Question 36: What is the panel’s experience with the latest technologies used to control the bromine index (BI) in aromatic streams, other than clay treatment? How many units are operational, and how were these justified?
DUBIN (Axens North America)
The most prevalent technologies for bromine index control are clay treating and selective hydrogenation of the olefins. At Axens, we recommend the selective hydrogenation route to control the olefins and trace diolefins that may appear in your aromatic streams. Last year at the NPRA Q&A, there was a similar question about the different reactive mechanisms between clay and selective hydrogenation. So, if you have any particular questions about that specific topic,
please refer back to last year’s transcript.
Going forward, we feel that the cost for handling and disposing of spent clay will increase. This spent clay disposal cost is going to make the economics for a selective hydrogenation unit more attractive. With any new aromatics plant, we recommend a selective hydrogenation unit where the capital cost to these units can be proportionally small compared to the overall cost of the complex. But for a retrofit, it will be much more site dependent as to whether the capital cost of installing a new selective hydrogenation unit will pay back over the
long-term cost of disposing of the clay.
ELIZABETH KACMAR (Marathon Petroleum Company)
If you do the selective hydrogenation in your sulfolane unit, would that not make further distillation more difficult? It is challenging enough to separate the toluene and xylene. So, if you are selectively hydrogenating certain species, would that make it even harder and result in an off-spec product?
DUBIN (Axens North America)
A selective hydrogenation unit should not make distillation significantly more difficult. The components you are hydrogenating are olefins and diolefins, so you are not increasing the non-aromatic content. Also, the boiling points of those streams’ components are fairly similar already.
ELIZABETH KACMAR (Marathon Petroleum Company)
When you clay-treat them, they normally are heavier, correct? So, if you are hydrogenating them, then their boiling points will still be very close to your product boiling points and you will use more energy to distill them.
DUBIN (Axens North America)
That is true. I wonder if you might have a site-specific issue that we could discuss after
this panel is over.
DUBIN (Axens North America)
Axens' current technology offer for the control of the olefins and traces of diolefins in the feed to aromatics plants is a reformate-selective hydrogenation application. In Question 33 of the 2011 Q&A, different reactive mechanisms between selective hydrogenation and clay treating were discussed; so that information will not be repeated here.
One advantage that selective hydrogenation units do offer besides the different reaction mechanisms is the reduced consumption of clay. The costs for disposal of spent clay should increase long term due to increasing environmental constraints, leading to improved economics for the selective hydrogenation of olefins in reformate.
Our reformate-selective hydrogenation technology is proposed with any new aromatics complex, where the cost of a new unit is proportionally small compared with the overall complex. As a retrofit, the cost/benefit analysis of clay versus selective hydrogenation will need to be based on local constraints that will obviously vary at each site. A detailed analysis of the economics to justify the installation of a reformate-selective hydrogenation unit will be required
for each refiner.
ZHEPENG LIU (GTC Technology)
Clay treating is a conventional method that removes the small amount of olefins in aromatics-rich streams. Under clay treatment, olefins go through an acid-catalyzed alkylation reaction with aromatic molecules to form higher boiling components which are removed in the downstream distillation sections. Some features of this process are negative due to moderately-high temperature (160°C to 200°C), non-regenerable clay, and frequent unloading, loading, and/or disposal [high maintenance cost with HSE (health, safety, and environmental) concerns].
GTC offers the market a novel adsorption method to remove the small amount of olefins in the aromatics streams. The process operates at ambient temperature with a regenerable sorbent in a treater bed. The outlet product achieves a Bromine Index of less than 5 during the entire operation cycle. Regeneration is made under mild conditions (temperature 200°C to 250°C) using nitrogen or methane. In many cases, clay treatment or adsorptive removal of olefins is not needed at all. The GT-BTX® extractive distillation technology for aromatics recovery and purification is very efficient at rejecting olefins from the aromatic product, such that the BI specification can often be met by only extraction without further treatment.
Break
Question 35: Does the panel have any experience intentionally producing a ‘heavy’ alkylate stream? What are the disposition options for this stream?
PIZZINI (Phillips 66)
Phillips 66 has one location that makes odorless mineral spirits (OMS), starting with the raw alkylate off of the deisobutanizer column. We take those bottoms through a rerun where we drive off to 340ºF, and then that stream goes right back into alkylation unit storage. The second fractionation step is done under a vacuum. Now we take the 410ºF overhead. Anything heavier can be routed to the mogas (motor gasoline); or if it has enough heavy content, we can put it in the diesel stream if we are getting a higher margin for diesel fuel than for gasoline. So that bottom stream off of the SOL (safe operating limit) column can go either way: to mogas or to diesel.
The stream that ends up as OMS is put through a molecular sieve which takes out the water and the sulfur. A typical yield relative to alkylate is about 3% for this unit. We have a few other sites that split alkylate generally to make aviation gasoline with a light fraction in at least one location that sends the heavy fraction to a third-party. The third-party then turns that into an isoparaffin solvent.
PIZZINI (Phillips 66)
One of the Phillips 66 sulfuric acid alky units produces odorless mineral spirits from heavy alkylate. A portion of the deisobutanized alkylate is routed to a fractionator where 340°F minus material is driven overhead and returned as alkylate product. The higher boiling point material is fed to a second fractionation column which separates material with an approximate boiling point of less than 410°F from material with higher boiling points. The lower boiling point
material is treated to remove water and sulfur and is sent to product storage as odorless mineral spirits (OMS), an FDA approved solvent. The higher boiling point material is sent to light gas oil or to gasoline. The odorless mineral spirits can also be routed into diesel if the economics dictate.
