AD-RI63 813 AFRRI (ARMED FORCES RADIOBIOLGOY RESEARCH INSTITUTE) 1.'1REPORTS JULY-SEPTEMBER 1995(U) ARMED FORCESRADIOBIOLOGY RESEARCH INST BETHESDA MD 199

UNCLSSIFIED F/O 61 NL

'p

111111. 1165L=2 132 2

196

1111I1 25 1114 ____

MICROCOPY RESOLUTION TEST CHARTNATIONAt BRRAU OF STANDARDS -1963-A

/ *~ -1

%r

UNCLASSIFIED .I

SECURITY CLASSIFICATION OF THIS PAGE *u. 1 3 &REPORT DOCUMENTATION PAGE

is REPORT SECURITY CLASSIFICAT ION 1b. RESTRICTIVE MARKINGS

UNCLASSIFIED2., SECuRITV CLASSIFICATION AUTHORITY 3. DISTRIBUTION/AVAI LABILITY OF REPORT *b.I*

2b OECLASSIFICATION/DOWNGRADING SCHEDULE Approved for public release; distribution .- \'unlimited.

4 PERFORMING ORGANIZATION REPORT NUMBERIS) 5. MONITORING ORGANIZATION REPORT NUMBER(S)

AFRRI SR85-26 through 85-32 lt6.. NAME OF PERFORMING ORGANIZATION b. OFFICE SYMBOL 7&. NAME OF MONITORING ORGANIZATION '

Armed Farces Radiobiology lit aititlcabie % P

Research Institute ARI'6c. ADDRESS IC06. Slate and ZIP Code) 7b. ADDRESS (City. State and ZIP Code)

Defense Nuclear AgencyBethesda, Maryland 20814-5145

Go. NAME OF FUNDING/SPONSORING 89. OFCSMOL 9PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER

ORFGANIZATION iaplcbe

Defense Nuclear Agency DNAac ADDRESSCI . State and ZIP Code I 10 SOURCE OF FUNDING NOS

Washington, DC 20305 PROGRAM PROJECT TASK WORK UNIT

11i Ti TLE 'Inciu.de Security ClawaficationI AXAFRRI Reports, Jul-Sep 1985

12. PERSONAL AUTHORISI

13a. TYPE OF REPORT D3.TM OVRD1 ATE OF REPORT (Y,. Mo., LDayl 5S PAGE COUNT

Reprints/Technical FROM _ TO___2_7 TO

16 SUPPLEMENTARY NOTATION

17 COSATI CODES 18 SUBJE CT TE RMS Won tinue on revers if necesa, and identify by block num berl

FIELD GROUP SUB OR N/A

19. ABSTRACT fConlinue on rYeere if ,leceta' and identify by b, it ,number,

This volume contains AFRRI Scientific Reports SR85-26 through 85-32 for Jul-Sep 1985.

20 DISTRIBUTION AVAILABILITY OF ABSTRACT 21 ABSTRACT SECURITY CLASSIFICATION

UNCLASSIFIED-UNLIMITED 2 SAME AS RPT . DTIC USERS 71 NLSSFE22a NAME OF RESPONSIBLE INDIVIDUAL i22b TELEPH-ONE NUMBER 22c OFFICE SYMBOL

(,Inclu1de 41ca ('.d,

Junith A. Van Deusen (202)295-3536 ADMG

DD FORM 1473, 83 APR EDITION OF I JAN 73 IS OBSOLETE UCASFESECURITY CLASSIFICATION OF THIS PAGE

CONTENIS

Scientific Reports

SR85-26: Burghardt, W. F., Jr., and Hunt, W. A. Characterization of radiation-'induced performance decrement using a two-lever shock-avoidance task,

81R85-27: Dagen, A. J., Alfano, R. R., Zilinskas, B. A., and Swenberg, C. E.-Fluorescence kinetics of emission from a small finite volume of a biologicalsystem;

SR85-28: Dorval, E. D., Mueller, G. P., Eng, R. R., Durakovic, A., Conklin, J. J.,and Dubois, A. -",Effect of ionizing radiation on gastric secretion and gastricL

SR85-29: Ely, M. J., Speicher, J. M, Catravas, G. N., and Snyder, S. L. -Radiationeffects on diamine oxidase activities in intestine and plasma of the rat,

SR85-1 Ledney, G. D Exu E. Dadackson, W. E., Ml.'-oniduealterations in survival of go irradiated mice% importance of wound timing,

SR185-31: Mickley, G. A., Stevens, K. E., and Galbraith, J. A. -Quaternarynaltrexone reverses morphine-induced behaviors.;

SR85-32: Pope, M., and Swenberg, C. E. -Electronic processes in organic solids.

y . I y

ARMEO FORCES RAOIOSIOLOGYRESEARCH INSTITUTE

SCIENTIFIC REPORT %i.

SR85-26RADIATION RESEARCH 103, 149-157 (1985)

Characterization of Radiation-induced Performance DecrementUsing a Two-Lever Shock-Avoidance Task

WALTER F. BURGHARDT, JR., AND WALTER A. HUNT

Behavioral Sciences Department. 4rmed Forces Radiohiologr Rev'earch Institute.Bethesda, Maryland 20814-5145

BURGHARDT, W. F.. AND HUNT. W. A• Characterization of Radiation-Induced Perfor- I..-.mance Decrement Using a Two-Lever Shock Avoidance Task. Radiat. Res. 103, 149-157(1985).

Rats were trained to perform a task involving responses on two levers. Responding on anavoidance lever delayed the onset of electrical footshock for 20 sec and responding on a " ,-%warning lever turned on a light for 60 sec. When the light was on. the task on the avoidancelever was changed from unsignaled shock avoidance to signaled shock avoidance by precedingthe shocks with 5-sec warning tones. The animals preferred the signaled avoidance condition.After 100 Gy of 'Co irradiation, the animals were less able to avoid shock, an effect fromwhich the animals recovered somewhat over 90 min. The response rate on the avoidance leverremained at or above control rates, while the response rate on the warning lever showed aninitial increase, followed by a decrease below baseline. The increase in responding on theavoidance lever occurred in bursts just after presentation of the shocks. The data suggest thatunder these experimental conditions a subject will not respond appropriately to avoid shock or ..

acquire cues that can facilitate the avoidance of shock. The effects, however, do not reflect aninability to perform the required movements but instead appear to reflect some characteristicof the task associated with a particular lever. i- i45 Academic Press. Inc

INTRODUCTION

Exposure to high doses of ionizing radiation can result in depression of thenervous system. In some cases, this is expressed as an early transient incapacitation(ETI). characterized as the inability to complete a task. and has been observed in anumber of species (1-4). This phenomenon occurs shortly after exposure tomoderate and high doses of radiation, with recovery to near normal performanceon some tasks occurring within approximately 30 min after exposure. A performancedecrement of some sort is often seen, even in the absence of an ETI (1). Frequently.some aspects of behavior remain unchanged or even enhanced, while others aredisrupted. For example. a decrease in general activity and food intake in monkeyshas been demonstrated, while learned task performance is maintained (5). In rats,maze performance is improved, while motor activity is decreased (6).

Ionizing radiation has been shown to disrupt active avoidance behavior (7. 8). Inthese studies, rats were trained to avoid successive electrical footshocks by jumping "

onto a retractable platform in response to an auditory cue. When the animals wereexposed to high-energy electrons or -y photons, their ability to avoid shock wassignificantly reduced in a dose-dependent manner within the dose range of 25-200

149 (X)33-7587/85 $3.0X)( tpsnght t14 th% Academic Prew Inc

lIl right% (it reprtduction in an% form resrscd

150 BURGHARDT AND HUNT .

Gy. This decrement was transient lasting less than 30 min. Although the animalsdid not avoid shock, they would jump onto the platform in response to shock (8).

In the present experiments, we attempted to further characterize the effect ofionizing radiation on active avoidance behavior by using a modification of aparadigm designed to measure an animal's preference for signaled versus unsignaledavoidance (9). We wanted to find out whether the irradiated animals would requestand use cues to avoid shock at a dose that would be expected to produce reliableand profound behavioral decrement, while not totally disrupting behavior (8). Thisparadigm is a two-lever, bar-press task in which animals may obtain auditory cuesas a warning for impending footshock. Pressing one bar (avoidance lever) results inpostponement of the shock. Pressing the other bar turns on a light. While the lightis on, a 5-sec tone precedes the shock. Using this auditory cue, rats will learn veryeffectively when to press the avoidance bar to maintain a low shock frequency(signaled avoidance). In addition, this paradigm simultaneously measures behavioron two different tasks and assesses the ability of the animals to maintain more than %one measured behavior after irradiation.

METHODS

Twelve Long Evans (Blue Spruce) rats (300 g) were housed separately and kept on a reversed 12-hrlight cycle. The animals were watered and fed ad libitum throughout the course of the study.

Operant-conditioning chambers approximately I I cm high. 25 cm deep. and 24 cm wide, each housedin separate sound-attenuating boxes, were used in these experiments. The floor consisted of aluminumrods through which scrambled electrical footshock (1.0 mA) could be applied using constant current acshockers and electromechanical scramblers. On one wall were two response levers. A SONALERT speakerwas centered on the wall to the left of the levers to provide, when required, a 1900-Hz tone atapproximately 68 dB/SPL. A clear jeweled I-W light was located in the middle of the ceiling. The unitswere maintained in a darkened climate-controlled room with white noise provided through a 16 cmspeaker to mask outside sounds.

Prior to the first training session, animals were placed in the operant chambers for at least 2 hr to

familiarize them with the apparatus. Thereafter, each experimental session lasted 4 hr.The animals were trained to avoid a 0.5-sec electrical footshock by responding on an avoidance lever

(10). A response on the avoidance lever postponed the onset of shock by 20 sec (RS 20). In the absenceof responding on the avoidance lever, shock occurred at 5-sec intervals (SS 5). A single response on aseparate lever (warning lever) turned on the overhead light for I min, during which shocks following aresponse on the avoidance lever were preceded by a warning tone during the last 5 sec of the response-to-shock interval (signaled avoidance) (11). If the animal responded on the avoidance lever during thetone, the tone was terminated and footshock was delayed 20 sec. Responses on the warning lever whenthe light was on were counted but had no scheduled effect. After I min of signaled avoidance, theoverhead light was turned off. A single response on the warning lever could then turn the light back onand reinstate signaled avoidance. Training was complete when the animals could successfully avoid morethan 901 of the shocks that could be presented (12/min) and when performance was maintained within101 a constant amount of time in signaled avoidance.

For programming of the experimental paradigms and data collection, a PDP-8E computer operating .under SCAT software was used. Data were recorded graphically by cumulative recorders and numericallyby the computer. The following parameters were continuously measured: the number of shocks received, the number of responses made on each lever, the distribution of responses in time, the number of warningtones used. and the latenc, of responses after cues.

After training. subjects were habituated to the effects of schedule interruption and transport thatoccurred for irradiations. After 2 hr of performance. the session was suspended with all stimuli andresponses disabled. The animal was placed in a Plexiglas restraining tube, transported to the "Co facility.and returned without being irradiated. The session was then resumed until there was less than a 10-"

*- .- 7. = " . "_ " -- - --

a.

PERFORMANCE DECREMENT AFTER IRRADIATION 151

difference in the number of shocks received and in the number of responses made on each lever duringthe next hour. compared with those during the hour before removing the animals from the conditioningchambers.

After habituation, the animals were randomly assigned to two groups of six animals each. One groupwas irradiated once, while the other group served as controls (sham irradiated). For irradiations or shamirradiations. the animals were handled as they were for habituation, except that each irradiated animalwas placed in the w°Co exposure room and received a single bilateral dose of 100 Gy of y radiation at arate of approximately 66 Gy per minute. Control animals were handled identically, with the exceptionthat they were not irradiated. The transport time from the radiation facility to the conditioning chamberswas less than 5 min. At the end of the study, all animals were sacrificed via barbiturate overdose.

For radiation dosimetry, paired 50-mI ion chambers were used. Delivered dose was expressed as a ratioof the dose measured in a tissue-equivalent plastic phantom enclosed in a restraining tube to the dosemeasured free in air.

For analysis, only the measurements made 60 min prior and 90 min after irradiation were used, periodswhen the performance of the animals was most consistent. The data collected were divided into six 10-min blocks before removal from the apparatus for irradiations, and nine 10-min blocks after resumptionof the session for analysis. For each lever, responses during each 10-min postirradiation period weretotaled and expressed as the percentage of the mean number of responses for the six 10-min periodsimmediately preceding irradiation. Responses from the sham-irradiated rats were similarly recorded. Allother measures were presented as totals for each 10-min period. The data were statistically analyzed usingmultiple factor analyses of variance with repeated measures on one factor (12). Radiation dose (0 or 100Gy) was one factor, and time after treatment was the repeated factor. The level for statistical significancewas 0.05.

RESULTS

While performing on the two-lever paradigm, the unirradiated rats avoided shock

very well, receiving less than five shocks over a 10-min period. They also spent

most of their time in signaled avoidance. In the hour before irradiation, the

experimental subjects spent an average of 83.7% of their time in signaled avoidance(range 78.3-95.5%). During a comparable period, control subjects spent an averageof 84.8% of their time in signaled avoidance (range 74.8-91.8%). Exposure toionizing radiation degraded the ability of animals to avoid shock. Figure I is arepresentative sample of the performance of one of the animals on the avoidancelever subsequent to irradiation. Note the low shock density, high number of intervals

of signaled avoidance, and overall rate and pattern of responding seen in the

preirradiation performance found in the upper two tracings. In contrast, the I. -..postirradiation performance clearly shows an increase in shock density, a progressive

decrease in the number and regularity of periods of signaled avoidance, and a smallincrease in overall response rates on the avoidance lever.

The numbers of shocks and warning tones received by the animals are illustrated

in Fig. 2. The control values presented are the pooled averages for the unirradiated r

control group during the 90 min after sham exposure. During the first 10 min afterirradiation, the animals received approximately eight times more shocks than thenonirradiated subjects. The number of shocks received decreased somewhat duringthe course of the session but remained at least three times control levels (effect dueto radiation, P < 0.05: effect due to time after irradiation, P < 0.01: interaction,P < 0.01). The number of warning tones progressively decreased after irradiationand remained below the number of shocks received during the same interval. This

. . . . . . . . . . . . . .. . . . . . . . . .- - . . .. . • .-~~~~~~~~~.. .....,.-...-.-...... .. .... ." .. .- .,. :.1.,...--,---.-.-.'... ..................... .. .. . ..-....-- ~~~~~~~.'.--.-.'. .. . . . . . ..'_. .-.-. .-...-.. . ..,. . . . . ...-"-.:,.. -.. ,'.' ' ''. ._. '..,. : 3..;" "-", "-.-., "- -,'"-"- '

~' b WT.~rw 5Wb ~rrr147 . wrnw wn Ik SlL IL X7MWSTLI11 %i7-z":C' *2IS r AICA! t- Q .7 .K AS-'t

152 BURGHARDT AND HUNT

irrOdiotion

100

ResDonses --

FIG. 1. Continuous cumulative record beginning in upper left comer from a representative subject on .-.

the day of irradiation. Downward vertical excursions on the lower tracing indicated periods when the

overhead light was on and signaled avoidance was in effect. Upward vertical excursions on the uppertracing represent responses on the avoidance lever; the pen is reset downward every 300 responses.Diagonal hatches on the upper tracing indicate shocks presented. Time proceeds from left to right. Timeand response scaling are provided on the lower left portion of the illustration. Time of irradiation ismarked on the figure with an arrow and a reset of the upper pen.

suggested that the animals were not responding for the tones and therefore couldnot use them as aids in avoiding shock.

Response on each lever after irradiation, expressed as the percentage of preexposurerates of responding for each animal, is presented in Fig. 3. When compared to thenonirradiated group, the irradiated animals responded approximately 30-40% more

40

35 -Tones 130 [~Shocks30 "°

( 25

4 T1 15 j"' "

U t0~.. -.. %0

Control 10 20 30 40 50 60 70 80 90

Tine afier lrradiaton (m nutesi

FIG. 2. Mean number of warning tones and shocks received by the experimental group after irradiation.plus or minus standard error of the mean. The control warning tone and shock rates consist of pooledrates of presentation for all subjects in the control group for the 90 min after sham treatment.

-p '° '~ ~ * ;:~1Js: ~ ~-~- -*.*** ** ***o-°o **o

W:.

PERFORMANCE DECREMENT AFTER IRRADIATION 153 .,

220-] AVOIDANCE LEVER0200-

2 WARNING LEVER

660.

4 10

0 r20, % ,

° .

20 ,P

COn:,ol 010 1020 2030 3040 4050 5060 600 770 o 090

MINUTES POSTIRRADIATION

Fi(i. 3. Mean responding rates on the avoidance and warning levers expressed as percentage of theaverage rate of responding for each subject during the six intervals before treatment. The control valuesare pooled from the control group's performance during the same periods before and after shamirradiation. All values are expressed plus or minus standard error of the mean.

on the avoidance lever. This effect was consistent throughout the session (effect dueto irradiation, P < 0.05: effects due to time after irradiation, interaction, n.s. at0.05). In contrast, the irradiated animals initially responded more frequently on thewarning lever, but within 20 min after irradiation, their performance droppedsignificantly below baseline levels and remained there during the remainder of thesession (effects due to irradiation, time after irradiation, and interaction all significantat P < 0.01). -.

One way in which to determine the distribution of responses during a session is . ".with interresponse time (IRT) histograms. Based on the requirements of the task, itwas expected that animals would respond on the avoidance lever most often justafter the onset of the warning tone (that began 15 sec after the last response on theavoidance lever), as can be found in the histogram of Fig. 4A. However, theirradiated animals responded more frequently after shock onset (that occurred 20sec after the last response on the avoidance lever in either signaled or unsignaledavoidance and 5 sec after the onset of the warning tone in signaled avoidance) and 1tended to continue responding once initiated (Fig. 4B). In addition, rates ofresponding were comparatively low around the time a warning tone could bepresented (beginning 15 sec after the last response on the avoidance lever).presumably because of the reduction in responding on the warning lever (Fig. 3).This would result in a lower number of warning tones available to assist performance P(Fig. 2).

IRTs on the warning lever are presented in Fig. 5. Control animals respondedmostly during the 20 sec after the overhead light was turned offl(indicating a changefrom signaled to unsignaled avoidance) (Fig. 5A). The animals presumably do this

to reinstate the signaled avoidance condition (Fig. 5A). These subjects also tend torespond in bursts (that is. interresponse time under 5 sec). Irradiated subjects makefewer overall responses with a much greater variation in IRTs and the appearance

.-.. ..

6. o,

154 BURGHARDT AND HUNT

[A-

15

'I''

10 * .. -

5

02

B 25-

S20-

15

10

005 2 4 6 8 10 12 14~ 1 18 A20

1205wl,,l Too. Shock,. .'

Interresponse Interval ,n Seconds

FIG. 4. (A) Interresponse time distribution on the avoidance lever for the control group after sham -.irradiation, plus or minus standard error of the mean. Arrows indicate the times of onset of warning toneand shock, when present (warning tones were presented on during signaled avoidance). (B) Interresponsetime distribution on the avoidance lever for the experimental group after irradiation, plus or minus the " ""-standard error of the mean. Scales foi both graphs are identical.

.7-

of very long IRTs (Fig. 5B). In general, these animals responded more randomlyrather than immediately after the light went off.

If an animal is using the warning tone appropriately as a cue, the latency betweenthe onset of the tone and a response on the avoidance lever should be short. In Fig.6A. the control group shows a characteristic patterning of responses with shortlatencies, indicating detection and appropriate use of the warning signal. Similarlatencies of the responses on the avoidance lever were found in the irradiatedanimals (Fig. 6B). However, the total number of responses to the warning tones wasreduced, consistent with the decreased number of tones presented (Fig. 2).

It should be noted that following irradiation with the dose used in this study (100(iA). the animals did not exhibit any gross abnormalities in spontaneous behaviorand were able to move about freely during the course of the study.

DISCUSSION

Consistent with previous reports, the results of this study clearly demonstrate thatexposure to ionizing radiation degrades performance of an active avoidance task ata dose of radiation that, based on previous studies using rats (7. 8), would beexpected to induce profound and reliable behavioral decrement. After irradiation, '

S . ............

.. . . . . . . ..- - - '- -- .. -" ..'.- : -; " -'' - -"- .- -- ' .- . - --.. -:-.:- . - -- -5-... -- . -.._..-_.- - .-,': " '., -"- i."-

PERFORMANCE DECREMENT AFTER IRRADIATION 155

30

20 -

0*

o

Avldnc ItevI

S30-

20~

0 ,~ 0 40 60 60 100 20 '0 160160 0°..

periods of signaled avoidance, after which the chambers were dark. (B) lnterrespons time diqtnaIutzen.

for the experimental group after irradiation on the warning lever, plus or minus the standaro error .,; tr, "

mean.

animals received a significantly greater number of shocks. However, it doe, notappear that they were incapable of responding to avoid shock. To the contrary, tiz --animals responded at a significantly higher rate on the avoidance ieve. A!'houghsome recovery in the ability to avoid shock was noted, an increased shock densityremained throughout the measurement period. This is in contrast to other reports -.-where animals recovered completely from behavioral decrements such as ETI within30 min, implying that the degree of recovery observed after irradiation may dependon the nature of the task.

The IRT distribution for the avoidance lever (Fig. 4) shows that irradiated anirnals-responded less at times appropriate to avoid shock, that is, just subsequent to thepresentation of the warning tone. This is due to the reduced number of tonespresented (Fig. 2). When tones were presented, the animals apparently used themby responding shortly thereafter (Fig. 6). In addition, the animals tended to respond _in bursts to the shocks. Such bursts of responses would contribute very little to thesuccess of subsequent shock avoidance. Maintenance of escape behavior withdepressed shock avoidance after irradiation has previously been demonstrated usingrats in a platform avoidance task (8).

In contrast to the sustained increase in the rate of responding on the avoidancelever, responding on the warning lever was substantially reduced. The IRT distributionon the warning lever (Fig. 5) became less regular and much more erratic in the

I ...

156 BURGHARDT AND HUNT -

40 -A.'S40'

Control -20

-IL%- toot°'

10 15 5 1 15 2 25 2 2 3 4 5 4 5

20 .,

d, , V l o iq n-t e<t

B.>-

25 5 12 1 5 22 2 5 3 5 3 5 4 5 4 5 5 "

L fl p ... .. ", ',"

Fi-K. 6. (Al Distribution of latencies of responses on the avoidance lever to the onset of the warningtone in the control group for the 90 min after sham irradiation, plus or minus the standard error of themean. (B) Distribution of latencies of responses on the avoidance lever to the onset of the warning tonein the experimental group in the 90 min after irradiation, plus or minus the standard error of the mean.These graphs include onk responses made during the presence of the warning tone and before the onsetof shock.

irradiated animals. This implies that the subjects did not respond to the appropriatevisual cue (lights off). Rather, they responded at a greatly reduced rate overall andwithout regard to any of the cues present.

The factors that underlie the radiation-induced decrement in two-lever activeavoidance behavior, such as sensory. motor, or cognitive factors. are uncertain. Onepossibility involving sensorv perception would be a reduced sensitivity for detectingthe electrical shock. This seems unlikely since the increased number and the IRTdistribution of the responses on the avoidance lever (Fig. 4) suggest that the animalsare responding to the shocks. In addition, the pain threshold after exposure to 100-Gy of -y radiation appears to be unaltered (13). The ability of irradiated animals todetect visual and auditory stimuli has not been examined.

The inability to perform the required movements to successfullh complete thetask also appears unlikely, since a general overall increase in responses on theavoidance lever was noted. A reduction in responses on both levers would havebeen expected if the animals were experiencing a gross motor deficit.

Perhaps the findings in the present paper are not the result of a gross sensor, ormotor decrement, but are dependent on the nature and number of' tasks requiredand of the cues associated with these tasks. i.e.. cognitive factors. The irradiatedanimals used the auditory cues (tones), but not the visual cues (lights out). Hogkever. _ _

the cues are not independent from each other in this paradigm. The visual cue isused to obtain the auditory cue. Also, the consequences of' not having these cues

-S i .::.:::: :: i - " -.-.- i :-:: :'::: -.:- : : ...: :i ; :: - ..:.:. .:..:,.:.: : -:-.---- .. .-

IPERFORMANCE DECREMENT AFTER IRRADIATION 157

are different. If the avoidance lever is not pressed at the appropriate time, theanimal is shocked. If the warning lever is not pressed, nothing happens to the ratother than the loss of the auditory cue. Therefore, each cue has a different level ofimmediacy. The animals first response is to attempt to avoid the shock. Secondarily, %they respond for peripheral cues, such as the auditory cue which aids in avoidingthe shock. Irradiated animals may be focusing more of their attention and actionson the shock and the immediate need to avoid it, rather than to initiate themovements required to obtain additional cues. Such a possibility was suggested ina previous study in which irradiated rats performed better in a maze, even thoughtheir general activity was reduced (6).

RECEIVED: December 6, 1984: REVISED: March 29, 1985

REFERENCES

1. A. BRUNER, V. Bocjo. and R. K. JONES, Delayed match-to-sample performance decrement inmonkeys after "Co irradiation. Radiat. Res. 63, 83-96 (1975).

2. A. P. CASARETT, Swim-tank measurement of radiation-induced behavioral incapacitation. Psichol.Rep 33, 731-736 (1973).

3. R. L. CHAPUT and R. T. KOVACIC, Miniature pig performance after fractionated supralethal dosesof ionizing radiation. Radiat. Res. 44, 807-820 (1970).

4. A. L. CASEY and W. L. BROWN, Incapacitation of the Goat fbllowing Massive Doses of MixedNeutron and Gamma Radiation. RTD TDR-63-3077. Air Force Weapons Laboratory, KirtlandA.F.B., NM, 1963.

5. W. L. BROWN, J, E. OVERALL, L. C. LOGlE, and J. E. WICKER, Lever-pressing behavior of albinorats during prolonged exposures to X-radiation. Radat. Res. 13, 617-631 (1960).

6 W. C. BLAIR. The effects of cranial X-radiation on maze acquisition in rats. J. Comp. PhYsiol.PVrchol. 51, 175-177 (1958).

7. G. A. MICKLEY and H. TEITELBAUM, Persistence of lateral hypothalamic-mediated behaviors after asupralethal dose of ionizing radiation. 4viat. Space Environ. Med. 49, 868-873 (1978).

8. W. A. HUNT, Comparative effects of exposure to high-energy electrons and gamma radiation onactive avoidance behavior. Int. J. Radiat. Biol. 44, 257-260 (1983).

9 P. BADIA, S. CULBERTSON, and B. ABBOTT, The relative aversiveness of signalled vs. unsignalledavoidance. J. Evp. .Anal. Behav. 16, 113-131 (1971).

1"), M. SIDMAN. Two temporal parameters in the maintenance of avoidance behavior by the white rat. -

J omp. Phtio/. P.svchol. 46, 253-261 (1953).11 M. SIDMAN. Some properties of the warning stimulus in avoidance behavior. J. Comp. Phtvxiol"

Pv1chol. 48, 444-450 (1955).12 B. J. WINER. Statistical Principles in Experimental Design. pp. 514-539. McGraw-Hill. New York,

1971.13 W. F. Bt'RGHARI)T and W. A. HuNT. The interactive effects of morphine and ionizing radiation on

the latency of tail withdrawal from warm water in the rat. In Proceedings ofthe Ninth S*vmpos'um(in Piht/ogi in the Department olDeh'ne.w pp. 73-76. United States Air Force Academ. 1984.

Z-.° .

• ... . . . . ... o. .

ARMED FORCES RAOIOUIOLOGYRESEARCH INSTITUTE

SCIENTIFIC REFORT

, SR85-27

48 .-

Chemical Physics 96 (1985) 483-488 483

North-Holland. Amsterdam

FLUORESCENCE KINETICS OF EMISSION FROM A SMALL FINITE VOLUME,* OF A BIOLOGICAL SYSTEM

A.J. DAGEN. R.R. ALFANO. Insttute for Ultrafast Spectroscopy and Lasers, Department of Phvsics, The Citv College New York 10031. USA

B.A. ZILINSKASDepartment of Biochemistry and Microbiologv. Cook College, Rutgers University. New Brunswick. New Jersey 08903, USA

and

C.E. SWENBERGRadiation Sciences Deportment, Armed Forces Radobiolog' Research Institute, Bethesda, Marland 20814, USA

Received 29 August 1984; in final form 19 November 1984

The fluorescence decay, apparent quantum yield and transmission from chromophores constrained to a microscopic volumeusing a single picosecond laser excitation were measured as a function of incident intensity. The fl subunit of phycoerythrinaggregate isolated from the photosynthetic antenna system of Nostoc sp. was selected since it contains only four chromophoresin a volume of less than 5.6X 10O A3. The non-exponential fluorescence decay profiles were intensity independent for theintensity range studied (5 x 1013-2 x 1011 photon cm- 2 per pulse). The apparent decrease in the relative fluorescence quantumyield and increase of the relative transmission with increasing excitation intensity is attributed to the combined effects ofground state depletion and upper excited state absorption. Evidence suggests that exciton annihilation is absent within isolated,8 subunits.

I. Introduction cal formalism for calculating observables whenelectronic energy is transported among molecules

Over the past 30 years energy transfer dynamics confined to small volumes. Their results demon-of excited electronic states has been extensively strate that time-dependent observables can be sig-

* studied and applied to a wide range of processes in nificantly altered in small systems relative to theirbiological and chemical systems [1 -8]. Most of the behavior for infinite systems. In particular thetheoretical and experimental studies have been fluorescence kinetics, in the absence of bimolecu-concerned with systems composed of chromo- lar annihilation, are fluence independent andphores randomly distributed in either solutions or non-exponential. An ideal biological system tosolids of infinite spatial extent. There are however, study the properties of electronic energy transfermany important molecular systems where the dis- in small domains is the P subunit of phycoerythrintributions of chromophores are limited to a small isolated from the photosynthetic antenna systemfinite tolurne, e.g., the chlorophyll light harvesting of the blue-green alga Nostoc sp. The phyco-pigments of the photosynthetic unit of green plants, erythrin pigment is one component of the phyco-chromophores incorporated into small micellar bilisomes. the well defined organelles. on the exte-units, as well as polymers which are constrained to rior surface of the thylakoid membranes of thesevolumes of microscopic dimensions. Recently. organisms. In trimer form phycoerythrin ofEdiger and Fayer [91 have constructed a theoreti- blue-green algae occupies a volume approximated.-.-

* 0301-0104/85/$03.30 , Elsevier Science Publishers B.V.(North-Holland Physics Publishing Division)

%' °.%.................................................... ,

484 A.J. Dagen et at. / Fluorescence from a small finite volume of a biological ststem

by a right circular disk of radius 60 A and height 5.0 to effect renaturation, and their identities as-30 A [10]. The basic monomer of the pigment sessed by absorption, sedimentation on linearconsists of two dissimilar polypeptide chains to gradients, and electrophoresis on SDS-poly-which chromophores are covalently bonded; these acrylamide gels. Details of these procedures are aschains, called the a and P1 subunits, contain two described earlier [15].and four chromophores and have molecular Experiments were performed on the /3 subunitweights of 16600 and 19500 dalton respectively component suspended in 0.1 M potassium phos-111.12]. In this paper we report the first picosecond phate at pH 5. The sample OD at the excitationfluorescence kinetic measurements of the frequency was 0.38. The effects of self-absorption

" phycoerythrobolin chromophores in the /3 subunit are minimized because the sample is frontally ex-" of phycoerythrin. We present evidence that the cited and most of the observed fluorescence is

observed decrease in the fluorescence quantum beyond 580 nm, where the absorption of the sam-yield with increasing laser intensity arises from pie is small. A frequency doubled (530 nm) 8 psenhanced transmission (with possible contribu- single pulse from a mode locked Nd:glass laser

tions from upper excited state absorption) and not was used to excite the sample at room tempera-bimolecular exciton annihilation 113]. The non-ex- ture. The intensity of the incident fluence wasponential fluorescence kinetic profiles were found varied by placing appropriate neutral density filtersto be intensity independent and could be fitted to in the excitation pathway. An RCA 7265 PMT p-..-either a double exponential or the Green function was employed for relative quantum yield measure- r""theory of Ediger and Fayer for energy transfer in ments, whereas the transmitted light was moni-small volumes. tored with a diode located behind the sample. A

streak camera and OMA system with a 12 psresolution was utilized to record the fluorescence *"" "'

2. Materials and methods kinetic decays which were digitized and stored in acomputer for later analysis.

Nostoc sp. (Strain Mac) was grown in a 14 iffermentor with pink or cool white fluorescent lightas described previously [14]. Phycobilisomes were 1.0------isolated from these cells according to the protocolof Troxler et al. [15]. Phycoerythrin was obtained '-from dissociated phycobilisomes by use of calcium 0.8[ A! 8phosphate chromatography and sedimentation on r ilinear gradients of sucrose as detailed by Zilinskas Z5

and Howell [14]. The smallest phycoerythrin ag- 0 0.6I

"" -

gregate so obtained (trimers (al)3 ) were removed -from the gradient, dialyzed exhaustively against I 'mM potassium phosphate. pH 5.0, containing , 0.4. '""

0.02 sodium azide, and were then lyophilized. W.

