Michelle K. Donnelly
2004
Citations
0
Influential Citations
2
Citations
Journal
Journal name not available for this finding
Abstract
This research investigates the stability of iodotrifluoromethane (CF3I) during storage. For CF3I to be used as a fire extinguishing agent, it must be able to be stored for long periods of time at high pressure in metal containers without degrading and losing its effectiveness. For these experiments, CF3I was placed into cylinders, along with various metal coupons. The cylinders were stored, some at elevated temperatures in an oven, for three years and then placed into storage at ambient temperature for five more years. Infrared spectroscopy was used to analyze the cylinder contents and concentrations at various times during the storage. This report focuses on changes to the cylinder contents during the final five year storage at ambient temperature by comparing spectra taken before storage to spectra collected after storage. Analyses of the spectra showed the volume fraction of CF3I agent remained stable to within the measurement uncertainty of ± 0.012 during the five years of storage at ambient temperature. For the infrared-active compounds that could be measured with the spectrometer, no new peaks developed in the spectra during ambient temperature storage. Evaluations of the other components showed that the amount of CO2 present in the cylinders decreased by 63 % or more and the amount of CF3H decreased by 29 % or more. No new F-alkene peaks developed, and changes to existing F-alkene peaks did not significantly affect the CF3I mass. The combination of both copper and elevated temperatures caused degradation of CF3I, and the presence of other metals affected the severity of this degradation. In particular, the combination of copper and Nitronic 40 resulted in a complete breakdown of the CF3I agent during storage at 150 ̊C. INTRODUCTION This set of experiments is a continuation of an effort to determine the viability of using alternative agents to extinguish fires. These alternative agents are needed because production of the commonly used halon 1301 was stopped in 1994 due to its adverse effects on the environment. Consequently, a number of replacement fire extinguishing agents were identified, including iodotrifluoromethane (CF3I). The Building and Fire Research Laboratory of the National Institute of Standards and Technology conducted a series of investigations to evaluate the performance and characteristics of CF3I and other replacement agents (Grosshandler et al., 1994, Gann, 1995). One important attribute that was investigated was the long-term stability of an agent during storage. Halon 1301 remains stable and effective as a fire extinguishing agent during years of storage in metal containers. For CF3I to work effectively as an extinguishing agent, it must also remain stable under storage. If the agent reacts with impurities in the system or with the storage container, it may degrade and could become ineffective and/or produce toxic byproducts. A study of CF3I agent stability during storage was originally investigated in 1994 (Peacock et al., 1994) and continued in 1995 (Harris, 1995). The tests examine the compatibility of CF3I with a variety of metals that may be used to store agents. The four metals selected for these tests were C4130 alloy steel, Ti-15-3-3-3 titanium alloy, Nitronic 40 stainless steel, and Inconel 625 nickel alloy. All of the storage containers were stainless steel cylinders with a volume of one liter and lined with polytetrafluoroethylene. Metal coupons made from one of the four test metals listed above were inserted in the cylinders. The coupons had roughly the same surface area as the inside of the cylinders to simulate the agent contact area if the cylinder had been made of that metal. Some cylinders were left without metal coupons as a control. In addition to the test 1 metals, half of the cylinders received metal coupons made of copper (type CDA 110) as a variable, since copper reacts with iodine to form the nearly insoluble cuprous iodide. Another variable was the moisture content. While some cylinders contained only dry materials, some cylinders received 100 μL of distilled water. The cylinders were then filled with a mixture of CF3I vapor and nitrogen to a pressure of 4.13 MPa as measured at 23 ̊C. A detailed description of the cylinder preparation and filling is described in Harris (1995). During the initial testing period, from 1993 to 1997, each cylinder was stored at its assigned temperature of either 100 ̊C, 150 ̊C, or ambient temperature (approximately 23 ̊C). The gas samples were evaluated using Fourier transform infrared (FTIR) spectroscopy. This process measures the ratio of the intensity of infrared radiation passed through a sample to the intensity of incident radiation as a function of frequency. The ratio is used to create an infrared spectrum of the sample, from which compounds and their concentrations may be determined. The FTIR analysis method was chosen as a simple and accurate