INTERIM REPORT PART II LIGHT SOURCE EVALUATION February 1970 1. FACTORS AFFECTING THE PERFORMAI%TCE OF PYROTECHNIC MIXTURES The object of this study is to investigate the feasibility of utilizing high intense light emitted from rapidly burning p otecrinic yr mixtures for optical incapacitation. Several 2mixtures have been selected for initial study. The compositions of these mixtures were itemized in a previous progress report. 1 It is useful to review the mechanisms of pyrotechnic burning and th e e important factors, r levant to light emission, so as to fully under- stand the utility of this approach. 2 1.1 DESCRIPTION OF BURNING PROCESS The basic ingredients of most pyrotechnic compositions consist of a fuel and an oxidizer. These ingredients should be stable under normal shelf conditions, yet they should be easy to ignite. Further, once ignition occurs, the heat released during burning should be sufficient- icl.-cuistain the 2buming process. It is useful to consider the following model to explain the burning behaviorof,--e-,pyrateehn Three thermal zones are established when an illuminating composi- tion is ignited and bums propagatively (see Figure 1). Air Flow (D 2 Tf B C Flame Ta x Tf = Flame Temperature T 1 a = Ambient Temperature x = Distance Below Burning Surface Figure 1. Profile of Combustion Zone Zone A is essentially the "burning su-face" Both exothermic and endothermic reactions talze place resulting in the formation of gaseous fuel and oxidizer intermediates. These intermediates react exothermically In the flame zone. Usually the pyrotechnic is designed to be fuel rich and the excess fuel.reacts with oxygen from the atmosphere.2 In these cases the flame size and intensity are somewhat dependent upon the required environmental oxygen and its availability. The energy required to form the reactive intermediates are h generated by the ekotherrhic reactions which occur in the flame zo e an 2 in Zone A. The dominant heat transport mechanisms are radiation feed- back from the flame and conductive transport from condensed phase reactions which occur in Zone A. Energy from Zone A is also transferred to Zone B which may be considered the pre-ignition zone. Directly below Zone B is the remainder of the unreacted pyrotechnic composition, or Zone C. 2 The radiative heat transport from the flame to Zone A can be described mathematically by the Stefan-Boltzmann equation 4 4 I c CF (T T f o where I is in terms of energy per unit area per unit time (e.g. ,cals/ cm2-sec)--i-,@e2p-,i-,sthe-eftriniv at-the-fta constant, F is a flame shape factor, Tf is the flame temperature and To is the temperature of the surface receiving the radiation. The temperature in Zone A is further affected by the local endo- thermic and exothermic transitions which occur in forming 2 the reactive intermediates. The important reactions which limit the burning rate of the pyrotechnic are the endothermic processes. Typically, these processes involve a. phase changes, and b. pyrolytic decomposition to generate gaseous combustion reactants. Under steady-state conditions the thin pre-6ignition zone (i.e.. Zone B) can be assumed to have a uniform temperature. For simple systems this zone has thic knesses in the molecular size range, and its temperature is governed.almost completely by the endothermic processes taken place. The tempcrak'ure profile within the body of the pyrotechnic (i.e. Zone C) can be approximated by the Rosenthal equation2 TX T + (Ts - T exp (-vx/a) (2) 0 a where Tx is the temperature at distance x2 below the reacting surface ambient temperature, T is the temperature within. (Zone B), Ta is the s Zone B, v is the burning rate, and a, Is the thermal diffusivity of the mixture: It should be noted that thermal diffusivity is related to more conventionalthermal.-Properties,.,i.e.-,2-,. 2/sec cL k/pc, cm (3) where k is the thermal conductivity, p is.density and c is the specific heat. 1.2 FACTORS-AFFECTING LIGHT OUTPUT The distribution of radiation in any spectral region is determined by t2he chemical nature and physical state of the products which emit in that regio4 and the temperature reached by these emitting species. The rate at which a pyrotechnic mixture burns depends an the amount and rate at which heat is evolved. Sufficient heat must be produced to raise reaction will be initiated and the reaction rate must be sufficient to more than compensate 2for heat losses in order for the burning to be sustained. Mathematically, the burning rate, v,,. can be related to the energy fedback from the exothermic processes and the rate determining endothermic process which must occur to produce the reactive inter- mediate s v E-(I)/p (AH + C AT) (4) 2 where Z (I) is the radiative, convective and conductive heat flux fedback to the pre-ignition zone, LH is the heat absorbed in the pre-ignition zone by the endothermic processes and C AT is the heat required to elevate the temperature of the pyrotechnic (i.e. , at the boundary between Zones B and C) to the reaction temperature. The latter (pC AT), can be best 2described as a heat loss term. The rate of burning, the products formed, and the flame tempera- ture are affected markedly by the compositiorr-of the mixture, as well as by the physical condition of the materials and the ambient conditions under which it is burned. Some of the more important factors which affect the performance of light producing pyrotechnics which were c2onsidered in our initial selection compositions are as follows, 3 heat of reaction 2 composition of emitters 3 particle size 4. consolidation 5 pyrotechnic diameter 6 container design 1.2. 1 Heat of Reaction The heat of a reaction is defined thermochemically as the 2 difference between the thermodynamic heats of formation of the reactants and products of the reaction. For example, referring to the energy dia- gram shown in Figure 2, one selects a pyrotechnic mixture having substituents which have a higher heat of formation than the reaction products and one which requires a minimal amount of input energy, &Ha, to initiate burning. 2 R Pyrotechnic Components (Reactants) P Combustion Products 0 >1 A Ha (D 0 R > Heat of Reaction 2 p Reaction Path Figure 2. Reaction Energy Diagram One of the important factors in deter-mining the luminous intensity of a light-producing pyrotechnic device is the temperature reached by the emitting species in the flame and produced by the burning of the pyro- 2 technic mixture. The texnperature reached depends, in turn, on the amount and rate at which energy is released by the reaction. Therefore, the energy released during combustion sh ducts ould be high and the pro formed should be stable at the high temperatures necessary to produce the desired lu inous intensity. 