use on medical patients - apps.dtic.mil · notices this final report was submitted by battelle...
TRANSCRIPT
Report USAFSAM-TR-83-39AD-A140 570
ENCAPSULATED DECON FOR USE ONMEDICAL PATIENTS
Jim P. Pfau, Ph.D.
Ben F. BentonDavid L. %•"0 ","nr Ph nl
I" RoIr E. Sharpe, B.S.Gordon E. Pickett, B.S., M.B.A.
4 Battelle Columbus Laboratories505 King AvenueColumbus, Ohio 43201
DTIC
December 1983 PR27I984
Final Report for Period June 1982 - June 1983
"• i Approved for public release; distribution unlimited.
"-Lu Prepared for
SE USAF SCHOOL OF AEROSPACE MEDICINE
Aerospace Medical Division (AFSC)SBrooks Air Forcc- Base, Texas 782-35
1 '84 04 21913 015
NOTICES
This final report was submitted by Battelle Columbus Laboratories, 505King Avenue, Columbus, Ohio 43201, under contract F33615-82-K-0612, job order2729-02-06, with the USAF School of Aerospace Medicine, Aerospace MedicalDivision, AFSC, Brooks Air Force Base, Texas. 'asu Tai Chen (USAFSAM/VNC)was the Laboratory Project Scientist-in-Charge.
When Government drawings, specifications, or other data are used for anypurpose other than in connection with a definitely Government-relatedprocurement, the United States Government incurs no responsibility or anyobligation whatsoever. The fact that the Government may have formulated orin any way supplied the said drawings, specifications, or other data, is notto be regarded by implication, or otherwise in any manner construed, aslicensing the holder, or any other person or corporation; or as conveying anyrights or oermission to manufacture, use, or sell any patented ir.vention thatmay in any way be related thereto.
The Office of Public Affairs has reviewed this report, and it isreleasable to the National Technical Information Service, where it will beavailable to the general public, including foreign nationals.
This report has been reviewed and is approved for publication.
-7
YASU TAI CHEN, M.S. .D.Project Scientist Supervisor
ROYCE MOSER, Jr.Colonel, USAF, MCCommander
___
UNCLASSIF!EDSECURITY CLASSII 'CATION OF THIS PAGE
REPORT DOCUMENTATION PAGEla REPORT SECURITY CLASSIFICATION lb. RESTRICTIVE MARKINGS
UNCLASSIFIEDI2a. SECURITY CLASSIFICATION AUTHORITY 3. DISTRIBUTION/AVAILABILITY OF REPORT
Approved for public release; distribution2b. OECLASSIFICATION/OOWNGRAOING SCHEDULE unl imi ted.
4. PERFORMING ORGANIZATION REPORT NUMBER(S) 5. MONITORING ORGANIZATION REPORT NUMBER(S)
USAFSAf1-TR-83-39
6,. NAME OF PERFORMING ORGANIZATION Lb. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION
Battelle Columbus Laboratories (If applicable) USAF School of Aerospace k!edicine (VNC)
bc. ADDRESS fCily. State and ZIP Code) 7b. ADDRESS (City. State and ZIP Code)
505 King Avenue Aerospace Medical Division (AFSC)Columbus, Ohio 43201 Brooks Air Force Base, Texas 73235
go. NAME OF FUNDING/SPONSORING leb. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION I(If applicable)__ R___ANI _ ____T____NJ_ (I __ F3361 5-82-K-061 2
Sc. ADDRESS ICity. State and ZIP Code) 10. SOURCE OF FUNDING NOS.
PROGRAM PROJECT TASK WORK UNITELEMENT NO. NO. NO. NO.
62202F 2729 02 0611. TITLE ltncudde Security Claetuication)
ENCAPSULATED DECON FOR USE ON MEDICAL PATIENTS12. PERSONALAUTHORIS] Pfau, Jim P.; Benton, Ben F.; Gardner, David L.; Sharpe, Robert E.; andPirkptt- (nrdnn F_Final re~ort FROM JUn 1982 ToJun 198 1983 Deceimber l1lPAECON
16. SUPPLEMENTARY NOTATION
17. l COSATI CODES 18. SUBJECT TERMS (Continue on reuerle if neceusary and identify by block number)
FIEýO GROUP SUB. GR. Chemical warfare defense; skin decontamination; detection;15,/ 02 encapsulated decontaminants; -0eCl encapsulated; fluidized
libed encapsulation; phase separation encapsulation;,STRACT rContinue on nrucre if neceuary and identify by block numbero
Encapsulated -OCI- materials (bleaches) and chemical warfare agent (CWA) indicators have beendeveloped and evaluated For use as a skin decontamination material for wounded and incapaci-tated personnel. More than 40 capsule samples were prepared, and the twelve best wereevaluated using live agents. The -OCl" materials selected as capsule core were calciumhypochlorite and sodium dichloroisocyanurate aihydrate. Cellulose acetate butyrate (CAB)
' and chlorinated ruboer were the candidate polymers that worked best as capsule wall materialsBoth -OCl- materials effectively neutralized HD, but neither fully neutralized GB nor GD.Dibromo-dimethyl-hydantoin (DBDMH) was found to be a useful indicator for HD as well aseffective decon for HD. Luminol and indole were studied as indicators for GB and GDrespectively. Those showed some promise but will require considerably more development,Three capsule samples were evaluated with each agent in pig skin tests. Wlth HD, capsulessignificantly reduced agent penetration. With GB, one capsule system prevented any furtherpenetration of the skin sample oy agent. With GD, capsules appeared not only to prevent....[
20. OISTRISUTION/AVAILABILITY OF ABSTRACT 1 21. ABSTRACT SECURITY CLS4ICATiON
UNCLASSIFIEO/UNLIMITED X SAME AS RPT. CO OTICU U rse C3 UNCLASSIFIED22s, NAME OF RESPONSIBLE INDIVIDUAL 2ib• TIEIPHONE NUMRER Ric, OPPM11" 2110.
fl11dMidt AI". COO*)Yasu Tal Chen, M.S. (512) 536-2921 USAFSM/YVKC
DD FORM 1473, 83 APR IDITION OF I JAN 73 18 0120,E1T, VU 5LA1FIM
SECURKITY CLAMSF iCATI'ON CI; 'hIS Pa,(I,
UNLLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE
18. SUBJECT TERMIS (Continued)
chemiluminescent metnod; fluorometric method; magnetic capsule; skin penetration testing;bleach materials; CIWIA absorption-desorption; indicator materiais; indicator encapsulation.
19. ABSTRACT (Continued)
further penetration of agent, but also actually to absorb/extract agent. However, no one
single capsule was effective against all chemical agents.MI
I
I
IDII
-j I
i;Ia:.L}, " C A
ii, Ia,1t1t e
UNCLASSIFPEU
SECURITY CLASSIFICATION OF THIS PAGE
EXECUTIVE SUMMARY
This program has been established to develop and evaluate encapsu-lated -OCI materials (bleaches) and chemical warfare agent (CWA) indi-cators for use as a decon on wounded and incapacitated personnel. An en-capsulated decon should help decontaminate CWA exposed personnel and pro-tect attending medical personnel without contaminating either with poten-tially hazardous decons. Encapsulated indicators would provide positiveidentification of CWA contaminated sites for treatment or special care.
Development of effective decon microcapsules was based on a series oftasks performed on this study. The preliminary tasks included a litera-ture search and review, polymer-agent absorption/permeation studies,materials compatibility and interaction studies, indicator studies, se-lection of materials for encapsulation, microencapsulation of -OCImaterials and of indicators, and preliminary evaluations of selectedcapsules.
This investigation culminated with evaluating selected microcapsuleson pig skin samples, with HD, GB, arid GD. Results appear encouraging.The best capsule performance observed with HD showed considerable decreasein skin penetration by agent. The best capsule performance observed withGD not only showed no further penetration of agent after capsules wereapp!ied, but also appeared to have absorbed a portion of the agent thathad already penetrated before capsules were applied.
Dibroino-dimethyl-hydantoin (DBDMH) has been identified and undergonepreliminary evaluation as an indicator for HD. Capsules containingCa(OCI), or sodium dichloroisocyanurate dihydrate as core material andchlorin',-d rubber as capsule wall material were modified by adding asecond wail coatina of a blend of chlorinated rubber and DBDMH. Thesecapsules provide both a color change and decon function when contactedwith HD. indicator systems investigated for GB and GD are not completelysatisfactory altnough results show some promise.
Leach test results show that some capsules are also resistant toimmersion in water for up to I hr or more, suggesting that there should belittle release of decon material into body fluids in short-term contact.in addition, a brief study showed magnetite can be incorporated into thecapsule wall to provide magnetic microcapsules that can be retrieved by amagnet. This capability may provide a method of recovery of capsulesafter decon acti.,ity is completed.
However, further development and optimization work is recommended.While the oreceding capabilities have been individually d2morstrated byvarious capsules, the desirable qualities/functions have not yet beencombined into a single effective formulation. Incorporation of these into
a preferred embodiment of the concept will require further development.Further study is also needed to identify and develop effective indicatorsfor GB and GD and to improve overall decon effectiveness with higherlevels of GB/GD destruction. Many results reported are based on a limitednumber of samples and/or analyses. Replication of such evaluations willbe required to provide fully reliable quaititative results and to deter-mine the range of product performance variability.
TABILE OF CONTENTS
P aq, e
INTRODUCTION .......... ....... ............................... 7
OBJECTIVE .......... ................................... . 7
WORK PLAN ......... ..... ... ................................ 7
SUMMARY ........ ..... ..... ................................. 8
RECOMMENDATIONS FOR FUTURE DEVELOPMENT ........ .................. 9
EXPERIMENTAL DETAILS .......... .......................... ... 12
Task 1. Literature Search anu Reiiew ........ .................. 12Task 2a. Polymer-CWA Compatibility Studies. ..... ............... 14
Permeability of Coating Films to Chemical Agents ....... .......... 14Solubility Parameters ....... ......................... ... 15Film: Agent Absorption-Desorption Studies .................. .... 15
Task 2b. Decon Coipatibility Studies ..... .................. ... 21Solvent -OCI- Comp&tibility ........ ....................... 21Bleach-Polymer Solution Compatibility Studies ...... ............. 26Decon-CWA (HD) Reactions .......... ....................... 26Decon-CWA (GD and GB) Reactions ..... .................... ... 29Further Study of Decon-CWA Reactions .... ................. .... 32
Preliminary indicator Materials Studies ..... ................. ... 35Indicator for HD Agent ....... ......................... ..... 35Indicator. for' G Agents ........... . ..................... 36
Selection of Materials for Encapsulation Stu4i t.s .............. ... 39Task 3a. Encapsulation Studies. ..................... 40
Organic 'hase Separation Encapsulation Studies ... ........... .... 40Fluidized Bed Encapsulation Scudies ....... .................. 41Encapsulation by Spray Drying ........................ a8
Indicator Development and Encapsulation Studies ..... .......... 48Indicator Materials Solubilities ........ ................... 48Preparation of Indicdtor Prills ........ .................... 50Evaluation of Luminol Prills .......... ..................... 50Evaluation of Indole Prills ........ ...................... ... 52Indicator Encapsulation Study ........ ..................... 57
Task 3b. Capsule Evaluations ....... ...................... ... 58Evaluation of Indicator Systems with Neat Agent ................. 58Evaluation of Encapsulated Oecons ........................ ... 60
Test Procedure ............................................ 60HD-Microcapsuie Test Results .................... 62Test Results of Microcapsules with Neat GB and GO ............... 65Capsule Leach Tests. ......................... 74Skin Decon Experimentation Using MicroencapsulatedDecontaminants .......... ... ........................... 79
3PREVIOUS PAGE• IS BLANK
TABLE OF COITFNTS ',Continued)
APPENDIX A. PATENI REI-LRENCE LISTS A AND B .................... .... 85
APPENDIX B. CHROMATOGRAMS FUR SKIN TEST STUDY ..... ............ 93
LIST OF FIGURES
Figur.: 1. Absorbence of HD by .?olymer Films . . . .......... 23Figure 2. Absoibence of GB by Polymer Films ...... .............. 24Figure S. Absorbence of GD by Polymer Films ..... ............. ... 25Figure 4. Schematic for Development of an Indicator Capsule System. . 38Figure 5. Schematic of 38388-5-3 Multi-Wall Inaicator-Decon
Capsule ...... ....... ........................... ... 47FicQure 6 DiffLsion Chamber ...... ... ..................... .... 80Figure B-i. Nondosed Skin Control ................ ......... ....... 94cigure B-2. HD Sk>, Ccntro!, 5-Min Exposure. . . ............ .95
Figure B-3. HD Skin Control, 35-Min Exposure ...... ... .............. 96Figure B-4. HD with 0. Z•38 g 38388-7-2 ...... ................. ... 97Figure B-5. HD with 0 4425 a 38388-5-2. ...... ... .................. 98Figure 3-6. HD with 0.400, 9 38388-18-2 ...... ... ................. 99Fiqur? B-7. GF SKin Con.rol, 5--Min Exposure ........ ............... 100Figure B-8. GB Ski n Control, 35-Min Exposure. . . . . ..... ...... .... 101• gure B-9. GB with 0.3110 q 38388-18-2 ........ ........ 102Figure B-1). GB with 0.3083 g 379'14-'43 ........ .................. ... 103Figure B-11. GB with 0.3021 g 37984-48 ........ .................. ... 104Fgure B-12. GD Skin Contro,, 5-Min Exoosure .......... .............. I05
i igure B-13. GD Skin Control, 35-Min Exposure ....... ............... 106Figure B-14. GD wi th 0.3183 g 38388-18-2e. ....................... ... 107Figure B-15. GD wij, 0.3192 g 379,04-46 . ................ 108Figure B-16. C- f,. 0.3054 g 37984-48 ........ .................. ... 109
LIST OF TABLES
lab](, 1. ',r.r i jq 1,.,LS for Perq e hbi Ii ty of I llriby Chemical Agents .......... ........................ ... 16
Table 2a. Puliished Solubility Parameter information forD0lymers ....... ........... ............................ 17
-ible 2b. Solubility ParamEter infurmation on Solvents and
Agenr. . ...... ................................. ... 18Toble 2c. Weight Chan-le of Polymer Fiim Samples Exposed to CWA Vapor. . . 22Tabip 3. Sunmnary, of 3 eacv.-SolIent Compatibility .............. .... 27|bir- 4. Sunin-nr, of Bleach/Polymer-Solution Compatibility ......... ... 28lable 5. Reactivity of Candidate Decon Materials with HD ........... ... 30Table 6. Rea:tivity of Candidate Decon Materials with GD ........... ... 31Table 7. R,'.a. vity of Candidate Decon Materials with GD and GB.....<= " " • • • 33Table 8. Reac tvity of Candidate Decon Materi•ls witn (MWA ........ ..... 34
4
LIST CF TABLES (Continued',
Page
Table 9. Encapsulation Runs: AU 56 .. .. .. ... ... ..... ........... 42Table 10. Encapsulation Runs: Olin HTH. .. .. ... ... ..... ... .......... 43Table 11. Encapsulation Runs: Calcium Hypochlorite (Reagent). .. .. ...... 44Table 12. Fluidized Bed Encapsulation Runs: Capsule Composition. .. ......46Table 13. Solubility of Luminiol and Indole in Various Media a,,
Room Temperatures .. .. ... ..... ... ..... ..... ... .......... 49Table 14. Indole Solubility in Varying Concentration of GlycerolTaland Ethylene Glvycol Solutions .. ... :.. ..... ..... ........51
Tabe 1. esut- ofLumno PrllLeaching Studies .. .. ... ... ........ 53Table 16. Res I ts of Indole Prill Leaching Studie ... .. ... ..... ......55idble 17. indicatior Reactions: Neat CWA's vs. SiMUidntS. .. .. .. ....... 59Table 13. Description of Capsules Selected for CWA resting. .. .. .. ..Table 19. HD-MicroCdpsule Evaluations: Comparison of Wall Materials:
Fluidized Bed Coated ACL56 Particles (300 to 595 i'm) with10%1 Wall Material .. ......... ... ..... ..... ........ 63
Table 20. HD-Micr-ocapsule Evaluations: Comparison of Fluidized BedCoated ACL56 and Ca( OCI) 2, Particles (300 to 595 pm)with Parlon6 and Par lon'ý -DBDMH Wall Materials.........64
Table 21. HD-Microcapsule Evaluations: Ca(OCl) Partic 'les Coatedwith CAB 381-20 (25%) by Phase Separai ion. Comparisonof Capsule Sizes. .. ..................... 66
Table 22. HD-Microcapsule Evaluations: Comparison of 10% and 25%CAB 381-20 Walls on ACL56 Particles (.300 to 595 l±m) .. .. ...... 67
Table 23. HD-11icrocapsuie Evaluations: Ca(OCl)., Particles Encapsulatedin Parlon,' S125 by Spray Drying Comip~red with Other SelectedVicrocapsules.............. . .... .. ... .. .. .. .. .. ....
Table 24. CWA-Microcapsuie Evaluations, Comparison of Wall Materials:Fluidiz.,d Bed Coated ACL56 Particles (300 to 595 ýIm) with 10%Wall Material .. .. ... ..... ... ..... ......................*
Table 25. CWA-Microcapsuile Evaluations: Comparison of Fluidized Bed CoatedACL-56 and Ca(0CI) 2 Particles (300 to 595 vim) with Parlon"' and 7Parlon®&-DBDMH Wall Materials....... ... .. .. .. .. .. .. .. .. 7
Table 26. CWA-Microcapsule Evaluations: Ca(OCl) Particles Coatedwith CAB 38l-20 (25%)0 by Phase Separation. Comparisonof Capsule Sizes. ................... **......... ....... 72
Table 27. CWA-Microcapsule Evaluations: Comparison of 10% and 25% 7CAB 381-20 Walls on ACL 56 Particles (300 to 595 iim).......7
F ~ Table 28. CWA-Microcapsule Evaluations: Ca(OCl) 2ParticlesEncapsulated in ParlonO S125 by Spray Drying Compared 7with CAB 381-20 by Phase Separation.............7
Table 29. Compar3t/ve Capsule L-each Rate Evaluation .. .. .. .. ... ...... 77Table 30. Effect of Wail Thickness (% Coating) on Aqueous
Leach Test Results. .. ................... 78Table 31. Control and Decontamination A*.ia~lytica~l Data' For Pigskin 8
Tests.............. . . ... . . ... .. ... .. .. .. .. .. ...... 8
5
ENCAPSULATED DECON FORUSE ON MEDICAL PATIENTS
INTRODUCTION
ADplication of chemical warfare agent decontaminants (DECON) towounded and/or incapacitated personnel helps to prevent further damage tothe patient from the agent; however, many of the more effective DECONmateriils are strong reactive chemicals that can cause severe tissue dam-age when applied tc skin or open wounds. Encapsulation of DECON withmembranes of suitaolp polymers would solve many problems associated withits use. Capsules designed to release DECON in the presence of organo-phosphorous or organosulfur compounds when a certain amount of agent hasbeen absorbed would eliminate unnecessary exposure to corrosive, oxidizingDECON for both patients and medical personnel. Incorporating an agentreactive/sensitiýve dye would aid in the identification of agent contami-nation and provide for continuous availability of DECON that had not yetbeen in contact with an ýcent.
