October 2022

Special Focus: Plant Safety and Environment

An economic evaluation for SO_x emissions reduction from FCC regenerator flue gas using an additive

Sulfur oxides removal has been a global trend in the refining and petrochemical industry.

Sulfur oxides (SO_x) removal has been a global trend in the refining and petrochemical industry. SO_x reduction additives and wet gas scrubbers (WGSs) are the two most widely used methods to reduce SO_x emissions from fluid catalytic cracking (FCC) flue gas in addition to feedstock hydrorefining. SO_x reduction additives can remove SO_x from the FCC regenerator flue gas with low cost, high efficiency and flexibility. In theory, flue gas SO_x could be reduced to an extremely low level if the proportion of additives was not limited.

However, in a practical FCCU, refiners must consider how to achieve SO_x reduction in the most economical way. Based on the application of a proprietary SO_x-reduction additive^a in 25 different FCCUs with WGSs, the economic benefits of the application of the SO_x-reduction additive were evaluated and the most cost-effective ways to use the additive under different regeneration conditions were predicted.

The cost of SO_x additives vs. the cost of caustic in WGSs

The main reaction in WGSs is SO₂ + 2NaOH = Na2SO₃ + H₂O, so a linear relationship exists between the apparent consumption of caustic and sulfur dioxide (SO₂) removal in a certain range of SO₂ concentration. The efficiency of additives to capture SO₂ is determined by what is called the pick-up factor [(PUF): the mass of SO₂ captured by additive per unit mass]. Assuming that the unit price of caustic with 30 wt% NaOH is P₁ and that of the SO_x-reduction additive^a is P₂, then the break-even point of additive can be calculated as follows (Eq. 1):

1/PUF × P₂ ≤ 1/64 × 40 × 2/30% × P₁                                                    (1)
PUF ≥ 0.24 × P₂/P₁

Therefore, it would be more economical to use SO_x additives than to consume caustic in WGSs when the PUF of the additive was greater than 0.24 × P₂/P₁ kg SO₂/kg additive. It should be noted that the actual cost savings would be even higher since the simplified model did not consider sulfur trioxide (SO₃) capture—it is generally believed that the additive is more effective than WGSs in SO₃ capture—the benefits of sulfur recovery, the cost of treating high-salt wastewater and the WGS plant investment.

Prediction of PUF and maximum cost savings

The PUF is closely related to the properties of SO_x additives and the operating conditions of FCCUs. The additive properties lie in effective magnesium oxide (MgO) utilization, the ability to promote the conversion of sulfur (S) compounds to SO₃ in the regenerator, and the ability to reduce sulfate to hydrogen sulfide (H₂S) in the reactor. A good SO_x-reduction additive should also balance the high reactant accessibility and low attrition index.

The operating conditions are embodied in the following aspects. Flue gas excess oxygen content is the decisive factor to some extent because it determines the conversion of S compounds to SO₃ in the regenerator. The initial SO_x concentration determines the mass transfer force so that higher SO_x concentration results in a higher PUF. Increasing the catalyst circulation rate increases the availability of fresh metal oxides for SO_x pick-up and, therefore, reduces SO_x emissions. Lower regenerator temperatures tend to favor SO₃ formation, while good air distribution and mixing in the regenerator enhance SO_x pick-up. The FCC catalyst itself and the presence of a carbon monoxide (CO) promoter also affect the SO_x capture efficiency of the SO_x-reduction additive, to a certain extent.

To enhance the efficiency of flue gas SO_x pick-up, the authors’ company developed the enhanced RFS additive^a, which specifically improved the content of the key active component MgO. At the same time, the content of oxygen storage components were adjusted to improve the SO_x removal efficiency under the condition of low excess oxygen, including partial burn regeneration. In addition, the preparation process of the additive^a was optimized to maintain good attrition resistance and avoid adverse effects on SO_x removal efficiency and unit operation. The authors’ company completed industrial trial production in 2014 and the additive has been successfully applied in more than 40 FCCUs.

The concentration of SO₂ and excess oxygen in the regenerator are the two key factors affecting the PUF for the same grade of additives. Here, the PUF of the additive was fitted as functions of SO₂ concentration at the regenerator outlet under different excess oxygen conditions based on the application of the proprietary additive^a in 25 different FCCUs. It was found that these units could be divided into five categories according to excess oxygen content and regenerator form. The fitting of each category presents a good linear correlation, as shown in FIG. 1.

