NOTE: THESE PAGES WERE SCANNED AND HAVE SOME ERRORS
SUMMARY OF EXPERIENCE IN DIRECT MEASURING INSTRUMENTS
FOR MISSILE PROPELLANTS
Philip Diamond, Certified Industrial scientist ‘
Anthony A. Thomas, M. D.
Aerospace Medical Research Laboratory, Wright-Patterson AFB
INTRODUCTION
Unsymmetrical Dimethylhydrazine, hydrazine and N02 are used as propellants in the Titan II ballistic missile system. These propellants are potentially toxic, and detection instrumentation operating on a continuous or intermittent basis is needed for the industrial Hygiene monitoring of these contaminants in air.
In the period of 1980-61, at the request of the SPO, commercially available instrumentation for UDMH and NO2 was evaluated. Since that time some instruments have been improved and are available with a wider range. Results are reported as of the time of testing.
METHODS
Known concentrations of UDMH and NO2 were produced using a metering-syringe, dynamic flow, dilution system (figure 1). This method of producing low level concentrations of volatile liquids and gases was developed by Dr. Robert Austin. A standard 100 nil hypodermic syringe was modified by fitting the plunger of the syringe with a finned cylinder of polystyrene so that when the syringe was mounted in a vertical position, the plunger could be made to rotate inside the barrel by directing an air stream at right angles to the polystyrene fin. A glass capillary was fitted to the end of the barrel and the flow from the syringe was adjusted by controlling the length and diameter of this capillary. The weight of the rotating. plunger then forces the cylinder contents out through the capillary. Calculated amounts of the toxic agent were placed inside the syringe and the contents diluted with dry air to the 100 ml mark and mixed. Some adsorption takes place on the syringe walls during the first two fillings, and equilibrium is achieved by the third filling. The resulting gas mixtures were then metered into the instrument sampling line. Different concentrations were produced by varying the amount of toxic agent placed in the syringe. Volume-volume dilutions of UDMH in benzene were prepared and suitable aliquots of these solutions were completely vaporize inside the syringe. The gaseous N2O4fl2NO2 was transferred to the metering-syringe using other hypodermic syringes of appropriate sizes. The system was allowed to reach equilibrium for’ each concentration prior to sampling. Accuracy of the generated concentrations of UDMH and NO2 was established using analytical colormetric procedures (Ref. 1. 2).
16
CALCULATIONS FOR SYRINGE METERING DEVICE
To calculate the amount of propellant to be added to the 100 ml. syringe the following information is incorporated in the calculations:
1. The critical orifice capillary delivers the contaminant gas at a contaminate uniform rate (ml/min).
2. The instrument air flow is constant (liters /min).
3. The total ml volume of the syringe used and the critical orifice site determine the total time of delivery (nun).
3 Therefore the total volume Of instrument air during the syringe delivery time can be calculated by the formula:
V=R v/r (1)
where: r • critical, orifice delivery rate (3. 33 ml/min)
It • instrument air flow (20 liters /min)
v total volume of delivery syringe (100 ml)
V s: total volume of instrument air flow during delivery of the total syringe volume (liters)
By substituting in the formula:
V=20x100/3.33
V=600.6 liters
5. The total quantity of liquid propellant to be vaporized in the syringe to produce x ppm of the propellant in the instrument flow is calculated as follows:
ppm x MW
ml =V k450 24450 (2)
Sp g r x 1000
(at 25°C and 760 mm Hg)
6. In calculating the amount of N204 2NO2 to be added to the syringe to produce a calibrating mixture, he volume expansion Which occurs with air dilution must be taken into account. The relative amounts of NO2 and N2O4 which exist at various pressures have been calculated from equili5riuni data and are summarized in Table I. At 1. 0 atmosphere pressure the undiluted equilbriurn mixture is 81.2% NO2 while after dilution with air the equilibrium at 1000 ppm Ii 90.3% NO2.
17
At 25~ C and 1 atmosphere the expansion factor is 1.7 and the amount of tank gas which must be used to obtain 50 ppm under the flow conditions described under formula I above would simply be:
50: 1,000,000 = X : 600,600
1.7
X = 51ml.
This 51 ml of tank gas is put in the syringe and diluted tà the 100 ml mark with dry air. This procedure Is repeated 3 times. to assure equilibrium conditions.
RESULTS AND CONCLUSION
The average percentage of error in generating UDME and NO2 Were + 4.8% and + 5. 6% respectively. Results of testing are detailed in table I and individual detectors are discussed Tables 3-12.
