AFPM Q&A—3 (Conclusion): Discussion turns to crude processing

What is your best practice and expected efficiency in removal of calcium naphthenates in higher calcium crudes? Has anyone experienced severe fouling in downstream unit equipment (i.e., vacuum heater, coker heater) when processing high calcium crudes? What are the concerns when processing resid containing high calcium content?

Parise: This question is about calcium naphthenate and calcium in crude, which is considered a metal for this discussion.

One of the key things you need to do when managing calcium naphthenate is to treat the crude unit as a total system. You are not looking at just treating the desalter; you are also looking at the overhead system, the preheat exchanger from a point of view of fouling, and also the waste water treatment plant impact as you are going to be trying to drive some of those calciums out of the crude oil and into the waste water treatment plant.

You will also be using chemicals. So, you will have an impact on the waste water treatment plant that you will have to address as well.

First, a metal-removing agent is required to react with the calcium naphthenate and form soluble calcium salt. It is acidic in nature. It is typically injected either into the crude or into the washwater upstream of the desalter.

It is absolutely critical that you optimize and make sure you have the best desalter operation. That means harassing your vendor on the unit to ensure that he has properly tested and defined your desalter capabilities, mixed valve studies, washwater studies, and level control studies to make sure that it is running at its best capability.

Most of the technologies used today to address the calcium naphthenates are simple carboxylic acids: acetics, citrics, and glycolics. Each of these has a particular drawback associated with it. Fouling tendencies are one factor, as well as corrosion.

We are aware that there are new proprietary chemicals on the market that are starting to show some success. Typically, one of the downstream impacts you are going to see is lower washwater pH. Sometimes you will need to do a total system approach and apply a corrosion inhibitor to help protect the effluent exchangers and your piping downstream.

Also, you may need to use an antiscalant to control any of the fouling evident in the feed-effluent exchangers. You can also see downstream fouling in the heaters due to high calcium naphthenate crudes.

We have seen some reports from manufacturers of increased furnace fouling. We are aware of several occasions, globally, where furnace fouling is not being observed, even over several years of processing high calcium naphthenate. This addresses one of the other questions.

People ask if we have seen furnace fouling. We have seen many sites that have run for quite some time with no furnace fouling.

Other parts of the question talked about processing resid containing high calcium content. We are aware of at least one location using a glycolic-based chemistry. What appeared at that site was a significant increase in preheat fouling. Also seen was some higher crude overhead corrosion necessitating more neutralizers in the overhead. There was also impact downstream on the FCC catalyst addition rates, particularly the equilibrium catalyst health. There was also some higher ash content in the FCC decant oil, requiring the use settling acids.

Duggan: [In a previous question] we were talking about using acids for amine removal. Now we are using acids for calcium removal. I thought it would be useful to point out a few of our experiences with some other acids because we have either been personally involved in or have monitored applications with these other acids.

Acetic acid: One of the biggest issues we have seen with acetic acid in applications is that it can generate a lot of COD [chemical oxygen demand] that is going to end up in your waste water treatment plant; so that is a concern.

Increased overhead iron solubility is probably [an] even bigger [problem] depending on corrosion in your overhead system. This shifts the solubility perimeter of iron sulfide and drives more iron into the water solution. The result is that you do not leave behind iron sulfide as a passive scale, which helps to reduce overhead system corrosion.

Systems like that see general acid corrosion rates exceeding 100 mils/year (mpy) compared with traditional oil that does not have a lot of acetic acid. You will see overhead corrosion rates of less than 5 mpy. It is hard to control that type of corrosion with traditional corrosion inhibitors.

Citric acid: The biggest negative is that its calcium citrate solubility is very poor. There are other side effects of citric acid, but the biggest one is that you can foul up your desalters in a very short period. That can lead to unplanned outage, which can be significant.

Oxalic acid is the same. It has very low solubility. If you look at the literature on calcium oxalates, it is not very soluble in water at all. So, [when] you heat that up to 250° F., precipitation is likely to occur.

I think I covered [earlier] some of the issues with sulfuric acid. You can make sulfate salts that foul the desalter and then [have] high SOx corrosion. To try to get around the corrosion issue, we designed our Excalibur amine contamination removal program to minimize those side effects.

We have had great experience with this program over the last 8 years with more than 20 applications worldwide where we have been doing it for calcium removal and others, like amine removal. I think what is interesting is that we actually see improved COD that is going into the waste water, not desalters. Even though all of these carboxylic acids used are going to generate, theoretically, a higher COD in the desalter water that goes to the waste water plant, you get improvements with these acids in reducing the phenols and other oils that are carried under. You actually end up with a net gain.

