Subcritical Water Extraction of Rosa alba L.—Technology and Quality of the Products

Featured Application

A novel type of extracts from the essential oil-bearing Rosa alba L. were obtained by the environmentally friendly technique of subcritical water extraction. The extracts were characterized for their volatile compounds, sugars, proteins, polyphenols, flavonoids and their antioxidant activity. Based on the analyses a conclusion could be made that the extracts could be a valuable additive for the formulation of foodstuffs and in cosmetic products.

Abstract

The white oil-bearing Rosa alba L. was subjected to green subcritical water extraction. The two factor modeling process revealed that 150 °C and 30 min treatment resulted in the maximum yield of phytochemicals, including essential oils, phenolic compounds, total carbohydrates, proteins, and simple sugars. A quantitative and qualitative analysis of the products was performed. The essential oil contained mainly phenylethyl alcohol, citronellol, geraniol, and hydrocarbons (paraffins). The phenolic substances were represented by phenolic acids (gallic acid: 30.92–113.37 µg/mL; ferulic acid: 44.50–99.96 µg/mL; rosmarinic acid: 25.27–80.47 µg/mL; protocatechuic acid: 13.05–25.48 µg/mL), as well as flavonoids (both quercetin and kaempferol: 8.35–8.56 µg/mL) and their glycosides (15.91–58.08 µg/mL). The monosaccharides were determined to include glucose (3.09–15.29 mg/mL), galacturonic acid (1.02–2.34 mg/mL), galactose (0.18–0.78 mg/mL), rhamnose (0.17–0.48 mg/mL), and xylose (0.07–0.17 mg/mL). The content of total phenols, flavonoids, and antioxidant activity were reported by the DPPH, ABTS, FRAP, and CUPRAC methods. The complex composition and activity of the extracts suggests their application directly as a food supplement or in cosmetic preparations.

1. Introduction

Extraction is a basic method for obtaining substances from aromatic and medicinal plants. Different variants of the process have been elaborated and implemented over time depending on the type of solvent used and the extraction parameters. In this way, the selectivity of the process could be modified and various products obtained. The conventional extraction procedure usually includes treatment with organic (n-hexane, petroleum ether, methanol, ethanol, diethyl ether, chloroform, ethyl acetate, etc.), water-based solvents or a combination of them. The solvent after the extraction is subsequently removed from the final product. In this case, usually some residual solvent amounts could be present and solvent removal could pose environmental issues [1]. Modern processing methods require minimal treatment of the raw material, and result in a quality product, economic efficiency, and environmental compatibility. This is especially difficult with the oil-bearing rose, a precious plant with wide application in perfumery and cosmetics, and having pharmacological importance, medical practice, and lifestyle quality [2]. Most of the efforts are directed towards Rosa damascena Mill. which include intensification of distillation to obtain the essential oil, use of liquefied gases in production of concrete and absolute products, and application of various solvents to obtain extracts, among others [3]. Rosa alba L. is the second most important oil-bearing plant for rose aroma production in Bulgaria and its aromatic products have unique qualities and are of strong economic interest. However, raw material processing is an even greater challenge because R. alba L. flowers contain half of the amount of essential oil compared to the Damask rose [4]. The essential oil obtained by steam-water distillation gives a 0.015–0.030% yield. Extraction with hexane gives a product called concrete, the yield of which could reach 0.21% [3]. Our previous study showed that the application of subcritical extraction with 1,1,1,2-tetrafluoroethane (freon R134 A) lead to a yield of 0.030–0.048% [4]. Among these methods, only steam-water distillation could be regarded as an environmentally friendly approach, but it is time and energy consuming. These observations gave us ground to consider water as the most preffered solvent for obtaining R. alba L. aroma products.

Subcritical water extraction (SWE), also known as hot compressed water, pressurized hot water extraction, or hydrothermal treatment, is a new, eco-friendly and “green” method for the extraction of biologically active compounds [5]. As the temperature increases, it’s dielectric constant and polarity of the water molecules decreases, and above 200 °C they ionize and become a very good solvent for hydrophobic organic molecules [6]. This highly effective technology allows for the use of minimal plant material with the maximum yield of beneficial nutrients [7]. As such, SWE yields pressurized hot water extracts which are sustainable, eco-friendly, and cost effective.

Our team and other scientists are working on applying the hydrothermal treatment technology to essential oil-bearing and medicinal plants, and proving that these extracts with bioactive compounds could be used in the food industry as food antioxidants, food supplements, and functional foods with potential health benefits [8,9,10]. To the best of our knowledge there is scant information in the literature regarding the application of subcritical water extraction to roses [11,12] and, for R. alba L., such data are completely lacking.

2. Materials and Methods

Materials: Fresh rose flowers of Rosa alba L. were used as raw material (2023 harvest). The plants were grown in the experimental field of the Institute for Roses and Medicinal Plants (IRAP), Kazanlak, Bulgaria. The rose blossoms were picked early in the morning (8.00–10.00 a.m.). The most appropriate flowering phases was used: semi-opened and full-opened buds. Half of the flowers had the petals separated from the calyx. Finally, whole flowers and petals were used as raw materials. They were processed immediately, in two repetitions.

Processing/Technology: InnoSolv Ltd. of Plovdiv, Bulgaria, has developed an automated system for the extraction of plant-based raw materials using pressurized hot water with a working volume of the extractor of 2 L. In the working chamber of the extractor, a metal mesh basket is placed. Its working volume is 2 L and it holds the raw material to be processed (200 g). The mesh basket has a mesh size of 100 µm. The basket is loaded in the extractor using a loading unit. To keep a constant extraction temperature throughout the whole process, the extractor is equipped with a tempering unit. The water solvent is heated up to the desired extraction temperature outside the extractor in a heating vessel with stainless steel electric heater. Two temperature regimens were used—mild (100 °C) and intense (150 °C). During each stage of heating the water, inert nitrogen gas is fed into the vessel from a pressurized cylinder to maintain counter-pressure. The counter-pressure value is aligned to the desired extraction temperature, and it should be higher than the saturation pressure of the water at that temperature. Thus, at the stages of water heating and extraction, the solvent is thermodynamically in a non-boiling liquid state.

