Color Photostability Assessment of Ultrasound-Assisted Extracts From European Blueberry ( Vaccinium Myrtillus L.) Obtained with The Use of Non-Toxic Solvents

the proportion 50/50 (v/v). However, the highest content of anthocyanins was recorded for the mixture of glycerin and ethanol in a ratio of 50/50 (v/v). It is shown that the addition of glycerin supports the extraction of anthocyanins to some extent; however, it results in a reduction of the color photostability of extracts used to dye cotton fabrics.


Introduction
Due to the biodiversity of flora, forests are a source of many valuable raw materials -not only wood, but also plants with a wide spectrum of uses, from food, through medicines to dyes.In recent years, there has still the primary method by which they are obtained.The solvents used are mainly polar organic liquids such as methanol, ethanol or acetone.They are usually acidified with organic and inorganic acids, and the use of strong acids may cause hydrolysis of the glycosidic bond in the dye molecule (Silva et al. 2015).Due to the toxicity and high costs associated with the removal of some of the aforementioned compounds, more environmentally friendly solutions are being developed; for example, the use of glycerin in the extraction process (Shehata et al. 2015;Kowalska et al. 2021).Among known improvements in this area, ultrasonically assisted extraction is particularly worthy of note.It has been proven that ultrasonic extraction not only increases the process efficiency, but is also economically and ecologically justified (Chemat et al. 2016).Factors supporting extraction also include the influence of microwaves (Liazid et al. 2011), the presence of an electric field (Puértolas et al. 2013), or the use of supercritical CO2 (Paes et al. 2014).
Anthocyanins are a group of dyes derived from plants, which among other things control the color of vegetables, fruits and flowers.They are formed from a combination of a sugar residue and anthocyanidin (aglycone) (Kayesh et al. 2013;Quina et al. 2009).They can take shades from blue and purple to red and orange, depending on the presence of hydroxyl groups in the molecule, the number and type of sugar moieties, and the aliphatic or aromatic carboxylates attached to them (Kong et al. 2003).Many research projects relating to the identification of anthocyanins have led to the recognition of several hundred compounds belonging to this group (numbers from 500 to 700 are mentioned in the literature) (Wallace and Giusti 2019).The most common naturally occurring anthocyanidins include cyanidin, delphinidin, malvidin, pelargonidin, peonidin and petunidin, which have chemical structures based on an identical carbon skeleton with various glycosidic substitutions at positions 3 and 5.The general anthocyanidin structure with possible substitutions is shown in Table 1 (Salehi et al. 2020).Anthocyanins are soluble in water, poorly resistant to high temperatures, and have strong antioxidant properties (Castañeda-Ovando et al. 2009;Turturică et al. 2018).They also change color depending on the pH of the environment: they turn red/pink, purple and green-yellow in acidic, neutral and alkaline solutions respectively (Michaelis et al. 1936).Depending on the pH, anthocyanins undergo structural changes.In an acidic environment they are transformed into carbinol, pseudobase, and quinonoidal structures, and at alkaline pH they are transformed into chalcone (Castañeda-Ovando et al. 2009).Their main source is fruit berries, including black chokeberry (Aronia melanocarpa Michx.Elliott), elderberry (Sambucus nigra L.), and European blueberry (Vaccinium myrtillus L.) (Wu et al. 2006).
Widely distributed and readily available, products rich in anthocyanins are used in a variety of applications.An interesting direction in the development of such extracts is their use for staining cellulose-based materials (Phan et al. 2020) or fibers made of cotton and silk (Vankar and Shukla 2011;Wang et al. 2016).Apart from being components of the human diet and environmentally friendly alternatives to synthetic food dyes (Coultate and Blackburn 2018), they can also be used to obtain intelligent packaging films.The most commonly used matrices include chitosan (Halász and Csóka 2018), cassava (tapioca) starch (Luchese et al. 2018), calcium alginate (Al-Qahtani et al. 2021), bacterial cellulose (Mohammadalinejhad et al. 2020), gelatin (Musso et al. 2018) and poly(vinyl alcohol) composites (Ma et al. 2017).They have also been found to improve the health-promoting properties of food (Mattioli et al. 2020).Another interesting potential application of anthocyanins is as inhibitors of metal corrosion processes (Kopyciński and Duda 2022).
This paper describes research concerning the color photostability of anthocyanin-rich extracts from European blueberry (Vaccinium myrtillus L.).Water, glycerin, ethanol, and their mixtures were used as environmentally friendly solvents.The color photostability of the analyzed extracts was determined by exposure to fluorescent lamps.The content of anthocyanins was assessed by comparison of three methods -two related to absorbance measurements, and one based on the colorimetric technique.

