1、The Technical Basis Document for API RP 941API TECHNICAL REPORT 941SEPTEMBER 2008The Technical Basis Document for API RP 941Downstream SegmentAPI TECHNICAL REPORT 941SEPTEMBER 2008Special NotesAPI publications necessarily address problems of a general nature. With respect to particular circumstances
2、, local,state, and federal laws and regulations should be reviewed.Neither API nor any of APIs employees, subcontractors, consultants, committees, or other assignees make anywarranty or representation, either express or implied, with respect to the accuracy, completeness, or usefulness of theinforma
3、tion contained herein, or assume any liability or responsibility for any use, or the results of such use, of anyinformation or process disclosed in this publication. Neither API nor any of APIs employees, subcontractors,consultants, or other assignees represent that use of this publication would not
4、 infringe upon privately owned rights.Users of this technical report should not rely exclusively on the information contained in this document. Soundbusiness, scientific, engineering, and safety judgment should be used in employing the information contained herein. API publications may be used by an
5、yone desiring to do so. Every effort has been made by the Institute to assure theaccuracy and reliability of the data contained in them; however, the Institute makes no representation, warranty, orguarantee in connection with this publication and hereby expressly disclaims any liability or responsib
6、ility for loss ordamage resulting from its use or for the violation of any authorities having jurisdiction with which this publication mayconflict.API publications are published to facilitate the broad availability of proven, sound engineering and operatingpractices. These publications are not inten
7、ded to obviate the need for applying sound engineering judgmentregarding when and where these publications should be utilized. The formulation and publication of API publicationsis not intended in any way to inhibit anyone from using any other practices.Any manufacturer marking equipment or material
8、s in conformance with the marking requirements of an API standardis solely responsible for complying with all the applicable requirements of that standard. API does not represent,warrant, or guarantee that such products do in fact conform to the applicable API standard.All rights reserved. No part o
9、f this work may be reproduced, stored in a retrieval system, or transmitted by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior written permission from the publisher. Contact the Publisher, API Publishing Services, 1220 L Street, N.W., Washington, D.C. 20005.Co
10、pyright 2008 American Petroleum InstituteForewordNothing contained in any API publication is to be construed as granting any right, by implication or otherwise, for themanufacture, sale, or use of any method, apparatus, or product covered by letters patent. Neither should anythingcontained in the pu
11、blication be construed as insuring anyone against liability for infringement of letters patent.This document was produced under API standardization procedures that ensure appropriate notification andparticipation in the developmental process and is designated as an API standard. Questions concerning