After reaching the set extraction temperature, the hot water is fed into the extractor chamber, which marks the beginning of the extraction process. A circulating pump is used to intensify the process and reduce the time for extraction. It allows the solvent to pass around the particles of the raw material in a closed loop during the extraction. The flow rate of the circulated solvent is 50 bed volumes/hour (BV/h). In our case, using a 2 L extractor volume results in a value of 100 L/h. After completion of the extraction for the specified time (15 and 30 min), the obtained miscella is transferred from the extractor to one of the miscella collectors passing through a recovery unit, where its temperature and pressure decrease. The clean water solvent is fed into the system through the recovery unit. Water is then heated there and fed into the heating vessel. Generally, the extraction process is carried out in the following sequence: heating of the solvent and tempering of the extractor/loading the raw material/extraction/draining the miscella/waste removal. All stages of the extraction process (except loading the raw material and waste removal) are fully programmable logic controller (PLC) automated. The operator pre-sets the desired extraction temperature, number of solvent changes, and extraction duration for each treatment.

In parallel with the extraction, steam-water distillation was performed on a Clevenger-type laboratory apparatus, with classic parameters: sample 200 g, duration 2.5 h and speed 3–4 mL/min. The aim was to obtain an essential oil comparable to the oil obtained by hydrothermal treatment extracts.

Chemical composition: The volatiles were obtained after the liquid–liquid extraction of 500 mL of the SWE extract with an equal quantity of diethyl ether. The extraction was performed three times. The combined ether extracts were dehydrated with anhydrous sodium sulfate, and evaporated on a vacuum rotary evaporator until complete solvent removal. The sample was tempered and weighed to account for the amount of volatile fraction. The essential oil and volatile fractions were dried over anhydrous sodium sulfate and stored in tightly closed dark vials at 4 °C until further analyzed.

GC-FID/MS Technique: The chemical composition of the rose oil and extracts was evaluated on an Agilent 7820A GC System (Agilent technologies, Santa Clara, CA, USA) coupled with a flame ionization detector (Agilent technologies, Santa Clara, CA, USA) and a 5977B MS detector (Agilent technologies, Santa Clara, CA, USA). The protocol was chosen according to ISO 9842 [13] for the gas chromatographic analysis of rose oil. The capillary column EconoCapTM ECTM-5 (30 m × 0.32 mm × 0.25 mm film of 5% phenyl, 95% methylpolysiloxane; MEGA S.r.l., Legnano, Italy) was used. Hydrogen (99.999% purity) was used as a carrier gas. The split ratio was 1:10, the inlet temperature was set to 250 °C, and the FID temperature was set to 300 °C. The component relative percentages were calculated based on GC peak areas without correction factors.

The identification of constituents was established by comparing the retention indices and MS spectra with those reported in the literature, as well as, whenever possible, co-injections with authentic compounds.

Neutral sugars, monosaccharide composition, and proteins: The amount of neutral sugars in the extracts was determined by the phenol–sulfuric acid method [14]. A total of 200 μL of the extract was mixed with 200 μL of 5% phenol (Sigma-Aldrich Chemie Gmbh, Schnelldorf, Germany) solution in water, and 1 mL concentrated H₂SO₄ (Sigma-Aldrich Chemie Gmbh, Germany) was added. The mixture was vortexed and the absorption at 492 nm with 1 cm cuvette was measured. The standard curve was prepared using D-galactose.

The quantities of galactose, rhamnose, glucose, and galacturonic acid were determined using the chromatographic system ELITE LaChrome (Hitachi, Japan) HPLC with a VWR Hitachi Chromaster 5450 refractive index detector (Hitachi, Japan) using Aminex HPX-85H column (Bio-Rad Laboratories, Hercules, CA, USA). The samples and standards were eluted with 5 mM H₂SO₄ (Sigma-Aldrich Chemie Gmbh, Germany) at an elution rate of 0.5 mL/min, a column temperature of 50 °C, and a detector temperature of 35 °C. The amounts of xylose and mannose were determined separately with the same chromatographic system using the Sugar SP0810 (Shodex^®) column (Resonac, Tokyo, Japan). The samples and standards were eluted with ultrapure water at an elution rate of 1.0 mL/min, a column temperature of 85 °C, and a detector temperature of 35 °C.

The protein content was determined employing the Bradford method [15] with an AMRESCO E535-KIT (AMRESCO, Solon, OH, USA) with bovine gammaglobulin as the standard. A total of 100 μL of the extract was mixed with 1 mL of the Bradford reagent, vortexed, waited for 2 min, and the absorption at 595 nm with 1 cm cuvette was measured.

Total polyphenolic and flavonoid content: The total polyphenolic content (TPC) of the R. alba L. SWE extracts was assessed according to the method of Ivanov et al. [16] using the Folin–Ciocalteu reagent (Sigma-Aldrich, St. Louis, MO, USA). The results were expressed as mg equivalents of gallic acid (GAE) per mL of extract. The total flavonoid content (TFC) of the R. alba L. SWE extracts was evaluated following the method described by Ivanov et al. [16]. The results were expressed as mg of quercetin equivalents (QE) per mL of extract.