This sample was dissolved at 10 mg/mI in 8.0 M ., rurea, 0.01 M potassium phosphate pH 8.0, and Z 0.20.01 M /3-mercaptoethanol and incubated at 37°Cfor 2 h. The denatured phycoerythrin was then

- applied to a DEAE-Sephacel column preequi- 04500librated with the denaturation solution. The a and WEN 5 80 6 0 780

,'-" ~WAVELENGTH Into) .. •,

S subunits were eluted (in that order) with a linear Fig. I. (A) Relative extinction coefficient for fl subunit. (max)- gradient of increasing NaCI in equilibrating solu- at 550 nm= 310(X) M cm . B) Relative fluorescence

., tion. The peak fractions were pooled separately, intcnite ver.rsus ra,.lngth excitation %avelength 530 nm.

- dialyzed against 0.1 M potassium phosphate. pH T 3() K,

[ '.;'.' .''.".-..".. , ./4 ....... . .'.. . ... "'..",," *S .,..I.' '., ... ,".'... ..- ;' y.c , . . i': .,., -. " '.. '''' . . "-''

A.J. Dagen et at / Fluorescence from a small finite volume of a biological sEstem 485

3. Results indicated by dotted curve) for low and high pho-ton fluence (5.4 x 10-1.7 x 10" photon cm -')

Fig. I shows the relative extinction coefficient are shown in fig. 2. The decays are highly non-ex-

and fluorescence emission spectra of the Pl sub- ponential as can be seen on comparisons withunit. These spectra are nearly mirror images of exponential fits shown in fig. 2A (curves I and 2).

"" each other and are similar to those reported by The decays are unchanged in shape over the inten- .Zickendraht-Wendelstadt et al. [121 for the fP sub- sity range investigated. Emission rise times areunit isolated from Pseudanabaena W1173. The within experimental resolution of the system, i.e.fluorescence kinetic profiles (experimental points less than or equal to 12 ps. The decays reach

one-third of their peak value in about 400 ps.Illustrated in fig. 3 is the relative fluorescence

,-"A , ,quantum yield (defined as the integrated fluores-C cence emission divided by the incident number of

photons, normalized to unity at low laser inten-2 sity) as a function of pulse intensity. Its form is

5 quite similar to quenching curves reported by other.4 workers for complexes of photosynthetic bacteria

[16] and green plant photosynthetic components[17]. Also shown in fig. 3 are the relative transmis-

S"--..0 ." sion data which are approximately a mirror image0 ...., . . . ."" of the quantum yield curve and offset the quantum-.

- . . . yield's apparent decrease at high intensity. Thetwo curves break from unity at 1.75 X 1014 pho- -" . ~Time (psi -'°

tons cm -.

i B . •.*C 0.4. WRO-0.5:.1 •C-0.2. R o 0.5 . '-2' :!C-0 . 4

.R/Ro O

- ." A: ISUBUNIT OF PHYCOERYTHRIN

% NOSTOC sp.

1.2"

.i.-i TRANSMISSION

0.0

0 5 0 0 1 , 0 0 1~o, .5 0 0 Q" 0 .8 , "1"50me ps,0 0 FLUORESCENCE "

Fig. 2. Fluorescence intensity decays for single pulse excitation. 0.6.Dotted curves experimental data. (A) Incident fluence 5.46 x cc

0lW"l photons cm 2 per pulse. Theoretical fits: (I I) exponential. Z 0.4.

A = l.Ox I09 . (2) exponential, k =1.6x 109 s . (3) double

exponential 0.65 exp( - k)+0.33 exp( - kt). k = 1.0 x 10 EXPERIMENTAL ERROR

s (4) Paillotin et al. [131. exciton annihilation theory withr 0.1, Z = I and k = 1.0 x 109 s 1. [corresponds to0.65 exp( - At) +0.32 exp( - 22kt)l. (5) double exponential......... ........ O.. "...... "...O.060.65 exp( - A,)+0.33 exp( -44kt)• k = 1.0x l10 s . (B) Inci- 1014 10dent fluence 1.75 x l101 photons cm 2 per pulse. Fits to the INTENSITY IPHOTON cm

2 )

expression II + (G,(N. V, i))exp( - ki). A = 1.72 x 101 sI Fig. 3. Relative apparent quantum ,ield and transmission"*":" N= 4 with 0 " = 0.4. R/R,, = 1.0; A C = 0.2. R/R, = 0.5 versus incident single-pulse fluence (photon cm 2). Solid line " ." -

and* C 0.4. RR,, =0.5. denotes theoretical fit to exciton fusion model with r = 0 1"131.

y

..............• ° , ,-........ ... o o.-"

.-.... 7- - .7. -

486 A.J. Dagen et al. / Fluorestence from a vniallfin;te olume of a blogual stvfem At

4. Discussion vided the incident intensity is less than the satura-tion intensity: the saturation intensity being de-

The kinetic curve profiles are intensity indepen- fined as that intensity required to obtain a fluores-

dent with an overall e ' time of 400 ps (see fig. cence yield decrease of 50% relative to its low*, 2). They can be made to fit the exciton annihila- intensity yield. Without a quantitative theory. con-

i e P i1 w ltributions of upper excited-state absorption to the *., -M.. tion theory of Paillotin et al. 1131 with values of a,., _

r= 0.1, k = 1.0 10 s and Z 1.0. where r is observed decrease in fluorescence quantum yield %Mopr01k=LX1 ' cannot Z= rulede isut.

the ratio of twice the monomolecular decay rate to cannot be ruled out.

the exciton annihilation rate and Z y(l + X). It is possible to fit the fluorescence profiles to

where the constant X < 1 and 1 is the average an expression of the form exp(-at" - kt) withk=4.4x 108 s ', a= 1.27 X10' s" : however,

number of excitons per domain at ts= 0. The the- such a fit must be considered fortuitous since thisory functional behavior is predicted under assump- ex.ssos"-fluorescence decay curves. For these values of r tion tha ior is pice une ,8su.

and Z. the decay is approximately a double ex- tions that are not applicable to the 3-subunit.ponential of the form namely energy transfer limited to two-dimensions

in a system of infinite extent [2]. Our data can best

0.65 exp( - ki) + 0.32 exp( - 22kt). be explained by ascribing the fluence independent,non-exponentiality of the fluorescence kinetics to

The observation that the decay profiles are inde- emission arising from more than a single chromo-pendent of excitation intensity however invalidates phore. A good fit to the decays can be obtained A

interpreting the non-exponential decay in terms of assuming a minimum two-component model (see <1

exciton fusion since a five-fold increase or de- fig. 2A, curve 3). The zero of time in figs. 2A andcrease in Z should produce observable shape dif- 2B was set at the peak of the fluorescence inten-ferences in the fluorescence curves. For the /3 sity. All theoretical curves were set equal to thesubunit, 50 = 310000 M-1 cm-' [12] implies ol maximum intensity.= vI at an incident fluence of 1.3 x 10

15 pho- It is interesting to note that Fayer's approxima-ton cm- 2 and therefore variations in Z are within tion for the solution to the rate equations govern- -..,-

the experimentally probed intensity range. The ing excitation on a finite number of chromophore,extreme sensitivity of the shape of the fluorescence randomly dispersed in a finite volume, also pro- .. .-

decays to changes in Z over the excitation inten- vides a reasonable quantitative fit to the kineti.sity range investigated ensures that we would have profiles. Several theoretical curves are shown in -noticed such changes if bimolecular exciton anni- fig. 2B. The key quantity in Ediger-Fayer theoryhilation processes were responsible for the non-ex- [9] for electronic excited-state transport among aponential decays. Furthermore, the experimental finite number N molecules distributed randomlydata in fig. 3 demonstrate that corresponding to in a finite volume V is the probability. G( N, V, t),the decrease in the apparent fluorescence quantum that an excitation is on the originally excited chro-yield with increasing pulse intensity there is an mophore at time t in the absence of decay due toincrease in transmission. The theory of exciton its finite lifetime. G,(N. V. t) is a function of bothannihilation as given by Paillotin et al. [13] is N, the number of chromophores in the domainknown to be applicable only if the transmission is and R/Ro, where R0 is the Forster critical radiusintensity independent. We cannot offer a quantita- [3,4] and R is the radius of the sphere which ..

tive explanation for the single pulse data reported approximates the volume of the domain (for the #"in fig. 3. However, the observation that the relative subunit R is - 22 A.transmission (TR) and apparent fluorescence yield In terms of the response function. G,(t ). the(4R) are approximately mirror reflection of each fluorescence decay F(t) we observe is given byother about the unit axis is strongly suggestive of F(t) = F,, (t) + F, (t)ground-state depletion since under steady-state ex- ep-[','.V.'(citation a three-level system gives OR Tjt pro-

, ... "

*A.J Dagen el al. / Fiuoresceno' from a ilnaII (iie tedunte 'a hi,~fjj ad w 6m .

where C is a constant which depends on the initial state, H-ence in contrast to the non-linear prticessexcitation distribution and the chromophores' in- of singlet-exciton annihilation ohsersed in largertrinsic anisotropy and F, and F, denote the size aggregates. of the a and 11 s-ubunits. Nmw ciiain-fluorescence emission viewed perpendicular and pie the c-phscocrsthrin 11l41. it appear% ihmte'parallel to the incident polarization direction. AN ton fusion is absent in %er% %mall domains [heiindicated in fig. 2B a value of R/R,, on the order critical size of a domain for the inoperatiseness ofof 0.5 with k = 1.72 x 109~ s 1gives good fits w~ith eiiciton annihilation in the general case presurna- .'

N = 4. The neglect of the rotational diffusion inl hlN depend% not ,nIN on the nurnher of chromo-eq. (1) is justified since T,, iVK 10 N SAt phores hut also on the rate prmptirtional to R,,. P1.room temperature assuming 17 is unit,.. The four Pi - N at Ahich encrg% ms transferred Aimong them. ~chromophores of the /3-subunit have definite A general theor. of %hen ex~iion annihilation isorientations relative to each other and therefore in inoperatise for a small aggregate of chroniophores 7a strict sense violate the assumptions emplo~ed in constitutes an intriguing theoretical probkm stillderiving eq. (1). Either the fits to eq. (1) are lacking a solutionfortuitous or for small R/R,) and N. eq. (I1) has amuch broader range of applicability than therandomization of chromophore distances (aver- Acknowted~ententaging) implies. This could be rationalized by as-suming that few configurations contribute for A. lDagen and R 4dfaiio thank NSF-. ALFOSR,R/R _< I with N small. There is also the possibil- NIH and PS( BIHf of ( t NN for support and B.ity that the solvent produces slight positional shifts ZilinskaN ackno-Aledges support of I SD). andof the #3-subunit chromophores thereby providing NJA[S state and t S Hatch I nds Wke thanksome degree of spatial randomness. For N fixed. M r. P. Sekuler for iechnical help and D~rs N1 1)eq. (1) constitutes a three-parameter k. R/R,,. and Ediger and M.D. Fa~er for their :mputer pro-C. For example. with R = 22 At (the radius of a grams. A. [)agen thanks Ms I Blurnan of Bi-ndi\sphere approximating the /3-subunit volume). Corp. for technical assistan.e.

RR,=0.5 as determined from the fluorescencedecay curves implies a Forster transfer distance

* R) on the order of 40 A. This is comparable to Reference%R,= 21 A determined for the a-subunit 1181.It' summary, the simplest model consistent with [1l t) L. Huber. Ph\, Re\ 20 19Yi9m 23~4C' 2o11 it I I

is 121 A. Blumen annd J Mani. J ( hem Ph\, s 65 1'h 11 XN'the experimental data reported in figs. 2 and 3 is 131 Trh. Forster. Z. Niurorch 4a (1949 121

that the domain size is too small to accommodate 141 R.S. Knovt. i Bioenergeic o.r phoioNsnihems. edmore than a single excitation, a view consistent Gosmndjee (Academic Press. Ne~ York. 14-751 pp IX0 221

*with the onset of bleaching at a fluence of -2 x 151 S.W. Hahn and R ?%%aniig. J (heni Ph\, tX i4l'Xi

* 1014 photon cm 2 (y 0.2). Our model suggests 17thatafte theabsoptio ofone hoto by sinle 6) R. F. Loring. H.( .Anderson and M t) Faser. J (hemvitha afer he bsoptin o on phtonby siglePhvs. 76 t1982) 2015.

chromophore the excitation energy is quickly dis- 171 P.Y. Lu. ZX Yu. R.R. Alfano and J I. Gersten. Ph%%persed among the other three chromophores com- Re%. A26 (1982) 36101.

prising the /3 subunit. The time for randomization 181 G. Porter. CT. Tredvwell. CTF.\V Stale and J Barhtr.

of the initial excitation is on the order of hundreds Biochem. Biophvs. ca 5011 11978) 232

of picoseconds. the time it takes G,( N. V. t) to qMt)EieradM).asrI.(hmPs 'iI932518.approach 0.25. The fas- rise-time ( < 10 ps) is (101 D).A. Brsant. G. Gughlmi. N T. deMarsac. .A M (aoteisindicative of the time for the initial emitting chro- and G. Cohen-Bazire. Arch. Microhiol. 123 i1979i 113

mophore to begin fluorescing. The bleaching indi- [III AN. Glazer. Mot. Cellular Biochem. 18 (1977) 125catd b fi. 3is ntepreed n trmsof he (121 B. Zickendraht-Wendelsiadi. 1. Friedrich and W Rudiger.inabiliy of 53Pholochem. Photohiol. 11 1980) 367..

excted 0 m potos t conec th dobly 1131 G. Paillotin. C'E. Swenherg. J. Breton and N.E. (icacin-

ectdstate of the /3aggregate to its first excited to% Biphs 1. 25 (1979) 513,

US.4 J Dagen et at / Fluoresence from a small finite volume of a biological sYstem 4

t141 A~ /lnski% ind 1) A fiowell. Plant Phvsjol. 71 (1983) by ultrafast laser spectroscopy. ed. R.R. Alfano (Academic1, Press. New York. 1982) pp. 158-189.

11- R' R I lroxler. I S (,rcen~ald and B.A, Zilinskas. I. Biol. 1181 A.J. Dagen. R.R. Alfano, B.A. Zilinskas and C.E. Swen-Shemi 2 S' (~~ 19404I berg. submitted for publication.RI ..in (rondelle. ( % Hunter. J.;( Bakker and H.J M. 119) D.F. Wong. F. Pellegrino. R.R. Alfano, and B.A. Zilinskas.Krrni kiw~hem Bioph % Acta 723 0143) M0. Photochem. Photobiol. 33 (1981) 651.

I \I (watinto% ind J Breton. in Biological e~ents probed

e-.

% %r

?-..,

ARMUED FORCES FOADMOI11OLOGY u,

St'* RESEARCH INSTITUTE

SCIENTIFIC REPORT

GASTROENTEROLOG SR85-28

Effect of Ionizing Radiation on GastricSecretion and Gastric Motility inMonkeys

ETIENNE DANQUECHIN DORVAL, GREGORY P. MUELLER,ROBERT R. ENG, ASAF DURAKOVIC, JAMES J. CONKLIN, andANDRE DUBOISDigestive Diseases Division. Department of Medicine. Uniformed Services Universitu of theHealth Sciences: Department of Physiology. Uniformed Services University of the HealthSciences: and Nuclear Sciences Division. Armed Forces Radiobiology Research Institute.Bethesda. Maryland

The prodromal syndrome of radiation sickness is emptying and gastric secretion, while increasingcharacterized by nausea and vomiting but the plasma levels of immunoreactive O-endorphin.pathophysiology and the treatment of this entity is Domperidone had no effect on vomiting or gastriclargely unknown. We investigated this problem by function either before or after irradiation, but itdetermining the effects of ionizing radiation on significantly increased plasma immunoreactive .3- - -

gastric function with and without administration of endorphin.the dopamine antagonist domperidone. We mea-sured gastric electrical control activity (waves per Emesis occurs immediately after whole body irradia-minute), fractional emptying rate (percent per min- tion and is the most obvious and the best docu-

. ute), acid output (microequivalents per minute), and mented prodromal symptom of radiation sickness (1,* plasma levels of immunoreactive Oendorphin. 2). Time of onset, duration, and intensity of this

Twelve conscious, chair-adapted rhesus monkeys vomiting depend on the species (3) as we!l as on thewere studied twice before, once immediately after. type. dose rate, and total dose of irradiation (2).and once 2 days after a single 800-cGy (800 rods) These symptoms are clearly different from those"°Co total body irradiation. In addition to causing observed during the intestinal syndrome, which oc-vomiting, total body irradiation transiently sup- curs 7-15 days after irradiation and is characterizedpressed gastric electrical control activity, gastric by diarrhea, often accompanied by intestinal bleed-

ing (4).Received May 25. 1984. Accepted February 5. 1985. We recently studied emesis produced in dogs byAddress requests for reprints to: Andre Dubois. M.D.. Ph.D.. total body v-irradiation and evaluated tie concur-

Department of Medicine. Uniformed Services University. 4301 - a e"d"c uJones Bridge Road. Bethesda. Maryland 20814-4799. rent effect of irradiation on gastric emptying. We also

The present address for E. Danquechin Dorval is CHR Tours. determined the efficacy of an antiemetic agent, theHopital Trousseau. 37044 Tours Cedex. France. dopamine antagonist domperidone. on vomiting as

This research was supported in part by the Uniformed Services well as its effect on gastric emptying. In this dog ,',.University of the Health Sciences Protocol No. RO-8342. model, gastric emptying was suppressed during the

The opinions and assertions contained herein are the privateones of the authors and are not to be construed as official or prodromal syndrome of radiation sickness. Preven-reflecting the views of the Department of Defense or the Uni- tion of radiation-induced vomiting with domperi- -,-.'formed Services University of the Health Sciences. done did not improve the suppression of gastric

The experiments reported herein were conducted according to emptying (5).the principles set forth in the **Guide for the Care and Use ofLaboratory Animals." Institute of Animal Resources. National The present studies were undertaken to furtherResearch Council. DHEW Publ. No. (NIH) 78-23. examine the relation between radiation-induced

The authors thank Dr. I. Long and M. Morton. lanssen R and D vomiting and stomach function in an animal modelInc.. N.J.. for their generous supply of domperidone and placebo. that appears to be closer to humans in terms of brainThe authors also thank M. Flynn. I. Stewart. I. Warrenfeltz. andN. L. Fleming for their valuable support in animal handling and Abbreviations used in this paper: DTPA. diethylene triamineradiopharmaceutical preparation. and I. Barchers for her expert pentaacetic acid; ECA. electrical control activity: FER. fractionaleditorial assistance. emptying rate; io-END. immunoreactive Oendorphin.

: . . . . . . . . . . . . . . . .' . .- - . .- - - ,

Auguist 1q85- GASTRIC, t-t \TRt)\ VOMITING; AND) RAIiATION 3175

1 Control Day 3. Irradiation Day polar recordings displayed periodic waves in) thle 31il> > range that have been shown to c:orrelate with gatstric. ECA

EE when gastric serosal electrodes were used in conjunctionU? with skiui electrodes (6.7). Each fasting anld postluiad trac-

ing was examined blindly by one of Os (E. 1). D.). First, the

2. Irradiation Day 4. 2 Days ed 'sas determined by visainpcon(g.Fgue1Postirradiation panels 1. 3. and 4).' Of the 96 tracings obtainedl (12

monkeys studied onl 4 separate days during fasting and~t~w> ~ -after the load). 77 (801),,) could be completely analyzed.

E 2 ~Within each tracing, artifacts due to intragastric mixing orLr movements of the animals (e.g.. retching and vomniting:

.e~ch~nqFigure 1. panel 2) made the recording inadequate (luringlnoirei I F~mpe if wri iding of vitfjtic a t firiil d.ti%-itN -I 10%i of thle time. Mean fasting and pustload frequencies

otied u-it n abdinl b ipl ar eletit riii ill I liol obtai ned in each animal onl each stud v das' were used toku's. 1. Coiiriil ita\: 2. slow at fiisit\ prv(iniedg si\ cmuethe gadmean ISE) for each t\ype of study' inri-l Itiug episitus (.hard( trerizist lv buirsts of sliasti( ecmpute grfandmlsfliiieRIWietS ot t1W Chi-SI Mrid Adh~Ul li rrovs hit- ea;gripofnmlsIi uvei biv dt sol his: 3. i rraiationl (it, noii vimitin bu t Vomiting wsas defined as a suc:cess ion of strong and briefshm isg ot itt-u tni aat ixi it\v: 4. 2 (lay salr irradi atioil: contrac~tions of t horacic and abdomninal muscles lead inmg tonoi ithange' i.oiiilildi with 1. the expulsion of gastric contents through the mouth:

retching wvas defined as a nonproductive vomliting (8).* ~organizationi andl gastric: futnction. We prodluced Durinig both events. recordings displayed a succ(ession of

vomiting in rhesus mocnkevs with af single dlose of brief bursts of high potential spikes (Figure 11 that were*total bodv' irradiation and we measured gastric acid clearly different from the movement artifacts that wvere

secretion. g'astric e mptving of liquids, and( gastric: sometimes superimposed.*electrical control a(lix'itv (Et,'A) before, during, and A previously described and validated marker dilution*after the acute prodroinal syndromle of radiation technique (9.10) was used conurmentlv to determine gas-*sic:kness. Ini add~ition, we detertmined plasma levels tric Seuretion and gastric: empty ing (luring a 40-nun fasting

*Of ifmmul~noreactive f.3-endlorphit1 (i[3-END) and evalu- period and for 60 miii after the injection of an 811-mt wvateratedtheeffct f dmnpridne n ech f tese load (postload period). Ini the present studies, this tech-atedtheeffct f dmpeidov o) vch f tese nique was slightly modified in that ""iiiTc-DTPA fdietliv-

p~arameters. tene triamille pentaacetii: acid) was used instead of phenolred as the mnarker. This intubation method requires oil]\-

Materals nd Mehodsthe sequential sampling of the gastrlic contents and permitsMatrias ad M thos onclurrent measurement oIf i utragastric volume. gastric.

Twelve consc ious. cha ir-ala ptedl rhesus mon key's empltyving, a ii d gastric sec~ret ion. A 12F" double- hI nenwere studiled ol 4 se parat e ulavs after an iovern ight fast: a) nasogastric tube was placed] ili thle stomach and its posicont11rot day after i A. admniin ist rat ion (If placeto, lb) co11nil t ion wvas verified by the water recovers' test (11). Starting(lav- after i . . ad in listratio of11 ut(111 ipern iie ( 1.0I ing kg 45 inl later, saul es of thte inli ed gast ni: .ont ett wei-retlodl wt(. ((:) irradiation dayt\ after administration ot either aspirated just befiire anl(l 'inmediately after i utragast nitp lacuebii ior (tom per) (lone. and (d ) 2 davs after i rrad ia tion adin istratin (If 5-20 fill ofif a ( i -DTPA test soluitioni(no itrug( gi veil). Tb is dlose of dom pen done was selected (30 pCi 1 t)m 10 H74 t7-C n( ee(itr ueba sed il preiou~is expertimenliots iii tilie (log show inig thIat The clear suplerinatanlt ot each sam plei was assaveil fursimilar (loses dlid not produclie ally* Sidle effect. Plaicebi anid Tc-)TPA' concentratioins uising anl Ultrogamiina auito-(11111pen done were given hlInd1l1v a 11(1 in railuni iorder lull gainma couinter ( LK 11 Ist rumienit s, Tourku. Fin Ian idad forcoiintriil days. Stud ies were performed ill the moilring alid titratalile acidity using eleictriinetric titratiuin toi ti 7.4started :101 iin after (drug adiiiinistratiiii and( 211 lii aifter (Radiometer. Copenhagen, Deiiniark). 'These uleteriiina-- -

either siani irraiation oii icintriil ulavs or after irraiationl lions were repeated every It) miii iluring tlii( basal pierioid-oil irradliatioin day. Onf .ontrol ulays. the aniimals were and after intragastric instillation of anl 8011il water louadl*

* ~~brouught to tile exposure rooimi arid thel doiirs were uluiseul (:iitainiig Tc-l)TPA (3 pCi 100 fil: pli 7.4: :37 Ql. --

- ~~foir :3 ruin. Off irradliation day. eachl nioiie was ex5poised toi iltragastri(: v(o1 lties oif fluoid (V, 1.'. . . . )and( arnoui s ill8001 i:(;%- (8itt rails) tiotal tiidv irraidiatioin dtelivereid at 5til ifT(:.DTPA (Ti:. To. I were idetermineid at tile tulle otcGv loin by two large. toi Ci ""Coi irraitors placed eaich sampling using the dilution priniciple (.t~ . )

* ,iil11teriuirly aunrd poistiriorly. Phanitomi studie's idemionstratedl Fractional emptying rate (g,( wais then dleterinedli fiir each:1* ~~that tlti midlin. ail(lien rerevivd 800t ~ anii thit tli 111-mm interval (P) bietweeni twit diluitionis. assuinig that* tichadl reiciveil filiti:O v eilultving was af first-order process (exponentiaul ( luirinig a

l':i~hironky ws isual Imoiniitioreod for :t II (lii iconitriil gie-i-ii nevl anid uisinlg the fuillowing equai~tioni:

* iavs anii Ii If oin irradiatio 1111 days. Hipoiuar tile( tril ill Illtil-IloT(-T )I.tials were reruirdeil from two abiidominial ulisposalli skill ug i 'ii)

* elvitriiies ol iinulltilIhaiiiel reciirder l leukiriali R612, Ber.ausi' g is allowed toi vary from intvrv~ul to1 interv~il. noIliu, kiman Iunstrumnitis. Sch illeIr Park. Ill. (. Abdoiiinial I(i - general assuminpti on has to be made regard inig emtvining

% %S

37tDANVECIN DRVA I- ALGASTMI *NTEROI.()(;Y %'I[ 8q. No.2

p. 37h I)-\~ CIN f JKAI D Elp'iol n rdai oil (ostric Elctrical Activiti fi Cvcles Per' M\inu te (.\ lu i i SE)4

Il(.)( o m. Placebo dJonera ( ft Iio

p.El _3 ii .I17 OA12 2.861 01.191 2.66 0 (.18' 3.11 (JM-

III 12) 111 41 (11 3 1 In 51 In = I

Inl III Illi 1 :) it 41 (n !)I

1) Iom- mup.'d l ith oroi-piint g %alO,' oil lolltrol ,tLo,.

over the totil lIuratili of the( experimnt. Net fluid outpuit Results(f, I i miIiliters per minute %\,is thien determined for thle

lmorresponin~ g inIterval . aIssuinjg that it remaline 1111.oil- ovmtn or retching was observed onlstant over the giv-en intervA and using the following (control davs. i.e.. after shiam irradiation. Onl irradia-

K'eqiti11:tion da " , episodles of retching or vomiting. Or 1)0th.B i'jil'211 g i's( 211.were o~bservedl andl recorded (Figuire 1) inl 5 of 6

Inlt ragust rict v oumes of f lid loll masses5 of ...... i)-TP-\ mon keys ill tlie placebo grou~p and inl all the ml) 1-

%%-tre (heln reflallilated, taking into aunli01lt these first keys inl the domiperidone group. Ini both treatment .

estimates o~f fri ia 111111011 t\ vinig andii flid iot 111t, w hi ii grotips. these episodes started 29 :s iiimin Imea 0

'eoill turn reicalcul11ateii.' [This iterative process 115 SE anrd ended 71 -9 mnin after irradiation. D~urin grepealtedlt Iitil tile- imlprovemlenlt of tile soltioin bec1a101 fasting. 1 ntke\y inl each grou ip exhlibited retching

P, 1p ter iteirat ion . Hav ingl prey 11u181v dleterm~linled int rigils- alone and1( 9 had retching assic iat ed withI vominit ing (4I ric 1.10ltrat niriills oif aid (Al . A. we thlen c.ah. u ated after p lacebo 80(d 5 afteor do ml eridliel: 4 mo(n0keys[lt aid ill tpult (t i13 \18 uing tie( to Illow ing equat11ioll: v'oni tell uil(:. 2 vom111itetd twice. 2 vo'li)ited t hreeJ

IV .A -p !0i [1 -p('01 times. adl 1 vornited four times. After intragastri(:c~-

I'leseI t~cilitins erepefored sig Ilo(iil\-devl- administration of the water load. retc'hing- 111(5 Ilier W* tes (11Ii li t i ils~vor ll~tl rlle I111 ~ga lii.111V lie - observed 8101ne and( only 6 oli 1 2 ml)keys halldllpe'( riigrai i a l'' it ili1 cmpulter ( Divisiioi oif Coln-

((lte Rs ' 11a 1( Ti 11111 I g. at11111 l~titilesut retching and vomn1iti ng: w hen p reset. voiti 11 g 01:-

Heath. Beithlesda. XMd.. Thie al.sumpltionls imlliiI\ l it\(' ciii rrd only once and alwayd\s ill mionkevs who hadbieenl (111ri I ei an d111iscu1.isselc1sewhelre M01 ant d 'Il1 re alreadx vorn ited do i rig, fasting. B\V 2 dayVs alter

1080111111 till, llrigini o nliitributionls b\ I lililes dliii Duinliip irradiationi. I otlititig lciurrel tin otilv I mitikeX':

f(2 I-l and Im Geo'lrge It. viii Ili otrast tii their meithodil. this was at the end1 of tillo stud\lV 1111(] after mo1(st lf it(h

hiioel.i'r. (lie piresenit tei(iniqiii illows i iirrei tiiii for (111(- stoimach~ contenits hail been aspirated.

tv iiig aiii set rwtilil oii ilrrilig dulring Ill' i-min five itill- Fastmno ''astril ECA is(!Iil ted ll om1 1-'lr '

tull iiiti'ri adl liii be aitpdo'il diiriiig lastingO. ( )i irridiil- was 3.3 1 01.10 per mlititI oil cointroil llivs Alter the(ion 1(i\. inte'r\,& %\(1 ilil firriit I vwn leo luoitin!_g wire nit tilachl and1 w\,is not) sigtifilly altered biv Illolmhet-

stairt adi it tile endi (ix'.. 2 Ii after sini irradiaitioll or trio'illil leesdale r-l~ltit.iii ''tItl( I

irradiationi) ilf em It ,tud\1 using a (lrvioulsk iil' rifled cotiroii level 2 days alfte'r irradliationt ([igurI' I and~

* n~~~ri iigmii lilt, rvgioli it 13-ENt)1 , I lirrt'spoliiiilg toi imi- lower after the i' lter loadi but (i iilhr'il'c(i \%ilt o

noi ill,1 throughi '-i , id .Ii'uuiiiu . et. lii .111i (-1-'NI aftertiii' w te 'lou1 c lit ll r' Iili lsii: ti

August 1985 GASTRIC liN(CTI()N. VOMITING. AND RADIATION 377 ,

Table 2. Effect of Domperidone and Irradiation on Fractional Emptying Rate. Acid Output. and Fluid Output

Control day Irradiation day

Domperi- Domperi- 2 days after ',.

FER {%o min) .%",Fasting (1) 4.1 t 1.3 3 0 1.2 0.4 -03 0 30 '34 ± 1.5Fasting (2) 5.5 t 1.2 19 1 :3 0 3 0 2" 06 - 0 V 3.3 1.2

Postload 11.8 = 1.8 11.3 ±13 08 0 " 1 3 (t4" 12.9 ± 2.2AO (AEq min)

Fasting(1) 8.5 : 3.0 8.8 : 3.3 54- 2 2 32 : 1 0 15.2 65Fasting (2) 9.7 : 3.8 8.1 - 4.5 0 - 0W 00 : 0 0" 14.6 ± 5.0Postload 24.4 = 7.0 21.1 : 8.4 6: I 6" 00 0.01 27.8 : 8.1

FO (ml min) [[['

Fasting (1) 0.21 t 0.06 0.16 t 004 040 - 0.12 030 - 0.06 024 : 0.06

Fasting (2) 0.20 t 0.04 0.14 = 0.03 0.23 - 006 0.27 - 006 0.24 : 0.06

Postload 0.30 :t 0.04 0.29 = 0.04 0.12 : 0.01 0.13 : 0.02 0.23 = 0.05

AO. acid output; FER. fractional emptying rate; FO, fluid output. Fasting (1) corresponds to the period 0-20 min after the start of thestudy; on the day of irradiation, this period was 20-40 min after exposure. Fasting (21 corresponds to the period 20-40 min after the start

of the study; on the day of irradiation this period was 40-60 min after exposure. Values are mean : SE. p < 0.05 compared with

corresponding value on control day.

found between gastric ECA and mean postload FER idone group. After the load. fluid output was

of the stomach, the best fit being obtained with a significantly suppressed in all monkeys (Table 2).

power curve (ECA = 3.0 x FER 0 21 ). As domperidone had no significant effect on emp-

Acid output was significantly stimulated after the tying, acid output, or fluid output, either in the

load compared with fasting on control days (Figure control state or after irradiation (Table 2), values

4): domperidone did not significantly modify either were averaged for all 12 monkeys and are depicted in

fasting or postload acid output. After irradiation, Figures 2-4. Two days after irradiation, all gastric

acid output was suppressed in all the monkeys who parameters had returned to control day levels, even

had secreted on control day; the suppression started in the monkey who had vomited on that day.