Materials
Fruits of European blueberry (Vaccinium myrtillus L.), hereafter referred to as VM, were used to prepare liquid extracts.The berries were picked by hand in the Lubliniec Forests (50°39'54.9"N18°37'58.7"E) in July 2021.The solvents used were deionized water (DI), ethanol (EtOH) (96% pure p.a.), and glycerin (GL) (anhydrous pure p.a.).Potassium chloride (pure p.a.) and sodium acetate (trihydrate pure p.a.) were used to obtain buffers at pH 1.0 and pH 4.5 respectively.The pH of both solutions was adjusted with hydrochloric acid addition.Undyed and unbleached raw cotton fabric, rich in cellulose, was used as a substrate for dyeing.All of the above reagents were purchased from Chempur (Poland).

Preparation of extracts
After harvesting, the berries were rinsed under running water, then air-dried and stored in cool surroundings before extraction.The preparation of liquid extracts began with the lyophilization of the VM fruits.Fresh berries were frozen at -80 ºC for 3 hours (ULUF 65, Arctiko).Each of the frozen batches was then placed into the lyophilizer chamber and dried under reduced pressure at 28 ºC for 96 hours (Alpha 2-4 LDplus, Christ).A reduction of approximately 88% in the weight of the fruits was recorded.The prepared lyophilizates were ground in an agate mortar, and appropriate solutions of the solvents were poured over them in a ratio of 1 ml per 0.1 g of dried VM (Table 2).The 24-hour maceration process of the lyophilizates was preceded and completed by 40-minute sonication of the samples in an ice bath (< 30 ºC) (Sonic-6, Polsonic).The macerates were centrifuged in a laboratory centrifuge (MPW 351/R/RH, MPW Med.Instruments).Each of the separated supernatants was filtered through quantitative cellulose filters (type 388, Ahlstrom Munksjö).The pH of the prepared solutions was measured (Delta 320, Mettler Toledo).The steps for preparation of the extracts as described above (as well as pH indicators) are shown in Fig. 1.

ATR-FTIR spectroscopy characterization
Infrared spectra of the prepared samples were acquired using a spectrometer equipped with a crystal diamond adapter (Nicolet iS 10, Thermo Fisher Scientific) in attenuated total reflectance (ATR) mode with Fourier transformation (FT).Before analysis, liquid extracts were dropped on a glass plate and evaporated in a drying oven at 28 ºC for 4 hours.The spectra were collected in the wavenumber range 4000-400 cm -1 at a scan number of 32 and a resolution of 4 cm -1 .To facilitate interpretation of the spectra (to suppress signals originating from the solvents used), before analysis the samples were dried at 36 ºC for 2 and 4 hours respectively for DI/EtOH-and GLbased extracts.

Weathering test and colorimetric measurement
A UV-accelerated weathering tester (Q-Panel, Q-Lab) was used to determine the impact of laboratory light exposure on the color photostability of the prepared extracts.The study was performed according to the ISO 16474-3:2021 standard.Method B ("daylight behind window glass") was applied, with the following parameters: cycle number -3, exposure period -24 h dry, lamp type -UVA-351, irradiance -0.76 W/m 2 /nm at 340 nm, temperature -40±3 °C.
Fragments of undyed, unbleached raw cotton fabric measuring 10 x 15 cm were dyed with each of the prepared extracts.The fabrics were dyed by dropping the extracts onto their surface and then applying them with a roller on both sides until the cotton surface was fully soaked.After drying (48 h, 25±2 °C, without access to sunlight) samples were fixed in the holders of the Q-Panel and tested.
The spectral photometric quantities in the visible range were determined, and the trichromatic components were described using a portable, handheld sphere spectrophotometer (Sp62, X rite) with a measuring geometry of d/8 (according to the PN-ISO 7724-2:2003 standard).The color coordinates in the CIELab color space were calculated (L* -brightness, a* -color from green to magenta, b* -color from blue to yellow).The color difference ∆E was determined using the following equation: For the determination of the color change of the samples, measurements were performed before and after the cycle in Q-Panel.The degree of dyeing was assessed primarily based on the value of L*.

