Help: How to analyze F-T synthesis products? ! ! ! ! !

This post was last edited by *aojungnft on 2011-11-2 at 15:00. A more convenient approach is to find a doctoral or master’s thesis to consult, as specific descriptions can usually be found there. For the analysis of the product, the product coming out of the reactor is usually divided into three parts: the gas phase, the liquid phase at room temperature, and the solid phase at room temperature. The separation of the liquid phase at room temperature from the solid phase at room temperature is achieved through two flash tanks with different temperature controls; the temperatures in these flash tanks are usually set at 150 degrees and 0 degrees (or room temperature), although these values can also be adjusted according to specific requirements. Those coming out of the high-temperature tank are generally considered to be wax products; offline chromatography analysis can be used, and the right chromatography column needs to be selected, such as HP-5 ； The products obtained from the low-temperature tank are product water and oil; they separate into layers after being left to stand. Before analysis, these two components are separated and analyzed individually. After separating the liquid and solid products at room temperature, what remains is the gaseous component at room temperature; this can be connected to an online chromatographic analyzer, although an offline chromatographic analyzer can also be used. For the qualitative identification of the products, simply by relying on the distribution characteristics of the products obtained through the Fischer-Tropsch synthesis, combined with chromatography column analysis databases (which can be found online or in the databases provided by chromatography column suppliers), as well as reference gases and reference samples (liquids such as n-alkanes of a certain carbon number), it is possible to identify all the products. More precisely, a chromatography-mass spectrometry analysis system can be used. The quantification of the products involves analyzing gases, using sampling and quantification loops for online chromatography, and analyzing oil and wax samples; the amount of each product is determined based on the mass fractions provided by chromatography and the mass of the sampled material. Finally, the amounts of the same components in the gas phase, oil sample, and wax sample are added together to obtain the amount of that component. For material balance, whatever goes in is divided into several components; by comparing the sum of the masses of these components with the amount of material fed in, an error within plus or minus 5% is generally considered an acceptable level of material balance.

The main difficulty in analyzing FT synthesis products lies in the wide distribution of the products, which makes it impossible to analyze them all together. Therefore, generally the dry gas is analyzed online, while different liquids are analyzed offline (sometimes it’s not even possible to analyze a single liquid thoroughly in one column; different analysis methods may be required for different sections), and the temperature is controlled by hot and cold traps; the overall analysis actually also needs to include the product water. The moderator above has already explained it in quite detail. The key lies in the material balance that follows – how to combine the results of online and offline analyses. It is essential to measure the dry gas accurately (this can also be done by using internal standards), as well as to weigh the offline products properly. Look up some literature by yourself and think it through. It’s not easy to achieve material balance.

Reply to 2# *aojungnft: Thank you very much for your guidance. The biggest problem I’m facing right now is obtaining a chromatogram; I can roughly determine what substance each peak represents qualitatively, but I’m at a loss when it comes to performing quantitative analysis of the concentration of these substances through chromatography. Firstly, due to the wide distribution of the products, the area normalization method is used; however, it is possible that the quantification of the product content is overestimated as a result of substances with high carbon numbers that do not produce peaks and thus go undetected ； Using the external standard method, it is not possible to ensure a consistent injection volume each time, which leads to poor repeatability ； It seems that the internal standard method is the best option, but with this method does one have to purchase standard substances with 1 to 60 carbon atoms in order to prepare standard samples and create a standard curve using chromatography? Ordinary alkanes and alkenes even have isomers; dealing with oxides requires an enormous amount of work. Moreover, sometimes the standard curve becomes inaccurate after too long, requiring it to be recreated, which means this task has to be done every day. . . . . Today I spent the whole day reading books on chromatography; they state that for TCD (with hydrogen and helium as carrier gases) and FID detectors, the relative calibration factors for the internal standard method can be found in the manuals, and the error resulting from using these calibration factors is within 3% (it seems that there’s no need to create calibration curves). The approach that comes to mind now is: for the detection of the gas phase, I might need to create a standard curve for each one; after all, hydrogen needs to be measured, and it’s not ideal to use hydrogen or helium as carrier gases. For the analysis of liquid-phase substances, I create a standard curve for that substance in order to quantify it accurately. Assume again that this product serves as an internal standard, and use the relative correction factor to calculate the contents of other products. I’m not sure if this method is practical; I hope experts can give me some guidance. There must be more precise quantitative methods; I would appreciate it if the teachers could provide some guidance. By the way, I have a minor question: where can one find the relative calibration factors for those internal standard methods? Books on chromatography say that the data accumulated in various manuals can be used, but so far I haven’t found any manual that contains such data. . . .