39 - 4 min after irradiation, persisted after load As shown in Table 3, plasma levels of i/3-END

stimulation, and was not affected by domperidone were elevated markedly by domperidone and by

(Table 2). Interestingly. acid oitput was not abol- irradiation with placebo; the effects of radiation and

ished in the only monkey who did not retch or vomit the drug were additive. Two days after irradiation, ,..

after irradiation. plasma concentrations of iG-END returned to basalFluid output was significantly stimulated after the values. Plasma levels of i3-END were significantly

load compared with fasting on control days. After (p < 0.05) and inverselh correlated with all gastric

irradiation, fasting fluid output tended to be in- parameters (FER: r = 0.63: ECA: r = 0.61; acid .'.',.

creased compared with control days. but the differ- output: r = 0.71).ence was statistically significant only in the domper-

100- = j,,adlotron Day

15 0, 90 1 fin absence of

Bass, 701 1 = 7 r --6Pos" yIo°-'ngt. "

12 5,

0 1 Co o = "

100- 00 1aa

aifeS a E

-a 2 DV

U 042450702 Da, n Currt"o', .%pos:,~a . s0 .,, r..., . ,

S30 D0 <0 800If: o Do % -

2o~ o 5,o~o 0o . . .,=2. ""-;

114 "0 20 30D0 0y0 s

C o n t r o l D a s 2 4 0 4 0 6 8 0 - . a t e,'

ON o s 6dotor. Figure Effect of irradiation on the percentage ot the loadM,,i.,, after I,,,do.o Fgr 3.Efc firdain o h ecnaeo h od . _

Figure 2. Effect of irradiation on fasting and postload gastric remaining in the stomach over time Values are mean .

fractional emptying rate. Values are mean - SE SE.

_1, ' *.- '.e -

378 lANQU'ECHIN DORVAL ET AL. GASTROENTEROLOGY Vol. 89, Nom 2 * '

37 DANL, %

4- agas first described in 1922 [quoted in Reference 15).

Fos,9 <o o0 05 84-1 .. The frequency of this cutaneous electrogastrogram is-30 - - .. correlated with the slow-wave frequency recordedPek Meian c I_,, ,2 F1 , from gastric serosal electrodes (6,7,16-18). In the

2 [0- T- present studies, the frequency of ECA was signifi- .

cantly decreased on the day of irradiation and hadL- returned to basal values 2 days later; this observationhas been confirmed by preliminary results obtained

2040 4060 70 20o 2 Days in monkeys with implanted gastric serosal elec-c . o aft ,..adv o PO.. S . .t.o trodes (7). The radiation-induced decrease of ECA

Figure 4. Effect of irradiation on basal and load-stimulated acid frequency accompanies a concurrent decrease ofoutput. Values are mean SE. gastric FER both during fasting and after the load. - .-

This latter finding is similar to that observed in dogsDiscussion (5,19) and in rats (20).

Our observations demonstrate that ionizing radia-In the present studies, we report the precise tion has a different effect on gastric acid output and

and objective measurements of the immediate occur- on nonparietal secretions. Immediately after irradia-rence of retching and vomiting in rhesus monkeys tion, acid output is suppressed both during fastingexposed to total body irradiation, as well as the and after a water load (Table 2 and Figure 4). Thisrelation between these events, gastric function, and suppression of acid could be due to ultrastructuralpituitary ip-END secretion. changes of parietal cells similar to those observed in

The visual distinction between retching and vom- the mouse within 30 min of exposure (21), but it isiting may be difficult in fasting monkeys, although it clearly different from the hypochlorhydria due tois easy in the presence of a large vomitus of food. gastric atrophy that appears several weeks after irra-Animals can either store a small vomitus into their diation (22-24). In contrast, fluid output is sup-cheek pouches and then swallow it. or they can emit pressed only after the water load, whereas it remainssome foamy saliva after a nonproductive retching. In unchanged or even tends to increase during fastingthe present studies, recording of skin potential (Table 2). Thus, fasting nonparietal fluid secretionshelped in differentiating between these two types of appear to be increased immediately after irradiation,events, demonstrating retching and vomiting in 9 of thereby masking the concurrent suppression of thethe 12 monkeys and retching alone in only 2 mon- parietal component of fluid output. As nonparietal -4keys. This dose of 800 cGy has been selected because secretions are not stimulated after a water load (9),it is twice the ED50 for vomiting as previously no change of fluid output is expected during the %determined by others for monkeys (1,8). Retching postload period if acid output is suppressed. In fact.and vomiting started after a delay of -30 min and the significant decrease of fluid output after thedisappeared after 70 min, which agrees with the water load indicates that the effect of irradiation on .'"

observations of Middleton and Young (1) for similar nonparietal secretions is biphasic, consisting of an .,. -

doses of exposure, but markedly differ from a delay initial stimulation followed in 40 min by an inhibi-of almost an hour with doses between 400 and 550 tion of fluid output. Two days later, however, bothcGy (1) and a delay of <5 min after a dose of 1200 parietal and nonparietal secretions have returned tocGy (8). Thus. the interval between irradiation and basal values.vomiting appears to be inversely proportional to the The relation between radiation-induced emesisdose received, and gastric inhibition is unclear. As acid suppres-

The incidence of vomiting after irradiation ap- sion starts at about the same time as vomiting and ....pears to increase if monkeys are fed solid food 1-2 h persists after its disappearance, these two symptomsbefore irradiation (8). In our study. however, a 16-h may be closely related: this possibility is also sup-fast before irradiation does not appear to have re- pduced the incidence of this "radioemesis": more- Table 3. Effect of Irradiation on Plasma Immunoreactie

O-Endorphin in Picograms per Milllter "-'"over. intragastric administration of a water load after (Mean n SE g p it "irradiation is associated with only 6 vomiting epi- control•sodes versus 18 during fasting, suggesting that the day 125 mn ater irradiation 2 daysincidence of radioemesis is actually decreased by Domperi. Domperi. after * -.

gastric distention with noncaloric liquids when irra- Placebo done Placebo done irradiation %diation is delivered at a high dose rate (500 cGv 229 = 48 881 - 88' 1447 0 22W 3389 0 702' 159 43mini. 'p < 005 compared with placebo p "001 (ompared ,with

Gastric ECA may be recorded via skin electrodes control day.

. "- . . . .

. - . -..--

Atigu~t 11085 CA-STK(:IC 7 I AHV MI tIIMIi .\\lJ R.\l)i t\ i7li

portedt hi. our aniecdtotal firlil.Ig that acid ou~tpUt Was stomach is respoinsible for tli' supjirvssion of oastriluiichdilgt'd iifttr irradliationi iii fte on~v iionkt'v that erriptviiig. The thrle of ooisu't of tatli sviiifioii atttrdlid riot retch or voil'iii. Ili contrast. rauliatioii-iuitdlier irra(Iiiii aiid tlhe transiencyx ot fit', it( fte ttrodtrtial

C.suppression of goastric einpiving and of ECA is ohl- svidronu' to radiation sickiness sugge. st th( inilvoivv-ser-ved even il tlhe 2 mioinkeys that dlid not vomnit: ilntuIlt Of 1MNo neurieohiriniial met( himisins. orfurthermore. ill the aiiials that did( v.omnit. g-astric both. or it recepjtor iliactivititon (32). The rt'l(',si it *

(tnptving, suppression starts blfore. aitid persists atl- pituitary it -JENI) dltiiitistrateul bv flu', pre'sent stood-I ter (lisappt'arailce of. eiiiesis. lTus. radiation-inl- les culdl ediate these effects '1i1(l. ill addtitioni.(Iluetl slowing' ofi tastric eiiptviiiig miav aippt'.tr ijidt.- clarly~ indicates aii altt'ratiioo oft brini tut titoi.

*penoi'nthv front voitfing. TI LIS. theV central ncervoius system aii)('ars to piv atnih,'nmechanism of radiatioin-inuo i'd emiesis and pivotal rule, inl flu( digestive symptoms thait iliiiiiidi- 7

- g1astric iniibiion is probaly inultitactorial. Thie atcfv follow total liodY irrauhiadtiii.* central nervous systemi appevars to playv a pivotal rileSinl these sx.Ill t Iulls. as suggevsted bv thei rise of

pdlam ij3-I'Nl oibservedl afti-r irrmoliatiiin. [his rise References-is similar to that ohserxvi after explisurt' to ph *ysical 1.tiddk~tiit (3. 'Iitw-i R\\ llmv'iv Ii mniiikt t(Illixviiin

* ~stress (25.2h)i nd could be ri'vsoiisibhlotfu the of)- I-\plisIf~ h- imll/ilu_ t1,iiii Ax ,1 Sl I 110irmn \l* servedl vonitimig and gastric inibit ion (27.28). !1is P _,,746i:170

- ~~~probaiv thuI to theV tact that irraldiatioN diutikvtSes l' tii~ > il.tiiktiiiii iiiiiiiiI.dx'cl

* peripheral til tif affireict nevrves or, alteratively. -,67 HHcauises flu( reilease of Immorail or tuxiu. suilstaout's. A I ( :iith, HtE Aiili-iiit' (titi~it fi~~tii-iiiii

* tlirec~t effect (it irradi~itioiii on flu', braii ippemrs I'Ji tt imixml I 's5.-V- S iii III i'ni'pa~ii i'\ til 'i.* ~~exclimtetl bv thei obst'rvaiiii thatt slivietin 01t u '12S-t~2t

- -. heiiirt'tceptiir trigge doll e its oot prevetnit radii- 2Idlii I "tII \ .[ d1,1r.1A It Ph%~i; mlt6. 110 12

tuotal hiidv irrailatit i. wliereas it is eftictJivi' Ii flue IIII- ITT 1118'tiuiitiiiix '4.1 444 8t* ilig o fu' fxu tajr reevaif i iiiin .VIt\. x.iii &qi S, 1:1f-. C13,11111i I[. %\ 111 t lad g15). could be' related toflv wo Omo'reev n t Ii i ii'v 'Iiiiiiii 1)'- Ill' Si I80.~i2) 1 7118

*differencevs that ate kiio%\u flit exist hetweemi these 7 1 It, I I_ tI l I Il I.. tI itrIkI A. Siitni K (].iiikl I I.two dlitiiiiil till els: the EDl),, fur radiatiunl-iium ed ihitiii A 01 (hIII~i l.I.tl(iin l x'i xiiII11 xttii ii I 'dix ii

* ~vuiiiitioit is fowi-(r inl fte ulu than ill ft'e nonkev (3ll:xiiifi'ii i ii/ln ~iiiiii i 1,1i1t.0-. i~il1 ,i,anld. inl idditioll. the dtlipni ii( re( et'tr itg,iiist j ij '52 W

* ~ap lolttul- ii-ii AliSrt' Voirri itilimg fl ltx'siiifing i tlitt n tH Iiilkii" V1 x)i ot,-li xi il MC tiii ii-i i t Itii I-iif-tx Ill

* treiilitiii Ito theif i~l'::'.iis) midlit tltiitg wuitii w:2 itsm il~ II- .w Ant I,.*

brod -bifi ii m barsrii, tuititit in ltst fi itig .u iwty 10 'l 'litm x V n \ I 'il ~ iit~rn iI i i (it hl

t'r. ~ ~ ~ ~ ~ ~ ~ n -.rt i"ml illsstti wit fltCl'iiisfittm lh iup- 'm

db* iorieo [wl' I h , ir it soloi. rio it m~i u Clp o- I - b. M- 'll- .It KI, 'IIi, wlt~t .pA10 I* ' ~ ~ phi % iggt 'st dg'a tIt t o iti rit \N itt fit , mnu oli vv Ti e flu'.1h.1M I ' l II Ill tl1t ~ IN-. -- Ii ll I -

* .t~~WI IANI'[(;IIN D1) VAI. E*I AL, (;A SIROE TIRol.f((; \'ill 8'). No 2 .

v\(.[inwMA tI, h-ii lx~crii ill it, (ablr) GastrolentirlogN 27, Iinliin A. Nateini (1Hliituiitimoutit iatrit tiu lii up1)8.1.4 120)8 prvsion tit numiikes ifter rvpeailed Irve-iiperant ai mii

Il Ginr R A. F th- I ofn m im riliatioi o igastrit, ein1,lviig ssins, Iihxcjiol P'eviulo 19 78.6 -,24-8,Ilifle ill Ili,- dogI Appl IPlix cii] I1:9:214--6 28. Stiea-IDinolitie FF. Ailani, N. Artuddli 1. IDnliiiiA F-x i-:t",1

21 lItlci FV (ilrii inipt% ing tit ratls after ;irt-liiit irrilialitimi \Iet-i nkiphalin and nalowneii oin gictri iiii\u i i and v ri.]ttIRadia it, Itil qiwi: 1(:521-32. tion tin rliis o iii kixs, Am I Iih\ciol P1q83:245( Ciff, 2i00

2- IleLinier Hi-: :at[\ Oelis tel ot -jidjalion oin fill uhistrcrm - 29f. Itrizzee KR Neati K.M. tWjlliiuii 11M, IThe lii'iiirci - 1 miI ire it gacltrli hniuc glaiil. Riliat Res 14615:26 ':244-0-1 trigger zonie for enieci, ti (ie( iiuuikv Ami lii c,d P455.

22 H,-iut C. N igier 'I . Ia-wcgiif A Stir It., 4tils re'dililit 180I:653q-12

li, irriliatiinc iiiliiscv dti Iliitiiiii I cor tic Ivijiln to 30). Sharp 13. Kaiiiin It. MaI,il 1). i it D.iiiupiridiin vi~liituitu iigstif ueti"rilliinvc lpar Ic vs i itx idt riiuitgiii A\rt 11 rat (ilasiiia 13-iiiulirphli- ifinlilinuuri I Ix ohio iw iliiii-E:1. I \Ied Iit12:2I1:3t2 1 -34 lveti heri iliraulx 1%utl nit mirai( vibr vi uriu iulit'.I Si I

ON l I ot \-1,1% s iiii u-ailaiir j ix illt, III[ ll Hw tje I~ o ru c II Shliri IB. Riic R. Lei K-. SoiixIr Dlpiiiniii wimiiliticinl gicti -ii otliun Raiitudigx Il2.l;I.:Vr- 4b. ine piaimi %-ini iiiiuI it% 1,- I, Ii liilm~ Ij~il

24 'aiiur \\IL. 'IiniplutiuiF I I(li It riulatiiin lheiix mi I IM2:1 10:1828-31)putrii wii rulo. I.-\1.- IlIll,1 12. 1429(- 14, 12 (Xurxu U.Kiiuijiuir I-S. Siver CI. -1t1 at .. \Iiieu 1/la

5, (;ujllemin R. \argu 1'. Roicr 1. t di 15-1-iiliurpulum iu auut ilurc tuur xiuauliv intctiniu uwidut IVIP it( nt ci ii~itruii (iriil 4mprillii au , omt ( it( miuli1111l Ilxli Ow ilu- dtitfriinwu h\ ruaiuii ii mm t jun mlatuli I G,~iuu I'll-it,irx uiliiiu. S, ii IN 7ii7ti:ttli7-, igv\ Ni82:82tit64

.it Hiuciur 1. -riiu It I- Rix hr C. Ling X. (;iiuliuuui 1<. fit, wuil IIuuuuctluu k iuuti -it stru- tit rvuiues 0-c.idlurpuin titiblo

but ot b~unN~tttr, 1117.20 ('8--

AREO FOfCES R^OIMeeoL(~yARMD8 §TIiCREPORT -,IR~EAIfCH IMSTITtJTIE v

SR85-29

~RADIATION RESEARCH 103, 158-162 (1985)

.Radiation Effects on Diamine Oxidase Activities '.,- in Intestine and Plasma of the Rat'

MELVIN J. ELY,2 JAMES M. SPEICHER, GEORGE N CATRAVAS,---.,AND STEPHEN L. SNYDER -- I.

i Buo henistry Departnent. Armed l~orces Radiohiology Research Institute,Be'thes'da. Mart/and 20814-5145

Ely. M. J.. SPEIctR, I. M., CATRAVAS, G. N.. AND SNYDER, S. L. Radiation Effects--- -'on Diamine Oxidase Activities in Intestine and Plasma of the Rat. Radial• Ret. 103. .-- '158-162 (1985)...-..

• ~~Diamine oxidase (DAO, EC 1.4.3.6) activity was measured in plasma and in ileal tissue- :-"homogenates prepared from male Sprague-Dawley rats euthanized at 1-15 days after acute

. ~whole-body irradiation with 14.5-MeV electrons. Animals irradiated with I Gy showed no""--diminutioni in plasma and ileal DAD activities through Day 13 relative to nonirradiated .- '-controls. Animals irradiated with 5, 10, and 12 Gy displayed marked declines in ileal DAO ' -

~~~~activity, w'ith levels reaching a nadir on Day 3. This was paralleled by a decrease in plasma .].•-DAO activity in all three dose groups. Recovery of ileal and plasma DAO levels was later seen .' '

aealasay4 in animals irradiated with 5- and JO-Gy doses, but animals receiving 12 Gydid not survive beyond Day 3. The relationship between radiation dose and levels of plasma

-.- ~and ileal DAO on Day 3• the time of maximum decrease at all doses, was also investigated."-...- ~~~Ileal DAD activity decreased almost linearl between 2 and 8 Gy. Plasma DAD activity closely .-.-•"paralleled the dose dependency of the ileal levels. These data suggest that plasma DAO activity -..- "• ~might be useful as a biologic marker of intestinal epithelial injury and recovery after acute"-'"

• ~~radiation exposure. ",: 105 Scade.... Press. Inc ir "

INTRODUCTION,

•"The levels of biological constituents found in the serum and urine of animals can • -..,. be markedly altered following whole-body exposure to ionizing radiation (1-3)• In..

some cases these alterations directly reflect radiation-induced biochemical damage• at the molecular or subcellular level (4. 5), while changes observed in certain serum .

• proteins most likely represent a nonspecific response to stress (6). Even thoughchanges in metabolites and proteins in physiologic fluids are potential indicators of -.

• ~~the extent of injury and of the total dose received, estimates of radiation trauma •(Ig ~~rely principally on clinical manifestation and hematologic changes (7). Furthermore.,-. .""'. none of the biological indicators of radiation injury studied to date directly reflect.- , ~the degree of mucosal injury and recovery following acute exposure. Also, they do.. [:

• ' ~Research was conducted according to the principles enunciated in the (iuide for the ('are and Usew of" .,.lahoratory .nImah. prepared b\ the Institute of Laboratory, Animal Resources. National Research. ."""

!,'.- Council.2 Present Address: Medical Division. tISS FRANK CA\BLE. (AS-40), FPO Miami. FL 34086.-...

- 0033-7587/85 $3.00 158.•-'.•.( .p"snght Iql5 hbr %cadcnm, Press~ Int ""

11 ight s (f reproduction n an form resers d -

S 92

MELVINJ. ~~~~ ~~- ELY ,, JA E M.SECEGOG N ARV

INTESTINAL AND PLASMA DAO 159

not permit differentiation between concomitantly developing bone marrow hypoplasiaand gastrointestinal damage. The ability to monitor the degree of damage to theintestinal epithelium (as distinct from other injury) is particularly important inassessing the status of an exposed individual, when one considers that (a) theintestinal epithelium is approximately as radiosensitive as bone marrow and (b) aclose association exists between the functional efficiency of the gastrointestinal andhematopoietic systems and the development of the full acute radiation syndrome(8. 9).

Several recent investigations (10-12) have shown that levels of intestinal mucosaldiamine oxidase [DAO; EC 1.4.3.6] decrease with progressive mucosal injury andincrease with subsequent recovery, and that DAO activity in the plasma follows aparallel course. Thus changes in plasma DAO activity may be a relatively specificmarker of intestinal injury and regeneration in the rat and in the human (10, 11).The present investigation was designed to assess whether the use of plasma DAOactivity might be useful as a marker of the integrity of the intestinal epitheliumafter radiation exposure. Such a marker would make possible, through bloodsampling, the clinical evaluation of radiation-induced gastrointestinal damage andthe efficacy of measures designed to minimize that damage.

METHODS

Male Sprague-Dawley rats (Taconic Farms). 150-250g. were irradiated with 14.5-MeV electrons , ;, ,

delivered from the AFRRI linear accelerator. Pulse duration was 4 lisec, and pulses were delivered at arate of 15/sec. Dose rate was approximately 7 cGy per pulse (60 Gy/min, I Gy = 100 rad). Dosimetrywas accomplished by using a 0.05 cm' tissue-equivalent ion chamber whose calibration is traceable to theNational Bureau of Standards. Animals were maintained throughout the experiment on Wayne Lab-Bloxrat chow and acidified water ad libitum.

Animals were euthanized by cervical dislocation, and blood was withdrawn by draining through the -vena cava into Vacutainer heparinized tubes. Plasma was obtained by centrifugation of the blood at

-9, 3000g for 20 min. The distal third of the small intestine, designated ileum, was removed. It was rinsedbriefly and homogenized in 9 vol of ice-cold Sorensen's phosphate buffer, pH 7.4. The tissue homogenateswere then centrifuged at 15,000g for 20 min to remove particulate matter They were then stored, alongwith plasma, at -800 C until use. Protein was determined by the method of Lowry et al. (13).

The assay of DAO activity was based on the method of Okuyama and Kobayashi (14) as modified byBeaven and Shaff(15). Briefly, tissue homogenates and plasma samples were incubated with [HIputrescine(New England Nuclear Corp., Boston, MA; 27 pmole/liter uiCi) in Sorensen's phosphate buffer (pH 7.4)for I and 3 hr, respectively, at 37°C. The concentration of putrescine was I mM for tissue homogenateassays and 0.01 mM for plasma assays; the final assay volume was 0.2 ml. At the end of the incubation,0.2 ml of an aminoguanidine solution (4 X 10-5 M) in Sorensen's buffer was added to stop the reaction.The labeled product was extracted from the incubation mixture directly into 2.0 ml of a PPO-POPOP'and toluene scintillation mixture, and a I.0-ml portion was assayed for radioactivity.

RESULTS AND DISCUSSION

Plasma DAO activity appears to come primarily from the small intestine in manymammalian species, including the rat and man (10. 16). In the rat, the mucosa isthe site of highest DAO activity in the small intestine. This mucosal activity isassociated with the mature villous tip absorptive cells rather than the proliferating ..

crypt cells (17). Radiation damage to the small intestine is characterized by injury

PPO = 2,5-diphenyloxazole; POPOP = 1.4-bis-2-(5-phenyloxazolyl)bnzene.

'a*. . . . ... . . .. . ..

9,--7

7' 77

160 ELY ET AL.

to the crypt stem-cell populations. resulting in a decreased population of maturevillous cells. As a result of this epithelial-cell attrition without renewal, one expectsto observe a decline in DAO activity of both the intestinal mucosa and the plasma,thus paralleling the decrease in population of mature villous cells. Our resultsconfirm this expectation. ''

Figure 1 shows the effect of acute radiation exposure on ileal mucosal and plasmaDAO activities in the rat. No diminution in these DAO activities was observedwithin 13 days after irradiation with a dose of I Gy. Following a dose of 5 Gy,however, mucosal DAO levels fell markedly to less than 40% of basal levels by Day3. with subsequent recovery by Day 5. This time course is similar to that observedfbr morphologic and physiological changes that indicate mucosal injury and recovery '2(18. 19). The loss and subsequent recovery of plasma DAO activity following a 5-Gy dose paralleled that of mucosal DAO activity. Although plasma DAO levels alsoreached a nadir on Day 3, complete recovery was not evident through the remainderof the 15-day observation period, with the maximum DAO activity at 80% ofcontrol levels on Day 9. A similar pattern was seen in animals irradiated with 10Gy. Both ileal mucosal activities and plasma DAO activities reached a minimumon Day 3 and then gave signs of recovery through Day 6. Death of irradiatedanimals prevented observation beyond this time. Similarly, animals irradiated with12 Gy failed to survive beyond Day 3. although the precipitous decline in levels ofboth mucosal and plasma DAO activity was again observed during these first 3 dayspostirradiation.

Figure 2 shows the relationship between radiation dose and levels of plasma andmucosal DAO activity on Day 3, the time of maximum decrease at all doses tested.Relative to nonirradiated control animals, mucosal DAO activities decreased almostlinearly between 2 and 8 Gy (presumably reflecting increasing mucosal injury), after

.LCa) IGy b) 5 Gy

PLASMA F PLASMA

O0 d)12Gy

PLASMA FPt ASMA

0

ILEUM ILEUM

100

0A00 5 10 is 0 5 10 1

DAYS POST EXPOSURE

Fici. 1. Effect of acute radiation exposure on ileal and plasma DAO activities in the rat. Each pointrepresents the mean of data from five animals: bars indicate ±I SEM.

INTESTINAL AND PLASMA DAO 161

" " "8

%

40

S20

00 2 4 6 8 10 12 0 2 4 6 8 10 12

DOSE

Fici. 2. Effect of increasing radiation doses on plasma (2A) and ileal (2B) DAO activities in the rat 3days after exposure. Each point represents the mean of data from five animals: bars indicate ± I SEM.

which no further decrease was observed. Except for the absence of a shoulderbetween 0 and 2 Gy, plasma DAO levels closely paralleled the dose dependency of

the mucosal enzyme. The dose-dependent changes in DAO activity shown in Fig. 2cannot be accounted for on the basis of loss of protein, since decreases in proteinwere observed only at the higher doses, and these appeared to be minimal. Forexample, the protein found in the gut homogenates decreased from a control value -1of 6.82 ± 0. 10 mg/mi to 5.80 ± 0.20 mg/mI or 15% in samples obtained from ratsexposed to 10 Gy. Likewise, protein found in plasma samples decreased by only 8%following exposure to 10 Gy.

These data suggest that DAO activity may serve s a useful plasma marker ofmucosal integrity in the irradiated rat. Although the significance of these data forirradiated humans was not investigated, recent studies have shown the potential ofplasma DAO activity as a useful marker in the human for studying intestinal --

mucosal toxicity of chemotherapy (11) and a number of pathological conditionsinvolving the small intestine (12, 20). It seems likely that plasma DAO levels wouldreflect the integrity of intestinal mucosa in irradiated humans.

RECEIVED: January 15. 1985; REVISED: March 29, 1985

REFERENCES

1/ H. M. PATT and A. M. BRUtES. The pathological physiology of radiation injury in the mammal. II.Specific aspects of the physiology of radiation injury. In Radiation Biology (A. Hollaender. Ed.),Vol. I, pp. 959-1028. McGraw-Hill, New York/London, 1954. [See especially pp. 964-976.1

2. S. L. SNYDER, Radiation-induced alterations in serum and splenic lysosomal hydrolases of rat.Radiat. Res. 69. 306-316 (1977).

3, M. DONLON, L. STEEL, E. A. HIFIrGUSON. A. SitIpp, and G. N. CAIR,W.\S, Radiation-inducedalteration in prostaglandin excretion in the rat. l.ile Sci. 32, 2631-2639 (19831.

4. J. PARIZEK, M. ARIENT. Z. DIENSTBIER, and J. SKoI. Deoxvcytidine in urine as an indicator ofchanges after irradiation. Nature 182, 721-722 (1958).

5. Z. DIENSTBIER and L. Bt rii, Biochemical indicators of irradiation. In 13th Internatlional (ontgr'"of Radiation, Madrid. Spain. Vol. I. pp. 604-607. 1974.

. - %- . -.i

J.~

162 ELY ET AL.

6. A. Koj. Acute-phase reactants and l sosomal enzymes in the blood of rats with experimentalinflammation or radiation injury. Biologica 18, 275-286 (1970). ' ~

7. V. P. BOND. T. M. FLIEDNER, and E. P. CRONKITE, Evaluation and management of the heavilyirradiated individual. J. Nuci. Med. 1, 22 1-238 (1960).

8. Life Sciences Research Office Staff Report: A Study' of the Aletabolic .4spects (if Therapy of RadiationInjur * in the Soldier. Federation of American Societies for Experimental Biology, Bethesda. 1969.

9. V. ARENA, Ionizing Radiation and Lil. pp. 414-43). Mosbv. St. Louis, 197).10. 0. D. LIJK, T. M. BAYLESS. and S. B. BANTUN. Diamine oxidase (H-istaminase): A circulating marker

for rat intestinal mucosal maturation and integrity. J ('n. invevt. 66, 66-70 (1980).IL G. D. LtJK, W. P. VAUGHAN, P. J. BURKE, and S. B. BAYtIN, Diamine oxidase as a plasma marker

of rat intestinal mucosal injury and regeneration after administration of l-WD-1-arabinofuranosyl-cvtosine. Cancer Re. 41, 2334-2337 (1981).

12. R. HESTERBERG, J. KLUSCHE, C.-D. STAMAzNF~lif, and K.-D. FEIJSSNER, The start of a programmefor measuring diamine oxidase activity in biopsy specimens of human rectal mucosa. AgentsActions 11, 33-37 (1981).

13. 0. H. LOWRY, N. J. ROSEBROUGH. A. L. FARR, and R. J. RANDAt.t.. Protein measurement withFolin phenol reagent. J_ Biol. Chem. 193, 265-275 (1951).

14. T. OKUY AMA and Y. KOBAN ASHI, Determination of diamine oxidase activity by liquid scintillationcounting. Arch. Biochetn. Bioph vs. 95, 242-250 (196 1).

15. M. A. BEAVEN and R. E. SHAFF. Study of the relationship of histaminase and diamine oxidaseactivities in various rat tissues and plasma by sensitive isotopic assay procedures. Biochem.Pharmacol. 24, 979-984 (1975).

1J. F. BUFFONI. Histamine and related amine oxidases. Pharinacol. Rev. 18, 1163-1199 (1966).17. K. M. M. SHAKIR. S. MARGOLIS. and S. B. BAY tIN. Localization of histaminase (diamine oxidase)

in rat small intestinal mucosa: Site of release by heparin. liekchein. Pharmacol. 26, 2343-2347(1977).

18. H. QUASTLER, The nature of intestinal radiation death. Radiat. Res. 4, 303-320 (1956).19. R. A. CONARD. Some effects of ionizing radiation on the physiology of the gastrointestinal tract: A

review. Radiat. Res. 5, 167-188 (1956).20. J. KusCHtE. T. BIEGANSKI. R. HF~siRB iRU. C.-D. STAtIt KNECHT, K.-D. FtEussNER. 1. STAHLENBERcI.

and W. LORENZ. The influence of carcinoma growth on diamine oxidase activity in humangastrointestinal tract. Agents Alctionsv 10, 110-113 (1980).

ARMED FORCES RAOSOI11OLOGYREIAPICH INSTITUTE

111ENYSFIC REPORT v

614 Lspernima 41 (191151. Itirkhauisir Verlag. CH 4010 BalSizrland 0

Wound-induced alterations in survival of "Co irradiated mice: importance of wound timing"*

6. 1). Ledncy. F. D. Exumn and W. F. Jackson Ill

* lnunumunologr Division. h.vperihtnnfl Ilcontologi t 1epar'ineirt Armicd f-rirt us Radiohiologv~ Ri-searc itsiai' Bethersda (Atarti/and20814, USA, .24 April 19N4

.Sunltori. Wounding mice shortly before or shortly after lethal "'Co irradiation enhanes survival. Survival of wounded-irradiated* mice may he due to enhanced hematopoictic recovery as measured by endogenous spleen colony (E-CFU-s) formation.

Key wtorils. Trauma. radiation: combined injury: endogenous spleen colonies.

