July 2021

Process Optimization

Analyze pressure loss during FCC standpipe catalyst transfer

The fluid catalytic cracking unit (FCCU) is the primary conversion equipment for modern refineries to obtain clean transportation fuels, along with high-grade petrochemical feedstock.

The fluid catalytic cracking unit (FCCU) is the primary conversion equipment for modern refineries to obtain clean transportation fuels, along with high-grade petrochemical feedstock. With increasing processing capacity, however, standpipe catalyst problems have become a bottleneck, restricting long cycle operation.¹

The ideal operating mode for industrial FCC standpipe is dense-phase fluidized solids flow, with a range of catalyst density of 500 kg/m³–600 kg/m³. This operating mode not only provides high pressure drop for the catalyst circulation, but also has a strong sealing function, preventing gas leakage from the slide valve.^2,3 However, catalyst flow patterns usually vary with operating conditions, standpipe geometry structure, catalyst properties, aeration flowrate, etc. It is difficult to maintain uniform fluidity. Erratic flow patterns often affect pressure buildup in the standpipe.^4,5

Matsen et al.⁶ proposed that friction formed in the packed-bed flow was the main reason for the pressure loss. According to the classic single-flow and double-flow pattern models proposed by Leung et al.,⁷ catalyst flow patterns in the standpipe can be conveniently determined. However, these models did not consider the effect of the slide valve opening value on catalyst flow patterns, and FCC engineers still could not explain some pressure-reversal phenomena in the industrial standpipe using the theory.

This article analyzes the influencing factors of pressure loss on a 1-MMtpy industrial FCCU by measuring the dynamic pressure in the regenerated standpipe under different operating conditions. A half-pipe flow model is established and a linear relationship equation between the catalyst circulation rate and the standpipe pressure loss is proposed, which can be used to guide the operating adjustment for the industrial FCC standpipe.

Experimental setup

The schematic diagram of the 1-MMtpy industrial FCCU is shown in FIG. 1. The riser terminal is connected with the vortex separation system (VSS). The regenerator contains a combustor, the dilute phase pipe and the disengager. The inlet pressure of the regenerated standpipe, p_i, is shown in kPa. The pressure above the regenerated slide valve is p_s. The outlet pressure of the regenerated slide valve, p_o, is shown in kPa. The pressure drop of the slide valve is expressed in kPa as Δp = p_s – p_o.

The key operating parameters are listed in TABLE 1. The corresponding catalyst circulation rate, G_s, changed from 435.4 kg × m^–2 × s^–2 to 857.4 kg × m^–2 × s^–2 between Case 1 and Case 4. The ratio of catalyst to oil was calculated according to the heat balance between the reactor and the regenerator system. The opening value of the regenerated slide valve increased from 45% to 65% with the increase in processing capacity, which was very large and not conducive to the operating adjustment. For Case 3 and Case 4, the processing capacity and reaction temperature were lower than in other cases. To ensure the gas velocity in the inlet of the cyclone separators and the VSS under an appropriate scope, the pressures of the reactor and regenerator were reduced to 165 kPa and 160 kPa, respectively.

Regenerated standpipe structure

The schematic diagram of the regenerated standpipe is shown in FIG. 2. The standpipe consists of two inclined pipes. The angle between the upper inclined pipe and the vertical line is 10°. The upper inclined pipe encompasses two parts with inner diameters of 760 mm and 630 mm, respectively, and a total length of 18.2 m. The top wider part is used to degas bubbles. The angle between the lower inclined pipe and the vertical line is 45°, with a length of 2.3 m and an inner diameter of 630 mm.

C1–C11 are the 11 axial cross-sections of the standpipe. One aeration nozzle is installed on each cross-section, and the angle between the nozzles and the wall surface is 30°. Nozzles on C2, C4, C6 and C8 are located at the bottom inclined pipe. The rest of the nozzles are installed on both sides, and the angle with the central busbar is 60°. The aeration gas above the regenerated slide valve is nitrogen with a pressure of 1.4 MPa and a normal temperature. The aeration gas below the regenerated slide valve is steam with a pressure of 1 MPa and a temperature of 265°C. The aeration flowrate is controlled by the flow limiting orifice with a diameter of 2 mm–5 mm.

