Cannabis sativa, Cannabis indica, and Cannabis ruderalis contain diverse chemical compounds termed cannabinoids. The medical and cosmetic applications of cannabis have changed in the last 3 years in the United States, South Korea, and Thailand [1,2]. Historical, medical, and scientific aspects of cannabis have been reviewed [3,4]. Additionally, extraction methods for making hemp seed extracts have been examined.

Currently, South Korean law permits the use of cannabis seeds without husks, stems, or roots. Traditional medical literature has been consulted regarding hemp seed extraction methods to investigate the presence of tetrahydrocannabinol (THC), the psychoactive component that induces hallucinations, as well as cannabinoids convertible to THC. The national literature on the cosmetic and skincare aspects of cannabis needs to be evaluated for future applications [5-7]).

In July 2020, Andong, South Korea was designated as a special zone for industrial and domestic cultivation of hemp products. Inadequate good manufacturing practices (GMP) facilities and foreign trade regulations have hindered compliance while unclear extracts or extracts derived from unknown seeds violate South Korean laws. Domestic research on cannabidiol (CBD) in cosmetics has raised concerns about cannabinoid content and cultivation standards [8] undermining clinical trust in South Korean hemp cosmetics. Medical trials are warranted to support compliance with 1% CBD content under South Korean law. Disputes among domestic and imported hemp suppliers may have political motives [9].

Recently, the South Korean Ministry of Food and Drug Safety included cannabis regulation relaxation in its ‘Top 100 Regulatory Innovations’ list. However, global debates over cannabis legalization persist. Hong Kong has tightened regulations on cannabinoids [10]. Moreover, Japan has enhanced the use of cannabinoids in health supplements but has enforced strict bans on compounds such as THC [11]. In Korea, debates focus on supporting patient rights and expanding general use, prioritizing safety and competitiveness.

This study explores the use of hemp seed oil in the domestic cosmetics industry. Many cannabis cosmetics manufacturers have emphasized its benefits while mitigating concerns about side effects [12]. Several premises underlie this approach. THC is extracted from cannabis leaves and flowers and is not used in cosmetics. Industrial hemp used for cosmetics has a low THC content. Cannabis seeds, roots, and stems contain a minimal amount of THC. Cannabinoid-based cosmetics manufacturers fail to mention the absence of THC in their product descriptions. Overseas products containing CBD risk noncompliance if imported without analysis.

The growing global interest in cannabis-derived products requires that these products be safe and comply with local and international regulations. By analyzing the cannabinoid content of hemp seed oil and understanding the implications of different extraction methods, this study aims to provide a scientific basis for the safe use of these products in cosmetics. We also aimed to connect traditional practices with current regulatory standards.

The requirement to obtain informed consent was waived.

Hemp seed oil preparation

To replicate traditional processing conditions, extraction was carried out using pure deionized (DI) water or oils to maintain the solution temperature below 80°C for 30 minutes (Fig. 1). The press plate was kept at 120°C to maintain the solution temperature at approximately 80°C while preparing cannabis seeds with more than 90% of their husks removed, while another batch had hemp seeds with unpeeled husk (Fig. 1). The ratio of cannabis seeds to water was kept at 1:4 and precisely measured using PK-16. The solutions used included DI water (Merck Millipore) and four types of oils: canola, palm, soybean, and Canadian cannabis oil. All the oils were sourced from a NongHyup Hanaro Mart store (a supermarket store owned by agricultural farmers) located in Andong city. The oil extraction process was carried out using specialized equipment designed by Biomedical 3D Printing Inc. that is currently covered by a pending patent. Each type of oil was added to the cannabis seeds at a ratio of 30% to 40%, based on the weight of the cannabis seeds, and then used for extraction. The mixing order and the ratio of DI water to oil are crucial for maintaining the consistency and effectiveness of the extraction process. The resulting oil was collected, and its quantity was accurately measured, ensuring precise measurement of both the input and collected amounts. Fig. 1 shows a schematic of the experiment, which utilized equipment based on traditional literature sources such as the Chinese medical document ‘Shengji Zonglu.’ Each separated component was diluted with PK0 at various ratios and then analyzed using liquid chromatography-mass spectrometry (LC-MS/MS) and high-performance liquid chromatography (HPLC) [13]. The PK-16 analysis protocol used was compared to the PK16 Reference Solution library for analysis. The analysis of samples diluted with PK0 was conducted following the protocols outlined in the South Korean Patent (Registered Patent: KP 102694442) and the Japanese Patent (C B D 3 D), which provide comprehensive details. The PK0 solution used in this analysis was supplied free of charge by Biomedical 3D Printing Inc. under another patent (Registered Patent: KP102355482 titled Cannabis Extractive Separation Method and Continuous Measurement Apparatus).

