Document reEgj1DqK1362ZBVzYykqawkG

MONSANTO CHEMICAL COMPANY TEXAS CITY, TEXAS December 31> 1959 To: J. S. Putnam Plant Manager's Safety Board Subject: Flash Fire at VCM Sample Tap - December 15, 1959 SUMMARY 4* On December 15, at about noon, while an operator was sampling VCM a small falsh fire occurred at the mouth of the sample bottle. The fire was extinguished by the operator closing the sample valve. No alarm was turned In. There was no Injury to personnel or damage to equipment. COMMENTS , Between 11:45 and 12:00 noon, an operator went out to catch a routine product VCM sample at 22P16-1, the VCM fractionator overhead pump. This sample is caught in a glass thermos bottle due to Its low boiling point. The sample tap used in this case is normally used to catch samples in metal bomb and was fitted with a short piece of polyethylene tubing attached to stainless steel tubing with swegelok fittings. General practice has been to.catch a thermos sample from the pump drain connection. Following normal practice, the operator opened the sample valve to set the flow rate, put a small amount of VCM in the thermos, rinsed the bottle and poured out the VCM. This step was repeated once. The operator then held the thermos up to catch the sample, still holding the bottle in his right hand and the end of the sample line in his left hand. Shortly after starting to fill the thermos, a flash occurred at the mouth of the thermos. (it Is believed that there was only a small amount of liquid in the bottle at the time the flash occurred.) The operator dropped the bottle and ran a few steps away. Some of the VCM vapor on the ground undoubtedly burned also. The fire on the ground subsided almost Immediately as the fuel was exhausted, but burning continued at the end of the tap. The operator closed the sample valve and the fire went out. samp SC 16640 LAM023909 2 COMMENTS There was no injury to personnel and no damage to equipment. Background information: a. Weather - humidity was 100$ (raining). Temperature was in the SO's. b. Data on VCM. 1. Explosive limits ^-22$ by vol. 2. Flash Point 20F c. Conditions at sample point. Pump discharge pressure - 60-65 psig. Temperature of liquid stl^am in pump -5C (23F) Atmosphere boiling point of VCM -l4C (7F) * The most probable cause of the fire was a static spark. Possible causes of the spark are (l) Flow through the polyethylene tubing with a charge collecting on the metal fitting at the end of the tubing. (2) Static generated by fall of the VCM liquid into the thermos. (3) Lack of grounding between sample tap and metal case of the thermos. The polyethylene tubing had been in service about two months. The number of times it was used to catch a thermos sample is unknown since most operators use the pump drain connection. This is the first flash fire in the area involving VCM since the unit was started up in 1952. A fire occurred in a control lab hood which was believed caused by smoking or a static spark. ACTION TO BE TAKEN 1. The polyethylene tubing was removed immediately. 2. A ground connection was made up to connect the metal case of the thermos to the stainless steel sample tap while sampling. Operators were Instructed through the log book on the necessity of grounding the metal case of the thermos and the stainless sample. SC 16641 LAM023910 -3- RECOMMENDATIONS The Safety Department is requested to review this incident with Bureau of Mines personnel to determine if their analysis of the cause and prevention of recurrence of this type fire is in agreement with ours. J. W. Kongable /Js 12-31-59 SC 16642 LAM023911 . 03 /.ft) *' :.'K ` 7t> ct ow H '& I ! .V-*" VO 5--' \vJo1 SC 16639 LAM023908