An advanced freeze-drying technology for HPLC fractions
Rob Darrington, Genevac Ltd., Ipswich, UK
Introduction
Drying high performance liquid chromatography (HPLC) purification of fractions (mainly water and acetonitrile) is a common and important experimental task in many laboratories. The requirement of the experiment is to dry the sample into a powder in order to accurately weigh the sample, perform a second sampling or dissolve again. Therefore, freeze drying is the preferred technique for experiments, but conventional large freeze freeze systems may be difficult to handle organic solvents because organic solvents can easily bump in the sample and thus damage the vacuum pump, causing sample loss. Actelion Pharmaceuticals using Genevac 1 company developed quickly lyophilization (LyoSpeed ™), using a Genevac HT-12 centrifugal evaporator to finish the job. The LyoSpeed ​​method controls the boiling of the sample through a centrifuge, concentrates the organic solvent, and avoids sample loss, and then extracts the organic solvent from the condenser to freeze the remaining water. This method is very effective for hydrophilic samples, but if the sample is insoluble in water, the sample from which the organic solvent has been removed will collapse and even form a grease. These samples require further processing to form the desired dry powder state.
Limiting factor
The average operating success rate of the standard compound library (completely lyophilized vs. oily compound) is approximately 50%. Depending on the properties of the compound, the minimum and maximum operating success rates can reach 30% and 90%, respectively. A simple way to achieve a higher operating success rate is to freeze the sample and leave acetonitrile in the sample. Unfortunately, existing devices are not able to achieve such operational results.
There are also certain limitations in the hardware, especially the temperature characteristics exhibited by the solvent condenser (VC3000) in the HT-12 evaporator. The lowest temperature that the VC3000 can reach is -50 ° C, which limits the vacuum that the evaporation system can achieve. In order to achieve a good lyophilization effect, the degree of vacuum should be as low as possible, and the most ideal state is to be able to reach below 0.1 mbar. For example, pure water has a boiling point of -32 ° C under a pressure of 0.5 mbar, but the boiling point of acetonitrile can be reduced to -61 ° C under the same vacuum conditions, so that the cold trap is blown during boiling and the vacuum pump is blocked, resulting in an evaporation system. The pressure is rising. An increase in pressure will increase the boiling point of the solvent, resulting in an unsatisfactory lyophilization effect and may even lead to a complete failure of the lyophilization process. The purpose of using the LyoSpeed ​​method is not to freeze the sample thoroughly (such as to produce a vaccine), but to produce a non-oily dry powder. Just like the real freeze-drying process, better experimental results can only be achieved by achieving a better vacuum.
Developed advanced solvent condensers
The solution to the problem is clear: the laboratory needs a more refrigerated condenser to concentrate the sample under low vacuum conditions and retain the acetonitrile component. The researchers worked with Genevac's engineering team to conduct a series of benchmark tests on the VC3000 condenser via a VirTis Benchtop K-85 °C freeze dryer (BTK) connected to the HT-12 evaporator. Based on the corresponding experimental results, the team developed a new HT-12 condenser named VC6000 (Fig. 1). The design of the VC6000 evolved according to the structure of the BTK, with stronger condensation, shortening the defrosting time, and the solvent resistance of the Genevac system.
Figure 1 – Genevac HT-12 Evaporation System with VC6000 Condenser
Materials and Methods
The Genevac HT-12 evaporation system was tested and connected to a VC3000, BTK or VC6000 condenser. The 10 ml sample component was a mixture of water and acetonitrile (volume ratio of 50:50), and the sample preparation container was a glass bottle having a diameter of 16 mm and a height of 100 mm, and the wall thickness was 1.2 mm. A solid PVDF plastic sample holder was placed on the flank of the HT-12 evaporator and fitted with 24 glass tubes. The solvent composition in each glass test tube is: non-compound, or 20 mg Diovan or commercially available 20 mg other chemicals, and the lipophilic material o-phthalo-β-alanine (AI028492). Both Diovan and AI028482 are more difficult to freeze from the HPLC fraction, so such a combination forms the "most difficult to handle" sample. Each condenser was tested under two load conditions. In one case, four samples (96 tubes) were used, and in the other case, 12 samples (full load of 288 tubes).
All samples and each condenser use the same method:
1. Control sample boiling/bumping with Dri-Pure® vacuum ramp for 1 hour
2. No hot full vacuum treatment, operation time is 3 hours
3. Full vacuum heat treatment of the sample holder at 40 ° C for 4 hours
4. Full vacuum without heat treatment, operation time is 18 hours
The processing time of the fourth stage is very long. The test is carried out in this way, and the actual drying time can be obtained from the recorded data chart.
result
After lyophilization, the state of the compound in each tube is evaluated to determine if it has been successfully lyophilized to a dry powder, or a sample containing a small amount of oil has been formed, or if there is still residual solvent in each tube. The drying time can be considered as the time at which the temperature converges on the sample holder and the sample.
Figure 2 – Freeze-drying results for each test condition
Test System: Test System
Number of Tubes: Number of tubes
Compound: compound
Drying Time (Minutes): drying time (minutes)
Count of Condition of Sample after Drying : Sample state after drying
Light Oily: a small amount of grease
discuss
If no residual sample is found after lyophilization, a test success rate of 100% is achieved. Therefore, the results of the experiment in which 2 of the 96 samples placed in BTK were not dried were unexpected. Since this result was not repeated under the experimental conditions of higher sample loading, no further investigation was made and was considered as an abnormal experimental result.
