UNCLASSIFIED

AD NUMBER

AD466061

NEW LIMITATION CHANGE

TOApproved for public release, distributionunlimited

FROMDistribution authorized to U.S. Gov't.agencies only; Administrative andOperational Use; Jun 1965. Other requestsshall be referred to the Office of NavalReseach, 800 North Quincy St., Arlington,VA 22217.

AUTHORITY

DDC, per IR ltr dtd, 17 Nov 1965

THIS PAGE IS UNCLASSIFIED

STEC HNIC AL REPORT NO *

OFFICE OF NAVAL RESEARCH

ADVANCED RESEARCH PROJECTS AGENCYARPA ORDER NO. 299, AMEND. 3

CONTRACT Nonr 4200(00)TASK NR 356-452

CHEMILUMINESCENT MATERIALS

JUL l bAMERICAN CYANAMID COMPANY

-I

CENTRAL RESEARCH DIVISIONSTAMFORD, CONNECTICUT I

MARCH I. 1965 - MAY 31. 1965

REPRODUCTION IN WHOLE OR IN PART IS PERMITTED FOR ANYPURPOSE OF THE UNITED STATES GOVERNMENT

p

NOTICE: When government or other draw.ngs, speci-fications or other data are used for any purposeother than in connection %ith a definitely relatedgovernment procurement operation, the T. S.Government thereby incurs no responsibility, nor anyobligation whatsoever; and the fact that the Govern-ment may have formulated, furnished, or in any waysupplied the said drawings, specifications, or otherdata is not to be regarded by implication or other-vise as in any manner licensing the holder or anyother person or corporation, or conveying any rightsor permission to manufacture, use or sell anypatented invention that may in any 'ray be relatedthereto.

4010

II SWeil ! 0 0 ,41-. W 6 1 o 1 6 A A N t

Cheailumineseeut fttoriala

4 OCICRIF-vive oe INOSI of ml pq and Inrlu.w( fee *0

Progreus Report No 8Preb 1, 11965 MaPy 31, 1.965)6 AUTHOX(8) (Lose nsova. hole name Inflid.I

Raubut, Milchael X.

S.RPOT 30 June 1.965 79 CO PSE 7 31.o.Rs

88 CONTRACT ON GRANT -40 90 ORIGINA?011111 MR00PORT NU14BER(S)

Nowx 4200(00) 06-14i32-25-46&PRtOJECT No.

ARPA order No. 299, Amn 396 gSb j0M '90() (Anyop re~ nmsh Atom be seeiw.o

code 3860oT~R~Od. Task MR 356-4&52 Nn

10- A VA IL ASPILITY/LIMITATION NOTICE&

None

I I. SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY

Nowe ARPA (monitored by 01'P

13. ABSTRACT

Progress in deteruining mechanisms of processes fundanental tochemi1uminescence is reported vith pa~rticular ref erenee to the chemilbwimeentsystems (1) oxalyl chloride-hydrogen peroxide-fJluorescer, (2) acyl-oxalicaahydride-lyd~rgen peroxide-fluorescer, (3) 9 -ehloroesrbwlcridiwivm malt-hydrogn peroxide, (14) tetracy aneblene-hydrogem peroxid.-fluorescer.EXPloratorT searches for ney chemilumines cent materials are also reported.

3eS ts11'able COPY~

D D , JN.A 1473 UnclassifiedSecurity Classification

"hostl I l osenrow

1I4STRUCTIONS1. ORIGINATING ACTIVITY: Enter the name and address imposed by security dassification. using standard statemenitsof the contractor, subc intractor. grant**, Department of De- such as:eons. activity or other organization (corporate author) issuing (1) "Qualified requesters may obtain copies of this

the repot. report from DDC."2a. REPORT SECUNTY CLASSIFICATION: Enter the over- (2) "Foreign announcement and dissemination of thisall security classification of the report. Indicate whether rpr yDCi o uhrzd"Restricted Date" is included. Marking in to be in accord-rpr yDCi ntatoie.ance with appropriate security regulations. (3) "U. S. Goverrnment agencies may obtain copios of

this repo,-t directly from DDC. Other qualified DIC2b. GROUP: Automatic downgr'ading is specified in DoD Di- users shall request throughriective 5200. 10 and Armed Forces Industrial Manual. Enterthe group number. Also, when applicable, show that optionalmarkings have been used for Group 3 and Group 4 as author- (4) "U. S. miulitary agencies may obtain copies of this

12od.report directly from DDC Other qualified users3. REPORT TITLE: Enter the conmplete report title in all *hall request througthcapital letters. Titles in all cases should be unclassified.If a meaningul title .awinot be selected without classifice-tion6 show title classification in all capitols in pareinthesis (5) "All distribution of this report is controlled Qual-immediately following the title. ifited DDC users shall request through

4. DESCR13PTIVE NOTES If appropriate, enter the type of_________________report. e.g., interim, progress, summary, annual, or final. If the report has been furrushed to the Office of TechnicalGive the inclusive dates, when a specific reporting period is Services. Department of Commerce, for sale to the public, indi-covered. catir this fact and enter the price, if knownt.5. AUTHDR(S) Enter the nasm(s) of author(s) as shown on IL SUPPLEENTARY NOTES: Use for additional explana.or in the report. Eart last name, first name. midde initial. tory notes.It military, show reek and branch of service. The nam-e ofthe principal aa~her is as shooluto minumun rqirmet 12. SPOF4UDR1ING MILITARY ACTIVTY: Enter the ame of

the departmental project office or laboratory sponsoring (pal-6.REPORT DATL Enter the dat* of the repast as day. rrr4 for) the research and development. Include -A-& e.

msonth, year or month. yew.h It more, than one date appearon the report. doe date of publication. 13 ABSTRACT: Enter arn abstruct giving a brief suid factual

summery of the document indicative of the report, oes though7.TOTAL NUMBER OF PAGE& The total pages count it may also appear elsewhere in the body of the techaical re-

should follow normal pagination procedures. iLe., enter the port. If additional space is required, a continuation sheet alolnumber of pages containing information, he attached.7b. NUMBER OF REFERENCE& Enter the total number of It is highly desirable that the obstruct of classified reportsreferences cited &ai the report. be unclassified- Loch paragraph of the abstract shall ead withS. CONTRACT OR GRANT NUMBER: If appropriate, enter an indic-ation of the suliiary security classification of the ia-the applicable number of the contract or paent under which formation in the paragraph, represented as (TS) fS). (C). az U)

the rpo" ws writetThere is no limitation on the longth of the abetract. How-88. , & d. PROJECT NU.MBER. Enter the apropriate ever, the suggested length is from I1SO t-3 225 words.imilitary, dapartasat identification. such as project number. 1 E OD.Kywrsaetcnclymaigu emsubproject naumber, system mimbees, task number. etc. 1 E OD Kywri r ehial ennfltm

or short phirases that characterize a report end any be used an9. ORIGINATIOWS REPORT NUMBER(S): Enter the offi. irdes entries for cataloging the report Key words must becial report number by which the document will be identified selected so that no security classification is required. Ideui-said controlled by the originating activity. Tis number most fiers. such s equipment model diesipeation trade seame. ilitarybe unitque to this report project c.do name, geogrephic location. may be ased as key96. OTHER REPORT NUMBERS): l1ithe report hue bee words hut will be followed by an indication of tochiucal c-on.assigned any other report numbers (either bY the originabor test. ThVe assignment of links, tles. and weights is optionalo* by the sponsor), also enter tise number(s).

10. AVAIL ABL 7Y/ L 16 A117ON NOTICFSk Enter any lint-Nations on fumher dissermination of the report. other than those

Secuiarty Clasificatiogf

The folloving scientits ade prtimry contributions

to the technical effort:

M. M. Raubut

L. J. Bollyky, R. A. Clarke, B. 0. Roberta,

A* M. Semsel, R. H. Whi nmn.

Hicronal~yeas were carried out under the direction of

Dr. J. H. Deonarine; mass pectroetric analyses under the direction

of Mr. T. E. Mee; cmputer analyses under the direction .)f Dr. D. W.

Bebnken; infrared analyses by Mr. N. B. Colthup; other apectroscopic

work was superised by Dr. R. C. Hirt.

_ Cest GOP

990ttun I Plowttiou Nbhaaums to f'lum.iheenro.

Part A - traSLyI Porti.i rwmmiiwm1useacrmu . . . . . . . . . . .

Itffect of Water on Reaction Rates MQan mtum Y1.14. . . . .5

Vrfot of Pdrogen Peroxide on the Rat1a Rate and

l~fect of 1l'drogen %iuiae and DPA Concentrations onthe Reaction Rat* sad Quantum Yield * * e e . * * # * # . .15

Zffect of 2,6-Di-t-butyl-.-met)'lphenol (DMP) an theReaction Rate and Quantum Yield . . . a * * e . . # * . . .15

Part A -xperimental . . . . . . . . . . . . . . . . . . . . . IS

Part B -Cheailuminescence from Mixed Oxalic Anbydrides . . . . 20

Acetic-Oxalic Anhydride . * * a 9 . . * . o . . . 21

Bistriphez~1acetic-Oxalic Anhydride . * . * * . . 22

Diphez~lacetic-Oxalic Ankqvdride o e o . . a e * . 38

Pivalic-Oxalic AWhydride . . . . . . 9 e * & . 38

Part B - FEprimental . . . . . . . . . . . . . . . . . . . . . 39

Part C - Cheuiliuinescence Derived from 9-CarboperoxyacridineDerivative3 . . * * . o . o e # * * * e * * * **94

Part C - ftxlmental . . . . . . . . . . . . . . . . . . . . . 62

Part D - Tetracyanoethylene Chmlumnescence . . * . . . 0 . . 63

Part D - Roerimental . . . . . . . . . . . . . . . . . . . . . 66

teellug I! Naperlmeta I. ol m

'~~*Et~~t'~ . . .Pa e~. .~ .. s . .ab .i.t..y . .. . . . . . . *. . .

Re ~ ~ ~ ~ ~ IVm ntabige. fa -. . . . . 0

DhtrlbutLus . . . . . . . . . 6 o * o * * o 9 o o o 9 * o7

,lt) Irone in IteilhII5n mechanlimis rf" vcemnem

rdamentai to chealLwalneeenro t. repurt~i with partic.ular r.'ferenc.

to tha cheat itlineecent My.seMS (1) oxa1yl chlortde- hydrogen pernxide-

fluoreacer, (?) acyl-oxalic anhydride-hydrogen peroxide-fluorescer,

(3) 9-chlorocarbotWlacridial m salt-hydrogen peroxide, (4) tetracyano-

ethylene-bydrogen peroxide-fluorescer. Exploratory searches for nev

chemilisinescent materials are also reported.

The effects of reactant concentrations on (1) have been

determined quantitatively and unambiguously. Water in small concentra-

tions vas found to have a pronounced effect in increasing reaction rates

and qiantum yields. The quantum yield increases with increasing hydrogen

peroxide and fluorescer, vhile the reaction rate is not appreciably

altered. The qiantum yield is seriouruy reduced by a phenolic free

radical inhibitor.

Reaction (2) was found capable of generating quantum yields

on the order of 10%, and is thus the most efficient non-biological

chemiluminescence system nov known. Effects of reactant concentrations

on the quantum yields is discussed and stability data for several

mixed oxalic anhydrides are provided. Reaction condition. studies on

reaction (3) have shown that quantum yields of 3.5$ can be obtained in

dilute solution, but that the efficiency decreases rapidly with increasing

concentration. Investigations relating to the mechanism of (4) have

resulted in the discovery of two nev chemiluminescent reactions of moderate

efficiency.

Jbissiun ,f light in chemlludnescence an In fluorescence

re!t4 rro)m the transition of an electron from an energetic antibonding

orbital In an excited molecule to a stale bonding or non-bonding orb!tal

(generally the former) corresponding to the ground state molecule. Thus

a chemiluminescent process must accomodate the formation of excited

molecules as a product of chemical reaction. Tvo requirements for chemi-

luminescence are immediately apparent: (1) the reaction must liberate an

amount of chemical energy at least equivalent to the energ difference

between a product molecule and its excited state (4] to 72 q/mole for

emission of visible light) and (2) the product either must be fluorescent

itself or be capable of transferring its excitation energy to a fluorescent

compound present in the system. Many, if not most, reactions meeting

these requirements do, in fact, generate a low level, barely discernible

chemiluminescent emission. Moderately bright emission, hovever, is

limited to a very few reaction systems. Clearly a third requirement

exists that an efficient mechanistic pathMy must be available for the

conversion of chemical energy to electronic excitation energ. It is

also clear that this third requirment is rarely met.