These acids were designed to have minimum overhead corrosion impact. It is fair to say that any acid you put in the system that takes bases out of the desalters is going to increase the neutralizer demand. In the overhead system, you cannot ignore the potential for any acid to do that. If that happens, you use the ionic model to guide you, in terms of how you are going to neutralize those additional acids in that additional acidity in the overhead system in place of the bases that you took out.

Now it is reasonable to admit that we have had one case with significant preheat fouling. It was manageable. I will not go into all the details. You can manage it with an offline acid washer or with dispersant antifoulant-type chemistries. But it is important to understand that once again, all acids suffer from this same potential problem. This particular case has very high calcium levels and, therefore, very high levels of acid to be able to bring the calcium out of the crude oil and into the brine water.

As a result, this was an extreme case, one that stands out among the norms for these treatment programs. However, all acids will suffer from this problem because, if you put this calcium into the brine water and carry a portion of that as your bs&w downstream, when that water evaporates in the preheat train, the calcium salts have nowhere to go. They have to deposit. The extent to which that is going to affect you really depends on your refinery, heat transfer, area availability, and your current levels of fouling and heat-transfer restrictions.

As I mentioned, you can actually mitigate this calcium fouling if you have the opportunity to put it offline. Do an acid wash to clean it out. It is very easy. This calcium is very water-soluble in acid conditions. Even dispersion chemistries have been proven in industry practice to function.

Gianzon: We started processing high-calcium naphthenate crude on one of our big refineries in Louisiana. At first, we were concerned about fouling in the convection section and upper radiant section of the coker heater from calcium naphthenate.

We initially set a very low target, as far as calcium concentration in the resid, and we were able to achieve that by using one of the panelists' chemicals. We found out that the fouling of calcium naphthenate, unlike sodium, is noncatalytic. We are slowly increasing calcium targets and reducing chemical usage.

Our pilot plant testing shows that the calcium is uniformly distributed on the heater, so it does not accumulate in the upper convection and lower radiant section as we previously thought.

Tariq Malik (Egyptian Refining Co.): Could you identify the country, source, or field of the crudes that have calcium naphthenates in them?

Gianzon: It comes from Chad.

Duggan: Some North Sea crudes are high in calcium. If you look around in the industry, you will see that there are other crudes that are fairly high in calcium. Some of the new production coming from offshore Brazil is reported to be pretty high in calcium, but I have not seen any firsthand data on that.

Sam Lordo (Nalco Co.): Tariq, some of the West African crudes, besides the one identified as Chad, are known to have elevated levels of calcium.

Jitendra Chaudhari (Reliance Industries Ltd.): Like calcium, do you have a program for arsenic, lead, or mercury because what comes along with the synthetic crudes are bitumen crudes? Arsenic, mercury, lead, and those kinds of poisons: How do you remove them from within the desalter?

Sam Lordo (Nalco Co.): There is some work being done. One of the major oil companies has already patented some processes for production-type removals of both arsenic and selenium and a little bit of mercury. I imagine you will start to see the impacts coming out of Southeast Asia and those crudes that are associated with it, but I am not at liberty to say at this time.

James Noland (Dorf Ketal Chemicals LLC): We agree with all of the acid drawbacks that were presented, but our experience indicates that there is one additional problem when you have high calcium crudes processed with high H₂S crudes as well.

You form an iron naphthenate layer in the desalter, and it creates a lot of desalter problems. We have a chemistry that mitigates that problem and prevents that rag layer from forming. Also, we have work going on right now in which we are commercially trialing some chemistry that is nonacid based. We are getting some good results. It does not quite get the same calcium removal as the traditional programs. It is more like in the 80% removal range, but it is very promising.

What are the impacts on coker operation (yields, capacity, energy, coke quality) of FCC slurry oil in the feed?

Gianzon: FCC slurry is very aromatic; so it acts almost like a recycle. From our experience, the incremental FCC slurry on a Conradson carbon residue basis has a much higher coke yield than on vacuum resid. The coke yield is approximately two times higher on a Concarbon basis vs. what we normally get from vacuum resid. The yield will depend on the pressure and temperature at which your unit is operating.

FCC slurry has the same boiling rate as heavy coker gas oil. It goes to a drum where most of it is flushed and then ends up in heavy coker gas oil (HCGO) section. If you are sending your HCGO product to your cat unit at a higher percent, let's say above 10% to 15%, and then the slurry can cycle up between the coker and FCCU units.

One of our units makes anode-grade coke. In order to meet anode-grade specification, we mix in 10% to 15% FCC slurry in the feed to meet the anode-grade metal specs and prevent shot coke formation. FCC slurry is known to help furnish run length. You are processing FCC slurry; you back out resid.

Ghosh: I have a different experience.