Individual phenolic acids and flavonoids: An HPLC analysis of the individual phenolic acids and flavonoids in the extracts was performed by HPLC with a UV-VIS detector (Waters, Milford, MA, USA). A total of 20 µL of extract were injected into a C18 column (Supelco Discovery HS; 5 μm, 25 cm × 4.6 mm) (Merck KGaA, Darmstadt, Germany) and eluted by 1% acetic acid (Phase A) and methanol (Phase B) and the following change for Solvent A was applied: 0 to 36 min Solvent A decreased from 90% to 78%; 36 to 37 min decrease from 78% to 70%; 37 to 47 min decrease from 70% to 60%; 47 to 58 min decrease from 60% to 54%; 58 to 59 min decrease from 54% to 40%; 59 to 71 min decrease from 40% to 20%; 71 to 72 min increase from 20% to 90%; 72 to 75 min hold to 90%. The 1.0 mL/min flow rate was used. The gallic, protocatechuic, vanillic, syringic, p-coumaric, and salicylic acids, (+)-catechin, (+)-epicatechin, and hesperidin were detected at λ = 280 nm, whereas the rosmarinic, chlorogenic, caffeic, and ferulic acids, rutin, quercetin, and kaempferol were detected at λ = 360 nm. Quantification was performed by retention times and calibration curves of the external standards.

Antioxidant activity:

DPPH Radical Scavenging Assay

The DPPH assay was performed according to the method of Ivanov et al. [16] using DPPH (2,2-diphenyl-1-picrylhydrazyl) reagent (Sigma-Aldrich, St. Louis, MO, USA). The antioxidant activity was expressed as mM Trolox equivalents (TE)/mL of the extract.

ABTS Radical Scavenging Assay

The ABTS assay was performed according to the method of Ivanov et al. [16] using ABTS (2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)) reagent (Sigma-Aldrich, St. Louis, MO, USA). The antioxidant activity was expressed as mM Trolox equivalents (TE)/mL of extract.

Ferric-Reducing Antioxidant Power (FRAP) Assay

The FRAP assay was performed according to the method of Ivanov et al. [16] using 2,4,6-Tris(2-pyridyl)-s-triazine (TPTZ) (Sigma-Aldrich, St. Louis, MO, USA). The antioxidant activity was expressed as mM Trolox equivalents (TE)/mL of extract.

Cupric Reducing Antioxidant Capacity (CUPRAC) Assay

The CUPRAC assay was performed according to the method of Ivanov et al. [16] using copper (II)—neocuproine (2,9-dimethyl-1,10-phenanthroline) (Sigma-Aldrich, St. Louis, MO, USA) as the reagent. The antioxidant activity was expressed as mM Trolox equivalents (TE)/mL of extract.

Statistical analysis: The experimental data (three replications) are presented as mean value ± standard deviation. For analysis, one-way ANOVA test (Tukey’s post hoc test; p < 0.05) was used with Microsoft Excel 2013 (additional XL Toolbox NG module installed).

3. Results and Discussions

3.1. Experimental Design

The experiments were organized as a two factors at two levels model. The factors were temperature and duration. The levels were 100 °C and 150 °C with a 15 min and a 30 min treatment. We carefully selected the factor levels based on our experience with the raw material and the data in the literature [10,11]. According to these data, the chosen factor levels are the most suitable temperatures and durations to simultaneously extract volatile fractions and phytochemicals without destroying their molecules. In addition, above 160 °C significant degradation of the polysaccharides could occur, and this complicates the process of extraction.

Table 1. Experimental design of subcritical water treatment of the R. alba L. plant material.

№	Sample Description	Abbreviation	Parameters		Extracts Obtained, mL
			Temperature, °C	Duration, min
1	Whole flowers	BRCC1	100	15	1937 ± 23
2	Whole flowers	BRCC2	100	30	2017 ± 15
3	Petals	BRV1	100	15	1978 ± 10
4	Petals	BRV2	100	30	2050 ± 15
5	Whole flowers	BRCC3	150	15	2112 ± 20
6	Whole flowers	BRCC4	150	30	2066 ± 12
7	Petals	BRV3	150	15	2053 ± 18
8	Petals	BRV4	150	30	2056 ± 21

presents the variants used in the investigation.

3.2. Extracts

As the temperature and duration of treatment increases with the sequence of variants, the amount of extract obtained also increases accordingly. The raw material (whole flowers or petals) has practically no impact on the amount of extract. In appearance, the variants at room temperature are a clear liquid with a yellow–brown color, and a pleasant rose aroma. The extracts at higher temperature are turbid, of a darker color, and also with a characteristic rose odor.

The volatile content in the extracts has been evaluated applying the extraction procedure aiming at concentration of the compounds, because of the low amounts of aroma substances. The methodology was developed and validated specifically for rose water, which also has a very low content of rose oil. Figure 1 presents both the results for the volatiles and the extractive substance content in the products of SWE. Although obtained by long-term distillation (2–3 h), rose water is the closest product suitable for comparison with the extracts. According to the standard, the content of essential oil in rose water is 0.025–0.050% [17]. Apparently, the SWE extracts the maximum amount of volatile substances for 30 min (Figure 1). The data showed that the essential oil content increases in the samples with the increasing temperature and duration of treatment. At a 150 °C extraction temperature, the petals yielded a higher amount of essential oil, an observation that confirms the well-known rule of oil localization in the rose blossoms [18,19].

The longer treatment resulted in increased amounts of dry matter in the average with 55%. The regimens applied having a higher temperature showed that the whole flower and petals have the same content and the same composition in the variants treated for 30 min, but have a drastic increase of the dry content. This contradiction with the observation could be explained by the disruption of the macromolecules at critical conditions and the release of additional substances in the extracts.

