oEPA
            United States
            Environmental Protection
            Agency
            Industrial Environmental Research
            Laboratory
            Cincinnati OH 45268
EPA-600/2-79-129
July 1979
            Research and Development
Evaluation of the
Ultraviolet-Ozone
and  Ultraviolet-
Oxidant Treatment
of  Pink Water

-------
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series  are:

      1.  Environmental Health Effects Research
      2.  Environmental Protection Technology
      3.  Ecological  Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

 This report has been assigned to the ENVIRONMENTAL  PROTECTION TECH-
 NOLOGY series. This series describes research performed to develop and dem-
 onstrate instrumentation, equipment, and methodology to repair or prevent en-
 vironmental degradation from point and non-point sources  of pollution. This work
 provides the new or improved technology required for the  control and treatment
 of pollution-sources to meet environmental  quality standards.
 This document is available to the public through the National Technical Informa-
 tion Service, Springfield, Virginia 22161.

-------
                                         EPA-600/2-79-129
                                         July 1979
     EVALUATION OF  THE  ULTRAVIOLET-OZONE AND
          ULTRAVIOLET-OXIDANT TREATMENT
                  OF  PINK WATER
                         by

                    Milton Roth
               Joseph M. Murphy,  Jr.
U.S. Army Armament Research and Development Command
     Large Caliber Weapon  Systems Laboratory
             Dover, New  Jersey 07801
              Grant No.  IAG-D6-0059
                 Project Officer

              Herbert S. Skovronek
      Industrial Pollution Control Division
 Industrial Waste Treatment Research Laboratory
            Edison, New Jersey  08817
  INDUSTRIAL ENVIRONMENTAL RESEARCH  LABORATORY
       OFFICE OF RESEARCH AND DEVELOPMENT
      U.S. ENVIRONMENTAL PROTECTION  AGENCY
             CINCINNATI, OHIO 45268


          U.S. Environmental >     '       •>-y
          Essie:) 5, Library '
          230 S. Dearborn Sti^^  ,       , _,
          Shicago, IL  60604

-------
                          DISCLAIMER
     This report has been reviewed by the Industrial Environmen-
tal Research Laboratory-Cincinnati, U.S. Environmental Protec-
tion Agency, and approved for publication.  Approval does not
signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute en-
dorsement or recommendation for use.
                               11

-------
                           FOREWORD


     When energy and material resources are extracted, processed,
converted, and used, the related pollutional impacts on our en-
vironment and even on our health often require  that new and in-
creasingly more efficient pollution control methods be used.
The Industrial Environmental Research Laboratory-Cincinnati
(lERL-Ci) assists in developing and demonstrating new and im-
proved methodologies that will meet these needs both efficiently
and economically.

     Munitions wastes present some unique problems related to
disposal or destruction, but are also similar to or representa-
tive of complex organic pollutants frequently encountered by
the chemical industry.  As such they offer an opportunity to
evaluate new and novel technology for water pollution control.
This report documents the investigation of two such innovative
schemes for the destruction of such pollutants, presenting data
on the effectiveness of three oxidants, ozone, hydrogen perox-
ide, and a proprietary peroxide salt when used in conjunciton
with irradiation by ultraviolet light.  The results of this
study, in addition to being of direct interest to the military,
will also be transferable to the control of a vast array of
suspect pollutants encountered by the chemical industry.  Devel-
opment of such technology will assist industry in eliminating
its chemical pollutants and meeting anticipated discharge re-
quirements.   For further information concerning this subject,
the Industrial Pollution Control Division should be contacted.

                                 David G.  Stephan
                                     Director
                   Industrial Environmental Research Laboratory
                                    Cincinnati
                              111

-------
                           ABSTRACT

     Pink water, a solution of trinitrotoluene (TNT)  and other
nitrobodies, is a major pollutant at ammunition plants engaged
either in the manufacture of TNT or in the loading, assembly,
and. packing of bombs and shells.  Purification of the water is
currently accomplished by carbon adsorption at the more modern-
ized installations where the carbon is used once and then
burned; the result is a high cost operation and a potential air
pollution problem.  Although thermal regeneration of activated
carbon with adsorbed explosives has been piloted, and appears to
be cost effective, it has yet to be implemented at Army Ammuni-
tion Plants.  As an alternative to carbon adsorption for the
treatment of pink water, two new methods are reported here, the
first involving the use of ultraviolet  (uv)-ozone and the sec-
ond using a uv-oxidant combination.

     In the first study, a 3.79 cubic meters per day  (m pd),
equivalent to 1000 gallons per day  (gpd), uv-ozone pilot system
was evaluated.  It was found that dissolved TNT and RDX were
reduced to less than one milligram per  liter  (mg/L) with no by-
products requiring disposal.  A design  for a  19 m3pd   (5000
gpd) pilot plant was proposed.

     In the second study, a uv-oxidant  process for treatment of
pink water was examined.  Commercially  available uv water puri-
fication units in conjunction with an oxidant such as hydrogen
peroxide or Oxone,  (a blend of potassium persulfate oxidants),
were evaluated for their efficiency in  treating pink water.
Variables such as film depth, dilution, uv wavelength,  and  oper-
ation of the units in series rather than in parallel were  ex-
amined.  A design for a 379 m3pd  (100,000 gpd) pink-water  treat-
ment plant was proposed.