We have two cokers, which operate with some amount of FCC slurry oil in the feed. In one, we are processing 3-4% slurry oil with feed. In the other, we are processing 10-12%. While we operate with 3-4% slurry oil, we do not find significant change in the coke production. But with greater than 10%, we can expect higher coke production, as high as 4 wt % additional coke over a no-slurry case.

From the unit capacity point of view, the drum volume may be limiting because the coke is porous. From an energy perspective, as Gary [Gianzon] already explained, the heater duty will increase because most of the slurry oil will remain in the vapor phase. The advantage of slurry oil is that it will keep the asphaltenes into solution, thereby potentially reducing the coking in the furnaces.

The coke quality: There will be less of a tendency for shot coke formation because of the slurry oil. Then, we can get needle structure on the coke. The other aspect is that more FCC slurry will increase the silica in the coke, which may be disadvantageous if we are going to produce anode-grade coke or coke for needle-grade due to the limitation on silica in these types of coke.

S. Jeyavel Sinha (Reliance Global Management Services (P) Ltd.): Processing the FCC slurry prevents coking inside the coker heater, which generally seems to increase the run length of the heater.

At the same time, there are a couple of case reports indicating that they have seen the FCC slurry condense the catalyst fines where it has also affected the heater length. So is there any experience from the panel on the adverse effect caused by higher catalyst fines in the FCC slurry oil?

Ghosh: We did not have any bad experiences.

So far, we have been operating this coker for the last 5 years with the FCC slurry going to the feed. Of course, we have a slurry filter in the FCC itself, which takes care of some of the bs&w.

We are also monitoring the bs&w of the FCC so that the slurry does not become catalyst fines. Of course, initially, the slurry oil was from the tank; now, we are processing directly hot.

Eberhard Lucke (Commonwealth Engineering & Construction): I have the same question, and I have a comment.

We process FCC slurry in a coker in Germany. We limit it to 15% of the feed because of cat fines in the stream and negative impacts on the heater run lengths. We had a two-tank system in which we allowed residence time to solve the slurry problem. Of course, then you have to cut open the tank every 5 years and get out all of those catalyst fines. It worked fine, though.

We definitely experienced a negative effect when we went down too low in the level of one of the tanks and carried over a lot of the fines. Within a couple of hours, the heater was basically coked solid. So, I am wondering if you had any kind of filter system or anything in place to prevent excessive cat fines from being carried with the slurry to the coker.

Ghosh: We have a filter in the FCC unit itself. There is no separate filter. But if there are FCC upsets, in those conditions, definitely the CLO [clarified oil] is not fed to the coker, and the CLO quality is continuously monitored.

Gianzon: We do not have any filters either. Our fouling rate was much higher, but we attributed that to processing heavy Canadian crude on a single-fired furnace. We know the impact of fines in the furnaces. Often, the economics for processing heavy Canadian crude warrant the more frequent shutdown that we are experiencing due to higher heater fouling.

Debangsu Ray (Indian Oil Corp. Ltd.): I just want to respond to what you said.

There are Pall filters working on slurry. Pall filters will actually remove the catalyst fines. The advantage of using slurry is that the aromaticity of the feed increases; therefore, the asphaltenes will remain in the suspended form and will not foul up the heater tubes. So, the advantage far outweighs the disadvantages.

Tariq Malik (Egyptian Refining Co.): One thing that has not been mentioned is that when there is heavy coker gas oil involved, it all goes back as cat feed. There is a conversion penalty in the cat cracker. That is why a lot of people have stopped this practice.

There are other means of staying in the shot-coke business by doing other modifications of the coker where you can handle this without backing out additional vacuum resid. Actually, if you do the economics on it, you will find that slurry is not a good option. What the cat gives up cannot be made up by doing this.

Mahesh Marve (Reliance Industries Ltd.): Actually, I agree with the opinion just expressed. We have also seen that.

When you process the slurry oil, the quality of the gas oil, as well as the distillates, the diesel cetane, deteriorates. The second effect we see is that when the slurry oil boils in the VGO range itself, unless you are operating the coke drum at higher pressures, most of the slurry goes directly into the vapor phase. So, you are essentially increasing the load on the system without converting it more.

During replacement of coke drums with larger diameter drums, what process and operational changes do you expect?

Lewellen: This actually fits in with our El Dorado [Kan.] experience.

We just recently upgraded our coke drums. The balance of the unit was left untouched; all that was necessary was to modify the support in the new drum system. The purpose of the modification was not to increase feed rate to the unit. We changed the feed rate to increase coke yield pretty substantially.

So, that's the background for our answers. I am going to walk through it system by system and the impacts we saw, starting with the fractionator wet gas system: More coke means higher wet gas volumes.