OBJECTIVE
The objective of this program, as stated in RFP F33615-82-K-0612, isthe development of a microencapsulated hypocrilorite-type chemical warfareagent decontaminant for use on medical patients. The microencapsulatedchemical agent decontarmJinant is to contain an agent reactive/sensitive dyeto provide a visual indication of the presence of a chemical agent. Ap-plication of the encapsulated DECON to medical patients will help preventfurther damage to the patient arid provide rrotection to medical personnelattending to them,.
WORK PLAN
The work plan, proposed by 3attelle Columbus Laboratories (BCL) inresponse to the subject RFP, was composed of the following tasks:
lask 1. (RFP Task number A.1) Literature search. Variousliterature sources were accessed and reviewed toprovide data on -OCY- source materials, polymer-OCI reactions and stability, polymer absorbencyfor agents, encapsulation of -OCl materials(bleaches) and agent sensitive dyes/indicators.
7 PREVIOUS PAGE
Task 2a. (RFP Task number 4.2.1) Absorbency/permeation study.Investigate at least 12 polymers and "waxes" (potentialcandidate capsule wall materials) and their interactionwith agents GB, GD, and HD.
Task 2b. (RFP Task number 4.2.2) Materials compatibilitiesstudies. Study compatibility and interactions of-OCI materils, pnolymprnr solvents, andagent-indicators. Evaluate nonencapsulated -OClmaterials for effectiveness in deconning GB, GD, andHD.
Milestone A. (RFP Task number 4.3) Based on preceding tasks, selectmaterials for use in subsequent encapsulation studies,.
Task 3a. (RFP Task number 4.4) Encapsulation studies. Usingthe selected -OCI- materials, polymers, solvents,and indicators, investigate at least two encapsula-tion processes for development of decon microcapsules.
Task 3b. (RFP Task number 4.4) Capsule evaluation. Evaluateselected microcapsules with GB, GD, and HD directly andwith skin tests.
Task 4. (RFP Task number 4.4.1 and 5.1) Final report. Reviewproject results, identify best capsule system(s), andprovide recommendations for further development.
SUMMARY
Results appear very promising. Of the more than 40 capsule samplesprepared, twelve were evaluated with agent (Tables 18-28). Several showgood decon effectiveness (up to 97% with HD, 80% with GB, and 75% withGO). Three capsule samples were evaluated with each agent in pig skintests. Cadaver skin was not readily available so pig skin was substi-tuted. Previous studies elsewhere have indicated that penetration of awide number of chemicals through pig skin is similar to penetrationthrough human skin. in this test, agent was placed on a prepared skinsample. Capsules were applied 5 min after agent was added and allowed toremain for 30 min. Capsules and unabsorbed CWA were washed off the sur-face of each skin sample into a beaker using about 10 ml hexane appliedfrom a wash bottle. This hexane was immediately decanted into a volu-metric flask and diluted to 50 ml with more hexane. The skin samples werethen processed and analyzed to determine residual (penetrated) agent.Controls with no capsules applied were similarly washed and analyzed.Capsules and hexane wash liquors were also analyzed.
With HD, capsules significantly reduced agent penetration. With GB,one capsule system prevented any further penetration of the skin sample byagent after capsules were added. With GD, capsules appear not only to
8
prevent further penetration of agent, but also actually to absorb/extract
agent already penetrated into skin samples at the time capsules were ap-plied.
Polymers selected for use as capsule wall materials were celluloseacetate butyrate (CAB), chlorinated rubber, polyvinyl butyral, and poly-vinylidine copolymer. The -OCI- materials selected as capsule coreswere calcium hypochlorite and sodium dichloroisocyanurate dihydrate. The
9 CAB and chlorinated rubber show greatest promise of the four polymersevaluated, Both -OCI- materials effectively decon HD, but neitherfully decon GB nor GD.
Dibromo-dimethyl-hydantoin was found to be a useful indicator for HD,as well as effective decon (only for HD). In contact with HD, a brightyellow or orange color is formed. Luminol and indole were studied asindicators for GB and GD. These show some promise but will require con-siderably more development.
Capsules used in tests with agent were also evaluated in a waterleaching test. The CAB and polyvinyl butyral, contrary to expectations,performed better than chlorinated rubber or polyvinylidine chloride co-polymer. The best samples showed only about 1% of the core material
Sleached after 1 hr in water. The leach test performan-e may be important~ to indicate whether capsule core materials will leach in the presence ofSI body fluids.
Two capsule samples incorporated magnetic iron oxide in the capsule! wall to demonstrate that magnetic capsules could be prepared. These cap-
sules were successfully sequestered in aqueous dispersion by a small barmagnet. No attempt was made to evaluate removal from whole blood or otherbody fluids. Further development is needed, but results suggest thatmagnetic based recovery might be considered for removal of spent capsulesfrom open wounds.
No one single capsule showed best results in all tests. Further workis needed to improve overall performance and bring together into one cap-sule system the various best performances in the different tests. Furtherimprovement in skin test decon performance is desirable, and can probablybe developed. Further development is needed for GB and GD indicators, aswell.
RECOMMENDATIONS FOR FUTURE DEVELOPMENT
Results of this feasibility study appear quite promising, but con-siderable further development will be required to uevelop a fully func-tional microencapsulated decon and indicator system which is completelyeffective against all three CWAs studied (HD, GD, and GB). The idealmicrocapsule system is perceived as having these characteristics:
9
* provides 100% effective decon of HO, GB, and GD absorbed into
capsule
* rapidly absorbs agent into capsule wall
* prevents any further agent penetration into skin as soonas capsules are applied to contaminated areas
R extracts/absorbs agent already penetrated into skin
s changes color on contact with agent to provide visual indicator ofcontaminated sites
o has good storage stability under standard storage conditions
s does not allow release/migration of absorbed agent, decon or othercapsule contents out of micrecapsules after application to con-taminated skin or open wounds
, does not allow leaching of capsule contents by and into contactedbody fluids (e.g., blood, perspiration)
a is safe to handle and easy to apply
* is readily removed from skin after agent has been absorbed
* is amenable to safe and easy disposal
s suitable for use in open wounds
* preferably all the preceding characteristics are demonstratedby each and all of the microcapsules in the system.
To develop a product approaching this ideal, several improvements andadditional study will be necessary. A follow-on prcram should include
W• the following tasks:
* identify an indicator or indicator system which is as effectivewith GB and GD as the DBDMH appears to be with HD.
* develop methods to incorporate these indicators into themicroencapsulated decon system and product.
* modify capsule wall composition to further increase absorbenceof CWA, minimize agent penetration into skin, and, if possible,extract and absorb agent already penetrated.
9* increase decon effectiveness. Several -OCl materials show deconeffectiveness for HD of 90% to 100% but only up to 70% to 80% for
10
GB and GD. A formulated capsule core composed of a mixture ofdecons may be necessary to provide thorough decontamination of GB,GD, and HD by each microcapsule. A composite core in all capsulesis probably more desirable than a blend of capsules, since thelatter would probably not completely decon contaminated skin.Another option to be studied is to incorporate one decon in thecapsule wall and others in the capsule core, especially ifmaterials are incompatible in processing, storage or application/use. Composite cores may be prepared by blending and agglomera-tion fine powders, prepared from hot melt, by prilling, or formedin layers by application cf encapsulation processes.
* investigate other capsule wall materials to increase CWA absor-bence. This effort should include additional polymers, blends ofpolymers, addition of plasticizers and/or absorbent inorganicmaterials which themselves do not form good capsule walls. Suchmaterials may be incorporated into the capsule wall during encap-sulation or be applied as a separate surface layer after encapsu-lation.
o incorporate magnetic materials, such as magnetite, in capsule walland/or capsule core to provide magnetic capsules. Such capsulesshould be easier to locate and remove/recover from open wounds andbody fluids. This magnetic material must be selected and incor-porated in ways that do not reduce capsule absorbency, deconeffectiveness, indicator effectiveness, or storage stability.
* incorporate materials and processes providing best performances inpreceding functions into a single microcapsule system.
* increase and extend evaluations. Key evaluations, such as pigskin tests, should be run in, at least, triplicate. This shouldbe followed by human cadaver skin tests and then by testingwith suitable animal models such as pigs or perhaps athymic("nude") mice with human skin grafts. Storage stability ofcapsules should be determined, quantitatively, under avariety of selected conditions.
A study based on these recommendations should lead to the development ofan improved encapsulated decon which will more nearly approximate the"ideal product described earlier in this section. Promising results of thecurrent study are interpreted as indicating a good probability of devel-oping a successful product.
k 11i
-P~- _ ____B
EXPERIMENTAL DETAILS
The following sections describe details of the experimental work doneon this project.
Task 1. Literature Search and Review
With respect to -OC1- encapsulation and handling procedures,existing BCL files were reviewed, and a computer search was made of Chem-ical Abstracts Index (.1972-81), NTIS (1964-82), Engineering Index(1970-82), and the U.S. Patent Index (1976-82). The literature searchprovided 65 references, about half of which appeared to be relevant (atleast marginally) to encapsulation and/or handling of -OCI materials.Copies of relevant references, articles, and patents were obtained andreviewed.
The patent literature examined provided relatively little data di-rectly and immediately relevant to this study. However, Reference List Ain Appendix A is a listing of patents with some limited relevance to thisstudy. References A-1 through A-17 and A-20 relate to coating materialsand processes for encapsulation of bleach particles. The function ofthese coatings is to provide delayed release of bleach in aqueous mediaand/or provide longer stability of component materials in detergent form-ulations. Reference A-15 (U.S. Pat. No. 3,647,523) provides a brief listof chlorinated solvents suitable for use in processing -OCI- materi-als.
Most references to coatings for the bleach materials relate to fattyacids, fatty acid esters, alcohols, glycols and paraffin waxes as thecoating material although polystyrene and styrene acrylic copolymers arealso mentioned. However, since these coated -OCI- materials are in-tended for use as laundry and dishwashing detergent additives, the func-tions of the capsule wall, and hence the wall composition, are quite dif-ferent from capsule walls developed for this study.
Many of the patents involve inorganic coatings or one class of or-ganic coatings; long chain fatty acids (such as stearic acid). Some var-iations involve an initial coating (capsule wall) of fatty acid followedby a second coating of other materials, such as sodium stearate. Theselected fatty acids are waxy solids at ambient temperatures. They aresprayed as a hot melt onto particles of selected -OCl- materials suchas potassium dichloroisocyanurate. Various processing systems may beused, but major emphasis is on fluidized bed coating (encapsulation) pro-cesses. Coated particles described are generally 1-2 mm, which is larger
Z •than the preferred size(s) for this project. The coated particles areclaimed to remain stable for extended periods of storage, to release-OCl (as a bleach) in hot/warm water and to prevent "pinholing"(damage to fabrics caused by solid bleach particles in direct contact).
j 12
However, release of the -OC- (bleach) decreases with thicker coatingsof fatty acids and with the use of relatively high-melting fatty acids.
Reference A-18 ("Method of Coating Chloramine and Product Thereof")describes coating of chloramine with sodium stearate, using a chloroformsolution and pan coating techniques. Reference A-19 describes the coatingof soda lime particles with sodium permanganate and as being "useful forthe absorption of certain gases used in warfare and encountered in theindustries, such as arsine, phosgene and other chlorinated organic com-pounds."
Patent Reference List B (Appendix) is a listing of patents that werereviewed but judged to have very little or no relevance to this currentstudy. However, some materials and processes, relating to increasin; thestability and shelf-life of -OCI- materials, are described. These maybe more relevant to a more-detailed Phase II, follow-on study of encapsu-lated -OCI- decon.
Only a limited amount of literature that discusses indicators foragents was found and reviewed. These are (1) "Military Chemistry andChemical Agents" TM3-215/AFM 355-7, Department of the Army and the AirForce (December 1963), and (2) "Government Report Literature Survey on theDetection of Selected CW Agents", by R. E. Wyant (of Battelle) to NavalEOD Facility (February 5, 1979). Three additional references are cited infootnotes on pages 5? and 54.
Information on the interactions between polymeric materials andchemical warfare agents was reviewed, especially with respect to chemicalwarfare agent/polymer absorption and permeation. The prior work has beenconcerned mainly with absorption of agent as measured by weight change andchange in physical properties of exposed polymers. Sources for this in-formation were primarily compilations of previous work carried out at theU.S. Army's Chemical Systems Laboratory (Aberdeen Proving Ground, Mary-land) and some recent work at Battelle's Hazardous Materials Laboratory(HML). Literature covering polymer solubility parameter data was alsoreviewed. Available information on (1) sorption and solubility propertiesof polymers with -OC- and solvents and (2) conditions required forforming films of the polymers were considered in final selection of poly-mers for evaluation in Task 2a.
To establish handling and evaluation procedures for compatibilitystudies of -OCY- materials with solvents, previous proprietary work atBattelle Columbus Laboratories on encapsulation of bleaches was also re-
viewed. Relevant procedures were adapted for use in this program.
"Some additional literature was identified and reviewed during thecourse of work on Tasks 2 through 4. These references are cited in theappropriate parts of following sections of this report.
13
I________
Task 2a. Polymer-CWA Compatibility Studies
Polymer film compatibility with HD, GB, and GO was studied throughthree comparative evaluations: (1) film permeability to CWA, (2) solu-bility parameter and hydrogen bonding indexes of polymers compared tosimilar data for CWA, and (3) CWA vapor sorption into selected polymerfilms.
Permeability of Coating Films to-Chemical Agent--Ten coating resinswere identified (from past work) as being sorptive to agents and have beenscreened for stability to the candidate decons. Films of these resinmaterials (approximately 25 vm thick) were prepared by casting on glassfrom sclvent solutions, Solvent-cast films were used because they morenearly simulate coatinos formed Lv the rticroencipsulation processes whichwere used later in this project. Most films were separated fron the glass(as free films) rather easily. Three materials (Sarans, Elvaxq, and Vitone)were separated more easily if placed in a freezer for a short period oftime, prior to the actctnpted separations.
The initial screening tests were run in a permeation cell on whichthe film (about ý 30 cm circle) was supported by a stainless steelfrit and chernical agent drops were placed on the surface of the film. Itwas found that many of the films softened and threatened to contaminatethe frits.
Because of these problems, the screening ýtudy was continued byplacing smaller portions of film (about 4-5 cm ) on white paper. A1 , drop of agent was then put on the film and observed through a magni-fying lens. The tests with HD were conducted in this manner and when theHD permeated the film a "wet" spot appeared on the paper. However, whenthe films were lifted a "hole" was observed in some films and softenedresin was attached to the paper.
For the GD and GB agents two samples of film were then evaluated, onein contact with the paper, and one suspended over the paper. This modi-fied procedure was used to determine if the films were soluble in theagents and a hole was formed spontaneously, or if the hole was caused bymanual removal of the film (adhesion of the softened film to the paper).
As can be seen from the data presented in Table 1, seven of the filmswere permeated by the HD in less than 30 min. GD and GB are more volatileso permeation and evaporation were competing processes. in a number of
- cases the film was softened and the drop appeared to persist. Probing theIM .apparent drop, however, showed it to be a swollen or softened area of the
film. Only one film appeared to form a hole on contact with the agents.7 This film was based on Viton® and the agent was GB. Some films, such as
polycarbonate, polystyrene, and polycaprolactone, appeared to be "wet" by
14
,AR
the agent with the droplet spredding out. This spreading either dilutedthe soften*ig effect or speeded evaporation of the agent (or both). Datashowing t'ese results with the GB and GD agents and the candidate filmsare also presented in Table 1.
Tvj other materials had been considered as capsule wall candidates.These are polyethylene and stearic acid. However, they are not includedin tiese tests become of potential problems in film preparation and en-cap'*alation processes. Polyethylene is only very slightly soluble in afe' hot solvents, such as xylene or trichloroethylene. In past work atBactelle, solid particles of inert materials have been dispersed in aS'sated xylene solution of polyethylene, and coated by cooling the solutionander controlled conditions. However, adding particles of strong oxi-dizers, such as the -OCl- materials of interest here, was deemed toohazardous for further consideration.
The stearic acid was studied in the compatibility studies describedlater in this report, but this material is quite brittle and not easilyhandled as a thin film. Consequently, it too was dropped from furtherevaluations.
Solubility Parameters--The literature was examined for solubilityparameters for the polymers used in this study. Table 2a lists the solu-bility parameter ranges found. No data was found for polycaprolactonePCL700 or for Viton 0 G. Solubility parameters and hydrogen bonding indexfor solvents and agents used in this study are shown in Table 2b.
Comparison of solubility parameters and hydrogen bonding indexes forpolymers (Table 2a) with that for CWA (Table 2b) inr'i.ates that the cell-ulose acetate butyrate and polyvinylidine chloride copolymers should besorptive of and/or permeable to HD. Polystyrene, polycarbonate, andchlorinated rubber are also possible candidates (i.e., based on the 10.6solubility parameter of HO and 2.7 hydrogen bonding index falling withinthe solubility parameter range of t'ie polymer as shown in the "weakly"column of Tabie 2a). A similar comparison for GB and GD suggests chlori-nated rubber, polyvinyl butyral, and cellulose acetate butyrate are likelyto be sorptive of and/or permeable to the GB and GD agents.
Film: Agent Absorption-Desorption Studies--Technical problems were" encountered in using a Cahn Electrobalance to screen the CWA absorp-
tion/desorption properties of the candidate polymeric capsule wall mate-rials. The Cahn Electrobalance procedure involves exposing a polymersample to CWA vapor and measuring the weight of the sample again as afunction of time. Efforts were unsuccessful in generating a constant CWAvapor "challenge" concentration over the absorption time period (severalhours).