     This report  is submitted by US ARRADCOM-DOVER, NJ  (formerly
Picatinny Arsenal) in fulfillment of Interagency Agreement  D6-
0059 with the US  Environmental Protection Agency.  The  work was
carried out during the period 6 June 76 to 30 Dec  77.
                               iv

-------
                          CONTENTS
Foreword ...............................................
Abstract ...............................................  iv
Figures ................................. ... ............  vi
Tables ................................................ \ vn
Acknowledgement .................................... ... viii

   1 .  Introduction ...................................... i
   2 .  Conclusions ....................................... 2
   3 .  Recommendations ................................... 3
   4.  UV-Ozone Treatment of Pink Water .................. 4
           Description of Ultrox Pilot Plant ........... ! ! 4
           Pilot Plant Operation ......................... 5
           Test Procedures .................. ............. 5
           Preliminary Testing ........................... 5
           Pink Water Tests .............................. 7
           Specific Analysis ............................. 7
           Discussion of Test Results ......... . .......... 7
           Design of 19 m3/d (5,000 gpd)  UV-Ozone
              Pilot Plant ................................ 11
   5.  UV-Oxidant Treatment of Pink Water ................ 18
           Description of UV-Oxidant System ............. '. 18
           Parameter Variations for Optimization
              of the System .............................. 19
           Chemical Reactions ............................ 26
           Additional Analyses ........................... 26
References
                                                         ,29
                             v

-------
                            FIGURES

Number
  1 - UV-ozone reactor assembly
  2 - Pattern of water flow through pilot
      plant reactor
  3 - The 920 m3pd (5000 gpd) pilot plant
      assembly
  4 - Effect of pink water dilution on
      TOC reduction
  5 - Basic design of the UV-oxidant
      system
                               vi

-------
                            TABLES

Number
                                                       Page
  1 - UV-Ozone treatement of TNT in water 	   6

  2 - UV-Ozone treatment of ARRADCOM
      pink water 	   8

  3 - Comparison of synthetic sample
      and ARRADCOM pink water tests 	   9

  4 - Analysis of synthetic sample
      and ARRADCOM pink water 	   10

  5 - Effect of oxidant in the UV-2000
      and UV-500 systems 	   20

  6 - Analyses of various concentrations
      of Oxone and Oxone and H~0^  	                 01
                              / 2           ...........   6±

  7 - Evaluation of UV-2000 system
      operating in series mode 	   23

  8 - Configurations for pink water
      pilot  plant 	   24
                             vii

-------
                        ACKNOWLEDGEMENT


     All experimental work for the uv-ozone portion of this
study was conducted by Westgate Research Corporation, West Los
Angeles, California.  The work performed on the uv-oxidant pro-
cess was conducted by the Naval Weapons Support Center at Crane,
Indiana.
                              Vlll

-------
                           SECTION 1

                         INTRODUCTION

     Pink water is a unique pollutant which is generated from
trinitrotoluene (TNT) manufacturing plants, and load, assemble,
and pack  (LAP) plants.  Pink water also arises from unloading
or demilitarization of munitions.  Pink water from manufacturing
operations may contain  Qf-TNT, TNT isomers, and dinitrotoluenes
(DNT).  Pink water generated from LAP and demilitarization oper-
ations may contain CX-TNT, cyclotrimethylenetrinitramine (RDX),
cyclotetramethylenetetranitramine (HMX), and wax.

     The volumes and concentrations of pink water streams vary
widely but, at full mobilization, volumes of 379 m3pd (100,000
gpd)^per line at concentrations of 100-150 ppm of TNT are not
atypical.  A complete review of TNT manufacturing and its wastes
can be found in reference 1.  Currently, activated carbon is the
most widely used process for pink water abatement.  The carbon
is used once and then burned; the result is a high-cost opera-
tion and an air pollution problem.  Although processes for ther-
mal regeneration and reuse of carbon have been tried for the ex-
plosive < laden carbon, all suffered from a high risk of explosion
and a high loss of carbon.  A new process for thermal regenera-
tion of activated carbon, using a rotary kiln with a reducing
atmosphere, has been piloted and appears to be safe and cost
effective.  However,  it has yet to be implemented.  Concurrent
with thermal regeneration, new processes for abatement of pollu-
tion from pink water are also being investigated.  This report
describes the uv-ozone and uv-oxidant processes,  two new tech-
nologies  for the abatement of pink water discharges.

-------
                           SECTION 2

                          CONCLUSIONS
UV-OZONE TREATMENT OF PINK WATER

     Data obtained from a 3.79 m3pd (1000 gpd) pilot plant dem-
onstrated the feasibility of using uv-ozone to oxidize TNT and
RDX in aqueous solution and produce an effluent containing less
than 1 mg/L of TNT and RDX.

     A design was proposed for a  19 m3pd  (5,000 gpd) modular
uv-ozone pilot plant for treating pink water.  The proposed de-
sign would-allow a single unit to be used for low-flow applica-
tions, or multiples to be used in series for full mobilization
requirements.

UV-OXIDANT TREATMENT OF PINK WATER

     From the studies performed, using commercially available
units for water purification via uv irradiation, the procedures
incorporating Oxone* are more efficient than #2®2 *~n reducing
the total organic carbon levels.

     A design for a 379 m3pd  (100,000 gpd) uv-oxidant treatment
plant was proposed.
*  Oxone, a product of the DuPont Corporation, is a mixture of
potassium monopersulfate, potassium hydrogen sulfate, and potas-
sium  sulfate.

-------
                           SECTION 3

                        RECOMMENDATIONS
UV-OZONE TREATMENT OF PINK WATER

     A  19 m pd  (5000 gpd) pilot plant should be evaluated at
an ammunition plant in order to determine the operating condi-
tions required for achieving the minimum fixed and operating
costs for a 379 m3pd (100,000 gpd) plant.

     An economic analysis should be performed on the 19 m3pd
(5000 gpd) unit after process parameters have been optimized.

