Characterization of Phenolic Constituents and Antioxidant Capacity in Rosemary Extracts Produced by Different Extraction Methods

1. Introduction

Rosmarinus officinalis L. known as rosemary is a perennial aromatic species of the Lamiaceae family, naturally found almost in all Mediterranean area (as well as in Albania) and widely used as both a culinary and medicinal plant. Beyond its traditional value, rosemary has become an important raw material for scientific and industrial purposes because of its diverse secondary metabolites and relevant biological effects. In particular, rosemary extracts are increasingly explored as natural antioxidant sources for food protection, nutraceutical products, cosmetics, and pharmaceutical formulations [1,2,3]. This growing attention is also linked to the current preference for plant-derived antioxidants that may replace or reduce the use of synthetic additives in food and health-related applications.

The functional potential of rosemary is mainly connected with its phenolic fraction, which includes phenolic diterpenes, phenolic acids, and flavonoids. Carnosic acid and carnosol are regarded as the principal lipophilic antioxidant diterpenes, whereas rosmarinic and caffeic acids are among the most relevant polar phenolic acids. Other compounds, including luteolin, apigenin, and hesperidin, may further enhance the bioactivity of the extract. These constituents are involved in radical-scavenging reactions, reducing activity, metal chelation, inhibition of lipid oxidation, and antimicrobial and anti-inflammatory responses [4,5,6,7]. The antioxidant behavior of rosemary is therefore strongly related to the combined action of carnosic acid, carnosol, and rosmarinic acid, whose effectiveness depends on their polarity, structure, and interaction with different oxidative systems [2,8].

The concentration of these bioactive molecules is not constant and may vary according to geographical origin, climate, growth conditions, harvest season, drying or storage practices, solvent type, and extraction procedure. Such factors can modify both the accumulation and preservation of secondary metabolites, producing extracts with different chemical profiles and antioxidant responses [9,10]. Consequently, the selection of an appropriate harvest period and extraction protocol is essential when rosemary extracts are intended for standardized products rich in target compounds.

Several procedures have been applied to isolate bioactive constituents from rosemary. Traditional approaches, including maceration, tincture preparation, Soxhlet extraction, and hydro-distillation, remain common because they are technically simple and accessible. However, they may require long processing times, large solvent volumes, and high temperatures, while also offering limited selectivity. Hydro-distillation is mainly appropriate for volatile essential oils and is less suitable for non-volatile phenolics. Hydroalcoholic extracts are more effective for polar compounds such as rosmarinic acid, whereas oil-based tinctures can be less efficient for many phenolic antioxidants because of solubility and mass-transfer limitations [5,11].

Greener and more selective technologies, particularly supercritical carbon dioxide extraction (SC-CO₂), have received considerable interest for aromatic and medicinal plants. Supercritical CO₂ is non-toxic, non-flammable, easily removed after extraction, and can be used at relatively mild temperatures, which helps protect thermolabile constituents. In rosemary, this technique is especially suitable for lipophilic diterpenes such as carnosic acid and carnosol. In contrast, more polar compounds, including rosmarinic acid, are recovered less efficiently unless polar co-solvents or combined extraction schemes are used [12,13]. Therefore, the comparison of conventional and advanced methods is necessary to select the extraction strategy that best matches the desired chemical composition and final use of the extract.

A reliable evaluation of rosemary extracts also requires analytical methods able to cover compounds with different polarity and structural properties. Liquid chromatography coupled with mass spectrometry (LC–MS) is well suited for determining phenolic acids, flavonoids, and diterpenes in complex plant matrices. Previous studies have shown that chromatographic and mass spectrometric techniques are useful for linking the phenolic profile of rosemary to its antioxidant performance [4,6]. Since antioxidant activity cannot be fully described by a single test, complementary in vitro assays are commonly applied. ABTS, DPPH, and FRAP are widely used to assess radical-scavenging ability and reducing power in rosemary extracts [11,14,15,16].

Although rosemary has already been widely investigated, comparative studies performed under the same analytical conditions remain useful, especially when both phenolic composition and antioxidant capacity are considered. They allow the identification of extraction methods that preferentially concentrate lipophilic diterpenes, polar phenolic acids, or broader antioxidant fractions. In addition, the study of samples collected in different seasons can clarify how harvesting period contributes to phytochemical variability and extract quality.