15
II C)
ICC
co 0. 4
r o -
'n I C C) ) 00LIn' 0V. Vn 100cm CA 0
t C) - C .. swI L4
0A 4.) Ep.., .,
42 00 C) - C C
VO) cm C) C'r)M.4 M.. 3- toJ 'a0 t C) 3-V
a) 0 m , C) 'A 0L mA 0. VL.C
CL 0LnV) V V n C ) :c. 'A 5~ C) 4 C
4UV I V S V VS V S -
0.eo C.,7 L ,0 USý
C) 'A- A - 1 3
>- IV I - ~ >S-. .43 - ) C00' CO. .0'. _ 3
C O4- 0 C C x)~
CL (A 'A C) CC
3.4. M VS. C) S-
V) LnC .ý.4 -m
a C)
U) 4 D
C E~C4
1 3 CD 4- >Cm) Ln C0 3- C- C S
I I - C) 3 -3
...4 4.v a) . C) .4. .4D C)C 5) s-
ii CL C) CL 3- 3- C) C) m )
CL CL C) C) 0. C) C) C) Cý 0
~~~.4 0. 0. 0 ) 0 . 0
.4a u) C V C 003- 3-S - 4.4 E
4.30 0 C I- C) C)0.0 . 0 C)
I~s - wC
Q 4 >4 3- C)-u
4.1 C) Q ) 0 V .- O
06 C) a,'0 0 - (D C
C~~~ ~ .- 0; 0 . '015) 4.
Ps.4 0I 0C6 -) C
16
TABLE 2a. PUBLISHED SOLUBILITY PARAMETER INFORMATIONFOR POLYMERS
Hydrogen Bonded Solventse
Polymer Name Weakly Moderately Strongly
Ethylene Vinyl Acetate (Elvax® 210) 7.3-9.8 0 0
SPolystyrene (Styrene® 6664) 8.5-10.6 9.3 0
Polycarbonate (Lexan ) 9.5-10.6 9.3-9.9 0
Chlorinated Rubber (Parlone S125) 8.5-10.6 7.8-10.8 0
Polyvinylidine Chloride Copolymer 9.5-11.1 10.8-14.7 0I "- (Saran® F220/F310)
Polyvinyl Butyral (Butvar® B76) 9.0-9.8 8.4-12.9 9.7-12.9m "7b
Cellulose Acetate Butyrate 11.1-12.7 8.5-14.7 12.7-14.5(CAB 381-20)
Polyvinyl Butyral (Butvar B73) 0 9.9-12.9 9.7-14.3
ia
Wa Weakly hydrogen bonded implies a hydrogen bonding index around 2.5,
moderately bonded is about 5.5, and strongly bonded is about 8.5.bBattelie's experience with CAB 381-20 indicates that this should be 9.3
to 12.7. For example, CH2 CI2 and CHC1 3 are solvents.
17
E_____
TABLE 2b. SOLUBILITY PARAMETER INFORMATION ON SOLVENTSAND AGENTS
Solvent/Agent Solubility Parameter Hydrogen Bonding Index a
Freon TF 7.2 2.5
n-Hexane 7.3 2.0
(yclohexane 8.2 2.2
T-luene 8.9 3.3
Di chloromethane 9.7 2.5
Ethyl acetate 9.1 5.2
Methyl ethyl ketone 9.3 5.4
Acetone 10.0 5.9
Ethanol 12.7 8.5
GG 9.0(9.10)b 5.4
GD (8.41) b c
HD 10.6 2.7
S: a According to DuPont system.
b The values in parentheses are calculated values. Others are experimentallydetermined values.
c Believed to be about 5.5.
J
L •18
MM- - -
In general, solubility and diffusivity of an organic penetrant (CWA)in a polymer are dependent on th'e concentrration of the penetrant in thepolymer. This penetrant concentration inr the polymer is, in turn, depen-dent on the challenge concentration. Tiilrefore, interpretation of ab-sorption data obtained with a varying challenge concentration is difficultand can be misleading. This difficuity is particularly pronounced whencomparing the orqanic liquid absorbing p.-operties of several polymerswhich is the main objective of the CWA/polymer absurptlon studies in thisprogram.
For these reasons, liquid CWA absorptioo/desorption experiments wereproposed and attempted instead of the Cahn Electroba'ance vapor absorptionstudies. Since the encapsulated decontaminants are being developed toabsorb liquid CWA, these liquid absorption/desorption experiments shouldprovide more valid "real world" results.
The following liquid CWA absorption/desorption procedures h".ve beensuccessfully used in other research programs and wee proposed for use inthis program:
(1) Test slabs (2.5 cm by 2.5 cm) are cut from each polymersample.
(2) An automatic pipette is used to contaminate each ofseveral (up to six) preweighed test slabs of a singlepolymer with 10 ý.l of CWA. The CWA is placed in stripsnear the center of each -ample and covered by a watchglass immediately after placement to reduce evaporationlosses.
(3) At selected intervals after CWA placement, the liquidR ,CWA remaining on the surface of a test sample is
blotted using filter paper and the sample is weighedto obtain the amount of absorbed CWA. The amount of
K •unabsorbed CWA is determined by extracting the filterpaper with chloroform and analyzing the extract by gaschromatography. The time periods between weighings area function of the amount of CWA absorbed by the firstsample weighed. The first sample is generally weighedabout 5 min after contamination. Also, after weighing,a test sample is extracted with chloroform, and the ex-tract is analyzed for CWA using gas chromatography. Theextraction supplements the absorption results obtained byweighing.
At the time the last sample is weighed in an absorption run, a con-taminated sample is prepared For desorption measurements. The CWA/polymer
1
M• 19
_W M A
contact time for the desorption sample corresponds to the contact time ofthe last absorption sample. The following procedures are then used fordesorption determinations:
(1) Unabsorbed CWA is removed using filter paper, and thetest sample is placed in a desorption test chamber.A I liter/min air stream is passed through the chamberfor a period of time (up to 170 min), and desorbed CWAis trapped for gas chromatocraphic analysis using chlo-roform bubblers. Several bubbler samples are taken atselected time intervals beginning with the initiation ofair flow.
(2) After the desorption test period, the test slab is ex-tracted with chloroform to determine the amount of CWAremaining in the polymer. Also, the filter paper usedin removing unabsorbed CWA is extracted with chloroform.The test slab and filter paper extract are analyzed forCWA using gas chromatography.
Attempts were made to measure the sorption of liquid HD, GD, and GBby films of the candidate polymeric capsule wall materials, following theexperimental procedure outlined above. Difficulties were encountered whenit was found that, in some instances, the liquid CWA softened and solu-bilized the film samp!,.s of the candidate polymers. For polymer sampleswhich became tacky and/or lost mechanical integrity, it was not possibleto blot off only the unabsorbed CWA prior to making weight gain measure-ments.
As a result, laboratory efforts were again directed toward Ch''A vaporsorption. Vapor sorption provides the same information as liquid sorptionin regard to relative CWA affinity of the polymer candidates but does notrequire the removal of standing (unabsorbed) liquid CWA prior to weightmeasurements.
The same general procedure was employed for CWA vapor sorption mea-surements as proposed for liquid sorption measurements with the exceptionthat the film samples were exposed to CWA vapor and not to CWA liquid.Vapor exposure was accomplished by placing each film sample of a polymercandidate on a watch glass which was then placed in a closed containercontaining CWA vapor at saturation level. The CWA vapor was generated bya wick (absorbent paper) which contained approximately 100 ul of sorbedCWA. Since the film samples were separated from the wick by a watchglass, they did not contact liquid CWA. Each film sample was one squareinch in size. Exposure times evaluated vdried according to type ofpolymer and CWA. Test temperatures were all 21.00 to 21.70 C.
Samples of each candidate polymer were removed after selected ex-posure times and were weighed to determine the amount of sorbed CWA. To
20
7 ---- --•-
NIM
eliminate the potential problem of disturbing the CWA vapor concentrationeach time a film sample was removed for weighing, each film sample wascontained in a separate CWA vapor atmosphere (closed 'ontainer). Thisexposure to CWA and weighing steps were repeated with separate filmsamples for several exposure timhes.
The Cahn Electrobalance, however, could not be used for vapor sorp-tion measurements under the condition of stagnant (not flowing) saturatedCWA vapor. Such measurements would expose the balance mechanism of theCahn to the potentially corrosive CWA vapor. Consequently a suitablelaboratory balance was used for obtaining sample film weights before andafter CWA absorption.
Based on these measured weight changes, sorption data were calculatedas a percerf weight increase and plotted as a function of time (see Table2c and Figures 1, 2, and 9). As shown in the graphs, the CAB and Parlono$125 were more absorbent than the ButvarO B73 and SaranO F310, especiallywith HD and GD. For GB, the Parlon® $125 absorption results are closer toButvar® B73 than to the CAB. This data agrees well with capsule testresults which also show that CAB and Parlon"' S125 generally providedfaster and more complete sorption of agent than Saran"> F310 and Butvar®B73. This data is also consistent. with the solubility parameters andhydrogen bonding index data presented earlier in this report.
Task 2b. Decon Compatibility Studies
Compatibility and reactions between the --OCi" decon materials andselected solvents, polyrer solutions, and CWA were studied. Efforts andresults are discussed in the followino sections of this report.
Solvent -0C1- Compatibility--A procedure was established fordetermining the compati6ility of candidate -0Cl- materials withselected solvents, such as those commonly used in various encapsulationprocesses. The procedure used was:
* place 0.5 g of -OCI material in a small via! (fittedloosely with polyethylene cap, modified to allow insertionof thermocouple)
# add 4 mi of selected solvent and immediately cap and insertthermocouple
* note temperature change, gas evolution, color change, anddissolution for several minutes
21
F. .
0o 1 0 11 i
C\J C' )C'
0 010ý I D I C I In O10*i I * I . II I - -
C)) -= 0C-0
C)J C ~ 0C z t CM
-C cl : l ' )Ci C) ) Cz \Cý 0ý III CII O l Oil
Cd') III I1I *I oIC/) C
L- U'-
M:~ 00 IO ' 'C0 IC) C'ý
0 E=00 0 0 0
7:J C
C.) UC) Ca.II I I I-Lu C*L 4-
C.DO :::- C):: -: j
CD L) 00
U')US- C~
= V 4-) uu ~ r
-0 0) 0 CV.
C4. CD
i -n -CT m0 I I I j I Y I OI
IS-C
0 04)
C) C C> ) 3l
3: m0 3 M o m moo m c c mom CUL) CCD C U~3 (D CDC CD 3CD ( -i06
E
0
0-0C)4
CU CQ m c S-
S.. -Ea) 0X C0. > 0.
m 0 toSSi S- 4> 0. (a'
0 < o(a:
22
M.-I
ILL -> C
0 S0
~1-x .. 0.
-I-- TCj W
.1..
*
C1 4
e~j C V.. c~.i
23
BILL-- ro >
co S.- 4-- 0
a-~ V L) C.0 E
0i 0
I i ICI
CL 0'I! LC) U
A-)
MR--
C) C)C) C C:)C)0.koJ .0i
U~~> L 46
1~~ g .0s 2
C\J
(NJ cco
S- I-0' 0 ~(
Sco L4-
S~--0
+- S..a
g C) ) V)2.-k -o
I ii-<
I 0)
LL.
1 __ '-4.I ~ ~aC
C) co LO 1i l I) CD.-
ULP 446La
25 -
* remove thermocouple and special cap
* secure with standard friction-fit cap and allow to standfor 18 hr
9 note changes in -OCI- material and solvent.
Ten different "bleaches" (-OCf- materials) were evaluated forcompatibility with 9 different "solvents." These solvents were selectedfrom among a list of organic fluids commonly used in encapsulation pro-cesses. Resulting data are shown in Table 3 and provide a 90 point matrixof bleach-solvent compatibility evaluations. Only two combinations showedtemperature increases: lithium hypochlorite with acetone or with methylethyl ketone. Several bleaches showed no reactions in any of the solventstested. Cyclohexane showed no reaction with any of the bleaches tested.
This study did not, by itself, eliminate any of the bleaches fromfurther consideration. The study, however, identified some combinationsto be avoided. Actual selection of candidates for Task 3 (encapsulationstudies) was made after the bleach-agent (CWA) reaction studies werecompleted.
Bleach-Polymer Solution Compatibility Studies--Another series oftests were run to investigate the compatibility of the candidate bleachmaterials with solvent solutions of ten different polymers and stearicacid. This test was similar to the one for solvents, except 5% polymersolutions (mostly in CH2Cl ) were used with the bleaches insteadof the solvents alone. Sa~an® F 310 was dissolved in methyl ethyl ketone(MEK) because it is the best conventional solvent for that polymer. Ethylacetate is the preferred solvent for Viton® B and was used in these tests.
Polymers were selected for this test based mostly on two factors: (1)past effectiveness as encapsulating materials, and (2) absorbency for oneor more chemical agents (as indicated by available literature). Table 4summarizes observations from this compatibility study. Nearly half of thedata points in the table matrix are nro (no reaction observed). Compari-son of the data presented in these tables with results from the studies ofthe interactions of bleaches with CWA and of permeability of selectedpolymer films to CWA helped provide the basis for selecting candidates forthe encapsulation studies. The BCL study of CWA absorption (reportedunder Task 2a) was not used as a basis for polymer selection because ofdifficulties and delays in generating the data. This absorption study datawas not available until later in the project and hence was used as asupplement.
Decon-CWA (HD) Reactions--The ten candidate decon materials werescreened tor eTficiency Ti reacting with HD under relatively dry condi-"tions. A solid powdered decon material currently in use, STB, was
26
N_
I- 3 C X X
oZ C.
*1 0 0m
4100 0 0
.0.. .1 -L
Ci 0
0.0 m - .0 mý mi m* m M0.
4) - - .
0~~C CiCi 41 4
A 0 >1 0>1 0V 0> 0
0. 0. m
101 .2!2!
00 0 0- Ci
V m 0
Lo-c
0 0 00. co ca
a 0 -- 3 0 27
.. � --
I -
.4. - C.
a.. � a.- a
H'0 C .4.4 C� 4.4. .4.4
2. . s . . . 0.a a.- C.C C-a
H �C CC*
I, aIIii �..C.3 C .. 4. c.I � C aC CC n.n
II .j'¼� U...........4.4 C-C.4.�.-4 -o a 8.0.0
II aI C
fl.............. �
aa- I .�a '�C. C.* C-a
aC
2.1 2� 2. 2. �.2
C a C
a ��-==a�-�-e 0 CC a 0- Ca .4
I .* -
� 'Is.. a a La2. �a � I
I a C- III S..
a IaC�c a
a, I
I "*'� -�
I.I a
'V a COg.
g-a
3 a 0£ S - .� - .� �
I- S CCV4
C i S2 aIc - a 40 Cii I
a 0.�
35 �
� -
28'�a .
-4.-C - -
included in the screening evaluations. For screening purposes, the HD wasexposed to about 20 times as much decon material (on a molar basis). Thesolid decon materials were weighed into small centrifuge tubes and I Il(1.27 mg) of HD was placed on the solid particles. Care was taken toavoid getting any of the agent on the tube walls so that as much as pos-sible would be available for reaction. After selected exposure times(described below) chloroform was added to the tube and the contents mixedthoroughly. The mixture was then centrifuged and the liquid portion wassampled for analysis by gas chromatography. Any changes in the liquid orsolid residue were noted.
The results of the screening tests are presented in Table 5. It wasthought that the reaction of agent with solid decon materials might beslow, so the materials were first exposed for 60 min. Since all of thematerials reacted (all but two almost completely), a 5-min exposure timewas used for the second screening. -he results of the shorter exposurewere essentially the same as for the 1-hr exposures, except for the chlo-rinated trisodium phosphate. Substantial amounts of HD were found afterreaction with potassium dichloroisocyanurate and chlorinated trisodiumphosphate at both 5-min and 1-hr exposures. The small amounts of HD (lessthan 6%) found in three samples may result from problems in handling the
W. small quantities of reactants. Conversely, calcium hypochlorite andlithium hypochlorite exhibited extreme reactivity with HD.
The dibromo- and bromochloro-substituted hydantoins gave an intensered-orange color in contact with HD. This intense color may serve as anindicator for presence of agent. However, it apparently is nonspecific,in that at least some coloration was observed in some mixtures of thesematerials and solvents as presented earlier in this report.
Decon-CWA (GD and GB) Reactions--The candidate decon materials werescreened for effectiveness in reacting with GD and GB under relatively dryconditions. The procedure followed was the same as for HD. The screeningtests were run first for 5 min, and then selected materials were againchecked after a 60-min exposure. The results of the screening tests withGD are shown in Table 6.
The 5-min exposure tests with GD showed sodium dichloroisocyanuratedihydrate to be singularly highly effective. Calcium hypochlorite and STBalso had significant decon action on GD. However, the 60-min exposure didnot confirm the effectiveness of the sodium dichloroisocyanurate dihydratethat was observed in the 5-min test. Using the same test procedure withGB, unexpected low results were observed even with the control.
These tests must be considered only as screening evaluations since* only 20-40 mg of solid decon candidate were treated with 1 ýLl (about 1 mg)
of liquid agent. Also, in this test procedure, the HD appeared to be morereactive than the G-agents. This judgment is based on observations of the
29
w c0 CCw
0 0 0
-V
- U
-c, 00 0
.. c 0 0 C
w,03 0 0 00 C0
Q 0
zC 10 ICn 10Cý
C CD CD m3-. . 0
Cv 01' -T)
V.~ L
.0 . C 0 0
47 10 a- CD -1-Cl
il I2-o
p .~, I-C
z k1- UCUIC.u
1-CS.. 0~o~ a a a a a - a U
257.
1-0
tk ---
eaa
a; (a
>
00S-V
CZ CL(D U3 s-3L
x (1 S-c) z n ci -r L) %
4 -
0)3
'4a --
c0 C0- *.- S.
cu 0-rL4
M~ S- m.
0< 0 0 M ýa#
W S - 'A .=
C) LeA I) LeA o LA3
10 a 0L CL
u -
0 L
U- MM U) 0n LA LC-L ) C- U LA
31L
reactants, especially with respect to the rapid development of the yellow-orange color with HD and the bromo-substituted hydantoins, and only aslight yellow color with the GO. The volatility of the G-agents isconsiderably greater than that of the HD, apparently resulting in thelower-than-expected and confusing values for the initial GO and GB testsusing these test procedures.
For these reasons, a modified screening test was investigated for theGO and GB agents. In these tests the decon materials were exposed tochloroform solutions containing the agent. The solutions were sampled andanalyzed for unreacted GB or GO after 5 min and after 60 min of exposure.These test procedures appeared to be less sensitive than the screeningtests involving solid dry decon materials and neat agents. Therefore theprocedure was again modified and the five decon materials which showedpromise in the first preliminary screening tests were reevaluated butusing larger amounts of reactants.
About 200-250 mg of solid decon was contacted with 5 jil of GD and GB.The procedure was 'e same as that described previously for HD except forquantities. The results of this screening are shown in Table 7. Againthe amount of agent recovered in the control samples was less than ex-pected (about i5% to 70% of applied agent). However, using the controlsas a base for calcilating decon effectiveness, calcium hypochlorite wasthe most efficieit decon material for GO and GB. HTH was as effective ascalcium hypochlorite with GB but not as effective with GO. In general,the decon materials appeared to be somewhat more effective against GB thanGO. Only very slight yellow color was observed when the 1,3-dibromo-5,5-dimethyl hydantoin was contacted with GO or GB. The color did intensifyon standing- after dilution with chloroform.