UV-OXIDANT TREATMENT OF PINK WATER

     Equipment should be found for the uv-oxidant process that
would be suitable for purification of large volumes of pink
water.

     Further studies of the process should be made only if more
cost-effective equipment is developed to process the volumes
and concentrations of pink water to be expected in a full-scale
plant.

-------
                           SECTION 4

               UV-OZONE TREATMENT OF PINK WATER


     The objective of the uv-ozone testing was to develop design
criteria and cost figures for a 379 m3pd (100,000 gpd) pink-
water treatment plant.  Tests were made in a 1000 gpd uv-ozone
reactor to estimate the conditions for minimum power demand and
retention time required to obtain an effluent containing less
than 1 mg/L of TNT and less than 1 mg/L of RDX.

DESCRIPTION OF ULTROX PILOT PLANT

     This pilot plant, recently developed by the Westgate Re-
search Corporation, is designed to demonstrate the practicality
and cost effectiveness of UV-ozone oxidation for destroying or-
ganics in wastewater.  The pilot plant can vary input and in-
tensity of uv radiation, ozone mass transfer, mixing, and water
flow rate.  The reactor is fabricated from 304 stainless steel
which is passivated and electropolished to reduce chemical
attack and increase uv reflectivity.  A separate NEMA cabinet
houses the ballasts for the uv lamps.

     The reactor can accommodate 30 low-pressure, 65 watt uv
lamps and has six operating stages.  From 0 to 30 lamps can be
turned on in a test run.  Ozone is uniformally diffused from
the base of the reactor through spherical spargers that generate
gas bubbles of 2.5 mm diameter.  The number of spargers can be
varied from stage to stage, so that the overall mass  transfer of
ozone can be changed as desired.

     The reactor is designed  for low-pressure operation  (2 psig,
maximum) to reduce the cost for pumping water and compressing
air for ozone generation.  Low-pressure operation  not only
provides greater safety but also reduces the thickness, weight,
and cost of materials of construction.  However, it reduces the
effective oxidant concentration.

-------
 PILOT PLANT OPERATION

      In each stage the water is contacted by the ozone and, in
 certain stages,  uv radiation.   The flow rate of the incoming
 pink water is measured by a rotameter located between the pump
 and the reactor  inlet.  The water is fed to the reactor by the
 use of a gear type,  sealless,  magnetic drive pump with integral,
 solid-state speed control.   The flow of pink water through the
 reactor can be varied by this  pump from 1.27 x 10~5 mVs to 1.27
 x 10~4 m-Vs (0.2 to  2 gpm)  with retention times varying from
 37.5 to 375 minutes.

      The purified water leaving the reactor overflows into a
 gas-water separator  in order to eliminate entrainroent of water
 in the exhaust gas,  and then drains into a sump.   No level con-
 trols are required within the  reactor.

 TEST PROCEDURES

      From previous experience  with pink water and waters of sim-
 ilar composition,  it  had been  determined that the following
 variables have the greatest influence on total power demand,
 reactor size,  and retention time:

      1.   concentration of 0. in sparging gas

      2.   intensity of uv radiation

      3.   placement of uv lamps  within reactor

      4.   temperature  of  incoming water  flow rate

      5.   composition  and concentration  of  solute

 PRELIMINARY TESTING

     Analysis  provided by ARRADCOM showed  that  the pink  water
 contained  140  mg/L TNT,  72 mg/L RDX,  10  mg/L  wax  and  68  mg/L of
 TOC.  The  synthetic pink water prepared  for the shakedown  tests
 was mixed  to contain  this concentration  of TNT.   An  in-line fil-
 ter was inserted at the  inlet of the  pilot plant  to  remove  resi-
 dual suspended solids.   The results of experiments with  the syn-
 thetic solution,  shown in Table 1, helped  to  estimate the opera-
 ting levels for the number of uv lamps per stage  and  the ozone
mass flow required to treat actual pink water.

-------
CTi
                        Table  1. UV-Ozone treatment  of  TNT in water





                                      Total  organic carbon (mg/L)a
Test No.
1022
1023
1024
1025
1026
Influent >
60
55
66
66
54
Stages 1-3
13
7
5
5
10
Stages 4-6°
3.0
2.5
1.2
2.0
6.5
           a.   Based on a flow rate  of  1200  + 100 mg of ozone per minute.



           b.   7 + 2 W uv per mg TOC



           c.   40 + 20 W uv per mg TOC



           d.   Temperatures were about  30°C  and residence time was 118 minutes.

-------
 PINK WATER TESTS

      Results  of pilot plant tests using ARRADCOM pink water are
 summarized in Table  2.   The first test (No.  1027)  was run at
 conditions similar to test No.  1026  for TNT  in water.  Greater
 resistance to oxidation  was found when using pink water than
 TNT  in water.   Therefore,  the  residence time and number of uv
 lamps were increased in  subsequent tests (1028 and 1029).

      Table 3  shows a comparison of conditions and results for
 TNT  in water  and pink water in  tests  1024  and 1029,  respective-
 ly.   The  ozone to organic  carbon ratios are  about the same for
 the  first three and  the  last three stages; however,  the ratio of
 uv input  power to carbon had to be increased in both the first
 three and second three stages  in order to  achieve 3  mg/L TOC
 after six stages for the pink water.

 SPECIFIC  ANALYSIS

     _Table 4  indicates that less than  1 mg/L TNT and 1  mg/L RDX
 remained  in the effluent,  but there was also some  unidentified
 residue both with TNT in water  and with pink water.   Test  No.
 1029  indicated that  the  TNT and RDX levels were below 1 mg/L
 after the pink water had passed through the  first  three stages
 of the reactor.  This result was most  encouraging, since it
 means that the residence time,  the number of uv lamps,  and the
 ozone input can be reduced by half of  the amounts  used  in  test
 No. 1029.