In previtous publications' ' we established that skin-wound mined 1) the survival of mice wounded prior to or after lethaltrauma 24 ht prior to graded doses of""Co radiation resulted in irradiation and 2) the number of E-CFLU-s in mice wounded l1

-survival of mice that ordinarily would die from radiation-in- prior to or after lethal irradiation.duced hemnatopoictic failure. We also examined the repopulation Materialys and ,nethods. Female and male (C57BL 6 X CBA)FIoft hemnatopoietic: centers with early proliferative cells [colony Cum BR mice were obtained from Cumberland View Farms,lforming unit --pleen.(CFU -s)I and committed progenitor cells Clinton. Tennessee. All mice were acclimated to laboratory con-(granufocy re-monocyte colony forming cell (GM-CFC) and mo- ditions as previously described".nocyte-macrophage colony forming cell (M-CFC)] of suble- Between the hours of 10.00 and 14.00. groups of mice were givenMallk irradiated-wounded mice. 4% skin surface wounds under light methoxyflurane anesthesia.While radiation has been employed in mice' ' and ratsr in combi- The technique for wounding was previously described"'.nation with surgical or wound trauma, the timing of wounding Whole-body irradiations of 40 rad~min (midline tissue) from%prior to or after whole body lethal radiation (loses has not been bilaterally-positioned "iCo elements were performed on micecomprehenisively studied in both sexes of a single species. Our placed in plexiglass restrainers. An ionization chamber cali-

*previous %ork' in combined injury (radiation wound traumna) brated against a National Bureau of Standards ioni~ation cham-and that of others dealing with the recovery from radliation her was used for dlose determinations. Radiation esposures wereinjury ' suggests the use of the endogenous-colony forming unit performed between 11.00 and 14.00) ht. Irradiations of trauma-spleen (F-C-I 1-%) assay ats at potential indicator o1 survival and ti~ed animals were appropriately timed in relation to skinrecovery from radiation damage in mice. Additionally, wound w ounding (see figs I and 21.trauma alone perturbs the clonogenic cell compartments of thehemnatopoictic sy stem. Since hematopoictic cells are inoved inrestoration after lethal radiation doses, we posited that woundtiming relatie to radiation exposure would effect survial friomthe combined injur . Therefore. in the present study we deter- 20-

10 belo's 18

100 Mouse B6CBF I? adiatonon n 16 pe' me pont 16

80.lO~

14 90G

no-lt ?Gy~lI~

datoradiation400

'301 -l0I11 Gy

8 W~ound f.

+ 7 +6 +5 +4 +3 +2 + 1 0 1 2 e -------- Wound t0.0Gy 10 before 10 afterDay of wouridnq btoe})ate -----rditin Wound 11.0 Gy

betie.)atel-l5 0radto 1 . - I -.- 1ii~

Figutre I. the 30l-daN survival percentagesofwoiinded irramdiated Hkl HF 1 7 +6 + 5 +4 +3 +2 +1 0 -1 -2femate mice At each tane point indlicated wounding wai prfrormed (in 16 D, yoftwounding betofel.)or after (-) 6Co radiation

% mie with the exception that 47 mice %ere injured I das aifter irradiationAll contriit-irradiated mice died (data nut shown) wtt the exception thait Figure 2. The mean survival time, (daisl of control irradiated indone mouse lived after) 9 t (;v All cintril-wiUnded miice survised 906v(i wounded- irradiatred mice Jhes' data reflect the sursisat of1 mice onlyIF- . 10 It (s 0- 0. II 1 0(, A -A within the 10-day observation pr iod See figure I legend for symbols *

I sp, Iiiia 4i 1 1910 . Birk IIis1CIN Mcrae. C I 1 41111 Ki,,cl St, II /o Ifind (,I' i

* In I A( I V - st udies, splecris scrc rernowd from nie W dat% fl~ aI signifticanil% p i 0] itierca sed orl liii Lir% I dJis hetorc.iiicr irraidiation. placcd in HBiutns solution for 4 h and tire irradiatiion Injur% Ia d.1 11ci Ilo (is wiitfc t -p (if),colonies couted (double-blind detertmnations) ['resented Ili rcduiced the stir is jt time lioni thait ohscrscd lt inadijied coil-fiigure 3 arc the mean %itucs SI-M) of iso independent (Fils 14 2~i9v 60 "i d.i,s Inturs either hci-ore or aftercoiunts 10r cach group otl spleens I I I (I\ haid no, signtificati IlisiiC .sr ncgaiise Influcnecc ont

MAdI% surs i t ::n I. ( -F-'-s cx pernti s sscrc si iitt tlcuICI sursi\,il pe rceiadges I si% I.11iI uie

performecd at 9,01.0l. and I 1 11 (h on miouse groups giscnl Ihe resultso I [-( H ~ [soa f let iaicrmixare lusiraicd tinciihcr I I \outid traumna and radiation. 2rad.itiOn 0111\ or tivur- I III all stir\ is in md F (II-scxpcrlInienis. Fl1cc 55crc

ssmi kund iraUnta onk Apprixinaic '00l mnice %kerC Used I' ll11 irradiated siulticiouiml\ I A( *F t "sssrc sicictd after ss ound- . -

hree cxpcriincnis Stirs is at pcrccn(iagcs sere based ont groupso Il irrdittiotim iliu cics ts, and radial ion dlocs ihit1 %%erc Issocl- .

If, niice ech. sshilc I-CI I. -s studies sscre done ssiih eight mice .ited wsill taltistilCtl s1IiL.1tcit IFinCease Ini suTi itPerenit-

per groulp. We,' or sun i sat lie tim,!c %inii t cxcpi on io ithis is as i hcii1 le surs iat daia \%ere anials /ed as ltons. [tic K ruskal-A allis nmice "sere ssouitdcd 11) nin bcfliie lit (Gs FIC I Is sscrc not

test is as uscd %%hen all mnice (lied is iii it)3 da s rhc C ox, .cicCicd esI cii t hogh siirI ~ai asF ibscrerd Ili imost in usi esI -test!' is as used " lhen sonic iie \A ii hiii an ex perirncnial I! ('roup ishcrc all t he conibi ncd-inui~ared aimalIs d icd ot u hec su i ialsurs iked I obsers atiolis cciisoredI. ice.. itnriinaied at 3tt da\ s times \&ere swigiiiait dcrcauscd contip.mred to irr-adiated con -

* I hc chi-sqiaarc iest suais used s hen comparing LTr is nil1 propor- trots, none ir scr% ten% IA-( I.-,L* coloniics \sei c touid I losses ci -

lions \iihiii radiation doses. 11) The caseC otl mice %%oiiided I tit\ pi ior ito I 1i 0, G\ % here no-R, ult,. 1-igUrC I 1ILusiratcs tic 3il-dt sirsi\.il percciigcso0 enhaneent il ,ur\ is at i as ltiid. [-(F t - "scrc pl csent

fcrnatc mie %% ounded beforc or aftier irradiation Tire nucan Di, iwi~iwm Ihe foregoing daia su~pport the contention ihait [ti,-stirs i atie iiics ofince d\ ing us itiuti cach ofl ihc i urcicui groups icgilat iiied \%iutd iraurnia rcsiilis inl I in11 unLcas LIT tie

*aire depicted iii fiurc 2. tiuLmber oA tmice stirs i% ing lethal irradiationi and 21I Ii increase Ii -

*\i 11 is. ssimi OUI nmr. performed 2' ILI\ s, t da~. 111 inin before c CIIOmmnouIS st cotonues 0 M I Ill irradiaited mice uhi mas a, -or Ili wini aftcr urradu.itioti rcsutted Illsuutla SWI I1 (1 P 1. 1iI OFi - 011111tfor th iecsdsr sa hree itiipirtittt a1ctors na.1 betimnenii tif siurs is i Itiis tasirtie Ior bott surs siia tics considered in iterpretitig ouLr data I hie\ ire t I sets titterenecs, Inl11 p 1 it 1 ti , asic,liuied b% tite ( ox I (t u tuifr sLitsi\s at pci-- response ito rudtatiiit. 211 soiund cotiaunatmon us itti bacti a

xitoYges pl - ti1 orI p -i lt nicasireut b\ the chi-squamre test and 31 tmechianisrns of- action teadinig ti sur\sI is t hr tnurtutit\-~ iipirtuttts. s iitiIri d.1 Alter Irradiationi O. olt47 dted I didIi Frs. the responises depicted miapilicati, I (figs t 31 %\ere uibtaittcd

nti sI1!nIIIITi(I\ ciontribute I,, ihc tmrttttt\ percentage "hieii usith cniate miuce. Ili other experufltititsdati lit shonii tuate

0111iroand (I that If the it iduated uitiriils t It- of 16 died). Illn mice if the same I gentictic l ieage \s ere used \%outI id Iit g het ire 1I IrIIFC 01e 'iDt tilt) ( is - 1ttiwl III sutaied eit her t da% iir 111 tiitiii prior it Iter 9lt0. 1ii 0, or1 t1 0 6% ( 1 1, resulted Ii siri tir surusi~a pa-

- t rauhuit i sId iccd a signtific.iit l Is I I td p - IF lI ); ratieters A lie maiti sfaititaiusIC tfititelrice tnoted \sas thtat bothr rccIi kch 1 I usr c.use in siursui pset1 enta 1 1ge Suuu IkIal tihe usm\kcc siriial percets and sn altimes "sere geuueratt grcter Itt

ittle nice xersusl 1i1.1t 101111du Ill tetnateC TIIte 111is mxt he dueI iiithe rid ioiresistm a I te i rac meiice (11) LD , S ( i\ I comtpa red tohit i tire tertiute truce 11 7), 1) (\ I Ilin oulahiirator\ ltcd-

Fie%. - uuupitIted mibserstmIIIons1. Itt - I I -s sitidies s ilb imie----- i tic.siiid tiunma 24 It btuirc and HI mitt alter 411 95 SG..... .vsle itmitutii~ tre coiiets thiti that found il irra-. -

* ituitmed ciintrit mice Thiis. us hite iit mice .irc inure radio-VCe1usili. OU iiir 11C.ta utisumi dIiirmna-eiiharcel surs isat oriiiituloi is simili Ili hiith scxe

Sesoisits 1. it sho1itd h, noted ilii Al usomiids can he poitctitiattIscuikIoii~iie11d \%ith hiu,Fi I iuseci . us itist sesstdiCteIdh% pmll fi"roioiuint etemuri a "~I iupi tote1 susefitiI \%,Is iii1 oh-sci secd Simc in idiut iii, depletes the itntiiiucroiat dceeisesui-

-. F I i u t 11i pus tIcl 111.1 %oiuuiditiri aite exposure toi tidiatuiti

'I ~I hue tesmci stusa fricius ind iriallites it IttiCeL, siuiif2fusIfterl 1ris11i1o0i Jrc Ill meill %kill ifis ldc~i

I tisseser ii It 11 if1u1 t i 10s ili ITh AtM hettefut~l11 'i dcitiiiiiCeital

- ~ ~ ~ HA 10 -elei i uut iusic m.us1 p.ictllts .. ru"iti'liliuucu sutli has,

I2 'I Ill 1,1 To 7: 1d ;2hlionPr l, % 5cuninc IrT9itf I..-

IAes sc~lllul Iiu ecisnenil eiihuik 11.iih11..lo i..is is11-

ITT[,, si Itile .ssmttIcil salieIfIII ss 1lluil"' l sp uses suIt1

v ~~ * I asil. the t c it l of uti o 1 tiiih LisissesX ii siursi

i IS I i Ci els ITT IcI mutsl"lII is 1

1 Iii Iii~ . p~I is. d ai i

1 'li 11 ' It'll St.-rl r ld t

: \'. 's I ' " 1 1.. 1 .... I~ Itraiitms .i1TIr.ils I tis I%% pfriT\r t II I "1f 1010\ti I "i i

it IiII-,l I5-.~ fl i I P'. Is di Ii it, irrutiiii.i htisKsti fsiiriicis It' ,ffruf . -e..-..- Wifidhi

.,MI'M. ... IL ,,, P1 1CO W 1 b ll I!LI~cI d -C Vl~ ll 11(c clllJu,'

616 Experientia 41 11985). Birkhauser Verlag, CH-4010 Basel Switzerland

mortality'5 Prostaglandins are known modulators of hemato- 9 Ledricy. G. D. Gelston. H M. Jr. Weinberg. S. Rt.. and Exumn. E. D..

pocss and elevated levels are seen after both trauma"5 and Experientia 38 (1982) 1228.

radiation". Thus, increased PG[. levels induced by trauma t os . og ) . el .adSrin.G.JLbdnshortlv before or shortly after radiation may enhance hemato- Med. 82 (1973) 727.

therefore~~~~~~ suvvl icesd nue I Ledne). G. D . Stewart. D. A., Gruber. D. F.. Gelston H M. Jr.poiesis andterfresrvvl whilecrae amounts [ued sum, E D . and Sheehy. P.A.. J. surg. Res. 38 1l985) 55both by radiation and subsequent late trauma may result in 12 Staiisical Methods for Survival Data Analysis. p. 133. Ed. FPT Lee .*

death associated with scrisis"'. We are currently directing our Lifetime Learning Publications. Belmont. CA 1980.research etforts toward delineating the role of these biological 13 Ainsworth, E.J. Larsen. R. M.. Mitchell. FA .and Taylor. J F , imodifiers in traumia-enhanced survival mortality with radiation Radition Protection and Sensitization. Proc. 2nd int. Symp on-exposure. Radiosensitizing and Radioprotective Drugs. Rome 1969. p 381.

Lds H. L. Moroson and M. Quintilani. Taylor and tFrancis. London* -

1 Supported h% the Armed [Fores Radiobiology Research tnsiituie. 1970,1.Deflense Nuclear Agency. under research Work Unit 100129 Views 14 Walker. Rt t and Porsaznik. M.. Life Set. 231(19781 231 Spresented in this paper ire those of the authors: no endorsement by, I 1062y.1. adWloRPrcSc x il.Md /i16he D~efense Nuclear Agency has beengsno should be interred. (6

2Research wa, conducted according to ihe principles enuniated in 16 Walker. R.L. Lednecy. G. D.. and Bertok. L..J I Trauma '? 0 1983ithec j;uide for ihe Care and (Ise of Laboraiory Animals'prepared h% 225.

*the Institute of Laboratory Research. Nitional Research Council. - 17 Kurland, J.. ,ind Moore. M A.. Exp. Iemat. 5 1 977) 3S7I ledncy. GiD). Siewart, D .A . [sum, E D.. and Sheti.. P.A.. Acia 18 Wang. [B.S.. Hecacock, E Hi.. and Mannick. JIA., Clin. Immuts Ini.rasliol. oneol 2)119X1)29 munopath. 24)(1982) 161.

4 tednes 6 1) . Lsum. L. 1), and Sheehy, P. A.. Esperientia 3'l 19S81) 19 Steel. I. K.. and (atrasas. G N.. tnt. J. Radial. Riot. 421 1982) 51191. 21) Shosrt. B L.. Gardiner, M., Walker. ft. I . Jones, S. Rt.. aiid Fletci.r

s tedneN. G, I). [sum. F D).. Stewart. D A . CGclstoi. I. M.. Jr. atid J ft .in: Advances in Shock Research. vol. 6, p. 27. Eds W. Schumier.Weinberg. S R . Exp ttemat. I0 Isuppl. 12) (19N') 261. J,.J Spitzer and B E. Marshall. A. Rt. Liss tnc.. New York 1981 -

6 Vachc~arl. J -' . and Feague. P 0) , Proc Soc esp tltol Mcd 14,.?1917si 109s

I..,neend,,ril. It , Messerschmidt. 1). nd Melching. H.. Sirahlethe-

Stromnberg. . W' Rt . Woodward. K. I Mahin. 1) t anitd 1)onati. 0014-4754 115 050614-031.501 0. 20 0Rt M . Attn Surg. /6- 196X111 N Birkbauser Verlag Basel, 1985

. . . . . . .

ARMIED FORCES RADIOIOLOGYRESEARCH INSTITUTE

Phd &h.ur. Vol. 35. pp 249-253. Copyright Pergamon Press Ltd.. 1985. Prited in Ihe U.S.A. SOOMFIC R.ORT

SR85-31

Quaternary Naltrexone ReversesMorphine-Induced Behaviors

G. ANDREW MICKLEY.* KAREN E. STEVENSt AND JAMES A. GALBRAITHt

*Behavioral Sc'ienc'es. Department, Armned Forces. Radiobiohogy Research Institute. Bethesda. MD 20814-5145and ,Dpartilentl ,ofB,,havioral Sc'ienc'es and Leaders hip. United States. Air Force Academy."%":' ..-

Colorado Springs, CO 80840 "'""

Received 13 April 1984

MICKLEY. G. A.. K. E. STEVENS AND J. A. GALBRAITH. Quate'rnar% naltc,ne rir i sc% morphoin-indh idhwiuoir.%. PHYSIOL BEHAV 3512)249-253. 1985.-This study explored the relative role of the peripheral and central

nervous systems CNSI in the production of morphine-induced behavioral changes. Toward this end we used a quaternar"derivative ofan opiate antagonist (naltrexone methobromide. NM) that presumably does not cross the blood-brain barrier.Nalire\one methobromide (20. 4) and 80 mgpkg. IP was used to challenge the stereotypic locomotion. analgesia and -.

elevated "Straub'* tail response observed in C57BL. 61 mice after a 30-mg kg (IP) iniection of morphine. The quaternaryderivative of naltrexone reversed the locomotor hyperactivity. "Straub" tail and analgesia normally observed in theopiate-treaed C57B1 6J motuse. The data reported here. if taken at face value. suggest an important role for peripheralopiate receptors in morphine-induced behavioral changes. However. these conclu,ions are contingent on further researchto more full evaluate NMs', capacity to cross the blood-brain barrier of the C57BL6J mouse.

Morphine Naltrexone methobromide Quaternary naltrexone Locomotion Analgesia'fail erection 'NS Peripheral nervous sstem Blood-brain barrier

THE discovery of the existence of central opiate receptors present study, we challenged these behavioral responsesand opioid peptides Ifor review, see 1221) has led inves- with naltrexone methobromide (NM). which presumably hastigators ito suggest several specific brain areas as mediators aI primary action of peripheral opiate antagonism 12. 15. 241. 1 ..ol a variety of opiate-induced behavioral alteration%. Forexample, morphine analgesia 117.291. catalepsy 19. II. 26.271. locomotor changes 17. 17. 23. 251. and elevated strauh METHOD

tail reactions 1131 are all apparently produced, at least in suhjictIpart. b) brain neurons. The role of peripheral opiate recep- The subjects were male C57BLo61 mice 15-20g) obtainedtors and endorphins in the production of these and other from Jackson Laboratories (Bar Harbor. ME). Animals werebehaviors has been addressed only recentl . Some of this initially housed in groups not exceeding 9 per cage

work has been accomplished through the use ofin locomotorquternaryderivatives of opiate antagonists. which presumably do notcros the blood-brain barrier. In this regard it has been activity testing were transferred to individual cageshown, for example. that methyl naloxone failed to alter x3x 19x 13 cm) at least 24 hours before activity testing was

morphine-induced antinociception unless administered in the begun. Mice:weremaintained on a 12-hour light-dark cyclecerebral ventricles of the rat 1101. Similarly. quaternary nal- (lights on 60) am.) in a temperature-controlled room

trexone did not block the stress-induced analgesia experi- (23'- . Behavioral observations were made at the same

enced b defteated mice 1151. Furthermore. methyl nal- time each daN in order to avoid circadian variations. Purinatrexone. when given sstemicall, did not antagonize the laboratory chow and tap %ater %ere available ad lib except

discriminative stimulus effect% of opiates: however, it did during behavioral measurements and drug injections.

work periphcrally to displace I:'H I etorphine from opiate re-ceptors and reversed the effects of morphine on the isolated Pon

" l. "

guinea-pig ileum 1241. Naltrc\one has also been shown to be Iw o studies were conducted to assess the effect of a iPC-over 15(X-fold more effecti%e than quaternarv naltreone in ripherallN acting opiate antagonist on opiate-induced loco-reversing morphine-induced cataleps 161. molor hyperactvit, elevated "Straub' tail and analgesia in

In a further analysis. the present investigation sought to the (57B1. 6J mouse. The terms quaternary naltrexone andclarif\ the extent of peripheral opiate involvement in the nalttexone melhobromide I NM) arc used interchangeabl\. Inproduction ofseveral morphine-induced beha%ior,. Prcvious the first experiment. 6(1 mice received an injection (IPI ofwork 13. 16. 231 suggests that morphine 1iP) produces a morphine ,ulflate 130 mg kg). and 60 control animal,, receivedstcrcot.pic ocomotor activat it continu'us movement . I 1 an equal % olume of saline. Fii\ e minutes after these in-largcl, in one direction around the cage perimeter). analgesia jections. locomotor acti ity w as recorded during two suc-and clealed 'Straub tail in the ('5ill. 6.1 mouse. In the ccsive I -minute periods using the Aultomex I) ,, stem of

S. . . , .,

.'-.., .... .... ~~~~~~~~~~~~~~~.... ......, ......,., .. ,... ....... ............ .. ... .:,,.,... ..-..

..- . .. . ..... .-.... i..?.i. *... *,". ,','...*.i.-

Columbus Instruments (Coumbus. OH).Halfof the animalsTEEN ANDy GLte I

of M 10, 0 o 80mg/kg. IP in O.Wl; saline: N = 10/groupt.ilhese doses of quaternary naltrexone were chosen because 2403 M-0P.,in, viti and receptor-binding studies suggest that this periph- 2000*erally acting opiate antagonlist is apparently 20-80 times less Mw-w..potent than its tertiary counterpart 124). A I-mg/kg dose of 1600 N. M

naltrexone reverses mouse stress-induced analgesia 1151.) sMThe remaining subjects were injected with an equal volume 120Stm~of saline. After a 5-minute wait, locomotor activity was re- 800oo.corded for twko additional IS-minute periods. During these Nfinal activit% measurements, observations of tail position 400

Ae also made. T%%entv and 35 minutes after the secondintection, t%&o independent observers categorized the tail0positions as (at) flaccid. (b rigid but positioned horizontally. 4m/gNh~o.Mto~meNc) rigid and elevated 5 -90' off the horizontal plane, or (d) 2400

rigid and elesated greater than 90" off the horizontal plane.Categories 3 and 4 are *'Straub" tail reactions 131. Observers 1 2000

%sere blind as ito the drugged condition of each subject.i~ 1600

Analgesia %%sas measured in at second experiment. SixtyKmice received an injection of 30 mg kg morphine sulfate (WP) r 120r .and 601 control subjects received anequal volume of saline.

Fifteen minutes after these initial injections, half of the - 80animals in each group received an injection of NM (20. 40o, 1 40SO) mg kg. IP: N- 10 group). The remaining mice received 40

control injections of saline equal in volume to those of oL

Wdquaternar naltrexone. Five minutes later. analgesia was W mg/kg Naftizane Mehobrmme (N.M

tested using the hot plate method Ill. Individual subjects 2400

esre placed on at hot plate maintained at SO0 I C. T~v indc- 20*pendent obscrsers noted the time after placement ss hen each 2000-

did not readil occur. latenc% measuremlents wvere ter'Mi_ 10

mouse "ick e (a) d forepa so:nde 1b2idp00 I as ik

loch oot i andanasi dalta ere 4anzd wihi a ihrnlte emthhoie(M 2) ( o )m g)o aieree*e tnhesa NOVA deign. Sepratet andV~ copare th e- Lr~52)mntsatrcalng neto)adlt 213 m

*four treatment groups (Morphine - saline, morphine -NM. te after challenge iti/ectioni activit,\ observations are sho%%n.*saline - saline, and saline N NMI at each antagonist dose Statistical comparisoins indicate results of ofle-wa ANOVAs and*and (in the case of locomotion) time period. Significant Ns a.el sihccmaios (1:.-dteetfo

saline - saline control gru:-different from morphine - saline*differences (p.Of II (were further described through the use :ontrol group. Variance bars, r'epresent the standard errors of theof Newman-Keuls post hoc tests 1281. mneans.

Morphine's 13(1 mg kg) locomotor-stimulating and"Straub" tail effects %here observed within the first 5 min-tes: thev reached their peak within the first 3(1minutes. and

lasted approximateN 2 hours. Subjects usually walked 8t0 mg kg: icion .ni 11 period: 1) 06)= 29.84. p- .00(81. andaround the perimeter of the cage, and rarel% changed direc. Nessman-Keuls. p .0(5, 4(0 mgkg: F(36)=58.4 p<0.(81l.t ion. This ssas a reliable effect: it was seen in each replication and Newvman-Keuls. pt 0.05. 80 mg/kg). Mice receiving 4(1

*and in eac.h time period, beginning ss ith the second 15-minute mg kg of NM often stopped their stcreot~ pie actis ity: the%baseline interval (one-vay ANOVAs. F1361 ranging from groomed and exhibited slowed locomotion, rearing and sniff-15.3 ito 216.24. p l.(l Ness man-Keuls post hoc tests. ing. Subjects receiving 8(0 mg~kg NM showed these same

p 0.05). changes, although. for some periods, of time. the% %%ere in-Baseline locomotor activity %sas similar for all morphine- mobile. This high dose of N M reduced the morphine-induced

-injected subjects. However, after at challenging injection of locomotion to such an extent that. quantitatively, it could*either 4(0 or 8(1 mg kg NM. locomotion decreased signifi- not be distinguished statistically from the spontaneous

cantly as compared to controls (first iiout period: locomotion of saline control subjects (see Fig. 11. The lowestF(36)- 30.85. p- .001,~l and Nessian-Keuls. 1)- 0.05. 40 dose (if NM used in this study did not reverse morphine

*mgkg: Ft36l=4l .6 6 .p-(lWWI). and Nessman-Keukp. (0(05. hvperactivity. In fact, a signifcant increase ssais seen in

- - -; l .- ' " - - " --- 7 1 - 7 1 "

REVERSAL OF MORPHINE-INDUCED BEHAVIORS 251

TABLE I

PERCENT OF MORPHINE-TREATED (30 mfgkg) MICE EXHIBITING VARIOUS TAIL POSITIONS FOLLOWING SALINE OR NALTREXONEMETHOBROMIDE (NM) CHALLENGE*

First Observationt Second Observation '

Elevated Elevated -.

Drug Treatment/ Rigid and Elevated greater Rigid and Elevated greater ., .Doses N Flaccid Horizontal 5' to 90' than 90' Flaccid Horizontal 5' to 90' than 90'

Morphine-saline 10 0 0 30 70 0 0 30 70Morphine-N,'. (20 mg!kg) 10 0 0 90 10 0 10 80 10Morphine -%aline 10 0 0 to 90 0 0 10 90 '="-.

Morphine-NM 140 mg/kg) 10 0 60 30 10 30 50 10 1(.

Morphine-saline 10 0 0 10 90 0 0 0 100Morphine-NM (80 mgkg) 10 60 20 10 10 50 30 20 0

*All subjects that received first injections of saline (i.e., saline-saline or saline-NM) are omitted from this table since all had tail positionsrated as flaccid.

t Two tail observations were made for each mouse: First=20 minutes after challenge and Second=35 minutes after challenge.

opiate-stimulated locomotion in both the time periods re- As expected, morphine-injected subjects were more anal-corded after 20 mg/kg NM treatment (fir.,t time period: gesic than were the saline-injected controls. This analgesiaF(36)- 113.42. p- ).ll. and Newman-Keuls, p- 0.05: wcoond has been shown to peak within 0.5 hour and to last at least 2time period: F(361-216.24: p- 0.001. and Newman-Keuls, hours (201. This response was observed on both the forepawp-- 0.05). measure (20 mg/kg NM replication: F(36)= 11.07. p<0.001; .s'

Quaternary naltrexone alone apparently does not produce 40 mg/kg NM replication: F(36)=19.58, p<0.001. allhypoactivity since post-hoc comparisons of locomotor ac- Newman-Keuls p<0.05) and the hind paw measure (20tivit, counts between the saline-injected subjects and those mg kg NM replication: F(36)=48.33, p<O.001:40 mg/kg NMreceiving naltrexone methobromide revealed no statistically replication: F(36= 39.29. p<0.001: 80 mg/kg NM replica-significant difference between these groups (Newman- lion: F(36t=9.17.p<0.00I. and all Newman-Keuls.p<O.05).Keuls. p- 0.0)5j. This lack of statistical difference is appar- See Fig. 2. This morphine analgesia could be reversed by anently not due to total inactivity of the saline-injected animals, injection of either 20 or 40 mg/kg of NM (Newman-Keuls.Although habituation to the testing environment did take P

< 0.05). However. when 80 mg/kg NM was given, miceplace, the saline-injected animals were still sufficiently often developed tremors, limb stiffness and lethargic move-active in the (.5-hour test period (average counts for all ments. which may have inhibited their paw licking.saline control groups= 228) to discern an even lower level oflocomotor activil had it occurred in the naltrexone-treatedmice. D)ISC'USSI(ON.-.-°.

All morphine-injected mice exhibited elevated "Straub" Naltrexone methobromide effectively challengedtails that were lowered significantly after treatment with 40 morphine-induced locomotor hyperactivity. analgesia. andor 80 mgikg NM (see Table 1). Tail positions were observed "'Straub" tail erection in the C57BL,/6J mouse. Previousand categorized as previously described. However, for the studies have indicated that this quaternary opiate antagonistpurpose of data analysis, observations of tail positions were does not readily penetrate the blood-brain barrier 12. 10. 15.grouped as non-"Straub" tail reactions (either "flaccid" or 241. Taken in this context. the present data suggest that pe-'rigid and horizontal*') or "Straub'" tail reactions (elevated ripheral opiate receptors may mediate, at least in part. thesefrom 5' to --90 off the horizontal planet. Fisher exact morphine-induced behaviors. It has been well documentedprobability tests were computed for each of the two obser- that seemingly central nervous system effects may some-vation times and doses of NM 1211. Morphine-induced times be mediated b. peripheral mechanisms (for review, see"Straub' tail reactions were significantly antagonized after 1141). These data do not, however. deny the already-challenges ofeither40or80mgkgofNM (allp- 0.01). Some demonstrated role of central neuronal mechanisms in theof the most dramatic reactions to the opiate antagonist were production of morphine-induced locomotion 17. 17. 23. 251.observed after 80 mg kg. The majority of these morphine- analgesia 117,291 and tail erection 1131.injected subjects went from exhibiting tail positions that One of the doses of NM (20 mg/kg) used here did notwere almost parallel to the vertebral column to completely reverse morphine-induced locomotion, but instead enhancedflaccid tails within 2(0 minutes. Although 20 mgkg produced the stimulating effect of the opiate. This is unusual but isa slight lowering of the tails (see Table Il This was not a consistent with some other reports suggesting that some " -

statiticall significant change within the present scheme of opiate antagonists (i.e.. naloxone) can display opiate agonistdata analysis. All subjects that received first injections of activits in a number of behavioral and in vitro test systemscontrol saline (i.e.. either saline and then NM or saline and (for review, see 1191),then saline had tail positions rated as "flaccid." In the present stud, . specific opiate-induced behaviors

* .:.-_:.-.**,,,

%. .:-°

-Al

-X6

52 MICKLEY, STEVENS AND GALBRAITH

29 .m/k high-dosed animals exhibited tremors and rigidity which may12 ,n 11 ab*ontiift have interfered with their capacity to lick their paws. Mo t120 IUt..1 of these subjects showed signs of pain sensation (e.g.. jumtp-

oo t ing off the hot plate), but their failure to perform paw lick.,,made them statistically indistinguishable from control

so morphine-injected mice (see Fig. 2). It may be the case thatso morphine-mediated antinociception was attenuated by

quaternary naltrexone but the mice injected with 80 mg/kg4o0 NM were physically incapable of demonstrating this effect

0 on the hot plate test. ".The interpretation that peripheral opiate receptors may

o partially mediate morphine-induced locomotion, analgesiaand -Straub" tail erection should not be made without somecomment regarding the likelihood that quaternary naltrexone

40 g/ko crossed the blood-brain barrier. The majority of the litera-1na2.0os.Uhob,,,,,a ture supports a primary peripheral action of NM. Still,

IN.U.i Brown and colleagues 161 have suggested that quaternary100 onaltrexone may be gradually leaking into the brain of rats -"-A

o ,after approximately 90 minutes. However. this explanation

SPm [,,may not address naltrexone methobromide's apparent ca-

C 60 pacity to rapidly (within 5-35 minutes) reverse themorphine-induced locomotion, analgesia, and tail erection

40 reported in the present study. In addition we found that early

20 * (within 15 minutes) reversal of morphine-induced locomo-tion by the antagonist is similar to that observed later (15-30minutes postinjectiont. This suggests that leakage, if it oc-curred. must be quite fast. Finally, if the behavioral reversalsreported here were caused by leakage of naltrexone metho-

80 -Wkg bromide into the central nervous system. then one might ex-Na =one pect to see a gradual onset of antagonism. which would begin

120 * , IN.U.I J at the receptors located nearest the weak points of the

10o blood-brain barrier 1141. The morphine-induced locomotion,analgesia. and tail erection responses measured in the pres-

so ent study are apparently mediated, at least partially, by quitedifferent brain areas 15.131. However, quaternary naltrexone0 "reversed all of these behaviors " ithin the same time period.

40 These data are in consonance with those of others thatsuggest that the failure of naltrexone methobromide to an-

20 tagonize heroin self-administration was consistent over a

0_1 -3-hour period 1121. Thus. leakage of quaternary naltrexoneW.POWIM Sam" into the CNS over time may not be the most likely explana-

Sam" NMW Sak" N.M tion for the results presented here.Another potential explanation of the present data is that

Ft 2. Mean latenc. to lick forcpaws and hind pa"s followming an rnary naltrexone may metabolize into naltrexone 161 orinitil niectton of morphine ll mg kgl or saline and then a challeng- quatery t n y a

ing dose of either naltremone methohromide 120. 40 or X0 mg kgI or some other opiate antagonist, and that it is this compound .-..-

-1.,lic St.,t11,cail comparion, indcilte resulls of tone wA A N(OV ,s (wAhich crosses the blood-brain barrier more readil% that ... ,

,ind Ncwnian-Keijls post hoc comparisons. p 0.05: ' different antagonizes morphine*s effects. Future studies are plannedfrom sitne • saline control group ; different from morphine - in an attempt to resolve these issues.sdlne control group. Variance hars represent the standard errors of It should be recognized that the effects reported here maythe mcnei,. also he species dependent. Others have suggested that the

blood-brain barrier may have different permeability charac-teristics in different mammals 141. Some evidence suggests.for example, that quaternary naltresone. at comparable

were antagonized by different doses of N M. Morphine's an- doses, fails to produce centrally mediated symptoms of nar-algeSic effects were significantly attenuated h) 21) mg kg of colic withdrawal in dogs or monkey s I 18 but may 1181 orNM mhereas opiate-induced locomotion and "'Sttaub " tail may not 1151 reverse mouse analgesia. The permeability ofwere c,sentially unchanged at this dose. Ihe,,e datt support the mouse blood-brain harrier has yet to be fully described.pre tis studies that highlighted the flact that the motor and The present data may be interpreted as suggesting either

41analgeic eflects obsered in morphine-treated mice are (a) a role for peripheral opiate receptors in the production of .clearls dissociated neurophy siologically and biochemicall. morphine-induced locomotion, analgesia and "'Straub'" tail11hi. Similrly . the same dose of NM has been obsersed to tieaction,. o (bi t CNS locus of action of naltre one metho-

recrsc caltalepsy 161h but not analgesia 1X of- the rat. bromide or its lipophilic metabolites. Future studies that\lthough morphine analgesia was effectielh. challenged measure the regional distribution of quaternary naltresone

b 210 and 4() mg kg of NM this anttagonism was not demon- and its breakdown products throughout the body shouldntitcd ftet X0rnig kg t,f the qUiternalt. compound. Ihese c larifs these issues.