Figure 1. A schematic of the process for producing hemp seed oils, followed by extraction, separation, and analysis using the PK16 (Perkin Elmer Protocol) method, as outlined in the four patents provided. DI, deionized; HPLC-PDA, high-performance liquid chromatography with photodiode array detection; THC, tetrahydrocannabinol; CBD, cannabidiol.

Analytical methods

To verify the legal regulations of cannabis, quantitative and qualitative analyses of 16 cannabinoids were conducted using the PK-16 method, which satisfies ISO 19025 standards. Sixteen cannabinoid reference standards were obtained from Cayman Chemical, to calibrate the Perkin Elmer LC 300 HPLC with photodiode array detection (HPLC-PDA) instrument, which was equipped with three different C18 columns (150×3.0 mm, 2.6, 3.0, and 5.0 μm), using Simplicity Chrom software. These standards were provided in individual amber glass ampules, each containing a 1 mg/ml concentration of cannabichromene (CBC), cannabichromenic acid (CBCA), CBD, cannabigerol (CBG), cannabinol (CBN), cannabidiolic acid (CBDA), cannabigerolic acid (CBGA), cannabinolic acid (CBNA), cannabicyclol (CBL), cannabidivarin (CBDV), cannabidivarinic acid (CBDVA), Δ8-THC, Δ9-THC, Δ9-tetrahydrocannabinolic acid (THCA-A), tetrahydrocannabivarin (THCV), and tetrahydrocannabivarinic acid (THCVA). Upon opening the glass ampules, the reference standards were swiftly transferred into amber glass vials with airtight caps and refrigerated at -4°C. HPLC-grade solvents, including methanol (MeOH), acetonitrile (ACN), and DI water (H₂O) from Supelco Inc., along with PK0 diluent from Biomedical 3D Printing Inc., were utilized for both the mobile phase and dilution of the samples. The preparation of the mobile phases followed the PK-16 protocol [14]. Mobile phase A comprised 1 ml of formic acid and 800 μl of 10 M ammonium formate in 1 L of PK0 diluent, while mobile phase B consisted of 900 μl of formic acid in 900 ml of ACN supplemented with 100 ml of mobile phase A. An isocratic process utilizing 21.1% mobile phase A, and 78.9% mobile phase B was employed for all analysis procedures as per the protocols of Perkin Elmer and PK0. Notably, in the PK0 calibration process, H₂O was used instead of the PK diluent while retaining the entirety of the calibration procedure. The primary distinction between PK0 and PK-16 lies in the pretreatment process and the application of a PK dilution solution optimized for HPLC equipment and tailored for the analysis of the 16 types of cannabinoids [3]. The 16 component peaks using the PK-16 protocol are shown in Fig. 2.

Figure 2. Sixteen component peaks of high-performance liquid chromatography 300 were detected using the PK-16 protocol (50 mg/L): CBC: ×, CBCA: -, CBD: ◇, CBG: ±, CBN: =, CBDA: ◆, CBGA: +, CBNA: *, CBL: #, CBDV: ◁, CBDVA: ◀, Delta8-THC: ◑, Delta9-THC: ●, THCA: ○, THCV: ■, and THCVA: □ [14]. CBC, cannabichromene; CBCA, cannabichromenic acid; CBD, cannabidiol; CBG, cannabigerol; CBN, cannabinol; CBDA, cannabidiolic acid; CBGA, cannabigerolic acid; CBNA, cannabinolic acid; CBL, cannabicyclol; CBDV, cannabidivarin; CBDVA, cannabidivarinic acid; THC, tetrahydrocannabinol; THCA, tetrahydrocannabinolic; THCV, tetrahydrocannabivarin; THCVA, tetrahydrocannabivarinic acid.