3.3. Chemical Composition of the Extracts

3.3.1. Gas Chromatography–Mass Spectrometry (GC-MS) Analyses

The chemical composition of the volatiles in the R. alba L. extracts is presented in

Table 2. Chemical composition of the volatile fractions of the R. alba L. essential oil and the SWE extracts determined by GC-MS.

№	RI	List of the Components/Classes	Essential Oil	BRCC1	BRCC2	BRV1	BRV2.	BRCC3	BRCC4	BRV3b	BRV4a
			Relative %
1	668	Ethanol	-	-	-	-	-	-	-	-	-
2	1031	Limonene	0.06 ± 0.00 g,1	3.31 ± 0.02 a	2.16 ± 0.05 d	2.94 ± 0.02 b	2.67 ± 0.02 c	0.08 ± 0.00	0.50 ± 0.02 f	0.48 ± 0.02 f	1.00 ± 0.00 e
3	1098	Linalool	1.29 ± 0.02 g	13.40 ± 0.04 c	18.42 ± 0.10 b	13.11 ± 0.02 c	7.35 ± 0.02 e	20.69 ± 0.01 a	2.98 ± 0.00 f	8.84 ± 0.00 d	2.83 ± 0.02 f
4	1118	2-Phenylethanol	0.16 ± 0.00 g	14.48 ± 0.05 b		10.55 ± 0.00 e	14.08 ± 0.00 c		6.69 ± 0.09 f	13.36 ± 0.02 d	25.05 ± 0.02 a
5	1109	Cis-rose oxide	0.04 ± 0.02 f	0.91 ± 0.00 b	0.60 ± 0.01 c	0.47 ± 0.02 d	0.64 ± 0.05 c	0.67 ± 0.02 c	0.15 ± 0.05 f	0.35 ± 0.01 e	1.26 ± 0.02 a
6	1134	Trans-rose oxide	0.02 ± 0.00 e	0.13 ± 0.02 d	0.13 ± 0.02 d	1.84 ± 0.08 a	0.12 ± 0.02 d	0.68 ± 0.02 c	0.04 ± 0.02 e	0.07 ± 0.01 e	0.92 ± 0.02 b
7	1228	Citronellol + Nerol	14.92 ± 0.06 a	0.97 ± 0.02 d	0.20 ± 0.02 f	0.8 ± 0.06 f	4.87 ± 0.02 b	3.28 ± 0.08 c	0.87 ± 0.00 d	0.38 ± 0.00 e	0.88 ± 0.05 d
8	1276	Geraniol	19.71 ± 0.00 a	7.81 ± 0.02 d	6.84 ± 0.05 e	9.08 ± 0.02 c	5.94 ± 0.04 f	0.68 ± 0.02 i	0.98 ± 0.00 h	3.41 ± 0.04 g	10.48 ± 0.00 b
9	1364	Eugenol	0.06 ± 0.02 f	2.47 ± 0.07 c	1.72 ± 0.07 d	1.36 ± 0.00 e	3.58 ± 0.05 b	8.33 ± 0.00 a	1.35 ± 0.05 e	1.83 ± 0.02 d	3.62 ± 0.04 b
10	1401	Methyl eugenol	0.05 ± 0.01 i	5.80 ± 0.04 b	4.12 ± 0.00 c	3.49 ± 0.04 e	7.80 ± 0.02 a	0.22 ± 0.02 h	2.99 ± 0.02 f	3.77 ± 0.07 d	1.93 ± 0.07 g
11	1678	Heptadecane	0.45 ± 0.01 c	0.05 ± 0.02 e	0.58 ± 0.02 b	0.30 ± 0.00 d	0.09 ± 0.00 e	0.69 ± 0.05 b	1.36 ± 0.00 a	0.62 ± 0.00 b	0.06 ± 0.02 e
12	1727	Farnesol	3.77 ± 0.04 a	0.08 ± 0.02 d	0.19 ± 0.02 c	-	0.10 ± 0.02 d	0.42 ± 0.02 b	0.38 ± 0.02 b	0.22 ± 0.02 c	0.05 ± 0.04 d
13	1874	Nonadecene	5.50 ± 0.05 a	0.46 ± 0.00 g	2.70 ± 0.02 d	0.27 ± 0.02 h	0.68 ± 0.02 f	1.73 ± 0.01 e	4.73 ± 0.00 b	3.22 ± 0.02 c	0.51 ± 0.00 g
14	1900	Nonadecane	13.21 ± 0.04 d	1.17 ± 0.01 h	15.56 ± 0.00 c	1.01 ± 0.00 i	1.65 ± 0.01 g	8.21 ± 0.00 f	24.96 ± 0.00 a	16.39 ± 0.01 b	10.04 ± 0.02 e
15	2000	Eicosane	1.39 ± 0.02 c	0.10 ± 0.01 f	1.50 ± 0.00 b	0.57 ± 0.00 e	0.15 ± 0.00 f	0.75 ± 0.01 d	2.04 ± 0.01 a	1.43 ± 0.03 b,c	0.04 ± 0.00 g
16	2100	Heneicosane	11.86 ± 0.00 a	0.58 ± 0.05 f	8.28 ± 0.00 c	0.18 ± 0.00 g	0.86 ± 0.00 e	3.54 ± 0.00 d	10.91 ± 0.01 b	8.40 ± 0.00 c	0.03 ± 0.01 h
17	2300	Tricosane	2.67 ± 0.00 a	0.08 ± 0.2 h	1.86 ± 0.00 c	1.72 ± 0.01 d	0.15 ± 0.02	0.83 ± 0.02 f	2.31 ± 0.02 b	1.53 ± 0.00 e	0.24 ± 0.00 g
18	2500	Pentacosane	1.18 ± 0.02 b	0.04 ± 0.02 h	0.71 ± 0.01 e	1.04 ± 0.00 c	0.14 ± 0.00 f	1.62 ± 0.00 a	0.72 ± 0.02 e	0.91 ± 0.01 d	0.11 ± 0.00 f
19	2700	Heptacosane	1.22 ± 0.00 c	0.04 ± 0.00 h	0.70 ± 0.00 e	3.49 ± 0.02 a	0.16 ± 0.02 g	1.66 ± 0.02 b	0.92 ± 0.00 d	0.30 ± 0.02 f	0.29 ± 0.00 f
		Monoterpenes	36.04	26.53	28.35	28.3	21.59	26.08	5.52	13.53	17.37
		Phenylethanol	0.16	14.48	18.42	10.55	14.08	20.69	6.69	13.36	25.05
		Rose oxides	0.06	1.04	0.73	2.31	0.76	1.35	0.19	0.42	2.18
		Phenylpropenes	0.11	8.27	5.84	4.85	11.38	8.55	4.34	5.6	5.55
		Sesquiterpenes	3.77	0.08	0.19	-	0.1	0.42	0.38	0.22	0.05
		Alkanes and alkenes	37.48	2.52	31.89	8.58	3.88	19.03	47.95	32.8	11.32
Total			77.56	61.20	66.27	52.28	51.03	54.08	65.88	65.51	60.14

