NERC-LV-539-1?.
             COMPARISON OF FREEZE-OUT AND
       ADSORPTION TECHNIQUES FOR COLLECTION OF
              ATMOSPHERIC TRITIUM AS KTO
£
Charles W, Forts Vernon. E. Andrews,  and Aaron  Goldmar
              Environmental Surveillance
        National Environmental Research Center
         U.S. KNfliaOJMSSTAL PROTJtCfjtW AGENCY
                  Las ¥egas, Nevada

               Published November  1972
      This study perforated under  a Memorandum of
            Understanding No0 AT(2.6- .1 )-539
                       for the
            Uc Sc ATOMIC ENERGY COMMISSION

-------
This report was prepared as an account of work sponsored by
the United States Government.   Neither the United  States nor
the United States Atomic Energy Commission,  nor any  of  their
employees, nor any of their contractors,  subcontractors, or
their employees, makes any warranty,  express or implied, or
assumes any legal liability or responsibility for  the accuracy,
completeness or usefulness of  any information,  apparatus,
product or process disclosed,  or represents  that its use would
not infringe privately-owned rights.
 Available  from  the National Technical Information Service,
               U. S. Department of Commerce,
                   Springfield, VA. 22151

          Price: paper copy $3.00; microfiche  $.95
054

-------
                                                   NERC-LV-539-12
             COMPARISON OF FREEZE-OUT AND
       ADSORPTION TECHNIQUES FOR COLLECTION OF
              ATMOSPHERIC TRITIUM AS HTO
                          by
Charles W. Fort, Vernon E. Andrews,  and Aaron Goldman
              Environmental Surveillance
        National Environmental Research Center

         U.S. ENVIRONMENTAL PROTECTION AGENCY
                  Las Vegas, Nevada
               Published November 1972
      This study performed under a Memorandum of
            Understanding No. AT(26-l)-539
                       for the
            U. S. ATOMIC ENERGY COMMISSION

-------
                              ABSTRACT
     A field study was conducted to compare the atmospheric concentrations




of tritium as HTO determined from a passive freeze-out moisture collector




and an air sampler using an adsorbent column of molecular sieve.  Effluent




from production test flaring of the Project Rulison experimental gas well




produced by a nuclear detonation was used as an environmental source of




atmospheric tritium.  Also studied was the comparability of absolute




humidity determinations based on two types of psychrometric measurements




and water recovered from a known volume of air by the molecular sieve.




From statistical analysis it was concluded that there was no significant




difference between tritium measurements or humidity determinations as long




as reasonable care was exercised in operating the samplers or making




psychrometric measurements.

-------
                          TABLE OF CONTENTS
                                                                  Page
 ABSTRACT                                                           i
 LIST OF FIGURES                                                   i±j-
 LIST OF TABLES                                                     iv
 INTRODUCTION                                                       1
 OBJECTIVES                                                          2
 EQUIPMENT AND PROCEDURES                                            2
   A.   Equipment                                                    2
   B.   Sampler Selection  Criteria                                  3
   C.   Experimental Design                                          4
        1.  Site Description                                         4
        2.  Sampler Calibrations                                     6
        3.  Sampler Preparation                                      6
        4.  Sample Collection                                        7
        5.  Sample Analysis                                          8
   D.   Calculations                                                10
 RESULTS                                                            11
   A.  Moisture Collection                                         11
   B.   Tritium Results                                             15
 DISCUSSION                                                         17
 CONCLUSIONS                                                        20
APPENDICES                                                         26

-------
                           LIST OF FIGURES



Figure                                                             Page

1.  Special Study Sampling Locations                                 5

2.  LLL Freeze-Out Sampler and NERC-LV Molecular Sieve Sampler       9

3.  LLL Freeze-Out Sampler Collection Rate vs. Absolute             13
    Humidity

4.  Absolute Humidity Calculated from Psychrometric Data            14
    vs. Absolute Humidity Measured by Molecular Sieve Sampler
                                    iii

-------
                           LIST OF TABLES
Table                                                                Page


1.  LLL Freeze-Out Samples:  Comparison of Rate of                    21
    Collection with Absolute Humidity

2.  Absolute Humidity Calculated from Psychrometric Data and          22
    Measured by NERC-LV Molecular Sieve Sampler

3.  Tritium Concentrations in Atmospheric Moisture as Collected       23
    by NERC-LV and LLL Samplers

4.  Tritium Concentrations in Air as Determined from NERC-LV          2 4
    and LLL Samplers

5.  Analysis of Variance - Calculated and Measured Absolute           25
    Humidity

6.  Analysis of Variance - 3H Concentration from LLL and NERC-LV      25
    Samplers
                                 IV

-------
            AN EVALUATION OF FREEZE-OUT AND ADSORPTION
      TECHNIQUES FOR COLLECTION OF ATMOSPHERIC TRITIUM AS HTO
                           INTRODUCTION


     Project Rulison was a joint Industry/Government-Sponsored nuclear
experiment conducted as part of the Atomic Energy Commission Plowshare
Program to develop peaceful applications for nuclear explosions.   The
underground nuclear detonation was conducted on September 10, 1969 to
investigate the economic and technical feasibility of using an underground
nuclear explosion to stimulate production of natural gas from the low
productivity gas-bearing Mesa Verde formation in the Rulison Field near
Grand Junction, Colorado.  Phase III of the experiment involved controlled
drillback into the cavity produced by the nuclear detonation followed by
flow testing to determine the cavity volume and the rate of natural gas
flow.  Natural gas released during the various periods of flow testing
was burned, or flared, at the top of a 30-meter stack near surface ground
zero (sgz).

