*3.2. Quantification of Curcumins I–III in C. longa and Herbal Formulation*

The curcumins I–III peaks in the *C. longa* and herbal formulation were identified by comparing their spots at Rf = 0.45, 0.52 and 0.61 with that of purified curcumins I–III (Figure 2) as well as the overlay UV spectra of the spots with the same Rf value (Figure S9). The amount of curcumins I–III present in the *C. longa* (Figures 3–5) and herbal formulation was quantified from the regression equation obtained from the calibration plot of HPTLC. The amounts of curcumins I–III in *C. longa* and herbal formulation are presented in Table 5.

**Figure 3.** HPTLC densitogram of fresh *C. longa* extract by (**A**) soxhlet and (**B**) maceration.

**Figure 4.** HPTLC densitogram of dry *C. longa* extract by (**A**) soxhlet and (**B**) maceration.

**Figure 5.** HPTLC densitogram of *C. longa* formulation extract by (**A**) soxhlet and (**B**) maceration.



#### **4. Discussion**

#### *4.1. Characterization of Curcumins*

The three purified curcumins I–III shared the same carbon chain composed of C-1, C-2, C-2 , C-3, C-3 , C-4 and C-4 with two trans-oriented methine carbons assigned for C-3, 3 and C-4, 4 , which have trans orientation based on the *J* value of 16.0 Hz between H-3, 3 and H-4, 4 (Supplementary Materials). Curcumin I 1H, 13C NMR spectrum, showed ABX spin system in a 1, 3, 4 trisubstituted aromatic ring (Figures S1A and S2A). Two overlapped OCH3 appeared at δ<sup>H</sup> 3.98 and δ<sup>C</sup> 55.84 ppm. High resolution electrospray ionization mass spectroscopy (HRESIMS) (Figure S3) showed an M+−1 at *m/z* 367.1188, M++1 at 369.1328 and M++Na 391.1147. The data were identical with those reported for curcumin I [32–34]. 1H NMR of curcumin II (Figure S1B) showed one system ABX system assigned for 6, 9 and 10 at δ<sup>H</sup> 7.19 (s), 6.80 (bd, *J* = 7.5 Hz) and 7.08 (d, *J* = 7.0 Hz) in a tri-substituted benzene ring and an AX system assigned for a para-substituted benzene ring at δ<sup>H</sup> 6.80 (bd, *J* = 7.5 Hz) and 7.46 (d, *J* = 7.5 Hz). Signals for one OCH3 at δ<sup>H</sup> 3.83 and δ<sup>C</sup> 56.04 ppm were also observed. HRESIMS (Figure S4) showed an M+−1 at *m/z* 337.1084, M++1 at 339.1226 and M++Na at 361.1043. The data were identical with those reported for curcumin II [32–34]. The 1H NMR of curcumin III (Figure S1C) showed only two aromatic proton signals at δ<sup>H</sup> 6.80 (d, *J* = 8.2) and 7.47 (d, *J* = 8.2) each integrated for four protons diagnostic for two identical para-substituted benzene rings. HRESIMS showed (Figure S5) an M+−1 at *m/z* 307.0977, M++1 at 309.1120 and M++Na at 331.0938. The data were identical with those reported for curcumin III [32–34].

### *4.2. HPTLC Analysis*

Standard curcumins are available as a mixture of the three main curcumins. For the purpose of quantification, the mixture was purified on chromatoron using normal silica gel and the resulted compounds were used as a standard after their identification spectroscopically. The calibration curve for curcumins I–III was observed as linear in the range of 50–500 ng band-1. The results demonstrated a good linear relationship between the concentration and spot area of curcumins I–III. The determination coefficient (R2) value for curcumins I–III was obtained as 0.998, 0.999 and 0.997, respectively, which were highly significant (*p* < 0.05) (Table 1). The R2 values represents the strength of the correlations

between the two variables. All these observations and results suggested that the sustainable HPTLC technique was linear and acceptable for the simultaneous estimation of curcumin I, curcumin II and curcumin III.

The preanalyzed sample of curcumins I–III were spiked with the extra 0, 50, 100 and 150% of the standard curcumins I–III, and the solutions were reanalyzed in six replicates by the proposed method. The % of recoveries of curcumins I–III were estimated as 97.22–98.67, 97.83–98.22 and 99.32–100.84%, respectively, using the HPTLC technique (Table 2). The estimated % recoveries within the limit that exhibited the HPTLC technique was accurate for the simultaneous determination of curcumins I–III.

The precision of the proposed method was determined at two levels, i.e., repeatability and intermediate precision. The repeatability was determined as an intra-day precision, whereas the intermediate precision was determined by carrying out inter-day variation for the determination of curcumins I–III. Intra-day and inter-day precisions (*n* = 6) for curcumin I, curcumin II and curcumin III were found to be 0.78–1.56% and 1.08–1.72%, 0.49–1.06 and 0.74–1.21 and 0.36–1.04 and 0.44–1.43, respectively (Table 3), which demonstrated the good precision of the proposed method.

The robustness of the HPTLC technique for the simultaneous determination of curcumins I–III was evaluated by introducing small deliberate changes in the composition of mobile phase components. The % RSD for curcumins I–III after this change were determined as 1.01–1.33, 0.36–0.75 and 0.45–0.55%, respectively. In addition, the Rf values for curcumins I-III were recorded as 0.43–0.46, 0.49–0.52 and 0.61–0.63, respectively (Table 4). The small differences in the Rf values and lower % RSD demonstrated the robustness of the HPTLC technique for the simultaneous determination of curcumins I–III.