¹ Results are presented as the mean of three measurements; ^a–i different letters in rows indicate statistically different values (Tuckey’s HSD test, p < 0.05).

. Compared to traditional rose oil obtained from R. alba, L. the volatile components found in the SWE extracts have the same qualitative profile, but a different quantitative composition. The essential oil is rich in monoterpene alcohols (geraniol (19.71%), citronellol and nerol (14.92%), and aliphatic hydrocarbons (nonadecane (13.21%) and heneicosane (11.86%)), while the extracts are characterized by a higher content of phenylethyl alcohol (from 6% to 25%) and linalool (from 3% to 20%). In the extracts obtained at a higher temperature (150 °C) and a longer extraction period (30 min), aliphatic hydrocarbons could be found in higher quantities. Basically, the products obtained have more phenyl derivatives: phenylethyl alcohol (from 14% to 25%), eugenol (from 2% to 8%), and methyl eugenol (from 0.2% to 7%). The levels of the eugenol and methyl eugenol were higher than in the essential oil (0.06% and 0.05%, respectively), but did not exceed the limits. This pattern has similarities to the composition of rose water [20], and confirms the results of Özel et al. [12]. Interestingly, Babu et al. [21] distilled R. damascena at elevated pressure and temperature, and the composition of the rose oil acquired had almost the same composition as the extracts obtained in the present work.

Antonova et al. [19] investigated and compared the composition of five aroma rose products: three essential oils (from three different distilleries in Bulgaria), one using a supercritical CO₂ extract, and one extract obtained using a subcritical freon (1,1,1,2-tetrafluoroethane, freon R134a) treatment. Their data suggested that the essential oils contained phenethyl alcohol in the range of 0.74–1.26 relative percentage, while the extracts obtained with liquefied gasses had a much higher content: 46.68 and 56.6 relative percentage for the CO₂ and freon products, respectively. The content of phenylethyl alcohol (from 14% to 25%) in the SWE extracts suggests that the aroma products obtained in the present study in relation to the phenylethyl alcohol concentration are somewhere between the rose oils and extracts produced by liquefied gasses. By contrast, the linalool quantities, ranging from 3% to 20% in the extracts, obtained as a result of the present study and in the aroma rose product investigated by Antonova et al. [19] are in the 0.08–1.31% range.

Ethanol is a component that is monitored and included in the international standard for rose oil [13]. It is associated with fermentation processes that occur in the flower during improper storage before processing. In our case, its content is so low that it is not found in the essential oil or the extracts. This is due to the immediate processing of the raw material. The total percentages for rose oil are higher than those for the extracts and this could be explained with the amounts of non-volatile components that were obtained during the SWE treatment of the plant material.

3.3.2. Total Neutral Sugars and Protein Content Determination

The extracts obtained using a polar solvent, such as water used in SWE, often also contain polar components, such as carbohydrates (polysaccharides and simple sugars) and water-soluble proteins. The next step in our research was to determine the content of total neutral sugars and proteins (

Table 3. Total content of neutral sugars and proteins in the R. alba L. SWE extracts.

№	Sample	Total Neutral Sugars, mg/mL	Proteins, µg/mL
1	BRCC1	0.52 ± 0.00 b,1	156.97 ± 3.99 e
2	BRCC2	0.49 ± 0,05 b,c	152.62 ± 2.18 e
3	BRV1	0.44 ± 0.01 c	170.56 ± 3.46 d
4	BRV2	0.45 ± 0.04 c	155.18 ± 0.73 e
5	BRCC3	0.41 ± 0.02 c	206.72 ± 1.81 c
6	BRCC4	0.52 ± 0.03 b	234.67 ± 5.08 a
7	BRV3	0.64 ± 0.02 a	221.85 ± 1.45 b
8	BRV4	0.63 ± 0.02 a	240.31 ± 3.63 a

¹ Results are presented as the mean of three measurements; ^a–e different letters in columns indicate statistically different values (Tuckey’s HSD test, p < 0.05).

).

The maximum amount of total neutral sugars was recorded when the petals were extracted for a short (15 min) and long time (30 min) at 150 °C. In these SWE conditions, the amount of extracted neutral sugars increased by about 30% (6.30–6.40 mg/mL) compared to the extraction performed at 100 °C (4.40–4.50 mg/mL). These data confirm that, in the rose flower, the main amount of sugars (oligo and polysaccharides) are localized mainly in the petals [22,23], which under subcritical conditions are hydrolyzed to their low molecular weight monomers (simple sugars). At 100 °C such processes were not observed to a significant extent. The data for protein content show that at 100 °C the amounts are the same in whole flowers and petals of the SWE extracts. At 150 °C the protein values were generally higher and the maximum values were reached at 30 min of treatment, but there is a clear distinction in the protein content of the petals and whole flowers. These results again indicate that, under subcritical conditions, degradation processes in high molecular weight metabolites occur.