This study evaluates the phenolic composition and antioxidant activity of Rosmarinus officinalis L. collected in the Durres area, Albania, during January and May 2025. Different extraction approaches were compared, namely SC-CO₂ extraction, Soxhlet extraction with and without ethanol, ethanol tinctures, oil-based tinctures, and hydro-distillation. Caffeic acid, hesperidin, rosmarinic acid, luteolin, apigenin, carnosol, and carnosic acid were quantified by LC–MS, while antioxidant potential was determined using ABTS, DPPH, and FRAP assays. This design enables evaluation of the relationships among extraction method, seasonal variation, phenolic profile, and antioxidant response, and supports the selection of suitable procedures for obtaining high-value rosemary extracts for food, nutraceutical, cosmetic, and pharmaceutical applications.

2. Materials and Methods

2.1. Sampling of Rosmarinus Officinalis Plants

The samples of Rosmarinus officinalis L. plants (leaves) were taken from the cultivated populations of Durres area (Lalez Bay). Rosemary plants were selected at 10 different stations of this area. The plants were sampled in January and May 2025. The plants were air dried in shadow in order to save their morphological characteristics. The dried plant material was grinded into small pieces < 0.5 cm for further step of analysis.

2.2. Isolation of Essential Oils for Rosmarinus Officinalis by Using HD Technique

A quantity of 100 g of dry plant of Rosmarinus officinalis L. was subjected to hydro-distillation for 4 hours by using the Clevenger type apparatus (recommended by European Pharmacopeia) for the isolation of the essential oil. The essential oil was dehydrated by adding anhydrous sodium sulfate and stored in dark vials at +4^oC. The essential oil of Rosmarinus officinalis was subject of LC-MS and antioxidant analysis [5,11,17].

2.3. Isolation of Essential Oil by Using Supercritical Carbon Dioxide (SC-CO₂)

A total of 50 g of ground aerial parts of Rosmarinus officinalis L. were extracted using the Super C apparatus (OCO Labs) with supercritical CO₂ at 150 bar and 32.5 °C for 45 minutes. After CO₂ evaporation, the extract was collected in a chromatographic vial for further analyzes.

2.4. Isolation of Essential Oils by Using Soxhlet Apparatus

Rosemary samples (10 g of dry leaves) were put on the thimble. A filter paper was used to cover the sample. A quantity of 200 ml solvent (water, ethanol and/or ethanol/water) was used as extracting solvent. Extracting process by using Soxhlet apparatus continued for 6 hours. After that, extracts of rosemary were concentrated up to 20 ml by using rotary evaporator. The essential oil of Rosmarinus officinalis was subjected to LC-MS and antioxidant analysis.

2.5. Base Oil Tinctures

Also, oil tincture were prepared from the leaves of the rosemary plant. Plant material was put on a jar, and cover with olive oil. The plant material and oil interact for 6 weeks, in the absence of sunlight. It were shaken every day. The mixture were transfer to a tincture bottle. This extract were used for further analyses.

2.6. Alcoholic Tinctures

One of the most common uses of rosemary in traditional medicine are ethanol tinctures. They were prepared from the leaves of the rosemary plant by putting on a jar, and cover with ethanol (80%). The plant material and ethanol act together for 6 weeks (out of sunlight). It were shaken every day and in the end of six weeks the mixture were transfer to a tincture bottle. The extract were subject of phenolic content and antioxidant capacity.

2.7. Preparation of Rosemary Extracts for LC-MS Analyze

Rosemary extracts (0.2 g) were dissolved with methanol (20 mL) and subjected to sonication for 3 min at room temperature, followed by filtration through a 0.22 μm PTFE membrane. After appropriate dilution, the extracts were injected into the LC–MS system for analysis. The oil-containing extracts were initially treated with DMSO to enhance solubility and were subsequently diluted with isopropanol to obtain suitable concentrations for LC–MS analysis.

