Valorizing ginseng residues using dimethyl ether: Recovering functional lipids and ginsenosides Authors: Sabrinna Wulandari, Ruqian Cao, Jongho Choi, Aye Aye Myint, Jaehoon Kim Published: 2026 ABSTRACT BackgroundRed ginseng marc (RGM), a byproduct of hot water extraction, contains valuable lipid fractions and bioactive compounds. Conventional extraction methods often require harsh processing conditions and additional solvent removal steps, leading to high operational costs.MethodsDimethyl ether (DME) was used as a green solvent to extract ginseng oil from RGM. Response surface methodology (RSM) was employed to investigate the effects of temperature (303–333 K), pressure (3–7 MPa), and extraction time (20–80 min) on extraction yield, total ginsenoside content, antioxidant activity, and total phenolic content (TPC) of the extracted oil.ResultsUnder optimal conditions (333 K, 7 MPa, 80 min), the extraction yield reached 3.37%, with total ginsenoside content of 4.22 mg g−1 dry RGM, and TPC of 1.09 mg GAE g−1 RGM. DME-extracted ginseng oil also exhibited higher antioxidant activity (9.04 mmol TE g−1 extract) and FAME content (14.93 mg g−1 dry RGM) compared to ethanol and hexane extracts. Notably, DME enabled direct extraction from wet RGM without prior drying.ConclusionExtraction from wet RGM using DME resulted in higher ginsenoside content and extraction yield, but lower FAME content, compared to dry RGM. These findings highlight the potential of DME as a versatile, energy-efficient solvent for extracting bioactive ginseng oil from both dry and wet RGM, offering practical applications in the food and cosmetic industries. Introduction Red ginseng is simply white ginseng (Panax ginseng) that has been steamed and sun-dried in a process that changes its color to reddish-brown, extends its shelf life, and alters its chemical composition [1,2]. Ginsenosides, the primary bioactive compounds in ginseng, are highly valued for their health-promoting properties. These compounds constitute approximately 4% of the dry weight of the ginseng root [3] and are known to exhibit anti-aging and anti-inflammatory effects, as well as the ability to heal skin wounds [4]. In addition to ginsenosides, ginseng contains phenolic compounds, phytosterols, and fatty acids, with a lipid content of 1%–3% [5]. The red ginseng marc (RGM) byproduct remaining following hot-water extraction contains significant amounts of bioactive compounds and lipids. The lipid-soluble RGM fraction has demonstrated potential as a value-added ingredient for use in the pharmaceuticals and cosmetics industries. Ginseng is conventionally extracted using toxic solvents, which requires long processing times and an additional heating step to remove solvent residues. These harsh conditions may degrade the valuable bioactive compounds within the RGM. Wang et al. used supercritical carbon dioxide (scCO2) with ethanol as a co-solvent to improve the extraction efficiency for ginseng oil and ginsenosides from ginseng root hair [6]. However, this method requires prolonged extraction times and the use of a cosolvent to deliver satisfactory amounts of ginseng oil, resulting in high operational costs. Although ethanol can enhance skin permeation, it is also associated with skin dryness, irritation, and potential toxicity in children [7,8]. In addition, there are increasing consumer demands for alcohol-free cosmetic products [8]. Lee et al. extracted ginseng oil from RGM under reflux followed by fractionation and reported that the use of methanol for extraction and ethyl acetate for fractionation delivered the highest phenolic content and antioxidant activity [9]. However, their study did not examine the chemical profile or fatty-acid composition of the extracted oil, which leaves a gap in our understanding of the full potential of RGM-derived lipids. Extracting RGM with subcritical water is an environmentally friendly technique that enables rare ginsenosides and antioxidant compounds to be recovered highly efficiently in short times and without the use of toxic solvents; however, it requires high-temperature and high-pressure equipment, leading to high initial setup costs [10,11]. Dimethyl ether (DME, CH3OCH3, MW = 46 g mol−1) is characterized by its low boiling point (−24 °C), high affinity for oils, and partial water miscibility [12,13]; hence, DME is an ideal solvent for extracting oils and other valuable compounds from biomass because it is easily removed from the extract by simple evaporation at room temperature. DME is considered to be a green solvent; indeed, it is classified as “generally recognized as safe” (GRAS) by the U.S. Food and Drug Administration (FDA) [14]. DME has also been approved for use as an extraction solvent for food ingredients by the European Food Safety Authority (EFSA) [15]. Lipids are more soluble in DME than scCO2 and shorter extraction times are generally required, thereby reducing the overall amount of solvent used [16,17]. Moreover, DME extraction can be conducted at low temperatures, which significantly reduces energy consumption and operating costs [18]. In this study, we extracted ginseng oil from RGM using DME as the solvent, with Soxhlet extraction also performed using ethanol and hexane for comparison purposes. The effects of the process conditions on the extraction yield, ginsenoside content, antioxidant activity, and total phenolic content (TPC) of the DME extract were evaluated using the response surface methodology technique. Additionally, the ginsenoside profile, overall chemical composition, and fatty-acid content of the ginseng oil extracted using DME were analyzed and compared with those obtained by Soxhlet extraction. This study also examined the feasibility of DME extraction using wet RGM, which highlighted its potential as a direct and efficient extraction method. Materials and methods Experimental procedure for extraction DME extraction was conducted using the view-cell apparatus schematically depicted in Fig. S1. RGM was Soxhlet-extracted for comparison purposes. The experimental results were statistically analyzed using analysis of variance (ANOVA) to assess the significance and goodness-of-fit of the model [19]. Detailed descriptions on the apparatus and experimental procedure were given in Supplementary Notes 1. Characterization The chemical composition, ginsenoside content, fatty acid methyl esters (FAMEs), antioxidant activities and phenolic contents of RGM extracts were analyzed by using gas chromatography–time-of-flight mass spectrometry (GC–TOF/MS), high-performance liquid chromatography (HPLC), GC augmented with a flame ionization detector (GC-FID), and multimode microplate reader, respectively. The detailed explanations are given in Supplementary Notes 2. Results and discussion RGM extraction yields Extraction yields obtained by extracting RGM (particle size <250 μm, MC: 4.68%) with DME are listed in Table 1, which reveals DME extraction yields of between 2.1 and 3.4 wt% under the experimental conditions, namely a 303–333 K temperature range, a 3–7 MPa pressure range, and an extraction time of 20–80 min. Fig. 1 reveals how the process conditions affect the extraction yield. The various process parameters and extraction yield were found to be mathematically related according to Eq. (1):(1)Extractionyield=−125.46+0.77T−21.48P+3.25t+0.14TP−0.02Tt−0.07Pt+0.45P2where T, P, and t are the temperature, pressure, and extraction time, respectively. The coefficient of determination (R2) and p-value of the model were calculated to be 0.96 and 0.03, respectively, indicative of a strong and statistically significant correlation between the three process parameters and extraction yield. A maximum ginseng oil yield of 3.4 wt% was obtained at 333 K, 7 MPa, and an extraction time of 80 min. Elevated temperatures enhance solvent penetration into the RGM matrix, which increases the extraction efficiency. DME penetrates into the biomass more effectively under pressure owing to its lower viscosity and ability to disrupt van der Waals forces, hydrogen bonding, and dipole interactions within the biomass matrix [20]. Additionally, prolonged extraction times provide sufficient time for the bioactive compounds to diffuse from the biomass into the solvent, resulting in higher yields.Table 1Yields, antioxidant activity, and TPC values of RGM-extracts using DME.Table 1RunTa (K)Pb (MPa)tc (min)Yield (%, dry basis)Antioxidant (mmol TEd g−a dry RGM)TPCe (mg GAEf g−a dry RGM)13183502.50.180.8023333802.60.250.8433185502.50.210.7643033202.20.130.6453185502.50.180.7463033802.90.140.8173333202.60.200.7883035502.20.120.6193185802.60.200.83103185202.40.130.80113185502.60.170.84123037802.50.150.71133335502.30.230.78143337803.40.301.09153337202.10.180.70163187502.70.220.76173037202.20.220.67aT, temperature.bP, pressure.ct, extraction time.dTE, trolox equivalent.eTPC, total phenolic content.fGAE, gallic acid equivalent.Fig. 1Response surface plots showing how the process conditions affect the yield of the RGM DME extract: (A) pressure and temperature, (B) temperature and extraction time, and (C) pressure and extraction time (n=3, p=0.03, technical replicate type, error=2.51 ± 0.12). Table 1Yields, antioxidant activity, and TPC values of RGM-extracts using DME.Table 1RunTa (K)Pb (MPa)tc (min)Yield (%, dry basis)Antioxidant (mmol TEd g−a dry RGM)TPCe (mg GAEf g−a dry RGM)13183502.50.180.8023333802.60.250.8433185502.50.210.7643033202.20.130.6453185502.50.180.7463033802.90.140.8173333202.60.200.7883035502.20.120.6193185802.60.200.83103185202.40.130.80113185502.60.170.84123037802.50.150.71133335502.30.230.78143337803.40.301.09153337202.10.180.70163187502.70.220.76173037202.20.220.67aT, temperature.bP, pressure.ct, extraction time.dTE, trolox equivalent.eTPC, total phenolic content.fGAE, gallic acid equivalent. Yields, antioxidant activity, and TPC values of RGM-extracts using DME. T, temperature. P, pressure. t, extraction