3.3.3. Monosaccharide Composition of the SWE Extracts

The monosaccharide composition of the aqueous extracts of subcritical water extraction has also been determined.

Table 4. Monosaccharide composition of the R. alba L. SWE extracts.

№	Sample	GalA, mg/mL (Galacturonic Acid)	Glc, mg/mL (Glucose)	Rha, mg/mL (Rhamnose)	Gal, mg/mL (Galactose)	Xyl, mg/mL (Xylose)
1	BRCC1	1.24 ± 0.11 c,d,1	4.09 ± 0.27 f	0.48 ± 0.08 a	-	-
2	BRCC2	1.02 ± 0.16 d	7.72 ± 0.12 d	0.28 ± 0.01 b	-	-
3	BRV1	1.36 ± 0.25 c	6.55 ± 0.14 e	0.18 ± 0.02 c	0.24 ± 0.07 c	-
4	BRV2	1.58 ± 0.18 b,c	3.09 ± 0.23 g	0.17 ± 0.02 c	0.47 ± 0.04 b	-
5	BRCC3	1.84 ± 0.14 b	8.53 ± 0.10 c	0.25 ± 0.07 b	-	-
6	BRCC4	2.34 ± 0.12 a	11.59 ± 0.21 b	0.33 ± 0.01 b	0.17 ± 0.01 c	0.07 ± 0.01 b
7	BRV3	1.29 ± 0.14 c	4.12 ± 0.26 f	0.22 ± 0.03 b,c	0.18 ± 0.09 c	-
8	BRV4	2.24 ± 0.18 a	15.29 ± 0.11 a	0.27 ± 0.05 b	0.78 ± 0.08 a	0.17 ± 0.01 a

¹ Results are presented as the mean of three measurements; ^a–g different letters in columns indicate statistically different values (Tuckey’s HSD test, p < 0.05).

shows more detailed data regarding the monosaccharide content in the obtained extracts. An HPLC analysis revealed the presence of five monosaccharides in the SWE extracts of the petals and the flowers, four neutral sugars (glucose, galactose, rhamnose, and xylose) and one uronic acid (galacturonic acid). The main monosaccharide determined was glucose, and the concentration varied from 4 to 15 mg/mL extract. Higher amounts of glucose were found in the extracts obtained at higher temperatures 150 °C and longer subcritical extraction times (30 min) (Table 4). This observation tentatively is related to the increased rate of hydrolysis of hemicelluloses and partially of cellulose. The subcritical water treatment, in addition to being a method for “green extraction”, is used for processing (often combined with enzymatic pretreatment) of lignocellulosic biomass for obtaining low-molecular carbohydrates. These sugars could further be utilized for ethanol production or as starting materials for the synthesis of biodegradable polymers [24,25].

Galacturonic acid and rhamnose are the characteristic monosaccharides of the pectic polysaccharides main chain. R. alba L. flowers are characterized by the presence of pectic polysaccharides [26]. In general, it was clearly seen that the highest monosaccharide concentrations were determined in the extracts obtained at 150 °C and 30 min processing time. The galacturonic acid content followed a similar trend, and the amounts determined were statistically similar in the extracts from the whole flower and the petals. The rhamnose amounts found are similar in all the obtained extracts. Galactose was mostly found in the extracts obtained from rose petals, and the longer treatment seemed to favor the hydrolysis rate of polysaccharides (higher galactose concentrations in BRV4 and BRV2). Xylose is determined only in the extracts obtained at 150 °C and 30 min extraction time. Galactose and xylose are characteristic monomers in the branched chains of pectin polysaccharides. Their concentration increases in the extracts obtained at an increased temperature and extraction time. Under these conditions, pectin hydrolysis also increases, which is the reason for the identification of these monosaccharides in the extracts [26].

3.3.4. Total Polyphenols, Total Flavonoids and Antioxidant Activity of the R. alba L. SWE Extracts

Polyphenolic compounds are characteristic for almost all the plants and hence present in the extracts obtained from R. alba L. flowers. These phenolic substances (phenolic acids and flavonoids and their glycosides) are associated with the antioxidant activity of plants and their extracts [27,28]. On this basis, the next step in analyzing the obtained extracts was to determine the content of total polyphenols, total flavonoids, and their relationship with antioxidant activity. The latter was determined by four different methods, covering all aspects of this biological effect. The results are presented in

Table 5. Content of total polyphenols, total flavonoids, and antioxidant activity of the R. alba SWE extracts.