Determination of this crucial mechanism for generating excited

moiecules is the primay goal of emil.minescence research. Once this

mechanism is understood, new chemiluminescent systems can be designed

having the efficiency and other characteristicc necessary for practical

lighting. Two approaches are being taken to achlevc an uie'andn . , uf

the chemical chemiluminescence mechanism. th f.. approach (Sectiur. 1)

involves direct mechanism studies of several known chemilumineseent

reactions. The second approach (Section II) involves exploratory studitrr

of new, potentially chemiluminescent reactions designed to test working

hypotheses regarding the chemiluminescence mechanism and to provide

structural criteria for chemiluminescent compounds.

To avoid excessive repetition, the objectives of a

particular study are described in detail only in the report where the

study ic begun. The progress of a specific investigattin ,'ai, heC f'1 .W-I

conveniently over periods longer than a single quarter by rererrltW ,

the Tables of Contents1l 2#3'4#5p6,7 -

NMMfON CUAW) IN CMaLwaCUC

PAN A

oUa l' poxite obmell±u imeee ,m is iluntted by the

roeutin ot oxaly1 chlortde with r4.rogen peroxide in an orgnic solvent

aontaia* a rluarescent ound

C l- - -el + loot-, CO + CO* + Ml + CI-9-el

te reactlun i uf mubstsantiaL interest became of the

Implied enera trasfer process vbereby chemioa. energy released by the

,tirimalwattun t peraidic Intermdat.es apears as sin" alet.itatLo

iberay In the r1ta.roscegnt copoundI. MD"Orver, the simplicity of the

wis tim* aterlalls aaW products ofer.s opportunity for detailed

motwialtic 1Isvestigaton in spite of the n- evident coplexty ofC~ho reacttin.

Aa dequ~te desription of the over-all mechanism requires

Iuwewrtl s riw.milntal questions dealing vith: (1) the chmical umschlm

f Ile. 1ee'wo 1nv')ilVrA uxaIlX1 hloride vhich leads to generation of

,I,.,i i ,.i tatL n nerly, sad () the Iechanis, of the process by

vwit,.t Itio t.pnrgy atjoears as the mingilet excited state of the fluorescent

, j' r. 'tr Pr, tor Is eiirrnt4y Investigating botn of these area.

Best t aable COPY

-5-

1. Effect of Water on Reaction Rates and Quantum Yields

Earlier work has indicated that water plays a major role in

the oxalyl chloride-ydrogen peroxide-fluorescent compound cheiiluminescent

reaction7. Indeed, the influence of water is such that we have encountered

considerable difficulty in reproducing reaction rate and quantum yield

results in experiments carried out at low water concentrations. The

maitude of the water effect has now been demonstrated urambiguously.

In Table I are sinnarized results obtained in experiments with constant

oxalyl chloride, hydrogen peroxide, and 9,10-dipherrlanthracene (DPA)

concentrations where the solvent ether was dried in various ways. It

is seen that reagent grade "anbydrous" ether is not nearly anhydrous

enough to avoid influencing the results. Even ether "dried" by passage

through an activated alunina column contains sufficient water to produce

an easily observed change in reaction rate and quantu yield. The

lowest rates and q'iantu yields were obtained using eter dried by

distillation over lithlum aluainum hydride under an argon atmosphere.

The water concentration in ether prepared in this way was below the

lhit detectable by the usual Karl Fischer titration. Dhvebrteless,

substantially different results ware cbtaLned from vo separately pre-

pared batches of hydride-dried ether. The difference between the two

batches was shown to be the difference in water con ent by experiments

(5) and (6 h vere water was deliberately added in concentrations large

enough so that the residual water content of the ether could be neglected.

An indicated in the Table, the rate and quantum yield results from the

two batches of equally wet solvent agreed within ejpeimental error.

-6-

TABLE I

The Effect of Ether from Various Sources on the ChemiluminescentReaction of Owclyl Chloride with K202 and DPAa

H2 0 conc- b K1 QuantumSource of Ether (moles 1:1) (sec- Yield

1. halinckrodt reagent-grade ankydrous ether 2.76 x lo 3.13 x 102 1.78 x 10"4

2. Mallinckrodt antrdrousether passed through a30 cM. A1203 (n*utral)column. 0.79 x 10-2 1.81 x 0-2 0.61 x 10- 4

3. b.llinckrodt ankhydrousether distilled overLAIH under argon. 0.39 x 10- 2 1. 9 3 x 10-2c 0.41 x 10- 4

4. all nckrodt ankydrousether distilled overLiA1I under argon -2atmosphere. < 0.39 x 10 113 x 10 2 c o.14 x I0 "4

5. Standard concentrationof aqueous ether added to#3. 4.96 x lO2 6.82 x 1o- 2 2.85 z 10-4

6. Standard concentrationof aqueous etkm added to 10 -2 2#4. 4.96 x lO2 6.58 x lo- 2 2.65 x 1O.-

(a) Reactions carried oyt with 2.50 x 10-3 N oxalyl chloride, 2.2 x I0-2 N

HgO2 , and 2.3 x 10" N DPA.

(b) Analyzed by Karl Fischer technique.

(c) The first minute of the log I vs. time plot fell below the extrapolatedlinear portion of the .' :V line.

-7-

Other results shoving the effect of added water on the reaction

are sum rized in Table II, and in Figures I and II. It is seen (Figure I)

that the pseudo first order reaction rate constant increases approximately

linearly with the square of the water concentration, and that (Figure II)

the quantum yield increases linearly with water up to a concentration of

about 3.7 x 10-2 solar and then levels off. The rate results might suggest

a rate equation of the form

dX 2- - k[X] + k[X], 2

dT

where k a 2.2 x 10 "2 see. "1 and kv a 18.6 1. mole "2 sece " 1. More likely,

however, the linear relationship in Figure I results from a coincidental

combination of several factors derived from the complex reaction mechanism.

The relationship between quantum yield and water concentration is of the

type discussed previously 7 .

2. WIfect of WAdroMn Peroxide on the Reaction Rate and Quar tum Yield

Reaction rate and quantum yield results from experiments where

the Wdrogen peroxide concentration wa vared independently of oxelyl

chloride, DPA, sad water are summarized in Table III. It Is evident from

the results that the reaction rate is independent of k'drogen peroxide

concentration. (The smLl ujiart trend in the pseudo first order rite

constant is alost certainly a consequence of residual water in the

~drrogen peroxide stock solution.) The quantu yield, however, increases

opproxImtely linearly vItb increasing t drogen peroxide concentra.ion

in the rane studied.

Results from a related series of experiments are siinerized in

Table IV.

-8-

Figure i

EFFECT OF WATER ON THE PSEUDO FIRST ORDER INTENSITY DECAY CONSTANT(DATA FROM TABLE 11)

-" I I I I I I I I I I I

12

II

9

'8

.7

S7 .4 ,r-[OxCl2 J -- 2.54 I0- _

oDPA] = 2.3x10- 4 M2 25" IN ETHER -

I L I I I I I I

5 10 IS 20 25 30 35 40 45 50 55

EH20J 2(x Io4 ) I

-9-

Figure If

EFFECT OF WATER ON QUANTUM YIELD(DATA FROM TABLE 11)

4

3

z

I2 3 4 56 7 8EN2 3 ( x 102)M

-10-

TABLE II

Effect of Water on Oxal1 Chlorde-ktrogn PerouideChomlinn.cencs ,

[Eo Ki 1 Quantum Yold(x 10, molar) (sec.' i9 (X 10'

< . 4b 193c (113d) 0.41 (.14d )

1.23 2.60 o.8o

2.46 3.53 1.59

3.69 4.75 2.33

4.96 6.82 (6.58d) 2.85 (2.65 )

6.15 9.04 3.-3

7.38 12.50 3.57

(a) The reactins were carried out vt.h 2.5 x 10 "3 N ozalyl chloride,2.13 x 10" EeO1 , and 2.3 x 10"_ DPA, at 25w-vbere the amuntof vater specifled wa added prior to injection of oxalyl chloride.

(4, lower lAit of analysis by Karl Faecr mthd.

(c) Log I vs. T plot vs# linear only %fter 30 se. of rewtlon.krliAr intensities w below the e po2.ated linear plots.

(d) Result fro separately prepared "anadrous ether".

(e) Log I vs. T plot vas linear after 15 sf. of reaction.

-11-

TABLE III

Effect of Hydrogen Perorlde Concentration on Oxalyl Chloride-Wdrogen Peroxide Chemiluminescence8

[H20] k1 Quantum Yield(x 102 Molar) (se. " x 1o2) (x 1o)

1.0 1.8 0.85

2.0 1.8 1.5

5.0 1.9 4.3

10.0 2.1 8.6

(a) Reactions were carried out with 2.42 x 10 - 3 X oxalyl chloride, and2.0 x i0-4 M DPA in alumina-dried ether at 25.

-12-

TABI IV

Effect of F&rogen Peroxide on Reaction Rate and Quantum Yieldin systemns Containi watera

(H202J QF

ix 102 mlar) (se.c' x L02 C:114 xi0' 4jj) QVT/QT

1.0 6.4 0.35 2.6 .59 4.14 5.1

2.0 6.5 0.66 4.6 .59 7.8 5.2

5.0 6.7 1.8 10.0 .55 18 4.2

5.0b 7.5 - - 6.6 1.5

10.0 7.6 4.o 13.2 .55 24 2.8

(a) Reactions verm carried out with 2.2 x 10- 3 oxalmyl chloride and2.0 x o- 4 X DPA in ether at 25". After 30 eec. of reactioq water

s injecte to provide a water concentration of 3.14-4 x 10' M.Q- a The fractional quantu yield observed before water injection;

-/ The fractional quantum yield observed after water injection;W v The fraction of total reaction measured after water injection;

Q T. The calculated lotal quant u yield for the reaction in thepresnce of wter; QA a The total quantum yield in the absence ofadded vhter.

(b) Water was added before oxalyl chloride injection to provide3.44 x 102 N watwr.

-13-

In this series hydrogen peroxide was varied in experiments

where a constant water concentration was introduced 30 seconds following

the onset of reaction. This procedure was adopted to minimize direct

reaction of water with oxalyl chloride. The observed effect of water

injection is shown in Figure III. As indicated in Table IV, the

fraction of total reaction measured in the presence of water (F W) varied

from 55% to 59% and this fraction wad used to estimate the complete

quantum yield of the aqueous system ( T). In agreement with the previous

series, it is seen that the hydrogen peroxide concentration has at most

a minor effect on the reaction rate, and in further agreement, the quantum

yield increases substantially as the hydrogen peroxide concentration

increases. A comparison of the k1 and -values with the quantum

yield and rate values in Table III, reemphasizes the effect of water on

the reaction rate and quantum yield. However, it is clear from the

ratios of QTW/QT that the influence of water on the quantum yield

decreases at higher hydrogeL peroxide concentrations.

Also indicated in Table IV is the result from an otherwise

identical experiment where the water was added prior to Injection of

oxalyl chloride. The observed quantum yield is not only smaller than the

calculated quantum yield for the experiment with delayed water addition,

it 1b even sml1er than the observed quantum yield (QwF) corresponding

to only 55% of the total reaction. The result clearly shows that at

the water and hydrogen peroxide concentrations Investigated, water

competes seriously with hydrogen peroxide at an early reaction stage to

divert a key reactant into a non-cheillninescent pathway. This "key reactant"

is probably oxa!yl chloride itself.

Figure III

TYPICAL DECAY PLOT SHOWING THE EFFECT OF WATER INJECTION2.2

I

"DPA 2 x 10- 4 M

-H202] 10x 10-2 M

0-OXCI23 = 2.42x 10- 3 M2.0

WATER INJECTED TO PROVIDE A,"CONCENTRATION OF 3.44x 10-3 M

1.8

z Log!

1.6I-

1.4

zw ANHY DROUS

z REACTION

1.2

1.0

20 40 60 s0 100TIME (sec.)