 DISCUSSION OF  TEST RESULTS

      The  mass  ratio  of ozone to  TOC in  test  No.  1029  for the
 first three stages was 13, which  is 1.6  times  the  stoichioroetric
 ratio for  oxidation  of carbon to  CO2.   Bench and pilot  plant
 tests on  a wide variety  of wastewaters  have  indicated that the
minimum stoichiometric ratio of  O3 to TOC varies between 1.3
 and 2.0,  depending on the original TOC  concentration  and the
 chemical  structure of the organic  contaminants.   (This  calcula-
 tion  does not  include hydrogen and nitrogen  oxidation)  Thus,
pink water behaves in a manner similar  to other wastewaters
with dissolved organic matter.

-------
00
                     Table  2. UV-Ozone treatment  of  ARRADCOM pink water






                                       Total Organic  Carbon (mg/L}



           Test No.e         Influent             Stages  1-3°         Stages 4-6d



             1027a                68                  22.0                17.0



             1028b                67                    6.5                 5.0



             1029b                70                    5.0                 3.0
            a.   Residence  time  -  118 min.



            b.   Residence  time  -  177 min.



            c.   1200  + 100 mg ozone per min;  7  +  2  W uv per mg TOC



            d.   1200  + 100 mg ozone per min;  40 + 20 W uv per ma TOC



            e.   Temperature 29°-30°C in each  test

-------
                  Table 3.  Comparison of synthetic and
                           ARRADCOM pink water tests


      Ozone/TOC mass ratio        Input, Watts/mg               TOC (mg/L)
      	(mg/mg)	  	  	
Test
 No.  Stages 1-3  stages 4-6  stages 1-3  stages 4-6  Influent stage 3 stage 6


1024a     15         198          7          60          66       5      1.2


1029b     13         180         11         140          70       5        3
a.  Synthetic sample


b.  Pink water

-------
      Table 4. Analysis of synthetic and ARRADCOM pink water
Test No.         Sample               TNT            RDX     Residue
                                    (mg/L)         (mg/L)
1020-1      Synthetic influent        76              -         -

1020-2      Synthetic effluent       <1

1023-2      Synthetic effluent       <1              -        11

1024-2      Synthetic effluent       <1              -         8

1025-2      Synthetic effluent       
-------
     Another way to examine  the  total  oxidation  of  TNT and RDX
is from the following  stoichiometric relationships:

     TNT

     C6H2CH3  (N02)3 +  18  0   ---- -0-  7 CC>2 +  3  HNO +  18 0   + H  0


     RDX
     C3H N3  (N02)3 + 18 03  ----- o  3 CC>2 +  6 HNO  +  18  0


     The pink water contained 140 mg/L TNT  and  72 mg/L  RDX.
According to the above equations , the theoretical amount  of
ozone required per liter to carry out the complete oxidations
is 813 mg.  Since the testing found that 910 mg/L was required
to obtain an acceptable effluent, the ratio of  actual to  stoi-
chiometric ozone is 1.12:1  (or an ozone efficiency of 89.3%).

     Although it was found that more uv lamps were required  to
oxidize the pink water than the solution of TNT in water,  the
number of lamps required per unit volume in each reaction  stage
has not been defined.  Further tests will be required to  estab-
lish the number of lamps that produce the optimum effect  at
least cost.

DESIGN OF 19 m3/d (5,000 gpd)  UV-OZONE PILOT PLANT

     With the data obtained from the 3.8 mjpd (1000 gpd)  reac-
tor,  the following criteria were recommended for the 19 m^pd
(5000 gpd) pilot plant.

Design Parameters

     Reactor volume .................. 2.6 m  (675 gal)

     Reactor dimensions (WxLxH) --- 0. 9x1.  8x1. 5m
                                       (0.3 x 6 x 5 ft)

     Water flow rate ................. 19  m3pd  (5,000 gpd)

     °3  .............................. 17  kg/day (37.5 Ib/day)

     uv  lamps (65 watts each) ......... 144
                              11

-------
Assembly

     The major components of the pilot plant are:

     (1)  reactor assembly

     (2)  NEMA ballast enclosure

     (3)  ozone generator

     Components 1 and 2 are assembled on one skid and component
3 is mounted on another skid.

     The reactor, shown in figure 1, consists of a stainless
steel tank with baffles, and a cover assembly made up of the
reactor cover, ozone diffuser, uv lamps, and supporting struc-
ture.

Construction

     The reactor tank is fabricated from 316 stainless steel
sheet about 0.5 cm  (3/16 in) thick.  All parts are certified
heliarc welded.  A  10.2 cm  (4 in) wide lip is welded to the top
of the tank to form a gasket flange containing a groove to ac-
commodate a Hypalon seal.  The tank is mounted onto the metal
skid by bolts.

     Five baffles are located vertically to create six reaction
stages.  Water flows in an undulating path from  stage to stage
as shown in figure  2.  The baffles are designed  for easy re-
moval to permit alteration of the number of reaction stages as
desired.