Based on the screening tests carried out in this study, calciumhypochlorite was the material of choice with respect to decontaminating GOand GB, as well as HD. The sodium dichloroisocyanurate dihydrate (ACL56)was earlier shown to be effective as a decon for HD but shows only partial
5 decon effectiveness for GB and GO. Consequently increased emphasis wasplace,! on encapsulation of calcium hypochlorite, with more limited empha-sis or the ACL56. Further work with HTH and STB was curtailed becausethese are non-homogeneous mixtures of different kinds of particulatematerials, unlike calcium hypochlorite and ACL56.
Further Study of Decon-CWA Reactions--The screening tests, discussedabove, were carried out as single experiments. To confirm the screeningresults and to gain some insight with regard to reproducibility of thesetests in light of the small quantities and restricted handling techniques,selected evaluations were run in triplicate. These evaluations includedcalcium hypochlorite with GD, GB, and HD; ACL56 with GD, GB, and HD; and
32
00
0 0
xa
o C 0 0 0 0-;; - a, 9 0D 0I
I 0
I 40
u ms
CLV 2r0 0o '- c-
0 0 0oM
.0 E
0)
'S 0 a, 0ý 0 % ~u'0 U
La" w0 50500
00
0 00 co CS "9-
w -0 CS Cc0* 0 00 IWO
a1 - - cOSS 0>a 0 U0
10 It ~~ 0
.00
1.0C 0
cs us o '
G ~ MI ES 0
.0 .0 .5 4.5 4
10 04.5 .5 0 v 00
03305
coO 0 I . I aI
'V'
10 00C '
I QIn0 00 coa! ~ ~ C. cn 0 o o 0 IZoC ýC )C
X.m C rK ) 01 4 1Ný Oc-4W a
u- 4.'
CD, m 00000
oI*~~ 0'=,C' C> I.> 0 0 0C)000 oa' (71 Jr ulý m I z~r~I- 4mVýe
10 1 'V OIC4)'
WIV WV z IV V S-4> >-
00 4)m> V - -
41 0)) .1
IV ~ IVS-. IV gI - I dIa jiii
34E
13-dibromo-5 5-dimethyl hydantoin with HD. The results of these evalua-tions, using 5 tl of CWA and about 200 mg of decon materials, are shown inTable 3.
Against HD, all decons evaluated were completely effective. Of thethree tests using calcium hypochlorite with HD, one resulted in no visiblereaction, one resulted in spontaneous generation of a quantity of whitesmoke, and one resulted in a discernible flash of flame and audible snap.Hiuwever, in each case, decon was complete. With DBOMH, an orange-redcolor was formed immediately on contact with HD.
With GD and GB, the calcium hypochlorite and ACL56 confirmed theresults observed in the screening tests. The calcium hypochlorite wasquite effective (78-88%) against both GO and GB. The ACL56 was onlypartially effective (18-24%) in deconning GB and GD.
It seems somewhat inconsistent that a material would decon -80% of aCWA and not react with the remaining 20% when about 20-fold excess ofdecon is present. The same argument could be used for the 20% reactionlevel. Possibly the "geometry" of the experiment may be a. limiting fac-tor. Five microliters of liquid forms a very small drop and that isapplied to relatively many small solid particles. The interfacial pro-perties of two contacting materials may not be favorable to completedispersion of the agent. Thus, a small quantity (say 20%) may not becontacted by decon. Where 20% of the CWA is deconned, reaction with theCWA may render the decon particles more sorptive of CWA and thus remove itfor further contact with other decon particles.
Preliminary Indicator Materials Studies
Although the major thrust of this project was the development of aneffective encapsulated -OCI- type decon, visual indication of thepresence of agent is highly desirable in order to identify contaminatedsites on the skin of personnel being treated. Consequently, some study ofencapsulated indicator systems was included in the BCL efforts. Thefollowing paragraphs discuss several candidates and approaches for theseindicator systems.
Indicator for HD Agent--The dibromo- and bromochloro-dimethyl hydan-toins have been found to effectively decon HD (mustard) and to also pro-vide intense color on contact with HD. This observation suggests thatthese materials may serve not only as a decon for HD, but also as indica-tors. To prevent contact with the skin of treated personnel, it willprobably be desirable to retain, in the capsule, any free bromine or otherreaction products formed in the color development. It is reasonable toconsider that the coatings selected to encapsulate the decon will also
S| serve to encapsulate the reaction products. Consequently, incorporationof dibromo- and/or bromochloro-dimethyl hydantoins in capsules was plannedfor Task 3 efforts.
35
Indicators for G Agents--The hydantoins also show slight colorationwith GD, but the color is not as intense as desired for an effective
indicator. Therefore, other indicator systems were considered and ex-plored for the G agents.
Both chemical and enzymatic methods have been used to detect thepresence or absence of G-agents. However, the most highly developedtechniques for the detection of G-agents are the chemical tests. Of thesechemical tests described in the literature reviewed, four tests appearedto offer the best chance of developing an indicator system; i.e., theSchoenemann reaction, a chemiluminescent method, a fluorometric method,and a diisonitrosoacetone method.
The reaction, known as the Schoenemann reaction, is based upon theprinciple that perphosphonates oxidize amines (or other types of indica-tors) at much faster rates than do the peroxides alone. The formation ofperphosphonic acid is considered to be the initial step (Step 1).
RO 0 RO 0
P + OOH--> P + X
R' X R' OOHO H+
OH ýT H (Step 1)
RO 0P
R' 0-
The perphosphonate amine, which is in equilibrium with the perphosphonicacid in alkaline solution, oxidizes an indicator to give a colored, fluo-rescent, or chemiluminescent product (Step 2), depending on the specificmaterials involved.
P + indicator > products (Step 2)
R' 00
36
MIR-_. -• • '''-
The perhydroxyl ions that can be considered are sodium perborate or sodiumpyrophosphate peroxide. The colorimetric indicators include o-tolidine,o-dianisidine, leucotriaryl-methane compounds, benzidine, and o-ethoxy-o-sulpho-p-amino diphenylamine (compound #34).
Mhe chemiluminescent test is similar to the Schoenemann reaction andinvolves sodium perborate, as the oxidant, and 5-amino-2,3-dihydro 1,4-phthalazine-dione (luminol) as the indicator.
The fluorometric method, which is also similar to the Schoenemannreaction, is based upon the formation of a highly fluorescent solution ofindoxyl by the oxidation of indole by perphosphonate.
In the diisonitrosoacetone method, the diisonitrosoacetone reactsrapidly with G-agents in a weakly alkaline solution to yield hydrocyanicacid and an unidentified, intensely colored magenta compound. The methodcan be made more sensitive by adding certain compounds, e.g., o-dianisi-dine, o-tolidine, p-aminodiphenylamine, and aminopyridine. Along thesesame lines, an agent indicator crayon was previously developed whichcontained diisonitrosoacetone, o-tolidine, urea, lithium stearate, lithiumchloride, and calcium oxide.
In considering these reactions and tests, BCL staff believed thatencapsulation technology may be of value in developing an acceptableindicator system. The plan for utilizing encapsulation technology in thedevelopment of an indicator system was as follows (shown schematically inFigure 4, using the Schoenemann reaction as an example).
(1) The amine would be encapsulated separately in prills,i.e., the amine would be dispersed in a polymer bead matrix.
(2) The prills obtained in (1) would be incorporated in asecond capsule, i.e., the second capsule system wouldcontain a suspension of the prills (containing the amine)dispersed in an alkaline perborate solution.
(3) Capsules produced in (2) would then be totally dehydratedto a dry powder.
Following application of the dried capsules to a contaminated surface (asa dry powder), a water mist would be sprayed onto the dehydrated capsules.The dried capsules would rehydrate and the G-agents, which are readilywater soluble, would permeate the capsule wall and react with the alkalineperborate solution. The prills would begin releasing the amine component,thus completing the reaction. The final indicator capsule system would bebased upon a colored by-product, as described above, a chemiluminescentcapsule product, or a fluorescent capsule product.
3
• : 37
7,W
44-)
ca >
00
U4-
alla)
A~~- .ý, 4 ~0C 0
U4 V
ý4.
to
Li -5-4 *,o
0 s-i-
0. 0..
s-4 C) $-L. ~I0IK-77
LT-. CC -3-
__:- -77-_: jý
Indicator capsules, as described, would need to be blended withseparate decon capsules. The latter would contain one or more of the-OCI materials. If both kinds of capsules are relatively small andare well mixed or blended, the mixture should provide bGth decon activityand visual indication of contaminated areas.
Selection of Materials for Encapsulation Studies
Three -oCF- materials were initially selected for the en-capsulated decon studies: sodium dichloroisocyanurate dihydrate, calciumhypochlorite, and STB. These materials were selected because they showedthe most effective decon actions with GD in preliminary screening tests,as previously discussed (see pages 32, 35), and because they are alsoeffective with HD. In addition, they did not show excessive reactivitywith solvents and solvent-polymer combinations. Efforts were also plannedand applied for the encapsulation of (1) the dibromo- and/or bromochloro-dimethyl hydantcin for use as an HD indicator, and (2) G agent indica-tors. Later in the program the encapsulation efforts with the Olin HTHwere curtailed because it appeared that this material is composed of amixture of several kinds of granules, rather than single granules con-taining the various chemical components including Ca (OCl) . A possiblesimilar limitation (non-homogeneity) was foreseen for the currentlycommercial STB (super tropical bleach) as a candidate decon capsule corematerial.
The candidate capsule wall materials selected for encapsulationtrials were (1) cellulose acetate butyrate, (2) chlorinated rubber (Parlon®'),1(3) polyvinylidine copolymer (SaranO), and (4) polyvinyl butyral (Butvare.These resins were chosen because they appear to absorb both HD and Gagents readily, and yet release it, as shown by paper spot test. Thistest, summarized in Table 1, indicated that films of these polymers eithersoftened on cePtact with CWA or allowed the CWA to pass through and wet anunderlying paper. These observations suggested that agents should bequickly absorbed by these polymers, yet readily diffused to theencapsulated -OCl material for decon activity.
These polymer choices appear to be essentially confirmed by solu-bility parameter comparisons discussed previously. The results of thestudy of polymer film absorption of CWA vapor, which was completed late inthe orogram, also confirm the choice of chlorinated rubber and celluloseacetate butyrate, but not of polyvinyl outyral or polyvinylidire chloridecopolymer.
39
Task 3a. Encapsulation Studies
Most encapsulation trials were made employing either a phase separa-tion process or a fluidized bed process, though additional brief studieswere made using spray drying encapsulation for coating small Ca(OCl) 2particles with chlorinated rubber.
Organic Phase Separation Encapsulation Studies--Organic phaseseparation works on the basis of solubility and the formation of apolymer-rich phase and a polymer-poor phase when a nonsolvent is slowlyaaded to a polymer solution. Many polymers are not sharply precipitatedfrom solution when selected nonsolvents are added. Across a limited rangeof nonsolvent-to-solvent ratios, a standing mixture will separate into twoliquid fractions. One will have a higher proportion of solvent andpolymer: but relatively little nonsolvent. The other will be higher innonso~vent with some solvent and relatively little polymer.
Phase-separation microencapsulation systems have been applied to avariety of polymers to prepare capsule walls. These polymers includecellulose esters (e.g., cellulose acetate, ethyl cellulose, and celluloseacetate butyrate), polyvinyl butyral, Sarano , chlorinated rubber, andpolylactide.
Capsule diameters may range from a few 1,m to greater than 1 mm. AtBCL, wall-to-core ratios have varied from less than 1:30 to about 1:3.These variations in particle size and wall thickness permit wide ranges ofrelease rates and/or core protection.
Usually the system is used with aqueous cores or solid particles, butit has been applied to oil-in-water type emulsions and to dispersions ofsolids in water. Agglomeration can be a problem with some materials,especially with very small capsules. Conversely, precipitation ofextraneous polymer can sometimes occur. It is occasionally possible toprevent agglumeration by spray drying tne capsular slurry at some selectedstage after the capsule wall has formed.
The phase separation encapsulation runs all used cellulose acetatebutyrate (Eastman CAB 381-20) as the capsule wall material. The firstseveral runs used sodium dichloroisocyanurate dihydrate (Monsanto ACL56)as the core material. Other runs were made with HTH and with Ca(0CI 2 )as core materials. Bleach particles were dispersed (by stirring) in apolymer solution. The nonsolvent was added slowly from a buret, withcontinued stirring. As phase separation occurred, the polymer rich phasecoated (encapsulated) the bleach particles. Tables 9, 10 and 11 showquantities of materials used in various runs. With addition of morenon-solvent, the fluid encapsulant hardened into polymeric capsule walls.
The ACL56 material was sieved to provide various particle sizefractions. The 300-595 jim fraction was used in these initial phase
4-E
40
separation runs. The CAB 381-20 was dissolved in various solventsincluding dichloromethane, chloroform, and ethyl acetate, alone, inmixtures, or combined with either hexane or toluene. A sample of the300-595 •m ACL56 particles was dispersed in the polymer solution bystirring. Selected nonsolvents (hexane, cyclohexane, petroleum ether,toluene, mineral spirits, or selected mixtures of these) were then addedto phase out the resin. The variables studied are discussed in thissection, and all of these variables can influence the effectiveness of theencapsulation process.
Encapsulation studies were continued with additional decon materials,i.e., HTH (proprietary calcium hypochlorite, Olin Corp.) and a commercialreagent grade of calcium hypochlorite (J.T. Baker Co.). The ACL56 and HTHdecon materials were initially sieved into various particle sizefractions. However, only the 300-595 pLm fraction was used in initialphase separation trials. With the calcium hypochlorite, several sizefractions were encapsulated, i.e., 45-74 pLm, 74-150 pLm, and 150-300 pm.
The encapsulation studies performed with each of the decon materialsare listed in Tables 9, 10, and 11. Table 9 lists the encapsulationstudies performed with ACL56, while Tables 10 and 11 list theencapsulation studies conducted with the two calcium hypochloritematerials. It can be seen from these tables that a number of variableswere examined in the encapsulation studies. These variables included (1)solvent type, (2) nonsolvent type, (3) ratios of solvent to nonsolvent,(4) polymer concentration in solution, (5) rate of nonsolvent addition,(6) particle size encapsulated, (7) stirrer speed, (8) core to wall ratio,and (9) effect of solvent/nonsolvent evaporation.
The use of cellulose acetate butyrate (CAB 381-20) as the wallmaterial and the organic phase separation process as the encapsulationtechnique resulted in the successful encapsulation of all of these deconmaterials.
Fluidized Bed Encapsulation Studies--Fluidized bed microencapsula-tion is performed using fluidized bed process equipment, often called,simply, a fluidized bed (FB). A typical FB is a modified vertical column,usually circular or rectangular in cross section. Near the base is aplenum chamber where air or fluidizing gas is admitted. Above this is adistributor plate, which may be a section of screen, porous metal, ordrilled plate. The distributor plate distributes the gas flow uniformlyacross the cross section of the column, and also prevents the bed materialfrom falling into the plenum chamber. A fluidizing chamber is situatedjust above the distributor plate. This is the zone where solid particlesare agitated and/or suspended in the moving gas stream. Above the fluid-ization chamber, the sides of the column taper outward, similar to aninverted, truncated cone or pyramid. The resulting, increasing crosssection causes a decrease in velocity of fluidizing gas. With decreasinggas velocity, suspended particles fall back into the fluidizing chamber.Gas velocity (flow) is regulated to maintain a selected degree of particleagitation.
41
mCCI
Cr n a .4 d, w 4 mCs (W
CL c 0.045-0 w0 0- 00La c
ac a uscL
I~~ .0 C.L - aatam - . 0 .a .a0011- 1.00 , .0 7;- l-iV'44ii~~~ %- 0- c . in . .
0. I .' on 04 JC4- 04 a -U- rx a a aid L a. 0, 000 sCL X UJCD 0"
4, >4S C I. >. -v >) - >-. >
I o a.4am -. C Jnn .0
0.0~~a am -amn
O04LsJC Lý c04 a CO Lous U>-o a CC MIw
W tid Ch> 4 L4 >54 .4as a M 0M0m30> - >a ha0 45 C w CL40 a,'-4mi2 0
> ;0 o- -- UI 0 > =0 > =
>> >0 >00 >0 W>~~0~ 2 L.D 0C -.'g . 0
So '0 EO a E -oaV
Em
0 A inz 10 G44 'DE442
-DC V- .- .0-I
a.. ol 4, vL 10'
I0 a c cm c c C C ap~ ~~~ ~~~~~ a- w4 4- . - - .~ ' 4
in C 0CC" ocE aaS .4~ CCO~ acmn acmn accmi I
- a OO CCa 00 >54 0' aaC . C~a aaa Coo I ,2 a 404a>X~i 0540 a~aO - c0J04 .0>N a> - >XN a>44' , 1>-~~~~~~~~~~~~ a- 3-- 4-w xi4 . cL- m s- Mca'n--.4~U
P *- ao 0 00 ansac CD ao a s 0 rIXx CD Lt -t4>L 4 i 45i ;
>i ~~C CD s oa > D z ~
) m Li i- Ia-n mmco aW cc M a IIo.
iE '. 4.4'C
.0 9 54 P42
-i
TABLE 10. ENCAPSULATION RUNS OLIN NIH
Run * Pulynmr Code Polyner. 9 Core Solvents Nonsolvents Stirrer Setting Ress'.ts
10 CAB 381-20 5 0 20 0 g 375 ml Chioro- a) 75 ml N-Nexane 2 0 (fast drive) Continuous wall observed.form added to solvent triple blade irregular, very thin, aggio-
5) 450 nil N-Neaane nwrutisn. lots of tiiuffadded over 61 nile.