time. TE, trolox equivalent. TPC, total phenolic content. GAE, gallic acid equivalent. Response surface plots showing how the process conditions affect the yield of the RGM DME extract: (A) pressure and temperature, (B) temperature and extraction time, and (C) pressure and extraction time (n=3, p=0.03, technical replicate type, error=2.51 ± 0.12). In comparison, Soxhlet extraction with ethanol or hexane afforded extraction yields of 22.6 or 3.6 wt%, respectively. n-Hexane is a nonpolar solvent that effectively extracts nonpolar compounds from RGM, whereas ethanol, which is polar, is more compatible with polar components. The use of polar solvents, such as water or an alcohol, to extract ginseng affords extracts with higher carbohydrate contents that increase in proportion to the polarity of the solvent [21]. Ginsenoside content The ginsenoside profiles of the ginseng oils extracted from RGM (particle size: <250 μm, MC: 4.68%) using n-hexane, ethanol, and DME are summarized in Table S3. The HPLC traces of the ginsenosides from the various extracts are shown in Fig. S3A, and potential transformation pathways of ginsenosides are shown in Fig. S4. The process conditions clearly affect the amount of extracted ginsenosides using DME, as illustrated in Fig. S5A–C. The ginsenoside contents of the DME extracts ranged between 1.2 and 5.1 mg g−1 dry RGM within the experimental conditions used (temperature range: 303–333 K, pressure range: 3–7 MPa, and extraction time: 20–80 min). The process parameters and ginsenoside yield were found to be related according to Eq. (2):(2)Ginsenosideyield=307.22−1.88T−141.56P−1.80t+0.089TP+0.01Tt−0.05Pt+0.09P2with R2 and p values of 0.96 and 0.03, respectively, indicative of a strong and statistically significant fit. The ginsenoside content was observed to increase with increasing temperature but decrease with increasing pressure and extraction time under the specified conditions. The highest ginsenoside content of 5.05 mg g−1 dry RGM was obtained at 333 K, 3 MPa, and 20 min (Run 7), with Re, Rf, Rh1, Rb1, Rc, Rd, Rg3, Rk1, and Rg5 as the extracted ginsenosides. A temperature of 333 K was previously identified as optimal for maximizing the ginsenoside yield, as excessively high temperatures may lead to thermal saponin degradation [22]. The high-temperature/low-pressure combination affords low-density DME [23], which enhances mass transfer within the reactor and promotes more-efficient ginsenoside extraction from the RGM matrix. Interestingly, prolonged extraction resulted in lower ginsenoside contents, which suggests that a short extraction time (20 min) at high temperature and low pressure is sufficient for extracting optimal amounts of ginsenosides with DME, highlighting the efficiency of DME as a solvent for the rapid and selective extraction of bioactive compounds under mild conditions. According to the ANOVA results in Table S5, temperature exhibits the strongest statistical effect on ginsenoside extraction (p = 0.01), followed by pressure and extraction time. This finding is consistent with the thermodynamic and kinetic principles governing solid–liquid extraction. Elevating temperature enhances the solubility of ginsenosides in DME by increasing the molecular mobility of solutes, while simultaneously reducing solvent viscosity and enhancing diffusivity. These factors facilitate solvent penetration into the biomass matrix and accelerate mass transfer of ginsenosides from intracellular compartments to the solvent phase. Moreover, higher temperature can disrupt weak hydrogen bonds and van der Waals interactions within the RGM matrix, loosening the structure and increasing extraction efficiency. In contrast, pressure showed a comparatively minor effect, likely due to DME's moderate compressibility and already favorable solvent density (650–659 kg m3) in the tested range (3–7 MPa). The amounts of ginsenosides extracted using n-hexane, ethanol, and DME are presented in Fig. 2A. Soxhlet extraction with ethanol yielded the highest ginsenoside content of 40.57 mg g−1 dry RGM and a broad range of compounds (i.e., Rg1, Re, Rf, Rg2, Rh1, Rb1, Rc, Rb2, Rd, Rg3, Rk1, and Rg5). In contrast, Soxhlet extraction with n-hexane produced only 2.65 mg g−1 dry RGM of ginsenosides that were limited to Rg3, Rk1, and Rg5. The highest-yielding DME extract (Run 14, 333 K, 7 MPa, 80 min) also contained only Rg3, Rk1, and Rg5, totaling 4.22 mg g−1 dry RGM. This selective profile is likely ascribable to the thermal- and pressure-induced transformations of polar ginsenosides (e.g., Rg1, Re, Rb1, Rc, Rb2, and Rd) into less-polar forms (e.g., Rg3, Rk1, and Rg5) during the extraction process [24,25]; these less-polar ginsenosides are known for their enhanced biological activities that are attributable to high cellular absorption rates [26]. Among these ginsenosides, Rg3, which is the dominant compound in the DME extract, is recognized for its strong