UV -Vis spectroscopy analysis and determination of total monomeric anthocyanin pigment content
To estimate the content of anthocyanins in the prepared extracts, quantitative analysis was performed by means of UV-Vis spectroscopy.First, the optimal dilution was selected by establishing the standard curve of sample VM_E1.Optimal absorbance at wavelength 520 nm was obtained for 32-fold dilution.Afterwards, absorbance at the same dilution was measured for the other VM extracts in order to determine the concentration of anthocyanins.Measurements were made with the use of a spectrophotometer (DU 640, Beckman) in polystyrene cuvettes (Ratiolab, 45 x 12.5 x 12.5 mm, inside width 10 mm).The absorbance intensities on the spectra obtained for the analyzed extracts were compared.
To calculate the total monomeric anthocyanin pigment content (TMAPC) of the prepared extracts, AOAC Official Method 2005.02 was used.Absorbance measurements of extracts in pH 1.0 and pH 4.5 buffers were made at wavelength 520 nm and 700 nm (DU 640, Beckman).Anthocyanin pigment concentration was expressed as cyanidin 3-glucoside (cyd-3glu) equivalents corresponding to the absorbance difference, as follows: where A (absorbance) = (A520nm -A700nm)pH 1.0 -(A520nm -A700nm)pH 4.5, MW (molecular weight) = 449.2g/mol for cyd-3-glu, DF (dilution factor) = 100, l = pathlength in cm, ε = molar extinction coefficient 26900 in L•mol -1 •cm -1 for cyd-3-glu, and 10 3 = factor for conversion from g to mg.Buffers were prepared using the following procedures: − Buffer at pH 1.0 (0.025 M potassium chloride) -0.186 g of KCl was weighed in a volumetric flask and 90 ml of DI was added.The pH was adjusted to 1.0 by adding around 0.6 ml of HCl.The solution was diluted to a volume of 100 ml with DI. − Buffer at pH 4.5 (0.4 M sodium acetate) -90 ml of DI was added to 5.44 g of CH3CO2Na•3H2O in a volumetric flask.The pH was adjusted to 4.5 by adding around 2 ml of HCl.The mixture was then diluted with DI to a volume of 100 ml.

ATR-FTIR characteristics
The spectra collected for the tested extracts were all found to be similar to each other (Fig. 2).Importantly, the absorption bands present at the same wavenumbers for individual extracts differ in intensity.This effect is probably due to the extraction of several different compounds from the VM fruits, and may indicate that these compounds are extracted in different proportions depending on the composition of the extraction mixture and the proportions of the components used (DI, EtOH, and/or GL).For the analyzed samples, fundamental spectral signals in the wavenumber range 4000-400 cm -1 were characterized.A band at 3359 cm -1 can be ascribed to the stretching of bonds of -OH groups (Haberhauer et al. 1998).The bands at 2936 and 2880 cm -1 indicate the stretching of C-H bonds in alkyl groups (Meissl et al. 2008).The spectral regions at around 1721-1601 cm -1 correspond to stretching of C=O bonds (carbonyl groups) (Schwanninger et al. 2004).Signals appearing at 1340 cm -1 come from symmetric CO2 stretching in carboxylates and/or symmetric COO stretching in oxalate (Nopp-Mayr et al. 2020).Bands at 1112 and 1044 cm -1 indicate stretching of C-O-C bonds (ethers) (Beć et al. 2017).Regions around 818 cm -1 can be associated with primary NH2 out-of-plane bending (aromatic) and/or CH bending in alkenes, while those at 781 cm -1 represent symmetric deformation of OCO (oxalate) (Nopp-Mayr et al. 2020).