Until further field data at missile sites become available, the requirements for sensing equipment must be arbitrarily established. Our interpretation of requirements for an “ideal sensor” for industrial hygiene field use Includes the following:
1. Response time, 90% of full response in less than 1 minute.
2. Recovery time, 90% of full recovery in less thai minute.
3. Sensitivity, ± 50% of Threshold Limit Value (or better).
4. Baseline Stability, + 25% of the Threshold Limit Value (in one hour).
5. Range 0—200 ppm (with range switch).
6; Cross-sensitivity, not responsive to ordinary atmospheric contaminants.
7. Weight — 20 lbs or less.
8. Polyethylene tubing and polyethylene parts wherever, possible for NO2 and UDMH.
B. Application to other propellant systems is very desirable.
18
TABLE I
Relative Amounts of NO~ and N2p4 at Various Pressures
**Oxides of Nitrogen
* Pressure
(Atmospheres) , % N02
.00025 99.7 .3
.001 99.3 .7
.01 93.8 6.2
.1 67.7 32.3
.2 55.9 44.1
.5 40.9 59.1
1.0 31.2 68.8
1.5 26.3 73.7
2.0 23.3 76.7
* Total Pressure NO2 + N304
** The values listed are the relative amounts of NO2 and N2O4 Above results are calculated from equilibrium data:
S. Olasstone, Textbook Of Physical Chemistry~ ID. Van Nostrand Company, 1940.
19
20
TABLE III
MINE SAFETY APPLIANCE COMPANY’S BORANE ANALYZER
FOR UDMH
Performance
Low Range $ ‘ High Range
Baseline stability acceptable ‘ acceptable
Rattle 0-1. 5 ppm 0-16 ppm
Sensitivity 0.2 ppm 1.5 ppm
Type of Response non-linear non-linear
Response time (90%) 10-16 mInutes 20-30 minutes
Recovery time (90%) 10—15 minutes 20-30 minutes
Interference tests not performed not performed
Comments
1) The response and recovery times of this instrument are such that its practical use for monitoring UDMH is not recommended.
21
MISSING GRAPH
Pic page 27 insert
22
TABLE IV
MINE SAFETY APPLIANCES COMPANY’S BILLIONAIRE
for (UDMH)
Performance
Baseline stability . acceptable
Range 0-8 ppm
Sensitivity 0.3 ppm
Type of response logarithmic
Response time (90%) 7. 5—10.5 sec. $
Recovery time (9*) 8-19 sec.
Interference tests: Materials Response
Aerozine positive
Hydrazine positive’
NO,2 negative
Pentaborane negative
Ammonia positive*
Pump oil negative
.TP-4 (kerosene) negative
Cigarette smoke positive**
water vapor negative
* response to hydrazine and ammonia was estimated by the MSA representatives to be ten times that of UDMH.
** Cigarette smoke gave a positive response when a lighted Cigarette was held close to the sample inlet of the instrument due to particles of appropriate site issuing forth directly from the cigarette.
Comments
1) The estimated price of this unit (without pump or recorder) was $2500. The detection system and associated electronics can be packaged in Roffman boxes (same as those used by Western Dynamics and American Systems) and would weigh between 100 and 200 pounds. Power consumption with thermostatically controlled heating elements would be. approximately 600-1000 wafts. Output for the digital read-out would be 5-10 millivolts per ppm of contaminant.
23
TABLE V
AMERICAN SYSTEMS, INC’S AUSTON TOXIC FUEL SENSOR, MODEL No. 4070
Performance
I.ow Range’ High Range’
Baseline stability acceptable acceptable
Range 0.50 ppm $ 0-200 ppm,
Sensitivity 1 ppm 5 ppm. $ $
Type of response linear linear
Response time (90%) . $ 10—30 sec 10—30 sec
Recovery time (90%) . 10—20 sec. 10—30 sec
Interference tests:
Material T Response
NO2 negative **
B5H9 positive
NH3 negative
pump oil negative
JP-4 (kerosene) negative
cigarette smoke positive”
*These ranges were chosen arbitrarily so as to determine most accurately the performance characteristics of the sensors and to provide compatible useful range characteristic. for th. recorder. These are not fixed ranges incorporated by the manufacturer. $ $ $ . $ . $$. $ $ . $
**N02 gave a negative response (one recorder division below sero on the I..OW range) at a concentration of 2000 ppm.. $ $
**‘Cigarette smoke gave a positive response (about. 5 ppm on the LOW range) when blown directly into the Meter
Comments
1) This sensor is a production model of the prototype previously submitted by Mioro-path Inc. It’s useful range has been increased to 300 ppm.
2) The instrument was submitted without any recorder or readout or panel, since this particular unit was intended for central read-out application. For this technical evaluation the signal from the sensor was fed through an appropriate resistance dividing network to a lO millivolt Brown recorder.