     The ozone inlet manifolds,  lamp venting tubes, and lamp
conduits are welded across the reactor and provide adequate
horizontal stiffening.  Vertical stiffening is achieved by 3
strips of stainless steel welded to the cover plate, the mani-
folds,  the gas vent and the wiring conduits.  A  diagram of the
proposed pilot system is shown in figure 3.
                               12

-------
  INLET
EQUALIZER
                 I   i
                                1.14m
                               (45.00 IN)
                fl*.    I  i   
-------
PINK
WATER
                     UV LAMPS
                      A




°r
o
o
o
o
vy
t k
fci ' i
-H 1

^
o
o
o
^
V
O
o
o
o
J
V
To
o
o
o
V.
°r
o
o
o
o
J
>°0
o
o
o


fiA
r-SEF
/
— o

5 '
PUR FIED
WATER
        SOLID STATE
        CONTROLLED
        GEAR PUMP
                                                                        SPENT
                                                                        -OZOI
                                                                        GAS
             FIG 2. PATTERN OF WATER FLOW THROUGH PILOT PLANT REACTOR

-------
PINK
WATER
                     OZONE
                     GENERATOR
                                                                         "^•DEOZONIZER
                                                                             GAS-WATER
                                                                             SEPARATOR

                                                                              EFFLUENT
                                                                             UV LAMP
                                                                            BALLASTS
                                  ED DISPLAY
                                 OF UV LAMPS
WATER
FILTER
                                                           UV-O3
                                                           -REACTOR
                         FIG 3. THE 920 M3 PD (5000 GPD) PILOT PLANT ASSEMBLY

-------
Cover Assembly

     The following openings are punched into the cover plate:
144 holes, 2.86 cm (1 1/8 in.) diameter, for the quartz tubes
that enclose the uv lamps.  Nipples are welded at the top sur-
face of these openings so that the quartz tubes are sealed to
the cover by compression nuts with 0-rings.  6 holes, 2.54 cm
(1 in.) diameter ,. for the spent ozone gas outlets.  6 holes,
3.81 cm (1 in.) square, for mating with the lamp support struc
ture.  6 holes, 1.59 cm (5/8 in.) diameter, for the outboard
lamp support and cooling air vent lines.  2 nipples, 3.81 cm
(1% in.) with NPT, for water inlets.
     The lower lamp support structure consists of 3.81 cm
in..) square tubes with 0.159 cm  (0.0625 in.) wall thickness.
Holes of 2.54 cm  (1 in.) diameter are drilled on the upper side
of the tubes at appropriate positions to install the quartz
tube support and sealing assemblies which are welded to the
upper side of the tube.  A 1.27  cm  (% in.) diameter hole is
drilled through the outboard end of the conduit to attach the
vent tube which also acts as a support for the end of the
square tube.  The center of the  square tube is supported by
welding the ozone line to the diffusers running parallel to  the
conduits .

NEMA Ballast Enclosure

     A standard 1.52 m x 0.91 m  x 0.31 m  (5 ft x 3 ft x 1 ft) ,
16 gauge NEMA cabinet is used to contain and cool the 72 lamp
ballasts.  The ballasts are mounted on racks within the cabinet,
A rotary air blower at the base  of  the cabinet directs the air
upward for cooling the ballasts.

     The cabinet door contains a LED display which indicates
the number uv lamps "ON" in the  reactor.  The door is sealed to
the NEMA housing by means of elastomer gasketing and spring-
loaded screw clamps.
                               16

-------
Ozone Generators

     A number of manufacturers can supply generators which meet
the 19 m pd (5,000 gpd) pilot plant requirement of 17 kg  (37.5
Ibs) of ozone per day.  Manufacturers, such as OREC or PCI
would supply two 9 kg  (20 Ib) per day ozone generators since
neither produces an 18 kgs (40 pound) generator.  The Welsbach
generator is oversized but can produce 18 kg  (40 Ibs) per day
efficiently by lowering the input voltage by means of a vari-
able voltage transformer.
                              17

-------
                           SECTION 6

              UV-OXIDANT TREATMENT OF PINK WATER
     Bench-scale studies have shown that short wavelength uv
and H202 is very promising for the treatment of explosive con-
taminated effluents (2).  This photo-oxidative treatment ap-
pears to be effective not only in decolorizing pink water, but
also in destroying TNT,  RDX, HMX, and other nitrobodies with a
concurrent reduction in TOC concentration.  These effects have
been determined by gas,  liquid, and thin-layer chromatography
(GC, LC, TLC), total organic carbon (TOC), and C-14 labeled
TNT assays.

DESCRIPTION OF UV-OXIDANT SYSTEM

     A small-scale pilot system was used in which four uv water
purification units were connected in series.  These were Model
2000 units, manufactured by the Ultradynamics Corporation of
Santa Monica, California, containing mercury vapor lamps  (254
nm) protected by quartz jackets that are continuously cleaned
by a hydraulically operated wiper assembly.  The capacity of
each chamber is approximately 22.8 liters  (6 gallons).  The
maximum film depth is approximately 5.72 cm  (2% in.).

     Pink water from a bomb loading and steamout operation at
Naval Weapon Support Center, Crane, Indiana was used in the
study.  Before being pumped into the system, the water was fil-
tered to remove suspended solids which may interfere with the
treatment.  Because of the intense color, which would reduce
the efficiency of the system, the pink water was diluted with
tap water to yield the following results:

                        Analysis, mg/L

                  1;1 Dilution   lj3 Dilution

             TNT  70.9           34.1

             RDX  72.4           27.0

             HMX   9.4            4.1

             TOC  52.6           27.0
                              18

-------
      From previous studies (2)  it was found that 0.1% H,.,O2 is
 the optimum concentration to use in treating pink water.   A 35%
 solution of H2O2  (Fisher Scientific Co.)  was added to the pink
 water to be treated to yield a  final concentration of 0.1% H-0..,.
 The water was  recycled from a reservoir through the uv units at
 a flow rate of 2.73 m3/hr (720  gph)  in a continuous flow  mode.
 The actual residence time of the liquid in any one unit is about
 0.5 minutes.   A Model 500 unit  (also manufactured by Ultradynam-
 ics Corp.)  with a  2.54 cm (1 in.)  film depth has been used to
 study the efficiency of the photo-oxidative treatment of  pink
 water with respect to film depth.   The capacity of its chamber
 is approximately 2.84 L (3/4 gal).   One gallon of pink water
 (undiluted or  diluted)  containing 0.1% H2O2 was recirculated
 through the unit and back into  a reservoir at a flow rate of
 approximately  0.4  m3/h (105 gph).

