Discussion Central

Expand all | Collapse all

Hydrogen induced corrosion

  • 1.  Hydrogen induced corrosion

    SENIOR MEMBER
    Posted 16 days ago

    Dear All

    In one of my client facilities, we have observed cracks and leaks in 2 new hydrogenation reactors (MOC SS 316L) after very few batches (< 20 number) within a month of commissioning. Inspection agencies pointed to Chloride Induced Stress Corrosion although none of the raw materials used during operations contain any chloride (confirmed by testing). Could the chloride in the water used during hydro-test (during and after fabrication) of the reactor be enough source of chloride to damage the reactors? Operating T & P of 200 deg C and 35 bar in non-aqueous system

    Thanks



    ------------------------------
    R. Muthukrishnan
    Chemical Engineering Consultant
    Mumbai/Vadodara
    India
    ------------------------------


  • 2.  RE: Hydrogen induced corrosion

    SENIOR MEMBER
    Posted 13 days ago
    Hydrogen embrittlement is a serious problem in certain processing equipment.  My father, who was also a chemical engineer, showed me the bottom 4 feet (approximately, 1.2 meters) of an amine stripper in the late 1960s that looked like wood with a severe case of "dry rot".  The stripper was made of about 4 inch (10 cm) of type 304 stainless steel.   He had that part cut off and replaced with type 316 stainless steel which includes molybdenum in addition to the iron, chromium and nickle of type 304.  According to my father, the molybdenum fits in a large interstice between the other atoms and blocks the diffusion of hydrogen.

    ------------------------------
    T. David Griffith, Ph.D.
    Blessing (Bay City), TX
    ------------------------------



  • 3.  RE: Hydrogen induced corrosion

    SENIOR MEMBER
    Posted 12 days ago
    Dear Muthukrishnan,
    I would exclude also hydrogen attack as root cause of the crack formation. From my point of view chlorides from test water cannot be excluded as cause for Cl induced SCC if the final hydrotest of the vessels has not been performed with demineralized water in case of designs which are not gap free or at least flushed with demineralized water after hydrotest if the design is gap free. I assume the crack formation is from inside the vessel. Hydrogenation is normally done with steam feed and as far I understand you process is a batch process which has some transient conditions where steam condensate might occur.
    We experienced SCC on a SS 321 nitric acid absorption tower even from outside. The column was equipped with a welded open drainage channel to drain any leaking nitric acid in case of a leak. There was never a  leak utilizing this channel but during normal operation chlorides were collected in this channel and caused SCC from outside. Unfortunately the main direction of wind was from the nearby sea and the cooling towers of the plant were located between the see and the column. Such the chlorides from the sea swell combined with the fumes from the cooling tower and got separated on the outer column wall and collected in the welded drain channel where they initiated SCC as the column was not coated.
    So I think the conclusion of the inspection companies is probably right. Maybe you can find chlorides in the cracks by electron microscopy.  The hydrogenation reactors I know were all made from ferritic materials fullfilling the API 941 conditions and having a PWHT for protection against hydrogen induced cracking. Probably in your case SS material was selected to avoid PWHT as  austenitic stainless steel is immune against hydrogen attack for the given operating conditions but has a risk of SCC in presence of chlorides. In so far the headline of your query  is a bit misleading.

    ------------------------------
    Reinhard Michel
    Mechanical Engineer
    Retired Head of Static Equipment Group, Fertilizer Division
    thyssenkrupp Industrial Solutions AG
    Dortmund, Germany
    Reinhard
    ------------------------------



  • 4.  RE: Hydrogen induced corrosion

    LIFE MEMBER
    Posted 12 days ago

    Dear Muthukrishnan, 

    May I suggest you to take reference of these valuable information available in websites pertaining to hydrogen corrosion / hydrogen embrittlement to steel. 

    https://www.hse.gov.uk/research/rrpdf/rr1134.pdf

    https://www.csb.gov/assets/1/20/metallurgical_analysis.pdf?15125

    http://xn--e1aabhzcw.xn--c1avg.xn--p1ai/dat/codes_doc_1891.pdf



    ------------------------------
    Ng H Kiang Lucas PE,CEngSpore
    General Manager
    Petrochemical Corporation of Singapore (private) Limited
    Singapore
    ------------------------------



  • 5.  RE: Hydrogen induced corrosion

    LIFE MEMBER
    Posted 12 days ago
    My recollection from my time in industry is that whenever hydrotesting was done for stainless steel equipment in the refineries, the water to be used was specified as steam condensate to assure that the chloride level was low enough to not be a cause of chloride stress corrosion cracking. 