1 m 77 1.2.2 Desirt-d Output Spuctra The ultimate Pyrotec@nic composition must emit intense light in regions most sensitive to the eye. It can be seen from the "standard observer curve" , shown in Figure 3 that the eye Is only sensitive to a very narrow region of electromagnetic radiation. 0 2 > 650 is* Wavelength, millimicron Figure 3.- Standarct ObserverZiirvp-- This region extends between approximately 400 to 700 m,-a (or 4000 to 7000 Angstroms). In Figure 4, the visual response curve of the human eye is shown. 2In this figure the effective radiation, In terms of photometric units (lumens) as a function of wavelength is shown. The absolute photopic luminosity is defined as the ratio of the electromagnetic flux sensed by the eye (in units of lumens) to the total radiant flux (in terms of watts). The most sensitive region in this narrow spectrum lies between 500 and 560 mp. Therefore, for e2ffects related to visibility of the light source, the pyrotechnic mixtures should be designed to emit strongly in this wavelength region. A separate question arises as to whether optical incapacitation effects are similarly correlated over the visible range. This answer is not known at present-but will be assumed to be positive for the present. Special design features must b2e included in a pyrotechnic to insure that a significant fraction of the total radiation emitted by the flame is in the visible region. The emission from a pyrotechnic flame is composed of line spectra, band spectra and continuum. The latter is directly dependent on the temperature of the flame. The continuum is essentially blackbody or greybody radiation. The distribution of the radiant 3 energy versus wavelength can be estimated from Planck's equation s I Ali, %4 Absolute Photopic Luminosity of Radiant Flux,, Lumen/Watt (D C:) 2 c:@ CD c:l C:) 0 In C) 0 En n 0 2Ca < 0 M En C) to (D (D (D 0 0 r+ CA) C:) C> 0 0 tri C) 0(D 0 r+ co L C:) col ta. 0 m 2 h 2 c 5 (exp'(hc/k X T) - 1) -,-j f energy per r unit t where I is the radiant flux, in terms 0 unit area pe time at wavelength X, emitted by a hot source at temper2ature T, h is zmann S constant Plancl.-I's constait, c is the speed of light, k is Bolt pical flux-wavelength distri- and r:, is the emissivity of the flame. Ty butions calculated from this equation are shown in Figure S. 2 350 0 MAX 0 30 C,] F- 250 - 200 >4 4-A 2 4 150. too - T 3000*K 2500*K lz Te2OOOOK *K T 2I 00 so- OOL, A "Z Ia 3 4 5 0 Wavelength, Microns Figure S. Planck's Law: Radiance as a Function of Wavelength for Various Temperatures It can be seen from this figure that very little of the energy emit2ted by blackbody radiation is distributed in the wavelength band most sensitive to the eye. Further, in.order to generate an intense source in the visible spectrum, a very high flame temperature will be required and even under iciency of the source in terms of the fraction of these conditions the eff visible light energy7 generated out of the total radiation energy would be very low. 7 e Pvrotechnic Ingredients 1.2.3 Des-irabl ,4 This observation leads one to recognize that pyrotechnics composed of organic fuels which have the highest heat of reaction can- not be considered since these mixtures produce flames having character- i stic ssimilar to blackbody emitters. This is p2articularly true for fuel rich compositions which have a tendency to generate significant amounts of carbon and aromatic soot particles. The radiant,emiss@iomin-the-v-isihl-t--canz-b(.-improved-by-in@c-ludirfg--,-@,',@-- chemicals in the pyrotechnic mixtures that will form thermally excit2ed reaction products capable of emitting radiation at desired wavelengths. This process can be described by the following equation, A + B + heat EC* C + hv (6) LC Pyrotechnic Reaction Thermally Light Components Product in Excited Emission Flame Product -ies Spec Many inorganic salts exhibit this phenomena. Several elements which react in pyrotechnic flames forming oxides, hydroxides and chlorides have been used to "co2lor" flames. These include, strontium which produce* A,-.re-d--czlDr,-ba green and orange) and copper (blue to green). Lithium (red), boron (green), thallium (green), rubidium (red) and cesium (blue) are also strong color producers but their use is not as practical because of cost, toxicity or the nature of their compoun2ds. The actual emitting species of these metals are known to be the di- and tri-atomic sp6cies which can exist at high tem peratures in the flame. For example, a. the red light produced by flares containing strontium and a source of chlorine is a result of SrCl emission (strong emission 2 near 640 MP). In the absence of chlorine, emission has been attributed to SrO. b. BaCl2 emits in the 505-535 MP region (green).. c. BaO emits over a broad spectrum, 400 to 800 MP 8 4 7:-7 d. The hydroxides of these melals also emit in the respective wavelength bands. 4 e . MgO emits at apprcximately 500 m 1.2.4 Color Intensifiers e Where possible, chlorides ar 2added to the pyro'Lechnix mixtures to enhance the color of the flame. Perchlorate oxidizers contribute the minimal requirements without reducing the efficienc)@- of -the energy out- -7, put. In some cases it has been found that the addition of chloro- organics significantly increase the color intensity. Substances such as hexachlorethane, hexachlorobenzene, polyvinylchloride are sometimes employed for this purpose. 2 . CRITERIA USED TO SELECT CANDIDATE PYROTECHNIC MIXTURES --were--ma@de in selecting candidate The following cori-siderations pyrotechnic formulationsl for the initial experimental investigations. Based2'on the,premis-ed, desirability of producing flames w-hichemit radiation in the visible wavelength region,primary consideration has been given to inorganic compositions. 2.1 OXIDIZER Comprehensive literature surveys by Shock Hydrodynamics and other investigators have shown that perchlorates are the most desirable 2 oxidizers. They contain a relatively high ratio of oxygen to total mole- oular weight, their heats or reaction with metals such as aluminum and magnesium are significantly better than other oxidizers and they generate only oxygen as a gaseous product. Perchlorates are gererally stable, m yet they are easily ig2nited, unlike oxidizers such as nitrates and etallic oxides. Nitrates'@uffer additional disadvantages in that they are not as exothermic and thus tend to have a slower burning rate with fuels such z as aluminum and magnesium (see Equation 4), and they generate a non- reactive gaseous product, nitrogen. Of the alkali perchlorates, lithium and sodium perchlorates h2ave the highest rates of oxygen to total molecular weight, viz., 64/106.4 and 64/127.45, respectively. However, these compounds are extremely hygroscopic. This characteristic decreases the storage life of a pyro- technic system. The absorption of water by these oxidizers is an'exo-. thermic process. Becatise of this factor one must also be concerned 7 with the design of special safety precautions. 