2.8. LC-MS Analysis

The major components of rosemary extracts were quantified using a Shimadzu Nexera HPLC system with a diode array detector and a single quadrupole mass spectrometer (LCMS-2020) featuring an electrospray ionization (ESI) interface. Separations were performed on a Poroshell 120 EC-C₁₈ analytical column (4.6 × 150 mm, 4 μm). The mobile phases consisted of solvent A (0.1% formic acid in water, v/v) and B (acetonitrile). The sample was eluted by gradient elution at a constant flow rate of 0.5 mL/min and a temperature of 35 ◦C. Gradients were programmed as follows: 0 min, 15% B; 5 min, 25% B; 10 min, 35% B; 28 min, 60% B; 28.1 min, 60% B; 35 min, 100% B, 35,01’ 85% A, 42’ 85% A. All solvents used in chemical analysis were of LC-MS grade and the analytical standards of phenolic compounds were purchased from Sigma-Aldrich (Steinheim, Germany). The applied MS conditions were: nebulizing gas (N₂) flow rate of 1.5 L/min, drying gas (N₂) flow rate of 15 L/min, interface temperature of 350 ^oC, heat block temperature of 200 ^oC, DL (desolvation line) temperature of 250 ◦C, and interface and DL voltage of −4.5 kV and 1 V, respectively. The injection volume was 10 μL and the quantitative determination of target compounds in the phenolic extracts was carried out in negative ion and SIM mode, by constructing calibration curves of the corresponding standard solutions at five concentration levels within a linear range of 0.01–4 μg/mL [9,18]. Data acquisition and processing were done using Lab Solutions LC-MS software (Shimadzu, Kyoto, Japan) and the results were expressed as mg per g of extract (mg/g).

Figure 1. LC-MS chromatogram for the sample 10, SC-CO2 extract.

Figure 1. LC-MS chromatogram for the sample 10, SC-CO2 extract.

2.9. ABTS Radical Scavenging Assay

The ABTS assay was performed according to the protocol of Re et al. (1999). 2,2-azino-bis(3-ethylbenzothiazoline)-6-sulfonic acid radical cation (ABTS^⋅+) was prepared by reacting 7 mM ABTS aqueous solution with 2.6 mM potassium persulfate (final concentration) and allowing the mixture to stand in the dark at room temperature for 12–17 h before use. The working solution of the radical cation was prepared directly before analysis by diluting the stock solution with water to obtain an absorbance value at 734 nm of 0.7 ± 0.02. For analysis, 100 μL extract was mixed with 3.9 mL of the ABTS^⋅+ solution and after 4 min the absorbance was recorded at 734 nm against a blank [9,19,20,21,22]. The ABTS^⋅+ scavenging effect (% inhibition) was calculated by using the following equation:

Inhibition (%) = (Ao-As)/Ao × 100

where: Ao is the absorbance of the blank sample and As is the absorbance of the sample at 4 min.

Trolox (6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid), a water-soluble analogue of vitamin E, was used for standard calibration. The ABTS values were the means of triplicate analyses and expressed as mg Trolox equivalents per g of extract (mg TE/g).

2.10. DPPH Radical Scavenging Assay

The DPPH (2,2-diphenyl-1-picrylhydrazyl) scavenging activity of rosemary extracts was measured according to a previous report (Yen & Chen, 1995) with slight modifications. Briefly, 100 μL of extract was mixed with 2.85 mL of freshly prepared 0.1 mM DPPH in methanol and the decrease in absorbance was measured at 516 nm after 5 min of reaction (i.e., the decrease in absorbance is proportional to the antioxidant capacity of the sample). The DPPH radical scavenging activity was calculated by using the following equation:

Inhibition (%) = (Ao-As)/Ao × 100

where: Ao is the absorbance of the blank sample and As is the absorbance of the sample at 5 min.

Trolox was used for standard calibration. The DPPH values were the means of triplicate analyses and expressed as mg Trolox equivalents per g of extract (mg TE/g) [19,20,21,22].

2.11. Ferric Reducing Antioxidant Power (FRAP) Assay

The FRAP assay was carried out according to the method of Benzie and Strain (1999). Briefly, the FRAP reagent was prepared directly before analysis by mixing 20 mM ferric chloride solution, 10 mM TPTZ (2,4,6-tripyridyl-s-triazine) solution in 40 mM HCl, and 0.3 mM acetate buffer pH 3.6, in a proportion of 1:1:10, respectively. Aliquots of 100 μL of appropriately diluted extracts were incubated with 3 mL FRAP solution at 37 ◦C and exactly after 4 min the absorbance was recorded at 593 nm [15]. Trolox was used as an antioxidant standard and the FRAP results were also expressed as mg Trolox equivalents per g of extract (mg TE/g).