№	Samples	TPC, mg GAE/mL	TFC, mg QE/mL	Antioxidant Activity, mM TE/mL
				DPPH	ABTS	FRAP	CUPRAC
1	BRCC1	0.57 ± 0.00 b,1	0.25 ± 0.00 a	5.81 ± 0.02 d	5.75 ± 0.01 c	5.49 ± 0.05 c	14.18 ± 0.02 c
2	BRCC2	0.51 ± 0.00 c	0.21 ± 0.00 c	5.17 ± 0.02 e	4.91 ± 0.08 d	4.67 ± 0.08 d	11.44 ± 0.11 d
3	BRV1	0.41 ± 0.03 e	0.19 ± 0.00 d	4.06 ± 0.12 f	3.69 ± 0.03 f	3.65 ± 0.01 f	8.66 ± 0.11 f
4	BRV2	0.43 ± 0.00 e	0.21 ± 0.00 c	4.44 ± 0.07 f	4.37 ± 0.19 e	4.14 ± 0.01 e	9.39 ± 0.04 e
5	BRCC3	0.48 ± 0.00 d	0.18 ± 0.00 d	5.62 ± 0.02 d	5.46 ± 0.08 c	4.87 ± 0.01 d	11.29 ± 0.04 d
6	BRCC4	0.57 ± 0.01 b	0.18 ± 0.00 d	6.27 ± 0.02 c	6.52 ± 0.51 b	6.16 ± 0.19 b	14.49 ± 0.02 c
7	BRV3	0.60 ± 0.01 a,b	0.23 ± 0.00 b	6.70 ± 0.22 b	6.71 ± 0.42 b	6.71 ± 0.25 a	15.18 ± 0.02 b
8	BRV4	0.63 ± 0.01 a	0.19 ± 0.00 d	7.56 ± 0.16 a	7.24 ± 0.01 a	7.22 ± 0.22 a	15.84 ± 0.02 a

¹ Results are presented as the mean of three measurements; ^a–f different letters in columns indicate statistically different values (Tuckey’s HSD test, p < 0.05). TPC—total polyphenol content; TFC—total flavonoids content.

.

As in the previous analyses in the present study of the obtained R. alba L. extracts, there is a clear trend for augmentation in the concentration of phenolic components with intensification of the extraction conditions: an increase in the temperature and extraction time (Table 5). The highest amounts of polyphenols were observed in the extracts derived from rose petals (0.60–0.63 mg/mL) for 15 and 30 min. The extracts obtained from the whole flowers exhibited similar polyphenols content. This might indicate that these compounds do not undergo significant degradation and are sufficiently stable at the applied extraction conditions.

The presence of flavonoids and their concentrations could not unequivocally be associated with a specific part of the plant, neither for the temperature nor for the duration of extraction. For the whole flower, a slight decrease of about 15% in the values was observed with longer extraction, which may be a consequence of the destruction of certain flavonoid structures. Verma et al. [28] reported that the flavonoid content in R. alba L. flowers is about 18 mg/g; apparently the subcritical water extraction applied almost completely extracts these substances.

As mentioned above, to date, this is (to the best of our knowledge) the first report on the SWE extracts from R. alba L., and for this reason they are compared with similar distillation products, including hydrosol and wastewater. Ilieva et al. [29] found that the wastewater from steam distillation of R. damascena had the highest amount of polyphenols (7.6 mg/mL) compared to those of R. damascena and R. centifolia.

The products of hydroalcoholic extraction of R. damascena have higher levels of total polyphenol and total flavonoid content, but are concentrated (or even lyophilized) and cannot be compared with our extracts [30,31]. A concentrated aqueous extract of R. damascena has a total polyphenol content of about 110 µg/mL and a total flavonoid content of about 176 µg/mL [32,33].

In terms of antioxidant activity, a clear relationship emerges between the content of total phenols and the biological effect. The phenolic compounds are recognized as free radical scavengers and they count for the majority of the antioxidant activity of plants. The mechanism of action is mostly derived from their metal ion-chelating and hydrogen-donating abilities. Antioxidant activity is usually measured by the DPPH method, but in the present study, the inhibitory power was investigated using the DPPH, ABTS, FRAP, and CUPRAC methods, each using different chromogenic redox reagents with different standard potentials. The DPPH and ABTS analyses are based on reaction with organic radical and provide simplicity and high sensitivity. The FRAP and CUPRAC use reduction with metal ions. They are fast and cost-effective, and do not require specialized equipment. The ABTS and CUPRAC tests can measure both hydrophilic and lipophilic antioxidants, the FRAP method only measures hydrophilic antioxidants, and the DPPH method only applies to hydrophobic systems. The experimental data from the analyses showed (Table 5) close but statistically different values. The antioxidant capacities determined by the DPPH, ABTS, FRAP, and CUPRAC methods were in the 4.06–7.56 mM TE/mL, 3.69–7.24 mM TE/mL, 3.65–7.22 mM TE/mL, and 8.66–15.84 mM TE/mL range, respectively. Similar antioxidant activity determined by the DPPH and CUPRAC methods, around 2640.53 ± 75.70 mg gallic acid/100 g and 3288.00 ± 53.31 mg TE/100 g, respectively, was reported for the R. alba L. flowers [27]. However, the extracts’ antioxidant capacities were significantly lower than the ascorbic acid (18.22 µg/mL) and rosmarinus subcritical water extract (11.3 μg/mL) [34].

In the SWE aqueous extracts, three major phenolic acids were identified: gallic, ferulic, and rosmarinic acids (

Table 6. Individual phenolic acids and flavonoids in the R. alba L. SWE extracts.