     The National Environmental Research Center-Las Vegas(NERC-LV) provided off-
site stationary, mobile, and aerial radiological surveillance during these
flaring periods.  An initial calibration flaring was conducted during which
NERC-LV and the Lawrence Livermore Laboratory2(LLL) collected samples of
atmospheric moisture in the offsite area for measurement of tritium ( H)
concentrations.
 At the time this work was performed, the Center was named the Western Envi-
 ronmental Research Laboratory.
2Formerly Lawrence Radiation Laboratory.

-------
     Techniques of moisture collection used by the NERC-LV and LLL were quite




different because of a basic difference in ultimate use of the resulting




data.  Since no information was available from the surveillance effort to




show comparability of data, the NERC-LV initiated a series of field comparison




studies at the Rulison site.









     This report describes this study and the results of that effort.









                             OBJECTIVES









     This study was performed with three objectives in mind:  (1) to deter-




mine the comparability of data collected by the two systems;   (2) to evaluate




the performance of the LLL sampler as an integrating sampler; and  (3) to




arrive at some conclusion pertaining to the applicability and use of each




system in a tritium surveillance program.









                    EQUIPMENT AND PROCEDURES









     A.  Equipment




         The LLL moisture sampling equipment consisted of a 9 1/2-inch pie




         tin holding a No. 10 can (equivalent to a three-pound coffee can)




         filled with crushed dry ice and placed in the open air.  The outer




         surface of the can was sandblasted to improve moisture recovery.









         The NERC-LV sampling equipment consisted of a battery-operated pump




         drawing air through a cannister containing 300 gm of Linde type




         13x molecular sieve and discharging through a dry gas meter for




         volume determination.   Flow rates were approximately 2 to 3 CFM.

-------
B.  Sampler Selection Criteria




    The different sampling techniques used by the two organizations




    resulted from different surveillance requirements and established




    procedures.  The NERC-LV  had the responsibility for measuring environ-




    mental concentrations of radioactivity resulting from the Project




    Rulison flaring activities.  Seven stationary sampling stations had been




    established at populated locations around the Rulison site.   These




    stations incorporated a dehumidifier providing water for rapid field




    analysis of 3H, a high volume air sampler for particulate radio-




    activity, and a low volume molecular sieve sampler collecting 48-hour




    integrated samples of atmospheric moisture for 3H analysis and




    CO- for 1I+C analysis.  Molecular sieve was chosen for its ability




    to co-adsorb H20 and CCL, its utility in collecting long-term




    integrated samples, and ease of handling.  The molecular sieve




    samples were returned to the NERC-LV for more precise analysis than




    could be performed on the dehumidifier samples at the field office




    in Grand Junction.









    The molecular sieve samplers collected water from a known volume of




    air, making it possible to determine the airborne 3H concentrations




    from sample volume, moisture recovered, and analytical data.  Because




    it was desirable to use a portable unit providing data comparable to




    that available from the fixed stations, a battery-powered sampler was




    designed using the same molecular sieve.  An additional consideration




    was that this type of sampler also provided the capability of collecting




    a particulate filter sample.  Because the laboratory equipment and




    techniques for analyzing molecular sieve samples already existed at




    the NERC-LV,  laboratory  support  required no  special  preparation.






    Lawrence Livermore Laboratory had a requirement to collect numerous samples




    for the purpose of determining plume trajectories and area covered as a guide

-------
    to conducting special environmental studies.  The most sensitive




    indicator of an increase of 3H above background is to measure the




    concentration in atmospheric moisture rather than the airborne con-




    centration since the airborne concentration is also a function of




    absolute humidity.  For this purpose an efficient collector of




    atmospheric moisture is perfectly satisfactory.  One of the simplest




    and most economical collectors to operate for that purpose is




    the system employed by LLL.







C.  Experimental Design




    1.  Site Description




        Project Rulison was conducted near the upper end of a deep




        valley extending to the southeast from Morrisania Mesa.  The




        valley bottom drops from the 8,154 feet mean sea level (MSL) eleva-




        tion of sgz to 6,400 feet MSL at the valley mouth, three miles




        northwest of sgz (see Figure 1).  Previous surveillance had




        shown that nighttime drainage winds carried much of the flaring




        stack discharge downhill along the valley to the northwest.







        For these studies, four sampling locations were chosen and




        marked along the road leading from the Rulison test areas down




        the valley to Morrisania Mesa.  The station nearest the test




        well was located directly across this road from the flare stack,




        while the farthest station was located at the old control point pad,




        2.5 miles northwest of the flaring stack at 6,800 feet MSL.




        The other two stations were placed at approximately even intervals




        between the flare stack and the  control point pad.  All subsequent




        sampling of the Rulison flare was conducted at these locations




        with two LLL and two NERC-LV samplers used at each location for




        each sampling period.