3. Results and Discussions

The present study evaluated the phenolic composition and antioxidant capacity of Rosmarinus officinalis L. extracts obtained through different extraction techniques, including supercritical CO₂ (SC-CO₂), Soxhlet extraction (with and without ethanol), ethanol tinctures, oil-based tinctures, and hydro-distillates. The quantification of major phenolic constituents (caffeic acid, hesperidin, rosmarinic acid, luteolin, apigenin, carnosol, and carnosic acid) alongside antioxidant activity assessed by ABTS, DPPH, and FRAP assays enabled a comparative assessment of the influence of extraction method and sampling month on extract quality. The findings aims to demonstrate variation/connections between phenolic content and antioxidant functionality across different extraction techniques.

3.1. Phenolic Composition of the Extracts

The quantitative analysis of phenolic compounds revealed pronounced differences across extraction methods. SC-CO₂ extracts (samples 10 and 12) consistently contained the highest concentrations of the key lipophilic diterpenes carnosic acid and carnosol, which are recognized as the primary antioxidant constituents of rosemary. Carnosic acid reached exceptionally high levels in SC-CO₂ extracts, ranging from approximately 289 to 446 mg/g depending on the sample and replicate. Carnosol concentrations also remained elevated, from around 21 to 29 mg/g. In addition to diterpenes, these extracts contained appreciable amounts of phenolic acids and flavonoids, such as rosmarinic acid (3.1–3.8 mg/g), caffeic acid (2.8–3.8 mg/g), and hesperidin (2.3–3.2 mg/g). These values substantially exceeded those obtained through any other extraction technique.

Soxhlet extraction performed without ethanol (samples 7 and 8) generated extracts with relatively high levels of rosmarinic acid, ranging from 5.7 to 8.1 mg/g, suggesting enhanced extraction of polar phenolics in this solvent environment. Moreover, the January sample (sample 8) exhibited unusually high hesperidin (4.2–5.0 mg/g), luteolin (3.3–3.5 mg/g), and moderately elevated carnosic acid (24–63 mg/g), leading to a balanced mixture of both polar and moderately lipophilic compounds. Although Soxhlet without ethanol produced meaningful levels of phenolic constituents, diterpene concentrations remained markedly lower than those observed in SC-CO₂ extracts.

On the other hand, Soxhlet extraction with ethanol (samples 3 and 6) yielded overall lower phenolic concentrations across all analytes. Carnosic acid levels exceeded 32 mg/g, and rosmarinic acid remained between 1.6–1.7 mg/g, substantially inferior to Soxhlet without ethanol. This suggests that the presence of ethanol during high-temperature extraction may reduce diterpene yield or promote degradation due to diterpenes’ sensitivity to prolonged heating.

Ethanol tinctures (samples 1 and 4) contained moderate quantities of phenolic acids and flavonoids but negligible diterpenes. Rosmarinic acid ranged from 1–5 mg/g, whereas carnosic acid levels were extremely low (3–7 mg/g). These results confirm that hydroalcoholic systems preferentially extract hydrophilic phenolics while being ineffective for lipophilic antioxidants.

Oil-based tinctures (samples 2 and 5) presented the poorest phenolic profiles, with most compounds remaining below quantification limits. Only minimal traces of carnosic acid (0.7–4.5 mg/g) were detectable. Given that rosemary diterpenes are moderately lipophilic, their poor solubility in the oil matrix suggests inefficiencies related to the extraction mechanism rather than polarity alone.

The hydro-distillates of Clevenger and SC-CO2 (sample 9 and 11) exhibited extremely low concentrations for all phenolics, consistent with the expectation that hydro-distillation primarily recovers volatile essential oils rather than phenolic constituents.

Table 1. Concentration (mg/g) of the major phenolic compounds determined in rosemary extracts.