Based on this, the PUF of the additive under different operating conditions can be predicted. It was found that the PUF value increased with the increment of excess oxygen of the regenerator and outlet SO₂ concentration. To identify the cases where using additives is more economical than using caustic in WGSs, the difference between the two costs was expressed as the following equation (Eq. 2):

φ = 80/64/30% × α × ω(SO₂) × Q × P₁ – α×ω (SO₂)× Q × P₂/PUF                                       (2)

where:
φ = cost savings using additive over caustic, $/hr
α = SO₂ removal ratio using additive
ω(SO₂) = mass concentration of uncontrolled SO₂ in the original flue gas, mg/m³
Q = flue gas flowrate under normal conditions, 10⁶ Nm³/hr
P₁ = unit price of caustic with 30 wt% NaOH, $/kg
P₂ = unit price of SO_x reduction additive, $/kg

Combined with the calculated PUF values under different operating conditions in FIG. 1, the maximum value of φ can be obtained.

FIG. 1. The relationship between the PUF of the proprietary and the regenerator outlet SO2 concentration under different excess oxygen conditions.

Three FCCUs are listed as examples in FIG. 2: two units in the CNOOC Refinery Co. Ltd. Huizhou refinery (HZ-1 and HZ-2) and one in the Sinopec QD Refining & Chemical Co. Ltd. (QD). The excess oxygen of HZ-1 is greater than 4%, and the excess oxygen of QD is between 2.5% and 4%, while HZ-2 adopts partial burn regeneration with excess oxygen of less than 0.1%. Details of the three FCCUs are described below.

FIG. 2. The expected economic benefits of the three refineries.

Using the fitting relationship in FIG. 1 can predict the optimal economic benefit of the three units, as shown in FIG. 2. It is predicted that the maximum economic benefit would be about $13/hr and $80/hr after the application of the additive^a in the HZ-1 and HZ-2 units while the removal ratio is about 80% and 47%, respectively.

Similarly, the QD unit is expected to save about $79/hr with a removal ratio of about 75% after the application of the additive. It should be noted that when the additive is initially added, the PUF is very high due to the high initial concentration of SO₂ in the regenerator; however, the insufficient amount of additive cannot remove enough SO₂ so the removal ratio is not high. In this case, it is more cost-effective to capture this part of SO₂ with the SO_x-reduction additive than with caustic in WGSs.

Once the additive amount exceeds a certain value when the excess oxygen is not high enough, the removal of SO₂ requires much more additive supplement than before—in this case, the cost of using additive has skyrocketed and is not cost-effective. This is reflected in the rapid decline of the curve in FIG. 2. The capture efficiency of additive is very low under marginal conditions. In incomplete regeneration, some S compound exists in the form of H₂S and COS, which are difficult to transform and capture by additive. These S compounds are transformed into SO_x in a CO boiler, so the refinery must use post-treatment like a WGS to meet the requirements of local environmental laws and regulations.

DETAILED ANALISYS OF APPLICATION CASES

Case 1: HZ-1 unit in the CNOOC Refinery Co. Ltd. Huizhou refinery

The HZ-1 is an MIP unit with a designed annual processing capacity of 1.2 MMtpy. The mass fraction of sulfur in the feedstock ranged from 0.35 wt%–0.41 wt%. The regenerator adopted full burn operation with a pre-combustor. The total air volume to the regenerator was 0.13 × 10⁶ Nm³/hr and the excess oxygen in the flue gas was 4.8%–5.5%. The mass concentration of SO₂ in the flue gas at the entrance of the WGS was 300 mg/m³–400 mg/m³ before the additive^a was used. After using the additive, SO₂ was reduced to less than 20 mg/m³ when the additive inventory was less than 2%, and the removal ratio was about 95% (FIG. 3).

FIG. 3. The SO₂ concentration in the flue gas during the additive application in the HZ-1 unit.

The caustic consumption decreased from 150 kg/hr to about 10 kg/hr, which is consistent with the removal ratio of SO₂, and sometimes even stopped filling caustic. Total dissolved solids (TDS) in wastewater decreased from more than 10,000 mg/kg to 1,300 mg/kg with the addition of the additive^a (FIG. 4). The SO_x in the flue gas was greatly reduced after being captured by additive, and the amount of caustic was reduced, thus reducing the salt content of Na₂SO₃ and Na₂SO₄ in water. TDS reduction not only benefits wastewater treatment and slows down equipment corrosion, but it also reduces the risk of system scaling, which is conducive to long-term stable operation of the unit.

FIG. 4. Consumption of caustic and TDS in liquid waste during the additive application in the HZ-1 unit.

Considering that the additive has a good effect on TDS and solves the problem of plume trailing subsidence from the WGS caused by SO₃ aerosol, it was used to reduce the SO_x of the flue gas before the entrance of the WGS to a very low level, rather than only reducing by 80% for the consideration of optimal economic benefit. Assuming that the price of caustic with 30 wt% NaOH was $173/MMt and the cost of consumption of the additive was considered, the total savings was about $4/hr in this case, which is still economical. This value is also consistent with the predicted value in FIG. 2.