-------
Figure 1.  Special Study Sampling Locations

-------
 2.    Sampler Calibrations




      Prior  to designing  the molecular sieve sampler, the NERC-LV




      staff  performed a series of tests on molecular sieve to relate




      collection efficiency and adsorber capacity for atmospheric




      moisture to absolute humidity, temperature, and bed residence




      time.  The sampler  volume and flow rates were selected on the




      basis  of those tests to give nearly 100 percent collection




      efficiency and sufficient collection capacity under the conditions




      to be  encountered.  The dry gas meters used for sample volume




      measurement were calibrated against the NERC-LV 500-cubic-foot




      spirometer and a calibration factor was assigned to each one.






      An experiment was conducted in Las Vegas to provide information




      on the collection rate of the LLL sampler over short intervals.




      Eight  samplers were used, two of which were collected every half




      hour and the volume of water collected was measured.  The moisture




      collection rate was determined to remain constant for the period




      of the test under the prevailing conditions of temperature and




      humidity.   Data from this test are included in Table 1.
3.   Sampler Preparation




     Molecular sieve used in the NERC-LV samplers was degassed at the




     Center in Las Vegas and packed in 300-gm lots in sealed bottles.




     Prior to sampling, the sieve was transferred from the bottles




     to sieve holders at the field office in Grand Junction.  The




     filled holder was then sealed in a heavy duty plastic bag which




     was then sealed inside a second heavy duty plastic bag.  In the




     field, the bags were opened, the sieve holder was attached to




     the sampler inlet, and the sampler motor was switched on.

-------
    Control sieves consisting of molecular sieve prepared and




    packaged as described above were handled identically to the




    test sieve, including being taken to the field,  except that




    they were not used for sampling.  This provided  information




    pertaining to extraneous moisture collection prior to and




    after the sampling period.  Suitable portions of sieve from




    each sieve batch were returned to NERC-LV, without  the above




    handling, for water extraction to ascertain water content prior




    to both exposure and handling.









    Prior to each sampling period the LLL sampling apparatus was




    washed to reduce any cross-contamination potential from one




    sampling period to the next.  Approximately 8 pounds of dry




    ice was crushed and poured into each can at the  field sampling




    station.  Each container was then placed in a pie pan and set




    on the ground.









4.  Sample Collection




    Design of the NERC-LV sampler, using quantitative moisture sampling




    from a known volume of air, eliminated the need  for making humidity




    measurements.  Use of the LLL sampler required determination of the




    absolute humidity to calculate the airborne concentration of 3H.




    Relative humidity measurements by NERC-LV for this project were generally




    made using a Princo sling psychrometer.  Because of the low tempera-




    tures in the Project Rulison area during this study, the sling psy-




    chrometer could not always be used.  Absolute humidity was then




    calculated from measurements of relative humidity, made with a Bacharach

-------
    Instrument Co. Model 22-4503 hygrometer, and dry bulb temperature.





    Sampling periods were of two-hours duration.  At the end of the


    sampling period, the frost on the outside of the cold LLL sampler


    was carefully scraped into the pie pan.  After all loose frost


    was removed, the dry ice was removed from the container.  A pro-


    pane torch flame was then used to slightly heat the inside of


    the container to melt the remaining ice allowing it to be collected


    in the pie pan.  The frost already collected in the pan was melted


    and the water carefully poured into a small polyethylene con-


    tainer labelled with time, date, and location of collection.





    Collection of the molecular sieve consisted of removing the sieve


    holder from the air mover, sealing it inside a plastic bag, and


    subsequently sealing it in a second bag.  At the field office


    in Grand Junction, the sieve was transferred from the holder


    into a one-quart plastic Cubitainer for shipment to the NERC-LV for

                         o
    water extraction and  H analysis.  Figure 2 shows the NERC-LV


    and LLL equipment set up for collecting atmospheric moisture.





5.  Sample Analysis


    Water on the molecular sieve was recovered at the NERC-LV.   The


    extraction technique involves heating the sieve to 350°C and


    passing dry helium through the heated sieve column to sweep


    water vapor into a condensation trap.  A vacuum is pulled on


    the enclosure to aid in the removal of water.

-------
Figure 2.  LLL Freeze-Out Sampler and NERC-LV Molecular Sieve
           Sampler.  Open sampler cartridge on top of NERC-LV
           sampler box shows molecular sieve used.

-------
        Water from both the LLL and NERC-LV samplers  was  distilled and



        a 5-ml aliquot added to a dioxane base liquid scintillation



        cocktail.   The samples were counted at ambient temperature



        for 100 minutes.
D.   Calculations


                                                        3
    Results of the water analysis were expressed as  pCi  H  per ml water



    for both type samples.   Because the total volume of  air pulled



    through the sieve is known and all water on the  sieve is  retrieved, a



    "measured" absolute  humidity  is readily  obtained as ml  water per cubic



    meter of air.