Code	Extract	Month	Caffeic acid	Hespe-ridin	Rosmari-nic acid	Luteolin	Apigenin	Carnosol	Carnosic acid
12	SC-CO2	May	2,811	2,312	3,127	0,233	0,979	29,443	383,091
			3,407	2,803	3,790	0,282	1,186	28,757	445,758
10	SC-CO2	January	3,237	2,634	3,156	0,335	0,988	27,638	289,691
			3,848	3,166	3,440	0,319	1,340	21,617	317,993
7	Soxhlet without EtOH	May	0,569	1,432	8,136	1,076	0,119	12,904	50,707
			0,657	1,353	7,405	0,954	0,124	11,339	63,161
8	Soxhlet without EtOH	January	1,001	4,198	7,747	3,473	0,175	65,306	23,987
			0,877	4,973	5,668	3,334	0,204	77,083	26,198
6	Soxhlet	May	0,315	0,435	1,732	0,257	0,094	2,623	24,324
			0,454	0,611	1,658	0,235	0,103	2,512	32,462
1	tincture of ethanol	January	0,185	0,492	4,923	0,244	0,025	3,398	6,833
			0,145	0,392	5,127	0,284	0,026	3,400	7,080
5	tincture with oil	May	0,341	0,030	0,041	0,463	0,017	0,209	4,522
			0,322	0,047	0,058	0,472	0,015	0,164	3,812
2	tincture with oil	January	0,018	0,003	0,006	0,102	0,002	0,142	0,737
			0,019	0,004	0,005	0,111	0,003	0,160	0,818
3	Soxhlet	January	0,041	0,263	0,031	0,198	0,004	0,032	0,212
			0,046	0,194	0,034	0,152	0,003	0,032	0,192
4	tincture of ethanol	May	0,225	0,248	1,037	0,640	0,037	0,037	0,098
			0,205	0,218	1,127	0,542	0,026	0,038	0,092
9	Clevenger hydrodistillates	May	<LOQ	<LOQ	0,001	0,001	0,001	0,001	<LOQ
			<LOQ	<LOQ	0,001	0,002	0,001	0,001	<LOQ
11	hydrodistillate after SC-CO2	May	0,004	0,011	0,016	0,008	0,001	0,011	<LOQ
			0,003	0,009	0,013	0,007	0,001	0,010	<LOQ

LOQ: limit of quantification.

Table 1. Concentration (mg/g) of the major phenolic compounds determined in rosemary extracts.

Code	Extract	Month	Caffeic acid	Hespe-ridin	Rosmari-nic acid	Luteolin	Apigenin	Carnosol	Carnosic acid
12	SC-CO2	May	2,811	2,312	3,127	0,233	0,979	29,443	383,091
			3,407	2,803	3,790	0,282	1,186	28,757	445,758
10	SC-CO2	January	3,237	2,634	3,156	0,335	0,988	27,638	289,691
			3,848	3,166	3,440	0,319	1,340	21,617	317,993
7	Soxhlet without EtOH	May	0,569	1,432	8,136	1,076	0,119	12,904	50,707
			0,657	1,353	7,405	0,954	0,124	11,339	63,161
8	Soxhlet without EtOH	January	1,001	4,198	7,747	3,473	0,175	65,306	23,987
			0,877	4,973	5,668	3,334	0,204	77,083	26,198
6	Soxhlet	May	0,315	0,435	1,732	0,257	0,094	2,623	24,324
			0,454	0,611	1,658	0,235	0,103	2,512	32,462
1	tincture of ethanol	January	0,185	0,492	4,923	0,244	0,025	3,398	6,833
			0,145	0,392	5,127	0,284	0,026	3,400	7,080
5	tincture with oil	May	0,341	0,030	0,041	0,463	0,017	0,209	4,522
			0,322	0,047	0,058	0,472	0,015	0,164	3,812
2	tincture with oil	January	0,018	0,003	0,006	0,102	0,002	0,142	0,737
			0,019	0,004	0,005	0,111	0,003	0,160	0,818
3	Soxhlet	January	0,041	0,263	0,031	0,198	0,004	0,032	0,212
			0,046	0,194	0,034	0,152	0,003	0,032	0,192
4	tincture of ethanol	May	0,225	0,248	1,037	0,640	0,037	0,037	0,098
			0,205	0,218	1,127	0,542	0,026	0,038	0,092
9	Clevenger hydrodistillates	May	<LOQ	<LOQ	0,001	0,001	0,001	0,001	<LOQ
			<LOQ	<LOQ	0,001	0,002	0,001	0,001	<LOQ
11	hydrodistillate after SC-CO2	May	0,004	0,011	0,016	0,008	0,001	0,011	<LOQ
			0,003	0,009	0,013	0,007	0,001	0,010	<LOQ

LOQ: limit of quantification.

Figure 2. Carnosic content in extract of rosemary samples obtained by different extraction methods.

Figure 2. Carnosic content in extract of rosemary samples obtained by different extraction methods.