3) Problems previously associated with the pump used with this detector appear to have been alleviated. The Instrument was operated continuously for a period of 2 weeks and no significant change was noted in the pump flow rate.
24
TABLE V (Cont’d) $ $
4) The estimated price of the Fuel sensor was $l2OO. This price is
exclusive of any recorder or read-out system; $ Dimensions are 20” x 16” x 7. 5’
weight is 60 pounds, end power consumption Is 150 watts at 115 volts and 60
cycles
$$
TABLE VI
BECKMAN INSTRUMENT COMPANY’S HYDROCARBON ANALYZER
for (UDMH)
Performance
Baseline stability acceptable
Range 0-20 ppm
Sensitivity 0.5 ppm
Type of response linear
Response time (90%) 0.1-0; 5 minutes
Recovery time (90%) 0.1-0.5 minutes
interference tests not. performed
Comments
1) The instrument is equipped with a very sensitive ammeter in series with several scaling resistors (scale reading lx, 3x, ,x1O, x30, x100, x300, x1000, and x3OO0). The meter has a linear sca1e marked from 0-100 and should be calibrated for the hydrocarbon of interest. A zero-adjust control allows one to zero this instrument to any constant background level of contamination, this zero, once set, is maintained through all positions of the attenuator control.
2) Higher concentrations (to 1000 ppm) than those used in the calibration were sampled by the instrument and the attenuator factors verified. In all cases the proper proportional reading on the meter was obtained, when the attenuator was repositioned.
3) This instrument is not specific for UDMH but responds to TJDMH because of the 2 methyl groups in the structure. It also responds to any vapor containing C-H bonds. Therefore, it is felt that if a significant number of interfering hydrocarbon substances (even at low concentration) are likely to be present, the validity of readings in terms of actual UDMH concentration would be questionable.
26
TABLE VII
GENERAL ELECTRIC COMPANY’S VAPOR DETECTOR for (UDMH)
Performance
Baseline stability acceptable
Range 0-5 ppm
Sensitivity 0.1 ppm
Type of resppse logarithmic
Response time (90%) 10-15 Sec.
Recovery time (90%) 10-15 sec.
Interference tests not performed
Comments
1) Concern over the limited range was partially alleviated upon receiving
a report from General Electric subsequent to our tests showing results of
double dilution technics In an effort to extend the range of the instrument.
Results indicate that the dilution technic is practical and that. a scale may
be chosen that best meets the application requirements.
2) Due to the very high volume air flow (approximately 30 liter/mm) contamination as far as 2.0-3 0 feet away from the sampling point was detected. by this instrument. This feature would reduce the number of point source instruments required at any particular installation.
3) The instrument has been subjected to rigorous shock and vibration tests and conforms to the Navy Military specifications.
27
TABLE VIII
MINE SMETY APPLIANCES COMPANY’S BILLIONAIRE
for (NO2).
Performance
Low range
Baseline stability acceptable . acceptable
Range , 0-10 ppm 0-80 ppm
Sensitivity . . 0. 5 ppm 6 ppm
Type of response logarithmic logarithmic
Response tine (90%) 8-fl sec. 7-il sect
Recovery time (90%) 8. 5—12.5 sec. 10—11.5 sec.
Interference tests:.
Material Response
Aerozine negative
UDMH negative
Hydrazine negative
Pentabrane negative
Ammonia negative
Pump oil negative
JP-4 (kerosene) negative
Cigarette smoke positive*
Water vapor negative
* Cigarette smoke gave a positive response when a lighted cigarette was held close &the sample Inlet of the instrument due to particles of appropriate size issuing forth directly from the cigarette.
1) The estimated price ‘of this unit (without pump or recorder) was $2500. The detection system and associated electronics can be packaged in Hoffman boxes (same as those used by Western Dynamics and American Systems) and would weigh between 100 and 200 pounds. Power consumption with thermostatically controlled heating element would be approximately 600-1000 watts. Output for digital readsout would be 5-10 millivolts per ppn contaminant.
28
TABLE IX
AMERICAN SYSTEM, INC’S AUSTIN OXIDIZER SENSOR, MODEL. 4075
for (NO2)
Performance
Baseline stability . acceptable
Range* 0-500 ppm
Sensitivity 5 ppm
Type of response linear
Response time (90%) 5 seconds
Recovery time (90%) 5 seconds
Interference tests:
Material Response
UDMH Neg
hydrazine Neg
pentaborane Neg
ammonia Neg
pump oil Neg
JP-4 (kerosene) Neg
cigarette smoke Neg
*TMs range was chosen arbitrarily so as to determine most accurately the performance characteristics of the sensor and to provide compatible rangb characteristics for the Recorder. This is not a fixed range incorporated by the manufacturer.