Commercially available cannabis cosmetics, imported cannabis seed oil, and domestically processed cannabis seeds with more than 90% of the husk removed (Fig. 1A) were subjected to simple heating at 80°C or higher, as mentioned in traditional medical texts, for a minimum of 2 hours, followed by the analysis and comparison of their internal components using HPLC-PDA. It was confirmed that all seven types of cannabis cosmetics and domestically distributed cannabis seed oils, not mentioned in this paper, are safe when extracted or used in cosmetics in the manner described in traditional literature. Further, THC was not detected or confirmed to be safe in either domestically produced products (Fig. 3A) or imported cannabis seed oil (Fig. 3B). Hemp seeds without husks (Fig. 3C) contained 3–5 mg/kg THC, which is within the safe range according to current Korean standards. Therefore, current advertisement descriptions that include CBD may weaken the use of cannabis cosmetics and dermatology by overemphasizing components that are absent. Basic experiments conducted using traditional methods revealed the presence of sufficient amounts of natural compounds and hydrophilic components in the 2-minute retention time duration compared to Canadian cannabis oil (Fig. 3B). Additionally, when examining the shape of cannabinoid peaks, it was found that the appropriate extraction method for domestically produced husk-removed seeds (Fig. 3C) contained more competing unsaturated fatty acid components in all areas compared to Canadian cannabis oil, and the safety of cannabinoids was confirmed.

Figure 3. High-performance liquid chromatography with photodiode array detection 220 nm analysis graphs depicting domestically distributed hemp seed products: (A) cosmetic oils obtained from processed domestic hemp seeds, (B) hemp seed oil imported from Canada, and (C) oil extracted from husk-removed hemp seeds purchased from Pungsan, Gyeongsangbuk-do. All samples were subjected to 2 hours of heating at 80°C in deionized water according to traditional methods in the literature.

One of the innovative approaches in this study is the emulation of past literature’s depiction of cannabis seeds at a level appropriate to their historical context. Presently, removing the husk from cannabis seeds originates from the traditional yet irrational practice of losing more than one-third of the core part of the seed through processes such as hulling, resulting in inefficiency, drying, and cannabinoid conversion due to mechanical friction. If domestically processed hemp seeds are properly treated, they can be processed into high-quality cosmetics or medicinal materials. Therefore, we question the necessity of husk removal.

Further, securing technology to selectively separate the content of natural substances in the 2-minute retention time duration is important. One legal approach involves kinetics. Instead of exposing cannabinoids for an extended duration, which may affect the useful components for cosmetics, a kinetics-based method was successfully applied to selectively remove cannabinoids within a short duration, as described in a patent.

Therefore, ancient texts, especially “Dongui Bogam,” containing claims that extracting cannabis seeds may lead to hallucinations, provide evidence from a current perspective that unpeeled cannabis seeds were used traditionally [15]. This raises reasonable doubts about whether the extract obtained through heavy stone pressing with hot water, once the cannabis seed husk had been removed, actually contained cannabinoids. To ensure the suitability of using hemp seed extracts for cosmetics and skincare, it is necessary to understand past practices and technologies. Extracting by using traditional protocols that employed imperfect hulling techniques raises new doubts about the existence and potential risks of cannabinoids when hemp seed husks are retained.

To mimic traditional processing conditions, the solution temperature was maintained below 80°C for 30 minutes using only pure DI water without alcohol, and the temperature of the press plate was maintained at 120°C to ensure >90% removal of the husks from the sample, as shown in Fig. 4A, while Fig. 4B shows that no husk removal occurred in the other sample. The adsorption intensity on the Y-axis was measured at 5,000 mAU. In close alignment with historical practices, we considered the process of pressing with hot stones at 120°C (solution temperature: 80°C) using water available at that time. Our findings confirmed the presence of THC in hemp seeds without husk removal and pressed with hot stones using water, with peaks related to THC at approximately 200 mAU (50 mg/kg) (Fig. 2), indicating the presence of THC among cannabinoids exceeding 300 mg/L. Further experiments probed the temperature, quantity, and safe removal methods of cannabinoids within husks, confirming scientifically that cannabinoids exceed Korea’s regulatory limits.

Figure 4. The high-performance liquid chromatography with photodiode array detection 220 nm analysis graphs depict (A) husk-removed (images in Fig. 1A) and (B) as-received without husk removal (image in Fig. 1B) hemp seeds. The samples were rapidly pressed at 120°C using deionized water to ensure that the solution temperature did not exceed 80°C for no more than 30 minutes according to kinetics theory before being processed.

