IMPROVED STANDARDS FOR IODINE ADSORPTION NUMBER NU M B E By George Joyce R (COLUMBIAN CHEMICALS COMPANY, MARIETTA, GA) INTRODUCTION During the ASTM International D24 Meetings in 2001 Columbian Chemicals Company (CCC) presented information indicating that carbon blacks, specifically some of the standard reference tread blacks, have not exhibited a stable iodine adsorption number. Laboratory testing with tread SRB’s indicated their iodine numbers dropped continuously over time, and such an observation casts doubt as to the validity of using such materials as references or standards with assigned target values. Because of the importance of iodine adsorption testing within the carbon black industry and the need for carbon black iodine number standards, a Task Force was formed to address this issue. In June 2002 a D24 research report was presented entitled “Improved Standards for Carbon Black-Iodine Adsorption Number”. This report documents not only the problem of iodine number instability with tread blacks but also the reason for the problem, and most importantly a solution to improving the standards. This research report describes the manufacture of two new experimental standards, HT-B6 and HT-E6, designed to provide a stable iodine number for use as an improved standard. HT-B6 and HT-E6 are partially graphitized carbon blacks prepared from the SRB’s B-6 and E-6. Initial round robin testing of the new experimental standards was completed in December 2002. This study included data from 36 laboratories world-wide providing target values for the new experimental standards along with updated values for the SRB’s B-6 and E-6. The target values for these new HT standards along with B-6 and E-6 are also located on ASTM International’s website at www.astm.org. IODINE NUMBER INSTABILITY The recent revision to ASTM D4821 contains both discussion and actual laboratory test data describing the potential problems typically associated with iodine number testing. These are found in sections 6.1.1 and Figures 1 and 3 of the standard. Figure 1, also shown below, is a short-term x-chart of the SRB B-5 (N330) showing a target or mean value of 77.7 mg/g. The original published target value for SRB B-5 was 79.1 mg/g. Both target values for B-5 were obtained through ASTM round robin studies based on short-term testing within multiple laboratories, but were obtained at different points in time. The original study led to the target value of 79.1 mg/g, and a few years later a second study resulted in a target value of 77.7 mg/g. ASTM D4821 Figure 3 is an x-chart of the SRB B-5 iodine adsorption number within a single laboratory, and exhibits a continuous drop in iodine number over a three year period. This trend chart is shown below in Figure 2. 3 Year Iodine Adsorption X-Chart of B-5 80 79.5 79 78.5 78 Iodine No, mg/g 77.5 77 76.5 76 75.5 75 Figures 1 and 2 (right): SRB B-5 x-charts Another example of iodine number instability was demonstrated with the SRB B-6. The original published iodine number target value for SRB B-6 (N220) was 117.9 mg/g. This material was reevaluated for iodine number after 14 months by several companies and the results are shown in Table 1. The average difference in iodine number observed during this time was -1.4 mg/g. Because the 3-sigma x-chart limits for this material were originally calculated to be ± 2.28 mg/g, a change of -1.4 mg/g could easily result in laboratories wrongfully testing outside recommended x-chart limits. Table 1: SRB B-6 Iodine Number Change in 14 Months Company # Labs Avg Iodine No Avg Iodine No Iodine No on Nov 2000 on Jan 2002 (14 months) CCC 1 119.3 117.4 -1.9 Cooper Tire 1 117.9 115.4 -2.5 Degussa 1 118.4 117.3 -1.1 Cabot 1 118.6 117.6 -1.0 Continental 3 116.4 115.2 -1.2 Sid Richardson 4 116.3 115.6 -0.8 Summary 11 Avg = -1.4 With these well documented cases of iodine number instability, a need for improved materials for use as standards was identified. The ideal material would preferably be a tread carbon black and exhibit a stable iodine number over time. IODINE NUMBER STABILITY Based on well documented research on modified carbon black samples produced by Columbian Chemicals Company (CCC) in the early 1990’s1, CCC proposed the use of heat treated or partially graphitized carbon blacks as candidate material for improved iodine standards. CCC