《NCHRP-WEB-139.pdf》由会员分享，可在线阅读，更多相关《NCHRP-WEB-139.pdf（143页珍藏版）》请在三一文库上搜索。
1、Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105, Determined Using CCRL Proficiency Sample Data National Cooperative Highway Research Program Haleh Azari Robert Lutz AASHTO Materials Reference Laboratory Gaithersburg, MD Contractors Final Task Report for NCHRP Project 09-26A S
2、ubmitted June 2009 NCHRP ACKNOWLEDGMENT This work was sponsored by the American Association of State Highway and Transportation Officials (AASHTO), in cooperation with the Federal Highway Administration, and was conducted in the National Cooperative Highway Research Program (NCHRP), which is adminis
3、tered by the Transportation Research Board (TRB) of the National Academies. COPYRIGHT PERMISSION Authors herein are responsible for the authenticity of their materials and for obtaining written permissions from publishers or persons who own the copyright to any previously published or copyrighted ma
4、terial used herein. Cooperative Research Programs (CRP) grants permission to reproduce material in this publication for classroom and not-for-profit purposes. Permission is given with the understanding that none of the material will be used to imply TRB, AASHTO, FAA, FHWA, FMCSA, FTA, Transit Develo
5、pment Corporation, or AOC endorsement of a particular product, method, or practice. It is expected that those reproducing the material in this document for educational and not-for-profit uses will give appropriate acknowledgment of the source of any reprinted or reproduced material. For other uses o
6、f the material, request permission from CRP. DISCLAIMER The opinion and conclusions expressed or implied in the report are those of the research agency. They are not necessarily those of the TRB, the National Research Council, AASHTO, or the U.S. Government. This report has not been edited by TRB. N
7、CHRP Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105 iii TABLE OF CONTENTS LIST OF TABLES v ACKNOWLEDGMENTS . vi CHAPTER 1. INTRODUCTION AND RESEARCH APPROACH . 1 1.1 INTRODUCTION 1 1.1.1 Problem Statement 2 1.1.2 Research Objectives . 2 1.2 SCOPE OF STUDY . 2 1.3 PROFICIENCY
8、 SAMPLES USED IN STUDY 2 CHAPTER 2. RESULTS OF ANALYSIS AND ESTIMATES OF PRECISION 4 2.1 TEST DATA 4 2.2 ANALYSIS OF THE DATA . 4 2.2.1 Silicon Dioxide (SiO2) 5 2.2.2 Aluminum Oxide (Al2O3) . 5 2.2.3 Ferric Oxide (Fe2O3) . 6 2.2.4 Calcium Oxide (CaO) . 7 2.2.5 Magnesium Oxide (MgO) 7 2.2.6 Sulfur Tr
9、ioxide (SO3) . 8 2.2.7 Loss on Ignition (LOI) 9 2.2.8 Sodium Oxide (Na2O) 9 2.2.9 Potassium Oxide (K2O) 10 2.2.10 Titanium Dioxide (TiO2) 11 2.2.11 Chloride (Cl) . 11 2.2.12 Insoluble Residue (IR) 12 2.2.13 Free Calcium Oxide 12 2.2.14 Carbon Dioxide (CO2) 13 2.2.15 Comparison of the Existing and De
10、veloped Precision Estimates 14 CHAPTER 3. CONCLUSIONS AND RECOMMENDATIONS . 15 3.1 COMMENTARY . 15 3.2 CONCLUSION 15 3.3 RECOMMENDATIONS . 15 NCHRP Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105 iv 3.4 PRECISION STATEMENT FOR AASHTO T 105, “CHEMICAL ANALYSIS OF HYDRAULIC CEM
11、ENT” 16 REFERENCES 17 APPENDIX A: SILICON DIOXIDE (SIO2) 18 APPENDIX B: ALUMINUM OXIDE (AL2O3) . 27 APPENDIX C: FERRIC OXIDE (FE2O3) . 36 APPENDIX D: CALCIUM OXIDE (CAO) . 45 APPENDIX E: MAGNESIUM OXIDE (MGO) 54 APPENDIX F: SULFUR TRIOXIDE (SO3) 63 APPENDIX G: LOSS ON IGNITION . 72 APPENDIX H: SODIU
12、M OXIDE (NA2O) 80 APPENDIX I: POTASSIUM OXIDE (K2O) 91 APPENDIX J: TITANIUM DIOXIDE (TIO2) . 100 APPENDIX K: CHLORIDE (CL) 109 APPENDIX L: INSOLUABLE RESIDUE 115 APPENDIX M: FREE CALCIUM OXIDE (CX) . 124 APPENDIX N: CARBON DIOXIDE (CO2) 133 NCHRP Web-Only Document 139: Precision Estimates for AASHTO