-15-

3. The Effect of IWdrogen Peroxide and DPA Concentrations Un the NLeati,.Rate and Quantum Yield

The results of a series of experiments where the yirotn ri, pr,,4L.

and DPA concentrations were independently varied over subaLntisa rane.

at approxiaately constant water concentration, are su arized in Table V.

It is seen that the reaction rate is essentially unaffected by the

htdrogen peroxide ol DPA concentrations. The quantum yield, however,

increases with increasing DPA concentration throughout the range stuillo.:

and increases with increasing hydrogen peroxide roncentration up to a

concentration of about 2.3 x 10-1 molar. These reaults are in essential

agreement with results of previously reported experiments6, " , where the

water concentration was not carefully controlled.

The Effect of 2,6-Di-t-butyl-4-methylphenol (DBMP) on the ReactionRate and Quantum Yield

The effect of the free radical inhibitor DMW on the reactlun

is indicated in Table VI. These reactions in dimethylphthalate solution

did not provide simple Log intensity vs. time plots, and lifetlims

are indicated with reference to the time required for the Intensity tu

decrease by 1/2 its maxirni value. It is clear from the aproimste

constancy of this half intensity period (as well as from the intensity

vs. time plots) that DBMP does not affect "he reaction rate. The quantu

yield, however, is drastically reduced with increasing DBP. As indirast

in the previous report 7 this result is not a consequence of fluorescence

quenching of the fluorescer by DBMP, and therefore sugests a in-rate a

determining free radical chain process. The possibility of interference

by an ionic route is unlikely, but han not been eliminated. Wo wvili

investigate this point.

-16-

14 J. K4) N

I .'

ax

q - F4 0 r4 P4 r4 4

%..0

I- tN o 0 c* t- A~

4A

9-44

0% ~ - 0 0 0 S S

'aa 6S O

* .*0 0 0

L a\oPj4 N '0 I . ' 4 4 .

0% 0- a - 'o a .4

A- N N '-4 *4 ANw x j0 00 ' t

A 5 a; S S

"4"4NP4

-17-

TABLE VI

The Effect of a Free Radical Inhibitor* on theOxalyl Chloride + H2 0 2 + Fluoreacer Chemi-

luminescence Reaction in DimetklphthalateSolution

Kolar Concentration First half-life Ft. Iamberts at

of Inhi :A-or in see. maximum intensity Quantum Yield

o 87 o.64 4.13 x IO2

0.67 x io-3 950..48 3.80 x 10 2

2.0 y 10"2 84 0.33 2.34 x 10' 2

1 x 10-2 65 0.09 0.65 x 10-2

1.5 x 102 73 0.08 o.48 x IO 2

ca. 2.0 x 10 - 2 68 0.05 0.29 x 10- 2

* 2,6-di-t-butyl-4-netkriphenol

DPA a 5 x 1C-4 M; H20 2 a 0.2 M; Oxalyl chloride a 2.42 x l03k.

The intensity during the first 2 nins. for all the above reactions fell below thelinear extrapolated log I vs. time plot.

-18-

SECTION I - PART A

MUMCRlENAL

Materials

AnkwrousEther. - (!ballinckrodt, Reagent Grade, Anhydrous) wasfurther dzi- -assag through a 30 cm. neutral aluina columnP, or bydistillation over lithium alumninum k~dride undier argon. Residual waterwas estimated by the Karl Fischer methodlO.

Oxalyl Chloride. - (Aldrich) was distilled through a 10 x 1cm. 1i rawx column under argon to obtain a fraction, b.p. 64" (1 atm.)(Lit.Li.,b.p. 63.5640 (763 im.). Standard solutions. of oxalyl chloridein ether or dimettylphthalate (1l11 x 10-1 N4 to 1.50 x 10-1 14) vere analyzedaravimetrically by conversion to oxanilide writh excess aniline. This reactionwas sh~own to be quantitative in preliminary large scale experiments.

Adrus Rydrogen Peroxide in ether vas prepared from 98%bydrogen peroxide (Becco Ceical Division, FMC Corp.) dissolved inether to a concentration of 25%p and dried by shaking 16 hours withexcess anlydrous magesiu sulfatel2. Water estimtion of the driedsolution by infrared analysis at 1640 cm. 1l indicated that residual waterwas below 2.9 x 10-2 N. Standard aolutions of tqdrogen peroxide (0.758 N4to 1.65 14) were prepar~ed from this solution anL ankydrous ether and wereanalyzed-iodowetrica.2Ly13.

9,10-Diph ewlanthracene (DPAJ (Aldrich) vas recrystaized fromabs. ethanol-chloroform, to obtain mterial, m.p. 250n-2510 (U4t. 1 , am.250-2511).

2,6-Di-t-buty'l-I-metkhvlphewl (Koppers Co. Air-) was sbiin vaco toD obtain material melting at 69-700 (lit.l3, &.P. 69-7O0).

Chisilimiescenc teriawr

In Leheral, the experiments were carried out by combiningapgropriate aliquo~t& of standard etberal solutions of khydroti peroxide,DPA and water with an appropriate volume of exuoer in a 3 &l. stirrecyliniri-al cuvette. The volume of ether used was chosen to nuke thefinal Volume 3.00 al. which exactly filled the curetto. The curettewas plaoced In the spectroradiometer and an appropriate aliquot ofstandard ethereal oxal4y chloride vas injected from an all-glass syringe.The intnsity of mission of a selected 5 up-wide wavelength span was,meaued as a function of time from the point ,' oxaJ41 chloride injection.The reaction temperature was 250 to within one aegree. The rudiamter,its calibration, aMd the calculation of rates and quantum. yields havebeen previously describedl,2?, 4 .

-19-

The experiments in Table I were all carried out with thesame standard solutions. The ether added to obtain the desired con-centrations vas treated a" described in the table and amounted to 83%of the f inal Volume.

The experimtents in Table IV were carried out by Injecting aconstant 0.20 al. of 5.16 x 10-1 N water in ether after 30 sec. ofreaction. "Fractional quaantum yiilds" from the ank~drous (Q7 ) and vet,(V/) parts of the experiments were calculated separately in the usu.alway?. The fraction of total reactions meaisured in the presence ofwater (NW) was estimted from the formula

Fw M 1--

where Q is th* quantu yield for the ank~drous reaction (Table III).

The total quantum yield for the aqueous part of the reaction

was estluted, from the formula

%T

-20-

SECTION I

PART B

Chemiuiminescence from Mixed Oxalic Ankiydrides

In previous reports we have described a series of efficient

new cheuiluminescent reactions based on the decomposition of monoperoxy-

oxalic acid and certain of its derivatives in the presence of a fluorescerv

such as 9,10-diphemylantlh-acene (DPA). While efficient, these reactions

are in general unsuitable for practical use because of their short emission

lifetimes. It is clear, however, that a reaction producin a wnoopernV-

oxalic acid derivative by a slov process, would accomoate tv b.al of

long-lived emission.

In the previous report 7 we described a series of exploratory

experiments, which exaned the possibillty of producxng o aop xyoxalic

acid derivatives by reaction of hydrogen peroxide with certain oxalic

acid derivatives including esters, anhbyrides, and Le is-base oxa l

chloride complexes. Results with triphezylacetic-ozaic anhydride were

particulArly encowagng, providing strong intensities vver exteadd

periods. A more detailed stuy of oxalle anb dride cmalumnasce.

is thus being carried out to determine the effects of electronic and

steric structural factors on arhydrlde stabill*, rate of reaction with

bydrogen peroxide, ese of hydrolyls, and cberziluinescence efficiency.

Three oxalic anbrdrIdes (Structure I) have been prepared

and exminmd in premiary fashion.

.-21-

Ia: RUtH3

b: P=(C 6H5) 3C

Ic: RNC6E5

The three ankdrides vere found to vary markedly in therma stability.

bydrolytic reactivity, and chimllu inescence efficiency. The poor

chemiluminescence results obtained fram Ic were discused in the

previous report 7 . Current results are smusrized below.

1. Acetic-oxalic anbqdride (IA) was readily prepared

from oxalic acid and ketene at -50"C. The product appeared to be

essentially pure and was stable over an extended period when stored as

a solid below its melting point of about -5". At higher temperatures,

however, the liquid anbrdride decomposed slowly to acetic anhydride

in agreement with earlier work by Edwards and Henly16 .

CH-.WCOO + mksI - CH3&O C1 3

IA

Cj:0ou % , CE 8E0s1 + CO,+ CO

IA

Therml decomposition of the pure liquid anbdride is

appreciable even at O and is complete after a few hours at 250C. owever,

the decomposition is significantly slow*7 in dilute solution. A study

of the decomposition In 1,2-diaetboxyetha solution was carried out

-22-

vhere the concentrations of IA, acetic anhydride ad acetic acid were

determined as a function of time by infrared analysis. The results are

smmarized in Tables I. As indicated in the Table, decositlon was

found to be catayzed by base (pyridine or potassium acetate) but was

not affected by free radical inhibitors.

Potential acid catalysis of decmposition is particularly

disadvantageous to the stability of IA since imli nowts of acetic

acid wll generally be present in 1A solutions as a consequnee of

hdrolysis. %n effort vas thus mde to stabilise IA In ether %W thm

use of excess ketene which -ould tie up acetic acid as acetic ankdrtde.

Substantial decomposition of ]A oecurred, however ev" under these

conditions, suWesting that IA is inherently unstable above its melting

point.

The instability of 1A in solution Me preveated quantitative

neasurement of its chabluminescene efficiency In reactioms with kdroqen

peroxide. Qualitative observations, hovew , Indicate that its cbmil-

luminescence efficiency is close to the U)% efflienc obtained with

1B (see below).

2. Distripb lacetlc-om c Anmu ft (a) me repared

frm poftstium trip lacetate a&M omlyl chloride In beimse solution.

2(c3~c . (c6w5)C3 c&=&(C6S) 3 + . 1

C655) 1%"l ___ (C6Y -2

-23-

-4 Vk o 0 mv ' 0 u

04r

0 0 0- H ul 0

c0 0 um 0

wo o c r 4 P4t- u

0 %-ocu ko

0 0fn 0 t- L^co W

OR 0

4A

0 ci

434

0 0

40#

Vj192

:it-00

-t 0

*p4 4 'F 'o p 4 I

a al

a 30aV4 p F-I r4 r404I ii

-25-

As indicated in Figure I, a prominent feature of the reactions

was the appearance of a variable but discrete induction period lasting up

to 19 minutes folloving admixture of reactants. The possibility that the

observed induction period was a consequence of a free radical inhibitor

present as a trace impurity vas investigated vith an experiment where the

inhibitor 2,6-di-t-butyl-h-methylphenol (DTW) vas deliberately added.

As indicated in Table III,hovever, the inhibitor appeared to reduce, rather

than lengthen the induction period, altL ugh the quantum yield was sub-

stantially reduced. Since DTBMP does not quench DPA emission 7 , the result

indicates tentatively that a key fast step my, indeed, be of free radical

chain character, but that the rate determining step responsible for the

induction period is an ionic process. Additional studies will be carried

out to bear on this point and to determine the cause of the induction

period.

The results of a series of rate and quantum yield measure-

ments, carried out in dietbylphthalate (DNP) are summarized in Tables

rV, V, and VI. The reproducibility of these results ts significantly

better than those reported above for reactions in 1,2-dimethoxyetbane.

Moreover, in contrast to the experiments in 1,2-dimethoxyethane, induction

periods vere usu.Uy absent in DNP except for several experiments in Table V1.

A typical light inteneity vs. time curve is given in Figure InI.

An Indeperlent infrared assay of the anhydride solution

vas prevented bf the strong absorption of the solvent DMP in the carbonyl

region. To minitize random variations resulting from varying IB concentrations,

each experimental set vas carried out from a single IB stock solution.

Comarimon of results from different experimental sets is of doubtful value.

-26-

Yields of crude product vere good, but considerable losses vere encountered

during purification steps leading to analytical grade material. The solid

product appears to be stable on storage for extended periods at room

temperature provided anhydrous conditions are maintained. In dilute solution

in 1,2-dimethoxyethane, however, infrared analysis indicates that IB is

slowly lost by conversion to tripbanyiacetic acid as indicated in Figure 1.

Results of other experiments bearing on the stability of

1B in 1,2-dimethoxyetbane solution are snmmwized in Table 3I. Triphezl-

acetic acid (TPA acid) apparently has little effect on the conversion of

IB to TPA acid, but pyridine initiates a rapid decomposition to tripheol-

acetic anhydride (TPA anbydride).