    Computer processing  of  data from the NERC-LV sample gave  total water



    collected, total volume of air sampled,  ml water per cubic meter


             3                        3
    air,  pCi  H per ml water,  and pCi   H per cubic meter air.
                                                     3

    Output  from the  LLL sample  results  gave  only pCi  H/ml water.  To


                                                       3   3
    convert to  airborne concentrations  of  tritium  (pCi  H/m  air)



    psychrometric  information was  obtained providing  "calculated"


                                              3

    absolute humidity  information  in ml water/m  air.
   To calculate absolute humidities, wet and dry bulb  temperature



   measurements were made.  If wet bulb temperatures were less than



   20 F, as was the case during some of these runs, direct reading



   hygrometer and dry bulb temperature measurements were made to



   provide the appropriate psychrometric information.  When wet and



   dry bulb data were available, they were used directly in calculating




   the humidity using a standard psychrometric chart and psychrometric



   formulas.  When the hygrometer and dry bulb were used, the equations



                            10

-------
               of Appendix A (derived at the NERC-LV from the Carrier
               Corporation's psychrometric chart and formulas) were em-
               ployed to calculate the absolute humidity.  Wet-bulb read-
               ings for use in this equation were extracted from the U.S.
               Weather Bureau Bulletin No. 1071 (Relative Humidity from
               Wet and Dry Bulb Thermometer).1

               Atmospheric 3H concentrations as pCi 3H/m3 air were determined
               by multiplying the 3H concentration in water from both LLL and
               NERC-LV samplers by the appropriate humidity values; i.e.
               "measured" humidity for NERC-LV results and "calculated" humidity
               for LLL results.

                                RESULTS

          A.    Moisture Collection
               The first comparison made was the approximate collection
               rate of the LLL sampler to absolute humidity and to de-
               termine whether or not the collection rate varies during
               the collection period.  Table 1 summarizes the collection
               rates and humidities observed during the field experiment
               at Project Rulison and the collection rate experiment at
               the NERC-LV.  As shown by the May 10 results at the
               NERC-LV, the collection rate remained constant over a two-
               hour period with an absolute humidity of 8.4 ml H»0/m3 air.
               Since that humidity was higher than any encountered
               during surveillance of Project Rulison flaring or
Handbook of Chemistry and Physics, 44th Edition, Chemical Rubber Company.
                                   11

-------
 during this field experiment,  it is reasonable to believe that




 a uniform collection rate would be the case under normal  circum-




 stances.   This should prevent  a short-term condition of either  a




 high or low activity exposure  from being over-weighted in deter-




 mining the average concentration,  as could be  the case with  a




 varying collection rate.









 Figure 3,  a plot of collection rate versus absolute  humidity, shows




 that over  the  range of humidities  encountered,  the relationship is




 linear.  Ambient temperature apparently had little or no  effect




 on the collection rate.   This  is most likely due  to  the large




 difference between ambient temperatures and the temperature  of  dry




 ice.   With knowledge of the relation between collection rate and




 absolute humidity,  field  humidity  measurements  can be used to deter-




 mine the length  of  sampling period required to  collect sufficient




 water for  analysis.









 The NERC-LV sampler had been designed  to be nearly 100 percent efficient



 as a water vapor  sampler.   Even so,  it was desirable to compare




 absolute humidites  "calculated" from field measurements of relative




 humidity and temperature  to absolute humidities "measured" from the




 air volume sampled  and moisture recovered  from  the NERC-LV sampler.




 The  comparison is  tabulated in Table 2 and is shown  graphically in




 Figure 4.  A statistical  analysis  of all data showed no difference




 between the two methods of  measuring absolute humidity  at the




99 percent significance level.   Investigation of  the  data  in




the "absolute humidity" columns of Tables  1 and 2





                          12

-------
   Figure 3.
    LLL FREEZE - OUT SAMPLER
       COLLECTION RATE
             VS.
       ABSOLUTE HUMIDITY
                                                           El
                                  ©
      © CALCULATED FROM
        PSYCHROMETRIC DATA
      H MEASURED WITH
        MOLECULAR SIEVE SAMPLER
                          EJ
                          ©
   m
   ©
   Q
   ©
                                                  I
0
0.05
0.1
0.15
0.2
0.25
0.3
0.3E
            COLLECTION RATE , ml H2O/ MIN
                         13

-------
    8
oo
  CM
 X
 e
 i
 8  4
 O
 <
 O
                   Figure 4.
                   ABSOLUTE HUMIDITY CALCULATED FROM PSYCHROMETRIC DATA
                                            VS.
                  ABSOLUTE HUMIDITY MEASURED BY MOLECULAR SIEVE SAMPLER
                 (NUMBERS BESIDE DATA POINTS INDICATE MULTIPLE IDENTICAL RESULTS)
    0
     0
                          MEASURED HUMIDITY, ml H O/m3AIR
                                        14

-------
    shows that the "calculated" absolute humidity is usually higher




    than the "measured" absolute humidity.  A biasing of the




    results for "calculated" absolute humidity on the high side




    is to be expected, since the commonest error in making psychro-




    metric measurements is to obtain a high wet bulb temperature.




    This would result in overestimating the humidity.









    The "calculated" and "measured" absolute humidity data were subjected




    to a statistical analysis of variance (See Table 5) assuming a two-way




    non-additive model.  There appear to be no differences among locations,




    no differences between calculated and measured humidity, and no inter-




    actions present at the .90 significance level.