way to evaluate the cylinder contents and determine if changes were occurring (Peacock et al., 1994). A preliminary investigation was conducted to examine the results of storage for one month at elevated temperatures (Peacock et al., 1994). Based on these results, a long-term evaluation was initiated. Samples were prepared in February and March of 1994 and were tested on a continuous basis through May 1995. Many of the cylinders kept at either 100 ̊C or 150 ̊C during the initial storage showed decomposition of the CF3I agent (Harris, 1995). Most of these cylinders developed CF3H peaks and many developed peaks attributed to partially fluorinated alkenes (F-alkenes) during the heated storage (Harris, 1995). In March 1997 interim measurements of these samples were made, but no formal report was issued at that time. Thus the salient findings are summarized here. The most notable result of the 1997 measurements was that the CF3I peak was completely gone for the cylinders stored at 150 ̊C and containing both copper and Nitronic 40 coupons. This is in contrast to the cylinder containing both copper and Nitronic 40 stored at ambient temperature, which showed no decrease in CF3I peak area. The 1997 results also showed the contents of cylinders stored at 150 ̊C and that contained only copper or copper plus Inconel 625 had CF3I peak decreases of 30 % to 60 %. The cylinders stored at 150 ̊C with copper and Ti-15-3-3-3, or copper and C4130 alloy steel had decreases in CF3I peak size of less than 15 %. The cylinders stored at 150 ̊C without copper, the cylinders stored at 100 ̊C, and the cylinders stored at ambient temperature did not experienced a measurable change in CF3I peak area. After the 1997 interim measurements were collected, all of the cylinders were placed into storage at ambient temperature until the final testing in March 2002. This paper focuses on comparisons between the 1997 spectra and the 2002 spectra to identify if further changes occurred to the contents of the CF3I cylinders. EXPERIMENTAL PROCEDURE For this investigation FTIR spectroscopy was again used to identify the type and concentration of components in the samples. In the previous investigation, a Galaxy Series 7000 FTIR Spectrometer was used to perform the analyses. Since those tests were completed, this spectrometer was damaged and no longer available, so a MIDAC Illuminator FTIR spectrometer was used for the data collection in March 2002. Because the infrared spectrum of a compound is assumed to be unique, spectral data collected with the two different FTIR spectrometers may be compared. The MIDAC instrument covers a spectral range from 600 cm to 4500 cm. 2 The samples were measured inside a 5.4 liter gas cell with a path length of 10 m. The cell was heated to 105 ̊C ± 1 ̊C. The procedure was to evacuate the gas cell and then place a small amount of the sample gas into the cell. The spectrometer was then used to send an infrared signal through the gas cell and a spectrum was obtained for each gas sample. A diagram of the testing assembly is shown in Figure 1. The cylinder to be tested was attached to the inlet valve of the cell. The cylinder valve was kept tightly closed while the inlet valve and micro-metering valve were opened to allow for evacuation of the tubing between the cylinder valve and the inlet valve. The gas cell and its connections were filled to atmospheric pressure with nitrogen and then evacuated to below 0.01 kPa three times. Following this, nitrogen was allowed to flow through the cell for 45 s to flush any residual contaminants. The gas cell system was again evacuated to below 0.01 kPa. The micro-metering valve was then closed while the inlet valve remained open. The cylinder valve was slowly opened and the micro-metering valve was used to control the flow of gas sample into the cell. The cell was filled to 0.05 kPa with the cylinder contents. One spectrum consisting of 128 scans was collected for the sample in the cell. The cell was then evacuated below 0.01 kPa, refilled to 0.05 kPa, and another spectrum collected. This was then repeated for a total of three fills at 0.05 kPa. These low-pressure fills flushed the cell with the actual sample to help obtain an accurate reading for the final spectrum. After the third scan at 0.05 kPa, more of the sample was added until the cell pressure reached 5.33 kPa. A final spectrum of 128 scans was taken for this higher fill amount. The cylinder valve was closed tightly and the cell system was then evacuated and flushed with nitrogen as described above to prepare for the next sample. To avoid any exposure to the cylinder contents, the connected cylinder cell was not removed until after nitrogen flushing was completed. A new cylinder was put in place, and the entire process, including cell evacuation and nitrogen flushing, was repeated for each cylinder. During the 1997 testing phase of the project, some of the cylinder valves for the cylinders stored at elevated temperatures experienced failures and once opened, these valves could not be clos