12、 theinterpretation of the content of this publication or comments and questions concerning the procedures under whichthis publication was developed should be directed in writing to the Director of Standards, American PetroleumInstitute, 1220 L Street, N.W., Washington, D.C. 20005. Requests for permi
13、ssion to reproduce or translate all or anypart of the material published herein should also be addressed to the director.Generally, API standards are reviewed and revised, reaffirmed, or withdrawn at least every five years. A one-timeextension of up to two years may be added to this review cycle. St
14、atus of the publication can be ascertained from theAPI Standards Department, telephone (202) 682-8000. A catalog of API publications and materials is publishedannually and updated quarterly by API, 1220 L Street, N.W., Washington, D.C. 20005.Suggested revisions are invited and should be submitted to
15、 the Standards Department, API, 1220 L Street, NW,Washington, D.C. 20005, standardsapi.org.iiiReport to API - The Technical Basis For API RP 941 TABLE OF CONTENTS 0HEXECUTIVE SUMMARY FOR THE TECHNICAL BASIS DOCUMENT . 361H1 1H1.0 ABSTRACT . 362H5 2H2.0 INTRODUCTION 363H6 3H2.1 What is Hydrogen Attac
16、k? 364H7 4H2.2 What is Fugacity and Why Is It Important 365H11 5H2.3 Details of the Attack Progression 366H12 6H3.0 ATTACK OF MATERIALS CONTAINING FE3C . 367H20 7H4.0 METHANE PRESSURE RELATIONS WITH CEMENTITE 368H23 8H5.0 STEELS WITH MORE STABLE CARBIDES 369H25 9H6.0 BEHAVIOR OF STEELS WITH COMPLEX
17、ALLOY CARBIDES . 370H26 10H7.0 RESEARCH STUDIES 371H27 11H8.0 STUDIES 372H28 12H9.0 EFFECT OF STRESS 373H32 13H10.0 EFFECT OF TEMPERATURE . 374H33 14H11.0 EFFECT OF HYDROGEN PRESSURE 375H34 15H11.1 Materials Issues . 376H34 16H12.0 EXPOSURE TIME OR STRESS 377H35 17H13.0 EUROPEAN STUDIES 378H36 18H14
18、.0 NEW ANALYSIS OF MPC WORK 379H40 19H15.0 DEVELOPMENT OF P/T LIMITS NEAR THE CREEP RANGE AND ABOVE 380H42 20H16.0 A TECHNICAL BASIS . 381H43 21H17.0 DISCUSSION . 382H45 22H18.0 FINAL COMMENT . 383H47 23H19.0 OVERVIEW AND CONCLUSION 384H49 24H20.0 REFERENCES . 385H51 25H21.0 FIGURES . 386H56 26HAPPE
19、NDIX A BACKGROUND TO THE NELSON CURVES . 387H124 27H1.0 HISTORY . 388H125 28H2.0 WHAT DATA IS AVAILABLE? . 389H128 29H3.0 WHERE DID THE DATA COME FROM? 390H129 30H3.1 API Publication 941 Fifth Edition Supplement 1 April 1998 391H129 31H3.2 Figure 1 Points C-0.5Mo Appendix A . 392H130 32H3.3 File Cor
20、respondence but Points Not Plotted 393H135 33H3.4 Points Plotted for 1-1/4Cr-1/2Mo and 2-1/4Cr-1Mo . 394H138 34H3.5 Attack / No Attack Points 395H139 35H4.0 DISCUSSION . 396H141 36H5.0 CONCLUSIONS . 397H145 37H6.0 FIGURES . 398H146 Report to API - The Technical Basis For API RP 941 38HAPPENDIX B COM
21、MONLY ASKED QUESTIONS . 399H157 39H1.0 HOW DO WE GET CARBON ACTIVITY AND CONTENT? . 400H158 40H1.1 How Much Carbon Is Necessary To Exhaust The Carbide Formers? 401H161 41H2.0 ARE THE API RP 941 CURVES WHERE THEY BELONG? . 402H162 42H3.0 WHAT VARIABLES MIGHT ADVERSELY INFLUENCE BEHAVIOR IN LONG TERM
22、SERVICE? . 403H163 43H4.0 WHAT OTHER FACTORS MIGHT PLAY A ROLE IN THE APPEARANCE OF HYDROGEN ATTACK? 404H164 44H5.0 WHY IS HYDROGEN ATTACK SEEMINGLY UNPREDICTABLE? 405H165 45H6.0 WHY HAVENT THE NECESSARY CRITICAL EXPERIMENTS BEEN RUN? . 406H166 46H7.0 WHAT IS ACTUALLY HAPPENING DURING EARLY STATES O
23、F HYDROGEN ATTACK? . 407H167 47H8.0 REFERENCES . 408H168 APPENDIX C ESTIMATING DAMAGE RATES FOR LIFE ASSESSMENT 169 48HAPPENDIX D EFFECTIVE PRESSURES OF HYDROGEN IN STEEL COVERED BY CLAD/OVERLAY AND/OR CORROSION PRODUCT 409H178 49HAPPENDIX E OBSTACLES TO UNDERSTANDING HTHA 410H184 50HAPPENDIX F APPL