B.  Tritium Results




    Tritium results for atmospheric moisture are expressed both in




    terms of concentration in the water and in air (pCi 3H/ml H-O




    and pCi 3H/m3 air).  The former is most useful in detecting small




    additions of HTO to the atmosphere while the latter is required to




    make dose calculations or to compare airborne concentrations to




    radioactivity concentration guides.  Table 3 lists the results of




    3H analysis on the NERC-LV and LLL samplers as pCi 3H/ml H20.  Tritium




    concentrations in air are given in Table 4.  Concentrations for




    the LLL sampler are the product of absolute humidities in Table 2




    and concentrations in water given in Table 3.  Concentrations for




    the NERC-LV sampler are derived from the total 3H activity collected




    from the measured volume of air sampled.
                              15

-------
        An analysis of variance was performed on the reported concentrations

        of 3H in air as determined from the two sampler types (Table 6).   A two-

        way classification non-additive model was assumed.   The F ratio was less

        than one in all cases, indicating no interactions,  no differences between

        samplers, and no differences among locations at the .90 significance level,

        Some data were excluded from the analysis. Reasons  for this omission

        will be given in the Discussion Section.



        As can be seen in Tables 3 and 4, the average LLL sampler results at

        location 1 for the February 10 and 12 sampling periods are almost

        three times those of the average NERC-LV sampler results.  Two expla-

        nations were advanced regarding the anomaly.  During both collection

        periods a rather brisk drainage wind was flowing, causing some

        drifting snow to be collected in the LLL sampler.  Snow samples

        were not collected in conjunction with these atmospheric moisture

        samples.  However, snow samples collected around the stack during

        this same period as a part of the NERC-LV surveillance program were

        found to have elevated levels of 3H.  These are documented in the

        NERC-LV Rulison surveillance report.*



        A second possible explanation was that the LLL sampler was collecting

        3H "rainout" in the immediate vicinity of the stack, whereas the

        NERC-LV  system  was  positioned with the molecular  sieve holder  inlet  pointed

        downward so that falling particles  or droplets would not fall on,

        and be  collected by,  the molecular sieve.  "Rainout" occurred in  the stack

        vicinity as a result  of water injection into the flare to dispose of

        water separated at the well-head.
* Off-site  Radiological  Safety  Program for  Project Rulison Flaring,  Phase III;
   National Environmental Research Center-Las Vegas. NERC-LV-539-15.
                                16

-------
          An attempt to verify the "rainout" theory was made on March 6 and 7




          (results in Tables 3 and 4 as location 1A).   Four NERC-LV




          molecular sieve and four LLL collectors were used on these two days




          at location 1.  On both days two molecular sieve holders were turned




          up so that falling water droplets or ice particles would be easily




          collected, and two sieve holders were turned down as usual.  The




          corresponding LLL set-up consisted of two collectors placed on the




          ground (packed snow) and two collectors placed off the ground on top




          of the wooden case of the NERC-LV sampler. The data show excellent




          agreement between the NERC-LV and LLL collectors and appear no different




          from the results of the standard set-up.  However, since they were




          collected in addition to the regular samples and were treated




          differently, it was felt that they, also, should be disqualified




          from the subsequent statistical analysis.  It should be noted from the




          data that the original discrepancy did not appear during subsequent




          sampling periods.  It was concluded that snow contamination had




          accounted for the difference in data.









                            DISCUSSION









     The following discussion of operational characteristics is primarily




subjective, based on the operating experience gained during this test.  The




basic conclusion was that although the LLL sampler is somewhat simpler to




operate, overall, the NERC-LV sampler is easier to operate in the field.









     Varying degrees and types of attendance to the LLL sampler are  required,




depending on weather conditions, in order to assure uniform sampling rate and




to effect satisfactory recovery of the collected moisture.  Contamination of




an open sampler such as this from precipitation or drifting snow can be



                                  17

-------
 serious,  but proper design can  eliminate this.    Problems such  as




 frost  blowing away while scraping the  can in  a  high wind  were found  to make




 the operation difficult and,  at times,  frustrating.   Daytime operation in




 warm sunlight can cause the dry ice  to  sublime  away from  the container wall,




 reducing  the collection rate  and allowing melting and possible  evaporation




 of frost.   Use of a shade would probably correct  this.  The necessity of




 using  a torch to  melt  the frost when ambient  temperatures are low  is another




 field  operation that may be required at times.  The LLL sampler used required




 a  minimum of two  hours to collect the required  sample under the lowest humidity




 conditions.   If a shorter sampling period is  desired,  use of additional samplers




 or redesign of the current model would  be required.









     One  distinct advantage of  the LLL  sampler  is the direct recovery of water




 which  can  be used as is  or distilled for liquid scintillation counting.  This




 reduces analytical complexity and cost  and shortens the time required to obtain




 a  measure  of the  airborne concentrations.









     Field  operation of  the NERC-LV molecular sieve sampler is very simple,




 requiring only that  the  sieve holder be  connected to  the  sampler inlet.




 Psychrometric  measurements  are made  and  a sampling time is selected to obtain




 sufficient  sample without exceeding  the  capacity  of the molecular  sieve.  No




 variation in operating procedures  is necessary  because of weather  conditions




 other  than selecting the  proper  sampling time.  Because of the  large difference




 between the  minimum  required amount  of water  and  the  bed  capacity, this time




 is quite flexible.









     Although  the NERC-LV sampler was operated at each location during the same



 two-hour period as the LLL unit, it would not normally need to  operate this




long to collect the minimum 5 ml of water required for 3H analysis, even




                                 18

-------
under the very low humidity conditions encountered.  These samplers were




operated at about one-half their normal flow rate and collected 10-15 ml




under the lowest humidity conditions of about 1 ml/m3.  The LLL unit collected




about 5 ml total volume for the lowest humidity.