H CAB 381-20 5 0 20 0 g 375 ml Chioro- a) 75 nil Toluene 2 0 (fast drive) Continuous wail Observed.foini added to solvent triple blade sligotly irregular, very thin.
b) 450 nil N-Neuane agg�osmratiomi. lots of coaf',addCd over 58 nuti l.l.l-TrlchlOrcethane causno
aggloileratlun in Quelith vials12 CAB 381-20 5 0 20.0 g 75 nil Ethyl a) 675 nil N-Neoane 2:3 (fast drive) Continuous wzll Obserees.
atetate, added over 91 nin trl7le blade regular. �ory thin, lots of375 ml Chloroform chaff. *,all "frizzy withpolynini' strands adherias.agglrueration Such rfdutfd fionrun- 10. 11
13 CAb 381-20 5 0 20.0 g 25 ml Ethyl a) 675 5-Neaune 2.0 (fast drive) Tv o. continuous touting observesacetate, added soer 22 mis triple blade ci.afC adhering. toating you
"frizzy" as in run 12. seriousagglcemeration Ohase during ron,greutly reduced at end
14 CAB 381-20 5 0 20 0 g 25 nil Ethyl a) 675 nil N-Neiane 2 0 (fast drive) Contivuoss, regular wall, noneachtate. aOOmd over 82 nun triple blade chaff adhering to particicO.375 nI Chloroform a�gliwnnratiov midway through run.
easing at end; lots of thaff i'isolution. amount of chaff adher-ing to particles lessened tniaurdhend Of run
15 CAB 381-20 0 200 g 25 nil Ethyl a) 675 nil N-Nenane 2 0 (fast drive) Continuous, regular, smooth unll,acetate, added over 87 mini, triple blade little if any chaff adnering 05lAd nil Chloroform. particles. aggloiniration nsjcr.187 ml Dithloro -______________
reduced BEST RUN SO FARnethane
if CAB 381-20 5 0 20 a g 25 ml Chloroform, a) 450 nil 5-Nenane 2.0 (fast drive) No coating us particles observed350 nil Oichlora- added over 60 mis triole blade throughout run, heavy polymermethane SuildoP on beaker
AB 381-20 5 3 200 g 200 ml Chloroform, a) 450 nil 5-Nenane 2 0 (fast drive) Continuous. unufurm, snouts sill.2C0 ml Oichloro- added suer 57 sin triple blade smue aggloimratiov during run.ressane mali not at thick us run 15
�0 CAt 281-20 5 3 23 3 70 nil Ethyl cetate.a) 675 nil 5-Neaune 2 0 (fast drive) Continuous, uniform. mostly150 nil Chlo oform. added seer 85 mis triple blade sisooth wall. slight chaff ad-150 n Outs oro- hering to particles. no sugnivi.nmthanie cant difference fror� run 15
19 CAB 381-20 3 75 15.0 g 19 ml EtHyl acetateta) 507 ml N-Neuane 6 5 (Sloe drive) NOTE 1000 ml Serzelius Soaker141 nil Chloroform, added Over 78 riir Ira covered durint run, con-140 nil Oichloro- ae inuous walls. not convletelyrmthane uniform and smooth-hone cha''
adhering to particles. soure
agglcntratlofl. thicker Sal' cyonran 15. distinct walls votobserved till later in run
20 CAB 381-23 75 lSQg 19 nil Ethyl acetate.a) 507 nil 5-Neaune 6S (sloe drive) NOTE 100301 9erueluas beaker151 ril Chloroform, added over 74 nm spiral-wound used, very thin continuous wuHo.100 n'l OichlorO- blade not cn'oletely s.nootii-ch.sffnethane adhering, sail very nonuniform
in niany cases. particles Observedon uhith no saIl could be
Observed. aggloneration
a iv all runs the core material used woo a 300-595 ni sieve fruction of Olin NTN
430
A ___ ______
_d_ 4-_ -0 > - -c
(o ~ ~ ~ ~ ~ ~ ~ -(ooS.c o0 )s.0r
-o Q). a)) LI) a)-v S. J l
a)0 - ( r a 0 - in C 0) I$._ Cj 0 >a) Ca CLa). C m -L) _ ) 0- a .-CC WI-_ý
E. W W4-tn E"- 4-0 m Wv (W a)0. N4 (-0 N4'-' C ca ~ 4->-= ) (D 'r
=~~~~~V =~C I~ -W O N .4-I~~~V 4.)0E.)a) O VL004- a) ) rW )SmSWQS_ P S_ 4- &-4-) S.- -0 S_ L 4- L) E0a)0 o * aS. )a 0W00) (L W0--.-U 0
0ý> C 0)-9- C 0> cO)QJVa C 0> P4- (n4-3 C%
"4-> (Z = .I >,0SZ W0 0 -.cI-
LLJ LLJ - 0 4-) M).0 Li. -0 0 4;.) 4- S_ Li V0 -- in U a) W
CDn o- 0iEOC c )oFooý C =o .S C:
c(-)
04 _0 -t W I
= - f0 0-> >0 _0 00
>_ 5-) 5- 4000s-
Cj o
W A IS- 04-'4
_j LO 0 C0 W04- V . _- (
4-)~~~4. W)rXC
>f 0) =a a)4-0 0 >r > S_ (
ca. t o0 a)- L'4'
<a -0 -S.C)
W a). E4 7E-0 WVL
0 0 cu a)>ý (
C 0O 0 oLo0oC) 0 m0)0 CLO-
co c -0 C) M
0W
u - ro -0
C)U CC4-C
Cf .4- E 'V N) N
(o (U ro ; -r44
There are several types of FB equipment. The one used for preparingmicrocapsules in this study is a modified Wurster column FB. The outercolumn of the fluidization chamber is circular in cross section. Centeredin the fluidization chamber is a second cylindrical section which acts asa separator, creating inner and outer compartments. This separator isadjusted so that its base is located about 5/8 + 3/8 inch above the dis-tributor plate. A spray nozzle is mounted through the center of thedistributor plate and protrudes up into the inner FB chamber. Gas flowsare adjusted so the solid particles flow upward in the center section,where they are sprayed with coating solution. The wetted particles arepropelled upward into the expanded, defluidizing zone and solvent evapo-
rates. These lightly coated particles fall downward, in a fountain orumbrella-like pattern, into the outer (annular) fluidization chamber.This outer chamber has a lower gas velocity than the inner chamber so thenext movement of particles is downward (though with agitation) in theouter chamber and upward in the inner chamber. Adjusting the height ofthe separator helps regulate the movement of particles from the outerchamber to the inner chamber. Repeated passes of particles through thecoating zone results in gradual coating buildup and microencapsulation.
Fluidized bed encapsulation runs were made in the modified Wurster®
fluidized bed (6 inch) using 900-1000 g batches of core particles. Forall runs, the core particles were sieved, and the 300-595 pm fractionused. Most runs used the sodium dichloroisocyanurate dihydrate (MonsantoACL56) as core particles, but the 38388-18-1, -2, and -3 trials usedCa(OCl) particles. All runs were made at I to 16%C using predriedair as •he Fluidizing gas. Solvent evaporation from the polymer solutionsprayed onto the fluidized particles caused a bed temperature drop ofseveral degrees centigrade below ambient (19 0 to 25 0 C). For all runs,fluidization was continued for about 10 min after completion of coatingapplication to evaporate residual solvents. Microscopic examination of
E capsules thus prepared showed clearly defined walls, especially obviouswith wall coating add-ons of 10% or more.
A total of 24 different capsule samples were prepared by this process(see Table 12). Runs 38388-5-2, -5-3, -18-2, -18-3, and -20-2 have DBDMHincorporated in the capsule walls. Figure 5 is a schematic of capsules
V from run 38388-5-3. The FB encapsulation series provides a basis forcomparison of four different coatings on the ACL56 core particles: cellu-lose acetate butyrate (Eastman CAB381-20), chlorinated rubber (HerculesParlon5S125), polyvinyl butyral (Monsanto Butvar® B73), and polyvinyli-dine copolymer (Dow Chemical Company, Saran® F310). It also providescapsule samples to compare core materials (38388-5-1) with ACL56 cores and38388-18-1 with Ca(OCl) cores with Parlon® walls), and capsule wallthickness (percent coating).
Runs 38388-20-1 and -20-2 were made to demonstrate that magnetite canbe incorporated in capsule walls. As expected, this technique was suc-cessful in making the capsules sufficiently magnetic that they are
45
M-u LO U' 0C) co-Z I I I I I 2 I I m C Y)C *
F- m ~00 c
Uu u< a ) a )L n
Mi cl CM r- c M )0 i
LiL
C: UýL)C) u) CM L) Sa) F CIS C') 0 C4 CM4 CM OT
1.0
10 _j
I.-~14 UL,*4~
V) 1.Q S)C Ifl0 C:)0Uf) UL)0LC) L C:) - t LL)0Lr) LU) C: t LflO C) 0 lzr~c 00 CCL. 0~ CMCMJ-- -S
LLJ~S. V-) -PS 4) (W- Cý 100
LL)~~~~~'c 00 )L * L O ,::
co C .. cCM')iU CCM' -CM rcM' -CMS- 111 II CDCC' rCM) .- CCý II II <a)a C') U') Vfl r-. cm CMV
K CL CM CM fu cc Mc c c<o S- cc cc 4-) S- a; cLL- Cl. u. co o- cc Wc -4-
%-Z C') a' C') CIO') C) '
a)46:
/ Core
())-Inner Wall
Intermediate Indicator Layer
[ Figure 5. Schematic of 38388-5-3 multi-wall indicator-decon capsule.
Core --- sodium dichloroisocyanurate dihydrate (Monsanto ACL56) particle
Inner Wall --- chlorinated rubber (Parlon® S125) equal to aoproximatelv 10%of core (by weight)
intermediate Indicator Layer --- one-to-one ratio (by weight) of Parlon®S125and DBDMH e ual to approximately3% of core (by weight i
Outer Wall --- Parlon® S125 equal to approximately 5% of core (by weight)
Note: ACL56 particles are not actually spherical, but are irregularshapes. Circles are used here to more easily depict the basicconcept.
4
47
attracted to a small magnet. Magnetic capsules may provide a means ofrecovering decon capsules from wounds.
Encapsulation by Spray Drying--Several trials were made in an effortto prepare microcapsules by spray drying using a CH19C12 solutionof Parlon® S125 as the wall-forming phase for Ca(OCI particles (< 74 ýLm)as the core material. Core to polymer ratios from 1:i to 3:1 were at-tempted at various levels of dilution with solvent. All trials producedsome threads, fibers, and irregular masses as well as some nicely coatedparticles and bits of superfluous polymer. None of the trials, however,produced a fully acceptable product alt.hou one system was selected fortesting with agent.
Indicator Development and Encapsulation Studies
Several indicator systems have been studied as part of the encapsu-lation efforts. The DBDMH appears promising for use as an HD indicator aspreviously mentioned on page 39. Also, 2-napthol has been found to de-velop a visible color with the indole-perborate system. The details ofthese efforts are discussed next.
Indicator Materials Solubilities--Indicator capsule systems basedupon either a chemiluminescent or fluorescent capsule product were se-lected for study. Luminol is the indicator associated with the chemilu-minescent reaction and indole is associated with the fluorescent reaction.The perhydroxyl ions (providing proper alkalinity) are supplied by sodiumperborate in both of these reactions. Thus, in preliminary studies, somebasic solubility data was obtained for both luminol and indole. Table 13lists the "gross" solubility data obtained.
Although 100 mg of luminol was only slightly soluble in a 1% sodiumperborate solution, it became readily soluble when trisodium phosphate wasadded. However, when this test was performed only 50 mg of luminol wasused in a 0.01% sodium perborate solution. Thus, although it appears thetrisodium phosphate increased luminol solubility, additional studiesshould be performed since the luminol and sodium perborate concentrationswere not equivalent in these tests.
Indole was also examined for its solubility in the following aqueoussolutions:
S(1) 1% sodium perborate/1% ethylene glycol
X (2) 1% sodium perborate/5% ethylene glycol(3) 1% sodium perborate/1% glycerol(4) 1% sodium perborate/5% glycerol(5) 1% sodium perborate/5% tetraethylene glycol
48
TABLE 13. SOLUBILITY OF LUMINOL AND INDOLE INVARIOUS MEDIA AT ROOM TEMPERATURES
Media Luminol Indole
Distilled water IS a isGlycerol IS b ISEthylene glycol SIS RSCI% Sodium perborate d SS e IS
a IS = insolubleSIS = slowly soluble
c RS = readily soluble0.1 g of luminol or indole was added to 9.9 g of a 1%
e aqueous sodium perborate solutionSS = slightly soluble
4
249
F
k
a9 .8 g of a 1% sodium perborate solution
0.1 q of ethylene glycol0.1 g of indole
Sb9 . 4 g of 1% sodium perborate solution0.5 g of ethylene glycol0.1 g of indole.
Indole was not soluble in any of the above solutions.
A follow-up study on indole solubility involved exposing 100 mg ofindole to varying concentrations of ethylene glycol and glycerol. Theseresults are presented in Table 14. The only method found for solubilizingI indole at a relatively rapid rate in glycerol was to first solubilize theindole in acetone or 200 proof ethanol. Glycerol was then added to theacetone or ethanol solution with rapid mixing. The acetone or ethanol wasthen evaporated with continued stirring. Residual solvent was removed byplacing the indole/glycerol solution in a vacuum oven for -1 hour. Aresultant clear solution of indole in glycerol was achieved, i.e., 10 mgindole/ml glycerol.
Preparation of indicator Prills--Six batches of indicator-containingprills were prepared for later incorporation into indicator system micro-capsules. Luminol and indole were selected as indicator materials.Disonitrosoacetone was also considered but later excluded because with GBand GD it produces hydocyanic acid as a reaction product, which is alsovery toxic and potentially lethal. Other indicator and color enhancingmaterials considered included o-dianisidine, o-tolidine, and amino-pyri-dine. These were excluded from further study and development because oftheir suspected carcinogenic potential.
Each of the s'x batches was composed of 1% (by weight) indicator and99% matrix (carrier). The three different matrix materials used were (1)polycaprolactone polyol (Union Carbide's PCP0260), (2) polyethylene glycol(Union Carbide's Carbowaxal 8000) and (3) Castorwaxe (a castor oil deriva-tive from Cas Chem, inc.) These three waxy materials have considerablydifferent degrees of water sensitivity. The Carbowax® is water soluble.The PCP0260 is water sensitive but not water soluble, and the Castorwaxis a good water barrier material. Each matrix material was combined inhot melt with each indicator (luminol and indole) to provide six combi-nations. These hot melt systems were then prilled to produce sphericalbeads. Each batch of prills was sieved into four fractions: <150 Pm,150-300 ur, 300-600 wnm, and >600 im. These prills were used for laterencapsulation studies.
Evaluation of Luminol Prills--One objective of these studies was todetermine if the indicator was released from these prills. The leachingtest for luminol indicator prills was performed by placing 1 g of theprills (containing 0.5% luminol) in 10 ml of the leaching solution. The
W• 50
C)) - if
S- S- a) 4
It P +-) 4-3 4-' 4- ,.
>) 4-) 4-) 4-) i
iic c c co o .
4J >
0-0
CAC
-CD V-
c CD 0
C0
C) C) CD --- (A
01~r ccr r~
u~i .,- tv
' ~ 4- J4 4-P 5 I 4.)
4 1 - 4- 4- 4~ . )
>_I 0D 0J 0-l 4- ' S- C~ o
I.n WJ -, -n M 0 .0 3m w- 0 ER
C:> (D - 1- 0) -) -- ( ) 4-) )
0 0 0 4-) 4- 4-) (0 a
C). M. -. 0 -1 L 0.0. 0 -o C
to .,- 0~I C ~ CD
ii0 c) 00 CDJ
-j OI CL CLV -coto t (n 'of 4-- ~- 0
14 0 0 0 4--tLO -) 4-a-0 C) 4-
0 00 0)4- CO)
-- 4-- C)
0 0 C=-
~- a) .
14--,L) LO~ C) 0 C)CC CS.- k- 00.J'-
51
size range of the prills tested was 150-425ý±m diameter. The leachingsolution contained 1 mg of sodium perborate and 10 mg of trisodium phos-phate/t0 m! of distilled water. Leaching tests were performed at 1, 2, 5,and 10 min of exposure.
One drop of a 1% cupric chloride/distilled water solution was addedto the filtrate collected at the end of the leaching test. Cupric chlo-ride gave a false positive test and the intensity of the chemiluminescentreaction was rated on a scale of I to 10, with 10 being ve-y intense. Theresults of the leaching tests for prills containing luminol are presentedin Table 15.
The results from Table 15 indicate that there is some luminol on, orclose to, the outer surface of the prill. However, there is no doubt thatluminol is being released from the prills, especially the PCP 0260 prills.The Carbowax 8000 prills released the luminol very rapidly, but theyreally had no integrity, i.e., they fell apart and dissolved.
Evaluation of Indole Prills--Based upon the indole solubility studiesinvolving ethnylene'glycol and glycerol (discussed previously in this re-port), efforts to develop a fluorescert reaction were carried out. Toobserve the fluorescence, a small hand-held UV light (Minerallight, UVSL-25) was used. The "lorg wave" setting was used in these observations. Toconduct these studies, 100 mg of indole was added to 10 ml s3lutions con-taining 9 ml of either a 5% or 80% ethylene glycol or glycerol solutionand I ml of a 1% sodium perborate solution. Each total solution wasagitated for 10 min. A series of four indole solutions was prepared byadding 100 mg indole and 1 ml of a 1% sodium perborate solution (aqueous)to 9 ml of eac.h of the following: 5% ethylene alycol in water, 80%ethylene glycol in water, 5% glycerol in water, and 80% glycerol in water.
Following agitation, the solution was filtered and each filtrate wassplit into two vials. One drop of a 0.1% cupric chloride solution wasadded to one of the vials and the fluorescence observed. The resultsindicated that the blank vial (containing no cupric chloride) also pro-duced fluorescence, i.e., the background fluorescence was considered toohigh to evaluate indole release from prills. Thus, based upon theseobservations, it was felt that another material, which would produce afalse positive test, should be investigated.
Si 52
CD rr WI ~4-)L
0 Q)4-) 3: S
>1 o00 4-)-)
I.~u C0-C 0)
S- CL)L S- CLo 3: a) C)~~~- 4- 0 )0) .
Q) C) 4--jý
cy 14-(5 CJ- O I-W - C 4-4- ) - 4-
W'- 0 U 0
LI4- ii q
C -
Lf) CiO S- * : (n 0
V) CL a)0) 0) u
I -' O 4- 0n
II T o tM -W ->
U-j m - )'~ 4- 4-
4- S. 0) -a -
< ~ 0- 0 00-
F- CL
53
CL 4-____S.-_0_S-_4-
In checking the literature it was discovered that phthalicanhydride produces a reaction sensitivity similar to that of chemicalwarfare agents, i.e., it produces a (false) positive test in the fluores-cent reaction. A similar test to that involving cupric chloride was con-ducted with phthalic anhydride. However, in these studies, a 0.1% ph-thal'ic anhydride-isopropanol solution was substituted for the 0.1% capricchloride. The results of these studies indicated that the filtrate pro-duced the desired fluorescence when 1 drop of 0.1% phthalic anhydridesolution was added to one of two vials, one vial serving as a blank. Thefiltrate tested was obtained from the following reactants:
9 ml of a 5% glycerol-distilled water solution1 ml of a 1% sodium perborate-distilled water solution100 mg of indole
As a result, the prills (containing indole) were evaluated for theirrelease Pf indole by placing 2.0 g of the prills (containing 0.5% indoleby wei§.t) in 10 ml of the test solution, i.e., 9 ml of a 5% glycerol-distilled water solution and 1 ml of a 1% sodium perborate-distilled watersolution. If all of the indole was released from the 2 g of the prilis,corresponding to 10 mg indole/lO ml of test solution, optimum developmentof the fluores-ent reaction shoul occur. This concentration of indolewas based upon literature sources as providing optimum fluorescence.The results of these leach studies are presented in Table 16.