A preliminary set of quantum yield measurents has been

completed on the tripbemylacetic oxalic abbydride-bydrogen peroxide-9 ,10-

dipbemylanthracene (DPA) system in 1,2-dimethoxyetsne end in dimewtb2htalate

solutions. The results of experiments in 1,2-dimthoxyethane are susin zed

in Table III and a typical intensity decay plot is horn In Figure II.

Since the anbydride vas bydrolyzed readily by the traces ot moisture

present even in freshly distilled 1,2-dimethoqethane infrared asa of

the ambydride vas carried out on the stock solutions used in the experimta

to monitor the actual starting concentrations of the aabydride and its

hydrolysis product, triphenlacetic acid.

In spite of this independent concentration analysis, the

quantum yield and rate results shov moderate scatter. Quantum liolds on

the order of 5 to 9 per cent vere obtained, however, substantiating the

high efficiency of the system.

-27-

.- 0 0

cux

'o I

cm rr-4

0I

cu

CuY

r+- 4

x 0%'

'.4

I 00I

I I 0 co Ipvr- - 5 0c

U.

0Uz

z

49FA4z

0

z4z CD

w z

U'U

Z =.

00

PiIL -

o

0

QUq 4 N

Irlm :E0"I-I33O ivi~l0

0 0VAN 31VON3Va3d~lG V313IM dV

-29-

q0

LU

4 "

-30-

0 N NN cuIod 0o 0 0 0r4 H- f- r4 r4

(1 * u %D UI% F-4

o1* o4' '0 t-ortor4 r

x x m(Id 0j CI CII

I,4Nl

'-4 0~~ -

V44

4a 01

0

,r444

4 W*14a

Ne k

a i

C4 c N Nm N aV

Cy J 4 M F.4 0r4

0 (I5 0 ; 0 3 0 V0

cuI N C, eN N

00 0 0 0304 r4 r4 r4 4 r-4

A N ~ No N N x

1, 1114A 4 A

,4 N 4-)3r4-

L)*. 4c a.3

C-

uiC'

UL

tuw

U P-

J ~ SIf MW~I)AjKJ

-32-

0

4S

'4 S C", C1 \ C

H0 0 0 0k 4 r 4 H4

4A v

V 4 A

4.)

0 : 0 .4I 8

v4.

V44

4AM)

10

A; A* A 0I0 1 0 1 4 6

-33-

H4 r-I 00 0 000 00004rI - q - r

Cu %00FIc 004 o% oc

00

9-4

mcm 0 V% irA UN0 0irLir U AL 00003

V94

0 El

r- Sfr0

- I -

.r0t

6 0

r4.4.4 a )

#24: 6 4 -0' d4 36 3 40

1 0

b 0 101- r 4 9.4

z c a31I.-

%.jU 61'6.

I

r -4 cuc UC UC

43 ~ ~ I Ic I~~I

a '4

0

r,'A

cu en rA4 on

.4 060 6 1r4 -4 r- r4 4 r4

* od -*

400'in 00

0i'

900 0 10 10

0'4 Pjor (idr

"3 I Ti

-35-

Experiments 8uarized in Table IV indicate that oxygen

and smal amounts of water nave negligible effect on the reaction, while

the oxidation inhibitor sodium diethyldithiocarbamate extinguishes emission.

As indicated in Table V, increasing hydrogen peroxide con-

centrations above 5 x 10 - 2 M result in a substantial decrease in quantum

yield. This is in sharp contrast to analogous experiments with oxalyl

chlorde-brdrogen peroxide reactions where the quantum yield was found

to increase with increasing hydrogen peroxid, in this concentration

reange. The result tentatively suggests the mechanism indicated below.

1) R9091g + H202 - ) ON)OOR + JgOE

2) OOHA R gOH + 2C0 2 + DPA*radical decomposition

3) RgOL E0 + R202 R0 + R0OMO0N

) IOuKI.cO - Non-chemiluminescent decomposition.

The first step of the mechanim is almost certainly

the formstian of the monoperomyoxalic acid derivative indicated. Additional

bhdrage pewroxde would be expected to react with this intermediate as in

step 3 to farm 4iperoj€zalic acid. Results from oxalyl chloride-bdrogen

perozade experiments (Section IA) indicate that under the a&vhdrvou con-

di;.ions ume in ,hese experiments, diperoxyoxalic acid is essentially

non-cnemil~uanescent. So we tentatively suggest that the peroxyoxalic

acid derivat ,ve, itself say be capable of concerted multJIple bond

cleavage dec iosi .on (step 2) which provides the excitation energy needed

for libt cams on.

-3

The effect of increasing DPA concentration on quantum

yield is indicated in Table V. The observed increase in quantum yield

is analogous to the effect of increasing fluorescer concentration in

the oxal yl chloride-bydrogen peroxide system (Section IA). Rovever, when

rubrene L used as the fluorescer (TaLle VI) the quantum yield is essentially

unchanged with rubrene concentration in the concentration range studied.

Additional experients vill be carried out to determine mre precisely

the effect of fluorescer and fluorescer concentration on quantum yields.

As indicated in Tables IV and V, small quantities of

aaditional vater have little effect on the reaction in di.et~lphthaeate

solution. Excess vater, hovever. seriously reduced the quantum yield

as indicated in Table VI.

Initial efforts have been mde to identify the chemcal

products from the triphalacetic-oxalic anbydride reaction with bydrogen

peroxide in dlmethlphtkalate. Infrared analyses of liquid phase products

Indizate that triphaelacetic acid is the predomina t reaction product,

althouh quantitative analyses have not yet bean carried out. Results

frm quantitative anmnitric and ms spectromstri analyses of gaseous

phase products are sumrized in Table VII. It is evident fr the

table that the predminant gaseous product Is carbon dioxide aon vith

a substantial mount of carbon unoviae. 1wwver, the total gaseous

product observed accounts at best for only about 33% of the oxalyl carbon.

The quantum yield data sumrized in Tables III, IV, and V demostrates

that the triphaeylacetic-oxalic a badride-bydroen peroxide system is

-37-

ou~i 'a C .- ~ cc0. U

4 .C

00

0~ ) 0\Q xEl CUCZm

00

.93

4

UN

_ 0

r-4)

6-0, C CO

mO r-4

-38-

far superior in efficiency to any previously known non-biological

chemiluminescent reaction. Additional work is planned to further define

optamum reaction conditions and the effect of reaction variables on

quantum yields.

3. Diphenlacetic-oxalic anhydride has been prepared in

an admixture with diphemylacetic acid by reaction of diphenylketene with

oxalic acid. The reaction proceeded sluggishly, indicative of the relatively

low reactivity of diphemyl ketene. The anhydride-acid mixture provided

chemiluminescence eussion when reacted vith hydrogen peroxide in the

presence of DPA. Other studies are in progress.

4. Pivalic-oxalic Anhydride. - Attempts to prepare this

anbydride from potassium pivalate and oxalyl chloride or from pivalyl

chloride and sodium oxmlate have provided discouraging results. MLxtures

of pivalic acid and pivalic anhydride were cutomarily obtained, vith no

indication of the formation of the desired mixed antdride. The formation

of substantial quantities of pivalic anhydride in these experiments

suggests that pivaeic-oxalic anhydride may undergo a decoposition

analogous to the decomposition repo'ted for acetic-oxalic anhydride.

-39-

SECTION IB

EXPERIMEMAL

Diacetic Oxalic Anhydride (IA)1 6 ,1 7 - Ketene, prepared bythe pyrolysis of acetone8 at the rate of 0.1 moles ketene per hour, vmsbubbled through a stirred solution of 4.5 g. (0.05 moles) of anhydrousoxalic acid in 100 ml. anhydrous ether at -50* under argon during 1.25hours. The reaction solution was then evaporated to dryness under reducedpressure at temperatures below -5". A white crystalline product w,obtained, whose infrared spectrum vas in agreement with that expected.Bands characteristic of acetic anhydride or acetic acid vere absent fromthe spectrun. The product was found to be very sensitive to moisture andto decompose at temperatures above -5o .

In a similar experiment, following completion of thereaction at -5oP the reaction mixture was maintained under a ketoneatmosphere at 00 for 5.5 hours, and then analyred by infrared spectroscopy.The resulting solution was found to contain mainly acetic ankdride togetherwith small amounts of diacetic oxalic anhydride and acetic acid.

Bisdiphemrlacetic Oxalic Anyd.nide - A zolution of0.27 g. (0.003 moles) of oxalic acid in 25 i. of ndrous ether mseadded dropwise under argon to a stirred solution of 1.14 g. (0.006 moles)of diphenyl ketene19 in 30 ml. of petroleum ether. The reaction wasnstirred 2 hours at 25; then 2 drops of concentrated sulfuric acid wasadded, and the reaction mixture was stirred 60 hours at 25. The reactionmixture was filtered under argon, and the filtrate was evaporated to Irynessunder reduced pressure to obtain 1 g. of product which van shown to be,by infrared analysis, a mixture of 30% desired product, 60% dipheiq1acot1racid, and 10% unreacted diphemyl ketene.

The product of the reaction was found to provide chesi-luminescence when reacted with bydrogen peroxide and DTA in l,2-dimotboy-ethane.

Potassium Pivalate - Potassium pivalate was prepared byadding 13.2 g. (0.234 moles) of potassium hydroxide to a solution of20.4 g. (0.2 mole) pivalic acid in 500 al. of absolute ethanol. Thesolvent was reduced to 50 ml. and 25 al. of benzene was added. Thevolume was further reduced until a solid began to precipitate. Thesalt was obtained in a yield of i4.8 g. (62%) m.p. > bOO" The Infraredspectrum was consistent with that expected for the product.

-40-

Attemupted Preparation of Dipival. c-Oxalic Anl~dride -A s~penson o Tj7(0.2 mole) of potassium pivalate in a mixture of

1,0 m-1. of 1.2-dimetboxyethane and 50 al. benzene vas heate to ref lux,and the suspension become gelatinous. The mixture ms then cooled to0* and oxalyl chloride, 16.8 al. (0.2 mole), was added. There was asubstantial evolution of gas as the addition proceeded. The mixture wasfiltered and the solvents removed, leaving a white, -mV solid. Thesolid produced a deak, short-lived Lig~ht when added to 5 al. of 1,2-dimethoxyethan containing a few mg. rubreneo, 4 drops 90% kdrogenperoxide and a pellet of potassium bydroxide. Subsequt infrred~analysis indicated that the product vas primarily a adztue of pival! .anhydride and pivalic acid with a trace of oxaLyl chloride.

Results of similar attempts to prepare the mixed anhydrideamre sinrized in Table VIII. AU were mrked by a substantial gS"evolution as the oxalyl chloride vas added.

Cue attempt vas made to prepare the anhydride from8.-01. g. (o. 06 mole) of sodium oxajate (which had been dried in vacuoat 500) and 6.03 9. (0.05 .ole)of pivalyl chloride in 50 al. of distilledhexane at 00. The starting usteriala were recovered unchanged.

Infrared Stu# of the Recamposition of Diacatic 0xali'APZjd!e - The experiments in Tables I& L'nd Th were carried out withsolutions i~f diacetic oxalic anhydride of varying concentrations,,prepared under argon in 1,,2-dinethozyothea freshly distilled overLtAlN. Solutions of reaction comonents vwr comined In flasks attachedto mercury manifolds and fitt:,, with serum stoppers for the withdraualof viamples. AUl ~Las9we v&a .horougbly cleaned, and drie. In anoven at UO0 for at least 1 hour. The flasks containing the reactionmixtures were placed in a eoastant temerture bath, and m&U aliquotsof each samle vere vithdravn pi odically a&M ainalyzed by infaespec troscopy for the dl sperme of diacetic oxalic ankdr1Ae bands so.1835, 1795, al 1767 ca.- =,adfor the apoace of on acetic oftdrideband at 890 cm.-l a&W an acetic acid band at 1740 cm.-l.

Infrared Stu of the R!mnition of Distriphoqlan~ic

bistriphe~lsectic omlic anhydridt ia l,,2-dimwtb~aethaw vas prepareduner argon with freshly distilied solvent. 7M sealm meItrdeinto a flask attached to a mrc-wy mifold &Md fitt , ith a earn stopperfor the wIthdrawal of .a~le.. other reactats were added when d&Glred.AD glassware vas thoroughly cleaned, amd dried in A oven at UO0 forat least 1 hour. The flask containing the sWlo ea placed in & oomstaat

Od 0 0 4 oV4 V 0 0*P

~ 00

0 0 0i o

0 0 0 10 0 0E-4 0U 0~ 0m 0 0

04 t-.

