Reproducibility of high throughput hydrocracking catalyst testing
Three pretreatment and two cracking catalysts were screened, with their performance benchmarked, and the differences were found to be statistically significant
Giada Innocenti, Jochen Berg, Kai Dannenbauer, Felix Hilpert, Ioan-Teodor Trotus, and Jean-Claude Adelbrecht hte
T he benchmarking of hydrotreating and hydrocrack- ing catalysts can be challenging due to similarities in performance among the commercial materials used. To rank catalysts provided by different vendors in a mean- ingful way, it is important to either load the reactors in repli- cates or rely on experience gathered over the years to know how much variability is possible for each target parameter. To build know-how and understand the statistical mean- ing of the results generated by multiple repetitions, as in this case, it is crucial to have a clear understanding of the possible sources of error in each measurement. The term ‘error’ in this publication does not refer to gross errors like feed channelling or incomplete catalyst wetting due to the incorrect loading of reactors, human error, or casual error. Instead, it refers to those intrinsic errors inherent to an experiment. These errors affect each test and were found to be com- parable among each unit used at hte. This statistical proof- of-concept study is important because it provides a good
indication of what types of differences in activity can be considered significant. Experimental setup Up to 16 combinations of catalyst and two different feeds can be tested in parallel with hte’s high throughput trickle bed units (X4500). It is possible to load up to 10 mL of cat- alyst per reactor while processing a maximum of 30-50 L of feed per week (see Figure 1 ). The same liquid and gas flow is fed to all reactors by imposing the same pressure drop ahead of each reactor using a capillary distribution system. Additionally, for the unit employed in the experi- ments reported here, each liquid capillary can be individu- ally heated to even out minor flow differences. The simultaneous exposure of different catalyst loads to the same liquid and gas feed allows for the collection of a large amount of data in a very short time, enabling confident decision-making. If key reaction parameters like temperature, liquid hourly space velocity (LHSV), and/or
PT1 PT1 PT1 PT1 PT1 PT1
PT2 PT2
PT1 PT1 PT1 PT1 PT1 PT1 PT3 PT3
HC1
HC1 HC1 HC1
HC2
HC2 HC2 HC2
1
2
3 4
5
6
7
8 9
10 11 12 13 14 15 16
Sampling position
Figure 1 Reactor configuration. Dashed red lines represent the heater size. All the hydrocracking catalysts had individual heaters, while each pair of pretreat catalysts shared heaters as enclosed in dashed lines. For simplicity, gas-liquid separa- tion, which occurs before sampling, is not shown here
101
PTQ Q3 2025
www.digitalrefining.com
Powered by FlippingBook