     Appendix B is a set of calculations used by NERC-LV field teams in estimating



absolute humidity for the purpose of adjusting sampling periods.  The




calculations are simple and are more than adequate for field use. These same




calculations could be used to advantage with the graph of Figure 3, to estimate




sampling times required for the LLL unit to collect desired water volumes.




The figures of Appendix B are graphical representations of psychrometric




equations.  A psychrometric chart must be consulted in addition to the figures




to arrive at the absolute humidity.









     Initial expense of the NERC-LV sampler would normally be greater than that




of the LLL sampler because of the need for an air mover, dry gas meter,




batteries, and sample holder.  Since NERC-LV already had the necessary air samplers




and laboratory facilities to handle the molecular sieve samples, very little




additional cost was actually required to conduct this type of surveillance.









     Two significant facts were established regarding moisture collection.




First, over a reasonable sampling period under the conditions of May 10




(Table 1) the sampling rate of the LLL sampler was uniform.  Second, with




reasonable care in obtaining field measurements absolute humidities calculated




from two types of psychrometric data agreed with those measured by the NERC-LV




molecular sieve sampler.  These facts assure comparability of data between




the samplers when they are  operated properly.






                                  19

-------
                             CONCLUSIONS








     Analyses of variance performed on data from this study on the LLL




collector and NERC-LV molecular sieve sampler showed that at the .90 sig-



nificance level there is no difference between atmospheric tritium con-




centrations measured by two sampling techniques nor between the calculated




and measured absolute humidities.









     It was concluded that with proper precautions, both samplers




provided reliable measurements of the airborne concentration of % in the form




of water vapor.  One sampler or the other might be chosen for a particular




application, or both might be used, with assurance that the results are comparable.




On the basis of initial and operating costs, some form of the LLL freeze-out




sampler would normally be the portable sampler of choice for collection of atmos-




pheric moisture.  When dry ice is not available or operators must remain in the




field too long to maintain a reserve supply of dry ice, the NERC-LV sampler is




especially useful.
                                 20

-------
                                       TABLE 1
                               LLL  Freeze-Out Sampler:
               Comparison  of  Rate of  Collection with Absolute Humidity
Location
1
1
1
2
2
4
4
WERL
WERL



Date
Collected
3/4
3/6
3/7
3/6
3/7
3/6
3/7
5/7
5/10



Sampling
Period
(Hours)
2
2
2
2
2
2
2
2
0.5
1
1.5
2
Collection Rate
(ml H_0/min) ' •"•)
0.18
0.04
0.08
0.06
0.06
0.08
0.08
0.22
0.30
0.31
0.31
0.35
f^LJJ V/ J_U L.
(ml H2
Measured (2)
2.7
1.0
1.7
1.1
1.6
1.2
1.5
7.0
NA<4>
NA
NA
NA
c. Liu.uiJLVJ.Jw ujr
0/m3 air)
Calculated^3)
2.9
1.0
1.6
1.3
1.7
1.1
1.8
8.0
8.4
8.4
8.4
8.4
(2)
(3)
(4)
Average of duplicate LLL samples.
Average of multiple NERC-LV sampler results.
Single psychrometric measurement at start of sampling period.
NA = Np Analysis
                                       21

-------
                                        TABLE  2
                 Absolute  Humidity  Calculated  from Psychrometric Data
                    and Measured by NERC-LV Molecular Sieve Sampler
                                              Absolute Humidity
Location
(altitude )
(feet, MSL)
1
(8200)




2
(7600)



3
(7200)

4
(6800)



Grand Valley
(5100)
Las Vegas
(2000)
Temp.
(°F)
7
10
16
18
34
34
12
12
19
27
36
12
27
38
12
14
23
29
38
28
45
63

Relative
Humidity
(Percent)
56
59
62
56
46
53
37
57
52
48
53
40
52
51
36
56
52
48
51
57
41
48

(ml H2
Measured ^ '
1.2
1.2
1.7
1.4
2.7
2.6
1.0
1.1
1.6
1.5
2.7
1.0
1.8
2.6
1.0
1.2
1.5
1.9
2.6
3.1
2.7
7.0

T
0/nT air)
Calculated (2)
0.9
1.0
1.6
1.5
2.9
3.2
1.0
1.3
1.7
2.4
3.6
1.0
2.0
3.6
1.0
1.1
1.8
2.8
3.6
2.7
2.4
8.0

Collection
Date
3/3
3/6
3/7
2/12
3/4
2/10
3/3
3/6
3/7
2/12
2/10
3/3
2/12
2/10
3/3
3/6
3/7
2/12
2/10
2/12
2/11
5/7

(1)
(2)
Average of multiple samples.
Single psychrometric measurement at start of sampling period.
                                          22

-------
      Location
         ,(1)
         i<«
         l(l)
         1
         1
         1
         1
         1
         1
         1
         1A(2)
         1
         1A(2)

         2
         2
         2
         2
         2
         2
         2
         2
         2
         2

         3
         3
         3
         3
         3
         3

         4
         4
         4
         4
         4
         4
         4
         4
         4
         4
                                         TABLE 3
                          Tritium Concentrations in Atmospheric

