Research Progress on the Separation of Alkaloids from Chinese Medicines by Column Chromatography

Alkaloids have a variety of bioactivities and great development value in the fields of pharmaceuticals, cosmetics and health food. Column chromatography is a common method for preparing alkaloids. In this paper, the research status of the separation and purification of alkaloids from Chinese medicines by column chromatography is reviewed, and the factors that influence the refining of alkaloids via a macroporous adsorption resin, ion exchange resin and silica gel are summarized. The thermodynamic and kinetic modeling methods for the static adsorption of adsorbents are also reviewed in this paper. It is suggested that the modeling method of the column chromatography process be deeply studied to establish a more stringent quality control method for sampling liquid and to strengthen the online detection of the chromatography process to improve the refining effect of alkaloids.


Introduction
Alkaloids are widely found in higher plants on land, especially in Caryophyllaceae, Annonaceae, Apocynaceae, Compositae, Berberidaceae, Boraginaceae, Buxaceae and other plants, but less frequently in lower plants and animals. Alkaloids are natural secondary metabolites that are generally synthesized through the biosynthetic amino acid pathway and the mevalonic acid (isoprene) pathway [1].

Properties of Macroporous Resin
Macroporous resins, as organic polymer adsorbents, generally have a macroporous network structure and large specific surface area. The macroporous resin does not contain exchange groups. The resin adsorbs molecules through van der Waals force, and the molecules are separated and purified after being eluted by a certain eluent. The physical properties of macroporous adsorption resins commonly used in alkaloid refining are summarized in Table 1. Some of their photos are shown in Figure 1. D-101 is opalescent spherical particles with a wide size distribution. The diameter values of some particles are larger than 1mm.
Compared with D-101, the particle diameter of HPD-100 or D-151 is smaller and more uniform. 001 × 7 are golden transparent spherical particles, and the particle diameter distribution is relatively narrow.  D-101 Nonpolar 9 -10 0.25 -0.83 480 -520 [6] D-3520 Nonpolar 8.5 -9 0.3 -1.25 480 -520 [5] HPD-100 Nonpolar 8.5 -9 0.3 -1.25 650 -700 [7] NKA-9 Polar 15 -16.5 0.3 -1.25 250 -290 [5] XAD-4 Nonpolar 5.8 0.49 -0.69 750 [8] X-5 Nonpolar 29 -30 0.3 -1.25 500 -600 [9] As presented in Table 1, the nonpolar or weakly polar macroporous resins are more widely used in the separation and purification of alkaloids than the polar macroporous resins. Table 2 lists the resins used by researchers in China and abroad in the separation of alkaloids. Table 2 shows that many researchers are more likely to use HPD-100, D-101, and AB-8 macroporous resins. The common characteristics of the three resins are that the specific surface area is greater than 480 m 2 ·g −1 , the pore diameter is in the range of 8 to 14 nm and the particle diameter is in the range of 0.25 -1.25 mm. A larger particle diameter is beneficial to reduce the liquid pressure in the chromatography process.

Chromatography Process
The basic process of refining the alkaloids of Chinese medicines with a macroporous resin is as follows: resin pretreatment, sample loading, washing, elution, and resin regeneration. Table 2 lists the sample loading, washing, and elution conditions and refining results reported in the literature.

Sample Loading
The main factors affecting the sample loading process are the properties of the loading solution, the loading solution volume, and the flow rate of sample loading. Table 2 shows that most of the work was performed with a water solution. The pH value of the loading solution has a great influence on the adsorption effect [9] [10], and many researchers controlled the pH value of the loading solution to be alkaline [9] [11]. In an alkaline solution, alkaloids often exist in molecular form, which is more favorable for macroporous resin adsorption. In general, a high concentration, large volume, and fast loading speed of the loading solution are more likely to lead to the leakage of alkaloids at the outlet of the chromatographic column. Most researchers controlled the sample loading speed to between 1 and 6 BV·h −1 [9] [12]. Some researchers controlled the loading volume according to the leakage of alkaloids in the liquid at the outlet of the chromatographic column [5] [9]. Compared with loading a fixed volume of the sample, this method enabled researchers to make full use of the adsorption capacity of the resin in the column. nearly 90% [8] Ps: in the case of no special instructions, the proportions in this table are volume ratios.

Washing
There was a washing step after the sample loading in most of the literature. The goal of washing is not only to remove impurities but also to minimize the loss of the target alkaloids. Considering that water is cheap and has a good washing ability for polar substances, such as sugars and salts, researchers chose water to wash impurities in most of the literature. From Table 2, we can see that the volume of water used to wash was generally between 2 -8 BV [10] [13], and the washing speed was generally between 1 -6 BV·h −1 [12] [13]. There is also a study in which an ethanol solution with a low concentration was used for washing, and the concentration of the ethanol solution was below 35% [14].