TABLE 2 shows the analysis of material properties of the equilibrium catalyst at 20°C. Pressure measurement was carried out with the selected aeration nozzles (C1, C2, C3, C6, C9 and C10) at the same time under different operating conditions. Measuring frequency and time were 1 Hz and 60 sec, respectively.

Results and discussion

In looking at the pressure profiles of the standpipe, FIG. 3 shows the axial pressure profiles of the regenerated standpipe from Case 1 to Case 4. The p_i depended on the regenerator pressure and the static pressure of the dense-phase bed in the regenerator, with little change under the different operating conditions. The axial pressure increased gradually along the standpipe, but the pressure between the C6 and C9 cross-sections was greatly affected with G_s.

When G_s was 857.4 kg × m^–2 × s^–1, the pressure at the bottom standpipe increased and had a maximum pressure of 255.6 kPa at the C9 cross-section. In the other three operating conditions, the pressure below the C6 cross-section decreased. When G_s was 435.4 kg × m^–2 × s^–1, pressure at the C9 cross-section had a minimum pressure of around 210 kPa. The p_o below the slide valve ranged in pressure from 180 kPa–223 kPa, which was mainly affected by the pressure drop of the riser, and increased with the increase of G_s.

FIG. 4 shows the dynamic pressure at the C6 and C9 cross-sections under different operating conditions. When G_s was 857.4 kg × m^–2 × s^–1, the mean pressure at the C6 cross-section was less than that at the C9 cross-section, as shown in FIG. 4A. In the other three operating conditions, the pressure at the C6 cross-section was greater than that at the C9 cross-section. Moreover, the pressure difference between the C6 and C9 cross-sections increased gradually with the decrease of G_s. When G_s was 435.4 kg × m^–2 × s^–1, the mean pressure difference increased to 30 kPa.

In analyzing the flow patterns in the standpipe, FIG. 5 shows the single and double flow pattern modes in the FCC standpipe proposed by Leung et al.⁷ Generally speaking, the sliding speed of the gas solids at the top standpipe is large, and the catalyst flow pattens are dense-phase fluidized solids flow. The fluidized gas volume in the bottom standpipe is reduced because it is compressed. If the aeration flowrate is approximately equal to the decreased gas volume, then the axial pressure gradually increases along the standpipe, as shown in FIG. 5A.

If the aeration flowrate is insufficient, then the pressure gradient along the standpipe decreases. The catalyst flow pattern on the top standpipe is in a fluidized state. Catalyst voidage, ε, decreases gradually, and the axial pressure increases continuously. However, in the bottom standpipe, when ε is less than the incipient fluidization voidage, ε_mf, the catalyst flow patterns change to transitional packed-bed flow or packed-bed flow, and the friction loss increases rapidly. In the transitional packed-bed flow, ε continues to decrease but is larger than ε₀. The friction loss is smaller than the static pressure, and the axial pressure continues to increase. The pressure gradient along the standpipe is negative but lower than that of the dense-phase fluidized solids flow, as shown in FIG. 5B, Curve 1. In the packed-bed flow, ε drops to ε₀, and the frictional loss is larger than the static pressure. The axial pressure gradually decreases, and the pressure gradient becomes positive, as shown in FIG. 5B, Curve 2.

Industrial FCC standpipes consist of vertical pipe and inclined pipe. Catalyst flow patterns in a standpipe are complex. Generally speaking, catalyst flow patterns in the top standpipe are in a fluidized state. When the aeration flowrate is insufficient in the middle pipe, the catalyst flow patterns change to the transitional packed-bed flow. The pressure loss increases and the pressure gradient decreases, which results in a decreased catalyst circulation rate. The actuator of the slide valve will increase the opening value to raise the G_s and ensure the stability of the reaction temperature. However, the height of the catalyst sealing column above the slide valve decreases with the increase of the opening value, which may result in gas leakage from the slide valve into the riser. The pressure gradient will become positive, as shown in the curve in FIG. 6C. However, if the catalyst sealing column is high enough, then the gas at the bottom standpipe will only flow upward and the pressure gradient is still negative, as shown in the curve in FIG. 6D.