."oI.

REVERSAL OF MORPHINE-INDUCED BEHAVIORS 253

ACK NOWLED[GiEMENTS%

Fhe authors w ish to thank Robert L Stwr.Jsp anle. ducted according to the principles enunciated in the-Guide for theUPaul Brenner. and Diane Pickle for their fine technical assistance Care and Use ot l-aborator% Animals~ prepared b% the Institute ofduring the course of these experiments. We wish to thank Dr. H. 1.aborator Animal Resoiurces. National Research Council. I'his e'.-Mer/ for his contribution of naltrexone methobromide. We thank periment %%aN conducted, in part. Ahile the first atuthor skas at theNlrs. NMarionGolightl for her help in t~pingthis manuscript and Ms. USAF Academ . and %;as supported bk l)efcnse N uclear Agcnc

,;Junith VanI~eusen for her- editorial assistance. Research %kas con- l)NAi uinder research project L W)MXMK. (00141

REFERENCES

I. Ankier. S. I. NekA hot plate tests toquantif antinociceptike and 15 Niczek. K V. M .1,1hompson and C. Shuster. ()pioid-likenarcotic antagonist activkities. Eur J Pharnis'/o 27: 1-4, 1974. analgesia in defeated mice klinm 215: 152(1- 122. 1982.

2. Bianchi, Gi.. R. Fiocchi. A -I is amy and I.. Manara. Quaternar N 16. Oh% erio. A.- and C. Castellano. (ienot vpe-dependent sensitis itnarcotic agonists relative abilitN to present antinociception and and tolerance iti morphine and heroin: Dissociation between5 gastrointestinal trainsit inhibition in morphine-treated rats as an opiate Induced running and analgesia in the mouse. Pmhio,-index of peripherail selectisits Life .Si .30: 1875- 1883. 1982. ;slou nm, oloc 39: 11-22. 1974.

*3. Hilbe% 1). L.. J. H. Sailem and M. H. Grosnan. [he 17,( Oliserio. A. (ienot~pe-dependent electroencephalographic be-* anatomnical basis o1 the straub phenomenon. BR. Pliapmwi ,l 15: has ioral and analgesic correlates of morphine: An anal sis in

541-543. 1960. normal mice and in mice -A ith septal lesions. Brain R,s 83: 135-- 4. Bradhur . M. I/I, ( on,, i iq a Blo od-Rriiii Bric,e Nev 141 . 191 S'

*York: John Wile% and Soins, 1979. pp. I f,- 136. IN. Russell. J.. P. Baiss. I_. 1. Goldberg. C. R. Schuster and H.5. Brase. 1). A.. H. H. Loh and F. L. Wai, Comparison oif the Menz Antagonism of gut but not central effects of morphine

effects of morphine on locomotor activit) - analgesia and pri- A it h quaternar% narcotic antagonists. Etr J Phatrmin * 78:mar% and protracted ph~ sical dependence in six mouse strains. -1-5-161. 1982..Piai ol Li;' I/icr 201: 3681-374. 1977. 19. Saw~nok. ._.C. Pinsk% and F. S. LaBella. On the speciticit oit

6. Broswn. 1). R_. M. J Robertson and L. I. Goldberg. Reikersal of naloxone as an opiate antagonist. Lit S, 25: 1621-1632. 19'9.miorphine-induced catalepsN in the rat bx narcotic antagonists 21). Shuster. I.G. W. Webster. G. Yu and B. F. Flethenisa. Aand their quaternars deris~atises.i'ri/iriihc 22: 317- genettc analI sis of the response ito morphine in mice- Analgesia

219193 and running. PNis,/iiUM, 01,cia 42: 2494-254, 19";7. B roii, n. R . G.. 1). C. Derri ngton and D). S. Segal . Comparison of 2 1 . Siegel. ,Voiiparainetrit Stauhiii % tM tt,, Behi i al SI i,

opiate- -and opioid-peptide-induced immobility .4 it 8i 24: New York: McGraw%-Hill Book Co..- 1956. pp. 171-I19.9313-942. 1979. 22. Szara. S. Opiate receptors and enydogenous opiates Panorama

8. Caste Ilano - C. and A. 4 lis erio. A genetic anal~ sis of morphine- of opiate research. Pl- S,,L! pi, /s 'hamu,, ,] 'I l Ru I /'

induced running and analgesia in the mouse. P%', hoiplir- ionx 6: 3-I5. 1982.mm, '('e 41: 197-2W8. 1975. 23. Teitelbaum. H.. P. Giammatteo. and G. A. Mickle Diffeiential

*9. IDunstan. R..- C. I . Broekkamp and K. G. t~lo~d. Involvement effects of localized lesions of n . Accumbens on morphine- -andof caudate nucleus. am igdala iir reticular formation in amphetamine-induced locomotor hN peractii,tii in the ( Yil. 6.1

-. ~~neuroleptic and narcotic c ataleps) . 1wp/im~a I R, hieu, Riia mouse. J Coimp P/is.i il Pik, /iil/ 93: 745-;. 1 I97~914: 169-174, 1981. 24. Valentino, R. J.. S. Herling. J. H. Woods. F. Medzihraidsk% and

W.Ferritt. P.. (G. Bianchi. A. I iani and L.. Manuira. Inhibition of H. Menz. Quaternar ' nalitrexone: Esidence to, the centraill~gastrointestinal transit and antinociceptise effects iif morphine mediation of discriminatise stimulus effects of narcotic .igonists

* and FK 33-824 in rats are differentli prevented b) nalo-,one and and anytagonists. J Phiarma- il Vier/u 217: 652-0s9 . 1981 .-

Iits N -met hNI quate rnarN analog. Re% -Coimuu Su/it 4I/,tcsv 25. Vany Abeelen, 1 . H. F. and C.M. Van den Heiisel Ilehasorail2:I-11. 1981. responses, to nosreliN in tsyo inbred mouse strains ifter

11. Koffer. K. B-. S. Berne~k and 0. Horn~kieyict. The role of the intrahippocampal naloxone and morphine. RB, at Blai Re, 5:corpus, striatumn in neuroleptic -and narcotic -induced cataleps) . 199-2(17, 1982.tin .1 I'/arimd ~ 47: 81-86. 197M, 26. Wilcox. R. F. and R. .A. I-c% titt. Naloxone re% crsal oif morphine

12. Koob. G. F..- H. 0. Petti. A. Fttenberg and F. Bloom. Effects catatonia: Role of caudate and periaqueductal gra\ . i'ansa, ,Iof opiate antagonists and thei, quaternar derisatises on heroin Rio,/,ei,, Rehac 9: 425-428. 1978.self-administration in the rat. .1 PJhnai o F I:it, I len 229: 481- 27. Wilcox,- R . F.. M . Boiart h and R . A. I esitt Re s eral oft

* 486. 1984. morphine-induced cataleps) b% naloxone microitmectiins intos13. Lee. H. K.. C. Y. (hai. Ni. J. Wakner. IL. C. Kao and P. Mi. brain regions with high opiate receptor binding: A preliminar%

Chung. Mesencephalic central gra) : Locus of morphine and report. P/,,rmnuis/ Bi,/'u Re/ti,, ISt1: 51-54. 1983electrical stimulation induced tail erecttion. I'/aru,,uol Rj,/iem? 2S. Winer. B. J.5 bStaii Il Prim, p/i's tit I .xsii,,u Jesw u Nse"Rehia, 9: 221-226. 1978. York: McGraw-Hill Book Co..- 1971. pp. 35-37. 262-283.

14. Miienberg. Gi. and W. H. Simmoins. Peptides and the blood- 29. Yeung. J. C..- T. L. Yalksh and F. A. Rud%. Effects of brainFbrain barrier, lift, N,j1 32: 2611-2623. 1983. lesiosns on the antinociceptise properties of morphine in rats.

(liii I'/inuit,d' P/in iol 2: 261-268. 1975.

ARMED FORCES RADIOGIOLOGYRESEARCH INSTITUTE

SCIENTIFIC REPORT

SR85-32

:::: ..-..-. ..

i Inn R,r Ph, (hern /9N4 I..? i ;<

.pmrh,4

1 ('4 16, 4,nuai RetieK 1n Inc r1,1qhts r'ertd

ELECTRONIC PROCESSES INORGANIC SOLIDS

Martin Pope

Radiation and Solid State Laboratory and the Depa-tment of Chemistry.New York University, New York, New York 10003

Charles E. Swenherg

Radiation Sciences Department. Armed Forces Radiobiology ResearchInstitute. Bethesda. Maryland 20814

-. INTRODUCTION

In recent years. the number of studies of electronic processes in organicsolids has assumed explosive proportions. In retrospect. it was inevitable.The increasing ability of organic chemists to design and synthesizemolecules and structures almost to order has made it possible to provide an -'}-'.

" experimental testing ground for what were once figments of the imagi-nations of theorists. Thus excellent approximations of ideal one- and two-

" -dimensional solids can be prepared with electrical properties varying from* those of an insulator, to those of a superconductor. Disordered systems can

. be prepared with a large range of nearest-neighbor interaction energies..- with and without additional trapping sites, making it possible to test almost

any conceivable theory of energy or charge migration in a random orregular network. Increasingly sophisticated computer simulations areproviding unusual insights into the microscopic details of dynamicalprocesses. such as carrier or energy transport. These serve as a check onanalytical theories and even a guide to indicate which assumptions arelikely to be reasonable. The simulations can probe situations that are as yetnot readily accessible to experiment, such as time domains in the

• femtosecond range. All of this new interest has been superimposed on whathas been an orderly growth in a relatively isolated field, the study of theelectrical and optical properties of polycyclic aromatic hydrocarbons

6130066-426X 84 1101-0613S02.00 ri

' ,.--. .--

614 POPE & SWENBERG

exemplified in anthracene. In the middle of 1982, our book appeared (1) inwhich most of the material to be presented herein approximately 40 pageswas discussed in 821 pages and, at that, with apologies for significantomissions. It is thus obvious that this review represents an even moredrastic condensation and omission of important work.

Books and reviews that have appeared since the beginning of 1982include that edited by Mort & Pfister (2) dealing with triboelectricity,charge storage, piezoelectricity, pyroelectricity, energy transfer, photo-conductivity, and polyacetylene. Organic semiconductors are discussed byHaddon et al (3). A wide ranging conference on the organic solid state washeld in 1982 covering such topics as solid state photorearrangement,chemistry, semiconductivity, superconductivity, the polyacetylenes, elec-tronic structure, and many others (4). Several lucid review articles have beenwritten by Duke and his colleagues. These discuss conductivity (5), --

electronic structure (6, 7), and electronic excitation (8). The preparation ofgood crystals of pure materials is fundamental to the development ofmeaningful theory. Two books dealing with this subject are the recent workof Sloan & McGhie (9) and the earlier work of Karl (10).

Mention must be made of the remarkable results that are being observedby the combination of tunable lasers and supersonic molecular beams ofisolated molecules in the study of intramolecular dynamics such ascollision-free vibrational energy decay. A review of this important field isgiven by Smalley (I I). The study of condensed phase molecular dynamics,including transport and trapping, has been reviewed by Fayer (12). Theeffect of pressure on molecular luminescence in solution is reviewed byDrickamer (13), and a critique of high pressure studies of molecular crystalspectra has been given by Berry et al (14). A review of luminescence innucleic acids has been prepared by Callis (15). Additional references toreviews and books are made in the text.

In the course of this review we attempt to provide references to subjectsthat cannot be adequately covered due to severe space limitations. Inaddition, we make many references to our recent book (1).

EXCITON PROCESSES

Exciton TransportDiscussion is limited to exciton diffusion, percolation, and exciton annihi-lation; the reader may refer to recent reviews or papers for discussions ofthe effeet of electric fields on exciton transport (15ab) polaritons (16-18),exciton-phonon interaction (18, 19), and exciton luminescence (19, 20,20a.b). In ordered molecular crystals. exciton transport is generally " --characterized by a diffusion constant. D (I, p. 102 et seq). Sensitized

.. . . . . . . . . . . . . . . . . . . . . . . . . . .

,'.. .m'.. .. . .............. ....... ...... .-..-...... '5

ELECTRONIC PROCESSES 615

luminescence (21) produced by exciton diffusion to a suitable trap anddelayed fluorescence resulting from exciton-exciton annihilation (1,pp. 135-47) have been used to determine D. In the usual interpretation, theexciton diffusion constant (D) is related to -, the annihilation rate constant,and K, the exciton transfer rate, by the relationships y = vy' = 8nRD and

% -K = pvK' = 41rRcD, where v is the volume per molecule of the crystal, p therelative guest concentration, and Rd and R, are the effective radii forannihilation by another exciton and capture by a trap, respectively. Kenkre& Schmid (22) have shown that the use of the above relationships, which arederived from coagulation analysis, would yield the same values for thediffusion constant only under special conditions; values of D inferred fromeither 7 or K could therefore be incorrect. A proper theoretical analysis ofthe experimental data can be obtained by noting that for trapping at a guestor defect site, the trapping time is the sum of the time to get to a trap, l/M,plus the time, 1/c, needed for the (local) capture process to occur; M is themotion rate defined as (22)

M { exp(-tr)TP(t) dt I.

where TPo(t) is the exciton self-propagator, and r the exciton lifetime. Forexciton annihilation a similar sum relationship holds, except that themutual annihilation time, lib, replaces the local capture time l/c. In termsof rates (-' and K') the following general relationships have been shown tobe valid (22):

= M- +b- (K') - ' =M +c -. 2.

It is only in temperature regimes where either y' or K' is dominant thatinformation about the magnitude of D can be inferred. These formulaewere applied by Kenkre & Schmid (22) to naphthalene and anthracene,where both y' and K' are known over a broad temperature range. For thespecial case of incoherent three-dimensional motion, where M = 4F, Fbeing the nearest neighbor transfer rate, a lower bound on the singletexciton diffusion constant, D = Fa'/6 can be inferred, where a is thenearest neighbor distance. In the particular case of the naphthalene singletexciton, at T = 300 K, D > 2.1 x 10- s cm' s-', while for anthraceneD 2! 4.3 x 10 cm' s. A similar type of analysis is possible for tripletexcitons provided there are temperature domains where either ," > K' orK' > ": under these conditions, it may be inferred from Eq. 2 that D > K'or D > ,".

An important technique for measuring D. particularly for triplet excitons,involves the use of Ronchi rulings (231 1, p. 173). It lends itself easily to thestudy of the anisotropy of D, and may be used to test for coherence in

. . . . . . . . . . . .

..... .... .....o -.) - o .• . " . .° • . ,-... . ..-..... •.. . . .. . .. . . . . . - % ° ." , , • ° . . - . , ' ° . " . . . . . . " ° "o ." .. °

616 POPE & SWENBER(i

exciton motion. Kenkre et al (24) have adopted the generalized masterequation IGME) approach to derive expressions for steady state delayedfluorescence as a function of Ronchi ruling period that hold for any degreeof coherence. The GME approach has also been adapted for use insensitized luminescence studies (25). %

Energy transport in mixed molecular crystals has attracted considerable .1,

attention. in part because these crystals. with their random distribution ofcomponents, provide useful models for testing theoretical concepts ofdisordered solids. The most extensively studied system has been thenaphthalene-4guest)-perdeuteronaphthalene (host) mixed crystal dopedwith a third component. fl-methylnaphthalene JBMN). present in concen-trations generally less than 10-3 mol fraction (26). The third componenthas fluorescing levels sufficiently low to serve as a sensor or supertrap forexcitons migrating among guest molecules. The experimental parametersare the time dependence and integrated intensities of the luminescence fromboth guests and supertraps, as a function of guest concentration.Measurements are performed at low temperatures ( -4 K) so that the hostsites act as barriers (antitraps) to energy transport among the guestmolecules. For both triplet and singlet exciton transport in CtoH,/.CoD, BMN the sensor emission depends in an almost step-functionfashion upon guest concentration, (CQ), increasing rapidly over a narrow(critical) range (27; 1, pp. 125-34).

Several theoretical models have been suggested to account for theseresults, each of which has some validity. In the percolation model, the sharptransition is attributed to the formation of a minimal macroscopicconnecting network of guest sites at the critcal guest concentration, C..Percolation analysis requires extensive computer simulations. Analyticalsolutions have been proposed by Blumen & Silbey (29) and by Loring &Fayer (30). Blumen & Silbey use a continuum kinetic equation formalism,and Loring and Fayer use a Green's function approach to solve the mixed-crystal kinetic master equation.

Several percolation models have been considered (26, 31, 32). Staticpercolation assumes that there is a cutoff on the allowable transfer distanceand neglects dynamics within clusters; thus for any cluster containing asupertrap. the probability is unity for capturing a guest exciton that landsanywhere in the cluster. The latter approximation is called the supertranslerlimit (26). If the supertransfer condition is relaxed so that not all excitons W_within a cluster containing a sensor are trapped because of limitationsimposed by the finite lifetime of the exciton and its finite rate of transfer,then quasistatic percolation ensues. For singlet excitons. Gentry &Kopelman (31. 33) have shown that the quasistatic percolation modelagrees with experiment at 1.8 K. but fails to account for data acquired at

C.

-' ..

ELECTRONIC PROCESSES 617 -

4.2 K. On the Other hand, the two-dimensional continuum models of bothLoring & Fayer (30) and Blumen & Silbey (29) are consistent with theexperimental results at 4.2 K but not at 1.8 K. In the Loring & Fayer ap-proximation (30) the trapping probability is given by P = 1 - KC(0,O = K d,) ,Vwhere K ' is the guest lifetime, and C is the Laplace transform of thetime-dependent part of the system Green's function for the probability of --

finding excitons somewhere within the guest manifold. This relationshipcan account for the T = 4.2 K data only if the FOrster distance (1, p. 100)Ro = 8 A and the transfer rate, W, has an octupole-octupole distance We._dependence, i.e. W x R ". A weakness of the model is that these parametersimply a transfer rate of 5 x 109 s-, whereas expected nearest neighborrates are the order of 102 sec '.The lack of agreement with the 1.8 K datahas been discussed by Gentry & Kopelman (33) and is attributed to theimplicit assumption in the Loring & Fayer model of the equality of theforward and reverse transfer rates, Wj = Wj; this equality is known not to Lg.hold for naphthalene singlet excitons at low temperatures (34).

The Blumen & Silbey model (29) calculates the average probability for anexciton to land on a supertrap site and the average trapping time. Theirresults, in the absence of back transfer from the supertrap, are

1 Xc {(C,/C 8 )C, dj 3.

Px0 {+ C I /C21C 4.

where the energy transfer rate constant is Ke, the sensor concentration isC,, the guest concentration is C. and C1, 2. is that concentration at whichP = 0.5; n is the multipole transfer exponent (n = 14 for octupole-octupole 2transfer) and d is the dimensionality of the system. Equation 4 providesan excellent fit to the 4.2 K data. however, it does not account forthe sharpening of the critical transition upon lowering the temperature.Furthermore, recent time-dependent emission data of Parson & Kopelman.- - -

(32) performed on the same mixed crystal system at 1.7 K demonstrate thatthe sensor emission profile is more rapid than expected from rate equationanalysis (29) for guest concentrations near the transition regime (guestconcentrations between 0.42 to 0.57 mol fraction).

The temperature and guest dependence of luminescence from the mixedcrystal system C 0 H8/C 0 D8 jBMN has also been studied by Brown et al -'-*

(35), who found evidence for a transition from coherent triplet exciton b 'motion at very low temperatures to hopping motion for T > 10 K. ..Transitions from band-like to hopping motion of triplet excitons have alsobeen studied by Gentry & Kopelman (36).

In addition to protonated and deuterated isotopically mixed crystals,orientationally disordered solids (37, 38) are excellent model systems for

* .,..**. . . . . . . . . . . . . .

- S... .-.. -

618 POPE & SWENBERG

studying the static and dynamic properties of electronic excitations indisordered solids. In these systems local disorder is provided by the randomdistribution of orientations of the substituent groups, e.g. the chloride andbromide ligands of l-bromo-4 chloronaphthalene (BCN) crystals.Orientational static disorder was studied by Morgan & EI-Sayed (39), whomeasured the zero phonon S, - Tt transition at :4.2 K in BCN singlecrystals. The absorption line shape was Gaussian, which is characteristic ofinhomogeneous broadening; the line width (HMHW) was -32 cm-',which is approximately two orders of magnitude larger than the linewidthsof the corresponding transition in single crystals of 1,4-dibromonaph-thalene and 1,4-dichloronaphthalene. In addition, Morgan & EI-Sayed (39)have shown how the temporal behavior of the phosphorescence emission atdifferent laser excitation energies in the long wavelength region of theabsorption band provides a measure of both the energy transfer rate and itsmechanism.

Exciton InteractionsExcitons can interact with other excitons, and with carriers, defects,photons, phonons, and essentially any other entity that can be reached by amobile electronically excited state (1, pp. 89-180). Exciton-exciton annihi-lation in neat molecular solids is characterized by an annihilation rateconstant y. The limitation of the assumption that the depletion of excitons(n) due to bimolecular annihilation is given simply by - 'n2 has beenreviewed by Kenkre (25). The role of boundaries and domain (or cluster)size on exciton kinetic rates is of considerable current interest [see Hatlee &Kozak (40) and references quoted]; confinement to small domains isexpected to enhance the number of collisions with other excitons, defects, ortraps. In the particular case of tetracene triplet excitons created in pairs insmall tetracene domains by singlet exciton fission, confinement results in anincrease in the fluorescence yield and singlet exciton lifetime as the domainsize decreases (41).

Mixed crystals are good model systems for the study of size effects onexciton reactions. Recently Klymko & Kopelman (42) have shown thatwhen guest cluster formation occurs, the delayed fluorescence (D) resultingfrom guest-guest exciton annihilation is no longer proportional to thesquare of the guest phosphorescence (P), even at low triplet excitonconcentrations. Using mixed CoH8 /C1 0 Dg crystals with dopant levels of0.1-0 to 200. CtoH 8 they found Doc Px. where x is a function of excitationintensity, time after excitation, and guest concentration. Values of x rangedfrom 2 to - 30 at T = 1.7 K and increased monotonically with decreasingnaphthalene concentration. As noted by the authors, these results areinexplicable in terms of standard diffusion kinetics (43). Klymko &

m

. . ° .

r.-. . . . . .. . . - '.. . '°,.... *' .*. -* . -. . . .. '.. - . , ' - b ' ". . . %. . ,. . *" ". ".. ".

*~~~ -7 .-. N-. av.-* --

ELECTRONIC PROCESSES 619

Kopelman (44) interpret their data in terms of fractal kinetics, which isappropriate for bimolecular processes in small domains. A discussion offractal properties is beyond the scope of this review; see Alexander &

Orbach (45) and DeGennes (46) for more information.Excitons interact with carriers (1, pp. 164 et seq), and in the process of

interacting, it may happen that a transient intermediate state is producedbetween the exciton and its collision partner. More recently the boundexciton + hole (excitonic ion) case was treated by Schilling & Mattis (47,48), who call the ion a trion as initially suggested by Thomas & Rice (49).

Here, the exciton could be either the Frenkel exciton or the Wannierexciton. The case of N excitons and one hole was solved by the sameauthors (48). Excitonic ions with Frenkel exciton parentage were discussedby Singh (50, 51) and ions based on excitons intermediate between Frenkeland Wannier excitons were treated by Gumbs & Mavroyannis (52).Experimental evidence for the existence of excitonic ions in benzene solidfilms is given by Sanche et al (53). Excitons interacting with trappedelectrons was discussed by Agranovich & Zakhidov (54), who deduced anattractive interaction, which is the order of e'Ao/2er, where Am is the change Iin the molecular polarizability of the excited molecule, - 10- cm', r is theseparation of the charge e from the exciton, and e is the local dielectric

constant.One intriguing possibility that has not yet been ruled out theoretically is

that of a Bose-Einstein condensation of Frenkel excitons in organiccrystals, particularly triplet excitons (55): attempts are being made to detectthis phenomenon, or at least the formation of exciton aggregates. A reviewof these experiments, particularly in the Soviet Union, is given byBrikenshtein et al (56).

Exciton Trapping

Trapping of excitons at either impurity or guest sites has been a subject ofconsiderable study because of the insight gained on the mechanism ofexciton diffusion (21, 25). Studies by Powell and co-workers (21) hadsuggested that the exciton transfer rate K was time-dependent, in signifi- :.-

cant departure from earlier results obtained in sensitized luminescenceexperiments (57). A careful study of the time-dependence of K has recently '''.

been made by in tetracene-doped anthracene by Braun et al (58). Theydemonstrated that for temperatures between 1.6 and 300 K and over adopant concentration range of 10 - to 2.3 x 10- mol fraction, that K wastime-independent at least for times greater than a few picoseconds. This is

-in agreement with previously reported studies of Campillo et al (59) andAI-Obardi et al (60).

A recent study of singlet exciton trapping in the system p-terphenyl

• .' .. S . . *,. .. . ..-

%',. . .- .. - - -, . . .. .. . . . . . . - . .. . .. .. ,-. -. . - -°' '. . -°,. . ,. .°.°.•

620 POPE & SWENBERG

(host)/tetracene (guest) (61) also showed that the transfer rate from host tohost was time-independent but, more interestingly, that it was thermallyactivated. It was concluded that the energy of activation was to beassociated with singlet exciton motion on the host lattice sites, anA that thetransfer from host to host was facilitated by a phonon-assisted icrease inco-planarity of their donor and acceptor molecules (62). This phenomenonis reminiscent of that observed by Meyer et al (63) described herein. Theconcentration of sites at which an increase in co-planarity might befacilitated (predimer or incipent dimer) can reach 10- ' mol fraction (64).

The theory of trapping of excitons at guest sites in the presence of low"p. trap concentrations has been considered by several authors (64-67). An

analytic theory that accounts for the guest and host fluorescence yield forall guest concentrations has recently been given by Kenkre (68) and Kenkre& Parris (69), who use the general master equation (GME) governingexcitation transfer (1, p. 108). They gave explicit expressions for both guestand host luminescence yield as a function of trap concentration (69).

In addition to exciton trapping at impurities and guests that have lowerexcited state energies than host molecules, it is also possible for self-trapping to occur in the host lattice (69a). Self-trapping of Frenkel excitonsin homomolecular aromatic organic crystals was unlikely, according toToyozawa & Shinozuka (70). However, under certain circumstances, theformation of an excimer that lies energetically beneath the exciton bandmay be viewed as a self-trapped exciton (1, pp. 85-89, 257-67).

The association of broad excimer-like luminescence with the existence ofa self-trapped state necessitates (a) simultaneous observation of both freeexciton and excimer luminescence. (b) an activated temperature dependenceof the emission bands, i.e. the existence of a small energy barrier between thefree state and the self-trapped state, and (c) the association of the emission ofthe trapped state with the intrinsic property of the crystal and not withdefects. Previously reported excimer emission in fl-9-10 dichloroanthracene(71), initially identified with a self-trapped exciton luminescence, is nowknown (72) not to satisfy the above criteria, since the monomer emissionarises from defects. This conclusion was reached after it was found thatcrystal annealing greatly reduced the monomer emission. Thus, in this case.excimer formation was not a thermally activated process. -.

Evidence of exciton self-trapping in x-perylene was initially provided byvon Freydorf et al (73). who observed a weakly structured emission (calledthe Y-emission) at low temperature (at slightly higher energy than thebroad excimer fluorescence band) that had a long first-order decay time(: 38 ns). In fi-perylenb crystals the temperature dependence of theluminescence spectrum provides evidence for at least two types of self-trapped states: at least one of them is in thermal equilibrium with free L"

. .

• ..- • .. . .% -, " . -. . ' -. 2 -, ., -i , - - '. . .- -.• . . -' .' : -' i • ; -. i .' ,- - , " ." , ' .. , -' . .- .

77 -7 .7o

ELECTRONIC PROCESSES 621

excitons 174). Additional evidence for self-trapping in both crystal phases ..- %comes from the work of Matsui et al (75).

SPECTROSCOPY

Lineshape Analt'sisA theoretical analysis of EPR. NMR. and optical lineshape data canprovide a wealth of information about structural and dynamical propertiesof solids. The inhomogeneous linewidth of optical lines is determinedprimarily by the degree of static disorder, and for this reason it is sometimesused as an indicator of the structural quality of crystalline materials. Thehomogeneous linewidth reflects intrinsic properties such as lifetime effects(T) 176) and the coupling strength of a particular excited electronic state tothe local dynamical (phonon) modes (T2). The homogeneous linewidth ofoptical transitions is not an easily accessible parameter; however, studies offluorescence line-narrowing (77), hole-burning (78), fluorescence saturation(79), and optical coherence such as the photon-echo (80) and stimulatedphoton-echo (81) are methods that can yield the homogeneous linewidth ofa particular vibronic state [see supersonic studies of (82, 82a)]. Hole-burning spectroscopy is a static method for investigating the dynamics ofmolecular vibronic states. In non-photochemical hole burning (N PUB) aparticular set of molecules tuned to the laser frequency are excited but notchemically altered. The excited chromophore produces a change in itsmicroenvironment, converting it rapidly to a new structural arrangementbefore deactivation; this decreases the number of chromophores havingexcitation energies at the laser frequency and increases the number ofchromophores having excitation energies at other frequencies. The netresult is that a hole is burned in the absorption band, the profile can beprobed by measuring the fluorescence excitation spectra. When spectraldiffusion processes are negligible, the hole shape is Lorentzian and its shapewill be a measure of the true homogeneous lineshape [FWH M = (n T2 ].

N PH B has been observed in a wide variety of amorphous matrices at lowtemperatures (83). Jankowiak & Bassler (84, 85) studied NPHB in tetracene-" 10' mol fraction) incorporated into amorphous anthracene at 4 K.Holes burned at T 2.5 K at ,. = 491.5 nm show a pronounced zero-phonon hole and two almost symmetric one-phonon sideband holes: insome materials multiple phonon sidebands can be detected (86. 87). Theratio of the zero- to the one-phonon sideband intensities gave an electron-phonon coupling constant of -- 0.9 and a phonon energy of : 40 cm -'forthose phonons which interact most strongly with the electronic state. Theholes disappeared after annealing at 40 K for one hour. The homogeneouslinewidth F (at FWMH) of the zero-phonon hole varied with temperature

.... ..............................

>.°'N'" -""" .". .... " _-_..__ .___ "_-_ -"-".".-___"___-_.-.-.-''_.-'"__""_."__•_-"-__"._._-.-.-"__"._"_ " "'-.--. "" .- " -

622 POPE & SWENBERG

as F(T) =- o + aT 2 where o 4.0 cm 1 .this corresponds to a dephasing

time on the order of 2.5 ps, which is 3 x 10' times faster than the radiativedecay of tetracene in solution [for a study of vibrational dephasing innaphthalene, see (87a)]. The temperature dependence was rationalized interms of the Reineker & Morawitz theory (88), which considers adistribution of tunneling states (TLS) coupled to the matrix acousticphonons. -,. .

In photochemical hole burning, a photochemical process, such asproton-transfer (detachment), occurs; the photoproducts in general haveabsorption bands that lie outside those of the inhomogeneous band of thereactant; and if the photochemistry is irreversible, there will be no loss inthe integrated hole intensity with time.

Fluorescence saturation methods have recently been applied byTreshchalov & Rozman (79) to determine the homogeneous linewidths ofthe first two inhomogeneously broadened vibronic states (400 cm-' and1400 cm-' phonons) of molecular anthracene, at 10-s mol fraction innaphthalene. The technique involves a measurement of the fluorescenceintensity (F) as a function of the excitation intensity (I.) and is applicable to r.--

guest molecules that are photochemically stable. For both steady-state andpulse excitation conditions, F ; 1. in the low intensity regime. At high

excitation intensities F c II.. Treshchalov & Rozman (79) reported anenergy relaxation time T, equal to 40 and 25 ps for the 400 and 1400 cm..vibrational modes of anthracene ; for comparison we note that in a fluorenehost, hot luminescence spectral measurement gave 28 and 22 ps, respec-tively, for these states (90).

The introduction of femtosecond time-resolution capabilities into thefield of spectroscopy will probably provide the ultimate measure ofrelaxation dynamics. So far, the shortest-lived molecular process that hasbeen directly time-resolved is the decay of the Rydberg 3R. state in gas-phase benzene; this is 70 + 20 fs, as measured by Wiesenfeld & Greene (91). "-

The decay mechanism was not established, but may involve transitionsdirectly into the high vibrational levels of the ground state. This time scale is -'-

so short that intermolecular collisions could be neglected. The ionizationrate is comparable to the relaxation rate.