    Even though the chloride content of the hydrotest water may appear to be low enough to not be of concern, evaporation of water left in the system between the hydrotest and commissioning can result in local pools of high chloride concentration. 

    Alfred M. Center
    Professor of Practice
    Robert Frederick Smith School of Chemical and Biomolecular Engineering
    Cornell University
    Ithaca, NY 14853






  • 6.  RE: Hydrogen induced corrosion

    SENIOR MEMBER
    Posted 12 days ago
    Mr. Muthukrishnan,


    Since hydrogen embrittlement of austenitic stainless steel is not susceptible to hydrogen embrittlement, you might verify the metal chemistry is in fact as expected.  That said, chlorides have been introduced to SS equipment via hydro tests in the past creating chloride SSC. You might dig in to understand the stresses on the metal since the number of cycles seems quite low.  An additional cause could be sea water or sea air depending on shipping methods and plant construction location. 

    The cyclic nature of the process being batch or semi-batch can increase the damage especially if the equipment was not designed for cyclic service.

    You and your client need a good mechanical design engineer and a deeply experienced metallurgical engineer, in my opinion.


    Regards,
    Gary Koehler
    Houston





  • 7.  RE: Hydrogen induced corrosion

    SENIOR MEMBER
    Posted 12 days ago
    Yes, hydrogen embrittlement is indeed a serious problem for many MOCs. For example the Syngas reformer pressure in ammonia plants is limited to ~ 3.5 MPa simply because of this H2 caused embrittlement at the temperature used. Higher pressure would have been ideal because it would have eliminated (or reduced the duty) of the booster compressor prior to the ammonia reactor. From what I know even for this low pressure the MOC (special alloy) is imported for Indian plants. Some of the above details are available in:

    Andrews SPS. (1977) Modern processes for the production of ammonia, nitric acid and ammonium nitrite. In: Thompson R, editor. The modern inorganic chemicals industry. The proceedings of a symposium organized by the Inorganic chemicals group of the industrial division of the Chemical Society, London, March, 31- April, 1, 1977. Soc. Chem. Ind., London, U.K.



    ------------------------------
    Vishwas Pangarkar PhD
    Retired Professor of Chemical Engineering, University of Mumbai, India.
    Currently: Independent Chemical Engineering Academic/Professional.
    Nasik-422013, India.
    ------------------------------



  • 8.  RE: Hydrogen induced corrosion

    SENIOR MEMBER
    Posted 11 days ago
    Dear all,
    I suggest you convert the operating data given into US units. 200°C operating temperature and I assume a design temperature of 250°C makes 482°F and assumed the operating medium is pure hydrogen with 35bar operating partial pressure makes 35x14.5=507.5 psia. This is avery conservative assumption as the real hydeogen partial pressure will be lower.
    When you insert these data into the Nelson curves on page3 of API 941 as provided by Mr. Lucas the clear conclusion is that even under these very conservative assumption post weld heated carbon steel as MOC would be sufficient. Thus hydrogen attack can be excluded for the selected MOC.
    Another hint to the possible source of chlorides. Even if the hydrotest in the workshop was done with fully demineralized water, sometimes a so called system hydrotest including vessels and connecting piping is performed at site based on decision of the construction manager or local legislation. This is typicaĺly performed with normal drinking water containing chlorides as demineralized water is not available. Thus a oerfect fabrication can be impaired by sit acrivities.

    ------------------------------
    Reinhard Michel
    Mechanical Engineer
    Retired Head of Static Equipment Group, Fertilizer Division
    thyssenkrupp Industrial Solutions AG
    Dortmund, Germany
    Reinhard
    ------------------------------