IC0

V I4 ~ -

temperature batb, and smll aliquots vere withdrawn periodically andanalyzed by infrared spectroscopy for the disappearance of the bistri-phemy1acetic oxalic anhydride band at 1760 cm.-l, and for the appearanceof a triphenylacetic abdride band at 1805 cm -1 and a tripheroylaceticacid band at 1730 cm. !n.1 The non-catalyzed decomnposition vas carried outtwice.

The absorptivities determined in these experiments werelinearly compared to absorptivities determined on standard solutions oftriphemylacetic acid and standard solutions of triphenylacetic-oxalicanhydride in l,2-dimethoi~ethane to determine absolute concentrations.The standard absortivities were: triphenlacetic acid at_1740 cm. -1505 1. mole-1 cm -, triphemylacetic anhydride at 1830 cm.i 1 370 1.mole1 cm.-l. The data for the experiment in Figure 1 are sunwrized,in Table rX.

Chemi1luminescence Quantum Yield and Reaction RPate Determin-ations for the Reaction of Trip e lacetic (halic Anyrde vith drg

ethane Solutions - Dry 1,2-dinetboxyethane vas prepared by distillationunder argon from 21thiu. aluminm h~ydride. A stock solution (3 x 10-2 K)of tripherylacetic oxalic anhydride was, prepared and transferred underargon to a serum-capped bottle to prevent reaction with atmospheric moisture.Stock solutions of 5 x 10-3 M DPA and 1.5 N H202 in l,,2-iethOxytewwwere also p-"pared. Aliquots of the anhyduide solution were vithdravu, bysyringe and combined with aliquots of the DPA solution in OP. rtdicmetercvi"o te 1 '2 . The wavelength of the radiometer was, set at 43J 1 atd the*1' opeued. The recorder vas started and aliquots of hydrogeu pertoxide

were injected from a glass syringe into the mongnetically stirre solutionin the cuvette.

Quantum yield calculations vere made analogously to thosedescribed previously for the oxalyl. chloride reaction2 . Results am,sawLrzed in Table III.

CbeNiluiinacone Qu~antu. Yield Mesrwta for Triphowl-acetic-Ozalic Auj!do &c4 ea Peroxide, Fluaruscer Reactions in 1

Ii*~it@ SonstO - iieWphAlate distilld from lti lmmkwd&ride vas used to prepare stock solutions of the ankydride,, hydrogenPeroxide and fluorescer. All solutions vore prepared and transferred tostrnu-capped bottle* In w dry box to minimize b*ydolyis. Aliquots ofthe reactants were traziaferred to the aegmtically stirred cuvette bysyring, 'with hydrogen Peroxide being the final solution added. The deesycurves and spectral distributions ~eerecorded And the quantu yieldscalculated as described prwriously . Both 9,lO0-diphenylanthrecene amdrubrone were used in different sets of experients as fluorescent acceptors(See Tables IV, V, azw VI).

.143-

TABIZIrX

Triphe~lacetic-xalle Ank~duide Decomposition Inl,2-Diethomyethane Solution as a Fimction

of Time

Time in B~ursfrom Preperation of Anhydride Concentratioa Triphexrlacetic Acid

__Solution (molar) Concentration (molarb)

0 2.55 x 10-2 0.89 x 1-

5 2.29 x io2 1.41 x 10-2

20 2.26 x 10-2 1.55 x 10-2

26 2.09 10-2 1.6o x l-

30 2.06 x 10-2 1.75 x 10-2

482.03 x 10-2 1.75 x 10-2

94 1.91 x 10-2 1.98 x 10o2

a V'amured by I.Jt. absorbance using 1830 Cm.'l for calculation.b JNasured by I.R. absorbance using 1740 cm.-' for calculAtion.

Solubility of Carbon Dioxide in Dimetylphthalate (DMP) -The apparatus (see Figure IV) used consisted of a 300 al. reaction vesselto which was attached either a 300, 500, or 1000 ml. gas reservoir. Asimple differential mercury manometer was included and could be attachedto either the reaction vessel or the gas reservoir. Evacuation of thesystem could be accomplished by means of the oil pump. Agitation was bymeans of a magnetic stirrer included in the reaction vessel. All volumesin the system were determined by calibration with water.

In a typical determination, 100 al. of freshly distilledanhydrous dimetbylphthalate was placed in the reaction vessel andthoroughly degassed via the oil pump to a pressure of less than 1 mm. Hg.The gas reservoir was evacuated and then charged to a predetermined pres-sure with carbon dioxide gas. Carbon dioxide was admitted into the reactionvessel and the uptake of the gas followed by the change in pressure indicatedby the manometer. When no change in pressure had occurred over a 2 hourperiod, the system was considered at equilibrium. The Henry's consantwas calculated from the known pressures, temperature, and volume,, ofthe system.

Henry's Constant for CO2/DMP System

Pressure (in. Hg.) Free Vol. Temp. Dissolved CO ka

Initial Equilibrium ml.) 9C (oles/l,.x ) (x 10 "4)

429.o 338.4 739.1 24.8 3.603 5.75

559.3 476.9 1291.2 25.7 5.678 5.14

a) k a P(m. Hg. at equilibri i)/ mole fraction dissolved CO)

Gaseous Products :r-i Triphe!Zlacetic-OxLlic AndrideReactions with * n Peroxide in Dimet!Z1hthaliate (DMP) -

apparatus used to measure the gaseous products from bistriphoWl-acetic oxalic anhydride was that of Figure III except that the gasreservoir (B) vas replaced by a pressure-equallzed dropping funnel whichhad been calibrated with ater. All of the glassware was care.i cleanedin a hot chrome-sulfuric acid wash, rinsed well in distilled water, anddried thoroughly at UOOC in the oven before use.

In a typical determination, alIquots of auhydrous bydrogenperoxide in ankydrous dimethylphthalate (DMP) and, where desired, rubreneor water in MP, were placed in the reaction vessel and cooled to OC byan ice bath. An aliquot of the ankydride solution was placed in the droppingfunnel and the entire system flushed with argon and then degassed to lessthan 1 m. Hg. via the oil pump. The ice bath was removed from the reaction

-45-

vessel and the anbydride soLution added to the peroxide solution. Gasevolution was folloved by the Increase in pressure in the reaction vessel.When no pressure increase bad been observed for a two hour period, thefinsa pressure was recorded and the Sas phase analyzed by mass spectro-scopy. The total reaction time vas about 18 hours. ]Results of the gasanalysis are sumrized in Table X.

-46-

0 0 r-4 -1 fnc

U

44

S Y~uN% a -1

45 004

v4 ~ 6\~

0 cV14 *- I~-cJ'4

1 4o t '- -,0 1% 3.

00 0

U'% CU Y!

W 00i 43 0 Q

0 . . -

%~- CU

A U'

001

0 9 u

0 --

cib

II 'I I II OA

00

ZECTION

PART C

Chemiluminescence Derived from 9-Cyrbo e aocridine Derivatives

Previous reports have described new chemtlufinescent systems

based on reactions of acridinium salts I7 and 116 with aqueous hydrogen

peroxide.

*

H3

-II - C1

Cl ClI II

Evidence vas provided in the last report 7 indicating that the chemi-

luminescent process involves the mechanism diagpaaed belov.

Cl" + IB20

N N N N Cl + Mj.

OCR2

*3------E2

+ W l C

1 00

OO 1+ N

III IV

VI

+ *C02 * B20

N O

Indeed the reaction was designed on the basis that pseudo base IV might

be capable of concerted multiple bond cleavage decomposition and thus

provide the substantial energy required to generate excited V as a

reaction product.

We have nov carried out a series of preliminary quantitative

light measurements to further characterize the reaction. The spectral

distribution of chemiluminescent emission from reactions of I with

aqueous hydrogen peroxide was found to vary substantially with the

initial concentrations of I as indicated in Figure I, where the spectra

are "normalized" to a comon intensity at 520 mp. The spectral change

with increasing concentration is evidently a consequence of reabsorption

of emitted light by the acridine derivatives present. The spectral

distribution was found, however, to be essentially unchanged with time

in a given experiment. The ctekiluminescent spectrum obtained at low

concentrations of I, where reabsorption (and reemission) would be small,

is compared with the fluorescent emission from l0-metbylacridone, V,

and from reaction by-product 9-carboxy-l0-metiylacridinium chloride, VI,

in Figure II. It is seen that the spectral match with VI is poor, while

spectral match of the chemiluminescent light with V fluorescence is

fairly good in spite of the different solvents used. (V is poorly

soluble in water under the chemiluminescent conditions.) The spectral

correspondence of chemilminescent emission with V fluorescence strongly

supports the astign ent of excited V as the primary emitting species in

-50-

NORMALIZED INTENSITY

C

m

z

z m

i

z rU' C

m d wi Liu

mA-4-

CD i

z 0

I)

-0 j

UJ U.

Imw

u >w

z

IC

CD C

LUS3N OSSWaZlWO

-52-

Notes to F'igure I

o Chemilumiinescence emission from 2 x 10-4 M V in 1 M R202 .

Fluorescent~e emission from 1 x lo-3 m vIII in ethanol containing0.02 M HCl.

*Fluorescence emission from 1 x 10-3 M IV in water.

3\Ja~\ao COP'

-53-

the chemiluminescent reaction. As noted in previous reports, however,

we never regard spectral correspondence between broad emission bands

as conclusive proof of the identity of emitting species.

The proposed chemiluminescence mechanism requires the initial

formation of peroxyacid III which is followed by a pH-dependent equilibrium

with key intermediate IV. In the previous report7 we showed that formation

of III from I and hydrogen peroxide in 1,2-dimethoxyethane is a slow

process, essentially complete only after 1 hour. To ex, -. the stability

of III and the proposed equilibrium step, a solution was prepared from

I and 90% ydrogen peroxide. The solution was essentially non-chemi-

luminescent under the conditions used, but provided strong chemiluminescent

emission on dilution with water. The chemiluminescence yield and light

decay rates of the diluted solution were determined under invarient conditions

as a function of the "ageing period" of the 90% hydrogen peroxide solution.

The decay plots are pictured in Figure III, and the results are sunmarized

in Table I.

As indicated in Table I, the chemiluminescence efficiency increases

sharply as the solution of I in 90% bydrogen peroxide stands, reaching

maximum yield at about 0.5 hours. Even after short ageing, the quantum

yield is higher than that from an equivalent experiment where I was

dissolved directly in 1.0 M hydrogcn peroxide. This effect is best

accounted for in terms of the following competitive processes.

_est e COP'

TABLE I

Chemiluminescence Efficiency of Aged,Slutions of I in 90%ffdrogen Peroxide(a)

Half-Intensity QuantumAgeing Period Period Yield Maximu= Brightness

(hrs.) (sec.) (x 102) (footlamberts from 1.0 cm. thickness)

0.Ob 35.0 1.3 0.08

0.1 22.5 1.9 0.15

0.5 c 22.5 3.5 0.24

1.0 c 25.0 3.4 0.22

1.5 22.5 3.2 0.26

2.0 25.0 2.8 0.21

3.o4 35.0 1.9 0.08

3.5 22.5 2.6 0.21

4.25 22.5 2.3 0.19

20 40.0 0.71 0.03

44 58.0 0.23 0.007

(a) A solution of 3.0 x 10-3 M I in 90% H202 was allowed to stand at 25.

Aliquots (0.1 ml.) were withdrawn periodically and diluted with 2.9 ml.

45% (by volume) aqueous t-butanol to provide chemiluminescent solutionscorresponding to 1.0 x io-4 M I and 1.0 M H2 02 .

(b) Solution prepared directly by dissolving I in 455, t-butanol containing

1.0 M H202.

(c) Reaction mixture contained 1.0 x 10-2 M 2,6-di-t-butyl-4-methylphenol.