The VC3000 condenser has the lowest operating success rate when processing actual samples. Benchtop K has better operating performance, but its internal volume is too small to freeze 288 tubes at the same time. The VC6000 condenser has the best operating performance, which is consistent with the expected results. The following two points are worth noting:
1. Why does the VC6000 condenser with Diovan exhibit different operating performance under both low load and high load conditions? The drying time required for the sample at a low load is also longer, which is quite different from the expected result.
2. In each test, the samples at each location were identical and taken from the same stock solution, but why is the drying between each tube so different?
When viewing the experimental data plots (as shown in Figures 3a and 3b), the reasons for the different results using the VC6000 for two samples (including Diovan components) were clearly identified. Figure 3a shows a lighter sample loading and the yellow line indicates the sample temperature, which does not intersect the pink line representing the sample holder temperature. This indicates that the sample is still wet, as shown by the experimental results.
Figure 3a – Data plot of the lighter sample load (96 samples) on the VC6000 condenser with Diovan
The reasons for the results of this experiment are not immediately known and require further investigation. Its insufficient vacuum of about 1.5 mbar caused the sample to freeze and dry completely. The root cause of the problem is a slight decrease in the performance of the scroll pump. For systems that handle organic solvent vapors, the Scroll pump is undoubtedly the best choice and does not create grease that can damage the vacuum pump. As with all vacuum pump operations, the pump must be regularly maintained to ensure optimum performance and the best possible results. For routine maintenance of the scroll pump, the end seals need to be inspected every 3 months to ensure good performance, ease of operation and reduced contamination.
Figure 3b – Higher sample loading (288 samples) data plot on a VC6000 condenser with Diovan
If you need to study the pressure curves of three different condensers that handle the same sample, as long as the pure solvent is used in the higher load, as shown in Figure 4, the larger and cooler condenser (VC6000) can maintain the pressure. Low level, which indicates that the condenser has a very strong condensing solvent function. The most important thing for achieving successful freeze drying is to maintain the required low pressure conditions, and the two essential elements that determine this condition are good vacuum and a powerful and fully cooled condenser.
Figure 4 – Pressure performance data for three cold traps at high load conditions
In each experiment, it was difficult to consider changes in drying performance between tubes. Figure 5 accurately reflects the experimental results of the VC6000. It can be clearly seen that the test tube at the edge of the sample rack is not easy to dry, and a part of the test tube near the center position is not dried successfully. Figure 5 accurately shows all the experimental results. Among them, one viewpoint that can explain such a phenomenon is that the test tube at the center position is more affected by the cooling of the adjacent test tube, and thus can be effectively frozen and dried. However, if this view is absolutely correct, then the drying effect of all peripheral tubes should be in an unsatisfactory state. Close observation of tubes that could not be dried did not reveal tidal lines, indicating that the samples had been frozen, lyophilized, and thawed or broken--can we conclude that these samples were only cryogenically cooled and evaporated? This view may be correct, but not sufficient to fully explain all experimental phenomena.
The staff carefully used clean new glassware, solvents and samples in each test, thus eliminating the possibility of contamination. Another explanation for the experimental results is that since the compound itself forms a film on the surface of the sample, it may hinder or completely block the evaporation process. The user of the freeze-drying system has carefully described the concept of "cake resistance", and the thickness of the film may also increase as the depth of the sample increases. The actual results of the dry compound library show that the film formed on the surface of the sample will hinder the drying, so that a few milliliters of liquid will remain. Sample research staff reported that a portion of the compound library with specific chemical properties is prone to form a film on the surface. For the experimental operation using AI02492 (Fig. 5), the reason why a small number of samples cannot be dried is still unknown. Although the formation of a film may be a cause of this result, it does not explain why only 14 samples cannot be dried, while another 274 samples are completely dry.
Figure 5 – Location of wet tubes after lyophilization using AI02492 (connected to VC6000) under high load test conditions
Description – yellow = <0.5ml red = <4.5ml white = dry powder
to sum up
The experimental results show that the VC6000 has superior freeze-drying performance (especially at high load) compared to other systems used in the test. The vacuum of the system is a key factor in achieving good experimental results, requiring a closed vacuum system and a powerful and fully cooled condenser. The VC6000 condenser has been used to perform routine workflows, building "real compound libraries" for a variety of compounds with highly variable dissolution patterns (some eluting in 90% organic solvents and some eluting in 90% water). In this case, similar evaporation experiments will occur. Even now some samples are placed in an aluminum sample holder for processing, and the experimental results obtained are still the same.
Acknowledgement
The author would like to thank Viktor Ribic, a senior laboratory researcher at the Chemicals Division of Actelion Pharmaceuticals in Alswell, Switzerland, for his long-standing research work and for providing valuable experimental data for this article.
references
1. Developments in Laboratory Scale Lyophilisation for Purification Laboratories.
Dr. Induka Abeysena, Rob Darrington, 2006, available at
An advanced freeze-drying technique for HPLC fractions (
128 times )
An improved lyophilization technique for HPLC fractions (
41 times )
Medical Mask
3 Ply Face Mask,Dustproof Face Mask,Civil Face Mask,Folding Face Mask
GULF STAR MEDICAL MANUFACTURING FZE , https://www.gulfstarmedical.com