The results shown in Table 16 indicate that there is some indole on,or close to, the outer surface of the prill. However, there is essen-tially no indole being released from the prill as evidenced by no increasein the intensity of fluorescence with time.
The difference in release observed between luminol and indole prilismay be related to the fact that when the prills were prepared, a
1 "Development of A Multipurpose Kit. Search of Open Literature"Research Report 63-939-532-RI, May 31, 1963, F. P. Byrne (ed.) K. W.Guardipee, 0. H. Kriege, R. J. McKeever and R. J. Nadilin, Contract No.DA13-108-AMC-115A.
2 "Detection and Estimation of Nerve Gases by Fluorescence Reaction",Bernard Gehauf and Jerome Goldenson, Analytical Chemistry, Vol. 29, No. 2.
5
• 54
4-) 4-) 5
II U4- S-
I I 4 -Q)
(T 4) 0 ).a)-~I-a
co 0 CL4-)E 0
L) i*,
V) M ~ 0 S-EU.1 -j05-P a) 4-)
SI C..) ~ u4-co= -- - s- >)
I 1 '. 0'.1 -
CL LC) a) 4-
C4-) C~) V)ra) 4-) (-
CD I -0 4-) -
U L)' -r0 S.-
-J W V)4-) 4- a
LU U) L'4- r 4-'
-a) I (U ) WM 4-W C au
4 l-~)(A -3 C) 4)
C 0 4- Q)
(2) 0-) a)'U fa>C 4-)
V/) (ZS3 L 0 - 1E C:LU fa 4-) co ( 0
-j LOI C)WU > U
- 0- 0o4 -4- *-) 4-
LU) U C., * ) -c V a) -
co (l ) C) c..j C\J .- - C S- a)< 4-) 4-)4- u- 0
F- *- -
II- Q>- C'L W C410 ~ 41 C )
x C (1) 31. c-
C:) 1~ 4-) ((/1 00- C -
ro x C) L) i! ( -
ro k S- )551
melt blegd of indole and polymer was f 8 rmed; i.e., the prills were formedat 65-93 C while indole melts at 52-54 C. In the case of luminol,it may not have solubilized at the temBeratures used in the preparation ofthe prills, i.e., luminol melts at 320 C. Thus, the luminol wouldhave been dispersed as a homogeneous suspension of discrete particles inthe prills. As the luminol is dissolved from the matrix, channelingprobably occurs and further luminol may be released. This is not the casewhere the polymer mdtrix may have more thoroughly blended with the indole,perhaps forming a true solution.
Further examination of the literature3 has disclosed the possi-bility of developing an indicator system which can be detected by a visi-ble color change. In order to investigate this possibility, work wasperformed on developing an indole/2-napthol indicator system.
This reference indicated that the following ratio of reactants mightbe feasible in developing a visible color change:
100 mg indole
200 mg 2-napthol reactants10 ml ethanol ITo examine this possibility, the following test was devised. One ml
of the above reactants was mixed with 2 ml of a 1% sodium perborate-dis-tilled water solution further diluted with 6 ml of distilled water. Thisvolume was then split into two equal fractions. To one of the vials wasadded 25 drops of 0.1% phthalic anhydride simulant. To the other vial, 2drops of glycerol were added and the solution mixed. This was followed bythe addition of 25 drops of 0.1% phthalic anhydride. The following finalconcentrations of indole, 2-napthol and sodium perborate were examinedusing the above procedure:
Sodium perborateSample Indole (mg/ml) 2-Napthol (mg/ml) (mg/ml)
A 1.1 2.2 2.2B 1.1 4.4 2.2C 2.2 1.1 2.2D 4.4 1.1 2.2
3 "Analysis of GA by the Metalloid-Halogen Reaction. PreliminaryReport", Bernard Gehauf and George B. Wilson, Army Chemical Center, T. D.M. R. No. 1307, Project A 1.13, Control Number 5004-1307, Feb. 28, 1947.
2f 56
The addition of 2 drops of glycerol was shown to enhance color de-R velopment. In addition, maximum color development (dark green) was ob-
tained when the ratio of reactants in sample C were used. Increasing theW number of glycerol drops, i.e., 4 and 6 drops, did not increase the color
intensity using the ratio of reactants in sample C. However, increasingthe concentration of simulant, i.e., phthalic anhydride, from 0.1% toeither 0.3 or 0.5% did increase color intensity with sample C reactants.Thus, it might be feasible to develop a capsule indicator system basedupon color development alone. Obviously, the color indicator capsuleproduct would be preferred if sufficient color develops for visualdetection.
Indicator Encapsulation Study--A core solution of the followingcomponents was encapsulated by an organic phase separation process usingcellulose acetate butyrate (CAB) as the wall material: 4.4 mg indole and2.2 mg 2-napthol in 2 ml of ethanol, 4 ml of a 1% aqueous sodium perbor-ate, 470 mg of glycerol, and 12 ml of distilled water. The microcapsuleswere humid-air-dried to remove residual solvent from the capsule wall.However, when a 0.5% phthalic anhydride-isopropanol solution was placed ona sample of the capsules, no color change was observed. Previously a darkgreen color developed with this agent simulant.
It appeared that the solution used to phase out the polymer hadprobably extracted one of the core components, e.g., 2-napthol. Conse-quently, the post-loading of microcapsules, which contained only a dis-tilled-H2 0 core was investigated (to avoid possible loss of indicatorsinto the encapsulation processing fluids). This was accomplished by pla-cing 1.5 of either humid air-dried or dehydrated capsules in 9 ml of thefollowing solution for 2 hr:
Ut indole (299 mg)1i ni 2-napthol (100 mg)
200 proof ethanol (10 ml)
2 m.l 1% sodium perborate6 ml. distilled water4 drops glycerol9 ml approximate total
However, once again, when the 0.5% phthalic anhydride-isopropanolsolution was placed on a sample of post treated, filtered capsules, nocolor change development occurred. Thus, one of the components may not betransferring across the CAB wall within the allotted time of the post-loading process, or the CAB wall was masking any color development.
It was later found that indole is much more soluble in the organicsolvents (used in the phase separation process) than in water (used incapsule cores). Consequently most of the Tndole, initially in the aqueouscore solution, was probably transferred to the organic fluids during the
57
4-
encapsulation process. Therefore, it was decided that a post-loadingtechnique would be used to imbibe indole into prepared microcapsules.CAB microcapsules were prepared using 1.1 g of polymer for wall formationand a core containing 9.0 g of a 5% glycerol/H 0 solution plus 1.0 gof 1% sodium perborate/H 0 solution. After thý capsules wereprepared and filtered, t&ey were humid-air-dried for several hours toremove residual solvents. In this process, residual solvent is removedwithout altering the core contents; i.e., the capsules retain theiroriginal diameter as achieved during their preparation. These capsuleswere then dispersed in a saturated aqueous indole solution (containing1% sodium perborate) in order to post-load the capsule core by anequilibration (diffusion) process. Triton® X-100 was used to disperse thecapsules in the aqueous solution. The capsules were rinsed with distilledwater to remove any surface indole using a filtering and suction appa-ratus. The capsules were then removed from the filter paper and humid-air-dried for 1 hr.
A sample of these capsules was spread as a thin layer on a glassplate. Three single drops of saturated phthalic anhydride (aqueous) wereplaced at three different points among the spread capsule sample. Thesample was then observed under UV light. After about 10 min, a definitefluorescence developed at the points where the phthalic anhydride solution
had been added.
The results of this study are not definitive, but the concept doesappear promising.
Task 3b. Capsule Evaluations
indicator and decon capsules were evaluated with neat CWA. In addi-tion, selected decon capsules were evaluated in leaching tests and in skintests. The following sections describe these efforts.
Evdluation of Indicator Systems with Neat Agent--The purpose of thispart of the effort was to evaluate the more promising indicator reactionsusing neat agent (GB and GD). Prior studies had shown that simulantsprovided the desired reactions with the fluorescent, color-change andchemiluminescent systems. The results of these studies, conducted atBattelle's Hazardous Materials Laboratory, are presented in Table 17. Ofthe indicator systems, the fluorescent system showed the most promise andappears especially promising with respect to a possible microcapsulesystem since a very small droplet in a glass capillary tube was easilydetected.
fThree different samples of earlon® S125 chlorinated rubber encapsu-
lated ACL56 were prepared by fluidized bed (FB) encapsulation: (1) ACL56particles coated with 10% Parlon®S125, (2) ACL56 particles coated with
z58
It S. "€ 0 coSI~~~~i (T3 0 (D-•. ,_(• . r
Si__- _'_- ____ _____ • ",
S 050 .r- 0
"-0UU
0 4-
-- 4-) 0
V)U
a) I- 4-ou L-0 to- CIO
) CA
•-~~- E= u Do•S •0'0
CV4.
4- ) CJ C
--- "--4-Z 40 0'
0•" I ' "C4-'( "(0 . ,. U -
v) 0(). Ii- U)VU) S-V
4 - * ,JI 0 S.
I-. - , " 41U r*,**- - U)
,, 0 ' ~ 0 00 F"0 2 0)
C>V) S-4- (1)U)) 0 0 5_) 0- --
2J C- a)
, a "- -- 4) 0-1 4-'ui~ E- a (L
0 ( V( 4 '-) (D U)(
o) 0) - Q CM U) a ) a* 4-)
U) ca E L 4-' C U )< 4 -U ) - a) E) C 4
W C)
~~~~- 4-- U U C oUC-, U -~ C~) ~ )CL U) to
4' S-U
F S-
C)( U)a)(Uoo- 4 -. 4- M
a)4-3 CO
rU) : a) In
UJ < (U 4--'
-4-) - 00 4-).0
__j I, 4-J 4-c).C 0 ( -co ~ ~~ C(OUa ca
Co S L)- .4 .'ot 130)(0 E>C- 7-M ( 5- 1 1 4-) C C4-0- S-000 M.- C\ -a-
=-4-. C C- = 3-(A S
CDL c-'-- 0'0U- U) V2
I-~ a0.j (A~0C 'A0
U0' '0-'- IU0- CS-- U) aU
q -P (A ) = CD C .- ( U) S.-
u (A a)Ua) O)C C- a)) 4-) uC U)(
04(t))CC0EC4-'C4-) j0 U0 --0V -0= 00 -ILIC I Ct'
r- 10U o'U -SC) COC-)~ -W C
59
Parlon~plus a second layer (3%) of a 1:1 ratio of Parlon®and DBDMH, and(3) ACL56 particles similar to (2) but with an added third layer (5%)comoosed of Parlon®. About 0.2 g of each material was treated with 5 Wlof HD. There was no visible color development with the particles coatedonly with Parlori. However, in both systems where DBDMH was present, animmediate yellow color was observed. This color appeared to deepen forabout 5 sec and then remained constant. There was no qualitativedifference in color development between the systems (2) and (3) describedabove. Thus, as anticipated from prior studies on this project theincorporation of DBDMH in the capsule wall appears to show considerablepromise as an indicating decon for HD.
Evaluation of Encapsulated Decons--Twelve capsule samples were se-lected for testing with agent and are described in Table 18. Capsulesamples 38388-5-1, -7-2, -8-2, and -10-2 wPrP qpiected to compare theeffects of four different coating materials. Other samples were selectedto compare the effects of capsule size or encapsulation processes on theperformance of the capsules when exposed to agent.
Test Procedure. To compare the effectiveness of -OCl-
capsules in absorbing and decontaminating CW agents, the following testprocedure was followed.
(i) ',lid capsule samples were weighed into conical-bottomcentrifuge tubes (quantity of -OCl material equalto about 20 moles per mole of agent).
(2) Five microliters of agent were added and the tubes capped.
(3) After 5 min 10 ml of hexane was added, mixed thoroughly,and allowed to remain in contact with the capsules for 1 min.
(4) 1.0 ml of the hexane solution was sampled and diluted with2.0 ml of hexane for analysis by gas chromatography.
(5) The remaining hexane solution was decanted from the capsulesan-' i0 ml of chloroform was added to the capsules inthe tube, mixed thoroughly and allowed to remain in contact withthe capsules for 5 min.
(6) The chloroform solution was centrifuged, if required, and1.0 ml of the solution was sampled and diluted with 2.0 ml ofchloroform for dnalysis.
j (7) Steps (1) through (6) were repeated for fresh samples,but 60 min was allowed between addition of the agentand the hexane wash.
60
&c4-) 4-) 4-) 4-) C ) )
CL) (3 a) 0) ~ - - - ( (0 (U (0'o -. _0 -0.0C 0. O- 0cu a) a)
C)0CL ) N N N N N N N C / I I I(1: -u
cn LeV- ~ - -) V) V) V)CL 0 -' -* -0 -0~ -0 ' > (3) 13) W)U.J0 a- :3 (U fu fu (`0L- t- - - L- Li- LL 0. CL 0- CL 0-
- - - - ci CMj IN CMi C\J
LU -
LL LLC) U) U) ) U')CM) CMC) CMI CD (lCMQ) V) V)
CM I \ C'~J (\.j ,.2:CO 00~ 0 : 0 00 00 CIOX7>- illV)C o Ili LIo <. t o C) m .(15 Cl L) <
LIC)
C ) IM ) m ) CM U ) m U ) C ) U*1) ' L' ) 0) 0-) U) LO 0) Lf CY)U) Ul U) C) C) C) U)C ) C') C:)
S- -- m oC) C) m m mV -t rI
IN IN IN>1 L OL I O CD CD CM C) CM CM in -C
CO~~~- (a rom r Il (
C\1 ' CMj -CM J CM j CM j
I
(D Ln LO N-C - - -- -El EEC C O C CO O COCO -ID
00 ol Io 00 0 I 0I I( COD Coo CO 00 COD CO CO CO CO CO CO CO
mI~ ' C') m' m' m' C' C' (V) m) IM m 0) Ir~co 0O CO C O CO CO O N- Niiii - N-iii Ni- IMM' m' ' C') C') C') C') m' C' C') CY) c
61
The hexane extraction provides a measure of absorption efficiency ofthe capsule coating (a smaller amount of agent in the hexane extractindicates higher absorption of agent by the coating). The chloroformextract measures the permeation efficiency of the coating and deconeffectiveness of the -OC1 material; i.e., a low agent content in thechloroform wash indicates good permeation and decon action. A highercontent of agent, however, probably is indicative of high absorption ofagent by the coating but low permeation.
HD-Microcapsule Test Results. Tables 19 through 23 detail theevaluation results obtained with the selected microcapsale systems withHD. Several capsule systems showed > 90% of HD deconned at 60 min and twoshowed > 85% of HD deconned at 5 min. The percent of HD from the hexanewash, plus the percent of HD from CHC1 wash, plus the "deconned"percent of HO equals 100% based on 500d pg HD in each test. (A series offour controls showed 5003 + 113 •g HD.) The hexane wash results includeany HD that may have been on test tube surfaces as well as surface wash ofthe capsules. As noted previously the CHCI wash percent HD is be-lieved to represent HD absorbed in the capsule wall but not yet reacted.The "deconned percent of HD" is calculated by subtracting the total of thetwo solvent wash values from 100%.
Table 19 shows a comparison of effectiveness of the four selectedwall materials used to coat (10% level) ACL56 particles (300-595 pm) in afluidized bed. In this series, the Parlon® S125 (chlorinated rubber)coated capsules show the best overall performance, i.e., the lowest hexanepercent of HO at both 5 and 60 min as well as the highest percent of HDdeconned at both time periods (77.3% at 5 min and 96.9% at 60 min).However, the CAB 381-20 (cellulose acetate butyrate) capsules show resultsnearly as ,avorable as the Parlong capsules. The Saran® coating wassignificantly less effective and the Butvar® B73 capsules showed poorabsorbency and little deconning, especially at 5 min.
The test results shown in Table 20 provide a comparison of Parlon®coated ACL56 and Ca(OCI) 2 particles with Parlon®-DBDMH coatings as asecond layer indicator. As expected, the capsules containing DBDMH showeda bright yellow color shortly after being contacted with HD (-15 sec).The color appeared to be the more intense with the Ca(OCI), capsulescoated with ParlonO and 3% of dibromo compound than with t e ACL56particles coated with Parlon® and 1.5% DBDMH. A white cloudy or smokyhalo appeared above the Ca(OCl) capsules on the sides of the tubes,especially with the 1-hr sampleý.
It appears that the incorporation of DBDMH in the surface layer ofthe coating (these two capsule samples have double walls) increased the
4 absorption rate of HD and the rate of the decon activity. The data alsosuggest that increasing the thickness of the Parlon®-DBDMH layer increasesthe effectiveness. As noted, the DBDMH also serves as a visible indicatorof HD because of its color change to a bright yellow when used in the
6S~62
-4-
00 a) U) -~
I-.
CN- CM
LUa 4--
c 0- CY)
LL- m (a CM
c/))C) C)
< KnzJU Cý CY)0 ) 00 mo toCC:) C I C4- - U,
C- -j
LLAUa) 4-
C) Ln C I0 0 *(n 0c0(1
-JO a) -
Fi ) F- CL
U C)
L~j
c. 0 -
to . c l - n
CIO~633
M m~
C4-)
-j<
CD v ý,0'- ro
N._ in 00LA iCDI . D i
0.L
-l IAii4 zclz CD '.
L)On
CL- .x V. .ý 0
C-: 0
FE_-I Ii Lo
C\j C\J i
Ln cn- 01F- u < 4-
<J '.0 c- . -
L. LI) 0 C-)0C) C-) C-> a
I C-, U' L ')>S.- C'j C' ~ 0
204-) V- - 0 .
tu -0o~ 0 CL 0 ~ *
LU '- - 'a-ll
ro ro C'J m. Ocoti a) - -- .
F- 'a-, I - I 0)*)
Sm - 64
Parlon® capsule walls. Reasons for the specific differences inperformance between the Parlono coated ACL56 and Parlon' coatedCa(OC1) 2 particles are not presently known.
Table 21 shows comparative data for three sizes of Ca(OCl) coreparticles coated with CAB 381-20 using phase separation encapsuiation.There appears to be a clear trend towards more and faster absorption of HDwith decreasing capsule size especially in the 5-min test. This sametrend is also seen at 60 min for the three Ca(OCl), capsules. Allthree samples of smailer size capsules show generally better deconning at5 min and less decon effectiveness at 60 min than the larger ACL56capsules. This latter observation may indicate that the quantity ofCa(OCI) in the capsule core of the smaller capsules is insufficientto provide complete deconn 4 ng of HD.