                     Moisture as Collected by. TiERC-T,V and LLL Samplers
   Date
Collected

   2/10
   2/10
   2/12
   2/12
   3/3
   3/3
   3/4
   3/4
   3/4
   3/4
   3/6
   3/6
   3/6
   3/6
   3/7
   3/7
   3/7
   3/7

   2/10
   2/10
   2/12
   2/12
   3/3
   3/3
   3/6
   3/6
   3/7
   3/7

   2/10
   2/10
   2/12
   2/12
   3/3
   3/3

   2/10
   2/10
   2/12
   2/12
   3/3
   3/3
   3/6
   3/6
   3/7
   3/7
3
H Concentration
ILL Sampler
17
21
75
86
85
82
3.7
4.3
3.8
4.0
23
22
24
24
23
24
27
26
1.6
1.2
5.4
5
44
45
64
64
4.4
4.2
0.59
0.92
17
17
64
61
0.99
0.70
15
14
55
56
8.3
8.3
7.1
7.2
(pCi 3H/ml H20)
NERC Sampler
8.9
8.4
25
29
76
77
4.2
4.7
3.7
4.0
22
20
23
22
24
24
25
24
1.3
1.1
6.3
7.1
52
52
59
54
3.9
3.7
0.69
<0.4
16
16
54
52
0.94
0.90
15
14
41
32
7.2
7.4
5.8
6.1
   Data not used in statistical analysis

(2)
   See discussion section; data not used in statistical analysis.

                                        23

-------
                                        TABLE  4

                           Tritium Concentrations  in Air  as
                      Determined  from NERC-LV and LLL Samplers
Date
Location Collected
1 2/10

2/12

3/3

3/4



3/6

(3)
(3)
3/7

(3)
(3)
2 2/11

2/12

3/3

3/6

3/7

3 2/11

2/12

3/3

ii vjvxu
LLL Sam
67
54
117
133
75
72
10
12
10
11
23
22
24
24
37
39
44
42
5.8
4.3
13
13
42
44
44
44
7.4
7
2.1
3.3
34
34
62
59
   Using pCi/ml and calculated ml/m .
   Using pCi/ml and measured ml/m .
3                     33
 H Concentration (pCi  H/m  air)

                     NERC  Sampler

                        24
                        22

                        41
                        37

                        90
                        86
                        12
                        13
                        10
                        10

                        21
                        23
                        33
                        24

                        41
                        35
                        43
                        46

                        3.3
                        3
                        10
                        9.9

                        51
                        51

                        77
                        53

                        6
                        6.1

                        1.9
                                                                 24
                                                                 32
                                                                 52
                                                                 55
                                                                           (2)
(3)
   Location 1A.
                                          24

-------
Date
Location Collected
4 2/11

2/12

3/3

3/6

3/7


LLL Samp]
3.6
2.5
41
39
52
53
9.4
9.4
12
12
                                   Table  4  (Continued)

                                     3H Concentration  (pCi  3H/m3 air)

                                                          NERC Sampler^
                                                               2.4
                                                               2.4
                                                              28
                                                              25
                                                              37
                                                              32
                                                               9
                                                               9.7
                                                               9.2
                                                               8.2
   Using pCi/ml and calculated ml/m3.
/ Q\
   Using pCi/ml and measured ml/m3.
                                  Table 5
                           Analysis of Variance
                   Calculated and Measured Absolute Humidity
Source
Location
Types
Interaction
Error
DF
4
1
4
32
SS
2.94933
.315571
.656762
24.6920
MS
.737333
.315571
.164190
.771625
F
< 1
< 1
< 1

                                  Table 6
                           Analysis of Variance
                 3H Concentration from LLL and NERC-LV Samplers
Source
Location
Sampler
Interaction
Error
DF
3
1
3
64
SS
1625.93
62.0773
50.4385
41432.5
MS
541.978
62.0773
16.8128
647.384
F
< 1
< 1
< 1

                                          25

-------
                            APPENDICES


APPENDIX                                                      Page

A.    ABSOLUTE HUMIDITY CALCULATIONS FROM WET BULB -
      DRY BULB MEASUREMENTS                                     27

B-l.  ABSOLUTE HUMIDITY FIELD CALCULATIONS                      28

B-2.  FIGURE A, AW^ VS. ALTITUDE                               29

B-3.  FIGURE B, AIR DENSITY VS. ALTITUDE                        30
                                 26

-------
                        APPENDIX A
              Absolute Humidity Calculations  from
                Wet Bulb - Dry Bulb Measurements
     Using Wet Bulb and Dry Bulb -  Calculate  for  Sea  Level
Conditions as follows:

WB S 26°
                0 nsAA TJR
     H20 = 4.9 e0'03^ W  - 1.375  (DB  - WB) grains/lb air
26° < WB g 32°
     H20 = 7.1 e°-04025 WB -  1.375  (DB - WB)  grains/lb air
32° < WB g 50°
     TT _   .. 1  0.04025 WB    1  c,  /T._   rTT>N     .   ...   .
     HO = 7.1 e           -  1.57  (DB  - WB) grains/lb air
50° < WB
     H20 =9.3 e°'0352 WB - 1.57  (DB - WB)  grains/lb  air

     To above value of H?0 add  correction,  A  HO,  as  follows:

WB g 32°
     A H20 = 0.000245 x (Altitude)  x e°'0472  WB grains/lb  air
WB > 32°
     A H20 = 0.000328 x (Altitude)  x e0'0381  WB grains/lb  air
W = H20 + A H20

Calculate local pressure from Altitude:   Calculate volume of  moist air:
p	29.92	     „ _ 0.754  (DB + 459.7)  f    W
F"          288	 5.256     V          P           L
      (288  -  0.00198 x Altitude)
                                  r\  O Q CTJ
Calculate moisture content:  M = —'—	 :   ml/m^
                                     27

-------
                             APPENDIX B

                Absolute Humidity Field Calculations
1.  Find T   , TL^ and altitude* of sampling location.


