ABSTRACT

Martensitic stainless steels continue to be one of the most widely used corrosion resistant alloys in oil and gas developments. Determining if a martensitic stainless steel is acceptable in an unproven environment requires testing to confirm, but predicting the outcome of a given test environment is often initially based on personal experience rather than a qualitative and quantitative assessment. An empirical method for improving the predictability of NACE TM0177 Method A Tensile tests on modified 13Cr 110 ksi grade martensitic stainless steels based on an H2S/Chloride/pH function has previously been developed based on published data in order to address this uncertainty. The environments considered by this function are only limited by the capabilities of the NACE TM0177 Method A test, and provide a method for rapidly estimating if a M13Cr 110ksi grade should pass or fail in multiple different environments. As a follow-on to the development of this empirical method, data points from new tests were used to check the general predictability of the H2S/Chloride/pH function. The general function was modified due to the addition of the new data, and subsequently checked again against a separate set of data. The nature and implications of these findings are discussed and conclusions drawn regarding the performance and value of the methodology for the evaluation of future materials applications

INTRODUCTION

13Cr-5Ni-2Mo 110 ksi grade martensitic stainless steels are some of the most widely used corrosion resistant alloys (CRA) in oil and gas projects due to their relatively low cost, ease of manufacture, and good CO2 corrosion resistance. These materials are susceptible to both sulfide stress cracking (SSC) and stress corrosion cracking (SCC) at various conditions, making their selection somewhat challenging, especially if there is little data to support or contradict use in said conditions. Currently, the best method to determine if this material can be used in a given sour environment is by domain diagrams and then by fitness-for-service testing. Several domain diagrams have been published over the past few years,and each diagram is valid for only one chloride concentration or content. This limits their use for selecting material, or determining if fitness-for-service testing should be performed and its potential outcome. Thus, experience becomes a very important aspect to extrapolate between the various diagrams and to account for differences due to specific materials. Recent work

developed an empirical model to aid in predicting the outcome of a NACE TM0177 Method A

5

tensile test of this material. This model was based on substantial published data, and is versatile enough to be used at any chloride content or H2S partial pressure found in a NACE TM0177 Method A test up to saturation and for pHs between 3.5 and 4.5. Unlike domain diagrams, this function represents a three-dimensional surface; below the surface, SSC is not expected, and above which SSC could potentially occur. This model can also incorporate new data and be modified as necessary based on experience and conservatism.

This content is only available via PDF.
You can access this article if you purchase or spend a download.