Decarbonisation Technology - November 2024 Issue

Cat 2 Cat 1 Cat 3 Cat 4 Cat 5 Cat 6 Cat 7

Feed 2 Feed 1

10 ˚C

20 ˚C

TLP Cloud Point (˚C)

Figure 4 Yield of SAF against total liquid product cloud point. Panel A: difference in SAF yields promoted by different catalysts. Panel B: difference in SAF yield promoted by different feeds

it is good practice to check for the presence of longer-chain paraffins (number of carbon higher than 30). The presence of large amounts of linear long-chain paraffins is detrimental to the final cold-flow properties of the total liquid product (TLP). The SAF yield as a function of cloud point for two feedstocks with a different hydrocarbon composition and different levels of S, N, and O is reported in Figure 1 panel B. The SAF yield, obtained by processing both feeds “ To ensure accurate quantification, it is important to perform an accurate calibration of the mass flow meters at the inlet and outlet of the unit ” with the same catalysts, decreased by ~7-8% at all cloud points, with the feed containing the least amount of S, N, and O (Figure 1 panel B). Hydroisomerisation processes are usually operated with a GTO above 300 Nm3/m3 to ensure good hydrogen availability. The hydrogen consumption is typically low, in the order of 0.1-0.3 wt% of feed. To ensure accurate quantification, it is important to perform an accurate calibration of the mass flow meters at the inlet and outlet of the unit. During the course of a test, the mass flow meters may also begin to drift slightly, showing a marginally increasing trend or perhaps moderately negative values. Negative hydrogen consumption might

also be observed due to the experimental error in the subtraction of two large numbers. These possible experimental errors make it difficult to compare hydrogen consumption among different tests in terms of absolute numbers. This type of additional complication in comparing performances can easily be avoided by employing high-throughput experimentation – where up to 16 reactors can be compared without run-to-run variability. In hte’s high-throughput units, after the reactor, there is a separator (condenser) that is used to remove and quantify the light ends of the fuels employing an online GC. The separator temperature is optimised from test to test, not only to fine-tune the mass balance but often also to keep the naphtha fraction, which would artificially lower the TLP cloud point in the gas phase as much as possible. A typical benchmarking test compares the performance of different catalysts at least at two to four different TLP cloud points. The temperature of each reactor is adjusted to the needs of the specific system loaded in it to match the target cloud point. After three days on stream, it is possible to automate the temperature adjustment of the reactors thanks to a continuous feedback loop between the myhte data warehouse and the test unit controlled by hteControl. The performance of the catalysts can be ranked as shown in Figure 4 panel A, and the data generated can be used