Thread Content
What is commonly referred to as passing through a column should be called column chromatography separation, or simply column chromatography. What we commonly use are adsorption columns with silica gel or alumina as the stationary phase. Due to the excessive number of empirical components in column separation, I will share some insights from my experience working with columns over the past few years, in the hope that it will be helpful. Columns can be divided into: pressurized, atmospheric pressure, and reduced pressure. Pressure can increase the flow rate of the eluent, reducing the time required to collect the product, but it reduces the number of theoretical plates in the column. Therefore, when all other conditions are the same, a column operated at atmospheric pressure offers the highest efficiency, but it also takes the longest time; for example, in the separation of natural compounds, it can take several months to use one column. A vacuum column can reduce the amount of silica gel used, seemingly saving half or even more of it; however, since a large amount of air passes through the silica gel, this causes the solvent to evaporate (sometimes water vapor condenses outside the column), and some substances that are prone to decomposition may not be recovered. Additionally, a water pump is required to create a vacuum (which generates a lot of noise and takes a long time). I used to use pressure-reduction columns extensively and had a strong fondness for them, but ever since I tried pressure-application methods, I’ve hardly thought about using pressure-reduction columns again. A pressurized column is a good method; similar to an atmospheric pressure column, except that the applied pressure speeds up the movement of the eluent. The source of pressure can be compressed air, a double ball pump, or a small air pump (one used for aerating fish tanks will do). It is particularly applicable to the separation of easily decomposable samples. The pressure should not be too high, otherwise the solvent will flow away too quickly, which will reduce the separation efficiency. Personally, I think pressure columns are quite suitable for the separation of ordinary organic compounds. Regarding the size of the columns, it is best to have them thick and long. As the columns grow, the corresponding number of trays increases. The column becomes thicker, and after sample loading the origin of the sample is smaller (which on the column means that the sample layer is thinner), thereby relatively reducing the difficulty of separation. Imagine if the column is 10 centimeters long while the sample is only 2 centimeters long; the difficulty of separation can be imagined – it will likely be necessary to use a solvent with very low polarity to carry out the process slowly. And if the sample layer is only 0.5 cm thick, it is easier to achieve complete separation of the various components. Of course, using larger columns means sacrificing more silica gel and solvent, but these costs are probably negligible compared to the value of the product (some argue it’s not environmentally friendly, but solvent recovery and re-distillation help reduce waste to some extent). The diameter-to-height ratio of the columns seen these days is generally between 1:5 and 1:10. The book states that the amount of silica gel should be 30 to 40 times that of the sample; the specific choice needs to be determined through careful analysis. If the difference between the desired component and the impurities is large (that is, when the RF value of the desired component is between 0.2 and 0.4, and the difference for the impurities is more than 0.1), less silica gel can be used, and a smaller column can be employed (for example, for a 200-milligram sample, a 2 cm × 20 cm column can be used) ； If the difference is less than 0.1, it is necessary to increase the size of the column; I think the diameter of the column can be increased, for example to 3 cm, or the polarity of the eluent can be reduced. An anhydrous and anaerobic column is suitable for products that are sensitive to oxygen and water and prone to decomposition. It can be a wet column or a dry column. However, the column must be saturated with the solvent at least once before using the sample, as the heat released when the solvent and silica gel become saturated could lead to the decomposition of the product; after all, what is being separated are sensitive substances, so it’s better to be cautious. It is also because the separated substances are quite sensitive; therefore, the receiving bottle must be sealable, and Schlenk techniques must be followed. As for whether it is under pressure, at normal pressure, or under reduced pressure, it depends on the requirements. Since it is a Schlenk operation, spotting the plate poses a problem. If the sample develops color, congratulations – there’s no need to spot the plate; one can simply observe the color bands on the column. If the sample is colorless, one has to prepare dozens of Schlenk flasks and test them one by one; however, after a few attempts it becomes clear where the sample is, which allows for saving some effort. Previously, when I worked with a column without water or oxygen, I needed six Schlenk lines; now, just one is enough to collect everything required. Alumina is commonly used as the stationary phase in anhydrous and anaerobic columns. Because there are many hydroxyl groups exposed in silicone, it is easy for the samples to decompose, especially metal-organic compounds and phosphorus-containing compounds. Alumina can be made in basic, neutral, and acidic forms, offering a wide range of options, but it is more expensive than silica gel. I heard there’s a method that involves using quartz as the column and HF254 as the stationary phase; by shining an ultraviolet light on the outside of the column, it’s possible to determine where the product is located. It hasn’t been tested yet. Those who have done it can share it so that everyone can take a look. Regarding wet and dry loading. The wet method is convenient; generally, a washing solution is used to dissolve the sample, and dichloromethane, ethyl acetate, etc. can also be used. However, the fewer solvents there are, the better, otherwise the solvents will end up functioning as washing solutions. Many samples are sticky before being loaded onto the column, but this is usually not a problem. However, some samples precipitate again on the silica gel after being applied; this usually occurs with larger amounts of sample, as the silica gel becomes saturated with the adsorbed sample. This phenomenon happens because the sample itself is a solid with good solubility. In such cases, recrystallization should