antioxidant, anticancer, and neuroprotective properties, as well as its role in improving liver function and reducing blood glucose levels [27,28].Fig. 2(A) Ginsenoside profile, (B) antioxidant activity, (C) TPC, (D) chemical profiles and (E) FAME content of hexane, ethanol and DME extracts. (A) Ginsenoside profile, (B) antioxidant activity, (C) TPC, (D) chemical profiles and (E) FAME content of hexane, ethanol and DME extracts. The ginsenoside-extraction efficiency from RGM is strongly influenced by the polarity of the solvent. Hexane, ethanol, and DME have Hansen polar solubility parameter (δp) values of 0 MPa0.5, 8.8 MPa0.5, and 6.1 MPa0.5, respectively [29]. Given that ginsenosides possess medium-to-high polarities, they are more soluble in polar solvents, such as ethanol, and are largely incompatible with nonpolar solvents, such as n-hexane [30], which explains the substantially higher ginsenoside extraction yields observed using ethanol over hexane or DME. Antioxidant activities of the RGM extracts We examined how the process conditions affect the antioxidant activities of the RGM extracts, the results of which are shown in Fig. S5D–F. Antioxidant activities were observed to range between 0.12 and 0.30 mmol TE g−1 dry RGM under the examined conditions (temperature range: 303–333 K, pressure range: 3–7 MPa, and extraction time: 20–80 min). The process parameters and antioxidant activity were found to be related according to Eq. (3):(3)Antioxidantactivity=15.48−0.01T−0.71P−0.29t+0.01Pt+0.12P2with R2 and p values of 0.97 and 0.01, respectively, indicative of a strong and statistically significant model. DME extracts that show high antioxidant activities are generally obtained at high temperatures, high pressures, and long extraction times. In the case of antioxidant activity, ANOVA results in Table S6 again indicate temperature as the most influential factor (p = 0.01), followed by pressure (p = 0.03) and extraction time (p = 0.04). This hierarchy suggests that temperature-induced changes in solubility and kinetics dominate the release of antioxidant compounds (e.g., polyphenols, ginsenosides) from the RGM matrix. Higher temperatures likely promote faster desorption and diffusion of phenolic antioxidants, while potentially enhancing the extraction of thermally generated antioxidant compounds such as Maillard reaction products or ginsenoside transformation products (e.g., Rg3, Rk1, Rg5). However, excessively high temperatures may also risk degradation, which highlights the need for balance. The maximum antioxidant activity of 0.30 mmol TE g−1 dry RGM was obtained at 333 K, 7 MPa, and 80 min. The antioxidant activities per extract are listed in Table S8. Fig. 2B displays the antioxidant activities of the extracts obtained using hexane, ethanol, and DME. The DME extract with the highest yield (extracted at 333 K, 7 MPa, and 80 min) exhibited an antioxidant activity of 0.30 mmol TE g−1 dry RGM; in comparison, the n-hexane and ethanol extracts showed lower antioxidant activities of 0.14 and 1.36 mmol TE g−1 dry RGM, respectively. Ethanol, as a more polar solvent than n-hexane and DME, is capable of extracting more polyphenolic compounds, which are primarily responsible for the observed antioxidant activity. Consequently, the ethanol extracts exhibit higher antioxidant capacities than those extracted using n-hexane and DME. TPCs of the RGM extracts Fig. S5G–I show how the process conditions affect the TPC of the RGM DME extract. DME extraction yielded TPC values in the 0.61–1.09 mg g−1 dry RGM range under the experimental conditions used (temperature: 303–333 K, pressure: 3–7 MPa, and extraction time: 20–80 min). The DME process parameters and TPC are strongly related according to Eq. (4):(4)TPC=0.77+0.08T+0.01P+0.07t+0.03TP+0.03Tt+0.03Pt−0.07T2+0.02P2+0.06t2+0.06TPtwith R2 and p values of 0.91 and 0.02, respectively, indicative of statistical significance. Increasing the temperature, pressure, and extraction time improved the solubilities of phenolic compounds in DME under the studied conditions (303–333 K, 3–7 MPa, and 20–80 min). As listed in Table S7, temperature once again had the most statistically significant impact on TPC (p = 0.00), followed by extraction time (p = 0.01). Phenolic compounds typically exhibit increased solubility and diffusivity with rising temperature, enhancing their release from the plant cell walls into the solvent. Meanwhile, longer extraction time allows for more complete diffusion and equilibration, though with diminishing returns. Pressure showed minimal influence on TPC, likely due to DME's already sufficient solvating power at moderate pressures. Higher temperatures are known to increase the solubilities and diffusion rates of phenolics [31], whereas elevated pressures improve DME density [23], thereby facilitating deeper solvent penetration into the RGM matrix, which promotes phenolic-compound extraction as a consequence. Fig. 2C compares the TPCs of