originally produced a series of N121 samples using an induction furnace to heat treat the carbon black at temperatures of 1000-1500°C under an inert atmosphere. These modified carbon blacks were thoroughly characterized for analytical and colloidal properties at the time of manufacture. This original information formed the basis for a 12 year study of iodine stability. The original analytical data for these samples are shown in Table 2. Table 2: Original Analytical Properties of Heat Treated N121 Samples Sample Iodine NSA micropore Lc Hydrogen Oxygen (mg/g) (m2/g) (m2/g) (nm) (ppm) (%) N121 Control 120.5 128 8 1.46 3046 1.69 N121 HT 1000°C 145.4 138 16 1.49 2820 0.69 N121 HT 1100°C 150.6 135 12 1.55 1965 0.47 N121 HT 1500°C 154.0 130 0 2.71 106 0.18 Both oxygen and hydrogen were measured by Leco using combustion analyzers. Micropore area was determined by difference of NSA and CTAB adsorption. These samples were characterized again in 2002 to observe the stability of specific properties including iodine number, oxygen and hydrogen content. These results are shown in Tables 2 and 3 and Figures 3 and 4. 1 J. A. Ayala, W. M. Hess, A. O Dotson and G. A Joyce, Rubber Chem & Tech, Vol. 63, No. 5, Nov 1990 Table 3. Iodine Number Stability of HT N121 Samples Sample Iodine No. Iodine No. Iodine No. (mg/g) (mg/g) (mg/g) 1988 1990 2002 N121 Control* 123.0 120.5 116.8 N121 HT 1000°C 145.4 136.9 N121 HT 1100°C 150.6 143.7 N121 HT 1500°C 154.0 153.9 * Manufactured in 1988. All samples were tested using 1.0g CB per 50mL iodine solution. Figure 2. Change in Iodine Number of N121 and HT N121 from date of manufacture. Table 4. Hydrogen and Oxygen Analysis of HT N121 Samples Sample Hydrogen Oxygen Hydrogen Oxygen (ppm) (%) (ppm) (%) 1990 1990 2002 2002 N121 Control 3046 1.69 3195 2.84 N121 HT 1000°C 2820 0.69 3176 1.98 N121 HT 1100°C 1965 0.47 2403 1.27 N121 HT 1500°C 106 0.18 179 0.13 Oxygen and Hydrogen measurements by Leco Figure 3. Change in oxygen content of HT N121 Samples over 12 years The data shown in Table 3 and Figure 2 indicate the N121 sample heat treated at 1500°C exhibits a stable iodine number during the 12 year period, within experimental error, while the N121 control changed -6.2 mg/g since manufacture. The iodine number instability observed with the N121 control and the samples heat treated at 1000 and 1100°C appears to be related to the increase in chemisorbed oxygen. For those skilled in the art, it is common knowledge that an increase in oxygen content of a carbon black due to oxidation will result in a reduced iodine number. Likewise, reducing the oxygen content of a carbon black by such means as devolatilization will result in an increased iodine number. Thus, the observed decrease in iodine number during the twelve year study for the N121 and N121 samples heat treated at 1000 and 1100°C is consistent with observed increases in oxygen content. A few other observations concerning iodine number of these samples are worth noting. First, the iodine number of the heat treated samples appears to increase with heat treatment temperature from 120.5 for the control to 154.0 for the HT 1500°C. This effect is related to the decreasing levels of oxygen on the carbon black. Another observation with these samples is the rate of change in iodine number for the N121 control was greatest within the first several years as indicated in Figure 2. Finally, the iodine number stability of the N121 HT 1500°C appears to be related to the fact that the oxygen content of this material did not change during the 12 year study, within experimental error. It should be noted that these samples were stored at ambient conditions in a warehouse without temperature control or special packaging. The stable oxygen content exhibited by the N121 HT 1500°C is believed to be the result of the observed change in micro-structure as measured by x-ray diffraction (Lc ). Based on these observations, a temperature of 1500°C appears to be the minimum temperature needed to impart a change in micro-structure that will render the carbon black surface resistant to oxygen chemisorption. HT-B6 and HT-E6 In January 2002 the D24.21 Iodine Number Task Force led by G. Joyce (Columbian Chemicals Company) and J. Wilson (Cabot Corp.) recommended the preparation of two heat treated carbon blacks, SRB B-6 (N220) and E-6 (N660). These