Assessment of degree of dyeing and color photostability
Based on the values of the L* and ΔEtot.parameters, the degree of dyeing and color photostability of the VM extracts were assessed (Table 3) The dyeing strength of VM extracts can be related to potential anthocyanin content, which is dependent on the type of solvent used.The use of pure glycerin resulted in a low efficiency of dyeing, which may be due to its high viscosity and the probable slowing of the process of penetration of the lyophilizates.On the other hand, glycerin contents of 15% or higher intensify the extraction process, especially a content of 50% in two-component systems with EtOH or in threecomponent systems with EtOH:DI.Although the presence of glycerin increases the retention of anthocyanins in extracts stored in the dark (Garzón and Wrolstad 2001), it also contributes to their degradation when exposed to sunlight.
The observed relationship corresponds with results on the use of glycerin for the extraction of anthocyanins from chokeberry and elderberry (Kowalska et al. 2021).In that study, anthocyanins were extracted with the use of 30%, 50%, 65% and 80% GL together with DI and EtOH at 20, 50 and 80 °C.Measurements at 20 °C showed that the extraction efficiency increased with increasing GL content up to 50%, but above that value the efficiency clearly decreased.GL has a higher viscosity (1.412 mPa•s) (Segur and Oberstar 1951) than DI (1.0016 mPa•s) or EtOH (1.074 mPa•s) (Rumble 2018), which is due, among other things, to the presence of three hydroxyl groups that can form hydrogen bonds.When the GL content exceeds 50%, the viscosity of the extraction system increases significantly, which probably limits the release and distribution of anthocyanins in the volume of the solution.Similarly, in another study (Eyiz et al. 2019), a system with 50% GL was found to be the most effective solvent.Despite its advantages, the use of glycerin as a solvent is rarely investigated.
The VM_E4 variant was selected as exhibiting the best light stability.This is in agreement with the results of previous work on the extraction of anthocyanins with EtOH (Nour et al. 2013;Ćujić et al. 2016).It was confirmed that aqueous solutions containing 50% or 60% EtOH have the highest extraction efficiency.A further increase in the proportion of EtOH results in a decrease in extraction efficiency.In the further part of this work, the sample dyed with VM_E4 was mounted on the inner glass of a window (southwest wall, 50º17'57.04"N18º42'23.627"E)and exposed to sunlight for three months in the summer/fall of 2021 (August-October).With direct normal irradiation in this period of about 280 kWh/m 2 , the fabric did not discolor completely after the testing period (ΔEtot.= 36.95)(Table 4).

Absorbance intensity and TMAPC measurement
The absorbance measurement results for the prepared extracts of VM are presented in Fig. 3 and Table 5.A relationship was observed between the intensity of absorbance recorded on the UV-Vis spectra and the type of solvent used.The intensity of the absorbance peak recorded at wavelength 520 nm was analyzed, because its presence in the spectrum statistically corresponds to the group of anthocyanins absorbing light in its vicinity (El Kouari et al. 2015).The highest absorbance intensity was observed for extract VM_E6, for which the solvent was a mixture of GL and EtOH in a proportion of 50/50 v/v (Abs.int.= 1.2077).The lowest absorbance intensity was recorded for VM_E3, for which GL alone was used (Abs.int.= 0.6502).For extracts VM_E5, VM_E8 and VM_E9, for which the solvents were mixtures of GL, EtOH and DI, the values of this parameter were practically the same (Abs.int.= ca.0.84).
The TMAPC measurement results for the VM extracts are presented in Fig. 4 and Table 5. Cyd-3-glu was used as an equivalent calculation reference because of its widespread occurrence in fruits and vegetables rich in anthocyanins (He and Giusti 2010).Fig. 4 shows one of the UV-Vis spectra, obtained for the VM_E4 extract solution in 1 and 4.5 pH buffers.Similarly as in the case of absorbance intensity, extract VM_E6 has the highest TMAPC value (1754.72 mg/L), while VM_E3 has the lowest (1187.96mg/L).However, for the remaining extracts, the values did not follow the same order as in the previous case.This discrepancy is due to the fact that in the TMAPC method only one of the anthocyanins is taken into account, representing a certain percentage of all compounds giving color to the fruit and to the solutions of VM.The cyd-3-glu extraction efficiency for the solvents used may vary.The values of TMAPC obtained for VM extracts in our study can be compared to the results recorded in the literature for different sources of anthocyanins.For example, previously reported TMAPC measurements for juice concentrates, including fruits such as raspberry, grape and elderberry, indicated anthocyanin contents of   et al. 2005).In another work, mixtures of water with ethanol or methanol in various proportions were used to obtain VM extracts, and the TMAPC values ranged from 1462 mg/L to 2953 mg/L for water/ethanol solvent and from 1462 mg/L to 3888 mg/L for water/methanol solvent.In that experiment, frozen VM fruits were used in a quantity of 0.25 g per 1 ml of solvent, and the extraction was carried out in an acidic environment (0.1% HCl) in an ultrasonic bath (Căta et al. 2010).It should be remembered that the results obtained by various research groups depend not only on the type of extract used, but also on the method of preparing the fruit and the ratio in which it was mixed with the solvents; the results are therefore often not mutually comparable.The pH of all extracts obtained was acidic (with values within the range 3.07-4.71).