The LOD and LOQ values for curcumin I were determined as 4.36 and 12.79 ng band-1, respectively. The LOD and LOQ values for curcumin II, were determined as 4.59 and 12.67 ng band−1, respectively. However, the LOD and LOQ values for curcumin III, were determined as 4.66 and 12.84 ng band-1, respectively. The recorded values of LOD and LOQ suggested that the HPTLC technique was highly sensitive for the simultaneous estimation and quantification of curcumins I–III.

The specificity was proven by comparing the Rf at 0.45 ± 0.02, 0.52 ± 0.02 and 0.61 ± 0.04 and spectra of the spots in the *C. longa* extract and formulation with that of the purified curcumins I–III. The proposed method was found to be specific by comparing the Rf of rhizomes of *C. longa* and herbal formulation and standard as well as the overlaid spectra at peak start, peak apex and peak end position of the spot showing λmax 423 nm for curcumins I–III (Figure S9). The specificity or peak purity for the proposed analytical method was suggested by the similar UV-absorption spectra, Rf values and detection wavelength of curcumins I–III.

#### *4.3. Quantification of Curcumins I–III in C. longa and Herbal Formulation*

Extraction methods selection is mainly based on the nature of the components to be obtained. Solvent extraction is the most widely used method. Solvent extraction includes maceration, percolation and reflux extraction. It usually requires a large volume of the organic solvents and long extraction time. Some modern extraction methods, such as the super critical fluid extraction, pressurized liquid extraction and microwave-assisted extraction, have been used in extraction. These methods lower organic solvent consumption, shorten extraction time and are more selective. However, they require more sophisticated and expensive instrumentation [35]. Maceration is a very simple extraction method with the disadvantage of long extraction time and low extraction efficiency. It could be used for the extraction of thermolabile components. On the other hand, the soxhlet extraction method integrates the advantages of the reflux extraction and percolation, which utilizes the principle of reflux and siphoning to continuously extract the plant materials with fresh solvent. The soxhlet extraction is an automatic continuous extraction method with high extraction efficiency that requires less time and solvent consumption than maceration or percolation. However, soxhlet extraction is not suitable for plant materials with thermolabile or volatile

components [35]. In the current study, for comparison, both the maceration and soxhlet extraction were applied for the extraction of curcumins from *C. longa* fresh rhizomes, dry powdered rhizomes and the formulation, using methanol as an extraction solvent. The amounts of the three main curcumins were estimated by applying the proposed and validated HPTLC method. The results of the analysis are presented in Figures 3–5 and Table 5. The content of curcumin I–III in fresh rhizome and dry powdered by maceration and soxlet extraction was determined as 0.44, 0.28, 0.65 and 0.53, 0.37, 0.76 and 0.23, 0.15, 0.33 and 0.34, 0.25, 0.47% *w/w*, respectively, using the proposed analytical method. The content of curcumin I–III in formulation by maceration and soxhlet extraction was determined as 0.26, 0.19, 0.41 and 0.28, 0.22, 0.46% *w/w*, respectively, using the proposed analytical method.

As expected in all cases, the soxhlet extraction was more efficient than the maceration and gives better yield of the extracted compounds. Moreover, the soxhlet saves much time and solvent consumption. The results also indicated that the fresh plant materials were superior to the dry powdered plants in term of the percentage *w/w* of the three curcumins. This fact may be explained by the loss of some contents during the drying and processing of the plant material by oxidation or other factors. The amount of curcumin III was the highest in all estimations followed by curcumin I, while curcumin II was present in a lower concentration.

#### **5. Conclusions**

In conclusion, curcumins I–III were purified and characterized by various spectroscopic methods from the standard curcumin mixture available on the market. An HPTLC method was developed, validated and applied to estimate the amount of curcumins in the fresh, dry plant materials and formulation containing *C. longa*. Two extraction methods were applied to obtain the target curcumins from the raw plant materials. The soxhlet extraction gives better yields of the three curcumins than maceration, indicating the thermal stability of the compounds under used conditions. The fresh plant materials contained more amounts of curcumins than the dry plant materials.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/separations9040094/s1. Figure S1. 1HNMR of the aromatic protons of curcumins I–III (A–C). Figure S2. 13CNMR of the aromatic carbons of curcumins I–III (A–C). Figure S3. Mass spectra of curcumin I (A: Negative Mode B: Positive Mode). Figure S4. Mass spectra of curcumin II (A: Negative Mode B: Positive Mode). Figure S5. Mass spectra of Curcumin III (A: Negative Mode B: Positive Mode). Figure S6. Calibration curve of curcumin I. Figure S7. Calibration curve of curcumin II. Figure S8. Calibration curve of curcumin III. Figure S9: Overlay UV absorption spectra of curcumins I–III and corresponding spots in *C. longa* extracts and formulations.

**Author Contributions:** Conceptualization, M.S.A.-K. and P.A.; methodology, P.A., A.A.S., K.A.A. and M.H.A.; software, A.I.F. and M.H.A.; validation, A.I.F. and M.H.A.; formal analysis, M.S.A.-K., A.I.F. and M.H.A.; investigation, P.A., A.A.S. and K.A.A.; resources, M.S.A.-K.; data curation, P.A., A.A.S. and K.A.A.; writing—original draft preparation, P.A., A.A.S. and K.A.A.; writing—review and editing, M.S.A.-K., M.H.A. and A.I.F.; visualization, M.S.A.-K.; supervision, M.S.A.-K.; project administration, M.S.A.-K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors would like to thank the Deanship of Scientific Research at Prince Sattam Bin Abdulaziz University, Al-Kharj, Saudi Arabia for supporting this research.

**Conflicts of Interest:** The authors declare no conflict of interest.