Table 22 compares two capsule systems with ACL56 cores and CAB 381-20capsule walls. One capsule system has a 25% CAB capsule wall and wasencapsulated by the phase separation process. The second system has a 10%CAB capsule wall and was encapsulated by a fluidized bed process. Theperformance/effectiveness at 60 min is similar for both samples. However,the 5-min samples show considerable differences, especially in the hexanewash (nonabsorbed HD) and the percent HD deconned. The complete reasonsfor these differences have not been established although the amount of wallpresent is obvious. It is suspected that the coating configuration isdifferent for the two processes and that these differences may be moresignificant than those related to percent coating (coating thickness).
Table 23 compares spray drieu capsules with capsules formed by phaseseparation. As described earlier ii this report, even the best spraydried capsule samples contained fiDers and irregular bits of polymer aswell as "good" capsules. The spray dried capsules show rapid absorptionof HD, similar to the smallest capsules from the phase separation seriesand similar totals for total deconning at 5 and 60 min.
Overall, the most effective deconning of HD is shown by capsules fromruns 38388-5-1, -5-2, -7-2, and -18-2 (see Tables 18, 19, and 20). Itappears that Parlon® is the best of the wall materials evaluated for bothrate of absorbency at 5 min and more complete deconning at 60 min, al-though CAB also looks almost as promising when coated on ACL56 particlesby fluidized bed (38388-7-2). Incorporation of DBDMH with Parlonein thesurface layer of the capsule wall further increases the absorption rate ofHD, as well as providing visual color indication of the presence of HD.Thus, 38388-5-2 and -18-2 are presently considered to be the most promis-ing capsules tested for HD deconning.
Test Results of Microcapsules with Neat GB and GD. Table 24 showstest results obtained with a series of capsules, with ACL56 cores andvarious polymer walls, treated with these agents. As expected from testswith non-encapsulated ACL56 and neat GB and GD, there was very little
65
~~, C: O-1cý
- 0; C\J C'
(~Lu
C))
I-L) C)
c~~aC~ LO LO
OL LOc) cL)I
0- C0
F-C LC) IC) Cj
V-) I,ý
CD a-
=) CL - D N-j 'VO U)*lI '
OLJQ. I; " a
apa)<u Li C-
(Y)/ C) CD c
-. 1
(NJ (N -z
K,- -(13% IUu- 14 O0- C c c N
MEg -LJ;-ý
Wa
Qw a
(Ii2C\j M
C~)
cIi a) .Zo
C>0) ~ M _rI
(l -J CU
Iý CD0)
F- r- -o
4-j( <- cr
(no </
- Cj
CII (a) a) 0
(3 C\J
co u a a)67
0 4-)W C- N
ua) 1 1 C-) V) C) i
C~) = MC~
(- 1
(n)U
C- L- C
L)C=)
0ý CL C~)
cz~ r
B4
9UC' -- (
ca~ 4-lF
0-l -~ a)
C)- iIL O C'; V
>0.. a-e~UJ(/)S *' C-
CD- SC
ALC) >0a
S- tz
n CL
0V)
C~~~~j~ .- CU >)D)
C> Q)
68
IWIU-,
4--
4-)00 11 (.0 (1)
LnC M
Li - Lo i )I a
(= I . .0li I t-_C)U
C) -- r- co.- ~CD \J r- c \ (.0 0~) ~ -0
UUa) co
C) =~- r I
4-)Ln)0 C-
M C I
-J (n~
LU! - I E
L-) (n LO \ -. C) U
(f)Z- (U1 )l
C-)m
c'to
UO UO 0 0Lw U C) (\j 0ý C') u S-
0jI C:- I,- a) 4-
I L-I LL.
co M' MU > (1 U0l
(0 4> C3 s- -) S. . 4)-
69
effective decon activity observed. The results shown in the hexane andCHCI- 3 wash columns suggest that SaranO F310 and CAB 381-20 (cellu-lose acetate butyrate) are more absorbent to GB and GD than are Parlon®S125 and ButvarO B73. All four of the polymers show higher absorbence forGB than GD. It is not known why some tests showed greater than 100%recovery of agent in the hexane and CHCI 3 washes.
The data presented in Table 25 provides a comparison of fluidized bedParlon®-coated capsules with those coated with Parlon&-DBDMH. The capsulecores are either ACL56 or Ca(OCl) as indicated. While the presenceof DBDMH in most of the capsule wZlls does not appear to significantlyaffect the total decon activity of the capsule, some of the hexane washresults do show definite decreases. Furthermore, the differences areareater with the Ca(OCl) core capsules than with capsule cores con-taining ACL56 (Runs 38389-18-1 and -18-2 compared to runs 38388-5-1 and-5-2 described in Table 18). It has not been clearly established at thistime whether these differences are due to the core material or to thethicker (6% vs. 3%) wall of 1:1 DBDMH/Parlon®. However, it appears likelythat with GB and GD, the action of the DBDMH is primarily to increase therate of agent absorption (i.e., less agent is "recovered" in the hexanewash especially at 5 min). These observations contrast with the observedPffects with HD which showed an increase in decon effectiveness when DBDMHwas included. These results, however, are consistent with results oftests with DBDMH and neat agents (HD, GB, and GD) as reported earlier,i.e., the DBDMH is an effective decon for HD but not for GB or GD. Theiicrease in absorption rate suggests that incorporation of other selectedmaterials in the capsule walls may further improve the absorption rates.
The data in Table 26 compares the decon activity of variousCa(OCl) 2 core sizes, coated with CAB 381-20 by the phase separationprocess. There is no clear trend of absorption rate (low hexane percent isinterpreted as rapid agent absorption) or decon effectiveness as a func-tion of capsule size with either GB or GD. The decon effectiveness at 60min compares favorably with non-encapsulated Ca(OC1) presented inTable 8, which shows 70% to 85% deconned for GD and 83% to 91% deconnedfor GB.
Table 27 compares ACL56-core particles coated with cellulose acetatebutyrate polymer (CAB 381-20) prepared by two different encapsulationprocesses, i.e., phase separation and fluidized bed. In the GD 5-min testresults, the hexane wash percent for the fluidized bed capsules is muchhigher than for the phase separation capsules, but the reverse is truewith GB. The percent deconned is low probably from the low decon activityof the ACL56 cores with both GB and GD. In the 60-min test, the hexanewash percent figures are more uniform between agents and between capsulesamples.
70
C'O O~ D UD c
cli k0 Lo - 00
C'CD
4-)I
Q) () 0d ) I NC) ~ 1 -<N r- 00 -)'r0 C )
ii.C) liLO4
CD - X V) r-- coJ (.0~.(- cc U)U).>U
Ijn - JU
Im U)) -c
L.u ) 4-- (M0) I U
cn4 4-.) C') '. C
:r (1) CU) 0) U) 0\ U\ ) 0. U)I' ~ci m;;imU) a) Cr)
MNJE (0a
. 4.)U)Jjr c i - * ~ j (\ *r=D V 4- -)- 4
Ca - 1- -1 to
> 0 -71
- C- U) mf 00~
MCLUJ <4- D
VI)
(0- M C LrU) qr to0
IiV X (A-i I
=v Lj i r- m. c I- N-i
- 0I-u.)
<w
LIr- 4-) .. - -
V) Lo .0- i
C) C)
-T-' co x tnC) k f C)
N. - CD C) CN-: C
S- N. -- rn I- I~ (CS
(0M4- 1C
72
C: CY) (u
4-)-
a)0 CY . N- (n
or ka)
0D -a
00T 00C 0
C:) m
~Ln
2= Le)
C~Lr j4) c ~ ol MC0 L) 0 U
u CJ) (1)
4--')U
21- 0<-
'U 4-) MU U'v-U C.- cn CD c
U)0~~0 a)0 ()0
0.5. V) Na.) N c)
cu C o u 4-c)
X d- ciL YC) C) 73
In Table 28, Ca(OCl) 2 particles encapsulated with Parlon® byspray drying are compared to Ca(OCI) particles encapsulated with CAB381-20 by phase separation. The 5-m~n hexane wash figures for Parlon• aresimilar to those in Table 24, but the 60-min test results for Parlone inTable 28 were much lower. This may be due to the smaller particle size ofthe spray dried capsules vs. the fluidized bed capsules. As noted earlierin Table 24, the CAB wall material shows faster absorption for GB and GDthan does Parlon®. In addition, the decon effectiveness at 5 min is muchhigher for the CAB capsules than the Parlon®, although the CHC1 3 washfigures are nearly equal, suggesting similar permeation rates.
As noted previously, the spray dried capsules include bits of fiber,filaments and "fuzz", as well as good capsules. Such polymer fragments(not containing any decon material) would probably provide minimal deconactivity. Consequently, some portion of the agents would likely becometrapped in such fragments, and this entrapment may account for the lower60-min percent deconned data for the spray-dried Parlon' capsules.
Tables 24, 25, and 27 show data for percent deconned from 0 to 29 forACL56 cores with GD and 0 to 41 with GB. Tables 25, 26, and 28 show datafor 1-hr decon percentages from 57 to 75 for GD and 65 to 80 for GB withCa(OCl) cores. This data correlates with earlier studies that showednonenca~sulated Ca(OCU) to be a more effective decon for GB (88%) andGD (78%) than for nonen~apsulated ACL56 (see Table 8).
Capsule Leach Tests. Encapsulated decon applied on incapacitatedpersonnel may contact body fluids such as blood or perspiration, or mayotherwise be subjected to presence of liquid water or other aqueous media.it is believed that release of large quantities of the -OCI- corematerials, especially into open wounds, would be generally undesirable.Consequently it was considered important to compare capsule samples forrate of leaching in water.
Ten capsule samples, previously evaluated with CWA, were selected foraqueous leach tests. These were 38388-5-1, -5-2, -7-2, -8-2, -10-2, and-18-1, each of which have calculated coating levels of 10% applied in th2fluidized bed and 37984-19, -43, -46, and -48, each of which have calcu-lated coating levels of 25% applied by phase separation. In this test, awe I .. ... n.. , of capsule u0.22 or 0.25 g) was placed in a beaker with
500 ml distilled water and stirred during the first hour. Aliquots (50S>ml) were taken at 5 min, 30 min, 60 min, and 24 hr, as well as 48 and 72
hr tests for selected samples. These aliquots were analyzed for "avail-able chlorine" (a standard measure of the -OCI- activity, i.e.,available oxidizer). The 5-, 30-, and 60-min periods were chosen, basedon the assumption that the decon capsules would usually be removed fromtreated, incapacitated personnel within an hour of their application. The
4• 24-, 48-, and 72-hr tests were run to provide additional comparative dataSon these capsules.
74
-o)
Q)
CC
0
00
< H-ji104
-AJ 0)
CD 0 co
~ ) 0 Ln cIC- X
C,, V) -LO a) -h0
u~ Ej
CDC
IO CD)
C)
00 CL 0 tom -0 S- o
O-4-'O4- 'C0
s..iL :3c cl I i C1
a- Cc. 4--<J 0.- m/ to .- ~ t
u 2: - -75
The analytical test used was titration with 0.1 N sodium thiosulfatesolution in the presence of dilute sulfuric acid, potassium iodide, andstarch. The various core materials used in preparing the selected capsulesamples were separately analyzed. These were 300 to 595 ým ACL56, 300 to595 ýLm Ca(OCI) 2, 150 to 300 •m Ca(OCl) , 74 to 150 Vm Ca(OCl) and 45 to74 ým Ca(OCl) For these core materi•ls, 0.20 g samples were placed in500 ml distil•ed H 0 and stirred for 5 min. Then 50 ml aliquots weretitrated as for caisule samples. Table 29 shows test results with percentleached calculated by comparing ml sodium thiosulfate required for cap-sules to that required for the appropriate core material.
Contrary to expectations, the -8-2 (Butvar® polyvinyl butyral) andl-7-2 (cellulose acetate butyrate) show better results than -10-2 (Saran®)
or -5-2 and -5-1 (Parlon®). The latter two polymers are usually con-sidered to be better water barrier materials than the other two. TheCa(OCl) appears to leach more rapidly than the ACL56 (-18-1 comparedto -5-1 for the fluidized bed coated samples. The fluidized bed CABcoated ACL56 showed slower leaching than the phase separation CAB coatedACL56 (38388-7-2 compared to 37984-19) although the latter has a (calcu-lated) higher level of coating applied. The presence of DBDMH in theouter layer of capsule wall (-5-2 compared to -5-1) appears to increasethe leaching rate during the first hour, as expected.
A comparison of capsules from runs 37984-43, -46, and -48 suggeststhat -43 and -46 (the two smallest sizes) were completely leached at 5min, whereas -48 showed only 44% leaching. The fact that leach resultsfor -43 and -46 did not appreciably increase after 5 min, though thenumbers are 56% and 67% respectively, is interpreted to mean that thesesmaller capsules have undergone loss of activity during processing (en-capsulation) and/or during storage.
The leach tests were terminated by 72 hours, at which time the -8-2,-7-2, and perhaps the -10-2 capsules were continuing to release. However,the low figures for these three at 72 hours may indicate that there hasbeen some loss of activity during preparation or storage. Further evalu-ations, would be required to determine this and should include capsuleswith various coating thicknesses, such as 38388-7-1 and -7-3. This PhaseI feasibility study was planned to determine a basic feasibility of theconcept of using encapsulated -OCI materials as decons, and was notintended to include a thorough evaluation of all significant properties ofall of the capsules prepared.
Another series of aqueous leach tests was also run on capsule prod-ucts from 38388-7-1, -7-2, and -7-3 which have calculated coating levelsof 5%, 10%, and 15% respectively (weight basis). Table 30 shows the
L results of this test series. As expected, the percent leached at varioustimes decreased with increasing capsule wall thickness. Past experiencewith fluidized bed encapsulation has shown that sometimes a rather sharpincrease in coating effectiveness is observed at around 8% coating (cap-sule wall) with particles about 0.3 to 1.0 mm. Results obtained in these
76
TABLE 29. COMPARATIVE CAPSULE LEACHRATE EVALUATIONSa
Percent of Available Active ChlorineTitrated at Times of
Sample Minutes HoursNumber 5 30 60 24 48 72
38388-8-2 0 1.4 1.4 26 39 47-7-2 0 0.5 0.5 41 70 79-10-2 2.1 12 20 59 68 70-5-1 0 21 47 92 96 95-5-2 8.8 59 82 84 -- ---18-1 4.7 90 100 102 103
37984-19 48 106 110 109 --..
-43 56 S6 57 58 57 --
-46 67 70 69 72 71 --
-48 44 82 95 101 100 --
aFigures shown are percentages based on comparison of;itrations of capsule leach liquor at times shown tosolutions prepared from nonencapsulated correspondingmaterials in estimated equivalent quantities (i.e.,based on 0.20 g core material in 500 ml distilled water).
77
II _ _
TABLE 30. EFFECT OF WALL THICKNESS (% COATING)ON AQUEOUS LEACH TEST RESULTSa
38388-7-) 38388-7-2 38388-7-g(0.21 q)b (0.22 g)b (0.23 g)
I CAB coatingb 5 10 15
5-min leach 1.2% 0% 0%
60-min leach 53% 0.5% 0.5%
24-hr leach 93% 41% 21%
crushedc not determined 97% 104%
I~a-Figures shown are percent of core material leached based on0.20 g core material (ACL56).
bCalculated percent coating. Actual percent coating probably
somewhat less. Sample size used for analysis was based onthis calculated percent coating to provide 0.20 g corematerial
c Sample crushed with mortar and pestle, dissolved in water
and analyzed.
7
.• 78
tests appear consistent with that experience, with a much greater diffe-rence seen between -7-1 and -7-2 than between -7-2 and -7-3, especia'ly at60 min.
Among the capsules evaluated in the leach test, the 38388-8-2 (10%Butvar• coating on 300-595 pm ACL56 cores) and the 38388-7-2 (10% CABcoating on 300-595 pm ACL56 cores) show especially excellent results up to60 min, and these low levels of -OCl- leached probably constitute aminimal hazard. Depending on the level of capsule application, the amountof fluids available, and -OCl- leach rate in these specific fluids,some of the other capsule systems may also provide adequate leach resis-tance. The very dilute, pure water leaching is probably more extreme thanmost actual conditions.
Skin Decon Experimentation Using Microencapsulated Decontaminants.Previous studies have indicated that the penetration behavior of a widenumber of chemicals through pig skin is similar to that through humanskin. Also, it has been demonstrated that there is a definite and repro-ducible relationship between the permeabilities of skins from differentspecies . Based upon these findings, pig skin was chosen for the skindecon phase of this study since it is extremely difficult to obtain cad-aver skin samples. This change in the basic testing procedures wasagreeable to the Sponsor. The skin samples were removed from a young pigwith an intact layer ot subcutaneous fat. The whole skin from the bellyarea was stored at -20 C.
Typically, diffusion cells are used in the measurement of chemical
penetration through excised skin. In a previous study conducted atBattelle, a diffusion chamber was designed (see Figure 6) to study CWApenetration through excised pig skin and the effect of various decontami-
nants.
The chamber was designed to hold the sample taut by a stainless steelretaining ring on a TeflonO locking2 ring. The sample was suspended withan ultimate exposure area of 1.3 cm
Prior to analysis, the skin sample was thawed and the subcutaneousfat removed using a surgical blade. The skin was draped over the samplelocking ring, and the skin retainer was slipped into place to secure thesample. Excess skin was trimmed away from the ring. The locking ring andsample were slipped into the decontamination chamber with the fatty sideof the sample resting on top of a stainless steel screen and held in placewith a Teflonering.
4 Tregeor, R. T. Physical Function of Skin, Molecular Movement: ThePermeability of Skin. New York: Academic Press, p. 1 (1966).
79
0-CLz
z -j
WH 0 1-
~3>
COC
ww
ZEEO
80
The apparatus was transferred to a hood and the epidermis was spikedwith 10 pI of neat agent. The agent was allowed to penetrate the samplefor 5 min. A premeasured amount of decon was then applied covering theentire surfa:e of the exposed skin. The apparatus remained undisturbedwithin the hood for an additional 30 min. The skin surface was thenrinsed with hexane to remove the microcapsules and any residual neatagent. The rinses and microcapsules were collected in a beaker. Therinses were quickly decanted into a volumetric flask to minimize extrac-tion of agent from the microcapsules. The rinse solution was diluted tovolume and the amount of residual agent was later determined by gas chro-matography.
The skin sample was ground in a homogenizer with hexane as the sol-vent and sonicated for 1b min to extract the penetrated agent. The solu-tion was concentrated to 1 ml. The sample extract was fortified with aninternal standard and subsequently analyzed by capillary gas chromato-graphy/mass spectrometry. The individual chromatograms from these analy-ses are shown in Figures B1 to B16 in Appendix B.