Conclusion
The emphasis on the development of environmentally friendly solutions is also visible in relation to the processing of resources of natural origin.The potential of this field of research can be confirmed by the considerable number of studies undertaken.In this study, an investigation was made of the color photostability of ultrasound-assisted extracts from European blueberry (Vaccinium myrtillus L.) obtained with the use of water, ethanol and glycerin as solvents.Assessment of the degree of dyeing and color photostability of the prepared extracts was performed in a UV-accelerated weathering tester.On the basis of UV-Vis measurements, the anthocyanin contents in the extracts were also estimated.
European blueberry (Vaccinium myrtillus L.) was selected as a raw material due to its fully renewable character and widespread presence throughout Europe, especially in Polish forests.In addition, its characteristics (including natural origin and nontoxicity) make this plant suitable for sensitive applications, including direct contact with food or cotton fabrics, as well as wood staining.The dyeing strength of all obtained extracts was found to be sufficient without the need to purify their remaining organic components, for example by means of chromatographic techniques.The use of the prepared solvents as a substitute for the traditionally used -and often toxic -liquids reduces the harmfulness of the extracts, thus enhancing their environmentally friendly profile.
The following conclusions can be given on the basis of the measurement results: 1.The extract obtained using ethanol and water in a volume ratio of 50/50 exhibits the highest color photostability, as it undergoes the smallest color change on exposure to UV radiation.This extract, which is also distinguished by a relatively high anthocyanin content, when applied to cotton fabric did not lose its color fully after three months of exposure to sunlight through window glass.2. The highest content of anthocyanins, as determined using UV-Vis measurements, was recorded for the extract obtained using glycerin and ethanol in a volume ratio of 50/50.In view of a number of advantages, such as low price, easy availability and non-toxicity, glycerin can be used as an additive medium enhancing the process of extraction of anthocyanins from plant materials.
Accordingly, due to their good ability to dye cellulosic materials, such extracts have the potential to be applied as dye additives, above all for wood stains.It can be assumed that the most advantageous solution from the point of view of application would be to cover the stained wood with varnish enriched with the addition of appropriate photostabilizers such as UV-absorbers or hindered amine light stabilizers (HALS).This arrangement will extend the color stability of the substrate and extend the life of the coated element.Moreover, they can be used as non-toxic food dyes, and as color concentrates for the preparation of pH-sensitive, biodegradable packaging films for monitoring the freshness of foodstuffs.

Fig 1 .
Fig 1. Steps for the preparation of VM extracts pH 3 pH 7 . It was observed that the extract with the highest dyeing strength was VM_E6, obtained with GL:EtOH in the volume ratio 50/50 (L* = 33.08).The lowest degree of dyeing was recorded for VM_E3, where the solvent was GL (L* = 61.51).For the remaining extracts, L* values ranged from 35 to 54. Analysis of the color change of individual fabrics after 24 hours in the UVA weathering tester showed that the extract most resistant to fluorescent lamps is VM_E4, based on EtOH: DI in the volume ratio 50/50 (ΔEtot.= 11.79).The abovementioned VM_E3 extract (ΔEtot.= 39.42)proved to be the most susceptible to light degradation.The ΔEtot.values for the other analyzed extracts were between 12 and 33.

Table 1 .
Selected anthocyanidin structure and its possible substitutions

Table 2 .
Types of solvents used for preparation of VM extracts

Table 3 .
Results of weathering test and colorimetric measurement for VM extracts

Table 4 .
Color change of VM_E4 extract in three months of sunlight exposure

Table 5 .
Results of pH, absorbance intensity and TMAPC measurements for VM extracts