Prolonged extraction at high temperatures (120°C for 2 hours) enables the removal of both natural unsaturated fatty acids and cannabinoids to significant levels. This is evidenced by residue analysis, demonstrating the effective application of both hydrophilic and lipophilic extraction methods. However, as shown in Fig. 5, this method is not effective for materials devoid of cannabinoids, such as husk-removed cannabis seeds. In the context of skincare cosmetics in Korea, the presence of natural unsaturated fatty acids may be even more crucial than the presence of cannabinoids. In studies extracting both hydrophilic and lipophilic substances according to kinetics, preliminary research utilizing the solubility of oil (Fig. 5A) and ethanol (Fig. 5B) was applied. While lipophilic extraction using oil generally releases cannabinoids slowly, ethanol has been found to rapidly extract cannabinoids, suggesting that extraction using this method may yield optimal results. Traditional literature such as “Seongje Chongrok” mentions the extraction of seeds using oil by roasting, supporting the significance of these research findings for skincare or cosmetic applications [13]. This historical practice lays the foundation for modern extraction methods such as supercritical and solvent extraction, aiding in the separation of cannabinoids for medical and cosmetic purposes.

Figure 5. The high-performance liquid chromatography with photodiode array detection 220 nm analysis graphs depict the following: (A) extraction with 30% oil added to hemp seeds, and (B) extraction with 20% ethanol added to hemp seeds, followed by rapid pressing at 120°C of hemp seeds including liquid components at a ratio of 1:4 deionized water.

A blend of oil and PK0 solution was used to prepare extracts from domestically sourced hemp seeds without husk removal to meet domestic regulations for optimal oil extraction. Seed extracts were prepared by different methods to ensure both the economic viability and safety of the solution, with particular attention given to maximizing the content of natural unsaturated fatty acids for use as cosmetic ingredients. Therefore, hemp seed extracts were prepared to ensure that the total amount of cannabinoids, even when converted to THC equivalents, did not exceed 10 mg/kg, and that the other ingredients were harmoniously manufactured. Fig. 6A illustrates the optimized extraction conditions using palm oil, while Fig. 6B depicts uniform extraction using commercially available hemp oil.

Figure 6. The high-performance liquid chromatography with photodiode array detection 220 nm analysis graphs depict the following: (A) extraction with palm oil and (B) extraction with hemp oil conducted at 50°C of unpeeled hemp.

Table 1 presents the extraction process results when extracting domestic hemp seeds under optimal conditions derived from a patent using oil [16-19]. In terms of cost-effectiveness, the use of canola oil proved to be the most effective, while the in-and-out ratio was highest when hemp seed oil was used as the processing oil, even considering the margin of error. Further, considering the internal safety and clinical efficacy of these four oils, there are advantages and disadvantages to their use in roasting extraction methods, indicating the need for further research based on additional clinical trials.

Table 1 . The ratio of the retained solution to the input amount for canola oil, palm oil, soybean oil, and hemp seed oil used for optimized oiling, along with the documented levels of omega-3, omega-6, and omega-9 content, and their respective prices.

Oil type	ω-3 (mg/g)	ω-6 (mg/g)	ω-9 (mg/g)	US dollars/100 ml	In-to-out ratio (out/in, %)	Standard deviation of out/in (%)
Canola [16]	9.1	18.6	61.8	0.7	24.5	0.7
Palm [17]	0.2	9.1	40.0	0.3	20.5	6.4
Soybean [18]	7.0	51.0	22.6	7.5	20.0	7.1
Hemp seed [19]	22.0	54.0	9.0	13.0	30.5	16.3

Mixing A and B from Fig. 6 at a 1:1 ratio and using them once daily for 2 weeks as a substitute for skincare cosmetics for dry skin did not result in any notable side effects [20]. Fig. 7A shows the images obtained from the oil roasting extraction, Fig. 7B shows the pre-images of the patient with dry skin, and Fig. 7C shows the images after 2 weeks.

Figure 7. Images of (A) hemp seed oil for cosmetics extracted at 50°C with palm oil and hemp oil, (B) the foot of an individual with dry skin, and (C) photographs taken after applying the aforementioned palm oil and hemp extract for two weeks.

