2.1 Formation of M_C mixture and complex (IP-MHA, IP-Mandelic acid)

Mandelic acid (R.L. Chemical Industries Pvt. Ltd, India) and L-Carnitine (ACTIVON Korea, South Korea) were prepared for the M_C complex (IP-MHA, IP-Mandelic acid) in a molar ratio of 1:1. Mandelic acid and carnitine were mixed in a 1:1 molar ratio in deionized water to achieve a total concentration of 40%. The mixture was stirred at 750 rpm and 55–60°C for approximately 20 minutes until a homogeneous yellowish solution formed without any precipitates. A simple mixture (M_C mixture) was also prepared by dispersing each substance in deionized water and mixing at room temperature until the final concentration matched that of the M_C complex. The pH was adjusted to approximately 4, and further pH adjustment using potassium hydroxide (KOH, Sigma-Aldrich, USA) was performed to evaluate the exfoliating efficacy under pH 4.5 conditions. No additional purification steps were performed.

2.2 Structure analysis of M_C complex

To verify the formation of the M_C complex, the following evaluations were conducted. First, changes in size distribution and zeta-potential of mandelic acid, carnitine, M_C mixture, and M_C complex were investigated by using photon correlation spectroscopy with a Zetasizer Nano ZS instrument (Malvern Instruments, UK). Scattered light was detected at an angle of 173° using the non-invasive backscattering (NIBS) technique. All solutions were measured at a standard material (mandelic acid) concentration of 0.5% to determine the change in particle size. Zeta-potential measurements were carried out using a Zetasizer Nano ZS instrument (Malvern Instruments, UK). Triplicate samples were measured three times each at 25°C. The measurement was initiated within 120 seconds after sample preparation. Second, Fourier-transform infrared spectroscopy (FT-IR, Spectrum Two, Perkin Elmer) analysis revealed energy changes in the functional groups due to the intermolecular interaction between mandelic acid, carnitine, M_C simple mixture, and M_C complex. Through complementary interpretation of the data, the difference between the M_C complex and others was elucidated, confirming the formation of the complex.

2.3 Measurement of exfoliating capacity using porcine skin

To evaluate the efficiency of exfoliation, we counted corneocytes, which are released from the porcine skin disks with dimensions of 2.5 × 2.5 cm (Micropig Franz Cell Membrane, Apures, Korea). After thawing and washing with Phosphate-buffered saline (PBS, GIBCO, USA), a porcine skin was prepared into 6 mm diameter disks by punching out and placed in a 96-well plate. Samples of the exfoliating agents were prepared in distilled water of various concentrations and pHs. The pHs of samples were adjusted by potassium hydroxide (KOH, Sigma-Aldrich, USA) and hydrochloric acid (HCl, Sigma-Aldrich, USA). These 100 µL samples with Triton X-100 (Sigma-Aldrich, USA) 0.1% were added to each well (positive control: gluconolactone (PHA, Roquette Frères, France) at pH 4). Incubation was carried out at 37°C with 50% relative humidity for 24 hours. Then, 10 µL of each well's supernatant was placed in a hemocytometer ((Paul Marienfeld GmbH & Co. KG, Lauda-Königshofen, Germany) to count the released corneocytes from the porcine skin disk.^{9, 21}

2.4 Clinical assessment

2.5 Statistics

All statistical analyses were performed using IBM SPSS Statistics 21.0 software (IBM, Armonk, NY). The results were expressed as the mean ± standard deviation (SD). We conducted paired t-test analysis to determine whether there were significant changes in the measured results of various skin improvements before and after the administration of the formula in the study participants.

3.1 Preparation and characterization of M_C complex

The ion-pairing method was employed to form the M_C complex. To verify the formation of the ion-pairing complex, we compared the M_C complex with Mandelic acid, Carnitine, and the M_C mixture in terms of size and zeta-potential changes. Additionally, we analyzed energy shifts due to functional group interactions using FT-IR analysis. Table 1 provides information on the particle size and zeta-potential changes for Mandelic acid, Carnitine, M_C mixture, and M_C complex. These measurements were taken at a 1% concentration. Although Mandelic acid and Carnitine are not polymers with large molecular weights, Table 1 shows that Mandelic acid exists in a highly aggregated form with a size of over 1000 nm. On the other hand, Carnitine exists in a smaller size of about 30 nm, indicating aggregation. When simply mixed, the M_C mixture has a size of around 200 nm due to the resolution of aggregation by Carnitine compared to Mandelic acid alone. However, the M_C complex formed by the ion pair method shows a significant decrease in aggregation, measuring around 100 nm. This finding is consistent with the resolution of aggregation that is often mentioned in ion pairing complex literature.

The variations in zeta-potential are crucial for confirming the formation of ion-pairing complexes, as these complexes arise from charge-charge interactions or hydrogen bonding between functional groups. Since the zeta-potentials of mandelic acid and carnitine have not been clearly mentioned in the literature, similar to the size measurements, the zeta-potential was measured in a 1% solution state. As shown in Table 1, mandelic acid exhibited values of −9.6 and 52.2 simultaneously. Carnitine showed a value of −14.2. The M_C mixture appears to be a simple mixture of both, showing an intermediate value between the negative values of mandelic acid and carnitine, and the positive value remains unchanged. On the other hand, it is known that ion-pairing complexes, such as the M_C complex, often show reduced influence of surface charge due to charge-charge interactions or interactions between functional groups, resulting in a value close to zero or measurement being impossible. In this case, the values of −1.03 and −4.7 in comparison to the values of other substances confirm the formation of ion-pairing complexes, as they are measured closer to zero.