materials were to be prepared in an induction furnace at 1500°C under an inert atmosphere precisely as the materials produced by Columbian in 1990. In April 2002, 50 lb. of HT- B6 and 100 lb. of HT-E6 were prepared by a US commercial carbon company. These experimental standards were initially forwarded to Columbian Chemicals Company for preliminary analysis. The initial analysis of the experimental HT-B6 and HT-E6 materials included oxygen content and iodine number. The results shown in Table 5 indicate the oxygen content of the heat treated carbons was 0.09- 0.10 percent. The SRB B-6 oxygen content was 1.32 percent, and E-6 was 0.60 percent. Based on this elemental analysis, the heat treatment removed most of the chemisorbed oxygen to levels slightly less than observed with the N121 HT 1500°C. Based on this elemental analysis, these two experimental standards exhibit a characteristic of partially graphitized carbon black. Table 5. Oxygen analysis of HT-B6 and HT-E6 and controls Material Furnace Oxygen Run Content (%) B-6 1.32 B-6 HT 1500°C 48 0.09 E-6 0.60 E-6 HT 1500°C 40 0.10 E-6 HT 1500°C 49 0.09 The next characterization of these experimental materials was to observe the level and uniformity of iodine number within each furnace run. Each of the three 50 lb. heat treatments were sampled for five aliquots then analyzed using 1.0g CB/50 cm3 iodine solution. The results shown in Table 6 indicate a mean value of 134 mg/g for the HT-B6 and 47.0 - 47.3 mg/g for the two runs of HT-E6. These values represent increases in iodine number of approximately 12-16 units based on the original target values of the SRB’s B-6 and E-6 as indicated in Table 7. Table 6. Iodine number uniformity of the Test B-6 E-6 E-6 HT 1500°C HT 1500°C HT 1500°C Run 48 Run 40 Run 49 1 134.0 47.2 47.5 2 134.1 46.8 47.2 3 134.1 47.5 46.9 4 133.8 46.7 47.8 5 133.8 46.8 47.2 Mean 134.0 47.0 47.3 Std Dev 0.15 0.34 0.34 Table 7. Change in iodine number from original SRB target values Standard Run Original Iodine Target No. * (mg/g) (mg/g) B-6 117.9 B-6 HT 1500°C 48 --- 134.0 16.1 E-6 35.3 E-6 HT 1500°C 40 --- 47.0 11.7 E-6 HT 1500°C 49 --- 47.3 12.0 Another issue discussed at this time concerning iodine number testing was related to the physical form of the sample prior to testing. Information was presented to ASTM D24.21 that indicated the size of the pellets used for the iodine test appeared to influence the absolute value and precision. Specifically, small bead sizes appear to result in higher iodine number than large beads. Therefore, this issue of physical form was addressed prior to distribution of the two experimental standards. Preliminary Precision Study to Determine the Physical Form of the Experimental HT Standards (Beads or Powder) A multi-lab precision study was designed to look at short-term precision of both beads and milled powder using the HT-B6 material. A second part of the study included a longer testing period within one lab to increase the number of observations (n) with both beads and powder. Forty eight aliquots of both beads and powder were distributed among 21 laboratories for a total of 96 samples. Each lab was asked to test both beads and powder in duplicate for two days using 0.5g CB/25 cm3 iodine (or 1.0g CB/50 cm3 iodine). A single lab analyzed both beads and powder in duplicate for 6 days. The results of the multi-lab study are shown in Table 8 and Figures 4 and 5. Table 8. Precision of physical form study Material Multi-lab Single lab Multi-lab Single lab Mean Mean Std Dev Std Dev n=4 n=12 n=4 n=12 Beads 133.8 133.7 0.32 0.32 Powder 134.2 134.6 0.26 0.40 Figure 4. HT-B6 iodine number study of physical form. Results indicate poor between-lab precision (R) for iodine number. The data shown in Figures 4 and 5 represent each lab’s iodine number for HT-B6, both beads and powder. The box plots in Figure 4 include both beads and powder. Figure 5 is presented as trend charts, one for beads and powder. The striking observation from this data is the very poor between-lab reproducibility observed, while data within labs tends to be very tight. This observation demonstrates very well the problem with iodine number testing within the carbon black industry today. Figure 5. HT-B6 iodine number study of physical form. Each lab’s mean iodine number is display in the trend charts. Conclusions