13、 Test Method T 105 v LIST OF TABLES Table 1-1- Sample Designation and Date of Final Report of Proficiency Samples 3 Table 2-1- Chemical Components and Number of Data Sets Analyzed . 4 Table 2-2- Summary of Statistics for % Silicon Dioxide (SiO2) 5 Table 2-3- Summary of Statistics for % Aluminum Oxid
14、e (Al2O3) 6 Table 2-4- Summary of Statistics for % Ferric Oxide (Fe2O3) 6 Table 2-5- Summary of Statistics for % Calcium Oxide (CaO) 7 Table 2-6- Summary of Statistics for % Magnesium Oxide (MgO) . 8 Table 2-7- Summary of Statistics for % Sulfur Trioxide (SO3) 8 Table 2-8- Summary of Statistics for
15、% Loss on Ignition (LOI) . 9 Table 2-9- Summary of Statistics for % Sodium Oxide (Na2O) 10 Table 2-10- Summary of Statistics for % Potassium Oxide (K2O) . 10 Table 2-11- Summary of Statistics for % Titanium Dioxide (TiO2) . 11 Table 2-12- Summary of Statistics for % Chloride (Cl) 12 Table 2-13- Summ
16、ary of Statistics for % Insoluble Residue (IR) . 12 Table 2-14- Summary of Statistics for % Free Calcium Oxide (Free CaO) 13 Table 2-15- Summary of Statistics for % Carbon Dioxide (CO2) . 13 Table 2-16- Comparison of the Existing and Proposed Precision Estimates . 14 NCHRP Web-Only Document 139: Pre
17、cision Estimates for AASHTO Test Method T 105 vi ACKNOWLEDGMENTS The research reported herein was performed under NCHRP Project 9-26A by the AASHTO Materials Reference Laboratory (AMRL). Dr. Haleh Azari and Mr. Robert Lutz were co-principal investigators on the study. Mr. Robin Haupt the CCRL Profic
18、iency Sample Program Manager played supporting roles and provided valuable comments in the study. NCHRP Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105 1 CHAPTER 1. INTRODUCTION AND RESEARCH APPROACH 1.1 INTRODUCTION Under National Cooperative Highway Research Programs (NCHRP
19、) Project 9-26, the AASHTO Materials Reference Laboratory (AMRL) is conducting a multi-phase research project to improve estimates of precision in AASHTO test methods for various highway construction materials. The report from Phase 1 of Project 9-26 includes precision estimates of selected volumetr
20、ic properties of HMA using non-absorptive aggregates 1. The report from Phase 2 discusses the results of an investigation into the cause of variations in HMA bulk specific gravity test results using non-absorptive aggregates 2. The report from Phase 3 includes a robust technique developed by AMRL fo
21、r analyzing proficiency sample data for the purpose of obtaining reliable single-operator and multilaboratory estimates of precision 3. The report from phase 4 includes two parts. Part one covers the precision estimates of selected volumetric properties of HMA using absorptive aggregates. Part two o
22、f the report investigates the effect of aging period on the volumetric properties of the absorptive aggregates 4. The report from Phase 5 includes update of precision estimates for AASHTO Standard Test Method T 269 5. This report includes the results for Part 3 of the 3 in Task 1 of NCHRP 9-26 where
23、 data from the CCRL Proficiency Sample Program (PSP) collected from the state laboratories are used to update precision estimates for AASHTO Standard Test Method T 105, “Chemical Analysis of Hydraulic Cement” 6. Data used in this study are from the chemical analysis of hydraulic cement samples that
24、were sent to the state laboratories participating in the CCRL Proficiency Program. The laboratories receive annual or biannual shipments of CCRL paired proficiency samples, which are tested according to specified ASTM test methods 7. The hydraulic cement samples for the chemical analysis were prepar
25、ed and tested according to the methods explained in ASTM C 114 8. The data from 14 chemical components were analyzed and are provided in this report. The proficiency samples included in these programs cover a wide range of test values and cement types. The technique developed by AMRL in Phase 3 was