the hexane, ethanol, and DME RGM extracts. Soxhlet extraction with n-hexane or ethanol afforded 0.92 or 3.50 mg GAE g−1 dry RGM, respectively; in contrast, DME extraction under optimal conditions (333 K, 7 MPa, 80 min) delivered 1.09 mg GAE g−1 dry RGM, which is higher than that achieved with n-hexane but lower than that observed using ethanol. These results reveal that the polarity of the extraction solvent plays a critical role in determining the phenolic-compound recovery efficiency from RGM. More polar solvents, such as ethanol, extract phenolic compounds more effectively because they are polar in nature. Chemical profiles of the RGM extracts The chemical profiles of the hexane, ethanol, and DME extracts of RGM (particle size: <250 μm, moisture content: 4.68%) were analyzed by GC–TOF/MS, the results of which are summarized in Fig. 2D and Table S9–S11, with the corresponding GC–TOF/MS chromatogram of each extract shown in Fig. S6–S8. The identified compounds were classified according to functional group, namely acids, aldehydes, alcohols, alkanes, aromatics, terpenes, alkenes, esters, alkynes, amides, terpene alcohol oxides, sterols, and alkanones. The n-hexane extract, which was obtained via Soxhlet extraction, is dominated by alcohols, which account for 37.8% of the total content and include spathulenol (23.9%), sitosterol (20.8%), and trans-3-undecene-1,5-diyne (14.2%) as major components. In contrast, the ethanol extract mostly contains amides, which represent 27.0% of the total extract, with 9-octadecenamide (23.8%), trans-3-undecene-1,5-diyne (11.2%), and spathulenol (11.2%) as key components. The highest-yielding DME extract (Run 14: 333 K, 7 MPa, 80 min) is also mostly dominated by alcohols that comprise 31.8% of the total extract, with sitosterol (17.8%), spathulenol (15.6%), and trans-3-undecene-1,5-diyne (12.0%) as principal components. Notably, the ethanol extract was found to contain fewer alcohols, terpenes, and sterols, but more esters and amides than the hexane and DME extracts. Additionally, it uniquely contained fructose derivatives, including 1,3,4-tri-O-acetyl-2,6-anhydro-α-d-fructofuranose, which were absent in both the n-hexane and DME extracts. Oleamide, an amide derivative of oleic acid, is recognized for its remarkable bioactivities even at low concentrations, including the ability to regulate sleep as well as its neuropharmacological, antiallergic, and hypolipidemic properties [[32], [33], [34]]. Although oleamide was detected in all extracts, it was most abundant in the ethanol extract, at 27.1% of the total composition. In contrast, only small amounts were detected in the n-hexane (1.4%) and DME (3.6%) extracts. Panaxydol and falcarinol (also known as panaxynol) are two important polyacetylenes found in ginseng oil that are endowed with health-promoting properties [35]. Both compounds were identified in the DME and n-hexane extracts, whereas only falcarinol was detected in trace amounts in the ethanol extract, consistent with their lipophilic nature; consequently, they are more soluble in nonpolar or weakly polar solvents such as DME and n-hexane, whereas they are less effectively extracted using ethanol, which is more polar. The DME extract contained the highest concentrations of both compounds (2.0% panaxydol and 3.4% falcarinol); in comparison, the hexane extract contained 0.6% panaxydol and 2.0% falcarinol, whereas the ethanol extract contained only 0.7% falcarinol, with no panaxydol detected. These results further highlight the suitability of DME for extracting valuable bioactive lipophilic compounds from RGM. Terpenes, which form a class of secondary metabolites that are involved in the defense mechanisms of plants, are known for their distinct flavors and fragrances, as well as their broad spectrum of bioactivities that include antimicrobial, anti-inflammatory, and anticancer effects, among other medicinal benefits [36]. The n-hexane, ethanol, and DME RGM extracts exhibited terpene contents of 7.0%, 2.01%, and 7.4%, respectively, with the DME extract containing the most diverse group of terpenes, which include limonene, bicyclogermacrene, elemene, clarene, aristolene, panansinsen, humulene, neoclovene, alloaromadendrene, gurjunene, selinene, germacrene B, cadinene, aromadendrene, neophytadiene, and squalene. Squalene is a triterpene that is naturally synthesized in the human body and found in both animals and plants. Squalene is recognized for its immunoboosting, anti-inflammatory, antioxidant, and skin-protective properties; it is efficiently absorbed through oral and topical administration and is widely used in pharmaceutical and cosmetic applications [37,38]. The n-hexane, ethanol, and DME extracts exhibited squalene contents of 1.3%, 0.7%, and 0.8%, respectively. Given that terpenes are nonpolar compounds [39], solvents such as hexane and DME, which are nonpolar and moderately polar, respectively, extract terpenes more effectively than