Catalyst density above the slide valve

For the parallel FCCU, Eq. 1 is used to calculate the catalyst density above the slide valve:

      A = (6.29Fs × 10^–2)/[C_s(∆p × ρ)^0.5]                                                                (1)

where:

      A = Actual catalyst flow area, cm²
      A_o = Full open area of the slide valve plate = 936.95 cm²
      F_s = Catalyst circulation rate, kg/hr
      ρ = Catalyst density above the slide valve, kg/m³
      Δp = Pressure drop of slide valve, kPa
      C_s = Flowrate coefficient.

For the single acting slide valve with a throttling cone, C_s equals 0.85. According to TABLE 1 and Eq. 1, catalyst density above the slide valve under different operating conditions can be calculated as shown in FIG. 7.

Within the operating conditions, catalyst density above the slide valve varied from 247 kg/m³–461 kg/m³, which was much less than ρ_mf . In Case 1, catalyst density above the slide valve was around 461 kg/m³, which had a strong sealing capacity to prevent gas leakage from the slide valve. However, catalyst density decreased gradually with the reduced G_s. In Case 4, catalyst density dropped to only 247 kg/m³. Catalyst flow patterns transformed into the lean-phase fluidized solids flow, and the catalyst sealing function disappeared. Gas leakage was the fundamental reason for the bottom standpipe pressure reversal.

According to the double flow pattern model shown in FIG. 3, pressure in the bottom standpipe decreased, which suggested that catalyst density above the slide valve would be close to ρ_mf. In reality, however, catalyst density above the slide valve from Case 2 to Case 4 was much less than ρ_mf, as shown in FIG. 7. This indicated that the gas leakage flow pattern had a similar pressure distribution to the packed-bed flow in the industrial FCC standpipe.

Pressure loss in standpipe

The average pressure difference between the C6 and C9 cross-section was defined as the standpipe pressure drop loss in kPa (Δp_loss = p_C6 – p_C9). According to the average value of dynamic pressure in FIG. 4, the relationship between Δp_loss and G_s was established as shown in FIG. 8. Within the operating conditions (G_s = 435.4 kg × m^–2 × s^–1 to 857.4 kg × m^–2 × s^–1), there was a linear relationship equation between Δp_loss and G_s, Δp_loss = –0.07554G_s + 58.51208.

Takeaway

The dynamic pressure in the regenerated standpipe was measured at different operating conditions in a 1-MMtpy FCCU. Process parameters and pressure distribution in the standpipe were used to determine the catalyst flow patterns. The main conclusions were summarized as:

1. Different catalyst flow patterns were present in the industrial FCC standpipe. Dynamic pressure characteristics could be adopted to determine the catalyst flow patterns.
2. Catalyst flow patterns in the standpipe varied with the catalyst circulation rate. Under the low catalyst mass rate, gas leakage appeared from the slide valve into the riser, which was the fundamental reason for the axial pressure reversal in the bottom standpipe.
3. Gas leakage reduced the standpipe pressure drop. A linear relationship equation between G_s and pressure loss was expressed as: Δp_loss=-0.07554G_s+58.51208, which could be used to calculate the catalyst mass rate. HP

ACKNOWLEDGMENTS

The authors acknowledge financial support by the 2021 Technical Talents Innovation Foundation Project of CNPC and the Natural Science Foundation of China.

LITERATURE CITED

1. Wolschlag, L. M., “Upgrade FCC performance,” Hydrocarbon Processing, September 2010.
2. Knight, J., “Maximize propylene from your FCC unit,” Hydrocarbon Processing, September 2011.
3. Chen, Y.-M., “Recent advances in FCC technology,” Powder Technology, 2006.
4. Raymond, W. M., “Troubleshooting FCC standpipe flow problems,” Catalgram, Fall 2009.
5. Phillio, K., “New and old equations tie together 75 years of FCC standpipe experience,” AFPM Q&A Technology Forum, 2016.
6. Matsen, J. M., “Flow of fluidized solids and bubbles in standpipes and risers,” Powder Technology, 1973.
7. Leung, L. S. and P. J. Jones, “Flow of gas-solid mixtures in standpipes: A review,” Powder Technology,1978.

The Authors

Related Articles

From the Archive

Comments