 PARAMETER VARIATIONS FOR OPTIMIZATION OF  THE SYSTEM

 Oxidizing Agents and Film Depth Studies

      To test the effectiveness  of  a  different oxidizing agent,
 a monopersulfate compound,  Oxone (Dupont  Chemical Co.), was
 substitured for the H202 in a number of studies.   Overall effi-
 ciency was  significantly increased by using 0.3%  Oxone in place
 of 0.1% H202.   Table 5  illustrates  the results of treatment of
 a 1:3  diluted  pink  water solution  in the  UV 2000  (4 unit  sys-
 tem)  at a flow of  2.73  m3/d (720 gpd),  and in the UV 500  sys-
 tem at a flow  rate  of 0.4 mj/h  (105  gph)  using either 0.1% I^O-
 or 0.3% Oxone.  The solutions were analyzed for wxplosives by
 liquid,  gas, and thin-layer chfomatography for dissolved  organic
 matter by TOC.

     From the  results in table  5 it  is  evident that Oxone is
 far superior to H2O2.   This  can be seen with respect to decolor-
 ization time,  TNT elimination,  and degradation of polynitroaro-
 matic  by-products with  corresponding reductions in TOC concen-
 trations.  Also illustrated by  these results is that,  with
 either  oxidant, the  2.54 cm  (1  in.)  film  depth of the  UV  500
 system  enhances the  efficiency  of the  treatment as compared
 with the  5.72  cm (2%  in.) film  depth of the  UV 2000  system.

     Studies were undertaken  in which  the  concentration of Oxone
 employed  in the treatment was varied.   Additional  studies  were
made to examine the potential of a combination H20;?/Oxone  treat-
ment.  Because of the limited time available,  it was only  pos-
 sible to examine the  effects  of using diluted  pink water  and
 0.3% and  0.2% Oxone in the UV 2000 units,  and  0.3% and  0.7%
Oxone in the treatment of undiluted  pink water in  the  UV  500
system.  A combination of 01% H20,> and  0.1%  Oxone was  also  ex-
amined in the UV 2000 system.   TaBle 6  illustrates the  param-
eters and results  of the treatment.
                              19

-------
                Table 5.  Effects of Oxidant in the
                         UV-2000 and UV-500 systems
System/Oxidant%

UV-2000
0.1 H000
UV-2000
0.3 Oxone
UV-500
0.1 H
UV-500
0.3 Oxone
UV Expo-
sure (hr)

  0.0
  3.0
  5.0
  0.0
  3.0
  5.0
  0.0
  0.5
  1.5
  0.0
  0.5
  1.5
                                        Water analyses (mg/L)
TNT
30.0
0.01
0.01
34.0
0.01
0.01
29.0
0.8
0.05
29.0
0.3
0.01
RDX
26.8
0.01
0.0.1
27.4
0.01
0.01
20.6
0.01
0.01
20.6
0.01
0.01
HMX
3.0
0.02
0.02
2.8
0.0
0.02
3.1
0.02
0.02
3.1
0.02
0.02
TOC*
24.8
5.6
1.6
24.2
0.1
0.2
23.0
17.4
8.0
23.0
7.6
1.8
* Includes other nitroaromatic  compounds

-------
             Table 6. Effects of oxidant concentrations
      Operating parameters
Systems/
Oxidant
UV-2000
0.3% Oxone
UV-2000
0.3% Oxone
UV-2000
0.2% Oxone
UV-2000
0.1% Oxone,
0.1% H202
UV-500
0.7% Oxone
UV-500
0-3% Oxone
Dilutions
1:4
1:2
1:4
1:4
0
1:4
UV Expo-
sure (hr)
3
3
3
3
6
3
TNT
28/0.01
64/0.01
23/0.01
23/0.01
138/0.02
29.0/0.01
RDX
26/0.01
55/0.1
19/0.01
20/0.01
3.1/0.02
20.6/0.01
HMX
27/0.02
8.5/0.02
2.4/0.02
2.5/0.02
15.0/0.01
3.1/0.02
TOC*
22.8/0
46.1/3
21.8/0
25.0/0
86.6/9
23.0/1

.3
.4
.4
.9
.0
.8
*Includes other nitroaromatic compounds

-------
      Treatment of 1:3 diluted pink water in the UV 2000 system
with 0.3%, 0.2%, or 0.1% Oxone plus 0.1% H202 are comparable.
The TOC levels were appreciably reduced, and no detectable
amounts of explosives or polynitroaromatics were found after 3
hours of exposure.  Treatment of an undiluted pink water solu-
tion with 0.7% Oxone in the UV 500 system is not as efficient
as treatment of a 1:3 diluted pink water solution with 0.3%
Oxone, but it is feasible if environmental trade-offs are al-
lowed.  The same is true of a 1:1 dilution of pink water treated
in the UV 2000 system with 0.3% Oxone.