9 Potassium parclilorate was selected as a primary oxidizer for the initial candidate mixtures. This oxidizer has an oxygen-total weight ratio of 64/138.55, which is somewhat lower than LiClO4 and NaClO4, however, it is a stable compound and its heat of reaction with aluminum, for example, is slightly greater than the 2reactions between LiClO4 or NaClO4 and aluminum. 2.2 FUEL COMPONENTS It has been found that aluminum and magnesium are the best fuels for use.in photoflash mixtures. The heat of reaction and peak light intensities resulting from Al/KC104 are much higher than equivalent Mg/KC104 compositions.2 The radiation emitted follows generally what would be expected for continuum radiation. Peak light intensities for stoichiometric mixtures of Al/KC104 and Mg/KC104 have been measured to be approximately 40 and- 18 million candles. respectively.- 2.3 CONSOLIDATION OF MIXTURE It has been shown, that the manner in which the fuel and oxidizer 2 are incorporated in the pyrotechnic device will greatly influence its performance.5 Consolidated compositions contain binders, usuall)@ -Waich-fnrTn a-itgid-@nt orga nic. -pol, mzr:* y dated composition however is a burning system which has a relatively large spa2tial separation between fuel and oxidizer. Thus, the burning rates are relatively slower than a comparative non-consolidated system. Non-consolidated systems under confinement usually have higher deflagration rates than consolidated systems and the intensity of the emitted light is greater. Most photoflash -systems are thus non- consolidated, and this type of system was selected for thes2e studies. 2.4 OUTPUT IMPROVEMENT (SELECTION OF STANDARD MI:KTURE) A standard photoflash composition, III..;A, was selected as a reference. The composition of this mixture is shown in Table I The addition of barium nitrate to this mixture increases the radiation output in the visible spectrum over that of the basic Al/KC104 mixture. As 2 discussed in a previous section, the BaO and BaCI2 formed in the burning processes emits strongly in the wavelength region most sensitive to the eye. 2.4.1 Mixture "D" During the Korean conflicts an experimental photoflash mixture having a very high fuel to oxidant ratio was developed having a peak 0 i n -ASH OF TABLE I. ClUiPP%.CTERISTICS TYPE III PliOTOF1 COMPOSITION6 Percen Ingredi2ents Specification Microns Aluminum, atomized JAN-A-289 is 40 PA-PD-254 24 30 Potassium Perchlorate PA-PD-253 147 30 Barium Nitrate pHysiCo-CHE AICAL DKTA: 2 Heat of Reaction, cal/g-/-.774 (calc) rature, OG-approx- 3500 Reaction Tempe Gas Volume, cr-/g-24 (caic) Tapped-1.67 OC, cc gas/40 hrs-O 16 Stab, 120 Vac SENSITIVITY DATA: 2 Impact: PA, inches-40 + Steel-Crackles; Fiber-No Action Friction Pend: 'DTA-NO Ignition Ignition Temp, OC: s sec value-610; p;,Hrs 24; Hygroscopicity2: 57% RH, room tem % Wt Gain < 0.1 Electrostatic Joule, Min 2.14; so% Pt-3.5, Sensitivity: -4.5; Temp-65OF; 100% Pt 6 % RH-40: Unconfined-Yes light output tv@iice that of the Type III mixture. This mixture consisted of 70% Al/30% KC104. Afurther improvement in performance is anticipated I W which will by replacing a portion of the I--ClO4 oxid'zcr ith Ba(NO3)2 act as an oxidizer and color enhancer 2(see Table II for compositions of candidate pyrotechnic mixtures). TABLE II Pyro* Arbitrary Type Designation Composition A B c D Designation Ingredients % % % % 2 Al 40 .50 25 70 KC10 30 40 30 20 4 30 10 10 Ba(NO3)2 Ca Si 10 Mg 2 3S TOTAL 7TO 0 100 100 100 *All mixtures prepared in accordance with PA-PD-267. TYPE A Follows the formulation given in PA-PD-267 for Type III Class A (Fine Oxidizers TYPES 7B, C, & D follows the same guidelines including particle size given in PA-PD-267. 12 2.4.2 Mixture "B" The addition of calcium silicide should have two ef" cts on the basic Al/KC104 reaction: 1. Because of the exothermic nature of CaSi in an oxidizing environment, the heat of reaction of this mix should be greater than the selected 2 reference-@i,.e-,.mixture A)-.--.This-incT-e4,se-An- the heat of reaction will raise the flame tempera- ture and the radiation intensity of the flame. 2. The flame emission of calcium is at 550 and 620 m@4 . This is in the yellow-green and ora2nge areas of the spectrum and as can be seen from Figure 4 should improve the flame luminosity in the most sensitive wavelength regions. A smaller percentage of calcium silicide than barium nitrate is required because of molecular weight differences. Calcium has an atomic weight of less than a third that of barium. Thus, on a2 weight basis cal-cium should be more efficient. 2.4.3 Mixture "C It has been observed that the use of magnesium-aluminum fuel mixture s re s ult s in fla s he s of I ong er duration. Thi s fuel c ombination i s also easier to ignite as compared with aluminum. Mixture 'IC " was therefore formulated for purposes of determining the 2differences in potential effectiveness with flash duration. 2.5 PARTICLE AND SHELL SIZE The same guidelines of particle size suggested for the reference photoflash mixture (designation "A") are being used for the other mixtures. This control is necessary so as to minimize the number of unknown experimental parameters. A2 representative array of shell sizes have been included in the experimental plan. These shells were described in Reference 1. AU of the casings are composed of aluminum. 2.6 METHOD OF INITIATION An addi*lional variable which was included in the experimental plan was the method of initiation. Simple central burster initiation'as 3 well as an imploding initiation system are included in the test Ian for p comparison. REFERENCES 2. D. Rosenthal, Trans. Amer. Soc. Engrs. 68, 849 (1946). 3. R. L. Tischer and K. Scheller, "A Gaseous System of High Luminous Efficiency", (Presented at the Western States Combustion Institute Meeting, 1967), ReportVV'SCI-67-12, Aerospace Research Labs. 4. 2 D. M. Johnson, "Proposed Kinetics and Mechanics of Illuminant Flares: Maximizing Efficiency", U. S. Naval Ammunition Depot, RD-TR-32 (June 1966), AD 627 649. 5 . 'The Compilation of -Flame and Shockwave Information Applicable to Photoflashes", Final Summary Report, Contract DAI-28-017- 501-ORD(P)-1096, Arthur D. Little Inc. (September 19055 6 Engineering Design Handbook, "Military Pyrotechnics Series", AMCP 706-185, U. S. Army Materiel Command (April 1967 4 14 L SUMMARY R-RPORT FINA PART 2 LIGHT SOURCE EVALUATION June 1970 'fAr)LL OF 00@,"IEIV'TS 1 11@TRODUCTION . . . . . . . . . . . . . . . . . . . . . 