(d) Diluted with water.

nest tkva'able COPvI

REkCTtONS O)F SC6-jTio4"S -r 11 V;" 0,0: AFTEP DILUTIGN RITH 4S' AQUEOUS BUTANOL

REACTION TIME IN 90'Tt422

70 I 05S HRS.2 1.0 t

654 2.05 3.S6 4.25S60 7 208 44

S5O

~4S

w4 0

EC

u.350

-~3 0zwAI.-z

4ulj20

is

10 A

I5 30 45 60 75 90 10 120 135 ISO 16S IS0TIME (sec.)

-56-

R3

*3H3 ~B2 02 N *

H3Cl III

H20 7I Cl-• C

H

VI

Thus, in 90% kadrogen peroxide, I is converted predominantly

to III provided sufficient time is allowed for the relatively slow process.

But in the presence of excess water an appreciable fraction of I ic con-

verted to non-chemiluminescent VI, reducing the chcmiluminescence efficiency.

As indicated in Table I, the cheuiluminescence efficiency

decreases only slowly with time beyond 0.5 hours, ;resumably the result of

a slow non-chemiluminescent decomposition of III. F en after 3.5 hours

the efficiency was sti1. 74% of that observed after 30 minutes, and an

appreciable emission was still observed approximately two days later.

The highest quantum yield observed in the experiments was

1.5%. This efficiency is of comparable magnitude to that obtained.in

oxalyl chloride-hydrogen peroxide-rubrene chemiluminescence 6 and is

substantially higher than the 1% quantum yields reported for 3-amino-

-57-

phthalhydrazide (luminol) chemiluminescence2. It is also evident from

the Table that higher chemiluminescence yields were obtained in aqueous

t-butyl tl.ohol than in water.

Several attempts to inhibit the reaction with the free

radical inhibitor 2,6-di-t-butyl--methylphenol are reported in Table I.

It is evident from the results that no appreciable inhibition or yield

loss was observed; thus, the reaction does not appear to involve a

free radical chain mechanism. This result is in sharp contrast to

free radical inhibition experiments with the oxalyl chloride-hydrogen

peroxide reaction (Section I A) where 2,6-di-t-butyl-4-methylphenol roducel a

marked reduction in efficiency.

A second series of preliminary experiments wa- tarried

out to determine the effects of reaction conditions on the quantum

yield and light decay rate. In these experiments I was treated with

aqueous hydrogen peroxide under a variety of conditions and the absolute

emission intensity was determined as a function of time. The results

are summarized in Table II.

Plots of log I vs. time were linear for three of the

experiments, which were carried out under pseudo first order reaction

conditions with an excess of hydrogen peroxide in an acetic acid-

sodium acetate buffer at pH 4. In all of the experiments, however,

except at the highest concentration of I, the plots closely approxi-

mated straight lines. Typical plots are shown in Figure IV. A

comparison of the quantum yield values in Table II indicates that (1)

the quantum yield increases substantially . h increasing hydrogen

-58-

TABLE II

Reactions of I with Aqueous Brdrogen Peroxide(a )

[I] [H2 02 k1 Quantum ield Half-life

___les 1:1) (moles 1:1) (see. " ) (x 10) (sec.)

5 x 10-4 8.6 x i0 - 2 1.21 x 10-2 9.8 57

2 x io-3 6.6 x 10-3 1.1 x 10 - 2 b 0.53 64

2 x 30-3 8.6 x 10-2 1.23 x i02 4.2 55

2x 10- c 8.6 x 10-2 8.9 x 10"4 1.0 780

2 x l0"3 8.6 x i0-1 2.44 x 10-2 10.6 28

6 x I0"3 8.6 x 1o-2 1.0 x 10-2d 1.5 70

(a) Reactions run ii acetic acid-sodium acetate buffer at pH 4 except

where noted. Reaction temperature = 25*C.

(b) Linear Log I vs. T plot after 90 sec.

(c) Reaction medium was unbuffered water.

(d) Linear Log I vs. T plot only after 120 sec.

-59-

Figure IV

REACTIONS OF I WITH HYDROGEN PEROXIDE IN AN AQUEOUS ACETATE BUFFER AT PH 4

2.6 1

2.4 1IMEASURED ATA 475 mu

2.2 [1] = 2X 10- 3 M

LH202] 0.86 x 10-1 M

2.0

1I.8

zS1.6

1.

1.01.4

0.

0.6

I2 3 4 5TIME (min.)

-6o-

peroxide concentrations (2) the quantum yield decreases substantially

with increasing concentrations of I and (3) that higher quantum yields

are obtained in buffered solutions at pH 4 than in unbuffered solutions

where the limiting pH is 2.4.

The effect of hydrogen peroxide on quantum yield is

easily understo-I in terms of the competition between hydrogen peroxide

and water for I as discussed earlier.

Moreover, the increase in reaction rate that is obsertred

only at the highest hydrogen peroxide concentration studied suggests

that at lower hydrogen peroxide concentration at pH 4 the rate is

determined primarily by the rate of the side reaction discussed earlier

between I and water and that at the highest hydrogen peroxide concen-

tration, hydrogen peroxide competes successfully with water so that

a substantial fraction of I is converted to III. Thus, with increasing

hydrogen peroxide the rate of disappearance of I increases and the

steady state concentration of III and IV increases causing the overall

rate increase with higher intensity and quantum yield.

The decreasing quantum yield with increasing concentration

of I results in part from serious reabsorption of primary emission at the

higher concentrations (See Figure I). However, reabsorption does not

account for the entire jield loss. The quantum yield for the experiment

in Table II with 6 x 10-3 M I was approximately corrected for reabsorption

by recalculation in terms of the spectral distribution obtained in the

.61-

experiment with 5 x 10 - 4 M I. This approximation does not take into

account fluorescent emission by the reabsorbed light at the higher

concentration and thus gives an erroneously hJgh quantum yield. In

spite of the h bias of the correction, the corrected quantum yield for

the 6 x lo-3 M experiment was substantially lower than the actual quantum

yield obtained at 5 x 10- 4 M I. The quantum yield values were: measured

yield at 6 x M0"3 m I = 1.54 x 10-4; corrected yield at 6 x i0 - 3 M I =

4.5 x i0-4; measured quantum yield at 5 x 10 - 4 M I = 9.8 x 10"4. Thus

fluorescence quenching of the emitting species or non-chemiluminescent

side reactions involving intermediates derived from I reduce the q-J.um

yield by a factor of at least two at the higher concentration of I

As indicated in Table II, buffered reactions at pH 4

were substantially more rapid than corresponding non-buffered

experiments. This is in agreement with previously reported qualitative

observations7 and is in agreement with the proposed equilibrium between

III and IV.

-62-

SECTION I - PART C

EXPERIMENTAL

The instrumentation, its calibration, and its use fordetermining fluorescence and chemilum'nescence quantum yields havebeen described in earlier reportsl,2,

6.

Experiments with Solutions of I in 90% Wdrogen Peroxide

A stock solution containing 0.0438 gins. of I (3 x I03 x)in 90% H202 was prepared. At appropriate intervals of time, 0.10 ml.of this solution was injected into the magnetically stirred cuvette con-taining 2.9 ml. of a 45% (V/V) tertiary butanol solution in water. Thedecay of the 440 mp band and a spectral distribution were recorded. Theresults are suarized in Table I and Figure III. For the experimentswith inhibitor present 2.9 ml. of a 45% (V/V) aqueous tertiary butanolsolution containing 1.0 x 10-2 M. 2,6-di-t-butyl-4-metiqlphe~ol wasused as the diluent.

Reactions of I iith Erdrogen Peroxide in Aqueous Buffer Solution of pH4

A pH 4 buffer was preparedi containing 1.7 x 10"1 Nsodium acetate and 1.0 M acetic acid. Appropriate quantities of I wereweighed into 25 ml. voluvetric flasks so as to provide after dilutionthe concentrations reported in Table II. Aliquots of 30% H202 wereadded to the flask and the solution immediately diluted to 25 ml. withthe buffer. After mixing, a 3 ml. portion was t ransfered to the cuvetteand placed in the radiometer. Spectral and decay plots were recordedas described previously 2 .

Figure I shows the effect of reabsorption mon the emittedlight and Figure IV ahovs typical decay plots of the 475 np band withtime. Table II summarizes the quantitative data obtained for this setof experiments.

-63-

SXION I

PART D

Tetracyanoetbylene Chemiluminescence

As described in the previous report 7 , the reaction of

tetracyanoetbylene I with alkaline, anhdrous bydrogen peroxide in the

presence of a fluorescer such as rubrene provides a moderately strong

chemiluminescent emission. Cyanates, carbonates, and bicarbonates were

identified as the principle products. It seemed possible that the

chemiluminescent emission might be derived from one of the steps in

the following sequence:

(1) (rC) 2 cc(cN) 2 + H202 0 (AC) c (C)2

I II

(2) (NC ,c(c)2 + HO-'- (-c)- 4)2

II IlI

(3) (N-4-q-N-2 1 2(E) 2 C-0 + NO-

(4a) (mc) 2c +, . C01" + 2CO"

IV

(4b) (c) 2Co + 2902- -0 aooo + 2C r

(5) HB OOH ? 2 +H2 0+C02

(6) cw- + H2o2 -+ o 2o

-64-

Thus the reaction of I with hydrogen peroxide is known

to produce epoxide II under certain conditions2 0 and the reaction of

II with the hydrogen peroxide anion to give hydroperoxide III is typical

of epoxide-nucleophilic reactions2 1 Moreover, the decomposition of III

to give cart yl cyanide (IV) would be typical of concerted bond cleavage

peroxide decomposition7 . IV might then be oxidized by bydrogon peroxide

directly to cyanate by a process probably involving concerted multiple

bond cleavage7 as in step 4a, or nucleophilic displacement by the

4 rogen peroxide anion might occur as in step 4b. Step 5 is bvpo-

thetical but it is known that the reaction of phosgene with hydrogen

peroxide in the presence of a fluorescer gives carbon dioxide, and is

not chemiluminescent . Cyanate would be an expected product in any

:aae, since it is a known prod4 t of the reaction of cyanide ion with

hydrogen peroxide 2

Accordingly, the possible intermediates II, IV, and CN-

were each tested for chemiluminescence under the conditions where

chemiluminescence was observed from I. In reactions with hydrogen

peroxide aud potassium t-butoxide or potassium hydroxide in 1,2-

dimetboxyetbane containing rubrene moderately strong cbesll.uminescence

lasting approximately ten minutes was observed from both II and IV, but

not from potassium cyanide. The chemillminescent reactions were remrkably

similar in intensity and lifetime to the analogous reaction with I.

-65-

The observed chemiluminescence from II and IV does not,

of course, establish that chemiluminescence from I is derived by the

process indicated above, since there is no evidence that the three

chemiluminescent reactions do not proceed by independent mechanisms.

Hovever, the proposed mechanism is reasonable in terms of the known

chemistry and the chemiluminescent results.

The chemiluminescent emission derived from the reaction

of IV with hydrogen peroxide suggests that direct oxidation of IV (step

4a) occurs rather than displacement of cyanide by peroxide (step 4b).

Step 4b is analogous to the reaction of phosgene with hrdrogen peroxide,

and the latter is not appreciably chemihminescent in the presence of

a fluorescer. The mechanism by which IV is xidized by hydrogen peroxide,

however, can only be speculative at present.

-66-

SECTION I - PART D

Experimental

Tetracyanoethylene oxide (II) was prepared by the methodof Reiche and Dietrich~v.

Carbonl cyanide (IV) was prepared using a modificationof the method described by W. J. Linn2 3 . A mixture of 7.2 g. (0.05 mole)of tetracyanoethylene oxide and 9.3 g. (0.05 mole) of diphenyl sulfidein 75 ml. of freshly distilled glyme was heated for three hours at 800.The product, ca. 2 ml. of a yellow liquid, was distilled from the blackreaction mixture at 270/18 ram.

An aliquot of the product was characterized by 'reatmentwith dimethlaniline in acetic acid to provide the known derivative,bis( p-dimethylaminophenyl)dicyanomethane. The crystalline solid isolatedfrom this reaction had au infrared spectrum consistent with that expectedfor the derivative and the m.p., 195-197, was in agreement with thelit. m~p., 194-196°.23

Chemilminescence Tests

Tetracyanoetbylene oxide (II). - About 5 .. of II weredissolved in 10 ml. of 1,2-dimethoxethane containing a few mgs. ofrubrene, and 5 mg. of potassium t-butoxide was added. The addition ofseveral drops of 90% hydrogen peroxide produced a medium intensitylight emission with a duration of about 10 minutes. Potassium ydroxidegave similar results.