Three types of control samples were analyzed. The first, a nondosedskin sample handled in the same manner as a test sample to determine thepresence of any of interferring compounds. No interferences were detected(see Figure B-I).
The other controls consisted of skin samples dosed with neat agentfollowed by a 5- or 35-min wait. The controls were then handled as werethe test samples. Tne 5-min delay sample was informnational on the amountof agent that penetrated before application of a decon. The extent ofpenetration pos:sible without decon added is evident with the 35-min expo-sure. These findings are listed in Table 31. As expected, the degree ofpenetration was proportional to the increase in exposure time.
All microcapsule samples tested showed decreases in penetrated agentpercent at 35 min compared to controls. For example, the 38388-18-2 testshows only 5% agent oenetrated at 35 min compared to 32% for control (and1% for control at 5 min). Thus only an additional 4% penetrated aftercapsules were applied, compared to an additional 31% for the control. Ananti-penetration effectiveness percentage was calculated from the data forpenetrated agent (see Table 31) with percent anti-penetration effec-
tiveness equal to 100 (I - test sample - 5-min control35-min control - 5-min control). Figuresgteater than 100% anti-penetration effectiveness indicate that the per-centage cf penetrated agent is lower for the capsule test sample than forthe 5-min control. These data imply that the capsules have not onlydeconned the skin surface but have also extracted scme of the agent that
dhad already penetrated into the skin whet. the capsules were applied.L
81
TABLE 31. CONTROL AND DECONTAMINATION ANALYTICAL DATAaFOR PIGSKIN TESTS
NonrecoveredResidual Penetrated and Deconned Anti-Penetrationb
Decon Agent Agent Agent EffectivenessSample Agent Wt (g) (%) (%) (%) (%)
5-rin HD 77 1 22
control
35-min HD 35 32 33
control
38388-7-2 HD 0.4238 28 13 41 61
38388-5-2 HD 0.4425 10 9 81 74
38388-18-2 HD 0.4001 57 5 38 87
5-min GB 76 3 21 -
control
35-min GB 2 127 0control
38388-18-2 GB 0.3110 30 17 53 89
37984-43 GB 0.3083 15 7 78 97
37984-48 GB 0.3021 16 3 81 100
5-min GD 81 21 0 -control
35-min GD 63 41 0 -
control
38388-18-2 GD 0.3183 63 9 28 160
37984-46 GD 0.3192 14 5 81 180
37984-48 GD 0.3054 11 13 76 140
aAnalysis was performed using gas chromatography or capillary gas chromatography/miss spectrometry (see page 81).
bCalculated from figures in the penetrated agent column as follows:%E 100 ( test sample - 5-min c3ntrol
Anti-penetration Effectiveness I -r5nin control 5-min control)
Figures greater than 100% indicate extraction of agent already penetratedwhen capsules were applied 5 min after agent was applied.
S82
While the 38388-18-2 appears best of the three capsule samples testedin anti-penetration effectiveness for HD, the 38388-5-2 capsules performedbetter in decon effectiveness (agent destroyed).
The 38388-18-2 capsules contained DBOMH in the capsule wall anddeveloped color with neat HD as described earlier in this report. Colorwas not observed in the skin test with these capsules and HD untilcapsules were washed off the skin test sample. However, the quantity ofcapsules applied formed a layer that was more than one capsule inthickness. Thus, capsules contacting skin and agent were obscured byother overlaying capsules. Yellow-colored capsules were observed whencapsules were removed during the wash step (after 30 min of capsulecontact with the HD contaminated skin sample).
Run 37984-48 capsules performed best in skin tests against GB, bothin decon activity and in anti-penetration effectiveness, and prevented anyadditional penetration of agent after capsules were applied. With GB, allthree types of capsules appear to extract GB from the skin sample, but3nly 37984-43 and 37984-48 show high-level decon activity.
Although time and money did not allow for replicate analyses, thedata, even based on limited experimentation, are encouraging and do indi-cate that decontamination is probable with the proper microcapsule sys-tems. A more extensive study is necessary to evaluate more precisely theefficiency of the microcapsules as decontaminants. Oifferent parametersneed to be varied to evaluate the effect upon the decons and replicateexperiments need to be done with a statistical evaluation of the data.Some of the recommended parameters to vary are as follows:
s Exposure Time. The critical time for application ofthe decontaminants needs to be delineated.
s Temperature. The temperature of the skin sample needsto be varied to simulate variances of actual typical skintemperatures of real subjects in different field conditions.
Application. The physical exertion used in applicationmight affect the efficiency of the decontaminant; i.e.,applied lightly versus impacting the decontaminant on theskin with force.
Identification of Unknowns. There are unknown peaks identifiedin the different chromatograms shown in Figures B-3 to B-17.The peaks may be metabolites and should be analyzed todetermine if they are toxic also.
83
. 1 •--- -- --
APPENDIX A
PATENT REFERENCE LISTS A AND B
85
APPENDIX A
PATENT REFERENCE LIST AMODERATELY RELATED PATENTS
1. Mazzola, L. R., "Encapsulated Bleaches and Methods for TheirPreparation", U.S. Pat. No. 4,327,151 (April 27, 1982).
2. Mazzola, L. R., "Encapsulated Bleaches and Methods for TheirPreparation", U.S. Pat. No. 4,136,052 (January 23, 1979).
3. Mazzola, L. R., "Encapsulated Bleaches and Methods for TheirPreparation", U.S. Pat. No. 4,126,717 (November 21, 1978).
4. Alterman, D. S. and Chun, K. W., "Encapsulated Particles", U.S. Pat.No. 4,124,734 (November 7, 1978).
5. Faust, J. P., et al, "Method of Inhibiting Scale Formation:', U.S. Pat.No. 4,087,360 (May 2, 1978).
6. Mazzola, L. R., "Encapsulated Bleaches and Methods for TheirPreparation", U.S. Pat. No. 4,078,099 (March 7, 1978).
7. Alterman, D. S. and Chun, K. W., "Encapsulation Process", U.S. Pat. No.3,983,254 (September 28, 1976).
8. Hachmann, K. et al, "Storage-Stable, Readily-Soluble DetergentAdditives, Coating Cempositions and Processes",
9. Alterman, D. S. and Chun K. W., "Detergent Compositions ContainingCoated Bleach Particles", U.S. Pat. No. 3,944,497 (March 16, 1976).
10. Alter,,an, D. S. and Chun, K. W., "Encapsulation Process for Particles",U.S. Pat. No. 3,908,045 (September 23, 1975).
11. Williams, S. D. et al, 'Stabiliting Peroxygen Compounds by Enveloping in aWiter-Dispersitle Layer", U.S. Pat. No. 3,847,830 (November 12, 1974).
12. Fries, W. er il, "Bleaching Assistants and the Preparation Thereof",U.S. Pat. No. 3,833, 506 (September 3, 1974).
13. Briggs, B. R., "Method of Combining Optical Brightners with Polymersfor Stability in Bleach and Encapsulated Product", U.S. Pat. No.3,666,680 (May 30, 1972).
14. Robinson, R. A. and Briggs, B. R., "Bleach Having Stable Brightners",U.S. Pat. No. 3,655,566 (April 11, 1972).
15. Horvath, R. J. and Parsons, C. G., "Coated Chlorine-GeneratingMaterials for Treating Fluids", U.S. Pat. No. 3,647,523 (March 7,1972).
S~86
16. Schiefer, J. and Dohr, M., "Bleaching Detergents and WashingAdjuvants", U.S. Pat. No. 3,459,665 (August 5, 1969).
17. Schiefer, J. and Dohr, M., "Bleaching Detergents and WashingAdjuvants", U.S. Pat. No. 3,441,507 (April 29, 1969).
18. Wilhelm, W. F., "Method of Coating Chloramine and Product Thereof",U.S. Pat. No. 1,950,956 (March 13, 1934).
19. McNeil, C. P., "Coating Granules", U.S. Pat. No. 1,480,561 (January 15,1924).
20. Alterman, D. S. and Chun, K. W., "Encapsulation Process", British Pat.No. 1,509,797 (May 4, 1978).
87
PATENT REFERENCE LIST BMARGINALLY RELEVANT PATENTS
1. Brubaker, G. R., "Encapsulated Bleaches and Methods of Preparing Them",U.S. Pat. No. 4,279,764 (July 21, 1981).
2. Saeman, W. C., "Round Multi-Layered Calcium Hypochlorite Granules",U.S. Pat. No. 4,276,349 (July 30, 1981).
3. Saeman, W. C. et al, "Granular Calcium Hypochlorite Coated With anInorganic Salt", U.S. Pat No. 4,201,756 (May 6, 1980).
4. Saeman, W. C. et al, "Granular Calcium Hypochlorite Coated With anInorganic Sait by Spray Graining", U.S. Pat. No. 4,174,411 (November13, 1979).
5. Brennan, J. P. et al, "Decomposition Inhibitors forChloroisocyanurates", U.S. Pat. No. 4,149,988 (April 17, 1979).
6. Saeman, W. C. and Coe, N. N. "Granular Calcium Hypochlorite Coated Witha Low Melting Inorganic Salt by Spray Graining", U.S. Pat. No.4,146,676 (May 27, 1979).
7. Gray, F. W., "Peroxymoonosulfate-Base Pleaching Detergent Compositions",U.S. Pat. No. 4,123,376 (October 31, 1978).
8. Saeman, W. C., "Granular Calcium Hypochlorite by Spray Graining", U.S.Pat. No. 4,118,524 (October 3, 1978).
9. Steward, D. A. et al, "Solubility Stable Encapsulated DiperisophthalicAcid Compo3itions", U.S. Pat. No. 4,094,808 (June 13, 1978).
10. Saeman, W. C., "Granular Calcium Hypochlorite Coated With a Low MeltingInorganic Salt by Spray Graining", U.S. Pat. No. -4,048,351 (September13, 1977).
11. Saeman, W. C., "Spray Graining Technique for Preparing GranularHydrated Alkalai Metal Dichloroisocyanurate". U.S. Pat. No. A,005,087(January 25, 1977).
12. Lamberti, V., "Process of Coating Calcium Sulfate Dihydrate DetergentFiller Particles", U.S. Pat. No. 3,997,692 (December 14, 1976).
13. Denaeyer, J. L. and Kegelart, W., "Sodium Chlorite ContainingGranules", U.S. Pat. No. 3,997,462 (December 14, 1976).
14. Saeman, W. C., "Process for Preparing Grarular Calcium Hypochlorite ina Fluidized Bed", U.S. Pat. No. 3,969,546 (July 13, 1976).
15. Faust, J. P., "Encapsulated Calcium Hypochlorite Granules", U.S. Pat.No. 3,953,354 (April 27, 1976).
.g788
16. Sakowski, W. J., "Process for Manufacture of Calcium Hypochlorite",U.S. Pat. No. 3,895,099 (July 15, 1975).
17. Kinne, W. E., "Production of Anhydrons Sodium Metasilicate in aFluidized Bed", U.S. Pat. No. 3,884,645 (May 20, 1975).
18. Berkowitz, S. and Roth, E. S., "Method of Stabilizing DichlorocyanuricAcid Salts", U.S. Pat. No. 3,853,867 (December 10, 1974).
19. Denaeyer, J. L. and Kegelart, W., "Process for Manufacturing GranulesContaining Sodium Chlorite in a Fluidized Bed Drier", U.S. Pat. No.3,844,726 (October 29, 1974).
20. Nielson, D. R., "Diperisophthalic Acids", U.S. Pat. No. 3,880,914(April 29, 1975).
21. Breece, J. and Brown, M., "Effervescent Tablet", U.S. Pat. No.3,821,117 (June 28, 1974).
22. Ries, C. R. and Smith, Jr., G. C., "Dishwasher Detergent Composition",U.S. Pat. No. 3,817,869 (June 18, 1974).
23. Corliss, D. S., "Stable Dishwashing Compositions Containing SodiumDichloroisocyanurate Dihydrate", U.S. Pat. No. 3,816,320 (June 11,1974).
24. Dychdala, G. R. and Custer, R. S., "Caiciur Hypochlorite Composition",U.S. Pat. No. 3,793,216 (February 19, 1974).
25. Weldes, H. H. and Ve3sey, E. W., "Process for Preparing CoatedDetergent Particles", U.S. Pat. No. 3,783,008 (January 1, 1974).
26. Inamorato, J. T., "Effervescent Granules"', U.S. Pat. No. 3,769,224(October 30, 197 3).
27. Marshall, D. E., "Metnod for Producing Laundry Products:', U.S. Pa.. No.3,761,549 (September 25, 1973).
28. Golton, W. C. ana Rutkie'-c, A. F., "Stabilized Solutions of N-HaloCompounds", U.S. Pat. No. i,749,672 (July 31, 1973).
29. runakoshi, Y. et al, "Process for Coating Granular Materials", U.S.Pat. No. 3,671,296 'June 20, 1972).
30. Faust, J. P., "'Preparation of High Test Calcium Hypochlorite", U.S.Pat. No. 3,669,894 (June 13, 1972).
31. Nielson, D. R., "Preparation of Diperphthalic Acids", U.S. Pat. No.3,655,738 (April 11, 1972).
89
32. Hudson, R. B., "Process for Preparing Products Having PreferentiallyInternally Concentrated Core Components", U.S. Pat. No. 3,650,961(March 21, 1972).
33. Van Dijk, A. B., "Fast Dissolving Non-Aqueous Built Liquid DetergentCompositions", U.S. Pat. No. 3,630,929 (December 28, 1971).
34. Piester, L. W., "Encapsulation Method", U.S. Pat. No. 3,622,366(November 23, 1971).
35. Robson, H. L., "Calcium Hypochlorite Compositions ContainingSpray-Dried Sodium Nitrate", U.S. Pat. No. 3,560,396 (February 2,1971).
36. Gray, F. W., "Bleaching Packets", U.S. Pat. No. 3,528,921 (September15, 1970).
37. Land, J. P. and Nielson, D. R., "Encapsulated Perphthalic AcidCompositior:s and Method of Making Same", U.S. Pat. No. 3,494,787(February 10, 1970).
38. Nielson, D. R., "Coated Perphthalic Acid and Method of Making Same",U.S. Pat. No. 3,494,786 (February 10, 1970).
39. Zimmerer, R. E. and Lyress, W. L., "Novel Bleaching Compositions", U.S.Pat. No. 3,393,153 (July 16, 1969).
40. Kaneko, T. M. and Schmolka, I. R., "Chlorine-Stable DetergentCcmpositions and Process f3r the Preparation Thereof", U.S. Pat. No.3,359,207 (December 19, 1967).
41. Wixon, H. E. "Bleaching Composition", U.S. Pat. No. 3,346,502 (October10, 1967).
4,. Olberle, T. M., "Process for th• Preparation of Storage Stable
IDetergent Composition" U.S. Pat. No. 3,306,858 (February 25, 1967).
43. Robson, H. L., "Manufactu'e of Calcium Hypochlorite Article", U.S. Pat.No. 3,234,141. (February 8, 1966).
44. Robson, H. L. and Shel t imire, W. H., "Calcium HypochloriTe Article andProcess", U.S. Pat. No. 3,154,495 (October 27, 1964).
45. Speak, R. C. and McConnelly, P. H., "Fabric Laundry Compositions", U.S.Pat. No. 3,154,494 (October 27, 1964).
46. Gleichert, R. D., "Process for the Minufacture of CakciumX7 Hypochlorite", U.S. Pat. No. 3,134,641 (May 26, 1964).
47. Morgenthaler, J. H. and Parks, T. D., "ProcesE for Making a BleichComposition", J.S. Pat. NO. 3,112,274 (November 24, 1959).
90
48. Jaszka, D. J. and Marek, R. W., "Coated Calcium Hypochlorite andProcess for Making Same", II.S. Pat. No. 3,036,013 (May 22, 1962).
49. Robson, H. L., "Production of Calcium Hypochlorite Product and Methodof Manufacture", U.S. Pat. No. 2,963,440 (December 6, 1960).
50. Robson, H. L. "Spray Drying Calcium Hypochlorite Slurry", U.S. Pat. No.2,901,435 (August 25, 1959).
51. Darbyshire, R. W., "Precipitation of Basic Calcium Hypochlorite", U.S.Pat. No. 2,436,745 (February 24, 1948).
52. Day, G. G., "Calcium Hypochlorite", U.S. Pat. No. 2,347,402 (April 25,1944).
53. Robson, H. L. and Petroe, G. A., "Granular, Dustless CalciumHypochlorite Product", U.S. Pat. No. 2,195,757 (April 2, 1940).
54. Taylor, M. C., "Production of Calcium Hypochlorite Product", U.S. Pat.No. 2,195,756 (April 2, 1940).
55. Pobson, H. L. and Petroe, G. A., "Production of Calcium HypochloriteProduct", U.S. Pat. No. 2,195,755 (April 2, 1940).
56. Robson, H. L. and Kaufmann, H. D., "Crushing Soft HeterogeneousMaterial", U.S. Pat. No. 2,195,754 (April 2, 1940).
57. Reitz, H. and Ehlers, H., "Method of improving the Handling and StoringQualities of Compounds of Calcium", U.S. Pat. No. 1,916,770 (July 4,1933).
91
L APPENDIX B
CHROMATOGRAMS FOR SKIN TEST STUDY
93
CD
CJ
toI
L.JI
C"-,-
CDa,
0 kl, C9
94w
00
CD
S-
.0)
L.)
CD \J
U,-
95,
C~
0ý
Ch
0.01IDk
- 4-)
LO -
"C4 0L '
c6i
96~
- wWM
QQw
inf
CY)
0 C00
C. -j
C'~
-15
0re
SWr
97
Ii..i
N -L
co~
CC--
(.J98
'Ol
inn
Z LM
oc ic,.- iot
.4-)Uc,
r,4 CD
0 -- J
Li.
e44
0-L
.WO
IL -
0 U
I9
-- ----
1(A
S0
S- -
•'-.,,.
00
400
U 'Lu
I Ev
I'M~
I 100
0 m1 m ••.) mmm km~mmm-w "" [[ ' 1
F Cl
0AO~
I-O V)-z Sn S
CD"
I',--
I~c 0
100
IC.ý
00
C.c,
vcf S
CZ.
102?
~,-.Q ~-
ICIDn
~CP
4-,
0)
C103
C.44 9
-- co
Cr
Q')
-S-
-A L)
R -i
104
crkU
Lm,
o 0 -
W) -
100
7-
* ~ -' ~ - - W =~y~-~ -~* -~--C
'AL
0x
0IZ
0 5L-
Af 106
51A.~
CC
U,
co
I co
w~ C
kn C'
00
UJ 4J
U-
C"-
C 4.
107
cI'
em li
" S CC)0
4J
(3) LA0
AS,-
t 3
-fur-
ISSR2108
z co
NN. ' )fý ?
.
CC
C:d)
co.
109