Flow Rates and Operations of Units In Series

      Two flow rates were examined in the UV 2000 system.  There
appeared to be no major difference in results between flow rates
of 2.04 and 2.73 rc3ph (540 and 720 gph) while operating in a
continuous flow mode.  If the ultimate treatment will be a di-
rect passage, and not a recirculation of the pink water through
the UV system, the flow rate will not be critical, but the total
UV exposure time or contact time of the solution in the units
will be.  Operation in a continuous flow mode through 1 to 4 UV
2000 units in series appeared to have no major effect on over-
all efficiency as is seen in Table 7.  This table illustrates
the results of treatmentof a 1:3 dilution of pink water solu-
tion with 0.1% H202 through 1 to 4 units in series mode.

      It would be extremely difficult to determine the overall
effect of units in series with just 4 units operating in a con-
tinuous flow mode with the small volumes used.  It is hypothe-
sized that as the color of the solution disappears, the effici-
ency of the system increases because there is no loss of energy
by color  absorption.  This can be demonstrated by diluting the
pink water and examining TOC reduction per unit time.  Table 8
illustrates the results of dilution studies.  There is a de-
crease of 91% in  the TOC level of the pink water after_3 hours
of exposure in the UV 2000 system when the pink water is diluted
1:7 but only a 38% decrease with a 1:1 dilution.  The relation-
ship among these  dilutions and the decrease in TOC observed
after the exposure period appears to be linear, as illustrated
in Figure 4.  These results confirm the hypothesis that the pro-
cess increases in efficiency with a decrease in color.

Comparison of Effect of Wavelengths

      Aliquots of undiluted pink water containing 0.1% H20~ were
exposed to radiation of either 254 or  375 nm, in static modes,
to determine if the wavelength had an  effect on the destruction
of explosives and a reduction in TOC.  The  film depth was  40 nm,
with  30 minutes of radiation.  After  the irradiation at 254  nm,
the pink  water was completely decolorized and the TOC had  de-
                               22

-------
                 Table 7. Evaluation of UV-2000 system in series mode




           	Parameters a	         Explosives content  (mg/L)


           Units (No.)      Exposure (hr)       TNT      RDX      HMX      TOC-


             1                 5             <0.01    <-0.01    
-------
        Table 8. Configurations for pink water pilot plant
            Parameters
Analyses, mg/L (Original/Final)
Dilution/
UV hrs
1:4/3
1:2/3
1:4/3
1:4/3

1:4/5
1:2/5
Volume
m3pd (gpd)
76
(20,000)
38
(10,000)
76
(20,000)
76
(20,000)

76
(20,000)
38
(10,000)
Oxidant
0.3%
Oxone
0.3%
Oxone
0.2%
Oxone
0.1%
Oxone
+ 0.1%
H2°2
0.1%
H2°2
0.1%
H2°2
TNT
28/ **
64/**
23/**
23/**

30/**
64/**
RDX
26/**
55/**
19/**
20/**

27/**
48/**
HMX
27/**
5/**
4/**
5/**

O/**
9/**
TOC
23/0
46/3
22/0
25/0

23/1.
45/13
.3
.4
.4
.9

6

 * Includes other nitroaromatics



** <0.05

-------
to
U1
   100


    90



    80

ss

tt  70
ixi  /u

<
5  60
*


I  50

111
w  40
UJ
cc

S  30
                  20


                  10 h
                                         _L
                                              _L
                                         1:2                  1:4

                                           PINK WATER DILUTION
                                                                  1:8
                                FIG 4. EFFECT OF DILUTION ON TOC REDUCTION IN PINK
                                      WATER AFTER 3 HOURS IN UV-2000 APPARATUS

-------
creased from 89 mg/L to 9 mg/L.  On the other hand, the color of
the solution exposed at 375 nm was intensified, and no loss of
TOC was noted.  However, the TNT level in the pink water dropped
to 0.02 mg/L after exposure at 375 nm but only to 0.15 mg/L
after exposure at 254 nm indicating that the higher wavelength
uv-light can alter the TNT molecule but cannot oxidize it.  The
significant reduction of .TOC and rapid decolorization observed
with the sample irradiated at 254 nm is a good indication that
this shorter wavelength has a pronounced effect on overall effi-
ciency of the photo-oxidation system.

CHEMICAL REACTIONS

     The proposed oxidation mechanism is the production of free
OH radicals which are ultimately responsible for the destruction
of the explosives in the water upon exposure to uv light.  The
reaction mechanisms are represented below with RH representing
the explosive.

     a.  Peroxide mechanism


         H2°2	"€s" 2*OH

         RH + *OH-«» R* + HO

         R*	e> cleavage of ring

     b.  Persulfate mechanism

         HOOSO  	1>  "OH +  "OSO
              •3                  J
         RH + "OH —o R* + H20
              UV
         R*	-o cleavage of ring

ADDITIONAL ANALYSES

     In addition to explosives analyses, treated samples with
high TOC levels  (2 mg/L) were  analyzed  for  nitrosoamines and
sulfonates.  No detectable  levels of these  products  could be
found by TLC procedures  (sensitivity 50 ppb) in samples  irradi-
ated in the UV 2000 or UV  500  system with either H2O2  or Oxone.

     In all treatments with K?0?' the Peroxide level also was
monitored.  The range of residual peroxide  was between 30 and
100 ppm after 3 or 5 hours  of  exposure  in the UV-2000  system and
1.5 hour in the UV-500  system.

     The pH of the pink water  solutions containing H20,  or Oxone
was monitored before  and after treatment.   Before  treatment, the
pH values  averaged 7.3  and  3.2,  respectively.  After treatment,
the values fell to 6.4  and  2.4.  If  the extremely  low  pH of  the

                               26

-------
water treated with Oxone poses a problem,  it can be economically
and simply neutralized by the addition of  lime after treatment
and before discharge.