24、ICATION AND SUMMARY OF PW PARAMETRIC MODEL 411H187 51H1.0 SUMMARY OF THEORETICAL KINETIC TREATMENTS ON BUBBLE GROWTH . 412H189 52H1.1 Diffusion Models - Grain Boundary and Surface Diffusion . 413H189 53H1.2 Coupled Diffusion and Creep Model 414H189 54H2.0 RELATIONSHIP BETWEEN THEORETICAL EQUATIONS A
25、ND ARRHENIUS TYPE EXPRESSIONS . 415H191 55H2.1 Grain Boundary Diffusion (GBD) Process . 416H191 56H2.2 Surface Diffusion (SD) Process 417H192 57H2.3 Coupling Between GBD and SD Processes 418H192 58H2.4 Power-Law Creep Process 419H192 59H2.5 General Form of All Processes . 420H192 60H3.0 DERIVATION O
26、F ENGINEERING PARAMETER PW 421H194 61H4.0 EXAMPLE OF EXPERIMENTAL RESULTS RELATIVE TO THE ARRHENIUS TYPE EXPRESSION 422H196 62H5.0 APPLICATION OF THE PW PARAMETER PREDICTION OF TIME DEPENDENT CRITICAL CURVES 423H198 63H5.1 Discussion of the Values of and Q . 424H198 64H5.2 Discussion of API RP 941 v
27、ersus Parametric Equations for Several Materials 425H200 65H6.0 APPLICATION OF THE PW PARAMETER ESTIMATION OF THE EFFECT OF APPLIED STRESS . 426H205 66H6.1 Derivation of Pw Including the Stress . 427H205 67H6.2 Effect of Applied Stress on the Critical Curve of 2.25Cr-1Mo Steel . 428H206 68H7.0 APPLI
28、CATION OF THE PW PARAMETER ESTIMATION OF EFFECT OF OVERLAY 429H207 69H7.1 Hydrogen Distribution in a Vessel Wall with Overlay During Operation 430H207 70H7.2 The Effect of Overlay on the Critical Curve . 431H208 71H8.0 CONCLUSIONS . 432H210 72H9.0 REFERENCES . 433H211 73H10.0 FIGURES . 434H212 Repor
29、t to API - The Technical Basis For API RP 941 74HAPPENDIX G OVERVIEW OF EUROPEAN RESEARCH 435H231 75H1.0 INTRODUCTION 436H232 76H2.0 NON-UNIFORM HYDROGEN ATTACK CAVITATION AND THE ROLE OF INTERACTION WITH CREEP . 437H235 77H2.1 Introduction 438H235 78H2.2 Polycrystal Model for HA 439H236 79H2.3 Meth
30、od of Analysis 440H242 80H2.4 Discussion 441H248 81H2.5 Conclusion 442H250 82H3.0 COMPARISON OF DELFT AND APPROACH REPORTED FROM JAPAN . 443H252 83H3.1 Introduction to the Hydrogen Attack Models 444H252 84H3.2 Calculation of the Methane Pressure . 445H253 85H3.3 Comparison Between Methane Pressures
31、446H254 86H3.4 Description of Void Growth Due to Diffusion 447H255 87H3.5 Comparison of Predicted Void Growth Due To Diffusion 448H256 88H3.6 Creep As An Additional Deformation Mechanism 449H258 89H3.7 Conclusions 450H260 90H4.0 FIGURES . 451H261 91H5.0 REFERENCES . 452H292 92HAPPENDIX H SAMPLE PROB
32、LEMS UTILIZING APPENDIX C AND DISCUSSION DEVELOPED BY SOME COMMITTEE MEMBERS DURING REVIEW OF THE REPORT 453H294 93H1.0 ESTIMATING DAMAGE RATES FOR LIFE ASSESSMENT: BASIC METHODOLOGY 454H295 94H2.0 EXAMPLES: HYPOTHETICAL CASES USING APPENDIX C 455H296 95H2.1 Case 1: Carbon Steel channel flange and s
33、hell welds discovered in 1.25Cr exchanger . 456H296 96H2.2 Case 1A: 225 psia at 620oF . 457H296 97H2.3 Case 1B: 225 psia at 750oF 458H297 98H2.4 Case 2: Carbon steel pipe . 459H297 99H2.5 Case 3: C-0.5Mo exchanger channel cover . 460H297 100H2.6 Case 4: C-0.5Mo Methanator Short-Time Excursion . 461H
34、298 101H2.7 Case 5: Mn-0.5Mo exchanger with austenitic weld overlay/clad 462H299 102H2.8 Case 5A: Consider Continued Operation at Current Conditions 463H299 103H2.9 Case 5B: Operations wants to increase the operating conditions . 464H300 104H2.10 Case 6: 1Cr catalytic reformer reactor 465H300 105H2.