3.2. Antioxidant Activity

The antioxidant capacity measured by ABTS, DPPH, and FRAP strongly mirrored the distribution of phenolic constituents, particularly diterpenes. SC-CO₂ extracts recorded exceptionally high antioxidant activity, confirming the biochemical relevance of their rich diterpene content. Sample 12 (May, SC-CO₂) displayed ABTS values between 289–312 mg TE/g, DPPH between 196–252 mg TE/g, and FRAP between 181–200 mg TE/g. Sample 10 (January, SC-CO₂) showed comparatively lower but still substantial values (ABTS 130–153 mg TE/g; DPPH 95–106 mg TE/g; FRAP 101–125 mg TE/g). These results reinforce that SC-CO₂ selectively enriches carnosic acid and carnosol—powerful radical scavengers and reducing agents known to dominate rosemary’s antioxidant activity.

Soxhlet without ethanol (with water) delivered intermediate antioxidant activity, particularly in sample 8 (January), where ABTS reached 160–167 mg TE/g and DPPH 101–125 mg TE/g. The elevated phenolic levels in this sample correlate with the moderate antioxidant values, supporting the established relationship between phenolic acid concentration and free radical scavenging efficiency. However, FRAP values remained modest (42–46 mg TE/g), potentially due to the lower diterpene content, given that FRAP often favors compounds with strong reducing potential, such as carnosic acid.

Soxhlet with ethanol extracted considerably less potent fractions, yielding ABTS activity as low as 10–12 mg TE/g and DPPH values near 7 mg TE/g. FRAP remained similarly low. These diminished activities reflect both reduced phenolic recovery and instability under high-temperature ethanol extraction conditions.

Ethanol tinctures and oil tinctures exhibited minimal antioxidant activity, with most values falling under 10 mg TE/g in all assays. Since these systems lack both polar phenolics (in the oil tinctures) and diterpenes (in the ethanol tinctures), low antioxidant activity was anticipated.

Hydro-distillates showed activity close to zero (<1 mg TE/g), consistent with the almost complete absence of phenolic constituents.

Table 2. Antioxidant activity (mg TE/g) of rosemary extracts determined by ABTS, DPPH and FRAP tests.

code	extract	month	ABTS	DPPH	FRAP
12	SC-CO2	May	312,107	251,823	181,096
			289,070	196,465	200,360
10	SC-CO2	January	153,306	106,170	101,650
			130,806	95,015	125,362
7	Soxhlet without EtOH	May	75,196	54,922	57,452
			78,596	60,239	60,895
8	Soxhlet without EtOH	January	167,249	124,678	46,357
			160,290	101,576	42,412
6	Soxhlet	May	11,726	7,641	7,769
			10,447	7,140	8,431
1	tincture of ethanol	January	10,936	7,325	13,521
			13,340	7,431	12,085
5	tincture with oil	May	3,079	2,255	-
			3,134	2,062	-
2	tincture with oil	January	1,298	0,570	-
			1,266	0,534	-
3	Soxhlet	January	4,785	2,655	3,805
			4,097	2,570	3,744
4	tincture of ethanol	May	7,859	4,262	6,429
			8,172	4,078	5,967
9	Clevenger hydro-distillates	May	0,270	0,089	-
			0,274	0,099	-
11	hidrodistillate after SC-CO2		0,765	0,049	0,200
			0,700	0,035	0,185

TE: Trolox equivalent.

Table 2. Antioxidant activity (mg TE/g) of rosemary extracts determined by ABTS, DPPH and FRAP tests.

code	extract	month	ABTS	DPPH	FRAP
12	SC-CO2	May	312,107	251,823	181,096
			289,070	196,465	200,360
10	SC-CO2	January	153,306	106,170	101,650
			130,806	95,015	125,362
7	Soxhlet without EtOH	May	75,196	54,922	57,452
			78,596	60,239	60,895
8	Soxhlet without EtOH	January	167,249	124,678	46,357
			160,290	101,576	42,412
6	Soxhlet	May	11,726	7,641	7,769
			10,447	7,140	8,431
1	tincture of ethanol	January	10,936	7,325	13,521
			13,340	7,431	12,085
5	tincture with oil	May	3,079	2,255	-
			3,134	2,062	-
2	tincture with oil	January	1,298	0,570	-
			1,266	0,534	-
3	Soxhlet	January	4,785	2,655	3,805
			4,097	2,570	3,744
4	tincture of ethanol	May	7,859	4,262	6,429
			8,172	4,078	5,967
9	Clevenger hydro-distillates	May	0,270	0,089	-
			0,274	0,099	-
11	hidrodistillate after SC-CO2		0,765	0,049	0,200
			0,700	0,035	0,185

TE: Trolox equivalent.