1 2 . DESCRIPTION OF EXPERIhIEl@,T!iL PIIOTOFL)\SH DEVICES 2 . . . . . . . . . . . . . . . . . 2 2.1 Photoflash Cartridge, Type S . . . . . . . . . 2 2.2 Photofla sli Cart-ridge-" Type SDE 2 2.3 Photoflash Shell, Type SS ... . . . . . . . . . 2 2.4 Photoflash S2hell, Type SSDE . . . . . . . . . 2 3 . DESCRIPTION OF EXPERIMENTS . . . . . . . . . . . . . 5 3.1 Experimental Arrangement . . . . . .5 3. 2 Detector Calibration . . . . . . . : .. ., - . .,. 5 2 3 .3 Experimental Errors . . . . .... . . . . . . . . 6 4. EXPERIMENTAL RESULTS . . . . . . . . . . . . . . . . 9 4.1 Comparison Between Reported and Measured Output of Type III Mixture . . . . . 9 4.2 Comparison -Bet2,.veen Charges ... . . . . . . . 9 5 DISCUSSION OF POTENTIAL FLASH BLINDNESS EFFECTS . . . . . . . . . . . 17 5.1 Permanent Eye Damage . . . . . . 17 5 .2 Recovery Time to Light Flashes . .2 . . . . . . 19 5.2.1 Review of Flash Blindness Experiments . . . . 19 5 .2. 2 Additional Observations . . . . . . . . . . . . 22 5 .3 Interpretation of Results . . . . . . . . . 24 5.3.1 Illumination of Pho'Loflash . 2. . . ... . . . . . 24 5.3. 2 Estimated Eye Effects . . . . . . . . . . . . . 24 6. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . 31 6.1 Cartridge Design . . . . . . . . . . . . . . . 31 6. 2 Pliotoflash Mixture . . . 0. . . . . . . . . . . 31 6 .3 Data Intorprctations . . . . . . . . . 31 REFERENCES . . . . . . . . . . . . .. . . . . . . . . . . . . . 32 APPENDIX I . . . . . . . . . . . . . . . . . ... . . . . . . . 33 1 INTIIO))IICTIOlq The objectivo of this investigation of pyrotechnic light sources was confined primarily to the evalucition of output illumination levels, the factors controlliiig them and comparisons with the existing criterizi for 2 optical incapacitation, since tlicse considerations determine ilie suit- ability of such a system to a variety of possible weapon applications. The effort was directed primarily at comparisons of existing threshold energies for incapacitation and damage, pyrotechnic output energies, and the'fabrication of test device-s with controlled parameters2, followed by estimation of their output energies by means of appropriate instrumentation. The approach permitted a comparison with existing data on energy levels corresponding to optical incapacitation and damage. this comparison was used to obtain first order estimates of the effective. ranges at which these pyrotdchnic illumination devices would-either 2 incapacitate or permanently damage exposed personnel. The four pyrotechnic light mixtures sho'@vn in Table I were evaluated during this investigation. In the follo@ving sections there are presented the methods and techniques employed, results obtained, and-the con'clu- sions regarding the feasibility of this approach.. -TABLE 2 0-r@EXPERIM-ENMt- C omposition Des.ignation A B c D Inqrc!dients % % % % Al 4b 70 25 50 KC103 30 7 20 30 40 BA(NO 3)2 30 10 10 Ca(Si) 10 2 M g 35 saw 2 DESCI'11"fION OF I'IfOTO'I:".f,,'@Sll J--)T@',Vj(:,ES A brief description of the charges was included in the Final Sumiiiiry Report - Part I. Two basic Pliotoflisli cartridges were empl2oyed. Design drawings of these cartridges are presented in Figurc-s 1 and 2. The per- forniance of each of the four selectcd pyr -otechnic mixtures xvds evaluated in cich cartridge design. The specifications of eacli of the miytures' were previously described (see Reference 1) . 2. 1 PHOTOFLTkSI2-j'. CARTRIDGE, TYPE S The outer diameter of the Type S photoflash cartridge is 2.7 in. and the length is 6.6 in. The outer shell is made of aluminum. A scale drawing of the cartridge is shown in Figure 1. A linear charge of DuPont PETN prima;:;ord, .22 in. diameter and 5.25 in. long, was surrounded by the photoflash mix. This centra2l bursting charge was ifsed to break the casing, and to ignite and disperse the pyrotechnic mixture. 2.2 PHOTOFLASH CARTRIDGE, TYPE SDE This cartridge was idenlical to the type S with the exception th-at DuPont Deta -sheet-explosive-was wrapped around the, exterior-cylindrical wall of the Type S cartridge. These sheets were cut, as show2n in Figure 1, to permit the complete coverage of the outer wall. The purpose of this nated, in an attempt to improve the light output (i.e. , as discussed in the last progress report, the rate of burning and light output should increase with the degree of charge compaction) .. The ignition train was designed so that the Deta sheet would be detonated before the central 2 bursting charge. 2.3 PHOTOFIASH SHELL, TYPE SS The Type SS shotgun shel I has an O.D. -of 0. 853 and a length of 3.20 in. It is also made of aluminum. Except for dimensions, this photoflash shell design is similar to the '&.7 in. photoflash cartridge (see Figure 2). 2.4 PHOTOFLASII SHELL, TYPE SSDE9 This shell is identical to the Type SS, with the exception that 'Dupont Deta sheet is wrapped around the cylindrical body in a similar manner as the -SDE.- 2 7Ai)APT--P (-r%fP) EN ri C-H@:i:T S PAC F- R I K I I.Io- r- Si-ir'E- '-HARI-,' CLOSUPt:"- CASE, -St@ lt@ZI-I 2 W cr:A T H v-,R i< 113 I C Pt WALL SP!-.- E A (-ryR) '5K I 11@) 2 (SDE DESIGN OIK.-LY),,,,. tt L k"k 'E3 PA- 7, C u S2!q Io,\'. - L3 & ;:=@ k,-" II 4,- K J PI 'TueE (T\@ -LI,'-rLAP, C--@IARGE (-r P) 1-< I I'Z7 I I -z e, TUBEILI@IEAR'@HA@Q%GE OUTPUT CHAPGE. FIGURE 1. 2.7@" PHOTOI'-LASH CAPTRIDC-9[-: CAL br.. I / I rAc T U -0 II' R'@ D t't P T E :ai@) I 11 S P/@, C L SKIII;'-- 5EAL C L t) 2 CA'@-)E (SSDE DESIGN OLN- LY) 0..U',rPUT C14A.RGF-- L 2 Cr r!j, %.. A.. r LtNr,AP, CI-IARGE 2 4AR&E 21 C L 0 5 U R E , 'I" I,- C' @,.K I I @T- Y P.* FIGURE 2. PI-AO-%C)7FLf-\r5lA C A P-'T 14 D -6 E. 3 C)j" The c:,:poriiiientzil tccliriiqiics employc@d and the sl)c.@cific experiments performed arc otitliiie-d iii tliis- s--ction. 3.1 EXPERIMENTP.L AP-,rlvl@NIGEMENT 2 The intensity of the liglit output as a function of time w&s measured and recorded using a silicon pl-iotodiode (United Detector Technology Inc., PIN-10) and Li Tel-,tronix 535 oscilloscope with a Polaroid filra attachment. The PIN-10 photodiodc has a half-value response between 3800 and 1050O X. The photodiode, PIN-10, was protected from fragments generated from the e:@-.plosive dissemination by a metal shield. The light from the flash was indirectly focused on-to the PIN-10 sensor using a mirror. A Corning 1-56 filter was placed in front of the PIN-10. This filter has transmission characteristics similar to the eye response. It has a 2 near gaussian trans m@ssion curve between 3600 and 7000 X,with a peak transmission at 5200 A. The ultraviolet is cut-off at 3600 A and less than 10% of the I-IZ radiation,(i.e., between 1 and 4.5 micror-s) can be trans- mitted through this filter. Depending on the expected output of each test 2 device, nolitral density filters were also used so that the PIN@10 would operate within its linear response regime (i.e. so that light intensit)@ versus output voltage would remair, linear) -%-;as fed into a Tektronix 535 The output sign,-:11 from the sensor 2 oscilloscope with a 50 ohm tormin;ltion. The signals were permanently recorded using a Polaroid camera attachment. The oscilloscope and event were triggered using a 5 KV firing panel. The events were also monitored photographically. A speed graphic single exposi-ire camera and a.Beckman & Whitley Dyiiafax motion picture came2ra were used with Polaroid filters. The Dynafax camera was operated at a framing rate of 3000 -frames per second.. Both cameras were protected -hind a barricade and from the blast. The Dynafax camera was located be received the light from the photoflash via a frbnt surface mirror. The speed 1graphic camera was located behind thick glass. 