Origins and Fates of Chlorides in Hydroprocessing Units – Part 1: Recognizing the Problem

Origins and Fates of Chlorides in Hydroprocessing Units – Part 1: Recognizing the Problem

chloride-processing-at-night.jpg

Introduction

This series of three articles explores the impacts chlorides may have on hydroprocessing units (hydrotreaters and hydrocrackers).  It provides a methodical approach to identifying the typical effects that point toward chlorides, the sources of chlorides in process feed streams, chloride-induced failure mechanisms, methods for identifying chlorides, strategies for chloride control, and a step-by-step process outline for dealing with a problem.  Some of the approaches and impacts here can also be applied to other halogens in hydrotreaters, such as fluoride.

This first article of the series focuses on recognizing a chloride problem in a hydroprocessing unit.  In the second article, we tackle how to identify the magnitude and source(s) of the problem.  The third article presents ways to address the chloride issues.

——————————————————–

Problems caused by chlorides are often missed or misdiagnosed in a refinery.  They impact not only the hydroprocessing units, but other units as well.  Sometimes the methods used to manage chlorides in upstream units, such as corrosion inhibitors, merely move the problem on downstream.  Partial solutions in hydroprocessing units may, in turn, just pass problems on to other units.  Comprehensive solutions require a wider understanding of the problem.

Locating a chloride root source is made more difficult if a problem has gone unrecognized or has been allowed to persist for a few months.  The chlorides will propagate throughout a refinery in multiple streams and in multiple forms which can obscure the original source.  There may be multiple sources.  Someone may have introduced the chlorides into a system without realizing it or without realizing the impacts.

The amount of material required to create a significant chloride problem is often very small.  To get 1 wppm chloride from perchloroethylene (PERC) contamination in a 50,000 barrels of naphtha hydrotreater feed, you only need about 1 gallon of PERC in the feed (0.0055 v% of the stream).  Introducing a barrel of PERC into the naphtha stream can contaminate it for several days.

So, how can you approach a chloride problem?  Methodical application of the steps below is suggested.  The balance of this article provides the background to execute the steps.

  1. Recognize the problem.  Recognize the chloride problem from the impacts observed in the plant. Where is the problem?  Are other units seeing problems?
  2. How big is the problem?  How much material are you looking for?  Calculate or estimate the amount.
  3. Identify the source(s).  Identify potential sources for the chlorides, both organic and inorganic.  Look especially at reformers and isomerization units where concentrated organic chlorides are present.  Use analyses to narrow down the possible sources. For organic chlorides, determine specifically what chloride compound(s) you are looking for (speciation).  If all you find is PERC, for instance, then you need to suspect the reformer, the isomerization unit, or solvent dumping.
  4. Manage the chlorides.  This may include physical changes or correcting practices and procedures.  You may need to use higher metallurgy in some equipment.  You may need to adopt a coping strategy rather than a complete solution.

Chlorides and/or their effects can be successfully controlled, once they are identified and understood.

We begin by looking at how to recognize a problem rooted in chlorides.

Step 1 – Recognize the Problem

Chlorides as a possible issue can be identified from its typical effects on hydroprocessing units.  There will likely be impacts in other units also which can serve to support your identification.

Figure 1 below illustrates several areas to look for indications of chloride impacts in a hydroprocessing unit. Impacts are sometimes also seen in other units of the refinery.

fig1 sources of chloride problems

The most common issues indicative of chlorides include:

  • Deposition of Salts in Reactor Preheat Exchangers and Charge Heater (A).  Seeing deposits of white salt in exchangers when opened often indicate a chloride problem.  This is seen in cokers and crude units, as well as hydroprocessing units.  If there is any entrained water in the hydroprocessing unit feed, or chloride salts above saturation are present, there will be fouling of the feed preheat exchangers.  The salts will simply lie down on the exchanger tubes as their solubility dictates.  If the tubes are austenitic stainless steel, stress corrosion cracking may occur.  Under-deposit corrosion is also a likely result.  In any event, there would be a loss of heat transfer and, eventually, high pressure drop.
  • Reactor Fouling (C). Chloride-containing salts that reach the reactors will decompose or hydrolyze at reaction conditions, releasing HCl and leaving metals fouling the catalyst.  If, somehow, sodium chloride is present in the feed, it will deposit directly in the catalyst bed without decomposing, forming a hard rind and causing high pressure drop.  Fortunately, most catalysts today are fairly resistant to poisoning, so the metals may not hurt much.  The HCl probably does not harm the catalyst and, in the case of hydrocracking, may even help catalyst activity a little.  More importantly, the HCl moves into the effluent train.
  • Reactor Effluent Fouling and Corrosion (B, E).  Heat exchanger tube and shell thinning or pitting, especially in reactor effluent exchangers, is often seen.  The most common corrosion location of concern is the reactor effluent side of the feed/effluent exchangers where ammonium chloride salts (NH4Cl) deposit in the exchangers, especially when wash water practices are inadequate.  If austenitic stainless steel is present, chloride stress corrosion cracking presents an additional metallurgical challenge.

Fouling or high pressure drop in effluent exchangers at higher temperatures than ammonium bisulfide laydown occurs (say over 250OF) also points toward ammonium chloride deposition.

