Q & A > Question Details
We have LVGO stream from vacuum column & processing it in VGO hydrotreater. In LVGO stream we encounter chlorides up to 20 ppmw (organic+inorganic) which is posing corrosion issues in VGO hydrotreater. I want to know how to remove these chlorides prior to enter downstream unit.
 
Answers
04/11/2010 A: Berthold Otzisk, Kurita Europe GmbH, Berthold.Otzisk@kurita-water.com
A perfect desalting will help to remove inorganic chlorides, but cannot remove organic chlorides. The analysed 20 wppm chlorides in your LVGO are significantly high and should be reduced as soon as possible. Kurita´s patented ACF technology is an elegant way to avoid chlorides entering the LVGO. Injected into the vacuum tower the ACF product will directly react with chlorides to form water-soluble liquid ACF salts. These salts or their thermal decomposition products are routed to the vacuum overhead section and will leave with the vacuum sour water.
24/10/2010 A: Morgan Rodwell, Fluor Canada Limited, morgan.rodwell@fluor.com
The inorganic chloride salts can be removed by desalting. The presence of these chloride in an LVGO stream may indicate poor desalting performance upstream of the Atmospheric, or a difficult desalting application. You can add a desalter to the LVGO stream to remove these salts.
Organic chlorides are chemically bound in the oil and there isn't a method to remove these prior to hydrotreating. Proper metallurgy and water wash design is a necessity if you have organic chlorides.
23/10/2010 A: Marcello Ferrara, ITW SrL, mferrara@itw.it
For reducing inorganic chlorides you should reduce water content of feed residue. We have developed a new demusifier which can solve the problem.
23/10/2010 A: Lindsay McRae, Pall Corporation, Lindsay_McRae@pall.com
Suggest first thing you do is go back to desalter and start there. For some reason, desalter is not doing its job properly so it could need revamp, upgrading, or debottlenecking. Maybe chemicals addition optimisation too. One interesting thing we see more with increase of heavy / difficult crude diets in recent times is that because density difference between crude and water (brine) is getting smaller, desalter operations are getting tougher so salt can carry to CDU or oil carryunder with desalter brine to WWTP. Combined with the more difficult separation, any capacity creep above name plate capacity of desalter further reduces the desalter residence time. These lead to increased salt carryover into CDU and corrosion in downstream units like CDU overhead and also in VGO HT like you observe.
If there is slops being fed back to desalter this can upset desalter operation too. In some cases, refiners have seen benefits in segregating slops and treating separately with PhaseSep LL coalescer.
Third thing to consider is there is there any streams with high chloride content going to CDU such as wild naphtha stream for instance. One refinery in Southern Europe had big concerns with a wild naphtha stream containing a lot of salt, ammonia, and water which was being fed to CDU. CDU overhead corrosion was feared and this problem was tackled by installing water wash with high efficiency PhaseSep Liquid Liquid coalescer to remove salt and other corrosive compounds. (mesh pad / gravity separator could not make an adequate separation so high efficiency LL coalescer and pre-filter was installed).
22/10/2010 A: Alan Goelzer, Jacobs Consultancy, alan.goelzer@jacobs.com
It is very important to distinguish between organic chlorine [organic halides] and inorganic chlorides in the vacuum gas oil stream. Petroleum laboratories [like Maxxam Analytics] can now sort out organic chlorine fraction.
If most of the 'chlorine' is inorganic, 'chain' of upstream steps need to be checked to be sure that cooling tower water is not leaking into the vacuum gas oil. It is possible that a water washing step needs to be added upstream of the filtration system within the feed supply section of the VGO Hydrotreater Unit. Practical hydroprocessing specialists can provide cost-effective designs.
Organic chlorine is not anticipated to be fouling upstream of the HT Reactor, but must be managed downstream of the Reactor---via appropriate water washing. Again, practical hydroprocessing specialists can evaluate the specific situation and provide practical solutions.
22/10/2010 A: Ralph Ragsdale, Ragsdale Refining Courses, ralph.ragsdale@att.net
The obvious answer is to install a desalter, and they have been in locations other than the crude charge, though rarely. If your crude desalter is achieving a typical target outlet of 1 lb./1000 Bbls, which by my calculation is 3 wppm, and if the chlorides are evenly distributed, you would have 3 wppm in the LVGO. On the surface, then, it would appear that the desalter operation needs improvement.