26、utilized for analyzing proficiency sample data. This technique is a four step methodology for shaving off extraneous results and analyzing the core data of a paired data set. The results of the analysis of the “core data” can then be used to obtain reliable single-operator and multilaboratory estima
27、tes of precision. The precision statement for Chemical Analysis of Hydraulic Cement Test Method in this study resulted from analysis of 107 paired data sets from 14 chemical tests on seven shipments of CCRL paired proficiency samples. Only the most recent proficiency samples were used in order to ac
28、count for changes in test precision resulting from recent improvements in the test methods. NCHRP Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105 2 1.1.1 Problem Statement AASHTO Standard Test Methods applicable to highway materials require periodic studies to determine estim
29、ates of precision. Some precision estimates become outdated as a result of improvements in the methods while other estimates need to be verified to see if they are still accurate. Some test methods need to be expanded to take into account a wider range of materials while other newer test methods may
30、 not have precision estimates of any kind. AASHTO T 105 covers the test methods for chemical analysis of hydraulic cements. Although a set of specific chemical test methods are suggested in T 105, any test method of demonstrated acceptable precision and bias may be used for the analysis of hydraulic
31、 cements. To ensure the reliability of the user selected test methods, it is important that the precision estimates of T 105 are updated. The precision estimates also need to be expanded to include chemical components that are not included in AASHTO T 105-06 but are frequently measured by in state a
32、nd private laboratories. 1.1.2 Research Objectives The objective of this study as Part 3 of the 3 in task 1 of NCHRP 9-26A study is to verify, update, and expand single-operator and multilaboratory precision estimates for the AASHTO T 105, “Standard Method of Test for Chemical Analysis of Hydraulic
33、Cement.” Only the most recent CCRL proficiency cement samples that were tested according to the latest version of the test method will be analyzed. 1.2 SCOPE OF STUDY This work is limited to an evaluation of the data collected from the laboratories participating in the chemical analysis of hydraulic
34、 cement for the CCRL proficiency sample program. There are a total of 107 data sets analyzed and included in this report. The resulting precision estimates will reflect a wide range of test values and cement types that are included in the scope of the CCRL Proficiency Sample Program. 1.3 PROFICIENCY
35、 SAMPLES USED IN STUDY Included in the study are the most recent CCRL proficiency samples that were tested in accordance to AASHTO T 105-06 (ASTM C 114-05), “Standard Test Method for Chemical Analysis of Hydraulic Cement.” The various cement types included in the study are Type I and Type I/II with
36、and without limestone and Type V with limestone. Table 1-1 provides the sample designation and sample type of the CCRL cement samples and the date of the final report on the chemical analysis of the samples. NCHRP Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105 3 Table 1-1- S
37、ample Designation and Date of Final Report of Proficiency Samples Sample Designation Cement Type Date of Final Report 147 therefore, the form of the precision estimates should be based on the sample standard deviation. The pooled repeatability sample standard deviation for the eight pairs of samples