ethanol, a polar solvent, which rationalizes the significantly lower terpene content of the ethanol extract. FAME analysis Fatty acids in the extracts were analyzed in the form of fatty acid methyl esters (FAMEs) after in situ acid-catalyzed transesterification, following the method described by Eder [36]. This derivatization step improves the volatility and detectability of fatty acids for accurate GC-FID analysis. The FAME analysis results for the n-hexane, ethanol, and DME RGM extracts are presented in Fig. 2E. Fatty acids are classified as saturated fatty acids (SFAs), monounsaturated fatty acids (MUFAs), and polyunsaturated fatty acids (PUFAs). Soxhlet extraction with hexane and ethanol led to total FAME contents of 12.06 and 5.07 mg g−1 RGM, respectively. In comparison, DME extraction at 333 K, 7 MPa, and 80 min yielded a total FAME content of 14.93 mg g−1 RGM that surpasses that of each Soxhlet-based method. The hexane extract was found to contain a diverse range of fatty acids, including capric acid (C10), palmitic acid (C16:0), elaidic acid (C18:1 trans), linoleic acid (C18:2 cis), linolenic acid (C18:3n3), methyl cis-11,14-eicosatrienoate (C20:2), erucic acid (C22:1), methyl cis-13,16-docosadienoate (C22:2), tricosanoic acid (C23:0), and lignoceric acid (C24:0). In contrast, the ethanol extract contained only unsaturated fatty acids, specifically linolelaidic acid (C18:2 trans) and docosahexaenoic acid (DHA; C22:1n6). On the other hand, the DME extract was found to contain a compositionally rich mixture of saturated and unsaturated fatty acids, including capric acid (C10), palmitic acid (C16:0), elaidic acid (C18:1 trans), linolelaidic acid (C18:2 trans), linoleic acid (C18:2 cis), linolenic acid (C18:3n3), and erucic acid (C22:1). The n-hexane and DME extracts contained similar amounts of saturated fatty acids (1.61 and 1.60 mg g−1 dry RGM, respectively). However, DME extracted significantly more unsaturated fatty acids, with MUFA and PUFA contents of 1.05 and 12.27 mg g−1 dry RGM, respectively, with a total yield of 13.32 mg g−1 dry RGM. In comparison, the n-hexane extract contained 0.85 mg g−1 MUFAs and 9.60 mg g−1 PUFAs, with a total unsaturated fatty acid content of 10.45 mg g−1 dry RGM. Linoleic acid (C18:2 cis), the major constituent in both the n-hexane and DME extracts, was present at 72.86% (8.79 mg g−1 dry RGM) and 76.09% (11.36 mg g−1 dry RGM), respectively. Linoleic acid is an essential fatty acid that is vital for skin health; it is a key component of intercellular lipids in the stratum corneum and is widely used in moisturizing formulations owing to its ability to restore the skin barrier, promote wound healing, lighten pigmentation, reduce the signs of aging, and stimulate hair growth [40,41]. Fatty acid polarity decreases with increasing chain length [42]. As a relatively polar solvent, ethanol ineffectively extracts saturated fatty acids from RGM. In contrast, DME exhibited superior extraction performance compared to n-hexane or ethanol owing to its unique ability to solubilize both polar and non-polar fatty acids. This dual-solubility characteristic enables DME to extract a broader range of functional fatty acids, resulting in a higher total fatty acid content in the extract. Extracting wet RGM with DME: A comparison with dry RGM The ability of DME to extract ginseng oil from wet RGM was also evaluated. Here, both fresh and wet RGM (MC: 86.63%) and dry RGM (MC: 0.30%) were used as feedstocks instead of ground RGM, with extraction conducted under the previously identified optimal conditions (333 K, 7 MPa, 80 min). Extracting wet RGM with DME afforded an extraction yield of 3.6 wt% dry basis along with a dewatering rate of 87.4 wt%. In contrast, extracting dry RGM with DME led to a lower yield of 1.5 wt% dry basis. In comparison, dry RGM was also Soxhlet-extracted using ethanol and n-hexane, which afforded yields of 3.5 and 0.9 wt%, respectively. These results reveal that the water present in the RGM enhances the extraction yield, which is likely ascribable to the higher polarity of the extraction environment, which facilitates solubilization of more polar components. The ginsenoside contents of the DME extracts of the dried and wet RGM are presented in Fig. 3A and Table S12, with the corresponding HPLC chromatograms shown in Fig. S3B. The dried-RGM DME extract was found to contain Rc, Rb2, Rg3, Rk1, and Rg5, totaling 1.32 mg g−1 dry RGM. In contrast, the wet-RGM DME extract afforded a broader profile that included Rf, Rc, Rb2, Rg3, Rk1, and Rg5, along with a significantly higher total ginsenoside content of 12.94 mg g−1 dry RGM. The enhanced extractabilities of the polar ginsenosides from the wet RGM are attributable to the higher polarity of the system associated with its water content. Notably, Rf, a polar ginsenoside, was detected only in the DME extract of wet RGM, while the concentrations of Rc, Rb2, Rg3, Rk1, and Rg5 were markedly higher than