35、11 Case 7: Over temperature Conditions in a 2.25Cr 1 Mo Reactor . 466H300 Report to API - The Technical Basis For API RP 941 1 EXECUTIVE SUMMARY FOR THE TECHNICAL BASIS DOCUMENT Before the first edition of API Publication 941 “Steels for Hydrogen Service at Elevated Temperatures and Pressures in Pet
36、roleum Refineries and Petrochemical Plants” appeared in 1970, there had been fundamental questions regarding the technical basis for the materials performance curves contained in the document (1-6). Based upon sparse laboratory data combined with plant experience, with only a few exceptions, the cur
37、ves have done an exceptionally good job at safely directing the refining industry in selecting materials based upon operating temperature, hydrogen partial pressure, and the metallurgy of the equipment being considered. However, in some cases, past editions of the API RP 941 document were of limited
38、 value, most notably for C-0.5Mo material. Today, with refining plants aging, engineers are seeking assurances that the curves are suitable for predicting continuing satisfactory performance for decades into the future. Of concern also, is the unusual shape of the 1.25Cr curve which appears inconsis
39、tent with the other curves without any obvious technical reason. Most important, engineers require technical justification for decisions made regarding suitability for service after process excursions that exceed the 941 “safe” limits. The API Subcommittee for Corrosion and Materials commissioned th
40、e 941 Task Group to provide a technical basis document that goes beyond empirical evidence to address three issues: 1. Do the curves given in the current edition have the correct shapes and locations? 2. Are the curves likely to “change” with time as our plants become older? 3. What methodology and
41、data can be used to handle process excursions? The Task Group met these objectives. The Technical Basis Document (TBD) that follows is the result of several years of effort incorporating the technical insights contributed by participating Japanese and European specialists with those presented to API
42、 from the United States. Dr. Martin Prager of The Materials Properties Council Inc. (MPC) prepared much of the TBD and developed the approach set forth in the main body of this document. Please see the acknowledgements of the contributions of the others noted below. Details can be found in the respe
43、ctive Appendix sections It is important in considering this work that it is a research report, not a recommended practice. Those workers most closely involved in this report believe some of the findings are so well supported that they can be immediately brought into the next edition of the 941 Recom
44、mended Practice. Examples are identified below. Other findings push the edge of our understanding and give very useful insight without yet being RP-ready. This is hard work, the complexity of which is matched by a frustrating lack of quality data in many cases. What goes into the next API RP 941 wil
45、l be the work of the overall 941 Task Group. Immediately below are highlights of the attached Technical Basis Document. For many, this level of detail along with selected portions of the document to handle specific issues will suffice. Those who delve into the entire document will be rewarded by Dr.
46、 Pragers elegant explanations that bring us down to the mechanistic level, and back out again to provide guidance in dealing with actual plant challenges. Report to API - The Technical Basis For API RP 941 2 Highlights of the Draft Technical Basis Report for API RP 941 1.) The shapes and locations o
47、f the curves in the most recent RP 941 are essentially correct. a) The carbon steel curve appears to be perhaps 30oF to as much as 50oF conservative. However, there is insufficient laboratory or plant data to justify adjusting the curve, and it is important to note that under special circumstances (
48、e.g., unusually high stresses) there have been failures even below the current carbon steel curve. b) All curves should have essentially the same shape. Therefore the unusual “kink” in the 1.25Cr curve is likely incorrect. The 941 Task Group should consider adjusting this curve. c) The shape of the
49、curves, where they go essentially “vertical” at low hydrogen partial pressures and become almost flat at high pressures, can be understood by taking into account both kinetics, thermodynamics and materials strength. To a remarkable extent, attack quantitatively tracks the hydrogen and carbon solubility, which are low at low temperatures. Nevertheless, at low temperatures the methane pressure formed from even small concentrations of these elements can be enormous, much higher than the strength of the materials. Fortunately the kinetics are very slow at low temp