Carboml Cyanide (IV). - One drop of carbonyl cyanide wasadded to 5 al. of p12-dimethoiethane containing a pellet of potassiumhydroxide, a few mgs. of .ubrene and several drops of 90% Iydrogen peroxide.The yellow light produced had a medium intensity lasting for severalminutes.

-67-

SECTION II

Exploratory Tests for Chemiluminesceut Reactions

Tests established7 for the screening of oxalic acid

derivatives were used to evaluate a number of other compounds, selected

primarily in terms of their availability, but also selected to establish

new guide-lines for the design of cheiluminescent systems. The test

results are compiled in Table I.

TABLE I

Exploratory Chemiluminescence Tests

Chemiluminescence Tests&

A B C D E F G

1. Trichloroacetic acid - -2. Trifluoroacetic acid - - VW3. Trifluoroacetic anhydride - - .4. Gycolic acid, calicium salt - - -

5. Chloral - VW - - W6. d-Tartaric acid . . . . . .7. d-Mannose .8. Itaconic u nydride .9. Tetrahydrozysuccinic acid,

disodiu salt .10. Malonitrile - . . VwU. Benzenehexol .12. 1, 3-Dihydroxy-2-propanone - . . .13. Netbylaainoethyltartronic acid - . .14. Dinetbyl acetylenedicarbo3ylate - . . .15. Chloroacetylchloride - .16. Dichloroacetylchloride - . . MW17. Cyclohexylphez711gycolic acid - . . .18. Diketene - . . W

Table continued

-68-

TABLE I (continued)

Chemiluminescence Tests

A B C D E F G

19. 03CU 0 5

20. 03C94~C~21. JLactic aCidP22. cC -Hydroxy isobutyric acidb23. di-Msndel-ic acidb24&. Benzilic acidb25. Benzilide26. Ethylene oxide27. Tetramethyletbaylene oxide28. Octarnethyloxamidimium chiorideb29. Tetra~heirylethuylene oxideC

FO(7TNOPM FOR TABLE I

a) Tests:

A) Approximately 3-5 mg. of the compon to be tested is added to5 ml. of a solution of about 1 mg. 9p20-diphezmylAwthracene (DPA)in paraffin (ESSO household vax) imaintained at 160-170 'c.

B) Approximtely~ 3-5 mg. of the compound to be tested is added to5 ml. of a solution of about 1 mg. DPA and "-5 mg. AIBN in paraffinmaintained at 85-90& under aro tm ee

C) Appro ,imtely 3-5 mg. of the compound to be tested is added to5 ml. of a solution of about 1 mg. DPA in 1,2-dinetboxyethaicontaining 5% vater at 60-65*c. About 5 mg. ft202 is M4dedimmediately.

D) Approximately 3-5 Mg. of the compon to be tested Is added to5 al. of a solution of 1 mg. DPA and 0.2 ml. CH9SOsN in 1,2-dimethoxyethane containing 5% water and mantained at 60-650C.About 0.5 al. 30% R20.? is added imediately.

E) Approximately 3-5 Mg. of the compound to be tested is added to5 ml. of a solution of 1. mg. DPA., and 5-10 mg. AI in diphezr1-methane maintained at 85.90'b under an oxygen atmosphere.

Footnotes continued

-69-

FOOTNOTES FOR TABLE I (continued)

F) Approximately 3-5 mg. of the compound to be tested is added to5 ml. of a solution of about 1 mg. DPA and 0.2 ml. anhydrousH2 02 in anhydrous 1,2-dimethoxyethane maintained at 60-65"C.

G) Approximately 3-5 mg. of the compound to be tested is added to5 ml. of a slurry of 1 ag. DPA, -'0.2 g. KOK (1 pellet) and0.2 ml. anhydrous H2 02 in anhydrous 1,2-dimethoxyethane maintainedat 60-65"C.

Qualitative intensities are based on the oxalyl chloride,hydrogen peroxide reaction taken as strong (S). Other designations areM a medium; W = weak; VW = very weak, barely visible.

b) In an additional test, a solution of 10-15 mg. of the compound in 10ml. 0.1 M K2 C0 3 containing 1-2 mg. of fluorescein was treated withI ml. of 30% H202 and 10 ml. of 6 x 10-2 M K2 S2 08. No chemiluminescencewas observed.

c) In an additional test, a solution of 10 mg. of t - compound in dimethylsulfoxide containing 1-2 mg. rubrene was treated with 0.5 ml. of 50% aqueouspotassium hydroxide and 2 2l. 30% hydrogen peroxide. No chemiluminescencewas observed.

-70-

SECTION II

EXPERIM4TAL

Tetrapbemylet ylene Oxide (TPEO)

TPEO was prepared by the method of Schmidlen24 ascited by Wang end Cohen25, m.p. 205-60, lit. 25 , m.p., 204-5 .

Anal. Calcd. C 2 0 0: C, 89.62; 1, 5.79. Found: C, 89.25;H, 6.02.

Bistripheznlacetic Carbonic Anhydride - A slurry ofi.C3 g. (0.005 moles) of potasaiu triphenylacetate in 50 ml. of 1:1(V:V) benzene-glyme was prepared at 25" under argon, and then 0.25 g(0.0025 moles) of phosgene in 4 al. of benzene was added slowly tothe stirred slurry. The reaction mixture was stirred 2 hours at 25*under argon, then filtered to remove insoluble material. The reacti.filtrate was evaporated to dryness under vacuum at 20-25 e to obtair1.2 g. of material, m.p. 179-192, soften 134', whose infrared spectrumis in agreement with that expected for the desired product.

-K-Chloroacetylchloride - The procedure of Bickel2

was used to obtain 10.9 g. (62%) of product, m.p. 4.85-50.0" (lit.m.p, 50').

Mixed Anhydride of Triphemylacetic Acid and CarbonicAcid, Monoethyl Ester - A solution of 2.9 g. (0.01 moles) of tri-pheqrlacetic acid and 0.8 g. (0.01 moles) of pyridine in 75 ml. ofanhydrous ether was prepared. The stirred solution was cooled to 0"under argon, and then 1.1 g. (0.01 moles) of etkyl chlorocarbonate in25 al. of anhydrous ether was added dropvise during 15 minz. Thereaction mixture was stirred 0.5 hours at 00, then filtered to removeinsoluble material. The reaction filtrate was evaporated to drynessunder reduced pressure to obtain 1.5 g. (41%) of vhite solid product,whose infrared spectrm strongly indicated mostly the desired producttogether with =all amounts of contaminants. The product of thereaction vas found to be non-chemiluainescent, and further purificationwas not undertaken.

Other MsteriaJ s - Benilide2 7 , tetramethylethylene oxide ,

and octamethyloxaiidiniva dichloride2 9, were prepared according toliterature recipes.

-71-

SECTION III

Toxicity and Thermal Stability Data

As a guide to possible hazards in the use of chemiluminescent

systems, data on toxicity and exothermic decomposition as measured by differential

thermal anal.sis will be published as they are obtained.

When it ii contemplated that systems of interest will be

stored and used as dilute solutions, the hazards involved would be those

of the solutions, not the pure choaiiluminescent constituents. It is

apparent, then, that toxicity or other hazards due to solvents should be

considered, especially for storage in poorly ventilated spaces, or areas

exposed to high temperatures.

For systems requiring non-aqueous solvents (i~e., the

oxalyl derivatives) those solvents used in the experimental work were

selected without regard to potential hazard. Since some latitude is

available in the choice of solvent, it seem unlikely that solvent

hazard need be a serious consideration. Two useful and relatv-ly

-on-volatile solvents which have been used extensively in studying

ox1ayl derivatives show very low levels of acute oral toxicity. Acute

oral MW50 in rate has been masu ed for dimetlwlphthalate (6.9 gs./kgm)3 0

and ethylene gly1 dimetbylether (.4.39 Su./kip)3 1 .

Toxicity of chemiltuinescent components is generally not

known. Hawever, the toxicity of 3-aminophthalbydrazide (luinol) was

examined in the Inviroamental ealth Laboratory of the American Cyanamid

Camma. As inbicated in the attached report, the pure mterial has an

-72-

LD5 0 greater than 10 ges./kgm. Conc .. rated bydrogen peroxide (90%)

is considered to have high acute local toxicity 31, although dilute aqueous

solutions are used as a skin antiseptic. Its behavior in non-aqueous

solutions is not known. Of the various oxalyl derivatives used in this

program, only oxalyl chloride is well known. No literature information

on oxalyl chloride could be found beyond the general observation that

pure acid chlorides hydrolyze rapidly to the corresponding carboxylic

acid and WE.' and possess the toxicity of their components. Oxalyl chloride,

however, hydrolyzes to CO, C02 and DC1 rather than to oxalic acid and l.

In concentrated form, oxalyl chloride should probably be considered highly

toxic and is certainly very irritating; however, itc behavior in dilute

solutions as used in this program (10"2 M maxzinm) is unknown.

No information on exothermic decomposition is yet available.

SOP '4 naV. ,-- .. 3 ,o

REPORT 64-96 DATE 17 Jul 1964 PAGE 367TCX ICITY DATA PROOucr 5-PM4~QO2o3-DIflRQ-l h-r JIPZIN DI0N;

ENVIRONMENTAL HEALTH LABORATORY L 0

CENTRAL MEDICAL DEPARTMENT

AMERICAN CYANAMID COMPANYWAYNE. NEW JERSEY 07470 0

EHL NO. 62-125 CI, NO. 3825 DISTRIBUTION 63SISINGLE ORAL DOSE SEX Male SPECIES Albino Rat LDO Greater than 10 g/kg

80o 1 PAGE 830 I(CF Nelson)CONDITIONS Dosed with a 20% aqueous dispersion.

DOSAGE ONSET OF S SIGNS', (0) DEAT, HOURS AND DAYS DIED MEAN WT. TIME OF (R) RECGVERYOA YS0-6 .I4 S 14 DOSED , T I 2 3 4 1 '-,,

I, 10.0 g/kg ' 77 1522. 5.0 0/5 10i 194a.

SIGNS OF INTOXICATION None observed.

GROSS AUTOPSY Normal.

10LE PAI|NTIRAL DOSE SEX SPECIES .D5s

BOOK PAGE CONDITIONS

AONSET OF (3) SIGNS. (0) DEATH. HOURS AND DAYS DIED MEAN WT. THI OF (R) ECOVER .DAYj0. 3. 4 1SS ? 1... DOSED I T I a: 3 4 s 4 7.4

3.

SIGNS OF INTOXICATION

GROSS AUTOPSY

SINGLE DIMAL DOSE SEX SPECIES .Oso

BOOK PAGE CONDITIONS

DOSAC, ONSET OF SSI(GN 0) DATH. H9LOL AND DAYS DIED. MEAN WT. TIM, UV0.6 014 a 3 4 1 7 14.14 DOSED T 4 $ 7,14

I.

4.

SIGNS OF INTOXICATION

SKIN IRRITATION

GROSS AUTOPSY

6

____ ____ ____ _ _ ____ ____ ___ "AGE

11I0.GLE INHALATION Tk,. '.IS LC

BOOK C-,NDITIONS

ChAMRE R ONSET OF CS) SIGNS. (D) DEATH MIN. AND HOURS DIED ME AN WT. DAYS TO FAI4 OR REC VERYI

CONCENTRATION 0.15 5s.3o0)0so z 3 1 4 5 1 7 1 DOSED T 0 1 1 13 4 5 6.14

2.

3.

4.SIGNS or INTOXICATION

GROSS AUTOPSY

RABBIT SKIN IRRITATION ICONDITIONS 0.5 Ml of a 5% solution of the compound applied underPAGE 11ian impermeable patch.