To analyze whether actual interactions occurred between functional groups, we investigated the energy changes of the functional groups using FT-IR analysis, as shown in Figure 1. By comparing the peaks of Mandelic acid, Carnitine, M_C mixture, and M_C complex, we observed changes at specific wavenumbers: 3300 cm⁻¹ (for -OH stretching vibration), 2900 cm⁻¹ (for O-H stretching of carboxylic acid), and 1475∼1450 cm⁻¹ (for N-H stretching of the amino group). These shifts indicate that interactions between the functional groups of Mandelic acid and Carnitine occurred within the M_C complex, confirming the formation of an ion-pairing complex.

3.2 Exfoliating Efficacy of M_C complex and Skin improvement Effects Through Clinical assessment

In our study within the cosmetics field, we employed a porcine skin evaluation technique to assess the exfoliating effects of M_C complex as a cosmetic ingredient. Mandelic acid, one of the alpha hydroxy acids (AHAs), is known for its mild exfoliating properties. However, it has been observed that the exfoliation efficacy of mandelic acid decreases when the pH shifts from mildly acidic to neutral. In light of this, our research team compared Mandelic acid with M_C complex and normalized the exfoliation efficacy to provide a more objective comparison. To achieve this, we used gluconolactone (PHA, 10%, pH 4), which is well-known for its excellent exfoliating effects, as the positive control group, setting its exfoliation efficacy as 100% efficiency. We analyzed the exfoliation efficacy of Mandelic acid and M_C complex at pH 4.5, considering different concentrations of mandelic acid (0.1%, 1%, 2%, and 3%). As shown in Figure 2, PHA (10%, pH 4) exhibited the highest exfoliation efficacy, and the exfoliation efficacy increased with higher concentrations of AHAs. Mandelic acid also demonstrated excellent exfoliation effects at pH 4, but at pH 4.5, even at 3%, its exfoliation efficacy significantly decreased to less than 40% compared to PHA. On the other hand, M_C complex showed limited exfoliation efficacy at 0.1% (similar to mandelic acid), but at concentrations of 1%, 2%, and 3%, it exhibited 2.5 to 4.1 times higher efficacy compared to mandelic acid. Carnitine, not primarily an exfoliating agent, contributed to these results through ion pairing complex formation. Notably, M_C complex demonstrated exceptionally strong exfoliation efficacy at 3% compared to PHA (93%) even at pH 4.5. This highlights its potential as a cosmetic ingredient with effective exfoliation properties, maintaining a mild acidic pH level compared to pH 4 while minimizing irritation.

To evaluate the potential efficacy of a low-irritation exfoliating substance, we conducted a clinical evaluation using the M_C complex in actual ampoule formulations. We divided the study participants into two groups and examined various skin improvement effects. As shown in Table 2, a total of 33 subjects applied the formulation twice daily (morning and evening) to their faces for 2-week. We analyzed changes in blackhead area (nose), whitehead area (chin), affected area of pores (cheek), and proportion of dead skin cells related to skin exfoliation efficacy (perioral) using video microscopy, image analysis software, and the ANTERA 3D device. Specifically, the skin exfoliation efficacy was evaluated and analyzed by assessing the degree of reduction dead skin cells ratio. After 2-week of use, the blackhead area decreased by 25.75% on the left side of the nose. Similarly, the whitehead area on the chin decreased by 36.10% compared to baseline. The affected area of pores, assessed on the cheek using ANTERA 3D, showed a 23.68% reduction after 2-week of use. Additionally, the proportion of dead skin cells, analyzed on the left perioral area, decreased by 83.04% compared to the baseline, and we observed an exfoliating effect on the skin. Overall, all four analyses demonstrated statistically significant improvements (p < 0.001) in various skin efficacy parameters, suggesting that the M_C complex has exfoliating efficacy for enhancing skin health.

In Table 3, a total of 31 participants were included in the study. They applied the test material to their faces twice a day (morning and evening) for 1-week. The ANTERA 3D device was used to analyze changes in various pore factors and skin texture over the course of the 1-week. First, pore volume decreased significantly by 16.72% (p < 0.01) after 1-week compared to baseline. Maximum pore depth also showed a statistically significant reduction of 12.1% (p < 0.05) after 1-week. Pore density decreased by 10.7% (p < 0.01), and the number of pores decreased by 10.62% (p < 0.01) compared to baseline. These results suggest that various pore factors improved by 10–17% within just 1-week, likely due to enhanced turnover from exfoliation and skin barrier recovery. Additionally, skin texture showed a significant reduction of 6.47% (p < 0.001) after 1-week, indicating overall skin improvement.

Both groups underwent adverse reaction assessments and surveys to evaluate skin irritation. Interestingly, no adverse reactions were observed. Clinical evaluations and surveys conducted in the two groups confirmed that the M_C complex had no irritant effects and that skin exfoliation was effective in improving blackheads, whiteheads, various pore factor, and skin texture.