3.2 DETECTOR CALIBRATION The PIN-10 photodetector was calibrated using a Nation al Bureau of Standards certified light solirco. The response of the detector was measured as a function of distance away from the standard source using the following equation K- F V D where I is the intensity of the light source in candle-pov@er, K is the calibration factor, 2640 foot-candles pcr 38. 07 millivolts response, F is the factor wli ich compensates for the transmission of the n2ouaal density filters, V is the output signal, volts, generated by the photo- detector, and D is the optical distance between the PIN-10 photo-, detector and the light source. The spectral response characteristics of the photodetector are shown in Figure 3. The relative spectral sensitivi2ty of the PIN-10 with the Corning 1-56 filter was calculated. These results are presented in Figure 4. The optical distance between the light sources and the photo- detector was 23.5 feet ii@i dach'experiment. The intefrisity of the source expressed in candle-power was calculated using Eq. (1). The'values of 2 candle-power estimated can be interpreted directly in terms of the luminous flux traveling through a normal plane which intersects the light path I foot away from.the source (i.e. , the density of luminous flux incident on a normal surface one foot away from a 1 candle-power source is 1 lumen/ft2). 3.3 EXPERIMENTAL ERRORS 2 An experiment was performed to determine the amount of light received by the photo-detectors via wall reflection. A General Electric No. 22 photoflash bulb was placed 23.5 ft. away from the photo-detector @a sh pyrot in a similar manner as were the photofl, echnic test devices. Black 2paper was placed directly in front of the flash bulb to prevent direct light transmission to the photo-detector. When compared with a control test in which the light bc-irrier was not used, it was calculated that less than 5% of the light received by the photo-detector was due to light reflections from the walls of the test chamber. As a precaution, in subse- 2 quent tests, all light colorr-@d objects were removed from the test chamber -iiiiber wc-re washed down before each experiment., The walls of the ch, after each test, also, to remove any debris which might increase the light reflection. 6 7 lpi Wcivelength, p 0.2 0. 3 0. 4 0. 6 0.8 1. 0 1.2 1.00 100 I I .- go 2 Relative - 80 0.80- Sensitivity 70 0. 60 - 60 U') 2 Quantum.------"C- Efficiency so CD 0.40 40 0 30 >1 2-4 4@ o 14 /...n-Absolute Sensitivity > 4- 0 --4 >1 0. 20 20 >2 0 @Q CD lo (D 0. 10 CY 4-J 9 2 O. 08 8 7 0. 06 - 6 0.04 - 4 2 3 0. 02 2, 000 4, 000' 6, 000 8, 000 10, 000 12, 000 0 Wavelength, A Figure 3. 0 Spectral Char,-ictcristics of the United Technology PIN-10 Photodiode 7 1. 0 0.9 Rela f4.ive Response @*--Relative Response o Eye Eye (Day) (Night) 0.8 0.7 0.6 r 2 elative Sensitivity f PIN-10 Phot odiode W, 0.5 h Corning 1-56 2 Filtter co C.) > '4'J 0.4 ri 0.3 0. 2 0.1 400 450 500 bbu 600 6 Wave Length (m P) Figure 4. Comparison of Eye. and Photodetector Sensitivity Curve. Elko 4 P,.UO,'Ujr.TS -------------- The MoOc-urccl out Puts versus tilne for each Perimental photoflzisli dcvic,@-,s are 8]io%,,,n in rig of the Lx re urcs 5 tlirough 8. I'llese figures we drax,..,n from t],e oscilloscope traces ol.)tained in ach experiment. The results of each photoflash cirtridge design wore combined so-tliat a.. 2 ready coinp,-Irison could be made regarding the relative perl'Ormance of the light output from each of the four pyrotechnic mixtures. The peak light intensities, the half width of intensity-time and the total integrated area under each curve (.i.e., total light output in candle-power-seconds) are reported in Table 11. T2he maximum cloud size generat.ed by each -device is also included in this table. The -latter data was obtained irom the film record@ taken. 4.1 COMI)APISON DENVE-EN REPORTED AND MEASURED OUTPUT OF TYPE III MIXTURR Mixture A has a composition identical to the standard Type III (I e. , mixtur2e P-@:) photofla-sh mix. This Type III plioloflash mixture has previously been used in a variety.of standard photoflash cartridges. Light output data was obtained for this mixture,' from other sources, and is summarized in Figure 9. . The peak inten.-,ities'and integrated candle-po er-secon@is total output are plotted as a function of char ie 2 g ze in pounds. Approximations of best-fit curves were drawn through each set of data as shown. The arrow on the abscissa represents the charge weight of the pyrotechnic mixture s in the Type S cartridge, 1. 4 Ibs'. The predicted peak 'ntensity and total output for this charge are, respectively, 2.3 2x 108 c. p. and 4.4 x 106 c. p. s. The experimental values obtained durin§ the tests- on this program were 1. 96 x 108 c .P. and 2. Ox 106 C. p s. (see Table II) . The7se data are considered to b e in rea- 2 sonable agreement and provide an additional check on the accuracy of the measuring techniques and the charge preparaitions. 4.2 COMPARIS(D'N 'BTTWEEN CHARGES Based on the peak intensity data', the overall performance of the Ty,pe A mixtures in.the various shells was the best. In order of decreas- Ing performance it was found that A > C >3 B > D The peak intensity of the output however is not the only criterion that should be used in evatua'&,ing the perfori,,ianc ' e of a mixture for this appli- cation. The dureition of the light pulse is of equal importance. From 9 200 180 - SA A, B, C and D refer to the type of pyrotechnic mixture used (see Table I). 160 - 0 140 120-- C2> SB x ioo SC >1 8 0 60 40- 20-; SD o 0 2 4 6 6 1 0 1 2 1 4 1 6 18 20 2 2 2 4 2 6 28 30 32 34 3 6 38 40 42 Time, milliseconds Figure S. Light Output From Series SPhotoflash Units A, B, C and D refer to the rotechnic mixture used (see py 30 Table I). 