Elution
The main factors affecting the elution effect include the composition of the elu-  [25]. The desorption of target alkaloids, the desorption of impurities, and the consumption of eluent should be considered in the optimization of the elution volume. In Table 2, the elution volume was generally 4-12 BV [9] [12] [26]. In the optimization of the elution flow rate, the consumption of eluent, time, and column working pressure should be considered. In Table 2, the elution speed was mostly between 1 and 5 BV·h −1 , generally not higher than the washing speed [7] [18] [20] [27].

Refining Results
In Table 2, the recoveries of alkaloids from macroporous resin column chromatography can often exceed 90% [11] [28]. This observation indicates that the optimal adsorption and elution conditions can reduce the loss of alkaloids in the chromatography process. The purities of some alkaloids obtained by chromatography reached 95% [8] [24], and the purities of some alkaloids were less than 5% [13] [29]. These results show that the Chinese medicine system is complex.
High-purity alkaloids may not be obtained by macroporous resin column chromatography alone because the working pressure of macroporous resin chromatography is not high and the processing capacity is also large. Macroporous resin chromatography can be used as a preliminary purification, and then, further purification can be carried out through crystallization and other methods.

Properties of Ion Exchange Resins
Ion exchange resins are organic macromolecular adsorbents with ion exchange groups and a network structure [42]. Ion exchange resins commonly used for alkaloid separation are strongly acidic or weakly acidic cation exchange resins with styrene or acrylic acid macroporous backbone structure, and their properties are listed in Table 3. Some of their photos are shown in Figure 1. Among them, the 001 × 7 strong acidic cation exchange resin is most commonly used in alkaloid refining.

Ion Exchange Resin Chromatography Separation Process
The main steps for the purification of alkaloids by ion exchange resins are the same as those of macroporous resins, which also include pretreatment, sampling, washing, elution and regeneration. The sampling, washing, elution steps and purification effects for certain research works are listed in Table 4.

Sample Loading, Washing and Elution
Compared with macroporous resins, the characteristics of a sample solution treated by an ion exchange resin include two points. First, the pH value of the sample solution is mostly less than 3, which means that the alkaloids are in the form of salts when they are loaded. Second, a higher proportion of ethanol is allowed in the sample solvent, which means that the eluate of a macroporous resin may again be refined by the ion exchange resin. In Table 4, the sample loading speeds are mostly controlled at 2 -4 mL·min −1 or 4 -6 BV·h −1 . The higher sample loading speed may be related to the larger particle size of ion exchange resins.
After sample loading, the ion exchange resin can be washed with water first.
The washing speeds listed in Table 4 shall not be lower than the sample loading speeds.  It can be seen from Table 4 that the eluent for refining alkaloids by ion exchange resin chromatography is usually an alcohol solution, ammonium hydroxide, sodium chloride solution, acid added solution, etc. The concentration of ethanol solution is generally more than 50%. A comparison of macroporous resins and ion exchange resins on the adsorption and desorption of alkaloids are shown in Figure 2.
Wang et al. [48] used the 001 × 7 ion exchange resin to separate and purify the total alkaloids of Uncaria and found that the 50% ethanol solution of 5% sodium chloride had a high elution rate for the total alkaloids. Peng et al. [49] found that 5% ammonia and 70% ethanol could wash out part of Leonurus alkaloids. In the literature, the range of elution velocities is 6 -8 BV·h −1 . In the literature, there are examples of not only a fixed elution volume but also the determination of the high performance liquid chromatography (HPLC) [50] and precipitation reactions [49].

Refining Results
From Table 4, it can be seen that the purity of alkaloids separated by ion exchange resin chromatography is between 0.3% and 78%, and the recovery of alkaloids is between 76% and 92%. Compared with macroporous resins, the purity of total alkaloids from ion exchange resins is not high.

Properties of Silica Gel for Column Chromatography
The silica gel for column chromatography is generally a transparent or milky-white  refine the alkaloids of Chinese medicines, as seen in Table 5. It can be seen from the table that most silica gel used for column chromatography is 200 -300 mesh, and its particle size range is 45 -75 μm. The smaller particle size is beneficial to increase the specific surface area for adsorption and separation effects, but it will also increase the pressure in the chromatography process.

Loading and Elution
When silica gel is used as an adsorbent, in addition to the wet method, the dry method is also often used. The operation of wet sampling is more convenient, but dry sampling can solve the problem of low solubility of the components to be separated in the sample solution.
For a positive silica gel, the eluant includes chloroform methanol [54] [55], methanol water [56], methanol [57] [58], etc., and sometimes ammonium hydroxide [59] [60] are added to adjust the pH value of the eluant. For a reversed phase silica gel, the eluant is generally an organic solvent water solution. Compared with macroporous resins and ion exchange resins, gradient elution is commonly used when silica gel is used as the adsorbent. Gradient elution is more complex than isoelution, but it is beneficial to obtain high-purity alkaloids.
The amount of eluent and the elution time should be considered when choosing the elution flow rate. The elution flow rate is usually 0.5 -1 BV·h −1 , which is substantially slower than that of macroporous adsorption resins and ion exchange resins. The main reason is that the particle size of silica gel is obviously smaller than that of the commonly used macroporous adsorption resin or ion exchange resin, and the operating pressure will be high when the flow rate is large.