While various benefits of cannabinoids from different parts of cannabis are mentioned for domestic medical health foods, cosmetics, and beauty ingredients, no medical evidence or clinical cases are provided in Korea. The strict laws on cannabis in Korea, coupled with limited business, research, and academic activities since the 1970s, have contributed to the undervaluation of domestic cosmetics and skincare products compared to international ones. Systematic results regarding the process of removing cannabis husks according to domestic laws are lacking, and there are no reports on whether the final products meet cannabinoid standards for medical use. Although the author’s research team mentioned extraction techniques and methods based on mathematical models reported in other papers, along with detailed technical reports or papers on cannabinoids from traditional extraction methods and the presence of husks in hemp seeds, regulations in Korea regarding cosmetics or dermatology are still unclear.

Even if there are claims that cannabinoid-containing cosmetics imported into the country enhance the effectiveness of skincare, there is no scientific evidence to support this in the realm of skincare. In Korea, cosmetics containing THC or derivatives of CBD that can be transformed into THC cannot be manufactured, as specified by the “Cosmetics Safety Act Enforcement Decree.” The current situation in the Korean cannabis skincare market, claiming safety due to the absence of cannabinoids, is an ironic scenario requiring scientific analysis and systems to provide consumers with reliable information. Efforts are needed to establish systems ensuring the safety of THC and CBD based on scientific evidence. Furthermore, premature conclusions regarding the safety of CBN, CBG, CBDA, and CBGA, which have not been proven safe for human consumption and could transform into THC, should be cautioned against. Therefore, efforts to challenge the management system of the total amount of cannabinoids, known as the “big six,” which are relatively common in cannabis, are necessary and have been attempted in this study. The Big Six includes CBD, THC, CBN, CBG, CBC, and THCV, which are concentrated and managed intensively due to their relatively high levels of cannabis.

The results of this study highlight the importance of analyzing cannabis constituents to provide stable, legal cosmetic ingredients, ensuring industry stability and development through scientific regulation. This aligns with the study’s objective to underscore the global use of cannabis-derived extracts as active pharmaceutical ingredients in cosmetics. By demonstrating that hemp seed oil can be safely used in skincare products when proper extraction methods and regulatory standards are followed, this study provides a foundation for further research and development in this area. The findings also suggest that Korea could lead in producing safe and effective cannabis-derived cosmetics, provided the regulatory framework continues to evolve and support industry growth.

The study focused on a few specific extraction methods and their impact on cannabinoid content. More research is needed to explore other extraction techniques and their effects on hemp seed oil’s safety and efficacy. Additionally, comparing the regulatory landscape in Korea with international standards would provide a more global perspective.

Future research should expand on different extraction methods and their impact on cannabinoid content and safety. Clinical trials are necessary to confirm the effectiveness of hemp seed oil in skincare. As cannabis product regulations evolve, ongoing research should adapt to these changes and offer updated industry recommendations.

This study aimed to analyze the safety and regulatory compliance of hemp seed oil for cosmetic applications by examining different extraction methods and their impact on cannabinoid content. Key extraction methods, including hot water extraction, low-temperature water extraction with ethanol, cold pressing, and roasting with oil, were assessed. The findings revealed that hemp seeds without husks showed safe levels of CBD (20 mg/L; 0.002%) and THC (10 mg/L; 0.001%). However, seeds with husks exhibited higher THC levels, up to 300 mg/L, depending on the extraction method. This underscores the importance of precise extraction techniques and regulatory oversight to ensure safety. Analysis of commercially available hemp seed oil and stem extracts from various countries revealed a lack of compliance with good agricultural practices and GMP. Establishing effective management systems to limit the total cannabinoid content of cosmetic ingredients to 500 mg/L could enhance regulatory compliance and prevent the distribution of non-compliant products. The hemp seed oil has significant potential for cosmetic use, provided that regulatory standards are strictly followed. The study supports the development of clear regulations and advanced extraction methods to ensure the safe and effective use of hemp seed oil in skincare products.

This study was supported by the Brain Pool Fellowship Program (Grant Number: 2022H1D3A2A0109646) of the Korean Research Foundation. Also, Dr. Namsoo Kim sincerely appreciates the support of Biomedical 3D Printing, Inc. for supplying the PK-16 solution under patent rules.

The authors have nothing to disclose.