The operators really keyed in on the higher volumes of gas that were much more difficult to manage. They had to be much more careful around the switch deck operation. Some shifting liquid yields were mostly due to slightly lower, sometimes higher drum pressures, lower drum outlet temperatures, and running along the drum cycle. On the charge heater system, we were able to lower outlet temperatures.

Our fire duty came down some. We are running longer drum cycles. Having less gas oil in the feed is really the benefit for the outlet temperatures. The larger drums, combined with a longer drum cycle, allows for lower drum outlet temperatures while maintaining the same VCM [vinyl chloride monomer] coke speciation. However, we did see a significant increase in our tube fouling rate in the heater.

Drum cycle length: We went from a 12-hr cycle to an 18-hr cycle. Of course, the velocity decreased significantly.

Foaming qualities: I have touched on this a few times. The overall foam front height went down but was more stable. We have had a lot more antifoam, we feel, due to the lower temperatures in the drum.

Also, operational impacts: We have to allow more time in our drum cycles for airframe pressure testing and warming up these much bigger drums. In the quench blowdown system, of course, it is intuitive. We are going to need much more quench water capacity. We produce much more sour water. There are more vapor rates in the blowdown system. We have generated much more oil and slop. We knew this system was going to be marginal, and it is been a little bit of a challenge to operate.

Moving on to the coke storage and handling: We are experiencing a significant increase in shot-coke production and shift to more shot coke. It was a major change for us. We utilized more surge volumes in our coke-handling systems, and this could be an issue for other units, especially those that utilize a coke crusher to slurry and hydro bin storage system.

We went to a feeder breaker system vs. a rail method in the decoking and conveying our coke. Also something that we almost missed in our facility with shipping logistics was that the additional coke has to be shipped in railcar truck and handled appropriately.

Finally, our biggest challenge was with our new jet pump because it was behind the rest of the project by 6-8 months; so we utilized a much smaller jet pump initially. The vendors will warn you, and it is absolutely true.

Our coke fines went up incredibly high: We experienced probably a two-fold increase. One reason is that you are trying to grind the coke out of the drum vs. cut it. The other one is that cutting times, as I said, doubled. Most of the time, we are actually tripling our cutting times.

The jet water pump is running that much longer than it was before. Of course, that led to reliability problems. Our first effect was the decoking valve followed by the cutting tools and nozzles; and eventually, the jet pump failed due to erosion. We did get the new jet pump in and created much less coke fines.

We did not get back to the previous smaller drum coke fine generation, but it is definitely much better. Cutting times improved. After the new jet pump, the decoking became limited by coke bridging in the dump chute/bottom unheading device. The new cutting system was no longer a rate constraint.

We did take the opportunity to add a little fine separation capacity in the project at that time to make the system much more reliable.

Eberhard Lucke (Commonwealth E&C): The extra capacity gained by installing larger diameter coke drums can be used by either increasing the unit throughput or by taking advantage of a longer drum cycle time at the same unit charge rate. Due to the larger diameter, steam stripping and quenching can be more challenging, and the probability of hot spots may increase.

Before installation of larger diameter coke drums, special attention has to be paid to the jet pump pressure. If the discharge pressure is lower than recommended for the new coke drum diameter, then coke cutting will take a lot longer and create significantly more coke fines due to the grinding effect of the broader water jet. The jet pump and the drill/cutting tool may have to be replaced or revamped to ensure proper coke cutting.

Also, the capacity of the coke-handling system—pit, pad, attached breaker or feeder-breaker system, slurry system—receiving a larger amount of coke has to be checked for adequate capacity.

Kevin Proops (Solomon Associates): Jeff, when you said that your fines did not go back to where they were with the new pump in, do you mean total pounds of fines or do you mean as wt % of coke produced?

Lewellen: Total fines produced on a daily basis went up on a percent of weight basis.

Kevin Proops (Solomon Associates): But you are making more coke, too?

Lewellen: Yes.

Kevin Proops (Solomon Associates): So, any idea whether it is a percentage of coke that can make the fines higher?

Lewellen: I would say that percentage-wise, it did increase.

Tariq Malik (Egyptian Refining Co.): What diameter drum did you go from where to where when you increased the drum diameter?

Lewellen: We went from 20-ft diameter drums to 26-ft diameter drums.

Tariq Mailik (Egyptian Refining Co.): And you did not increase feed rate or anything else?

Lewellen: No.

Tariq Malik (Egyptian Refining Co.): Then the only thing that was increased were the hours of the coking cycle? So, then by how much time were they increased?

Lewellen: From 12 to 18-hr drum cycles, we changed feed composition. We upgraded our vacuum tower upfront and did a significant cut point improvement in our VTB [vacuum tower bottoms].