Compound	Concentration, µg/mL
	BRCC1	BRCC2	BRV1	BRV2	BRCC3	BRCC4	BRV3	BRV4
Gallic acid	44.45 ± 1.01 e,1	39.07 ± 0.86 f	30.92 ± 0.98 g	40.79 ± 1.05 f	68.17 ± 1.05 d	113.37 ± 1.35 a	104.34 ± 1.35 c	108.85 ± 0.99 b
Protocate–huic acid	NF *	NF	NF *	NF	NF	13.05 ± 0.94 c	19.22 ± 0.67 b	25.48 ± 0.86 a
Ferulic acid	99.96 ± 1.80 a	47.21 ± 1.62 e,f	77.49 ± 1.02 c	87.56 ± 1.68 b	51.36 ± 1.85 e	44.50 ± 1.47 f	84.26 ± 1.57 b	67.63 ± 1.77 d
Rutin	47.13 ± 1.21 a	19.25 ± 0.99 e	33.04 ± 0.89 c	37.76 ± 1.02 b	16.58 ± 1.35 f	13.53 ± 0.86 g	31.01 ± 1.14 c	25.00 ± 0.94 d
Hesperidin	10.95 ± 0.94 a	2.51 ± 0.87 d	7.42 ± 0.96 b	6.12 ± 0.85 b,c	7.68 ± 0.88 b	5.77 ± 1.02 c	10.95 ± 0.94 a	2.51 ± 0.87 d
Rosmarinic acid	65.12 ± 1.74 b	25.27 ± 1.16 f	50.06 ± 1.02 c	46.95 ± 1.11 d	79.41 ± 1.34 a	80.47 ± 1.20 a	65.12 ± 1.74 b	25.27 ± 1.16 f
Quercetin	NF	NF	NF	NF	8.35 ± 0.89 b	8.25 ± 0.94 b	NF	NF
Kaemphe–rol	NF	NF	NF	NF	ULOQ **	0.311 ± 0.08	NF	NF

**—Under the limit of quantification; *—Not found. ¹ Results are presented as the mean of three measurements; ^a–g different letters in columns indicate statistically different values (Tuckey’s HSD test, p < 0.05).

). Protocatechuic acid was detected only in the extracts obtained at 150 °C. Gallic acid was more completely extracted at higher extraction temperatures and a longer extraction time. The concentrations determined were about two times higher in the extracts obtained at 150 °C (104–113 µg/mL) compared to those obtained at 100 °C (30–44 µg/mL). The behavior of the other two acids, ferulic and rosmarinic acids, is similar, but at 150 °C and an extraction time of 30 min their amounts decrease due to hydrolysis and oxidative processes. (+)-Catechin, (−)-epicatechin, chlorogenic acid, vanillic acid, caffeic acid, syringic acid, p-coumaric acid, and salicylic acid were not detected in the R. alba L. SWE extracts and the essential oil.

Two flavonoid glycosides were mainly identified in the extracts: rutin and hesperidin (Table 6). Rutin is in three–four times higher (15–40 µg/mL) concentrations than hesperidin (2–10 µg/mL). Higher yields were observed in the extracts obtained with a shorter extraction time (15 min) compared to those obtained for 30 min, due to hydrolysis processes occurring in flavonoid glycosides. Quercetin was only found in whole flower extracts obtained at 150 °C (about 8 µg/mL). Quercetin is most likely a hydrolysis product of rutin during the extraction process. Our results confirm the study of Ko et al. [35] on the relationship between flavonoid structure and SWE, namely that the optimal temperature for extraction of glycosides is 50 degrees lower than that for aglycones.

4. Conclusions

Subcritical water extraction, a promising “green” method for extraction of raw plant materials was successfully used as an efficient method for obtaining extracts from Rosa alba L. flowers and petals. To the best of our knowledge this is the first report in the scientific literature on obtaining of R. alba L. SWE aroma products. The experimental design included investigations at two factors and two levels. The factors chosen were temperature (at 100 °C and 150 °C levels) and duration (15 and 30 min). Extraction with superheated water for 30 min extracts almost all of the volatile substances (0.010–0.060%), bearing in mind that, according to the standard, the content of essential oil in R. alba L. flowers is in the 0.025–0.050% range. The analyses for determination of extracts chemical composition revealed they exhibited a complex composition. Compared to traditional rose oil obtained from R. alba L., the volatile components found in the subcritical water extracts have the same qualitative profile, but a different quantitative composition. The essential oil is rich in monoterpene alcohols (geraniol (19.71%), citronellol, and nerol (14.92%), and aliphatic hydrocarbons (nonadecane (13.21%) and heneicosane (11.86%)), while phenylethyl alcohol (6–25%) and linalool (3–20%) predominate in the extracts.

In addition to volatile substances, the extracts contain proteins from 152.62 ± 2.18 µg/mL to 240.31 ± 3.63 µg/mL. The monosaccharides found included glucose (3.09–15.29 mg/mL), galacturonic acid (1.02–2.34 mg/mL), galactose (0.18–0.78 mg/mL), rhamnose (0.17–0.48 mg/mL), and xylose (0.07–0.17 mg/mL). The phenolic substances determined were represented by phenolic acids (gallic acid: 30.92–113.37 µg/mL; ferulic acid: 44.50–99.96 µg/mL; rosmarinic acid: 25.27–80.47 µg/mL; protocatechuic acid: 13.05–25.48 µg/mL) and flavonoids (both quercetin and kaempferol: 8.35–8.56 µg/mL; and their glycosides: 15.91–58.08 µg/mL). The phenolic type of compounds possess free radical scavenging activity and their presence count for the majority of the antioxidant activity of plants and their extracts. The antioxidant capacities determined by four methods, namely the DPPH, ABTS, FRAP and CUPRAC methods, were in the 4.06–7.56 mM TE/mL, 3.69–7.24 mM TE/mL, 3.65–7.22 mM TE/mL, and 8.66–15.84 mM TE/mL range, respectively. Similar results for antioxidant activity (determined by DPPH and FRAP), around 6 mM TE/mL, exhibit 70% ethanol for extracts of R. alba waste rose flowers (obtained after steam-water distillation and subcritical extraction with CO₂).

The analyses performed suggests that the extract composition differs from the main aroma products derived from essential oil-bearing roses (rose oil) and rose extracts obtained by liquefied gasses (supercritical CO₂ extract and subcritical freon extract). The extracts obtained by SWE are close in chemical composition to rose hydro distillates, which are valuable raw materials used in cosmetics. The results from the present investigation suggest that SWE could be regarded as a promising method for the extraction of R. Alba L. and the extracts could be utilized as food supplements and for cosmetic products.