Reply to 3# robinbird: Thank you for your help. I’m currently trying to figure out how to determine the quantitative amounts of various substances from chromatography data, but I don’t know how to do it. I’ve read some literature, which generally states that xxx is used as the internal standard, and XXX columns are employed for chromatographic analysis. Next, various formulas are listed to calculate the conversion rate and the selectivity of products with different carbon numbers. In fact, it is not difficult to derive these formulas based on the definitions of material balance and selectivity; however, the specific chromatographic analysis methods used for quantification are not mentioned in the literature I have seen. What I don’t know how to handle isn’t which formulas to use for the calculations, but rather the values of each individual term within those formulas (and what kind of chromatographic quantification method I should use to obtain accurate values). I hope the teachers can offer some guidance on this matter.

This post was last edited by fossil-zhang on 2011-11-3 at 16:53. Actually, this issue was already explained very clearly by two people earlier; since you still have doubts, I’ll say a few more words. First of all, you need to understand that due to the wide distribution of Feixiao synthesis products, with commonly used catalysts, the products can range from C1 to C70; these include not only hydrocarbon compounds but also various oxidized compounds (such as alcohols, aldehydes, and esters). In particular, these oxidized compounds are present not only in water but also in oils and waxes, including those with long chains. Therefore, the analysis of Fischer-Tropsch products must be performed in stages. Speaking in segments, let’s first discuss the gas-phase products. Since the composition of these products is quite complex, including not only low-carbon hydrocarbons but also unreacted hydrogen, carbon monoxide, and carbon dioxide, it is impossible to analyze all of them completely using just one chromatography column. At least two to three chromatography columns are required for analysis using column switching technology; generally, hydrogen, nitrogen, methane, carbon monoxide/carbon dioxide, ethylene, and ethane are separated using molecular sieve columns (although these groups can also be analyzed using a single column). The remaining C3–C8 light hydrocarbons are then separated using another alumina or OV-1 column (although C7 and C8 are liquids at room temperature, they are present in the gas-phase products as well). The specific order of peak emergence can be found in many chromatography column standard databases. (The order of elution on the molecular sieve column is the order I mentioned above; I will discuss the OV-1 order later.) The carrier gas used is argon or helium, and the detector is a TCD. As for quantitative methods, the external standard method can be used: two standard gases of different concentrations (one at a high concentration and one at a low concentration, covering each substance present in the gas phase) are purchased, and a standard curve is used for quantification. However, such standard gases are relatively expensive; if conditions do not permit this, relative correction factors (which can be found in books and literature, such as analysis chemistry handbooks) can also be used for quantification. The accuracy of both methods depends on the quality of the standard gases and the accuracy of the relative correction factors. The liquid-phase products are separated into two fractions; during collection, a hot trap and a cold trap are used respectively. One fraction is an oil with a carbon chain length of around C5–C22, while the other fraction is a wax with a carbon chain length of around C8–C70. The oil fraction can be analyzed completely using the commonly used HP-5 (or OV-1) column, whereas the wax fraction requires a column capable of withstanding high temperatures. The wax fraction also needs to be dissolved and diluted in solvents such as xylene before analysis; generally, a column that can withstand temperatures above 450 degrees is required (although those commonly used can only handle up to 350 degrees). The detector used is a FID. The qualitative identification of these two types of products can be carried out using a chromatography column standard library or gas chromatography-mass spectrometry. Generally, compounds are grouped by the number of carbon atoms; within each group, the order in which peaks appear is isomer – terminal alkene – n-alkane – trans-diene – cis-diene. The retention time increases as the number of carbon atoms increases, and for compounds with a high number of carbon atoms, n-alkanes dominate. The quantification of these two fractions of products can be carried out using a normalization method, since for each fraction of product, every component elutes on the chromatography column. Moreover, assuming that the corresponding precursors for each component are identical (all being hydrocarbon homologs), the percentage of the peak area for each component corresponds to its weight percentage. The error resulting from such an assumption remains within acceptable limits, and it can be said that almost all those working on the Fischer-Tropsch process analyze things in this way. It should also be emphasized here that if an iron catalyst is used, there may still be some long-chain alcohols present in the oil and wax; the peak positions are generally located near those of the n-alkanes with a difference of a few carbon atoms. Gas chromatography-mass spectrometry can be used for qualitative analysis, while relative correction factors can be employed for quantitative analysis. Most other oxygen-containing compounds are in water, and they can be analyzed using highly polar columns (the columns commonly used for analyzing wine will work as well). For quantification, both external standard and internal standard methods can be employed, following the same procedures as those used for analyzing gas-phase products. For cobalt catalysts, oxidized compounds are originally present in small amounts; there are very few in oils and waxes, and they are mainly found in water. With the chromatography analysis results, I think you know how to calculate the product distribution. In our laboratory, the gaseous products are analyzed using an Agilent four-channel micro online chromatograph; each channel is equipped with its own column and detector – namely the MS-5, PLOT-U, Al2O3, and OV-1 columns. Different products are analyzed in each channel: the first two channels are used to separate hydrogen, nitrogen, methane, carbon monoxide, carbon dioxide, ethylene, and ethane, while the last two channels are used to analyze low-carbon hydrocarbons with C3–C8 structures. The quantitative method is the external standard method using a standard gas. The other oils, waxes, and water were analyzed using 3 offline Agilent chromatographs, with HP-5, HT-5, and INNOWAX as the chromatography columns respectively. GC-MS is used for qualitative analysis, while the normalization method and external standard method are used for quantitative analysis.