25 2 0 SDEA 20 to x >1 SDED 41 SDEB S DEC lo lb 2 41b ,10 0 2 4 6 8 10 12 14 16 .18 20 22 24 26 28 30 32 34 36 38 40 42 Time, milliseconds r- I citp*e 6lirjht Outyput F'rom 'geries 4grpr- 'Pl@otaflosh ttv@it4 I T 4.8 A, B, C and D refer to the 4. 4 pyrotechnic mixture used (see 2 Table I). 4.0 3. 6 3. 2 u to 2.8 C) 2. 4 'SSG 2. 0 1. 6 S SA 4J 1.2 2 SSB 0.8 SSD 0.4 ....... ............ ........... ",SSD ......... 6 0 0 1 2 3 4 5 6 7 8 9 10 'll 12 13 14 15 16 17 1C 19 20 21 Time, milliseconds Figure 7. Light Output From Series SS Photoflash Units 4.8 A, B, C and D refer to the SSDEA pyrotechnic mixtured used (see 4. Table I). 4.20 0 3. 6 3. 2 CD SSDEB 2. 8 SSDEC x 2 . 4 2.0 1. 6 i-4 S'S D' ED 2 1.2 0.8 0.4 0 10 11 12 13 14 .15 16 17 18 19 20 0 1 2 3 4 5 6 7 8 9 2 1 0 Time,'millisec.onds Figure B. Light Output From Series SSDE Photoflash Units 'rAl3LL@' 11. SUj"lilVil@,T',Y 01@ TEST iIE@-!ULTS Sample Pc-a k I-ii I f t Totiil Poak Type C,.-I II @ic-Po%,jer Duration 2 Output Cloud Size (1 0 cp) (msec) (1 0 3 lft - --------- SA 19 5. 7 9.8 2010 7.4 S B 9 7.8 15. 5 2 1579 7. 5 SC 10 3. 6 17.8 2 02 1 7. 1 S D 9. 5 40. 4 371 3i5 S DEA 29. 8 20. 7 52 6 7. 0 2 SDEB 19. 5 20. 1 375 7. 0 SDEC 20. 3 18. 3 49 3 7. 3 SDED 20. 5 16.4 354 4. 3 SSA 1.99 4. 0 2 10. 3 --- SSB 0. 99 6. 0 7.8 1.8 ssc 2.81 5.4 17. 1 2.4 SSD .0. 57 20. 0 8.8 1. 5 2 SSDEA 5. 59 8. 3 52. 9 2.8 SSDEB 2. 64 9. 1 29. 0 3. 5. SSDEC 3. 02 12. 7 40. 2 3. 2 SSDL--D 2. 714 13. 0 28. 6 2. 7 *S and SDE refer to the type of photofla-sh cartridge used. A, B, C and D refer to the pyrotechnic mixture used (see Table I). 14 Aik, whom 100 1000 Peak Light Intens 0 ri 2 C> Z<- Integral Light Output x C:> 0 41 .,-I 100 2 lo I I fit' -I' f I 1-1 III I I fill lo, 10 100 Powder Weight, lbs. Figure 9. Peak Intensities loduced by Typ0e III Photoflash Mixtures as a Function of Charge Weight it V,7aS that tile light pulses fro@ -the plots of iiiteiisity-vcr-us tinic the "C" mi.,.tLirc, were in niolt cises of 'oiicrcr (ItirzLtion. Based on the total ii-itegratr-,d light olitput t2he "C" and "A" mi>,.tLtres were equally effective. i.e. C -- A > B > D It could be concluded that of the mixtures studied, the "A" and "C" mixtures providc-d the most intense and total light oti+.put. The "BI, and "D" mixtures g7oneracted, sigi-lificantly, less light in the visible spectrum. 16 5 DISCUSSJO'.,\' 01'1-'OTEI,@Tl@'l@L rlltsl-l 1.,LINDT@ISS EFRECI'S Several. important questions bacl to be answcrc,-d before these data could be interpretc@d in terms of potential flzisli blindiess off,-.cts: 1.2 What is the level of flasl-i luminosity that will produce permanent eye injury? 2. What is the level of flash luminosity that will produce. transient flash blincliiess? 3. If the measure of the transient flash blindness is expressed in terms of functional requirements 2 calling for viewing and resolution of specific objects, how does the luminosity of the objects to be viewed bythe observer affect recovery time from non-injurious flash blindness? S. 1 PERMANENT EYE DI@.MAGE It has been reported that the threshold energy level for permanent eye damage is between 02.2 and 1.6 cals/cm2.3-6 Zaret,3 expresses the requirements for permanent eye injury in terms o'L the fraction of the photopigments which are bleached by the light flash. This bleaching process involves the ph-otochemical transforma'tion of 11-monocis retinene to,trans-retinene. 'The 11-7monocis retinene complexes with th2e opsin enzymes to form the active photosensitive pigments. Upon light 'excitation the 11-monocis olefin is transformed to the more chemically stable trans isomer via an electronically or vibra- 7-9 -retiziene, a yello,,,@,, pigment, which tionally excited state. - The trans' is formed is not compatible w2ith the opsin enzyme and, thus, does not form a photosensitive pigment.' Recovery of the bleached pigment is dependent upon a chemical transformation back to the 11-monocis isomer cc-talysed by retinene isomerizase. The concentration of visual pigment only begins to bd significantly affected by light intensities of the order of 105 troland-seconds*2 and decreases rapidly with further increase in intensity. The concentration of bleached pigment as a function of light intensity can be expressed as follows' *The troland is the unit of retinal illumination. It is equal to the product of the lumiriance o' the object viewed in candles/(nicter)2 and the area of th1e pupil in mrn2 17 -lemon Cb /Co 1 - exp ('-.ay It) (2) whcre c ure I b is the concentration bicachcd at expos c0 is the origiiial pigment concentreti on, I t is the retinal total irradia2nce in trolar;d-seconds, and ay is the photosensitivity expressed in (td-sec)-l In the case of the human pigments the value of a, y is approximately 10-7 (td-sec)-l. Zaret estimates that as the fraction of pigment bleaching approaches unity permanent retinal damage, occurs. Within the time frame of the flashes which2 were produced in the experiments reported here, the 2 (4 x 1 01 td-sec) threshold damage irradiances are approximately 0.4 cals/cm and 1.6 cals/cm2 (1.6xlOlO td-sec)10-32 fcr,- exposures of I and 100 msec, respectively. According to Br2o,%,,@'n,' these irradiance levels must be delivered at flux levels of at least 0.7 cals/CM2-sec or the rate of heat dissipation in the eye tissue will be sufficient to prevent an elevation or temperature to the degree where thermal burn will occur. At 5550 angstroms the wavelength of maximum'sensitivity to the eye, 2 one watt of radiant energy corresponds-to 672. 1 lumens. Assuming that.all of t@e light emitted from the pyrotechnic flash devices tested during the program is at this wavelength, the light intensities required to affect thermal damage to the retina would be 1. 75 and 6.99 x 104 lumen-sec/ft2 for exposures of I and 100 msec, resp2ectively. The above estimates take into account the fact that the light received at the cornea is-intensified when it arrives at the retina (i.e. , the imag-- size is reduced): Ham' noted that an irradiance received at the cornea of a rabbit eye is intensi- fied by a factor of 60 times when it reaches the retina. With respect to the experiments perfor2med in this investigation, these values are considered to be lo-,,v. ;issuming that the light emitted from the flash units'have the same spectral characteristics as the sensor, than the luminous efficiency of the light output is only 28 percent that of 5550 ang- strom light. The threshold light exposures would then be 6.26 x 104 and 2.48 x 105 lumen-sec/ft2 for exposures between 1 and 100 msec, respec- tively. Based on the luminous intensity data, sho%,in in Table II it can be seen that the small photoflash charges (i.e. the SS and SSDE series) are cliarges not czipcblc of dcliveriiig diimz,.cing light flishes. The 1,-irger (i.e. the S aii(i SD'i" series) czin produce permlinciit eye damage. Further discussions rcgarding the possibility of permanent eye2 injury are presented elsewhere in this re,,.)ort (see Section 5.3.2). 