Typical OMS yield on alkylate is around 3% (see simplified flow diagram below). Three of our other sites generate a light alkylate cut as an aviation gasoline blend stock. Two send bottoms to mogas, and the third sends bottoms to a third-party to make isoparaffin solvent 355°F to 410°F.
Question 37: What is the panel's experience in operating chlorided isomerization units, which were not designed specifically for benzene saturation, in a benzene saturation mode? How are operational parameters adjusted for the different operation?
MUEHLBAUER (Valero Energy Corporation – Benicia Refinery)
This is a good question; because as we blend more ethanol into the gasoline pool, a lot of refineries become octane-long and RVP (Reid Vapor Pressure)-long as well. So, the incentives to isomerize are diminishing as you blend more ethanol. Within Valero, we looked at one of our isomerization units and did a project to convert it into a benzene saturation unit but without capital improvements. We accomplished this by operating a unit designed for isomerization in benzene saturation mode. The key component is the cooling of the first reactor effluent. The benzene concentrations are normally maximum 4% to 5% in the isomerization feed. For each percent of benzene in the feed, you get about a 20°F exotherm on the reactor. So, the more you can cool the first reactor effluent, the more favorable the iso-to-normal equilibrium will be in the second reactor.
In terms of ways to cool the effluent, increasing unit recycle is one method, assuming there are not any hydraulic or space velocity limitations. Essentially, the benzene in the reactor feed, compared to the unit feed, is much more dilute. The other method would be trying to improve cooling by installing an inner reactor quench. I have heard that some facilities will bypass some of the feed around the first reactor into the second reactor to get exotherm and perhaps a little more cooling that way.
DUBIN (Axens North America)
We have seen that the quantity of benzene in the isomerization feeds has gone up due to the current regulations on benzene in the gasoline pool. For those units that do not have a dedicated benzene saturation reactor, the impact is a little more significant. The rule of thumb that Joe mentioned, 20ºF exotherm for every vol% (volume percent) of benzene, is applicable.
When processing feeds with a benzene content higher than design, we recommend reducing the inlet temperature to your first isomerization reactor to lower than typical. The goal is to have the temperature at the outlet of what we term the benzene saturation zone (that is, the top layer of the first reactor where all of the benzene is hydrogenated) at the ideal isomerization reaction inlet temperature.
Those refiners who are targeting maximum octane improvement, as opposed to just benzene destruction, may still need a higher overall temperature through that first reactor as you are bringing in more naphthene rings that will have to be opened before you can turn them into isoparaffins.
PATRICK BULLEN (UOP, A Honeywell Company)
We have had some licensees who do not have octane as a priority in their Penex unit but who still need to remove the benzene. In several cases, they have cut out their chloride injection in the Penex unit. They still typically see around a five upgrade across the unit, even without the chloride injection. They are able to handle the loss in activity by raising reactor temperatures to around the 400°F range for peak temperature. The system seems to be fairly stable. Of course,
you get the normal deactivation from oxygenates in the feed, for example, that eventual kill the catalyst. But for that type of operation, it seems to work well for them.
Our customers have seen some sulfur upsets in this type of unit operation. Reinstating their chloride injection seems to help recover some of the activity and still allows them to continue getting that minor octane upgrade they wanted. I agree with the other comments that, generally speaking, using the start-up line from the bottom of the stabilizer as a dilution source certainly dilutes the benzene that is in the feed as long as the hydraulics of the unit will allow it. We have some customers who have put in the separate benzene saturation reactor in front to try to minimize the impact on the Penex unit itself as a capital option.
DUBIN (Axens North America)
Axens has a number of units operating in benzene saturation mode even if the original design did not call for the same level of benzene reduction. With increasing benzene content in the isomerization feed, an increasing percentage of the isomerization catalyst in the first reactor is now utilized for benzene saturation. The refiner should lower the inlet temperature to the first reactor so that the temperature at the outlet of the ‘benzene saturation zone’ in the first reactor is
in line with the desired temperature. An increased overall temperature profile in the first isomerization reactor may still be required for ring opening of the additional naphthenes in the system brought about by the increased benzene content in the feed. The typical rule of thumb for liquid phase benzene saturation is a 20°F exotherm for every 1 vol% of benzene saturated.
AREIN LAZONDER (Albemarle Corporation)
Benzene saturation is possible in an isomerization unit loaded with platinum promoted chlorinated alumina catalyst. The bi-functional catalyst only needs the platinum function to saturate benzene. The acid sites on the catalyst have no additional function for the hydrogenation of benzene.
Operating in a benzene only saturation mode with no additional isomerization of the paraffins can be done in one of two ways: (1) by the injection of a chloriding agent to conserve the acid sites to be able to isomerize the paraffins once the benzene saturation mode is over or (2) without the injection of a chloriding agent if the unit will continue operating in a hydrogenation mode. With the latter operation the acid sites will be lost in the course of time and the isomerization activity cannot be fully recovered. Chloride inhibits the hydrogenation activity of platinum partly. For optimum hydrogenation activity, ceasing chloride injection can be considered. In both cases the operating conditions are moderate, at a RIT (reactor inlet temperature) = 120°C/250°F benzene is already hydrogenated. An excess of hydrogen is mandatory to prevent the catalyst from coking. At the reactor outlet the H2/HC (hydrocarbon)ratio should still be 0.1 molar ratio. Benzene saturation is exothermic so, if large amounts of benzene are processed, the heat release can be significant. To prevent temperature excursions, it is recommended to keep the maximum benzene concentration to be processed at less than 7 wt%.