Reply to 6# fossil-zhang: The explanation was very detailed; thank you so much. I would also like to ask you: you mentioned that hydrogen, nitrogen, methane, carbon monoxide/carbon dioxide, ethylene, and ethane can also be analyzed using a single column. I’m not sure what kind of columns would be better; those two columns used in your laboratory – could they be separated if the chromatography conditions are adjusted? Today, Shimadzu’s customer service recommended a Plot Q column; I’m not sure if it’s possible to use separate columns. After all, we only have two chromatographs in our laboratory, so it’s better to use as few columns as possible

This post was last edited by *aojungnft on 2011-11-3 at 19:18. The Plot Q column is an excellent choice; it can separate all the light hydrocarbons in the gas phase. In other words, usually two columns are required to analyze the C1-C8 components (one column for C1-C2 and another for C3-C8), but this can be achieved using just one Plot-Q column. I remember this column being quite special; it was a capillary column made of aluminum oxide. Analyze in combination with FID. The remaining gases, namely hydrogen, nitrogen, carbon monoxide/carbon dioxide, can be fully analyzed using another column in combination with a TCD; our laboratory uses a Carboxen column. Additionally, in the analysis of waxes, it is not possible to analyze all long-chain hydrocarbons, which leads to inaccuracies in calculations using the area normalization method. To address this issue, first, try to increase the temperature of the injector as much as possible; second, increase the analysis temperature inside the column chamber as much as possible (without exceeding the maximum temperature of the column). When the maximum value for either of these two parameters is only, for example, 350 degrees, the impact of the resulting values on the overall material balance calculations is not significant; this is the level of analysis used in the vast majority of laboratories working on FTO processes. Therefore, the area normalization method can be used. Furthermore, the spectra obtained from the analysis of wax always show a clear downward trend in the range of components with higher carbon numbers. This is not necessarily due to an issue with the analysis; it could also be that such is the composition of the wax itself. As for why hydrocarbons with higher carbon numbers deviate from the alpha distribution, one theory suggests that it is due to the cracking of long-chain hydrocarbons. Therefore, from this perspective, the injector temperature and the final temperature of the column oven should also not be set too high. Sasol’s H-1 wax has a starting cracking temperature of around 260 degrees.