5.2 REC O@n7ll'Y TIME TO LIGI-ITrLAsi-irs Before oiie can estimate the recovery period after light exposure it is important that the important recovery measurci-nent conditions be defined. Clearly,2 the intensity of the light which the observer is exposed to will determine recovery time. An additional consideration is the illuminance of the object which the observer needs to detect after exposure for func- tional reasons. A- review of the literature was made in order to define the.dependence of recovery on the two factors noted above. The resul2ts of this search and the subsequent analyses are shown in Figure 10. 5.2.1 Review of Flash Blindness Experiments Metcalf and Horns conducted flash blindness experiments using the high intensity flashes from a carbon arc. The experiment was designed to determine the effect of light exposures li'Al-lely to be encounter2ed during nuclear operations. Each of the four subjects had their pupils dilz,,Lted prior to exposure. A 6mm artificial pupil was used in order to maintain constant pupil size. The sub ects were exposed for 100 msec to illumina- .j tion ranging from 70 to 12, 000 lumens per square foot. Fo2llowing this exposure, the subjects were required to detect the flashing of a 17 minute visual angle circular patch. The luminance of the test patch was varied between .07 and 71 foot-Lamberts. A summary of a complete set of this data at a flash luminosity- of 5'x lo5 iumen-sec/ft2 is shown in Figure 10. The time require2d-to recover visual sensitivity following exposure to high intensity, short duration adapting flashes also has been investi6ated by Chisum and Hill.r-' , 14 Adapting flashes of 33 to 165'gsec and 9.8 msec tion with luminances from lx 104 to 5x 108 lumens/ft2 in dura 2 were used. Visual sensitivity was determined by the resolution of gratings requiring acuities* of 0. 13 and 0. 33 at display luminances between approximately .004 to 200 millilamberts. The 0.33 acuity level requires the function of cones while the 0. 13 acuity level can be resolved by rod vision. The light pulses used by Chisuni and H2ill which best represent the flashes *Acuity is defined as the relative ability of the visual organ to resolve detail. It is usually expressed as the reciprocal of the, minimum angular sepiration in minutes of two lines just resolvable as separate. -Woo"" Values Associated With Each Curve are in Units of 2 Lumens-sec/2ft 100 6, 14 5 x 1 0 5(2) (1) Chisum and Hill 6, 13 2 (2) Metcalf and Horn 90 (3) Severin6 80 5 x 1 01 (1) The vertical dashes correspond to the four display lumina2nce levels shown in Table III. 70 60 50 0 40 4 IXIO 30 20 3 5xlO 1) 2 3 89-lxlo 10 .......... (1, 3) ......... 0 7 7 ri5-- 7 -i- 7 10-2 lo-' loo 101 102 Display Luminance, millilamberts (mL) from the pyrotechnic devices tc-..qtcd (luring this program were selected "or the comparisons made in Figure 10. Also the data for the acuity level 0.33 v,,as uscd, since the effects to cone vision are the most critical to this study. It should be noted that recovery from ro2d saturation is a much faster process than recovery from cone saturation. The latter also requires more energy for satitratioii.3 It was observed 2that the recovery time.- for the light illuminations of 5 x 105 lumens-sec/ft , reported by Metcalf and liom, and 5 x 104 lumens- 2 2 sec/ft , reported by Chisum and Hill were almost identical. This is not 4 t.3 too surprising after one reviews the discussions by Brown and Zare Namely, both postulate that the relation between the energy of an adapting 2 flash and recovery time for a specific visual task is similar in nature to that sho-,.vn-in Figure 11. 100 80 - 60 - 40 - > 2 >20 - 0 a 0 1 Reaction Time Log Adapting Flash Eriergy Figure 11. A hypothetical curve illus-Lrating the relation between energy of a blinding flash and time required2 for detection of infor- mation in a visual display. The minimum detection time at low flash energy corresponds to visual reaction time. Detection time approaches infinity as flash energy approaches a value which will cause irreversible injury. For very low adapting flash energies there is 2very littlet if any, effegt on the visual capability, and recovery time is minimal. As energy is increased, there is an increase in recovery time at an increasing rate. The form of this function depends on the nature of the visual task. As the energy of the adapting flash reaches a level wliich corresponds to a maximum possible bleaching of the2 photosensitive pigments of the retina, the rate of increa-se of recovery time may be expected to decrease. It is postulated as shonvn in Figure 11 that recovery time may actually assume a constant value over some range of adapting flash energies beyond that at'which maximum bleaching occurs. *Adaptii5ig flash energy usually refers to the total energy to VIliich the subject is exposed and from -%vhich the subject must recover normal vision. t2 It api)czlrs tlizit the data o' Cliisuri Znd ljjll -It Sx 104 luircnc,-so /f and 1\,Ietco-If and Ilorii at 5 x 105 lumcns-sr-,c/ft2 cz:ii be exple,,iiiccl by this 2 quzilitt-,tive cirguiiient. Althougli there are some differences between the individual experiments, the very close co.-relation in measured recovery times, together with the fact that the saturated bleach levels are being approached or perhaps' exc-eeded@n both, indicate that the intensity p2la- 4. sec/ft2. tea u is in the rL, iige of 1 0 ' to 1 0 lumen- The data reported by Chisuin and Hill at the illumination level of 1 x 103 ltimen-sec/ft2 was extended as shown in Figure 10 using data reported by Severin." Severin reported re2sults in which the display luminance of test patches were between 0.06 and 0.013 foot- amberts. The results obtained ,-t an adapting flash intensity of 8. 