Refining Results
It can be seen from Table 5 that the purity of the products obtained from silica gel separation of alkaloids is mostly over 90%, sometimes even close to 100%.
Compared with ion exchange resins and macroporous resins, the purity of the product is high. The overall recovery in Table 5 is not high, which may be due to the sacrifice of the recovery to obtain high-purity target alkaloids. If the purity of the alkaloids is required to be high, macroporous adsorption resins or ion exchange resins can be used for a crude separation, and then, silica gel can be used for refining.

Modeling Method
Modeling is important for the chromatography process optimization. In general,

Modeling Based on Statistics
Dr. Yu suggested adopting a design of experiments in research on pharmaceutical processes [67]. Some researchers adopt the Taguchi design to study the technological parameters of column chromatography [9] [12] [14] [35]. This design requires a few experiments, but the obtained data can only be modeled by linear equations. The Box-Behnken design [20], by contrast, enables researchers to model second-order polynomials with quadratic terms and interaction terms, although the design requires more experiments. The second-order polynomials modeled are beneficial to obtain the optimum conditions in the research scope after optimization. The model form is generally as shown in Formula (1).
where Y is the response variable; a 0 is a constant; a i , a ii , and a ij are the linear, quadratic, and cross-product coefficients, respectively; X i and X j are different parameters; and m is the number of parameters.

Thermodynamic Model of Static Adsorption
In static adsorption, the adsorption capacity of a macroporous resin for alkaloids can be evaluated by adsorption isotherms. There may be many forms of adsorption isotherms. Certain examples are listed in

Kinetic Model of Static Adsorption
The static adsorption rate is an important index when optimizing the resin for refining alkaloids. The results can be fitted by using a variety of models, as seen in Table 7.
The pseudo-first-order and pseudo-second-order kinetic models are the most commonly used empirical models for alkaloid adsorption [73] [74] [75]. The pseudo-first-order kinetic model is more accurate in fitting the initial stage of adsorption when the initial concentration of adsorbate is high, while the pseudo-second-order kinetic model is suitable for fitting the subsequent stage of adsorption when the initial concentration is low, and the pseudo-nth-order model dynamics is the generalization result of these two models. The mixed order model combines the pseudo-first-order kinetic model and the pseudo-second-order kinetic model, so it can be used to fit the whole adsorption process of any initial concentration.

Conclusions
In conclusion, there are many studies on the purification of alkaloids by chromatography. The most commonly used adsorbents are macroporous adsorption resins, ion exchange resins and silica gel. In the separation and purification of alkaloids, a nonpolar macroporous resin is often used. The purity of alkaloids from an ion exchange resin and macroporous resin is not high, but the purity of alkaloids from silica gel refining is high. Compared with silica gel, macroporous resins and ion exchange resins are cheaper and have a lower operating pressure, so they are more suitable for the preliminary separation of alkaloids.
The authors think that future research can be carried out in the following directions: Firstly, the modeling method of the column chromatography process needs to Y. Q. He et al. Advances in Chemical Engineering and Science be further studied. The dynamic adsorption process is usually described by the general rate model [76]. Xu et al. [77] used the model to simulate the chromatography of a simulation system containing Puerarin and Daidzein, and the results were in good agreement with the experimental values. It is difficult to describe the phenomenon of competitive adsorption because of the complexity of the components in Chinese medicine extract. To date, there has been no research report related to the general rate model of alkaloid chromatography. Therefore, it is still necessary to develop effective modeling methods to describe the chromatography process of Chinese medicine extracts.
Secondly, stricter quality control methods should be established. There are differences in the content of components in different batches of Chinese medicine extract solution, which affect the effect of chromatography. However, there is no research focus on the influence of the quality change in the sample solution on the chromatography effect so far. To control the consistency between different batches of alkaloids, it is suggested to set the quality standard of the sample solution. Pan et al. [78] established a quantitative model of process parameters, raw material properties and evaluation indexes of column chromatography eluent and then calculated the quality standards of raw materials according to the requirements of the evaluation indexes. This idea can be used for reference to establish the quality standard of a sample solution for alkaloid column chromatography.
Thirdly, online detection of chromatography processes should be strengthened. At present, in academia, spectral technologies combined with multivariate statistical methods are often used to detect the content of indicators/major components in Chinese medicines or to detect the process trajectory [79] [80] [81]. This approach has the advantage of real-time online quantitative detection.
However, it has not been reported that it can be used in the chromatography of alkaloids, so it needs further development.