Photoelectron Spectroscopy

With the advent of synchrotron light sources, the energy region between far _UV and X-ray has been made accessible to experimenters. A discussion ofthe principles of this light source and a description of recent photoemissionexperiments on phthalocyanines (Pc) is given by Koch & Gurtler (92). Anexamination of photoelectron energy distribution curves in Zn-Pc showsthat the excitation of 3d to 4s states in the metal Pc is enhanced relative to

-,.

'-- -. ,-- --

. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .

" . ..... ... . ...... ....... . . ... .. . . .f

ELECTRONIC PROCESSES 623

that in the free metal atom because 4s states in the metal Pc are transferredfrom the metal site to the ligand, increasing the availability of empty 4sstates for occupation from 3d levels. In a similar way it was found that thereis a smaller number of empty 4s states in Cu-Pc than in a Zn-Pc, because the3d - 4s transitions are stronger in the Zn-Pc. This hybridization of metal3d electrons with the jr-electron structure of phthalocyanine ligand isparticularly important in understanding the high conductivity of theporphyrinic molecular metals (93). In many studies of photoemission, the ," ",three-stage model of Berglund & Spicer is used (I. p. 533) because of itssimplicity. A good discussion of this model has appeared, using anthracenedata, showing that structural defects must be taken into consideration (94).

A singularly clear and powerful demonstration of the utility of photo-electron spectroscopy coupled with the application of a simple theoreticalmodel to determine the electronic structure of solids, particularly polymers,continues to emerge from the laboratory of C. B. Duke. A CNDO/S3Icomplete neglect of differential overlap, self-consistent field method)model, originally developed to rationalize the photoemission and UVabsorption spectra from polyacenes. was extended (95) to include infinite(periodic) macromolecules, as for example polyacetylene. In a companionpaper (96) this method was used to rationalize the ultraviolet absorptionand emission spectra of pyrrole and polypyrrole. In the calculation, it wasfound by varying the nymber of pyrrole units in the polymer that some ofthe essential features of the polypyrrole photoemission spectrum appearedwhen 4 to 6 pyrrole units were connected, implying that the photoemissionhole state extends over at least 4 to 6 pyrrole units, or - 14 A or more. In asimilar way, it was found that a bonding t -+ n* transition extending over atleast 4 pyrrole units is responsible for the 3.0 eV peak in the absorptionspectrum. By simulating the twisting of the pyrrole units, it was found thatthere would be no major changes in the photoemission or absorptionspectra relative to that of a planar molecule, so no information abouttwisting may be deduced from the spectra. In another paper (97) theelectronic structure of poly(p-phenylene) (PPP). poly(p-phenylene vinylene-(PPV), and poly(p-xylylene) (PPX), as well as the previously studiedpolyacetylene and polypyrrole was calculated. [For ab initio calculations,see (97a).]

CARRIER GENERATION MECHANISMS

IntrinsicAmong the homomolecular polynuclear aromatic hydrocarbons (PAH)compounds. anthracene has long been the hydrogen-atom for theoreticalcalculations. As one proceeds from naphthalene to pentacene, for example.

*1

7-77

Z_- -

I .,--

' . . . .. . . . . .. . . ............. ...-°-•°°... . , ....

624 POPE & SWENBERG N

the optical absorption spectrum shows a significant increase in thecontribution of states that have ionic character (98). This increase in ioniccharacter is paralleled by an increase in the quantum efficiency ofphotogeneration. However, there is not yet unanimity on the association ofthis increased ionic character with a specific mode of ionization.r One hypothesis has it that carrier generation requires the excitation of aprecursor Frenkel exciton that dissociates when its energy is degenerate

with that of a pair of free carriers. This process is referred to asauto ionization (AI). It does not follow that all Al transitions lead tocompletely uncorrelated carrier pairs. The electron thermalization distanceis - 60 A I1, p. 489). Thus, most electrons that are created inside the solid willthermalize within the Coulomb capture radius of the geminate positive ion,forming a transient charge-transfer tCT) state that can either decay to theground state or dissociate by the absorption of ambient energy. Forexcitation energies exceeding that of crystal ionization, electrons arephotoemitted with the maximum kinetic energy (100. 101) as calculated bythe Einstein photoelectric equation, or with kinetic energies diminished byamounts equal to the energy of excitation of the remaining positive ion.This is proof that no intermediate CT state is necessary for free carriergeneration.

% However. there is strong experimental and theoretical evidence thatdirect optical excitation to a CT state takes place in homomolecular solids(102. 103). particularly in the more highly colored members of the polyacenefamily and particularly for excitation energies less than that of the band gap.The proposition that direct CT generation was the primary mechanism forphotoconductivity in most. if not all, PAH crystals was put forth in a seriesof theoretical papers by Bounds & Siebrand (102, 103) and more recently byBounds. Petelenz & Siebrand (BPS) (104, 105).

The fundamental premise of the theoretical work of (BPS) (based onanthracenel is that the direct excitation of the CT exciton is made possibleby the coupling of this low oscillator strength state, with that of the highoscillator strength (f Z: lS 3(1 B) Frenkel state located at 4.63 eV. Inanthracene the S 3 state is about 0.5 eV higher in energy than the anthraceneband gap Eq : 4.1 eV. and all optical excitations in the energy region up to4.63 eV would produce CT states (albeit, vibrationally excited), in otherwords, there would be an energy of activation for photoconductivity due tothe dissociation of the intermediate CT state even when the optical energynv > Eqo this is observed (106). The energy EcT of CT states of varyingseparation distance between the charges was calculated quantum mechani-cally by BPS: the CT energies were capable of being described qualitativelyby a simple Rydberg-like expression of the form ,.

E.r(n) = E,-pe " 8(hrron)2

1'..!

I:.-...

* PV

ELECTRONIC PROCESSES 625

where ju is the effective mass of the electron-hole pair, Eo the dielectricconstant parameters, and n is a principal quantum number, related to thedistance r between electron and hole. The E-T(n) values can be related to thedifferent activation energies E,(hv) found for carrier generation by light of -

energy hv by the equation E. = E,- Ec4n). The explicit assumption ismade here that while it is possible to excite vibrational states of CT excitonsfor each value of E., only the 0-0 vibrational state is available for thermaldissociation, the excess energy being dissipated in a time shorter than thatfor dissociation or recombination. The agreement between theory and

-' experiment (107) for the E. values is good, but not so good for the quantumefficiencies of carrier generation 40 (hv). It was possible for BPS to estimatethe contribution of CT states to the overall absorption spectrum; inanthracene this contribution was small but significant, while in tetraceneand pentacene, the CT contribution was dominant.

Considerable experimental support for the CT exciton mechanisms forcarrier generation has been provided by work on electric field modulatedabsorption spectroscopy in pentacene (108. 1, p. 574). Similar studies weremade on anthracene by Sebastian et al (109), although it proved to be moredifficult to detect the CT states. These authors correlated the energies of thepeaks that were attributed to the CTexciton electroabsorption spectra withthe size of the CT exciton using a Coulombic equation. The extrapolatedvalue of E, proved to be 4.4 eV, in contrast with the accepted value of about4.1 eV; they justified this discrepancy by identifying their CT values withthe vertical (Franck-Condon) transitions from a Frenkel ground state tovibrationally excited CT states. Furthermore, in opposition to a basicpremise of the BPS treatment, they postulated that the excess vibrationalenergy for any particular CT state could be used to separate further themembers of the ion-pair state. This postulate, in a sense, combines the .

processes of direct CT exciton generation with the ballistic carrierseparation mechanism that is implied in the Al mechanism. The identifi-cation of the structures found by Sebastian et al (109) in the electroabsorp-tion spectrum with CT states of different separations rCT has beenquestioned by Siebrand & Zgierski (109a). These authors calculated thespectrum of CT states in anthracene and attributed the structure observedby Sebastian et al (109) to the vibrational overtones of the nearest-neighborCT state (110). A resolution of the discrepancies between theory and

t experiment is not yet at hand.The Al mechanisms of ionization coupled with the ballistic model of

electron-hole separation has been discussed in detail by Silinsh et al (111),who studied'tetracene and pentacene. Here, Al steps are succeeded by theescape of the hot electrons into the lattice, where they thermalize byacoustic phonon scattering, producing the series of bound CT states. Usinga rather simple scattering theory, Silinsh et al (I 11) calculated the optical

7- .--. .-

626 POPE & SWENBERG

energy dependence of the thermalization distances, ro(hv). which in turncorrespond to specific separations of CT states. They found good agree-ment between their calculated and experimental values. Calculations werealso made of the photoconductivity quantum efficiency, and agreement wasfound between calculated (using ballistic theory and a modified Onsagertheory) and measured values in the energy region E(hv) > Eq. In the region ,.E(hv) - E., they found a discrepancy that they attributed to a contributionfrom direct CT state absorption. In the higher regions of photon energy, "..'they found another discrepancy that they attributed to excitation to adifferent Al state.

In addition to the cases considered above there is another interestingexperimental result in which it appears that the efficiency of photoconduc-tion is independent of photon energy (112). This work was carried out withX-metal-free phthalocyanine crystals embedded in a polymeric matrix. Itwas found that all the ionization proceeded from the first excited singletstate, S1 .

Summarizing the results presented herein regarding the relative roles ofAl and direct CT exciton formation, it appears that there are materials andenergy ranges in which one or the other, or both, mechanisms prevail, thispoint has been made by Silinsh et al (t 11). The actual determination of therelative roles of Al and direct CT exciton formation might be resolvable ifphotocurrent rise time measurements could be made. This suggestion was -.

made by Bounds et al (105). '"" "

ExtrinsicCarrier injection by means of a suitably chosen electrode is a convenientand widely used method for producing a mobile carrier density in organiccompounds (1, pp. 273 et seq). The time-resolved study of the injectionprocess, which makes it possible to expose its microscopic details, can befacilitated by the fact that the rate constant for the recombination of theinjected carrier with the electrode can be made arbitrarily small bychoosing an electrolyte as the injection electrode. Using this fact, Eichhornet al (113) have carried out the first time-resolved measurement of theescape of charge carriers by diffusional motion out of a Coulombic wellnear the injecting electrode. This well is created by the attraction of theinjected carrier to its image in the electrode. The complete model consistedof a hole injected into the top layer of molecules in the crystal, which couldrecombine with the electrode with a rate constant k..., or decay to a traplevel with a rate constant ktr. The trapped hole could either recombine withthe electrode with a rate constant k,,rec or be activated back into the freestate with a rate constant kd. The rate of escape of the electron out of itsCoulombic well was k,. A fairly good agreement with experiment was

D.'

......................................... . -

ELECTRONIC PROCESSES 627

found using 4 A as the effective boundary distance from which carrierinjection proceeded, and the values kd = 5 x 10 -', k, 1 = 4 X 107 S,.k, = 3 x 10's - . c z 0. Other conclusions that were made were that thehole mobility in the c' direction was independent of electric field up to atleast 5 x l0' V cm - 1 and also at 1 x 10' V cm (114). The recombinationwas assumed to be nongeminate, and to involve an OH - ion. There is a

N' problem with this assumption because the energy required to discharge anOH- ion in a neutral aqueous solution is --6.8 eV, referenced to thevacuum zero (115), whereas only 5.8 eV are available as a result of thedischarge of the anthracene hole. Chemical attack by a water molecule is amore likely route for this process.

CARRIER RECOMBINATION

Time Independent (Theory)The Onsager theory of geminate recombination (116, 117) was designed as asteady state solution for a pair of oppositely charged geminate particlesmoving in a condensed phase continuum in the presence of an externalelectric field. Nevertheless, it has been widely and rather successfully used tointerpret the electric field dependence of the quantum yield of photo-generated carriers in crystals, where the carriers undoubtedly are movingon a discrete lattice 118) (see 1, pp. 481 et seq). Recently, the Onsagercontinuum theory was extended to include the transient case (119), and theoriginal Onsager assumption of a point-sink at r = 0 was refined into arecombination sphere of finite radius and recombination velocity (120). Aseminal theory has recently been introduced by Rackovsky & Scher(12 la)in which the problem of geminate recombination in a face-centered cubic(fcc) lattice is treated as a random walk on lattice sites in the presence ofboth an external field and a Coulomb field. The solution is given in terms oflattice Green's functions, modified by the presence of an applied electricfield. In this model, one carrier is fixed at the origin and the other,oppositely charged. carrier moves on the lattice in accordance with afunction 011, t) that gives the probability per unit time that it will leave a siteI in a given direction. This function is determined by relative transitionrates to the neighbors, and the transition rates in turn are determined byvariables such as wavefunction overlap, energetic differences, and electron-vibration interactions. An important factor is the role of competingprocesses, such as the decay rate of the electronically excited precursors tothe ion-pair state that dissociates. What emerges from this treatment willhave profound effects on the interpretation of experiments using theOnsager formalism. The Onsager theory has two parameters that arespecific to the system under study. These are the electron thermalization

4 ....-.

. . . . . . . . . . . . . . . . . .4•* -

- -- - - - - - - - - - - - -- - - - - - - - - - - -- - -. .. . , , *

- . . ,I.-,

628 POPE & SWENBERG

distance ro, and the initial quantum yield of geminate pair generation P'o'These parameters can be evaluated if an assumption is made regarding theinitial distribution of geminate pair distances (1, pp. 489 et seq). Scher &Rackovskv show that depending on the relative magnitudes of somemolecular parameters described in Figure 1 and using a fixed initialdistribution of charges (hence, a fixed r,), the values of 00 that would havebeen calculated using the Onsager formalism on the yield versus field plotscan be considerably different from each other. Thus, consider thecalculation shown in Figure 1 illustrating the photogeneration efficiency 11dependence on reduced electric field strength. Here E, = Ei'kT where E,is the difference in energy between the first electronically excited state at theorigin and the state consisting of a hole at the origin and an excess electronat a nearest neighbor. This could be a few tenths of an eV. The term X is thestrength of the Coulomb field between the carriers in the geminate pair, R . .gives the ratio of the relaxation rate of the initially excited state to the rate ofintermolecular charge transfer, and the reduced field strength " = eFa/2kTwhere F is the electric field and the lattice spacing is a. Notice that in the toptwo curves, only f, has changed. this could be produced for example bylocal energy fluctuations, which would be significant in the case of

. l~l35- ..a 305034 0 5 "'"-"• 5 0 5D0 0 3 .l I

'' '"""

S/ t' Figure I Log-log plot of the photogener-'/ ation efficiency I vs the reduced field strength

illustrating the effect of varying t, with y/ and R essentially fixed. From Scher &

Rackowsky (121).

.7,"

LOG

i .- . % °.%iV %%"-"*

V V ,, - '

ELECTRONIC PROCESSES 629 %

molecularly doped polymers. In this calculation, the initial distribution isfixed (i.e. the average value of ro is constant). At low fields, it may be seenthat for the lower value of E,, i has decreased by an order of magnitude. Ifthe Onsager formalism were used then for the same value of 00 the apparentvalue of r0 would have to be altered in order to compensate for thedifferences in q. For a greater Ee, and a slightly larger value of R, ?I changesby another factor of 10, ro would have to change even more, and no.'

*i saturation is evident at high field, in contrast with what is seen for the topcurve.

Another important experiment is the measurement of the temperaturedependence of the yield. According to the Onsager theory, the temperaturedependence of ol at zero field should follow an Arrhenius relationship inwhich the activation energy is E. = e2 !41rgero. In the notation of Scher &Rackovsky, this would read E, = azkT/2ro, or r0/a = XkT,"2E,. Thisprediction is not realized. When R is large, a linear Arrhenius plot isobtained but with Ea a function of applied field. When R is small there is nolinear plot.

In addition to the analytical treatment of Scher & Rackovsky, a MonteCarlo simulation was carried out by Ries et al (122) of field and temperatureassisted dissociation on a perfectly isotropic 3D lattice, and perfectly linear(ID) systems. It was found that for a perfectly ID system, the zero-fieldintercept of the dissociation probability would be zero, and the extent towhich this intercept is not zero is a measure of the degree of anisotropy ofthe carrier motion. The power of the simulation approach is that any degree ,..

of anisotropy between I D and 3D can easily be inserted.An important, and not self-evident, assumption made in the simulations

is that of field-independent hopping rates in ID and 3D systems, i.e. the fielddoes not affect either the initial pair distribution, or the prefactor to theexpression containing the barrier height (which is modified by the field).For intersite hopping energies < kT, the Onsager result emerges.

An important test of the field dependence of the hopping rates was madeby Seiferheld et al (123) using crystalline polydiacetylenes (see 1, pp. 673-99for a description of these compounds) as model compounds. Thesecompounds (PTS and DCH) are formed by the solid-state polymerizationof crystalline acetylenic monomers, as a result of which a highly crystallinesolid is formed composed of long conjugated polymeric chains heldtogether by van der Waals forces. The strong covalent forces along thechain and the weak van der Waals forces between the chains lead to a W7markedly anisotropic electrical behavior. This makes these crystals goodexamples of quasi-i D systems. The degree of anisotropy may be quantifiedby means of the ratio u,/p,, which are respectively, the mobilities parallel

Now

.: .

p . *

630 POPE & SWENBERG

and transverse to the molecular chain. The deduced ratio for themicroscopic anisotropy was > 10' in opposition to the value of - 103

indicated by current flow measurements (125, 126). The conclusion that theanisotropy is so high follows from an application of the Onsager theory ofgeminate recombination as modified for true I D behavior. In a true IDsystem, in the absence of an external electric field, the degree of dissociationof a hole-electron pair must be zero (127) because a particle executing arandom walk on a 1 D lattice must return to the origin with unit probability.In the present case, the experimental results are in accordance with theOnsager theory for a I D system for applied fields down to 30 V/cm.Furthermore, according to a computer simulation carried out by Ries et al(122), the low-field intercept of the dissociation yield is a direct measure of elthe mobility anisotropy, i.e. limF_00 (F) tai/,, where e"4 is theprobability of an electron escaping geminate recombination at the appliedfield F; no deviation from the Onsager ID behavior appeared even when

Another important conclusion of this paper follows from the low-fieldagreement with the predictions of the Onsager ID theory. According to thecomputer simulation results (122) the Onsager equations will be followed aslong as the mean free path of the carrier between scattering events iscomparable to the distance between lattice sites (< 10A). The low-fieldagreement with the Ohsager predictions then imply that diffusive transportin PTS and DCH is accompanied by strong scattering. However, for DCH, amean free path of z 100 A for carrier scattering follows from an m* = 0.05 m,(128) and a scattering time of 8 x 10- " s. This apparent contradiction maybe removed by the application of theoretical findings of Movaghar & Cade(129, 130), who have shown that carrier transport in PTS occurs vialocalized polaron states whose mean free path for scattering could be aboutthe distance between polymer repeat units. During the initial hole-electronformation, during which the Onsager formalism is not operative, theelectron could travel in the wide conduction band for a distance of about100 A before falling into a polaron state. This distance could be associatedwith the thermalization distance, which is calculated by - 65 A from otherresults.

The proposal that a polaron state is created in - 10 -t, s has otherimportant ramifications. According to Donovan & Wilson (131), electronmean free paths are -1 m in length, and the electron mobility is > 2 x10 cm' V s-'. If the mean free paths are so long, then geminaterecombination should not be observed, whereas it is. The high mobilities(132) also become puzzling. This paradox was resolved by Movaghar et al(133), who showed that standard linear response theory as applied totransport was not applicable in these I D systems.

. . . . . .. •

t - .,,.

ELECTRONIC PROCESSES 631 -%A most significant result is that at high-field IF > 104' V cm - the F--'anticipated saturation does not occur. This failure to saturate had already

been noticed in anthracene by Geacintov & Pope (134). Seiferheld et al (123)concluded that their results can only be explained by a primary ionizationyield 0o that is field dependent.

Time Independent (In Polymers)Evidence for exciton fusion as a mechanism for the intrinsic generation of

free carriers in molecularly doped polymers has been reported by Orlowski& Scher (135, 136). The system studied consisted of solid solutions of theamine TTA [tri-(p-tolyl)amine] in the polymer Lexan (bisphenol-A-polycarbonate) in concentration between 30 and 400/ by weight. Bulk (asdistinguished from surface) photoconductivity was induced in 10 pm thickfilms. The singlet exciton decays by intersystem crossing (0,, 0.95) to thelowest triplet state IT) of TTA. The phosphorescence of TTA peaks at 2.8eV, so two triplet excitons can supply 5-6 eV; this is sufficient to inducephotoionization. It was found that the photogeneration yield, q/A, variedsmoothly as the light intensity P2 at low light intensity, as I at high lightintensity, and at some intermediate value in between. With an increase in

field strength F from 5 V/pn to 20 V/pum, q/A increased by a factor of about250 at 1 = 10' photons cm - 2, and in general the I dependence itself was Fdependent. The scheme used to explain the results was as follows:

k2 '".-

,71 k, kcpTT I(TT) - P- CI~~ ~~~~ k k[a< .:,-.1k

so sosoi(TT)* = T* is an associated triplet-pair state, S, is the ground electronicstate, and C is the hole on the TTA molecule generated at the rate Kcr from Pwhich is a charge-separated state in which the polymer contributes anacceptor state. The constant y refers to the diffusion-limited process oftriplet exciton fusion, and applies to the diffusion-limited separation ofT*" K, refers to the creation of P, K 2 describes the annihilation of T* toproduce T and So. and K, represents the recombination of hole andelectrons to produce a ground state. The quantity feff is to be comparedwith the quantum yield obtained from data representing the linear responseof qA to 1, and 7,ff corresponds to what would be measured as the ..

diffusion-limited rate constant for triplet-triplet fusion to produce T*.It was found that ,ff increased from 0.16 x 10- '2 cm3 s' at F 0. to

. .--. -.....

632 POPE & SWENBERG

160 x 10-' 2 cm 3 s' at F=20 x 104 Vcm 1. At F=5 x 104 Vcm-,q,,( was I x 10 4 , while at F = 38 x 104 Vcm-'., of was 24 x 10-.

The Onsager field effect found in many systems !1, pp. 484-95) applies tothe relative magnitudes of Kcp and KR, with P given by some initialefficiency 01o, and says nothing about any of the other quantities in theschematic diagram. However, the Onsager theory cannot explain themagnitude of the observed results even if 0, is made field dependent (I l).The picture that emerges is that the T* state is probably a configurationconsisting of a mixture of a triplet exciton pair, an excited S* state, anexcited charge-transfer complex and some other configurations as well. Theelectric field would have the effect of increasing the tunneling rate of the CTstate to the P state, thus enhancing the value of /.ff- The assignment of CTcharacter to the intermediate triplet-pair state has been made in the past(137, 138). In systems in which singlet excitons play a dominant role inphotoionizations, an extension of this mechanism would propose theexistence of an intermediate S* state that has CT character: the electric fieldshould have a similar effect in this system and, indeed, had already beenrecognized by others (139, 140), who found this field effect on the ratio ofcarrier generation yield to fluorescence yield. Popovic (140) studied X-metal free phthalocyanene powders dispersed in Lexan. He concluded thatthe electric field increased the rate of dissociation of the lowest-lying excitedsinglet state S, into a CT state, which subsequently dissociated thermally,assisted by the external field. His work also provided an example of a case inwhich the same S, state acted as a precursor to CT state formation eventhough the initially excited state was higher in energy.

It has now become clear from the work of Popovic (140). Scher &Orlowski(135. 136). Silinsh and co-workers(141). and Rackovsky & Scher(121a) that the field dependence of photocarrier generation in organicmolecular solids is not capable of being interpreted entirely by thecontinuum Onsager model (116. 117). Depending on the energy ofphotoexcitation, a state is produced in these materials that has a complexcharacter including, for example, singlet-singlet (142), triplet-triplet (143),and a considerable charge-transfer component. The effect of the highexternal field is to shift the character of the intermediate state to one that ismore ionic, and easier to ionize. Following the dissociation of theintermediate state, the usual Onsager formalism applies.

Time Dependent

Monte Carlo computer simulations of physical processes have also played amajor role in the elucidation of the time-dependent geminate recombi-nation. This is particularly important because it is difficult to carry outexperiments in the picosecond time scale. The time dependence of geminate

% -

-". - %-°b°....

ELECTRONIC PROCESSES 633

recombination was simulated by Ries, Sch6nherr. Bassler. and Silver (144)and compared with the analytical solution given for the isotropic con-tinuum model by Hong & Noolandi (HN) (119, 120) and Hong, Noolandi &Street (145). The simulation was not restricted to the isotropic continuumcase, but considered 1, 2. and 3D motion on a discrete lattice of variablesize and graininess. One of the conclusions of HN was that at long times. thesui vival probability of a geminate pair in 3-D follows a t- 3, decay law.That is. according to HN,

R 3D(t > r) r. exp(-rr 0 )/(4rDt3 )1 2

where

rO = r '4D, rD = e2 i4n 0kT

and r, is the Coulomb capture radius. D is the diffusion coefficient, R 3 is thegeminate recombination rate in 3D. The simulation indicates that the t 2

behavior, if realizable at all in an actual experiment, would be observed onlywhen the signal amplitude (as for example. that for recombination "-,luminescence) has decayed to 10- of its peak value. For practicalsituations, the exponent x of the decay function t - is > 3,'2, the value of xwill depend also on the dielectric constant of the medium (through rc) andthe dimensionality of the motion. In anthracene, r,,/ro -t 8 (118), whereas for2-Si. where r = 11.5 (compared to i = 3.2 for anthracene), riro is smallenough so that the HN theory works well in explaining the 2-Siluminescence decay rate (121).

Another startling result of the simulation is that in the first decade of the ' '-"'"decay process (t > T0 ) the simulated decay rate exceeds the HN results ifextrapolated to short time limits. Thus, by using the HN theory torationalize an observed recombination luminescence life time, one wouldhave to use an inordinately large diffusion coefficient. The simulated andanalytic results converge when the number of surviving particles hasdecayed to 10 6 of its peak value. Again, the HN theory predicts that fort > to

N 3 D = exp(- r,'ro) I1 +rc/(47rDt) 1 21

where N3D is the fraction of surviving pairs, whereas the simulation resultsshow a more complex function of time.

Another parameter that was varied in the simulation was the frequencyof the final jump to the recombination center. This is particularly relevantbecause of the accumulating experimental evidence for the existence of along-lived electron-hole pair state in organic crystals (142, 146-152). So farno attempts have been made to rationalize such behavior theoretically.Ordinarily one would expect the final recombination step to be rapid.

~ '~.. -. -.°~

634 POPE & SWENBERG

Although no picosecond time-resolved geminate recombination studieshave been made on anthracene or its homologs, Braun & Scott (152) didcarry out such a study on a hexane solution of anthracene at aconcentration of 4.5 x 1015 cm -. At room temperature, the decay rate ofphotogenerated geminate cation-electron pairs is characterized by a firsthalf-life of no more than 9 ps. The recombination process is stronglynonexponential, as would be predicted by the theory of Hong & Noolandi(120) and the recent work of Ries et al (144). Thus, the second half-life is 28ps and the third half-life is 125 ps. The work in liquids indicates thatgeminate recombination can be rapid. However, there are conditions underwhich the final jump can be slow. To explore the consequences of such asituation, Ries et al (144) generated a "waiting factor" fw for the final jumpsuch that the inverse probability of the final step is t,, = fwto, where o isthe hop time for an isoenergetic or exoenergetic jump. Thus, if the finalrecombination took place at a defect site, there might be an energeticbarrier UB to overcome, to which could be related an fw = exp(U JkT).Carriers trapped in defect sites may exhibit abnormally long dwell times(see this review under CARRIER TRAPPING).

CARRIER TRANSPORT

In Molecular CrystalsCarrier transport in van der Waals molecular crystals, such as anthraceneand naphthalene, has been the subject of intensive study during the lastseveral years. The adequacy of band theory as a proper framework forinterpreting experiments had been questioned as a result of its failure toexplain electron mobility in the c' direction in anthracene and naphthalene.For a review of pertinent experimental data and theoretical models prior to1982 see (1, pp. 337-78). A collection of data on mobilities in organicmolecular crystals was published by Schein & Brown (153). In this review,we present the most recent experimental data and theoretical model.Requiring explanation are (a) the band-to-hopping transition observed(154) in naphthalene at T 7 100 K for electron transport (j-) along the c'crystallographic direction, and (b) the almost complete lack of temperaturedependence in y- in anthracene from 78-479 K (155) and in naphthalenefrom 100-325 K (156).

In contrast to electrons, hole transport (pu") characteristics are inaccordance with band theory predictions (1, p. 349); p+ increases withdecreasing temperature in the temperature region, in which shallowtrapping is not dominant. Because of the weak intermolecular interactionsin van der Waals crystals, the nearest-neighbor transfer energies, which areon the order of 10-2 eV (157, 158), are comparable to the conduction band

'..............................,...... . '.. . . . . . . . . . . . . . . . . . .- . . . . . . ... .-.

j . • .-.--

ELECTRONIC PROCESSES 635

broadening due to dynamic disorder; strong dynamic disorder tends to ,,,localize carriers. A major theoretical problem has been to isolate thedominant terms in the complete transport Hamiltonian, which so far hasbeen intractable. Sumi (159-162), in a series of papers, assumed that p - wasdominated by electron-libron scattering. A recent analysis of mobility datasuggests that a more conventional approach is fully capable of accountingfor the temperature dependence of the electron mobility for T < 100 K.

Specifically Andersen et al (163) have demonstrated that a Boltzmann

equation analysis with longitudinal acoustic phonon-electron scattering W ,can account quantitatively for the electron mobility in all crystallographicdirections, in which case p oc T - 3/2. Figure 2 illustrates how well p c T - 3/2fits the experimental data. The lack of a temperature dependence in l"for T > 100 K is, however, still unexplained. Furthermore, as noted byAndersen et al (163), there is still the difficulty of reconciling the conven-tional band picture with the short mean free path (A) of the carriers in the c'direction at temperatures 40 < T < 100 K (A a).

Although saturation of the electron velocity along the c' direction has notbeen observed (164), field-dependent hole mobilities for F parallel to the a-axis at low temperatures have been reported by Warta & Karl (165).

100oElectron Drift Mobilities ' Naphtha lene

70- T-3 2.."•7 0o _\ * .• "b axis

50.\ a axis

3.

, . Caxis .

o-" '.;,....-.

E

0 7-

"0 5•

V: -. - %

30 50 70 100 140 200 300

Fiqure 2 Measured electron drift mobilities along the a. b, and c directions in naphthalene asa function of temperature. From Anderson, Duke & Kenkre (163).

3.,'

r:;

636 POPE & SWENBERG

Presumably the freezing out of phonon modes at low temperaturesenhances the carrier mean free path sufficiently so that momentum statesnear the edge of the Brillouin zone are significantly occupied.

In Poly-t-electron PolymersInterest in the electrical conductivity of polymers is high and will certainlybecome greater with time. Several conferences have been devoted almostentirely to the subject of these poly-7r-electron polymers (166, 167). A reviewof polyacetylene has been given by Etemad et al (168). Doped poly-acetylenes have conductivities that are metallic (- 1000 0) ' cm- ) andthey have been fashioned into p-n junctions (169). In this review, we discussonly two polymeric systems that consist of long chains of conjugatedmolecules; these are the polyacetylenes and the polydiacetylenes, whichhave been the model systems used for the study of carrier transport.

POLYACETYLENES The polymer polyacetylene has received the greatestattention both theoretically and experimentally due, partly because it is thesimplest conjugated polymer and also because the trans form can support *

mobile solitons. A soliton is an uncharged defect or a kink in the trans-polyacetylene (t-PA) chain consisting of two single bonds adjacent to eachother, instead of the usual single and double bond alternation. This kink, orsoliton, has one unpaired electron associated with it and is free to movealong the t-PA chain. A charged soliton would have no unpaired electrons,

-.- and would not exhibit electron spin. No soliton exists in cis-polyacetylene '.

(1, pp. 701, 780).Structural changes usually associated with the cis- to trans-thermal

isomerization have been observed in polyacetylene by Robin et al (170).who used synchrotron X-radiation (1.34 A); the isomerization process wasfound to proceed homogeneously throughout the polymer and not inisolated amorphous or selected crystallite regions. That doping alsoinduces a cis-to-trans isomerization has been demonstrated by Hoffman etal ( 171 ). The soliton has an EPR Lorentzian line-shape with a linewidth lessthan I Gauss (172, 173); that is suggestive of motional narrowing at least atroom temperature since estimated linewidths from hyperfine interactions

. with neighboring protons is expected to be approximately 23 Gauss (174,175). The observation of an Overhauser effect (176) is consistent with theparamagnetic defect being highly mobile. Furthermore ENDOR studies(177, 178) show a coalescing of the two-line spectrum into a single line onincreasing the temperature. The ENDOR 2-line spectrum has been shownby Heeger & Schrieffer (179) to express the qualitative features of a mobilesoliton. Evidence for solitons, independent of optical studies ( 180), has beenprovided by elastic tunneling conductance measurements ( 181 ). -1 ,,

,'. ), :..-.,. ,, -, '...-..,...",.,'...,,...,.,,..,.,, ,. .:'.'., "-." .,,...............................'........,.-.:..'.....................•..,,.,

ELECTRONIC PROCESSES 637 N%

A direct measurement of the time evolution of a photogenerated 4:electron-hole state into charged kink-antikink states in cis- and trans-PAhas been made by Shank et al (182). They showed that the electron-hole

, state decays in less than 1.5 x 10 -3 s. This result supports the theoreticalcalculation of Su & Schrieffer 0 82a). From the decay curve of the chargedkink state in cis- PA. it appears that the photogenerated state converts into a %

metastable state with a lifetime greater than 10a sec. The long lifetime forrecombination in cis-PA is reminiscent of that found in the polyacenes(146-151). In t-PA, the recombination is complete in -200 ps, in whichnearest neighbor jump times of - 10- 3s at 300 K and 10- 2 sat 20K werecalculated.

