scan was acquired. The oil sample was placed directly on the crystal, and a pressure gauge was applied to ensure maximum contact. The background scan and the oil samples were sequentially measured from 4000 to 400 cm-1. The spectra were obtained with 24 scans at 4 cm resolution using the OPUS software (Bruker, Karlsruhe, Germany).

Computational Modelling of FTIR Spectra

The Gaussian 16 software (Frisch et al. , 2016) was used for all computational modelling and calculations. Oleic acid, palmitic acid, and lauric acids were used as the fatty acids present in TOs required to simulate the photooxidation of TOs (Cheng et al., 2018). The initial geometries of the fatty acids were optimised, and the vibrational frequencies were obtained. The geometry optimisation and vibrational frequency calculations were performed with the density functional theory (DFT) in the gas phase (Pereira et al. , 2017). The basis set used was Becke’s three parameters, and the Lee-Yang-Parr nonlocal correlation functional (B3LYP) at the 6-311++G (d, p) basis set. The final geometries of the fatty acids after the vibrational frequencies were visualised with GaussView (version 6.0.16) (Dennington, Keith and Millam, 2016). A scaling procedure (Palafox, 2019) was implemented to improve the wavenumbers and transmittance of the simulated spectra to match those of the experimental data.

Statistical Analysis

All data were obtained from at least three measurements, and each replicate was reported as the mean standard\(\pm\) deviation. OriginPro (version 2020, Northampton, MA, USA) and SPSS (IBM Corp., 2020, version 27, Armonk, N.Y., USA) were used for the one-way analysis of variance (ANOVA). The significant difference was evaluated by the P-values (P < 0.05).