those observed for the dried sample. These findings highlight the beneficial role that moisture plays in promoting the extraction of polar bioactive compounds using DME.Fig. 3Comparing the DME extracts of dried and wet RGM: (A) ginsenoside content, (B) chemical profile based on functional-group composition, and (C) FAME content. Comparing the DME extracts of dried and wet RGM: (A) ginsenoside content, (B) chemical profile based on functional-group composition, and (C) FAME content. The chemical profiles of the DME extracts from the dried and wet RGM samples are presented in Fig. 3B, with the corresponding GC–TOF/MS chromatograms shown in Fig. S9 and S10, respectively, and detailed chemical compositions were listed in Table S13 and S14, respectively. Sterols (35.4%) and alcohols (22.8%) are the dominant chemical components of the dried RGM extract, whereas the wet RGM extract contained higher proportions of alkanes (24.2%) and sterols (19.4%). Notably, the wet RGM extract contained significantly more alkanes, slightly more amides, and markedly lower levels of sterols and alcohols than the dried RGM extract. Sitosterol was found to be the dominant compound in both extracts, accounting for 27.3% and 17.0% of the dried and wet RGM extracts, respectively. In addition, the dried extract contained much more falcarinol (18.8%) than the wet extract (5.2%), which indicates that lipophilic compounds are better preserved in low-polarity extraction environments. The FAME profiles of the DME extracts of the dried and wet RGM are shown in Fig. 3C. DME extraction of the dried RGM afforded a FAME content of 8.16 mg g−1 dry RGM, with linoleic acid accounting for 72.9% of the total. In contrast, wet RGM extraction yielded 6.31 mg g−1 dry RGM of FAMEs, with a higher proportion of linoleic acid recorded (83.0%). The dried RGM extract was found to contain a broader range of fatty acids, including capric acid (C10), palmitic acid (C16:0), stearic acid (C18:0), elaidic acid (C18:1 trans), oleic acid (C18:1 cis), linolelaidic acid (C18:2 trans), linoleic acid (C18:2 cis), linolenic acid (C18:3n3), eicosadienoic acid (C20:2), docosadienoic acid (C22:2), lignoceric acid (C24:0), and docosahexaenoic acid (DHA, C22:1n6). In comparison, the wet RGM extract contained fewer fatty acids, namely palmitic acid (C16:0), elaidic acid (C18:1 trans), oleic acid (C18:1 cis), linoleic acid (C18:2 cis), and linolenic acid (C18:3n3). The dried RGM extract contained more SFAs (13.6%) than the wet RGM extract (7.2%); this difference reflects the lower polarity of the dried RGM extraction environment, which favors the extraction of saturated and nonpolar fatty acids, whereas the water present in the wet RGM increases the overall polarity of the system, which favors the extraction of more unsaturated and polar lipids. Comparing extraction methods Table 2 compares the various ginseng-oil extraction methods, highlighting the differences in feedstock, operating conditions, solvent systems, and extract compositions. Among the evaluated techniques, DME extraction stands out for its moderate operating temperature (333 K) and low pressure (7 MPa), requiring only 80 min to deliver a ginsenoside content of 4.22 mg g−1 dry RGM and a TPC of 32.38 mg dry GAE g−1 extract. Despite affording a somewhat lower oil content (3.4%) than ultrasonic or scCO2 extraction (16.9% or 10.7%, respectively), DME offers distinct advantages in terms of solvent-removal simplicity and energy efficiency, as it requires no post-processing steps, such as centrifugation or freeze-drying. While extracting ginseng-root hairs with ultrasound or scCO2 led to higher ginsenoside concentrations (729 or 153 mg kg−1 absorbent, respectively), these methods involve longer processing times (150 min or 4 h, respectively) and higher pressures, especially in the case of scCO2 (31.2 MPa). Moreover, while extraction with subcritical water yielded the highest TPC (114.9 mg GAE g−1 extract) and ginsenoside content (14.4 mg g−1 RGM), it operates at a high temperature (473 K), which poses potential the risk of partial degradation or isomerization of ginsenosides and oxidation of fatty acids at elevated temperatures, particularly in extended extractions or when trace oxygen is present and consumes large amounts of energy. Green metrics were calculated to quantify the energy consumption of DME extraction in comparison with ultrasonic extraction, scCO2 extraction, and subcritical water extraction. The estimated energy consumptions were 7, 15, 90, and 19,221 Wh g−1 dry RGM for DME, ultrasonic, scCO2, and subcritical water extraction, respectively. These values were roughly estimated based on the energy requirements of the solvents and equipment involved, and further detailed analysis is needed for more precise predictions. Due to the mild processing conditions and the absence of an additional solvent removal step, DME extraction exhibited significantly lower energy consumption compared to the other methods. Overall, DME extraction demonstrates a favorable balance between selectivity, process simplicity, and functional-compound recovery, which is particularly advantageous when extracting lipid-soluble compounds from RGM, thereby providing an efficient and scalable alternative to conventional solvent- or pressure-intensive techniques.Table 2Comparing ginseng-oil extraction methods.Table 2Extraction methodFeedstockSolventT (K)P (MPa)tEnergy consumed (Wh g−1 feedstock)1)Yield (%)GinsenosideTPC (mg GAE g−1 extract)Ref.DME extractionRGMDME333780 min73.44.22 mg g−1 RGM32.38This workUltrasonic extraction followed by centrifugationGinseng root hair (<105 μm)H2O/EtOH (54/46)3337.6150 min1516.9729 mg kg−1 absorbentN.A.