38、 analyzed is 0.119 percent. The corresponding pooled reproducibility sample standard deviation is 0.196 percent. The pooled estimates are derived using the following equation from Ku 9: ()()() knnn snsnsn s k kk p + + = . 1.11 2 22 22 2 11 1 (Equation 1) Where: = p spooled standard deviation = k s k
39、th standard deviation = k n number of laboratories analyzed resulting in kth standard deviation Table 2-2- Summary of Statistics for % Silicon Dioxide (SiO2) Sample Number No. of Labs Average Results Repeatability Reproducibility Reproducibility Odd Samples Even Samples 1s Odd Samples Even Samples O
40、dd Samples Odd Samples Even Samples Even Samples CV% CV% 1s CV% 1s CV% 147 therefore, the form of the precision estimates should be based on the NCHRP Web-Only Document 139: Precision Estimates for AASHTO Test Method T 105 6 sample standard deviation. The pooled repeatability sample standard deviati
41、on for the eight pairs of samples analyzed is 0.073 percent. The corresponding pooled reproducibility sample standard deviation is 0.110 percent. The pooled estimates are derived using Equation 1. Table 2-3- Summary of Statistics for % Aluminum Oxide (Al2O3) Sample Number No. of Labs Average Results
42、 Repeatability Reproducibility Reproducibility Odd Samples Even Samples 1s Odd Samples Even Samples Odd Samples Odd Samples Even Samples Even Samples CV% CV% 1s CV% 1s CV% 147 therefore, the form of the precision estimates should be based on the sample standard deviation. The pooled repeatability sa
43、mple standard deviation for the eight pairs of samples analyzed is 0.029 percent. The corresponding pooled reproducibility sample standard deviation is 0.051 percent. The pooled estimates are derived using Equation 1. Table 2-4- Summary of Statistics for % Ferric Oxide (Fe2O3) Sample Number No. of L
44、abs Average Results Repeatability Reproducibility Reproducibility Odd Samples Even Samples 1s Odd Samples Even Samples Odd Samples Odd Samples Even Samples Even Samples CV% CV% 1s CV% 1s CV% 147 therefore, the form of the precision estimates should be based on the sample standard deviation. The pool
45、ed repeatability sample standard deviation for the eight pairs of samples analyzed is 0.199 percent. The corresponding pooled reproducibility sample standard deviation is 0.384 percent. The pooled estimates are derived using Equation 1. Table 2-5- Summary of Statistics for % Calcium Oxide (CaO) Samp
46、le Number No. of Labs Average Results Repeatability Reproducibility Reproducibility Odd Samples Even Samples 1s Odd Samples Even Samples Odd Samples Odd Samples Even Samples Even Samples CV% CV% 1s CV% 1s CV% 147 therefore, the form of the precision estimates should be based on the sample standard d
47、eviation. The pooled repeatability sample standard deviation for the eight pairs of samples analyzed is 0.049 percent. The corresponding pooled reproducibility sample standard deviation is 0.070 percent. The pooled estimates are derived using Equation 1. NCHRP Web-Only Document 139: Precision Estima
48、tes for AASHTO Test Method T 105 8 Table 2-6- Summary of Statistics for % Magnesium Oxide (MgO) Sample Number No. of Labs Average Results Repeatability Reproducibility Reproducibility Odd Samples Even Samples 1s Odd Samples Even Samples Odd Samples Odd Samples Even Samples Even Samples CV% CV% 1s CV
49、% 1s CV% 147 therefore, the form of the precision estimates should be based on the sample standard deviation. The pooled repeatability sample standard deviation for the eight pairs of samples analyzed is 0.047 percent. The corresponding pooled reproducibility sample standard deviation is 0.076 percent. The pooled estimates are derived using Equation 1. Table 2-7- Summary of Statistics for % Sulfur Trioxide (SO3) Sample Number No. of Labs Average Results Repeatability Reproducibility Reproducibility Odd Samples Even Samples 1s Odd Samples Even Samples Odd