Comments
1) This sensor is a production model of the-prototype previously submitted by Micropath, Inc. The shortcomings of this instrument.that rendered it unsuitable for long-term, unattended operation, have been overcome. The response and recovery times .have been significantly decreased.
2) An attempt was made to find the maximum concentration that could be measured
accurately by this sensor without overloading. This occurred at approximately
2000 ppm.
3) This instrument was submitted without any kind of recorder or read-out panel since this particular unit was intended for central read-out application. For this. technical evaluation the signal from the instrument was fed through an appropriate resistance dividing network to a 10 millivolt brown recorder.
4) There was no lag time associated with initial low level concentrations; this appeared to be a problem with the prototype instruments previously evaluated. Problems associated with the pump used in this detector appear to have been alleviated. No noticeable corrosion of the metal fittings was experienced
29
TABLE IX (Cont’d)
during the period of evaluation.
5) The estimated price of the Oxidizer Sensor was $1000. This price is
exclusive of any recorder or read-out system. The dimensions are
20” x 16” 1 7-1/2” weight ii 60 pounds, and power consumption is 150 watts,
at 115 volts and 60 cycle.
30
TABLE X
BECKMAN INSTRUMENT’S FLOW ~COLOR!METER FOR NITROGEN DIOXIDE
Performance
Baseline stability acceptable
Range 0-100 ppm
Sensitivity 5 ppm
Type of Response linear.
Response time (90%) 35-50 sec
Recovery time (90%) 30-50 sec
Interference tests:
Material Response
Aerozine negative
pentabiorane negative
ammonia negative
pump oil negative
JP-4 (kerosene) negative
water vapor negative
cigarette smoke positive
Comments -
1) The instrument submitted for this evaluation was a “breadboard” or prototype model; the manufacturer stated that the unstable baseline would be eliminated in any production model purchased by the Air Force.
2) The principle of operation of this instrument is that of a colorimeter and the color of NO2 is measured at its peak absorbency. Only colored gases and. vapors having similar absorbencies’ would be expected to give interfering responses. Of course, smokes (such as that from a cigarette) and fogs will give interfering responses since they will physically block the light beam of the measuring cell and consequently will reduce the amount of light reaching the phototube.
3) The estimated price of each unit was $3200 (in quantities over 100, $3000). Weight was estimated at 100 pounds each. The dimensions of the final production model were estimated to be 41 x 2’ x 1. 5l~ No power consumption figures were given; however, it is estimated that a thermostatically controlled unit would consume approximately 500-1000 watts.
31
TABL.E’XI.
GENt~AL ELECTRIC COMPANY’S TOXIC VAPOR DETECTOR*
for (NO2)’
Performance
Baseline stability not reported
Range 0-100 ppm
Sensitivity 5 ppm
Type of response logarithmic
Respons. time (90%) not reported
Recovery time (90%) not reported
Interference tests not reported
*Thjs instrument was not evaluated by this Laboratory. The data presented was submitted by the manufacturer.
32
TABLE XII
MAST DEVELOPMENT COMPANY’S OZONE METER
for (NO2)
Performance
Baseline stability not acceptable
Range 0-25 ppm
Sensitivity 2 ppm
Type of response linear
Response time not recorded
Recovery time not recorded
Interference tests not performed
Comments
I) Except for the two lowest concentrations monitored, readings were off—scale With the range switch of the recorder on LOW. Therefore, all readings were taken with the range switch on HIGH.
2) Modifications should be made on the air sample flow control valve so that a more uniform air flow can be achieved. Without. this modification this instrument cannot be left unattended for any extended period of time.
33
TABLE XIII
E-24 R ALARM for (NO2)
Performance
Baseline stability not acceptable
Range ‘ 0-25 ppm
Sensitivity 2 ppm
Type of reapoñse non-linear
Response time (90%) 30 sec.
Recovery time (90%) 30 sec.
Interference tests not performed
Comments
1) The principle involved in the operation of this detector is that Of a conductivity cell. It is therefore considered to lack specificity, thereby rendering it unacceptable for use in detecting specific water soluble vapors.
34
REFERENCES
1. Pinkerton, M. C., Lauer, J. M., Diamond P., Thomas, A. A., A Colorimetric Determination for 1,1 Dimethylhydrazine in Air, Water, and Blood. ASD Technical Report 61-708, Wright,Patterson Air Force Base, Ohio, Dec. 61.
2. Saltzman, Bernard E., Colorimetric Microdetermination of Nitrogen Dioxide in the Atmosphere, Analytical Chemistry Vol. 26, 1949 (Dec. 54).