PTQ Q4 2023 Issue

4.74

34.00

4.72

Side reboiler unit steam

Bottom reboiler unit steam

33.50

4.70

33.00

4.68

4.66

32.50

4.54 4.60 4.58 4.56 4.62 4.64

32.00

31.50

31.00

30.50

281

282

283

284

285

Column bottom temp. (˚F)

Figure 8 DIB reboilers field testing results

temperature. Heat balance shifting between the side and bottom reboilers was necessary to improve the required total reboiler duty. In addition, the DIB bottom temperature needed to be reinstated. The calculated side reboiler duty already exceeded the original design maximum value. Field testing was arranged to check both reboiler per- formances. Objectives of the test were to identify the true maximum side reboiler duty and sensitivity between the DIB bottom temperature and shifted reboiler heat balance. The maximum steam rate was set as the ‘baseline’ value for the bottom reboiler. Steam rates for both reboilers were varied on a ‘step basis’. An infrared thermal imaging cam- era was used to verify the accuracy of temperature meas- urement. Measured reboiler steam rates and DIB bottom temperatures were gathered and plotted in Figure 8 . Field testing verified that the side reboiler duty could be increased further and the target DIB bottom temperature could be achieved. Increased side reboiler duty could circumvent a fouled section and increase the total reboiler duty. Case study 2: Feed point relocation Rebalancing and optimising between the side and bottom reboiler could increase the total reboiler duty and contribute to increasing the DIB feed rate. Nevertheless, this reboiler

optimisation alone could not meet the target DIB feed rate and RVP of the bottom product because the side reboiler duty reached the maximum limit. The required column inter- nal vapour and liquid traffic was still too high to be handled through fouled trays. Is there any other optimisation point to further reduce column vapour and liquid traffic? Further process evaluations were continued to meet the DIB targets. It was found that the existing feed point was not optimised at the latest feed composition and target degree of separation. The DIB was originally designed and constructed with three different feed points. These feed points are illustrated in Figure 6. Only the feed point at tray #21 was utilised historically. Meanwhile, feed points at trays #9 and #15 had never been used. Key ratio plotting was used as a major methodology to identify the optimum feed point. Key ratio is usually expressed as the mole frac- tion ratio of light keys to heavy keys in a semi-logarithmic scale chart. Non-optimum feed location can be visually identified through a constructed key ratio profile. Multiple key ratio plots with different feed points can find the opti - mum feed point. Since key ratio plotting shows only binary key component behaviours, non-key component compo- sition profiles through the column should be reviewed for multi-component distillation.² A DIB key ratio plot with tray #21 as the original feed point was constructed. Iso-butane (IC4) and normal butane (NC4) were defined as light key and heavy keys, respectively. Light and heavy key compositions through the column were obtained from a simulation model. The model was validated with base test run data. The key ratio graph with tray #21 as the original feed point is shown in Figure 9 . The figure disclosed that a substantial amount of rectification trays did not contribute to IC4/NC4 frac - tionation. On the other hand, all stripping trays were used for IC4/NC4 fractionation. The model predicted that an increased number of stripping trays could reduce column internal vapour and liquid traffic at the same degree of fractionation. Switching the feed point from tray#21 to tray #9 could simply convert ‘wasted’ rectification trays to extra stripping trays. The total number of trays was unchanged. Minimising

Feed at #21 tray

Column top

Column bottom

Figure 9 DIB key ratio plot

PTQ Q4 2023