Design Parameters of Full-Scale UV-Oxidant System

     If pink water is to be treated effectively it must be fil-
tered to remove suspended solids before irradiation.  The sep-
arated explosive crystals could be reclaimed, if desired.

     After eliminating the suspended solids, the liquid is di-
luted, if_necessary, and pumped into a mixing tank where me-
tered^additions of hydrogen peroxide or Oxone are made.  To be
certain that a homogeneous solution is prepared, the pink water
and oxidant are further mixed by an in-line triblender after
which the mixture is pumped into the UV-system for treatment.
Figure 5 illustrates the general design of the system.  After
treatment,  the water can be discharged directly into a sewer
line or, if H202 was used as the oxidizer, the water can be re-
cycled for dilution of additional pink water.  Recycling of the
treated water originally containing Oxone is not allowable
without further treatment since residual amounts of potassium
and sulfate ions in the treated water would increase with each
fresh addition of Oxone.
                             27

-------
M
00
          PINK
        WATER
           \
                        FRESH WATER            RECYCLED WATER
                      V(FOR DILUTION)      OR    (FOR DILUTION)
            CONCRETE
              SUMP
OXIDANT
HOLDING
 TANK
                                                                TRI-
                                                              BLENDER
               DISCHARGED
                 WATER
                    i ,
                                                                                 VV
                                                                             IRRADIATION
                PUMP
                              FIG 5. BASIC DESIGN OF THE UV-OXIDANT SYSTEM

-------
                      REFERENCES

J. Patterson, N. I. Shapire, J. Brown, W. Duckert, and J.
Poison,  "State-of-the-Art: Military Explosives and Propel-
lants Production," Environmental Protection Agency (E.P.A.)
Report 600/2-76-2/3 a, b, c, October 1976.

C. C. Andrews and J. L. Osmon, "The Effects of Ultraviolet
Light on TNT and Other Explosives in Aqueous Solution,"
WQEC/C 77-32, Naval Weapon Support Center, Crane, Indiana,
JL _/ / / »

E. Zuesh, G. and J. Sherma, "CRC Handbook of Chromatography,
Solvent 3," CRC Press, Cleveland, 1972, p. 445.

E. Merck, "E. M. Reagents: Dying Reagents for Thinlayer and
Paper Chromatography," Darmstadt, W. Germany, 1975, p. 37,
p. 83.

Ed Hais, I. M. and K. Macek, "Paper Chromatography,"  Aca-
demic Press, New York, 1963, p. 637.

Standard Methods'• for the Examination of Water and Waste-
water, 14th edition, American Public Health Association,
Washington, DC,  1976, 493-495.
                          29

-------
                                  TECHNICAL REPORT DATA     .
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.
   EPA-600/2-79-129
                                                          3. RECIPIENT'S ACCESSION
4. TITLE AND SUBTITLE
  Evaluation  of the Ultraviolet-Ozone and Ultraviolet-
  Oxidant Treatment of Pink Water
             REPORT DATE
              July 1979 issuing date
            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
  Milton Roth
  Joseph M.  Murphy, Jr..
                                                          8. PERFORMS
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  U.S. Army  Armament Research and Development  Command
  Large  Caliber Weapon Systems Laboratory
  Dover, New Jersey  07801
            10. PROGRAM ELE
              1BB610
            11. CONTRACT/GRANT NO.

              IAG-D6-0059
12. SPONSORING AGENCY NAME AND ADDRESS
   Industrial Environmental Research Laboratory-Cinti,
   Office of Research and Development
   U.S.  Environmental Protection Agency
   Cincinnati, Ohio  ^5268                       	
            13. TYPE OF REPORT AND PERIOD COVERED
              Final  6/6/76—12/30/77	
            14. SPONSORING AGENCY CODE

              EPA/600/12
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT  pinjj water, a  solution of trinitrotoluene  (TNT)  and other nitrobodies, is a
  major pollutant at ammunition plants engaged either  in the manufacture of TNT or in
  the loading, assembly, and packing of bombs and shells.  As  an alternative to carbon
  adsorption for the treatment  of pink water, two new  methods  are reported here, the
  first involving the use  of ultraviolet (uv)-ozone and the  second using a uv-oxidant
  combination.                                            _
       In the first study,  a 3.79 cubic meters per day (nHpd), equivalent to 1000 gal-
  lons per day (gpd), uv-ozone  pilot system was evaluated.   It was found that dissolved
  TNT and RDX were reduced to less than one milligram  per liter (mg/L) with no by-
  products requiring disposal.   A design for a 19 m3pd (5000 gpd) pilot plant was pro-
  posed.
       In the second study, a uv-oxidant process for treatment of pink water was ex-
  amined.  Commercially available uv water purification units  in conjunction with an
  oxidant such as hydrogen peroxide or Oxone, (a blend of potassium persulfate ox-
  idants), were evaluated  for their efficiency in treating  pink water.  Variables such
  as film depth, dilution, uv wavelength, and operation of  the units in series rather
  than in parallel were examined.  A design  for a 379  m3pd  (100,000 gpd) pink-water
  treatment plant was proposed.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                                                                           COS AT I Field/Group
   Ordanance  Disposal Tools, HMX, RDX,
   Trinitrotoluene, Oxidation, Ultraviolet
   Radiation, Ozone, Demilitarization
 Loading, Assembling
 and Packing Plants
13B
 18. DISTRIBUTION STATEMENT

    Release to Public
19. SECURITY CLASS (Tl
  Unclassified
  38
20. SECURITY CLASS (Thispage)
  Unclassified	
                           22. PRICE
 EPA Form 2220-1 (9-73)
                                             30
                     * U.S. GOVERNMENT PRINTING OFFICE 1979  657-060/5360

-------