Figure 3. Antioxidant activity of rosemary samples.

Figure 3. Antioxidant activity of rosemary samples.

3.3. Influence of Sampling Period

Across all extraction methods, extracts produced in May tended to leave behind those collected in January, though this pattern was most apparent in SC-CO₂ extracts. Seasonal variation affects rosemary’s phytochemical accumulation, with spring growth typically associated with higher secondary metabolite levels. For SC-CO₂ extracts, May samples contained significantly more carnosic acid and consequently much stronger antioxidant activity. Interestingly, in Soxhlet without ethanol, the January sample exhibited higher phenolic levels and antioxidant activity, suggesting that the seasonal influence may vary depending on the extraction solvent and its affinity for different compound classes.

3.4. Correlations Between Phenolic Profile and Antioxidant Activity

The strong correlation between diterpene concentration (especially carnosic acid and carnosol) and antioxidant capacity is consistent across the dataset. SC-CO₂ extracts, which yielded the highest diterpene content, showed the strongest radical scavenging and reducing power. Polar phenolics such as rosmarinic acid and luteolin also contribute to antioxidant activity, particularly in the ABTS and DPPH assays, but appear less influential in FRAP performance. Oil tinctures and hydrodistillates, which lack phenolics almost entirely, demonstrated antioxidant activities near zero, reinforcing the central biochemical role of these compounds.

The data clearly shown that SC-CO₂ extraction as the most effective approach for producing antioxidant-rich rosemary extracts, due to its superior ability to preserve and concentrate lipophilic bioactive diterpenes. Soxhlet without ethanol, although less potent, remains a viable method for extracting hydrophilic phenolics. Hydroalcoholic tinctures, oil tinctures, and hydro-distillation produce extracts with poor phenolic yields and negligible antioxidant activity. Seasonal effects are evident but secondary to extraction method, with spring harvesting generally offering better phytochemical enrichment. The strong quantitative relationship between phenolic composition and antioxidant activity underscores the functional importance of diterpenes in rosemary and highlights SC-CO₂ extracts as promising candidates for nutraceutical, cosmetic, and food preservation applications.

Figure 4. Correlation (Heat-map) between phenolic compounds and antioxidant activity for rosemary extracts.

Figure 4. Correlation (Heat-map) between phenolic compounds and antioxidant activity for rosemary extracts.

4. Conclusions

This study demonstrates that the chemical composition and antioxidant activity of rosemary extracts clearly depend by the extraction technique and season of collection. Among all methods tested, supercritical CO₂ extraction (SC-CO₂) consistently produced extracts with the highest concentrations of bioactive diterpenes, particularly carnosic acid and carnosol, as well as appreciable levels of phenolic acids and flavonoids. These compositional characteristics translated directly into superior antioxidant performance across all assays (ABTS, DPPH, FRAP), confirming SC-CO₂ as the most efficient approach for obtaining potent antioxidant extracts from rosemary. Soxhlet extraction with water yielded intermediate results, performing notably better than Soxhlet with ethanol and tincture-based methods, especially in the recovery of rosmarinic acid and selected flavonoids. Although the antioxidant activity of these extracts was moderate, it remained substantially lower than that of SC-CO₂ extracts, largely due to reduced diterpene content. In contrast, Soxhlet with ethanol, ethanol tinctures, and oil-based tinctures produced extracts with very low phenolic content and correspondingly weak antioxidant capacity, highlighting their unsuitability for applications requiring high concentrations of phenolic antioxidants. Seasonal variation influence directly extract characteristics/phytochemistry. Generally, samples collected in May shown higher levels of phenolic compounds and superior antioxidant activity compared to those obtained in January. This trend was particularly pronounced in the SC-CO₂ extracts, suggesting that the accumulation of secondary metabolites in rosemary peaks during the late spring. However, in specific extraction systems such as Soxhlet with water, the January extracts occasionally exceed the May samples, indicating that solvent–matrix interactions can interact with seasonal effects in a more complex ways. These findings confirm a strong positive correlation between total phenolic content (especially diterpenes) and antioxidant capacity. The marked superiority of SC-CO₂ extraction underscores its potential as the optimal method for producing high-value rosemary extracts for nutraceutical, cosmetic, pharmaceutical, and food-preservation applications. This work demonstrate that both extraction method and harvest season are critical parameters that must be considered when optimizing rosemary extract quality.