2.  Consult psychrometric chart for sea-level humidity conditions,

    grains water per pound air.


3.  Consult Figure A for water correction factor at your altitude

    using T   from Step 1.


4.  Add the values of Step 2 and 3.
5.  Consult Figure B to find volume of moist air at your altitude,
      3
    FT  air per pound a:


6.  From Steps 4 and 5.
  3
FT  air per pound air using TD  from Step 1.
    Total Weight H0/# Air m Grains H
       FT3 Air/# Air          FT3 Air
7.  2.3 X Step 4 _ ml H 0/m  Air
       Step 5
*Request from Control by radio after arriving on station.
                                 28

-------
2       3
Figure A:
  567
ALTITUDE , MSL X 1000'
8
10
                                 29

-------
                   VOLUME OF MOIST AIR
                              TDB=100°F
                                                      •)  TDB=50°F
                                                      •>  TDB=°°F
                                                                          I
11
  0         2
    Figure B:
     6          8
MSL ALTITUDE, FT X 1000
10
12
14
                                     30

-------
                              DISTRIBUTION

 1  - 13   National  Environmental  Research Center, Las Vegas, Nevada
     14   Man!on E.  Gates,  Manager,  NVOO/AEC, Las Vegas, Nevada
     15   Robert H.  Thalgott,  NVOO/AEC,  Las  Vegas, Nevada
     16   Henry G.  Vermillion, NVOO/AEC, Las Vegas, Nevada
     17   Chief, NOB/DNA,  NVOO/AEC,  Las  Vegas,  Nevada
     18   Robert R.  Loux,  NVOO/AEC,  Las  Vegas,  Nevada
     19   Donald W.  Hendricks, NVOO/AEC, Las Vegas, Nevada
     20   Technical  Library, NVOO/AEC, Las Vegas, Nevada
     21   Mail  & Records,  NVOO/AEC,  Las  Vegas,  Nevada
     22   Martin B.  Biles,  DOS, USAEC, Washington, D.C.
     23   Director,  DMA,  USAEC, Washington,  D.C.
     24   Harold F.  Mueller, ARL/NOAA, NVOO/AEC,  Las Vegas, Nevada
     25   Gilbert J. Ferber, ARL/NOAA, Silver Spring, Maryland
     26   Stanley M. Greenfield,  Assistant Administrator for Research & Monitoring,
          EPA,  Washington,  D.C.
     27   William D. Rowe,  Deputy Assistant  Administrator for Radiation Programs,
          EPA,  Rockville,  Maryland
     28   Dr.  William A.  Mills, Dir.,  Diy. of Criteria  & Standards, Office of
          Radiation Programs, EPA, Rockville, Maryland
     29   Ernest D.  Harward, Acting  Director, Division  of Technology Assessment,
          Office of Radiation Programs,  EPA, Rockville, Maryland
     30   Bernd Kahn, Chief, Radiochemistry  & Nuclear Engineering, NERC,  EPA,
          Cincinnati, Ohio
31  - 32   Charles L. Weaver, Director,  Field Operations Division, Office  of
          Radiation Programs, EPA, Rockville, Maryland
     33   Gordon Everett,  Director,  Office of Technical Analysis, EPA,
          Washington, D.C.
     34   Kurt L. Feldmann, Managing Editor, Radiation  Data & Reports, ORP,  EPA,
          Rockville, Maryland
     35   Regional  Administrator, EPA,  Region IX, San Francisco, California
     36   Regional  Radiation Representative, EPA, Region  IX, San Francisco,  California
     37   Eastern Environmental Radiation  Laboratory, EPA, Montgomery, Alabama
     38   Library, EPA, Washington,  D.C.
     39   William C. King,  LLL, Mercury, Nevada
     40   James E.  Carothers, LLL, Livermore, California

-------
 DISTRIBUTION (continued)

     41   Charles I. Browne, LASL, Los Alamos, New Mexico
     42   Harry S. Jordan, LASL, Los Alamos, New Mexico
     43   Arden E. Bicker, REECo, Mercury, Nevada
     44   Savino W. Cavender, REECo, Mercury, Nevada
     45   Carter D. Broyles, Sandia Laboratories, Albuquerque,  New Mexico
     46   Robert H. Wilson, University of Rochester, Rochester,  New York
     47   Richard S. Davidson, Battelle Memorial Institute,  Columbus,  Ohio
     48   J. P. Corley, Battelle Memorial Institute, Rich!and,  Washington
     49   Frank E. Abbott, USAEC, Golden, Colorado
     50   John M. Ward, President, Desert Research Institute, University of
          Nevada, Reno
51  - 52   Technical Information Center, Oak Ridge, Tennessee (for public
          availability).

-------