[6]Supercritical fluid extraction followed by vacuum dryingGinseng root hair (<105 μm)CO2+6% EtOH33331.24 h9010.7153 mg kg−1 absorbentN.A.[6]Reflux extraction followed by fractionationRGM powderMethanol, ethyl acetate333N.A.2 hN.A.N.A.N.A.3.79[9]Subcritical water extraction followed by the use of vacuum pump and convection ovenRGMWater4731015 min192214814.4 mg g−1 RGM114.9[10]T: temperature; P: pressure; t: extraction time; DME: dimethyl ether; RGM: red ginseng marc; N.A.: not available.1) Energy consumption was estimated based on the energy required to heat the solvent—calculated as the product of solvent mass, specific heat capacity, and temperature change—as well as the rated power consumption of the equipment used in the study. Table 2Comparing ginseng-oil extraction methods.Table 2Extraction methodFeedstockSolventT (K)P (MPa)tEnergy consumed (Wh g−1 feedstock)1)Yield (%)GinsenosideTPC (mg GAE g−1 extract)Ref.DME extractionRGMDME333780 min73.44.22 mg g−1 RGM32.38This workUltrasonic extraction followed by centrifugationGinseng root hair (<105 μm)H2O/EtOH (54/46)3337.6150 min1516.9729 mg kg−1 absorbentN.A.[6]Supercritical fluid extraction followed by vacuum dryingGinseng root hair (<105 μm)CO2+6% EtOH33331.24 h9010.7153 mg kg−1 absorbentN.A.[6]Reflux extraction followed by fractionationRGM powderMethanol, ethyl acetate333N.A.2 hN.A.N.A.N.A.3.79[9]Subcritical water extraction followed by the use of vacuum pump and convection ovenRGMWater4731015 min192214814.4 mg g−1 RGM114.9[10]T: temperature; P: pressure; t: extraction time; DME: dimethyl ether; RGM: red ginseng marc; N.A.: not available.1) Energy consumption was estimated based on the energy required to heat the solvent—calculated as the product of solvent mass, specific heat capacity, and temperature change—as well as the rated power consumption of the equipment used in the study. Comparing ginseng-oil extraction methods. T: temperature; P: pressure; t: extraction time; DME: dimethyl ether; RGM: red ginseng marc; N.A.: not available.1) Energy consumption was estimated based on the energy required to heat the solvent—calculated as the product of solvent mass, specific heat capacity, and temperature change—as well as the rated power consumption of the equipment used in the study. Conclusions Ginseng oil was successfully extracted from RGM using DME, with the highest extraction yield (3.37%) obtained at 333 K, 7 MPa, and an extraction time of 80 min (Run 14). These process conditions resulted in an extract that contained a total ginsenoside content of 4.22 mg g−1 dry RGM and a TPC of 1.09 mg GAE g−1 dry RGM. Moreover, the DME extract (Run 14) exhibited an antioxidant activity of 0.30 mmol TE g−1 dry RGM and a FAME content of 14.93 mg g−1 dry RGM, which exceed those of the corresponding hexane or ethanol extract. Linoleic acid is the main fatty acid in the ginseng oil extracted using DME. DME can also be used to extract ginseng oil directly from wet RGM, which reduces the cost and energy associated with drying the RGM. Extracting wet RGM with DME resulted in a higher extraction yield compared to that obtained from dry RGM. The water present in the RGM increases the polarity of the extraction environment, thereby increasing the extraction yield and the ginsenoside content in the extract while decreasing the amount of FAME extracted from the RGM. Accordingly, DME is a promising green solvent for extracting ginseng oil from RGM, even in a wet state. Conflict of interest statement Manuscript title: Valorizing Ginseng Residues using Dimethyl Ether: Recovering Functional Lipids and Ginsenosides. The authors whose names are listed immediately below certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or nonfinancial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript. Author names. Sabrinna Wulandari, Ruqian Cao, Jongho Choi, Aye Aye Myint, Jaehoon Kim. The authors whose names are listed immediately below report the following details of affiliation or involvement in an organization or entity with a financial or nonfinancial interest in the subject matter or materials discussed in this manuscript. Please specify the nature of the conflict on a separate sheet of paper if the space below is inadequate.