Soliton theory has been quite successful in explaining a wide variety ofphysical properties in cis- and trans-PA. Some of the subtle difficulties withthe soliton theory in explaining EPR and NMR measurements and photo-induced changes in optical absorption have recently been summarized byBaeriswyl 1183). We note here only the problem in explaining the transportcharacteristics. The Kivelson theory (184) of phonon-assisted hoppingbetween charged and neutral solitons, although quite successful in makingspecific predictions regarding conductivity as a function of temperature,frequency, and concentration, is not without difficulties. The theory -

requires the presence of both neutral and charged solitons and is valid onlyat low concentrations, in the regime where Y < 0.005. This assumption ishard to reconcile with the observation of spinless conductivity at highdopant levels (at 2"(, dopant, all neutral sol.itons have been converted tocharged solitons). Furthermore, spinless conductivity is not a uniqueproperty of trans-polyacetylene; it has been observed in poly(p-phenylene)(PPP) and for this polymer it is known that solitons cannot exist since thebenzenoid form is significantly more stable than the quinoid form, i.e. PPPdoes not possess a degenerate ground state. In fact, PPP doped with AsF,has been reported to have a conductivity comparable to that of doped-PA(-500 D-' cm at dopant levels of 0.42 mole of AsF 5 per mole ofmonomer (185)). Thus, a theory of spinless conductivity that is based solelyon charged solitons, even for trans-(CH)x is considered highly artificial.

A more promising model for spinless conductivity is the polaron-bipolaron theory of Bredas. Chance & Silbey (186, 187). This model doesnot exclude solitons: in factor neutral solitons are the main extrinsic defectsin undoped trans-(CH),. In (CH), the polaron is viewed as a pair consistingof a free radical (neutral soliton) and an ion (charged soliton) with anestimated binding energy of 0.05 eV, a value quite comparable to the pairbinding energy of 0.03 eV calculated for PPP. An antisoliton is the matingkink to a soliton. An antisoliton will annihilate a soliton. producing adefect-free chain. An antipolaron is a combination of a charged soliton and

3. A .'.%

r'. 7. - 'o -.. • h , ,./ - . . - 1 - - t j1 -. G -. .- '!' t

,. -,, " .P " " c. . J..' "[o%,' '"d -,."d". " -r .l W r . . . M- a,.. sr ". 'r2 r- '--7P, . C 1 7% 71- Y Y, -

638 POPE & SWENBERCI

a neutral antisoliton. The evolution of states within the gap as a function ofdopant concentration can be summarized as follows: Low doping intro-duces both bonding and antibonding localized polaron states within thegap. At higher doping levels the localized polaron states coalesce to formtwo polaron bands. In PA the interaction of a polaron with an antipolaron %of the same charge results in two free charged solitons. This process occursonly for PA since the neutral soliton-antisoliton pair annihilates and the .%..rdegenerate ground state allows the charged solitons to move apart. In PPPand other conducting polymers that lack a degenerate ground state, the twocharged defects cannot separate and bipolaron formation ensues.Bipolarons are correlated charged soliton-charged antisoliton pairs. Ateven higher dopant concentrations bipolaron bands are formed. Evidenceto support the existence of bipolarons and a polaron band in doped-PPPhas been provided by electron energy loss spectroscopy (188).

Within the bipolaron model of transport, the neutral solitons that arerequired on neighboring chains are produced by the formation on aneighboring polymer chain of a virtual neutral soliton-antisoliton pair.After the charged portion of the bipolaron jumps to the virtual soliton-antisoliton pair on a neighboring chain, the residual soliton-antisolitonpair annihilates. This mechanism requires only a small activation energy, isapplicable not only to trans4CH), but also to other conducting polymers,and does not eliminate the phonon-assisted interchain hopping mechanismof Kivelson in trans-(CH), in dopant regimes where sufficient neutralsolitons are available.

First principle calculations have shown that as in PPP, polaron statesform within the band gap in polypyrrole at low doping levels (189). Opticalabsorption due to polarons in polyacetylene and in other polymers(polypyrroles, poly-paraphenylenes) has been considered theoretically byBredas et al (186) and Fesser, Bishop & Campbell (190). Experimentalsupport for bonding and antibonding polaron states in the low dopant limitin t-PA is reported by Etemad et al (191). It appears that in t-PA bothpolarons and solitons coexist at low doping concentrations and are thecurrent carriers, whereas in polymers lacking a degenerate ground state andthereby precluding soliton formation, polarons are the exclusive carriers.

POLYDIACETYLENE The unusual features of carrier transport in PDA-TSare the apparent saturation of the drift velocity Vd at -- 2 x 10' cm sdown to fields as low as I V cm -' [giving a lower bound for M z 2 xl0 cm 2 V ' s '(131)] and the photocurrent decay rate following a shortlight flash" this decaying photocurrent I(t) obeys a relationship of the formt - where x - I over 6 decades in time (10 -' s to I s) (144). This decaylaw is often found in amorphous materiais, whereas in PDA-TS one would

r-

ii ...... . ...... . ..... .....- ?

•. .. . .. .... . .. .'-: .... . - . . L . . . .'''% '''' '' '' " "'''' " :

ELECTRONIC PROCESSES 639 'V..'..

expect very few scattering or recombination centers. Additional experi-ments by the authors in the time domain for which 1(t) oc I shows that Vdis still field independent and 2 - 0.85.

P According to Seiferheld, Ries & BAssler (123), the anomalously high

carrier mobility (pu : 101 cm2 V - s - ') deduced for PDA-TS (131) is

inconsistent with the strong scattering required to explain their successful

*7 use of the Onsager 1D carrier recombination formalism. The paper by ,-'..'

Movaghar et al (133) resolves these difficulties and shows that there need be

no qualitative conflict between the results of Donovan & Wilson (131) and

those of Seiferheld et al (123).The explanation for all of the observed results lies in the breakdown of

linear response theory as applied to transport problems in a broad class of

I D systems (192). A model is used that is suitable for materials like PDA-

TS, with its low concentration of crystal defects. Using the ID diffusion

theory of Alexander et al (193) one may deduce a time (t) dependent current

j(t) relationship of the form

j(t) - ea t -( '2 -, t - x. 6.

Here ) is the applied reduced electric field and need not be defined here

other than to state that for small fields, 2q/= eFa/kT, where a is the lattice

spacing. However, as explained by Movaghar et al (192), linear-response

theory breaks down in the long time domain and therefore Eq. 6 is wrong.

The correct form for j(t) at long times isfit) "-2eaml Wot' "C(2),]I 2Pqt'l 7. . .'."

where C(x) is constant and Wo is the hopping rate in a defect free ID chain.

Sincej(t) is proportional to the carrier drift velocity c, then according to

Eq. 7 and the definition of r, vd will vary sublinearly with the electric field.

Furthermore. since x- - 1, it follows that the drift velocity will be almost .. ,

independent at F. ., C

The authors make the important point that behavior of the type

described by Eq. 7 should appear in a wide class of ID transport problems.

Thus, for PDA-TS. the authors derived, for t> t,

vIt. P1) - (2xP1C(2, x),x)Wo' 1 2r1t 8.

where x is the fraction of jumps that are difficult and tc is the time required

for the carrier to drift to a defect; t - (2qxWo)- '. Equation 8 applies to the

PDA-TS experiments where 2 = 0.85, and t'd x F0 '5 . which looks like a

saturated drift velocity. In addition, the photocurrent will decay as t - until

carrier recombination (or collection at the electrode) takes place. It can be

also shown that a lower bound for the mobility along the regular portion of

I

C .'p

...

t .)l,

640 POPE & SWENBFRG

the chain will be uf > 10a, 2xF. If Vd remains saturated at 2 x 10' cm s- ,down to F = IV cm -. as stated by Donovan & Wilson (131). then pf >

2 x l0 cm V I s 1. Thus. with a = 1.2 A x could be - 10 or fewer thanI defect per 10' A: this was also stated by Donovan & Wilson (131). Using ,.-.

the data of Seiferheld et al (123). who only went down to F = 50 V cm -'.

and taking 1-d = 2 x IO cm s ', one gets If _ l0 cm 2 V s andx - 10 which is still a high mobility. Another interesting calculationis the time required to cross half the sample length, or Lo!2. This turns outto be

to = (2t1Wo) -'(Lox/2a)' 't-,' 9.

Withr,= 2 x 10cms ',F= I Vcm-', L =0.3cm, x=0.85. a - 5and x = 10-5, then to - 10-2 s. For x = 10- , one obtains to - 10 s.Thus, to is extraordinarily sensitive to x and o. It is possible to increase x bycreating radiation-induced defects; this has been done and very longrelaxation times have indeed been observed (194).

In Amorphous Systems

As mentioned above, the introduction of computer modeling as a quasi-experimental technique is having a profound effect on the course of bothexperimental and theoretical studies. Consider the study of carriertransport in amorphous systems, of which polymeric conductors are aprime example. In amorphous systems, carrier transport clearly is byhopping. It has been shown by Bissler (195) and Sch6nherr et al (196) thatby assuming a Gaussian distribution of hopping site energies, a wavefunction overlap factor exp(-2.7r), where r is the distance betweenneighboring sites, and a Boltzmann factor to accommodate hops toenergetically unfavorable sites. one obtains the following: a temperaturedependent trap-free mobility of the form u(T) = u , exp -(To, T)2 } whereA is the ideal mobility in the absence of disorder, and a field dependentmobility of the form u(F) = u(O) exp(FrFo). The field dependence is ofinterest because it shows that no charged traps need be invoked to explain afield dependent mobility (1, p. 78); such a field dependence had already beenobserved, and charged traps had been conjectured (197). In a series ofexperiments (198) carried out with hole-transporting solid solution of aphenylmethane derivative in a plastic matrix, complete agreement wasfound between experiment and simulation, without invoking charged traps.As for the temperature dependence of the mobility in amorphous systems.this was studied experimentally by Lange & Bdssler (199), who usedamorphous tetracene films. The use of amorphous films of an otherwisecrystalline material permits a study of the effects of disorder withoutchanging chemical composition. Tetracene films deposited on substrate

.- .

ELECTRONIC PROCESSES 641

held at 120-180 K are amorphous. A trap-free mobility obeys a relationshipof the formu(T) = ,. exp 1-(ToiT) 2},whereju . 0.4 to0.Scm2 V s-validating the Monte Carlo computer simulation. The Gaussian widthgiven by [o is .z 0.1 eV, and is a function of sample preparation temperature OltT, and hence of the degree of disorder. Important results of this paper arethat the Gaussian width is a consequence of disorder-induced splitting ofthe transport band into a distribution of localized site energies: this effect islarger than that produced by the dielectric relaxation of the environmentaround a carrier. The success of the computer simulation strengthenssupport in its assumptions, and provides an alternate explanation for theresults reported by Mort & Pfister (200).

Another interesting result was the observation of an almost temperatureindependent trapping time. although p varied by more than an order ofmagnitude. Their explanation is based on the notion of trapping at anincipient dimer site. The density-of-states' profile in tetracene as deduced in _ ...these experiments exhibit a 0.1 eV width; this is in contrast with widths of0.5 to I eV found in photoelectron spectra of molecular glasses (6). Anexplanation may be found in the differences between the processes ofphotoemission (which depends not only on the inhomogeneous broadening .

of the valence states, but lifetime broadening of the final state) andtransport. In addition, transport probes the volume density of statedistribution, whereas photoemission is essentially a surface phenomenonand the polarization energy changes markedly in this region (201. 202;see also 1, p. 63).

Another example of the power of a computer simulation is that carriedout by Silver et al (203), who showed that while a Gaussian distribution ofhopping site energies gives little dispersion, an exponential energy distri-bution gives rise to dispersive transport indistinguishable from multipletrapping. This introduces another mechanism for the interpretation ofexperimental results.

CARRIER TRAPPING

Experimental TechniquesA general technique for the determination of the concentration, energeticdistribution, and composition of carrier trapping sites in organic solidsdoes not yet exist. It remains as a challenge, and the rewards for solving theproblem are great. since relatively small concentrations of traps radicallymodify the transport properties of the carriers. Present techniques includespace charge limited currents (SCLC), thermally stimulated currents (TSC),optical detrapping, and electric field detrapping (1, pp. 267 et seq).

An important comparison of the results obtained on trapping states by

.........................................................................

'-. ~~. '

642 POPE & SWENBERG

SCLC and TSC was made by Taure et al (204). They studied thin films oftetracene (Tc and pentacene (Pc) in the form of oriented crystallites. TheSCLC results were consistent with the existence of at least two sets ofshallow traps of Gaussian distribution: In Tc the binding energy of one was pE, = 0. 1 eV. with a distribution parameter ranging from a = 0.08 to 0.14eVin different samples, and a total trap density N, = z 4 x 1015 cm- 3 . forthe other, E, -0.3 eV, a = 0.05 to 0.1 eV and N, - 10" cm 3.The TSCresults were in fair agreement with those of SCLC as shown in Figure 3, butgave higher resolution and probed deeper traps. Trapping cross-sectionsfor the hole traps in Pc and Tc range from 10- 16 to I0- 14 cm2 for theenergy range 0.2 to 0.5 eV, and are about 1014 cm 2 for deep electron traps inPc.

Some recent work involving the use of thermally stimulated currents(TSC) is that of Samoc et al (205), which also contains many references toimportant work in this field. This paper tests a recent theory of Plans et al 1-(206) that deals with the field and sample thickness dependence of the peakat T,,. the temperature of the maximum that appears in the TSC glow curve.Samoc et al (205) added phenothiazine in a concentration of 5 x 10' molfraction to anthracene. Using their technique, they found that pheno-thiazine created a hole trap centered at 0.62 eV above the valence band andthat there was sufficient distribution of energy around this value to inducedispersive transport in the doped material. This trap depth may be

081

• ---- -------.,>. 04-

02 10 1

• 101, 1014 1015 1016

N(E cm- 3 eV -1)'-

Fiqure 3 Density of electron trapping states N(E) in pentacene layers as a function of trapdepth E as determined by SCLS (solid curves) and TSC (dashed curves) techniques. From Taureet al 1204).

%, °-A.

.j..>:, -

ELECTRONIC PROCESSES 643

compared with the value of 0.57 eV deduced by Karl et al (207) fromphotoemission studies.

An experimental technique for studying multiple deep trapping levelsusing only a single thermal scan was developed by Yoshie & Kamihara

(208), and several intuitively appealing theoretical treatments of multiple "4. Vltrapping have appeared (209, 210). In these papers, an exponentialdistribution of traps with relatively fixed parameters was assumed. In arecent paper, Monroe & Kastner (211) relax these restrictions by allowingvariation in capture cross-section, release times, trap distribution, andenergy. The final result is a classification scheme in which the currenttransients may fall into any of five basic types; this scheme may be used as adiagnostic tool to identify the relevant trapping parameters operative in theexperimental situation. The analytical results of Monroe & Kastner canalso serve as a check on the results of increasing popular computersimulations. Mention should also be made of another comprehensive j.treatment of multiple trapping in amorphous systems by Arkhipov et al(212).

An interesting aspect of electron trapping was exposed by Meyer et al(63). They found that the electron trapping lifetime for transport in the c'direction over a range of temperatures (81-374 K) and applied electric fields(10' to 4.7 x 10' V cm ')was activated, and that this activation energy isindependent of the field. The activation energies, Ea, ranged from 30 to 84meV on crystals all prepared from the same sample, indicating that thebehavior is extrinsic. The electron transport in the c' direction was notshallow-trap controlled; it was nearly independent of temperature. Theauthors conclude that the most likely source of the activation energy wasthe cross-section of capture. A parallel experiment was carried out by -.

Arnold & Hassan (213), who measured the effect of pressure on the tripletexciton lifetime in anthracene. They found that the cross-section of capturewas increased markedly (by a factor of four, in going from I atm to 6.4 kbar),without changing the trap concentration. Arnold & Hassan concluded thatspecific mechanical defects, referred to as preexcimer sites, were the sourceof these traps, and that triplet excimers were formed at these sites. Theyattributed the pressure effect to the creation of a more favorable inter-molecular orientation. The role of predimer or incipient dimer sites as

trapping centers was discussed by Pope & Kallmann (214; 1, pp. 283 et seq)and the existence of such sites was supported by the work of Thomas et al(215) and was also discussed by Lange and Bdssler (199) and, at great length,by Williams & Thomas (216; 1, pp. 50-53). More recently, Zboinski (217)has examined the possibility that a deep trap can be created by thelocalization of a carrier at a pair of approximately parallel anthracenemolecules.

- . '., -.

I-0

. . . .. .. . . .. ..

7T 7 --.7-1717L

644 POPE & SWENBERG

A novel explanation of what can account for apparently deep (-0.7 eV)traps in pure molecular crystals has been presented by Petelenz (218). Inpure crystals, one can account for shallow ( -0.3 eV) traps by the increase inpolarization energy surrounding a defect in the crystal lattice. However,one cannot use lattice distortions to explain a trap of depth -0.7 eV.Petelenz points out that the detrapping of a trapped electron involves notonly supplying the binding energy E, of the trap, but the electron musttransfer itself away from the trap in a time short compared to thevibrational relaxation time of the excess electron. This requirement is thesame as that for autoionization of a neutral state. In a sample calculation,Petelenz showed that detrapping could be accomplished by excitingvibrational degrees of freedom, rather than the carrier itself. He also showedthat for a trap only 10- eV deep by virtue of lattice distortion, the apparentdetrapping energy could be 2!0.75 eV due to the requirement of excitingeffective vibrational modes.

Field EffectsOne of the many differences between 3D systems and those of lesserdimensionality (1, p. 617) is the electric field-dependent reduction in the -charge carrier trapping time in the I and 2D systems. In a 3D system,trapping is generally a first-order process with a time-independent rateconstant. This is a consequence of the trapping probability increasing :";. ' |

linearly with the number of new sites visited. The number of new sitesvisited by a randomly diffusing particle varies as n in 3D, n/In (n) in 2D, andn' /2 in ID, where n is the number of steps taken (1, p. 122). If the hopping.:rate is constant, then the trapping rate will become time dependent in 2Dand ID motion. Movaghar et al (219, 220) showed analytically that in aID system, at long times, the relaxation of an excitation follows anexp[-(t/) 13] law. This law was shown to apply to ID trapping kineticsby Hunt et al (221) in a disordered polydiacetylene polymer, which shall bereferred to as PDA-10H; this polymer contains a pendant CH2OH groupin every repeat unit of four carbon lengths. This paper reports the firstmeasurement of a ID relaxation law. The specific phenomenon is the decayof a transient photocurrent pulse in the PDA-IOH in the time range I to 2x 10" s. Although the more widely studied PTS has an exceedingly low

dislocation density (222), the concentration of deep traps is - I per mm ofpolymer chain (131) in PDA-10H; due to the presence of extensivehydrogen bonding between the CH 2OH groups, there is a closer packingand greater internal strain in the final crystal, which is not as perfect as thePTS single crystals. A sample was excited by a square wave pulse of lightand it was found that the decay follows an exp(-bt) law.

i- ,%-- '.i

• • ~,... °- .\-

-5,

. . . . ... . . . .

- . ..... .

ELECTRONIC PROCESSES 045

In the presence of an electric field, the motion of the particle will be .* ,

anisotropic and thus, in a ID system, more new sites are likely to be visitedthan would otherwise be encountered. This implies that the trapping ratewill increase with field strength. This effect was first observed by Haarer &M6hwald (223; 1, p. 616), in the charge-transfer complex phenanthrene-pyromellitic acid dianhydride. An exact solution of the time dependence of .

the trapping probability in the presence of an electric field F was derived by '%"'

B. Movaghar, B. Pohlmann, and D. Wuirtz (MPW) (unpublished). Theyshowed that as t -- oc, a simple exponential law should be approached, thefree carrier concentration decay time going as F 2 below a critical field F.,and as F above F. This theory was tested by Seiferheld et al (194), whoused PTS; this polymer is an excellent model for a I D crystal because its .... *

mobility anisotropy ratio/ I, /p, is - 10s, where/ I, is the mobility along thechain and pu is the mobility transverse to the chain direction. The PTS .' -_polymer is also unusual because it contains essentially no recombinationcenters for carriers (131), so the decay of a transient carrier population isgoverned by the kinetics of hopping and detrapping from traps of - 0.7 eVand discharge at the electrodes. In this experiment, traps and recombi-nation centers were produced in a PTS crystal by bombardment with100 KeV He + ions. It was found that at long times, and for fields of < 1.7 x10' V cm - ', the current relaxation went as exp[-(t/t)" 3 ]. As thefield increased, the long time relaxation curve became steeper, approach-ing a simple exponential, exp[-(t/r 2)]. Moreover, r2 varied as Fbelow Fr % 10' V cm -' and as F-' above F. The effective carier jumprate W can be calculated from rt1 , giving -1 s - 1; this implies that the trapdepth is , 0.8 eV.

SUPERCONDUCTIVITY

The dramatic discovery of superconductivity in organic crystals was madeby Bechgaard et al (225) and Jerome et al (226) in a family of isostructuralcompounds of the general formula (TMTSF)2X (also referred to asBechgaard salts) where X is an anion (C10, TaF6, AsF 6 , SbF6, andReOn) and the cation is tetramethyltetraseleno-fulvalene. Only the C10 4

derivative exhibits superconductivity at atmospheric pressure; this occursat a critical temperature T : I K. The other compounds all require theapplication of external pressure (8-12 kbar). Reviews of this field bave beenwritten by Jerome & Schulz (227) and by Friedel & Jerome (228). Inaddition, the proceedings of two conferences are particularly useful (229).

In the search for a general mechanism for superconductivity in organicmaterials, attention has been focused on the unit cell volume V,, and on the