REACTIOi RABBIT NUMBER. VALUE ___MEAN [WAS STRUCTURE OF THE TISSUE AT THE SITE OF CONTACT

1 4 1 0 1 6 VALUE 'DESTROYED OR CHANGED IRREVERSIBLY IN 24 HOURS OR'- Y THEM -1L--~.~ESS? YES C3NLINTACT 0 0 0 0Q H SCORING SYSTEM USED HEREIN FOR SKIN AND EYE

'd4ACT 0 0 0 0 0 0 0 IR~iT.T:ON REACTIONS IS THAT OF DRAIZE, J. H., WOOD-a ~ o o a 0 ARD, G., AND CALVERY. H. 0.: J. PHARMACOL. EXPTL..iwAOEO 0 U 0 0 0 0 THERAP., 12:377, 1944, THE MAXIMUM POSSIBLE VALUE

0 0 0 0 0 0 0 FOR A SKSN REACTION (EXCLUDING NECROSIS) IS54. THE"PRIMAR1Y IRRITATION $CORE" IS THE SUM OF THE MEAN

ALT 0 0 0 0 0 0 0 VALUES DIVIDED BY 4. THE MAXIMUM POSSIBLE SCORESACT 0 0 0 0j 0 0 FOR EYE IRRITATION REACTIONS (AGAIN EXCLUDING

ABRADED 0 0 0 0 0 0 0 NECROSIS) ARE. CORNEA, 80: IRIS, 10; CONJUNCTIVAE. 20.

ABRADED 1 -LQ. ulo 0 PORIMRY IRRITATION $CORE.....

RABB:1 FYE IRRITATION T ONDITIONS Dosed with 0.1. ml of a 5% suspension in H20.BOOK I~PAGE 401TIME. TUTX RAUIT NU MUER VALU MEAN jAT ANY Of THE READINGS, MADE AT 24,48 AND 72 HOURS,

HOURS WAS THERE:1kU- DISCEPNIBLE OPACITY OR ULCERATION OF THE CORNEA

CORNEA o0 0 0 0 0 OTHER THAN A SLIGHT DULLING OF THE NORMALiI LUSTER? YES I No al

24 1IRIS 10 0 0 0 0 0 -INF.AMMATION Of TH4E IRIS OTHIER THAN SLIGHT DEEP-'CONJUNCTIVAE 0 0,1.2 0 0 0 ENING OF I HE FOLDS. OR SLIGHT CIRCUMCORNEAL

CORNEA 10 0f 0 0tol 0 INJECT-.--*? YES '-Z NO'I IIS 00 0 -A DIFFUSE, DE EP-CRIMSOIN RED APPEARANCE Or THEASIIISf 0I CONJUNCTI"AE. WITH INOIVNIAL VESSELS NOT EASILY

CONJUNCT 1VA C)0 0 o 0 DISCFRNIBLE? YESC0 No ~

CORNEA 0 -AN OBVIOUS SWELLING OF THE CONJUNCTIVAE, X

IRISP1 0 0 0 0 0 Of THE LIDS? YES No

CONjUNCTIVAl 0 0 0 0 WAS THERE DESTRUCTION OR IRREVERSIBLE CKANGE OFj,-. ___ I - ANY 71".~LISUE IN 34 Hai, OR LESS? YES = No ~

CODMMENT

T. .>-pcurd is considered to be practically non-toxic cy ingection in( uuses. It Is not expectti to present any significant hazards toi n ordinary insdustria~l handling,.

1.4 *.. * 4'o0 , a S3 i-6 9 o- TV0040 toSv4 6 0Ib * ol ?u 4,gho~a t& . %.4a voog eoveIak. "to*,, ototevo~ov. avotsa vaato-P... 41t.07004'.49 * fitS 40.84'. v .. @ .0 01,10664 tooe o00 VP* 4 t5.lsa1? se C4960 4 hf 8afta""40 v.b

C... ~ ~ 11 f5~ ,I0'S " S .80 0S63 to&UY 1Epst 01Ok.tIoh SG0*S*-ft w 4 OSbCv e. TS46 m0USeaYo S. Tt0I., 'I *a too *OOM v ot.

-75-

References

1. American Cyanamid CompaW , "ChezlminDescent Materials," TechnicalReport No. 1, Contract Nionr 4200(00).

2. American Cyanamid Company, "Chemiluainescent Materials," TechnicalReport No. 2, Contract onr 4200(00).

3. American Cyanamid Company, "Chemiluinescent Materials," TechnicalReport No. 3, Contract Non. 4200(00).

4. American Cyanamid Compazy, "Chiilaidnescent Materials," TechnicalReport No. 4, Contract Nonr 200(00).

5. American Cyanamid Compazn, "Chemiluainescent Materials," TechnicalReport No. 5, Contract Nonr 4200(00).

6. American Cyanmid CompazW, "Chemiluminescent Materials," TechnicalReport No. 6, Contract Nonr 4200(00).

7. American Cyanamid Compazy, "Chmihuinescent Materials," Technical

Report NO. 7, Contract Nonr 14200(00).

8. E. A. Chandrass, Tetrahedron Letters, No. 12, 761 (1963).

9. W. Dasler and C. D. Bauer, Ind. Et. Chen., 18, 52 (196).

10. J. Mitchel and D. Sith, Aquametry, Interscience, New York, 1948.

11. H. Staudinger., Ber., 41, 3558 (1906).

12. 7. D. Greene and G. razan, J. Org. Ch=., 20, 218 (1963).

13. I. M. KolthofT and E. B. Sandeli, Textbook of Quantitatie IPOegaicAisis, The macmillAw Co., Nev York, 1945, p. 630.

14. G. Wittig and R. W. Noff-a, Der., 95, 2718 (1962).

15. C. E. Boozer, G. S. Dsmond, C. S. RuIdton, and J. N. Sen, J. An. Chem.So_., .U, 3233 (1955).

16. W. R. Edwards and W. N. genley, J. Am. Chum. Soc., 7, 3587 (1953).

17. C. D. Rurd and H. F. Dull, J. Am. Chew. Soc., 54, 3427 (1932).

18. J. W. Williams and C. D. Rurd, J. Ore. Chm., 5, 122 (1940).

19. H. Staudaiger, BeT., 38, 1735 (1905).

20. A. Peicbe and P. Dietrich, er., 96, 3P44 (1961).

21. R. E. Parker, Chn. Revs., 2, 737 (1959).

-76-

22. 0, Msson, J. Chem. soc., 91pi 140 (1907).

23. W. 3. Linn, U. S. Pat. 3,115,517.

24. J. Schmidle,, !Sr., 43, 1157 (1910).

25. C. B. Wang and S. G. Cohen, J. Org. Chie., 26, 3301 (1961).

26. B. Bickel, Bar., 22, 1537 (1989).

27. B. T. Arnold, W. E. Partma, and R. M. Dodson, J. An. ChM. SOC.Th,24.39 (194.9).

28. E. L. Eliel and N. N. Rerick., J. An. Che.. Soc., 8L2, 1362 (1960).

29. We f. i ny and J. Sheeto, Paper presented at "9yoposiu on Ch -

lminescence" at U. S. ArM Research Office, Durban, N. C., April 1,1965. (Preprints, p9. 223).

30. Industral fi cue anW ToxicoloW, F. A. Patty, Ed., Interacience Pub.,kNw York, 1962.

31. M Properties of Indutrial Materials, N. Irving Sax, ReinholdNb. Co., Nsw York, 1957.

-77-

DITM I LIST

commanding Officer -' a--U ndng OfficerOffice of Naval Research Branch Office ar-ry Diamond Laboratories219 So. Dearborn Street ngshington 25, D. C.Chicago, Illinois 60604 (1) a Utn: Library (1)

Comanding Officer Office, Chief of Research &

Office of Naval Research Branch Office Development

207 V. 24th Street Department of the ArsWNev York, New York 10011 (1) Washington 25, D. C.Attn: Pt1eical Sciences Div. (1)Commdiug OfficerOffice of Naval Research Branch Office Chief, Bureau of Ships1030 East Green Street Department of the NavyPasadena 1, California (1i Washington 25, D. C.

Attn: Code 342 (2)Commanding OfficerOffice of Naval Research Branch Office Chief, Bureau of Naval WeaponsBox 39 Navy #100 Fleet Post Office Department of the NavyNew York, New York (7) Washington 25, D. C.

Attn: Technical lbrary (4)

Director, Naval Research Laboratory Mr. R. Dviggins (RIUO-3) (1)Washington 25, D. C.Attn: Technical Information Officer (6) Defense Doctentation Center

Chemistry Division (1) HeadquartersMr. R. Kirk (Code 64O) (1) Cameron Stetion

Alexandria, Virginia (20)

Chief of Naval ResearchDepartment of the Navy Director of ResearchWashington 25, D. C. U. S. ArvW Signal Research &ttn: Code 425 (2) Development Laboratory

Fort Mouthp Xew JerseyDDRU Attn: SEKA/OR (1)Technical Ltbrar7Room 3C-128., The Pentagon Naval Radiological Defense tab.Washington 25, D. C. (1) San Francisco 24, California

Attn: Technical Library (1)Research DirectorClothing & Organic Naterlals Division Naval Ordnace Test StationQurtermater Research & China Lake, California

Etgineering Comad Attn: Ewd, Chemistry Div. (1)U. S. Arw Dr. 0. W. Leonard (1)Natick, Massachusetts (1) (Code 4501)

Mr. S. D. Little (I)Air Force (Code 353k)Office of Scientific Research (SRC-E)Washington 25, D. C. (1)

-78-

Cou-anding Officer Chief of Naval Research

Army Research Office Depar nt of the Navy

Box CM, Duke Station Washington 25, D. C.Durham, North Carolina Attn: Code 421 ()Attn: Scientific Synthesis O-'fice (1) Code 440 (1)

Code 450 ()

Brookhaven National Laboratory Code 461 (1)Chemistry DepartmentUpton, New York (1) Dr. Charles E. Waring

University of Connecticut

Atomic Energy Commission Dcparnt of Chemistry

Division of Research Storrs, Connecticut (1)Chemistry ProgramsWashington 25, D. C. (1) Dr. Emil H. White

Department of Chemistry

Atomic Energy Counission Johns Hopkins UniversityDiv. of Technical Information Extension Baltilore, Maryland (1)Post Office Box 62Oak Ridge, Tennessee (1) Dr. Ferd E. Williams

Department of Physics

U. S. Army Chemical Research and University of DelawareDevelopment Laboratories Newark, Delaware (1)

Technical LibraryArny Chemical Center, "yland (1) Dr. J. R. Castle

E. I. duPont de Nemours & Co.Director, ARPA Central Research DepartmntThe Pentagon Wilmington, Delaware (1)Washington 25. D. C.

Attn: Mr. Fred I. Edwards (2) Dr. Daniel GrafsteinGeneral Precision, Inc.1150 McBride Avenue

Inspector of Naval Material Little Falls, New Jersey181 Middle StreetBridgeport 3, Connecticut (1) Dr. A. W. Berger

Monsanto Research Corp.

Naval Awunition Depot Boston LaboratoriesCrane, Indiana Everett, Massachusetts 02149 (1)Attn: Code 314 (1)

Dr. Henry Linschitz

Naval Ordnance Laboratory Department of ChemistryWhite Oak Brandeis UniversitySilver Spring, Maryland Waltham, MbssachusettsAttn: Mr. B. E. White (Code WE) (1)

Mr. R. T. ReillyLimited War LaboratoryAberdeen Proving Grounds, Maryland (1)

_nest/k f\ab~e COPY

Office of the Chief of Research and

Dr. F. H. Johnson Development

Department of Biology Department of the Army

Princeton University Washington, D. C. 20310Princeton, New Jersey (1) Attn: Lt. Col. Irving C. Hughes

Special Warfare Office (i).

Dr. W. H. UrryDepartment of Chemistry Bio-Chemical Division

University of Chicago Chemical Branch (ATCC)

Chicago, illinois Attn: Lt. J. R. ClaryEglin Air Force Base 32542

Remote Area Conflict Information Center Florida (i)Battelle Memorial Institute505 King Avenue Mr. Phillip Zirkin

Columbus, Ohio 43201 Pyrotechnic LaboratoryPicatinny Arsenal

Remote Area Conflict Information Center Dover, New Jersey (I)

Battelle Memorial Institute1755 Massachusetts Avenue, N. W. Dr. Mura S. CohenWashington, D. C. 20036 (1) Mawiger, Chemistry Department

Thiokol Chemical CorporationDr. Vincent J., Keenan Reaction Motors DivisionInstitute for Defense Analysis Denville, New Jersey 07831 (1)hWO Arx-Navy DriveArlington, Virginia (1)

Colander, Naval Weapons LaboratoryDeblgren, VirginiaAttn: Code WWB (1)

Comanding OfficerPicatinn ArsenalDover, Rev JerseyAttn: SMUPA-DR4

Mr. E. Deuberg (1)

%G0911