r y»5ClA SIFlED L M Redman INTER-OFFICE MEMORANDUM July 6 1945 TO K T Bainbridge FROM J 0 Hirschfelder and John Mage»-ii 2iiL- y- - SUBJECT IMPROBABILITY OF DANGER FROM ACTIVE MATERIAL FALLING FROM CLOUD v -vyCt -c In a previous memorandum to you dated June 16 1945 we showed that there was the possibility of a dangerous amount of active material sedimenting down onto nearby towns IF all of the active material were to condense onto cold sand particles with the same distribution of activity versus pArtiole size that H L Anderson found in the 100 ton Trinity shot We have tried to examine this hypothesis in some detail It i appears lilcely that a considerable fraotion of the active material will be oo-preoipitated with vaporized silica in the form of very small particles which remain suspended in the form of smoke This smoke should gradually diffuse and cause no health haeard unless it were washed down in a thunderstorm f Unfortunately we cannot make a quantitative estimate of what fraction of the active material will be oo-precipitated in the smoke and what fraction will plate onto the cold sand CM of the meohaniim involved In this memo we will try to present a picture Our present arguments would seem to indicate that CM the amount of active material sedimenting onto a nearby town may be less by Of a factor of from 2 to 10 than the amount estimated in our previous memorandum a As near as we can tell the sand which causes the danger comes up from the CM crater rather than from very large distances CM I MIXIHO OF ACTIVE MATERIAL WITH VAPORIZED SAND AND STEEL Originally the active material is located in the outer fringes CLASSIFICATION CANCELLED UMCUl niED 0 UHOLASSiFIFD of the ball of fire Ths ball of fire when fully expanded at the end of the one tenth of a second has a radius of around 500 feet Because of the inter- action with the ground aixi the reflected shook the ball of fire will be quite flat on the bottom and still almost round on top i e almost a hemisphere It will be sitting immediately on top of the crater The crater according to present estimates will be only 190 feet in radiu» see LA-292 or 60 feet according to Penney's estimates so that all of the sand rising from the crater will pass through the ball of fire The total amount of dirt con- tained in ths crater as expected by UacMillan and Wilcox is 50 000 tons or 5000 tons according to Penney It is reasonable to expect more than 250 tons of dirt will rise into the ball of fire The energy required to heat one gram of sand up to its boiling point 2500 C and vaporized it is approximately 2700 cals This figure would apply for pure silica Thus the energy of 2 7 tons of T N T would be required to vaporize one ton of sand If the ball of fire contained lOjt of the energy or 500 tons T N T equi valent it could vaporize something less than 250 tons of dirt The length of time to vaporize the sand is very short once it comes into the high p temperature region Assuming that the surface of the dirt is black and the CV radiation is black body it follows thati CM dr dt r CM body CM CM -2 T lO 000 om seo « d i u a of a sand partiele and T is the temperature of the black The steel tower will be dissolved at approximtely one fourth this rate because of its greater density In any case the vaporization processes are extremely rapid and should serve very effectively to cool the ball of fire down to around 5000°C iiNi i iD II CONDENSAHON OF ACTIVE MATERIAL The maximum rate at which the ball of fire could cool by o emission of radiation would require 0 6 of a second to reach 2000 C This is calculated for a black body radiating into a vacuum Actually the emissivity of the ball of fire is probably somewhat greater than 0 1 which is a reasonable value for a non-luminous gas such as CO- Therefore a reasonable upper limit on the cooling time would be 6 seconds At the 2000° temperature all of the solid material will presumably be condensed The dirt which rises when the blast wave first hits the ground is given a large horizontal velocity most of it making an angle of around 15 This dirt has been pulverized by the blast and the upper portions are rapidly vaporized There is no rapid method of transferring the silica vapor to the central and upper portions of the ball of fire This process will continue until approximately 100 tons of sand is vaporized and the ball of fire is chilled below the condensation point is then rapidly precipitated in the form of a fine smoke This silica The upper and central portions of the ball of fire contain so little solid material that the rate of condensation is very slow Convection currents are set up at the bottom of the ball which mix the smoke and the sand and these turbulent eddies e t into the ball of fire The active material will ad- here inqjartially toiiiatever solid it happens to hit and sand scours the 49 and fission products In this way the smoke The relative amount of these substances whioh adhere to the smoke and to the sand depends only on the relative surfaces of the two components mate this ratio quantitatively It is impossible for us to esti- If there were the same weight of silica in the form of smoke with mean diameter of 0 5 micron and of sand with mean diameter of 20 microns the smoke would have 40 times the surface and therefore pick up 40 times the active material Actually there will be much more weight of cold sand and smoke in this mixture possibly 4 to 40 times as much Thus there should be 10% to 50 i of the activity plated out onto the cold sand The active material sedimenting onto a nearby town should therefore be less by a factor of from 2 to 10 than we anticipated in our previous memorandum The high rates of chilling make the chemical nature of the various components completely unimportant so that there will be co-preoipitation of the active rrBterial with the sand in the smoke Furthennore 49 is known to adhere to sand and scrubbing with sand is used to remove 49 from surfaces At first a fraction of the fission products ' 20% are in the form of noble gases which transmute to alkali metals within a minute and these of course are easily absorbed on the smoke particles The large amounts of ionization in the smoke cloud will tend to prevent agglomeration and thus help to dispose the active material over a larger area This ionization is due both to the radio activity and to the rapid chilling fran the high temperatures where ions are stable We wish to thank Robert Kamra for his experimental assistance He showed that Trinity sand vaporized in either a carbon arc or a tantalum crucible condenses into the form of a very fine bluish-white smoke He also detentined the distribution of particle site in normal Trinity sand whioh we used in our previous memorandum Because of the difficult nature of this problem we felt it advisable to discuss a number of the technical points with experts Drs F G Cottrell and Bernard Welch Western Precipitation Co gave their opinion that the 100 micron particles about irtiioh we are concerned should sediment according to the normal Stokes' Law as we had assumed in our previous I memorandum Cement dust of the same sijje falls in noticeable quantities at distances up to 50 miles Drs Sage and Lacy of C I T studied the problem of condensation of the silica and felt that practically all of it will come out in the form of smoke We should also acknowledge many helpful discussions with Drs G I Taylor C S Smith W G Penney and V Weisskopf J 0 Hirschfelder John Magee jah Oppenheimer Aebersold Eethe Carlson Hempelman Capt Jones Robert Kamm Capt Nolan Parsons Penney Reines Segre C S Smith Sugarman CM CM CM CM Turkevioh Col Warren Weisskopf