PTQ Q3 2025 Issue

8 0 9 0 100

1,600

Relief load (kg/hr) MeOH in sump (wt%) Pressure (kPag)

1,400

1,200

30 20 40 50 60 70

1,000

800

600

400

200

120

150

180

210

240

270

300

330

Time (min)

Figure 6 Methanol distillation column dynamic relief results

Figure 5 ). The relief scenario investigated was a total loss of cooling duty in the overhead condenser due to the loss of cooling water. The basis for the preliminary UBH calcu- lation can be summarised as follows: u Overhead condenser duty is set to zero. v Reboiler duty during relief is also set to zero because the boiling point of water at relief pressure is higher than the LP steam temperature. In other words, credit was taken for reduced LMTD based on constant bottoms composition (~99% water), resulting in zero reboiler duty. The preliminary relief load by UBH was calculated to be 0 kg/hr (no relief) due to the lack of heat input from the reboiler. This prompted the engineer to investigate the assumption of constant bottoms composition. The UBH relief load was re-calculated with the bottoms composi- tion assumed equal to a) the column feed stream and b) the reflux stream. Both options resulted in significant relief loads but could not easily be validated. A dynamic simulation was prepared to estimate the relief load once it had been shown to be highly dependent on the composition of the sump inventory at relief. The basis for the dynamic simulation can be summarised as follows: u Overhead condenser duty set to zero, initiating the relief scenario. Reflux to the column continues running for some time but will be lost once the reflux drum runs dry. v Reboiler modelled as a UA exchanger, taking credit for reduced LMTD as column bottoms temperature rises. Results of the dynamic simulation of this scenario are shown in Figure 6 and compared against the UBH relief

loads in Table 1 . The dynamic relief load was found to be 1,517 kg/hr, invalidating the ‘no relief’ result from the pre- liminary UBH calculation. The difference can be attributed to the change in the composition of the sump liquid at relief conditions, which was found to be 65.5 wt% methanol dur- ing peak relief. As a result, the boiling point of the sump liquid at relief conditions was less than the LP steam temperature, and reboiler duty was high enough for relief to occur. The reason for the shift in bottoms composition can be explained as follows: • Upon loss of condenser, system pressure begins to rise as vapour accumulates in the column without condensing. • Reflux liquid (~99% methanol) continues to be pumped into the top of the column while overhead product flow continues until the overhead drum runs dry. In this system, the inventory in the overhead drum was very large relative to the column, and the reflux ratio was high, so the reflux was able to continue for hours. As column pressure rises, bottoms temperature rises with it, and reboiler duty is reduced by exchanger ‘pinch’ (reduced LMTD). As the boil-up rate falls relative to the reflux rate, methanol content in the sump liquid increases. • Over time, the sump liquid composition shifts such that its bubble point at relief pressure leaves a margin below the LP steam temperature. In this case, the typical UBH assumption of constant bottoms composition was not valid mainly because of the large inventory of liquid in the overhead drum relative to the volume of the distillation column, which allowed continued reflux during the upset scenario to impact the composition of the sump liquid. Further sensitivity analysis shows this effect was exacerbated by a high reflux ratio and a large difference between operating pressure and PSV set pres- sure, which provided the time required for the composition to shift. Recommendations The UBH method remains a valuable tool for column relief load analysis, especially for grassroots design of common

Methanol distillation column dynamic and UBH relief results

Bottoms composition basis Calculation method Bottoms composition (wt% MeOH)

Normal

Feed

Reflux Dynamic

UBH

Dynamic

0.1

60.0

99.0

65.5

Relief load (kg/hr)

2,550

8,340

1,517

Table 1

PTQ Q3 2025