from this study were largely based on the data presented in Table 8. These experiments indicate that test precision (standard deviation) for beads and powder were not sufficiently different to justify altering the physical form of the beads. In fact, the single-lab data (that has much higher degrees of freedom) indicates the beads exhibit slightly less variation than powder. Based on a one-way analysis of variance (ANOVA) of the multi-lab study, the difference in mean level for beads and powder was not statistically significant at a 95 percent confidence level, though the powder exhibited a slightly higher mean than beads. Based on these observations the data indicates there is no justification to alter the physical form of the experimental standards, and the two materials were subsequently distributed in their pelleted form. Initial Target Values for HT-B6 and HT-E6 Thirty-six laboratories participated in an ASTM International D24.21 Task Force Study to ascertain target values for the HT-B6 and HT-E6 standards and also obtain updated target values for SRB B-6 and E-6. Each lab was asked to test HT-B6 using 0.25g per 25 cm3 iodine solution and the HT-E6 with 0.5g per 25 cm3 iodine, in accordance with D1510. The resulting data was treated per D4483 for outliers and calculation of precision statistics. The precision data is presented in Table 9. The control chart target values and limits are presented in Table 10. This information is also posted to the ASTM website at www.astm.org. Table 9. Precision data for HT standards and updates for SRB’s Standard Mean Sr (r) SR (R) B-6 117.1 0.39 0.94 1.63 3.93 HT-B6 135.6 0.43 0.89 1.04 2.18 E-6 35.4 0.39 3.12 0.59 4.70 HT-E6 46.0 0.35 2.13 0.51 3.16 Table 10. Control chart limits for HT standards and updates for SRB’s Standard Target 3s LCL UCL B-6 117.1 1.17 115.9 118.3 HT-B6 135.6 1.28 134.3 136.9 E-6 35.4 1.17 34.2 36.6 HT-E6 46.0 1.04 45.0 47.0 The target iodine number for HT-B6 was observed to be 135.6 mg/g, and HT-E6 was 46.0 mg/g. These values represent significant increases compared to the original untreated carbon blacks, and as discussed above, is related to the removal of most of the chemisorbed oxygen from the carbon black surface. The precision data observed within this study indicates, as anticipated, a significant improvement in between-lab precision (R) for both experimental standards as compared to the SRB’s B-6 and E-6. The major contributor to this improvement in between-lab precision is the uniformity or consistency of the heat treated experimental standards. The current iodine number observed for the SRB B-6 as compared to the original published value represents a change of - 0.8 units. The current mean value is 117.1 compared to the original value of 117.9 mg/g. This change or decrease in iodine number is directionally consistent with changes observed with other tread blacks. In general, the decrease in iodine number for carbon blacks appears to be time dependent, and is very dependent on the total surface area. The effect of surface area is demonstrated within this sample set. The current iodine number of the SRB B-6 was 35.4 mg/g compared to the original published target of 35.3 mg/g. This difference of +0.1 mg/g is statistically insignificant, and no change is observed in the iodine number of the E-6 standard. In conclusion, only the tread grades have exhibited the instability with iodine testing, and the HT-B6 represents the first material produced to address this problem for the carbon black industry. SUMMARY The problem of iodine number instability with tread blacks has been demonstrated using carbon blacks manufactured by industry leading producers, and also using ASTM International Testing Standards that are recognized world-wide within the carbon black industry. The problem has been shown to be related to the slow process of chemisorbing oxygen on the surfaces of carbon black at ambient conditions. A solution to this problem using partially graphitized carbon black has been recognized, demonstrated, and pursued by D24 committee from a testing perspective. . RECOMMENDATIONS The D24 committee recommends use of these precision statistics and control chart limits as they are most current data available concerning the ASTM SRB B-6 and E-6 standards, along with the experimental HT-B6 and HT-E6 standards.