%. .-

1.. A.'

~~~~... ...... ,........-............... . ,•.,...*., ..'.1.'. .-. "-7..

646 POPE & SWENBERG

interstack Se-Se distances (230). As has been found to be the case with thehighly conducting ion-radical salts of the TTF-TCNQ class (1, pp. 581-91),the nearly planar, almost parallel TMTSF molecules are arranged in stacksalong the a crystal axis (1, p. 638). In addition, the TMTSF molecules inneighboring stacks are sufficiently close to each other along the b direction Iso that infinite sheets are formed in the ab plane, separated by columns of

anions. The short interstack, and intrastack Se-Se distances (d < 4.0 A; LWthis is less than the van der Waals radius sum for Se-Se) lead to stronginteractions that result in the high conductivity in these materials. As thetemperature is lowered from 298 to 125 K, anisotropic structural changestake place in which the interstack distances can decrease almost twice asmuch as the intrastack distances (231). A linear correlation was foundbetween the unit cell volume, V, and the average interstack Se-Se distance(dae) ; this is shown in Figure 4. The correlation shows C10 4 , FSO3, andBF, salts clustering around the minimum values for V and d.,, , withalmost identical values for d.,,. This suggests that these compounds will 'have a similar Se atom network geometry and similar low temperatureelectrical properties barring the onset of anion ordering (232); suchordering introduces a new crystal symmetry, and is considered to be aprerequisite for the attainment of superconductivity in (TMTSF)2 CIO1(233), in direct opposition to previous beliefs. In addition, those compoundsrequiring external pressure to achieve superconductivity have d,, valuesgreater than d.,,(CIO). One may therefore venture to predict the anionsize most likely to produce the desired VP ( V = predicted cell volume V).These results suggest that good results should be obtainable with thoseanions whose sizes are comparable to C0; these include not only BFT

70AsF

o 690

EZ. PF60A >FSaD ReO4>60 FS03A ~0

U C10 4670

3.69 3 71 3.73 3 75 3,77 3 79 3 81 3 83

Average Interstack Se-Se Distance, A-

Figure 4 Plot of observed unit cell volume (V) vs the average interstack Se-Se distance forvarious (TMTSF)2X metals at 125 K. From Williams et al (231).

.°V

ELECTRONIC PROCESSES 647

and FSO, but anion alloys embodying all possible combinations of thesetwo anions.

Most organic molecules in crystals that exhibit metallic conductivity areplanar, of D2h symmetry, and pack in parallel layers in a stack, along which

A the conductivity is a maximum (1, pp. 588 et seq). In this quasi- ID crystal, itwould therefore be expected that there would be intrastack bonding in thisdirection; this expectation is supported by the observation that in theorganic metal TTF-TCNQ, the TCNQ-TCNQ distances are ; 3.17 A as ,.compared with 3.45 A in the pure TCNQ crystal. In the case of thesuperconductors in the (TMTSF)2 X family, it has been concluded that Se-Se bonding within a stack and bvtween stacks creates a quasi-2D structure,which is chiefly responsible for the high conductivity. This conclusion wastested by making X-ray diffraction studies on perfect crystals of(TMTSF)2 AsF6 prepared by WudI (234). An accumulation of electrondensity was found above and below the TMTSF molecular plane in theregion between the Se atoms, implying that there is bonding between theTMTSF molecules within the stack. probably as a result of the Se atominteractions. In addition, there was considerable electron density betweenSe atoms at the edge of adjacent stacks in the stacking direction. This couldbe evidence of a one-dimensional conduction band. In view of thesubstantial conductivity observed in the direction (b axis) perpendicular tothe stacking axis (a) of the TMTSF molecules, electron density was lookedfor between Se atoms in neighboring stacks. An electron density was foundbetween the Se atoms at the edge of adjacent stacks, but surprisingly, onlybetween Se atoms with the longest interstack distance, -4.15 A. This waspredicted by Grant (235). The continuum of electron density from Se to Sebetween the stacks may represent a conduction band, Thus, there exists abonding originating from the Se atoms, between molecules within andbetween the stacks of (TMTSF)2AsF 6 and (TMTSF)2 PF 6 .

In the search for other superconducting compounds a new, and only thesecond, family of organic conductors was discovered (235a) based on thesubstitution of S for Se in the cation. The newly discovered compound iscalled (BEDT-TTF)4 (ReO) 2 where the cation is bis(ethylenedithiolo)-tetrathiafulvalene, and the anion is the perrhenate ion. This compoundbecomes superconductive above 4 kbar for T 2 K. A metal-insulatortransition takes place at a pressure < 7 kbar, and it may be associated withan anion rearrangement. These BEDT-salts have a variety of crystalstructures and stoichiometries (236), in distinction to the Bechgaardsalts, so it may be easier to locate the origin of superconductivity in thesematerials. These discoveries are particularly valuable because the mech-anism for organic superconductivity is still poorly understood. Onearduously sought for goal is to find a compound that has a high T.

,:..-.

,. -. ,--.---.--.. .,...- .... .- .,***,,.'.'... . . . . .. . . . . . . . .... '... ..... .. ... "- .'.. . .,... .-' - *. ?-", .* .,"**', *-. *' "".

648 POPE & SWENBERG

MISCELLANEOUS

In this section, we would like to mention work that cannot be adequatelydiscussed due to space limitations.

An example of the power of picosecond spectroscopy to unravelmechanisms of chemical and biological reactions is given in a review articleby Rentzepis (237). In the wings stands the even more potent tool offemtosecond spectroscopy, as outlined by Shank & Greene (238). Opticalpulses as short as 30 fs have been attained, which is shorter than somevibrational periods. One can anticipate the use of such pulses to follow the .-

evolution of energy transfer in a coherent fashion between two degenerate .'.:states. Another impressive experiment showing the fission of a singly ......

excited state into multiple excitonic states from which quantum yields andautoionization efficiencies are obtained has been carried out by Klein (239).

A discussion of electroluminescence in organic crystals is given byKalinowski (240); the effect of pressure and temperature on the lumines-cence of tetracene single crystals was also studied by Kalinowski et al (241).In this latter study, a surprising feature of earlier work on tetracene wasreexamined; this was the observation of an unusually large Stokes red shift(-500 cm- 1) between the 0-0 fluorescence and absorption transitions,whereas in anthracene it is 100-200 cm '.It now appears that this shift isan artifact caused by the compensation of the red shift resulting from thetemperature modification of the exciton state and a blue shift caused bydecreased overlap of the fluorescence and absorption spectrum. The Stokesshift of the reabsorption free 0-0 transition at 528 nm is 260 cm - t, in goodagreement with the value 280 cm found at 4.2 K (242).

A recent review of the important subject of energy transfer has been givenby Ki6pffer (243). This review discusses basic concepts, measuring tech-niques, and results, mainly in polymeric systems. In the same book is anexcellent discussion of triboelectricity in organic materials (244). Thissubject has not yet been put on a sound theoretical footing because of theenormous experimental difficulties in creating reproducible contactingsurfaces. Important insights have been provided by Duke (245) and Duke &Fabish (246). Triboelectricity and triboluminescence may one day bestudied in a more straightforward manner in outer space. Tribolumi-nescence is thought to arise from the creation and annihilation of mobilecracks (247, 248), which in piezoelectric crystals can result in the crea-tion of oppositely charged neighboring surfaces. The intense electric fieldat the tips of the cracks may facilitate charge recombination. In Mort &Pfister's book, a review of piezoelectricity and pyroelectricity by Wada(249) brings the field up to date from the last reviews prepared by Kepler1250) and Davies (251). ,

* -. *.=.....*]

ELECTRONIC PROCESSES 649

In the first of a series of photoconductivity experiments on PDA-TScrystals, Donovan & Wilson (131) found that the low field mobility washigh, p> 2 x 10 cm 2 V -' s - and that the drift velocity saturated at a lowvalue v. = 2 x 10 cm s even for fields down to 1 V cm-'. On the otherhand. Spannring & Bissler (252) measured SCLC in PDA-DCH usingohmic electrodes; they found that J o V2 for F < 26 V cm-, Vd o F,andA = 6 x 103 cm 2 V- I s- 1 .This large discrepancy has been removed bythe development of a SCLC theory for ID materials. The conclusions are asfollows:

1. A trap-limited SCLC shows a J oc V2 dependence even if v, is saturated.2. If F increases, then for some critical value F, if F > F, the SCLC will

become trap-free instead of trap-limited if vd is saturated.3. A trap-free SCLC in a ID material shows a J oc V dependence if v, is

saturated.

With these findings, the discrepancies between the Donovan & Wilsonresults and those of Spannring & Bassler can be reconciled. The properlyinterpreted data of Spannring & BAssler yield a calculated p for PDA-DCHof 1.6 x 10 cm 2 V' s- , very similar to that in PDA-TS.

CONCLUSIONS

This has been an active period for this field and promises to become evenmore so. The enormous skill of the organic chemist is being harnessed forthe creation of a cornucopia of compounds with novel electronic properties.The discovery of superconductivity in more than one type of ion-radicalsalt greatly increases the prospects of determining the mechanism(s) ofsuperconductivity, and hence of synthesizing compounds of higher trans-." \ition temperature. The synthesis of compounds that behave as quasi 1- and2D materials has provided a field day for theorists who can now find exactsolutions to transport problems. The use of computer simulations has - -

assumed major proportions,' and is already dominating fields such asamorphous solids, in which transport takes place by hopping. Thecontinued development of ultra-short laser pulses of precisely definedwavelength has made possible the excitation of specific vibrational modes,and the study of their intrinsic relaxation rates; homogeneous linewidthsare being measured and the mechanisms of line broadening elaborated.

Carrier generation in the homomolecular polyacenes has become much

'Journals devoted to a discussion of simulation techniques include Mathematics andComputers in Simulation, published by North Holland, and Simulation, published by Society ofComputer Simulation.

. . •-. ~ - -. -. . - -. -. ~ -. ~ ~- ,°c.i. °

'• A-: :

650 POPE & SWENBERG

better understood with the recognition of the importance of direct optical -excitation of charge-separated (CT) states, and of precursors to CT states. Itwas certainly satisfying to view the evidence that at least at lowtemperatures, all carrier transport processes in anthracene and un-doubtedly in essentially all of the polyacenes, is understandable in terms ofa band theory of mobility. There is still the problem of the almost zerotemperature dependence of electron mobility in the c' direction in Nanthracene, but this should give way before the next review of this field. Thedevelopment of a generalized master equation (GME) approach to thestudy of exciton transport has revealed instances in which significant errorshave been made in interpreting experimental data. The study of carrierrecombination was enlivened by the discovery of novel high field effects thatpoint to the existence of a field sensitive process for producing CT states,and by indications that the evaluation of the thermalization distance andthe initial ionization yield from Onsager theory is a more delicate operationthan previously thought. This latter conclusion followed upon the develop-ment of analytical and computer simulation techniques for following therecombination of geminate carriers on a discrete lattice.

The surface has not been scratched in the study of electronic processes inorganic solids.

ACKNOWLEDGMENTS . -

One of us (M. P.) wishes to acknowledge support of the Department ofEnergy. We have benefitted from correspondence with H. Bissler, C. B.Duke, A. J. Epstein, and W. Siebrand. We express our appreciation to Ms.M. Menzel and Ms. A. Lunsford for their cheerful and indefatigable typingeffort.

Literature Cited

I . Pope. M., Swenberg, C. E. 1982. 5. Duke. C. B. 1982. FestkorperproblemeElectronic Processes in Organic 22:21-34Crystals. New York: Clarendon Press, 6. Duke. C. B. 1982. In Extended LinearOxford. 821 pp. Chain Compounds, ed. J. S. Miller.

2. Mort, J.. Pfister, G., eds. 1982. 2:59-125. New York: PlenumElectronic Properties of Polymers. New 7. Ford, W. K., Duke, C. B. 1983. See Ref.York: Wiley 4,93:327-54 ,- -- -

3. Haddon. R. C., Kaplan. M.L.,WudI, F. 8. Duke, C. B. 1983. In Electronic1982. In Kirk-Othmer Encyclopedia of Excitations and Interaction. ProcessesChemical Technology, 20:674-98. New in Organic Molecular Aggregates, ed.York: Wiley. 3rd ed. P Reineker. H. Haken. H C. Wolf.

4. Adler. G.. ed. 1983. Proc. 6th Int. Conf. 49:14-19. Berlin Heidelberg New York: !' -on Chem. of Organic Solid State. Mol. Springer-VerlagCryst. Liquid Cryst., Vols. 93. 96 9. Sloan, G. J.. McGhie, A. R. 1984. Melt

%A . .

% ' ."' ., .

'.3f

ip

,: '. ..'...... ".:.. ....-. '...,.........-.. . . ..". ......... .".-..... .-...... ... .... . .....-.. ...., .. ..- .. . ..., .- .,,....,--..,

ELECTRONIC PROCESSES 651 4u

Crvstalliz-ation Techniques. New York: 28. Deleted in proofWiley 29. Blumen. A.. Silbey. R. 1979. J. Chem.

10. Karl. N. 1980. Crystals: Growth Ph vs. 70:3707-14Properties and Applications, pp. I-100. 30. Loring, R. F., Fayer. M. D. 1982. Chem.Berlin: Springer-Verlag Ph vs. 70: 139-47

11. Smalley. R. E. 1983. Ann. Rev. Phys. 31. Gentry, S. T.. Kopelman. R. 1982.Chem. .14:129-53 Chem. Phys. Lett. 93:264-66

12. Fayer. M. D. 1982. Ann. Rev. Phys. 32. Parson. R. P.. Kopelman. R. 1982.Chem. 33:63-87 Chem. Phys. Lett. 87: 528-32:.-.

13. Drickamer. H. G. 1982. Ann. Rev. PA vs. 33. Gentry, S. T.. Kopelman, R. 3983. J. .-

Chem. 33:25-47 Chem. Phys. 78:373-8214. Berry, D. E., Tompkins, R. C.. Williams, 34. Monberg. E. M., Kopelman. R. 1980.

F. 1982. J. Chem. Phys. 76:3362-70 Mol. Cryst. Liquid Cryst. 57:271-31215. Callis. P. R. 3983. Ann. Rev. Phys. Chem. 35. Brown. R., Lemaistre. J. P., Megel. J..

34:329-57 Pee. P.. Dupuy, F.. Kottis, P. 1982. J..-15a. Hanson. D. M. 1983. In Molecular Chem. Phys. 76:5719-26

Electronic Devices, ed. F. L. Carter. pp. 36. Gentry, S. T.. Kopelman. R. 1983. Phys.89-1l1. New York: Dekker Rev. B 27:2579-82 SydI15b. Hanson. D. M., Patel, J. S.. Winkler, 1. 37. Prasad, P. M., Morgan. J. R., El-aydC., Morrobel-Sosa, N. 1983. In Spec- M. A. 198 1. J. Phys. Chem. 85: 3569-71troscopy and Excitation Dynamics of 38. Morgan, J. R., EI-Sayed. M. A. 1983. J.Condensed Molecular Systems. ed. V. M. Phys. Chem. 87: 383-8 5Agranovich, R. M. Hochstrasser, pp. 39. Morgan, J. R., EI-Sayed, M. A. 1983. J.621-79. Amsterdam: North Holland Phys. Chem. 87:2178-85

16. Johnson, C. K., Small, G. 1. 1982. In 40. Hatlee. M. D.. Kozak. J. J. 1981. Phys.-Excited States. ed. E. C. Lim, 6:97-216. Rev. B823:1713-18New York: Academic 41. Arnold, S.. Alfano. R. R., Pope. M.. Yo,

17. Turlet, .1. M., Kottis. P.. Philpott. M. R. H. P.. Selsby. R.. Tharrats. J., Swenberg,1983. Adv. Chem. Phys. 54: 303-468 C. E. 1976. J. Chem. Phys. 64: 5104-14 Z

18. Agranovich. V. M.. Hochstrasser. R. M. 42. Klymko. P. W., Kopelman. R. 1982. J.1983. Spectroscopy and Excitation Ph vs. C'hem. 86: 3686-88Dynamics of Condensed Molecular 43. Blumen. A.. Klafter. J.. Siibey. R. 3980.Systems. Amsterdam: North Holland J. Chem. Ph vs. 72: 5320-32

19. Agranovich, V. M. 3982. Electronic 44. Klymko, P. W.. Kopelman. R. 1983. J.-- Excitation and Energy Transfer. Phys. Chem. 87:4565-67

Amsterdam: North Holland 45. Alexander. S., Orbach. R. 1982. J. Phys.20. Knox. R. S. 1983. In Collective Lett. 43: L625--L631

Excitations in Solids, ed. B. DiBartolo. 46. DeGennes, P. G. 1983. C.R. Acad. Sci.pp. 183-245. New York: Plenum Press Paris 296(2):881-85

20a. Aaviskoo. J., Liidja. G., Saari, P. 1982. 47. Schilling, R., Mattis, D. C. 1982. Phys.Ph vs. Status Solidi B 110: 69-73 Rev. Lett. 49: 808-11

20b. Saari, P.. Rebane, K. 1981. J. Phys. 48. Schilling, R., Mattis, D. C. 1983. Phys.Colloq. 42(C6): 502-4 Rev. B 27:3318-23

23. Powell. R. C.. Soos, Z. G. 1975. J. 49. Thomas, G. A.. Rice, T. M. 1977. SolidLumin. 11: 1-45 State Commun. 23: 359-63

22. Kenkre. V. M., Schmid, D. 1983. Chem. S0. Singh, J. 1981. Phys. Status Solidi BPhys. Lett. 94:603-8 1036:423-28

23. Em., V.. Avakian. P.. Merrifield, R. E. 51. Singhj. 1981. J. Chem. Phys. 75:4603-1966. Phys. Rev. 148:862-67 11

24. Kenkre. V. M.. Fort. A.. Ern. V. 1983. 52. Gumbs. G., Mavroyannis. C. 1982.Chem. Phys. Lett. 96:658-66 Solid State Commun. 41: 237-40

25. Kenkre. V. M. 3982. In Exciton 53. Sanche. L., Bader, G. B.. Caron, L. 3982.*Dynamics in Molecular Crystals and J. Chem. Phys. 76:4017-27; 77: 3166-

4ygregates. ed. G. Hohler. Berlin: 70Springer-Verlag 54. Agranovich. V. M., Zakhidov, A. A.

26. Kopelman. R. 3976. In Topics in 3979. Chem. Phys. Lett. 68:86-89Applied Physics. ed. F. K. Fong, 55. Sugakov. V. 1. 1979. Sot'. Phys. Solid15: 297-346. Berlin: Springer-Verlag State 21 : 332-35

27. Kopelman. R.. Monberg, E. M., Ochs, 56. Brikenshtein. V. Kh.. Benderskii. V. A..F. W_. Prasad. P. 1975. Phys. Rev. Lett. Filippov. P. G. 3983. PA vs. Status Solidi34:1506-9 B 117:9-39

652. WofrHkC 167 I vnes ntoi 81. HessenlinI, W. H..Wiersma. D. A.

Yr:Academic 82. Amirav. A.. Even. 0.. Jortner. J. 1982. J.58. Braun, A.. Mayer. U.. Auweter. H.. Ph vs. Chem. 86: 3345-58 ~.

Wolf. H. C., Schmid, D. 1982. Z. 82a. Godzik. K., Hays. T. R., Henkre. W. E..Narurforsch. Tedl A 37: 1013-1023 Selzle. H. L.. Schlag. E. W. 1982. InI59. Campillo. A. J., Shapiro. S. L.. Laser Chemistr " 1:59-75. London:Swenberg. C. E. 1977. Chem. Phys. Lett. Harwood Headem Press52: 11-15 83. Small. G. J. 1982. In Molecular

6.AI-Obaidi. S. J., Birks. J. B.. Birch. D. J. Spectroscopy: volume of ModernS.. Hallam, A.. Imhoff, R. E. 1978. J. Problems in Solid State Physics, ed. V.Phys. B I1I: 2301 -I1 M. Agranovich. A. A. Maradudin. -

61. Davies. M. J., Jones. A. C.. Williams. J. Amsterdam: North-Holland0., Munn. R. W. 1983. J. Phys. Chem. 84. Jankowiak. R., Bassler. H. 1983. Chem.87:541-43 Phys. Lett. 95: 124-28

62. Shinohara, H.. Kotani, M. 1980. Bull 85. Jankowiak. R., Bassler. H. 1983. Chem.Chem. Soc. Jpn. 53: 3171-75 Phys. Lett. 95:310-14

63. Meyer. K. E.. Schein, L. B.. Anderson. 86. Jankowiak. R., Rassler. H. 1983. Chem.R. W., Narang, R. S.. McGhie. A. R. Phys. Lett. 101 :274-781983. Mol. Cryst. Liquid Cryst. 87. Olson, R. W.. Lee, H. W. H., Patterson.

K101 :!99-218 F. G.. Fayer, M.DShelby. R. M..64. Crisp. G. M., Walmsley. S. H. 1982. Burum. D. P.. Macfarlane, R. M, 1982.

Chem. Phys. 68:213-22 J. Chem. Phys. 77:2283-8965. Huber. D. L. 1980. Ph vs. Rev. B 87a. Hess. L. A., Prasad. P. N. 1983. J.

22: 1714-21,24:1083-86 Chem. Phys. 78:626-3166. Kenkre. V. M.. Wong, Y. M. 198 1. Phvs. 88. Reineker, R.. Morawitz. H. 1982. Chem.

Rev. B 23: 3748-55 Ph vs. Lett. 86: 359-6467. Lakatos- Lindenberg, K.. Hemenger. R. 89. Deleted in proof

P.. Pearlstein. R. M. 1972. J. Chem. 90. Freiberg, A.. Saari, P. 1983. IEEE J.Phys. 56:4852-67 Quantum Electron. QE19:622-30

68. Kenkre. V. M. 1982. Chem. Phvs. Lett. 91. Wiesenfeld. J. M., Greene. B. 1. 1983.93: 260-63 Phys. Rev. Lett. 51:1754-58

69. Kenkre. V. M. Parris, P. E. 1983. Phys. 92. Koch, E. E.. GuIrtler. P. 1983. InRev. 8 27: 3221-34 Photoph'v-sics and Photochemistry in the

69a. Khizhnvakov. V. V.. Sherman. A. V. Vacuum U V, ed. S. P. McGlynn. G. L.1980. Sot. Phys. Sol. State 32: 1904- Findley, R. H. Huebner. Dordrecht.9 Holland: Reidel

70. Toyozawa. Y., Shinozuka. Y. 1980. J. 93. Hoffman. B. M, Ibero. J. A. 1983. Acc.Phvs. Soc. Jpn. 48:472-78 Chem. Res. 16:15-21

71. Mayer. U,, Auweter. H., Braun, A.. 94. Zagrubskii, A. A., Petrov. V. V..Wolf. H. C.. Schmid, D. 1981. Chem. Lovcjus, V. A. 1981. Izv. AkadNauk.Phys. 59:449-65 Lat. SSR Fiz. Tver. Tela. 5:29-33

72. Ludmer. Z., Berkovic, G. E.. Zeiri. L.. 95. Ford. W. K.. Duke, C. B.. Patson. A.Muhle. W.. Wolf. H. C. 1982. Chem. 1982. J. Chem. Ph vs. 77:4564-72Phis. Lett. 90: 245-46 96. Ford. W. K.. Duke. C. B.. Salaneck. W.

73. Von Freydorf. E.. Kinder. J., Michel- R. 1982. J. Chem. Phys. 77: 5030-Beyerle. M. E. 1978. Chem. Phys. 3927: 199-209 97. Duke. C. B., Ford. W. K. 1983. Int. J.

74. Matsui, A.. Nishimura. H. 1982. J. Phys. Quant. Chem. 17: 597-608Soc Jpn. 51:I1711-12 97a. Ladik. J. 1981. In Theoretical

75. Matsui, A.. Mizuno. K.. lemura. M. Chemistry. ed. C. Thomson. 4:494if1982. J. Ph vs. Soc. Jpn. 51:1871-77 London: R. Soc. Chem. London

76. Berk. N. F.. Rosenthal. J., Yarmus. L. Butterworths1983. Ph vs. Rev'. B 28:4963-69 98. Rice. S. A.. Jortner. J. 1967. In Physics

77. Szabo, A. 1971. Phys. Ret. Lett. and Chemistry ol'the Organic State. ed.27. 323-26 D. Fox. M. M. Labes. A. Weissberger.

78. DeVries. H,. Wiersma. D. A. 1976. Ph vs. 3:201-497. New York: WileyRet. Lett. 36:91-94 99. Deleted in proof

79. Treshchalov. A. B.. Rozman. M. G. 100. Pope. M., Kallmann. H.. Giachino. J.1983. Opt. Commun. 47:262-67 1965. J. Chem. Phys. 42: 2540-43

80. Olson. R. W.. Patterson, F. G.. Lee. H. 101. Vilesov, F. I., Zagrubskii. A. A..W. H.. Faver. M. D. 1981. Chem. Ph vs. Garbuzov. D. Z. 1964. Sot. Phvs. Sol.L~eft 401-7 State 51460-.64

I b

ELECTRONIC PROCESSES 653102. Bounds, P. J.. Siebrand. W. 1980. Chem. 1973. Chem. Plays. Lett. 23: 128-30 *

Plays. Lett. 75:414-18 128. Sebastian, L., Weiser. G. 1981. Phys.103. Bounds. P. J., Siebrand. W. 1982. Chem. Rev. Lett. 46:1156-59%

4Ph vs. Lett. 85:496-98 129. Movaghar. B.. Cade, N. A. 1982. J.104. Bounds. P. J., Petelenz, P., Siebrand. W. Plays. C Solid State Plays. 15: 1-807-

1981. Chem. Phys. 63:303-20 L813105. Bounds. P. J.. Petelenz. P.. Siebrand. W. 130. Cade, N. A., Movaghar, B. 1983. J.

1982. Chem. Ph s. Lett. 89: 1-3 Phys. C Solid State Phys. 16: 539-50106. Chance, R. R.. Braun. C. L. 1976. J. 131. Donovan, K.. Wilson. E. G. 1981.

Chem. Phys. 64:3573-81 Philos. Mag. B 44:9-29V107. Kato, K., Braun, C. L. 1976. J. Chem. 132. Spannring, W., Bissler, H. 1979. Ber.

Plays. 72:172-76 Bunsenges Plays. Chem. 83: 433-36l08. Sebastian, L., Weiser, G., Bissler, H. 133. Movaghar B., Murray. D.. Donovan.

1981. Chem. Plays. 61:125-35 K. J.. Wilson. E. G. 1984. J. Phys. C109. Sebastian. L.. Weiser, G.. Peter, G.. 17: 1247-55

Bassler, H. 1983. Chem. Plays. 75: 103- 134. Geacintov, N. E., Pope. M. 1971. In14 Proc. 3rd Int. Conf. Photoconductivity,

109a. Siebrand. W., Zgierski. M. Z. 1983. In ed. E. M. Pell. pp. 289-95. New York:Electronic Excitations and Interaction PergamonProcesses in Organic Molecular 135. Orlowski, T. E., Scher. H. 1983. Plays.

2'Aggregates. ed. P. Reineker. H. Haken, Rev. B 27:7691-7702H. C. Wolf. pp. 136-44. Berlin: 136. Scher. H., Orlowski, T. E. 1983. Plays.Springer-Verlag Rev. Lett. 50:775-78

110. Hinchljiffe. A.. Munn, R. W., Siebrand. 137. Jortner. L. Choi, S. I., Katz J. L., Rice,W. 1983. J. Phvs. Chem. 87:3837-39 S. A. 1963. Plays. Rev. Lett. 11: 323-26

Ill. Silinsh. E. A.. Kolesnikov, V. A., 138. Petelenz, P. 1980. Chem. Plays. Lett.Muzikante. 1. J1., Balode. D. R. 1982. 76:186-89Plays. Status Solidi I I 3b: 379-9 3 139. Yokoyama. M.. Matsubara. A..

112. Popovic. Z. D., Sharp. J. H. 1977. J. Shimiokihara. S.. Mikawa. H. 1982..Chem. PlaYs. 66:5076-82 Polymer J. 14:73-75

113. Eichhorn. M., Willig. F. Charle. K. P.. 140, Popovic, Z. D. 1983. J. Chem. Plays.Bitterling. K. 1982. J. Chem. Plays. 78:1552-5876:4648-56 141. Silinsh, E. A,. Kolesnikov, V. A..

114. Eichhorn. M. 1982. PhD thesis. Muzikante. 1. J., Balode, D. R., Gailis,Technical Univ.. Berlin A. K. 1981. Izv. Akad. Nauk. Latv. SSR

115. Slotnick. K. 1983. PhD thesis. New Ser. Fiz. Tekhn. Nauk. 5:14-28York: Univ.. New York 142. Klein. ., Martin. P.. Voltz, R. 1981. J.

116. Onsager. L. 1934. J. Chem. Plays. Lumin. 24'25: 99-1022:599-615 143. Moller. W.. Pope. M. 1973. J. Chem.

117. Onsager. L. 1938. Phlays. Rev. 54: 554- 57 Plays. 59:2760-61118. Chance. R. R., Braun. C. L. 1976. J. 144. Ries, B.. Schonherr. G.. Baissler. H..

Chem. Ph vs. 64: 3573-81 Silver. M. 1983. Philos. Mag. B 48:87-119. Hong. K. M.. Noolandi. J. 1978. J. 106

(hem. Ph vs. 69: 5026- 39 145. Hong. K. M., Noolandi. L.. Street. R. A.120. Noolandi. J.. Hong, K. M. 1979. J. 198 1. Plays. Rev. B823: 2967-76

Chem. Ph ' s. 70: 3230-36 146. Pope. M.. Burgos. J. 1966. Mol. Cryst.121. Scher. H.. Rackovsky. S. 1984. J. (hem. 1:395-415

Play s. In press 147. Altwegg. L., Pope. M., Fowikes. W.121a. Rackovsky. S., Scher, H. 1984. Plays. 1984. Chem. Plays. 86:471-82

Rer. Lert. 52:453-56 148. Fuinfschilling, J., Samoc. M.. Williams.122. Ries. B.. Sch6nherr. G.. Bissler. H.. D. F. 1983. Chem. Ph vs. Lett. 96:157-

Silver. M. 1983. Philos. Maq. 48:87- 60106 149. Klein. 0.. Carvalho. M. J. 1977. Chem.

123. Seiferheld. U.. Ries. B.. Bassler. H. 1983. PlaYs. Lett. 51:409-12J. Phi.vs. C Solid State Plays. 16:5189- 150. Popovic. Z. D. 1983. Chem. Ph vs. Lett.1201 100:227-29

124. Deleted in proof 151. Altwegg. L., Davidovich. M. A..125. Lochner, K.. Reimer. B.. Bissler. H. Funifschilling. J.. Zschokke-Grinacher.

1976. Chem. Ph vs. Lett. 41:388-90 1. 1978. Phys. Lett. Rer. B 18:4444-53126 Siddiqui. A. S.. Wilson. E. G. 1979. J. 152. Braun. C. L., Scott. T. W. 1983. J. Plays.

Phivs C Solid State Plays. 12:4237- (hem. 87:4776-7841 153. Schein. L. B.. Brown. D W. 1982. Mol.

27. Haberkorn. R.. Michel-Beycrle. M. E. (rst. Liquid (rtst. 87:1-12

.t ~- %J~W~~T ~ - '~'v , .7

654 POPE & SWENBERG

154. Schein. L. B.. Duke, C. B., McGhie, A. 179. Heeger, A. J., Schriefier.J. R. 1983. Solid .-

R. 1978. Phys. Rev. Lett. 40:197-200 State Commun. 48: 207- 10 -

155. Schein. L. B. 1977. Chem. Phys. Lett. 180. Suzuki, N.. Ozaki. M., Etemad. S..48:571-75 Heeger, A. J.. MacDiarmid. A. G. 1980.

156. Schein. L. B.. McGhie, A. R. 1979. Phys. Phys. Rev. Lett. 45:1209-13pRev. B820:1631-39 181. Leo. V., Gusman. G., Deltour, R. 1982.

157. Katz. J. L.. Rice. S. A.. Choi. S.. Jortner. Ph vs. Rev. R 8 26: 3285-88J. 1963. J. Chem. Phys. 39: 1683-97 182. Shank. C. V., Yen. R., Fork. R. L..

158. Silbey. R., Jortner. J.. Rice. S. A.. Vala. Orenstein. J.. Baker. G. L. 1982. Phys. V .M.. T. Jr. 1965. J. C'hem. Phys. 42: 733-37 Rev. Lett. 49:1660-63

159. Surm H. 1978. Solid State Commun. 1 82a. Su, W. P., Schrieffer, J. R. 1980. Proc.28: 309-12; 29:495-99 Nat!. Acad. Sci. USA 77: 5626-29

160. Sumi. H. 1979. J. C'hem. Phys. 70: 3775- 183. Baerisviyl. D. 1983. Heir. Ph vs. Acta85 56:639-53

161. Sumi, H. 1979. J. Chem. Phys. 71: 3403- 184. Kivelson, S. 1981. Phys. Rev. Lett.11 47:1549-53; 1982. Phys. Rev. B

162. Sumi. H. 1979. J. (Chem. Phys. 75: 2987- 25:3798-382193 185. Shacklette, L. W., Chance. R. R., Ivory.

163. Andersen. J. D., Duke. C. B.. Kenkre. V. D. M., Miller. G. G., Baughman, R. H.M. 1983. Phys. Rev. Lett. 51:2202-5 1980. Synth. Metals 1: 307-20

164. Schein, L. B., Narang, R. S., Anderson. 186. Bredas, J. L.. Chance, R. R., Silbey, R.R. W., Meyer. K. E., McGhie. A. R. 1982. Phys. Rev. B 26: 5843-541983. Chem. Phys. Lett. 100: 37-40 187. Bredas. J. L.. Chance. R. R., Silbey, R.

165. Warta. W., Karl. N. 1982. 10th Mol. 1982. Mo. ('ryst. Liquid Crys. 77. 319-Cryst. Symp.. St. Jovette, Canada 32

166. Epstein. A. J., Conwell. E. M.. eds. 188. Crecelius. G., Stamm, M., Fink. J.,1981182. Proc. Int. Conf. on Low Ritsko. J. J. 1983. Phys. Rev. Lett.Dimensional Conductors, Mo!. Cryst. 50:498-500Liquid Cryst. Vols. 77, 79.81,.83,85. 86 189. Bredas. J. L.. Themans. B., Andre, J. M.

167. Int. ('onf. sur la Ph vs. C'hem. des 1983. Ph vs. Rev. B 27:7827-30Polymeres et Conductors, 1983. J. 190. Fesser, K., Bishop, A. R.. Campbell, D.Physique 44: Coll. C-3 L. 1983. Phys. Rev. B 27: 4804-25

168. Etemad. S.. Heeger, A. J., MacDiarmid. 191. Etemad, S., Feldblum, A., MacDiarmid.A. G. 1982. Ann. Rev. Phys. ('hem. A. G., Chung, T. C., Heeger. A. J. 1983.33:443-69 J. Phys. Coll. 44(C3):413-22

169. Nigrey. P. J., MacDiarmid. A. G., 192. Movaghar. B.. Murray. D., Pohlmann.Heeger. A. J. 1982. Mol. Cryst. Liquid B. Wuirtz. D. 1984. J. Ph ' s. C 17:1677-83Cryst. 81:309-17 193. Alexander. S., Bernescorn. J., Schneider,

170. Robin. P., Pouget, J. P.. Comes. R., W. R. 1981. Rev. Mod. Phys. 53(2): 175-Gibson, H. W., Epstein. A. J. 1983. Phys. 98Rev. B 27: 3938-94 194. Seiferheld. U.. Baissler. H., Movaghar.

171. Hoffman. D. M., Gibson, H. W., B. 1993. Phj's. Rev. Lett. 51:813-16Epstein, A. J.. Tanner, D. B. 1983. Phys. 195. Bassler, H. 1981. Phys. Status Solidi BRev. B 27:1454-57 107:9-53

172. Francois, B.. Bernard. M., Andre, 3. J. 196. Schonherr, G., Baissler. H.. Silver. M.1981. J. ('hem. Phys. 75:4142-52 198 1. Philos. May. B 44: 369-81

173. Weinberger. B. R., Ehrenfreund. E., 197. Hirsch, J1. 1974. Phys. Status Solidi APron. A.. Heeger, A. J1., MacDiarinid, A. 25:575-80G. 1983. J. ('hem. Phys. 72:4749-55 198. Baissler. H., Schonheff, G., Abkowitz,

174. Weinberger, B. R., Kaurer, J., Heeger, A. M., Pai. P. M. 1982. Phys. Rev. BJ.. Pron, J.. MacDiarmid, A. G. 1979. 26:3105-13Phys. Rev. B 20:223-30 199. Lange, J., Bassler, H. 1982. Phys. Status

175. Roth. S.. Eichinger. K., Menke, K. 1984. Solidi B 114: 561-69In Quantum Chemistry of Polymers, ed. 200. Mort, J.. Pfister, G. 1982. In ElectronicJ. Ladik. M.. Andre, pp. 165-90. Properties of Polymers, ed. J. Mort, G.Dordrecht: Reidel Pfister, pp. 215-65. New York: Wiley I

176. Nechtschein. M., Devreux, F., Greene, 201. Salaneck. W. R. 1978. Phys. Rev. Lett.R. L., Clarke. T. C.. Street. G. B. 1980. 40:60-63Ph vs. Ret'. Lett. 44:356-59 202. Duke, C. B., Fabish. T. J., Paton, A.

177. Thomann, H.. Dalton. L. R., 1977. ('hem. Phys. Lett. 49: 133-36Tomkiewicz. Y., Shiren, N. S., Clarke, T. 203. Silver. M., Schonherr. G., Baissler. H.C. 1983. Phys. Rev. Lett. 50: 533-36 1982. Phys. Rev. Lea. 48:352-55

178. Kuroda, S.. Schrieffer, J. R. 1983. Solid 204. Taure, L. F., Silinsh. E. A.. Muzikante, 1.State Commun. 43: 591-94 J., Rampens. A. 3. 1983. In Taqungshand I

....................... 7 ~.%

ELECTRONIC PROCESSES 655Organische Festkorpen. Brandenburg: D. L., Loesing. D. M. 1983. J. Am.DPR Chern. Soc. 105: 643-45

205. Samoc, M., Samoc, A., Sworakowski, J.. 231. Williams, J. M., Beno, M. A..Karl, N. 1983. J. Phys. C Solid State Appelman, E. H., Wickel. F.. Aharon-Ph-vs. 16:171-80 Shalom, E., Nalewajek, D. 1982. Mo!.

206. Plans. J., Zielinski, M., Kryszewski, M. Cr ' st. Liquid Cryst. 79:319-261981. Phys. Rev. B823:6557-69 232. Jacobsen, C. S.. Pederson. H. J.,

207. Karl. N.,.Sato,.N., Seki, K.,.Inokuchi. H. Mortensen, K., Rindorf, G., Thorup, N.,1983. J. Chemn. Ph vs. 77:4870-78 Torrance, L1. Bechgaard. K. 1982. J.

208. Yoshie.O0., Kamihiara, M. 1983. J. Appl. Phys. C 15:2651-63Phys. Jpn. 22:629-35 233. Pouget. J. P.. Shirane. G., Bechgaard.

209. Tiedje. T.. Rose, A. 1981. Solid State K.. Fabre. J. M. 1983. Phys. Rev. 8Commun. 37:49-52 27:5203-6

213. Orenstein. J., Kastner. M. A. 198 1. Solid 234. WudI. F. 1981. J. Am. Chtern. Soc.State Commun. 40:85-89 103:7064-69

211. Monroe. D., Kastner, M. A. 1983. 235. Grant. P. M. 1982. Phys. Rev. B -

Philos. Mag. B 47:605-20 26:6888-95212. Arkhipov. V. I., Popova. J. A.. 235a. Parkin, S. S. P.. Engler, E. M.,

Rudenko. A. 1. 1983. Philos. Mag. B Schumacher. R. R., Lagier. R., Lee. V.48:401-10 and references therein Y., Scott, J. C.,Greene, R. L. 1983. Phys.

213. Arnold, S.. Hassan, N. 1983. J. C/tern. Rev. Lett. 50:270-73Ph vs. 78: 5606-11 236. Parkin, S. S. P., Engler, E. M.,

214. Pope. M., Kallmann, H. 1972. lsr. J. Schumacher, R. R., Lagier, R., Lee, V.C/tern. 10: 269-86 Y., Voiron, J., Carneiro, K., Scott, J. C..,

215. Thomas, J. M.. Evans. E. L., Williams. J. Greene, R. L. 1983. J. Phys. Colloq.0. 1972. Proc. R. Soc. Ser. A 331:417-27 44(C3): 791-97

216. Williams, J. 0., Thomas. J. M. 1973. In 237. Rentzepis, P. M. 1982. ScienceSurface and Defect Properties of Solids, 218:1183-90pp. 229-49. Vol. 2, Specialist Periodical 238. Shank, C. V., Greene, B. 1. 1983. J. Am.Re port. London: Chemical Society C/tern. Soc. 87: 732-34

217. Zboinski. Z. 1983. C/tern. Phys. 75: 297- 239. Klein. G. 1983. C/tern. P/tys. Lett.304 95:305-9; 97:114-18

218. Petelenz, P. 1981. Mat. Sci. 7:285-90 240. Kalinowski, J. 1981. Mat. Sci. 7:44-50219. Movaghar, B.. Sauer, G. W., Wurtz D.. 241. Kalinowski. J.. Jankowiak, R., Bassler.

Huber, D. L. 198 1. Solid State Commun. H. 198 1. J. Lumin. 22: 397-1839:1179-82 242. Kolendritskii, D. D., Kurik, M. V.,

220. Movaghar, B., Sauer. G. W., Wurtz, D. Piryatinskii, Yu. P. 1979. Phys. Status1982. J. Stat. P/tys. 27:473-85 Solidi B 91:741-51

221. Hunt. I., Bloor. D., Movaghar, B. 1983. 243. Klopffer, W. 1982. In ElectronicJ. Phys. C Solid State 16:2623-28 Properties of Polymers. ed. J. Mort, G.

222. Dudley. M., Sherwood. J. N., Bloor. D., Pfister, pp. 161-2 13. New York: Wiley .:'

Ando, D. J. 1982. J. Mat. Sci. Lett. 244. Ritsko, J. J. 1982. In Electronic '1:479-81 Properties of Polymers, ed. J. Mort, G.

223. Haarer. D., M61hwald. H. 1975. Phys. Pfister, pp. 13-58. New York: Wiley -

Rev. Lett. 34:1447-50 245. Duke, C. B. 1983. In Physicochtemical224. Deleted in proof Aspects of Polymer Surfaces, ed. K. L.225. Bechgaard. K., Jacobsen. C. S., Mittle, 1: 463-75. New York: Plenum

Mortensen, K.. Pedersen. H. I., Thorup. 246. Duke. C. B.. Fabish. T. J. 1978. J. App).N. 1980. Solid State Commun. 33:1119- P/tys. 49:315-2125 247. Alzetta. G.. Chudacek. I., Scarmozzino,

226. Jerome. D., Mazaud, A., Ribault. M., R. 1970. Phys. Status Solidi 1: 775-85Bechgaard. K. 1980. J. Phys. Paris Lett. 248. Nowak. R., Krajewska. A., Samoc, M.41:1,95-1-98 1983. Chem. Phvs. Lett. 94:270-71

227. Jerome. D., Schulz. H. J. 1982. Adv. 249. Wada. Y. 1982. In Electronic PropertiesPhys. 31: 229-490 of Polymers, ed. J. Mort, G. Pfister. pp.

228. Friedel. J.. Jerome. D. 1982. Contemp. 109-60. New York: WileyPhtvs. 23:583-624 250. Kepler, R. G. 1978. Ann. Rev. Phys.

229. Colloq. lht. CNR sur la Phys. et Ia C/tern. 29:497-18Chem. des Metaux Synthetique et 251. Davies. G. R. 1980. In Physics ofOrganiques. 1983. J. Phys. Colloq. Dielectric Solids. ed. C. H. L. Goodman.44(C.3) pp. 50-63. Conf. Scr. No. 58. Bristol:

230. Williams, 3. M.., Beno. M. A.. Sullivan. J. Inst. Physics Techno HouseC.. Ranovetz. L. M., Braam. J. M., 252. Spannring, W.. Baissler. H. 1981. Chern.Blackman. G. S.. Carlson, C. D.. Greer, Phys. Lett. 84: 54-58 i

- - - ~ ~ r

i

L

FILMEDS -

~

A

-'4

DTIc- ~-rr~-~