be carried out first to obtain most of the product, followed by column separation. If recrystallization is not possible, then it can be ignored and the sample can simply pass through directly, as it will dissolve as the eluent flows. Some samples have poor solubility, and the solvents that can dissolve them cannot be used for column loading (e.g., DMF, DMSO, etc., which will move along with the solvent, resulting in a long tail in the color development); in such cases, dry loading onto the column is necessary. There is a suggestion that the ratio of sample to silica gel should be 1:1; I think the less, the better, but it is necessary to ensure that no visible solid particles remain after centrifugation (as that would indicate that some of the sample has not been adsorbed onto the silica gel). Selection of solvent. Of course, it’s the cheapest, safest, and most environmentally friendly. Therefore, petroleum ether and ethyl acetate are mostly used. The literature mentions the use of n-hexane, but it’s too expensive; unless it’s truly necessary, it’s not worth the high cost – the amount spent on it is even greater than that spent on rinsing agents. However, due to its low polarity, it is sometimes the only option available. Diethyl ether can also be used, but it makes one drowsy easily; be careful to stay awake and avoid letting the solvent run out, otherwise the column will not function properly. Dichloromethane is also useful, but it should be noted that its adsorption by silica gel is an exothermic process; therefore, bubbles often form inside the column in summer, while the situation is better in colder weather. Methanol is said to be able to dissolve some silica gel; therefore, if elemental analysis is to be performed on the product, care must be taken, and subsequent treatments such as recrystallization should be carried out. Other solvents are used less frequently, and their choice depends on individual needs. For some reason, the rinsing agents used are usually in large packages (as they’re cheaper). Here we use plastic tanks of 10 or 25 liters, and it’s important to note that the purity of these industrial products is relatively low. Colored impurities can often be seen at the bottom of the large barrels delivered, and other impurities can be inferred from this; therefore, solvent re-evaporation is necessary during more stringent column separation processes. Of course, this step can be skipped when using raw materials, as there are purification methods available later on. Additionally, it is advisable to recycle the solvent after it passes through the column; this is beneficial for the environment and helps save some costs, though it does require some manual labor. It should be noted here that reduced-pressure rotary evaporation is generally carried out simultaneously during column chromatography. The ratio of petroleum ether to ethyl acetate leads to changes in polarity due to their different volatilities; this usually results in an increase in polarity, which is suitable for gradient elution, as the polarity gradually increases. After passing through the column, normal pressure should be used for the final recovery of the solvent, as partial low-boiling-point impurities are removed during vacuum rotary evaporation; using normal pressure reduces this phenomenon. It would be disastrous if those impurities reacted with the sample you intend to process next. Regarding the operation issue. 1 Install columns. The piston below the column must not be lubricated, as the cleaning agent could carry lubricant into the product; a PTFE valve can be used instead. There seems to be no difference between dry and wet column packing; as long as the column is properly packed, that’s sufficient. The filled columns should be moderately compact (if they are too compact, the eluent flows too slowly); they must also be uniform (otherwise the sample will flow down at an angle from one side). The book states that bubbles must not be present, but I think in most cases small bubbles don’t cause much of a problem; once pressure is applied, they all disappear. Of course, if the bubbles keep appearing in the columns you install, it means you need to practice more*. But what columns fear even more is cracking; whether it’s vertical or horizontal, it will affect the separation effect and may even render them useless! 2 Add sample. Add the sample using a small amount of solvent. After adding it, open the piston below; once the solvent layer has dropped to the level of the quartz sand, add another small amount of low-polarity solvent, then open the piston again. Repeat this process two or three times, and generally the quartz sand will become basically white. Add the eluent; do not apply pressure at first. Wait until there is a certain distance between the solvent of the dissolved sample and the sample layer (2–4 cm is sufficient), and then apply pressure. This prevents the solvent (such as dichloromethane) from carrying the sample downward rapidly. 3 Selection of rinsing agents. It seems that it is better to set the required value around Rf0.2~0.3. Don’t think that a larger difference in Rf values means better separation on the plate; when passing through the column, that polarity comes into play. If the Rf value is 0.6, even a difference of 0.2 won’t result in easy separation on the column, as the column operates in a manner similar to multiple passes over the plate. Through formula comparison, it’s clear that the degree of separation achieved with 0.6/0.8 is certainly less than that obtained from (0.2/0.3) raised to the third or fourth power. Collection of 4 samples. The principle behind using silica gel as the stationary phase in column chromatography is a balance between adsorption and desorption. So if the sample has a strong adsorption to silica gel, it will not flow out easily. That’s what happens: the later points appear first, while the earlier points appear later. Alumina can be used as the stationary phase at this time. Additionally, the size of the test tubes used for collection should be determined based on the amount of sample; especially with small amounts of sample, if large test tubes are used, three samples might end up in one tube, wuwu. If small test tubes are used for everything, the workload will be too large. 5 Final processing. For the product obtained after column separation, since a large amount of solvent is used, impurities can accumulate in the product as well. Therefore, if it is to be sent for analysis, it is best to wash it with a small amount of solvent, as most of the impurities are dissolved in the solvent; washing it away will basically eliminate those impurities. Recrystallization may be necessary if required