Case 2: HZ-2 unit in the CNOOC Refinery Co. Ltd. Huizhou refinery

The designed annual processing capacity of the HZ-2 FCCU in the Huizhou refinery was 4.8 MMtpy with an S content in the feedstock between 0.35 wt% and 0.39 wt%. The unit adopts a two-stage, coaxial up and down incomplete regeneration process with excess oxygen of less than 0.1%. The total air volume to the regenerator was 0.32*10^6 Nm³/hr and the CO volume concentration at the outlet of the regenerator was 3.2%–4.5%. When no additives were used, the mass concentration of SO₂ in the flue gas at the entrance of the WGS was approximately 1,400 mg/m³–1,800 mg/m³. The SO₂ removal ratio was 55% when using 3.8% additive^a in the inventory, and the consumption of caustic decreased from 2,200 kg/hr to 1,100 kg/hr (FIG. 5), saving about $72/hr. This value is close to the predicted optimal economic operation in which the maximum economic benefit was $80/hr with a 47% SO₂ removal ratio.

FIG. 5. Consumption of caustic during the additive application in the HZ-2 unit.

Case 3: QD unit in the Sinopec QD Refining & Chemical Co. Ltd

The QD unit is an MIP-CGP unit with a designed processing capacity of 2.9 MMtpy. The feedstock oil is mainly hydrogenated VGO with an S content of 0.35%–0.40%. The regenerator adopted full burn operation with a pre-combustor. The total air volume to the regenerator was 0.336 × 10⁶ Nm³/hr and the excess oxygen in the flue gas was 2.1%–3.2%. The mass concentration of SO₂ in the original flue gas at the entrance of the WGS was about 600 mg/m³ without additives. The SO₂ concentration decreased rapidly to about 250 mg/m³ with the addition of the additive^a at the end of July. When the additive was not added due to dredging pipeline and catalyst storage tank replacement in August, the SO₂ concentration came back to about 400 mg/m³. After the normal addition was resumed in September, the mass concentration of SO₂ decreased rapidly again and stabilized to about 100 mg/m³. The results showed that the removal ratio of SO₂ reached 80% when the sulfur transfer agent was added at 2% of catalyst inventory.

Under the same pH value of controlled circulating liquid in the WGS, the total consumption of caustic decreased gradually from about 800 kg/hr to about 130 kg/hr–200 kg/hr with the addition of the additive^a, which was consistent with the removal ratio of SO₂ in flue gas, as shown in FIG. 6. The economic benefit is about $71/hr, almost the same as predicted. In addition, the problem of plume tailing from the scrubber was well controlled. The TDS in the circulating liquid decreased from about 4% to about 1%. With the addition of the additive, the mass concentration of chemical oxygen demand (COD) decreased from 100 mg/l–150 mg/l to 20 mg/l–30 mg/l. This is also attributed to the reduction of SO₂ in the original flue gas and the reduction of sulfite and its corresponding COD in water.

FIG. 6. The SO₂ concentration in the flue gas and consumption of caustic during the additive application in the QD unit.

An economic model for predicting maximum cost savings using the proprietary additive^a over caustic under different operating conditions was established in FIG. 7.

FIG. 7. Maximum cost savings using the additive over caustic.

Assuming that a mass concentration of uncontrolled SO₂ in the regenerator flue gas was 1,000 mg/m³ and the price of caustic with 30 wt% NaOH was $173/MMt, the additive would provide a maximum cost savings of $0.72/MMt of feedstock, with excess oxygen greater than 4%, $0.41/MMt of feedstock with excess oxygen between 2.5% and 4%, and $0.25/MMt of feedstock with excess oxygen between 0.5% and 2.5%. If the concentration of uncontrolled SO₂ was higher, greater economic benefits would be obtained.

Takeaway

The PUF of the additive has been fitted as a function of SO₂ concentration at the regenerator outlet and good correlations have been established based on the application of the additive in 25 different FCCUs under 5 different operating conditions. To identify the cases where using the additive is more economical than using caustic in a WGS, a model was built to show the difference between the two costs, and the economic benefit of the additive was evaluated. The most cost-effective way to use the additive in FCCUs with a WGS has been predicted under different regeneration conditions. Three FCCUs are listed as examples to demonstrate the accuracy of the model. The actual total cost savings from the application of the additive are consistent with the prediction of the model. The additive also has the benefit of reducing total dissolved solids (TDS) and chemical oxygen demand (COD) in liquid waste and solves the problem of plume tailing caused by SO₃ aerosol in scrubber stacks. HP

NOTES

  ^a Sinopec’s RFS09

The Authors

Related Articles

From the Archive

Comments

Comments

{{ error }}

Add Comment

{{ comment.comment.Name }} Editor • {{ comment.timeAgo }}

{{ comment.comment.Text }}