In the reactor effluent train, ammonium chloride will deposit below the temperature indicated by:

TDep = 523 * EXP(0.0507 * Ln(Ksp))                    (Ref. 4)

Ksp = PNH3 * PHCl     where

TDep = Deposition Temperature, OF
PNH3 = Ammonia Partial Pressure, psi
PHCl  = Hydrochloric Acid Partial Pressure, psi.

Note that the feed nitrogen is just as important in this equation as the chlorides.  The use of amines upstream to control corrosion or scavenge H2S will aggravate a chloride problem.  Once deposited, the ammonium chloride increases pressure drop, reduces heat transfer, and causes tube damage by under-deposit corrosion.  This effect is intensified as water begins to condense in the reactor effluent.  The first drop of water will be rich in acid gas and very corrosive.  Ammonia generated from feed nitrogen or injected with wash water can help reduce the pH impacts; but ammonia is not as soluble at high temperatures as HCl, so the HCl tends to control the pH.

  • Stripper/Fractionator Feed Preheat (A).  Some of the most difficult exchanger conditions are presented when fractionator or stripper feed is preheated by high-pressure reactor effluent.  The fractionator feed has a small amount of residual free water that is carried into the fractionator preheat exchangers.  This water contains dissolved ammonium chloride.  As the stream is heated, the water eventually evaporates, leaving ammonium chloride salt deposits on the tubes where it evaporates.  The exchanger where you can expect trouble can be identified using flash calculations, if you can estimate the water slip out of the upstream separators.

    The deposit insulates the tubes, raising the temperature on the underside of the deposit until the ammonium chloride breaks down into ammonia and hydrochloric acid.  The presence of trace amounts of water in the hygroscopic ammonium chloride deposit promotes acid attack of the tubes in the form of pitting.

    Dry attack under deposits is also possible as a corrosion cell is set up between the clean and the fouled metal surfaces.  Attack is, again, in the form of pitting.

  • Stripping and Fractionation Impacts (D, G, H).  Corrosion of the upper trays in a hydroprocessing unit stripper or fractionator may indicate a problem.  Crude units will see essentially the same effects from chlorides.  Wet H2S can also show a similar effect. Testing for corrosion products and pH wherever water collects downstream of a suspected problem may help.  Deposits of iron sulfide will be observed in product rundown coolers and tanks when chloride is active in a system.  A metallurgist frequently helps sort out these effects.

    In the fractionation system, chlorides will follow the water and ammonia, just as they do in the crude units.  Expect to see corrosion anywhere a liquid water phase may be present.  Problem areas frequently include upper trays, tower walls, and overhead condensers.  Chloride salt deposition in draw lines and exchangers has also been observed.  This laydown follows the equation presented above, although application of the equation is difficult because it is hard to determine partial pressures for ammonia and acid.

  • Compressor Issues (F).  Makeup hydrogen compressors and recycle compressors will experience chloride salt deposition on machine surfaces.  In reciprocating compressors, deposits form on valves, resulting in high valve failure rates.  Centrifugal compressors will experience loss in efficiency.
  • Impacts Observable in Other Systems and Units: 
    • Amine Systems – Several hydrotreating units have high-pressure amine scrubbers to remove H2S from recycle gas.  These scrubbers also effectively remove chloride from the gas.  While the hydrotreater may not have problems, the amine regenerator tower and overhead system may suffer accelerated corrosion.
    • Fuel Gas Systems – Chlorides have been detected in fuel gas streams where fouling and corrosion are issues.  These are believed to be active in promoting corrosion of the fuel gas piping.
    • Reformer Water/Chloride Balance Problems – Reformers that do not need chloride makeup to maintain a residual are probably getting chlorides through the feed.  This may or may not be a problem, depending on the unit.

As a general comment, chloride issues are often missed because evidence may come in the form of iron sulfide (FeS) deposits in equipment, which may be attributed to sulfidic corrosion.  The FeS may actually come from wet NH4Cl under-deposit corrosion via the following reaction route:

chloride equations

Once you realize you have a chloride issue, you need to determine the magnitude and find the source(s) of the chlorides.  These are the subjects addressed in the second article of this series.

If you have a question relating to this blog, you may post a comment for the author at the bottom of this page. If you would like to submit an Information Request please click below:

Request Info from Becht

|

About The Author

Contact:
Steven Treese began his professional career with Union Oil Company of California in 1973 as a Research Engineer with a BS in Chemical Engineering from Washington State University. He remained with essentially the same company through several “heritages” before retiring from Phillips 66 in 2013 after over 40 years in industry.  He continues to provide process consulting serrvices by request. His range of experience includes operations, hydroprocessing, hydrogen plants, catalyst development, design, process safety, utilities, sulfur recovery, geothermal, shale oil, nitrogen fertilizers, procurement, and process licensing.  Steve has a handful of publications and was on the 1994 NPRA Q&A Panel. He has been an inventor on patents in diverse areas, including vessel internals, enhanced oil recovery, and hydroprocessing. He was lead editor for the “Handbook of Petroleum Processing, 2nd Edition” (Springer, 2015) and is the author of an upcoming book on measurement history. Steve is a licensed Professional Engineer, a senior member of the American Institute of Chemical Engineers, and a mentor for FIRST Robotics Team 3049 from Bremerton, WA.

Authors Recent Posts

Origins and Fates of Chlorides in Hydroprocessing Units – Part 1: Recognizing the Problem
Let Becht Turn Your Problem
Into Peace of Mind