9 x 10 lumen- sec/ft2, very close to I x 103 lui-nen-sec/ft 2, were used to extend the Phisum and Hill data. One reservation about Severin's data is that the duration o2f the adapting fla.shes was 150 msec. This is slightly greater than the normal blink tifne of 100 msec. Nonetheless, Severin's data does appear to agree with the general trend at high display luminances found by Chisum and Hill. 5.2.2 Additional Observations' 4 to 2 An experiment was performed by Brown determine the dependence of recovery time from flash blindness on the luminance of objects viewed. The test subjects were exposed to' light intensities of 3 x 10-2 to 3 x 102 ft-lamberts for durations-of 0.9 seconds. The display consisted of a - grating pattern and observers were req@iired to identify its orientation. A - 2 timer was started and the grating was illuminated with presentation of the flash. As soon as the grating orientation was defected, the observer depressed a switch that turned off the timer. Detection times were recorded only for correct identification of grating orientation. The results of this study are reproduced in Figure .12. The families of curves in the upper part of t2he figure represent results with a grating, the individual lines of which subtended a vis@ial angle of 3.8 min. According to Brown, the detection of the orientation of this target display depended on cone vision. The curves in the lower two graphs were obtained 'for a grat-.:Lng which represented a visual angle of 12.5 min. Rods may serve in detection of the orienta2tion of this coarser grid. It is clearly shown, again', that recovery times are dependent on the luminance of the object which the eye attempts to detect. There is one major uncertainty, regarding the quantitative nature of Brown's data, hoiv- ever. This experiment wis performed by exposing the test subjects to a 0. 9 sec light pulse duration. Si7nce the blink time is of the order of 0. 1 soc it is zcjssible that the observers viere not exposed to the full d'uration 22 L-Y, Vistiai i-,@-cui' -.,@ 0.26 60 - 60 5.0 2 so - 4 5 lo; ti-L so- 4 0 log lt-L t) a 3.5 lol fl-L 0 o 30 log 1,-L 40 2 40 0 2 leg fl-L 30 30- 4J 0 -4 20 - 4a 20 2 10 - I o 0 0 -2.5 0.5 2.5 -2.5 2 -0. 5 0. 5 2.5 Log Display. Luminance Visual Acuity 0 08 6 0 6 0 2 5 0 5 0 40 - 40 ,n 30 - 3 0 2 20 c 20 0 10 0 1 0 0 2 0 r -2.5 0.5 2.5 -2.5 0.5 2.5 Log Display Luminance Figure 12. Relations of Perception Timp- to Log Display Luminance 3 (ft-L) for Each of 6 Adapting Flash Luminances4 23 of the light pulse. Tlicrcfurc thcrc is an iii-icertai,lity in the total exposure duration of cacli subjcct to tl-ie light. Thcrofore tliesc dati were not used to correlate results obtaiiied -in the prc-scnt study. 5.3 INTI:RPRI-:TATIONorRESULTS The me2-2surcd light iiitensities and durations were interpreted in terms of possible irreversible and revcrsi))Ie eye effects. 5.3.1 Illiimination of Photo'Llish The experimentally measured illuminance of each pyrotechnic test device as reported in Table II wzls estimated as a function of distance from the flash origin using the inverse square l2aw. These estimates are shown in Figures 13 and 14, The expected illuminance which observers would received from exposure to;@ G.E. No. 50 flash bulb ar(; also shown in these figures. The value for the total output of this -flash bulb, 1 x 105 lumen- sec/ft2, was obtained from General Electric specifications. 5.3..2 Estimated Eye Effects 2 Some of the iso-illuminance curves shoivn in Figure 11 were extra- polated to a d'splay illuminance level of 0. 1 millilamberts (or 0. 093 lumens/ft2). The recovery time for each of the reported ada-oting flash energies shoan in Figure 11 were estimated for each of.four display lumi- nances; 0. 1, 0. 2, 0.5 and 1. 0 m2illilamberts. These estimates are tabu- lated in Table 111. TABLE III. ESTIMATED P,.t,'@C;OVERY TIMES AS A FUNCTION OF ADA@TIN G LIGHT ENERGY AND TARGET DISPLAY LUMINANCE Adapting Flash Energy -Recovery Time (sec) (lumen-sec/ft2) Target Luminance 2 (millilamberts) 0.1 0.2 0.5 1.0 s x los 82 55 35 26 1 x 104 47. 31 18 @12 5 x 103 24 17 12 9 7 .9 - I x 103 8_ 5 4 3 24 Curves iclenticil for SD tind SDED A, C, S and SDE refer to the type of pliotoflash cartricige used. B J A, D, C and D refer to the pyrotechnic mixture used 106 (see Table 1). 2 A C; B D 5 10 C,4 2 4-A 4 1 0 4 3 I 0 2 Lj 10 3 2 101 100 10 130 Saparation Distance, ft. Figure 13. Light Illumination From Type S and SDE Photoflash 5 10 SS ar-ci OOSDE refcr to the photo'Llzlsh c,-irtridg@-, used. A At B, C aiicl D rr-,fcr to the pyroteclinic mi>,ture used (see Table 1). 2 B,D c 4 10 A s B, D 4-2 3 2 10 fA4 2 10 4 10 10 3 2 101 0 1 0 I 0 4 Separation Distc-ince, ft. Tlicsc cs'Liiiiz@tcs -ire clso iiidico.L(-,cl. by the vcrticz',l dashes crussing eacli of the curx;cs iii FigLirc 1 1 For eacli pyrotochiiic dovice tested, the recovery times for a stibjc-ct -isli were estima2ted using the data iii Figures 11, 13, c>.poscd to the light fl, and 14 and Tcible III. These estimatcs were made as a functibn o'L distance away from the flash and the luriiinaiice lovelz of objects which the observer might attciiipt to detect after exposure. These estimates are reported in Figures 215 and 16. As expected the large photoflash devices (i.e'. , the S and SDE series) should be the most effective as far as separation distance is concerned. An obsc,,-rver separated from the flash by 50 feet can be affected if an SA charge is employed. At separation distances less than 7 feet there would be the possibility that irreversible e2ye damage could be affected using the SA charge. Significant flash blindness effects are exp8cted within this distance range for all oi the charges. Recovery times of as long as 60 seconds are predicted for the detection of objects which are very dimly illuminated. The effects of exposure to a G.E. 50, flasl-i bulb were also 2predicted., It can b@- seen that the estimated effects are not as great as for the S and SDE ser-i-es photoflash units. By further comparison with a recent report by Tiller et al. Ir' (ARPA Contract DAAK02-69-C-0338) the estimates made for the G.E. 50 flash bulb appear to be reasonable. Tiller et al. evaluated 'the effects of exposure to this flash b2ulb to subjects performing mili'Lary tasks. The subjects were exposed to a flash at distances between 6 and 19 feet. After exposure the subjects were required to detect ground emplaced m'ines or detect and fire upon a test target. All of these tests were performed under ;@,arious night time conditions to which the subjects' had adapted before being exposed to -the light flash. It was 2found that the subjects, all trained Marines, were able to resume their assigned task with the sanie efficiency after an average recovery time of 5 to 20 seconds. No indication of reflected luminances of the objects detected were made. It is felt, however, that the predictions of recovery times for dimly lit displays (viz., 0. 1 and 0. 2 millilamberts) agree with the results obta2ined by Tiller et al. Between 6 and 10 feet it is predicted that'exposure to the G.E. 50 flash should take approximately 8 to 18 seconds. No predictions beyond 10 feet for the G.E. 50 were made because of lack of data. How- ever there is much indication to suggest that at longer distances (i.e. , lower flash energies) the recovery times versus distance dec4reases at a very small rate. 27 Target Display Luminance (mL) 62 90 3 26 Expected iye Damage (ie. at distances 80 corresponding to the intercepts of the rve) 70 arrow and each.cu 2 S and SDE refer to photoflash 60 cartridge design. A, B,C and D refer to pyrot'echnic 50 mixture used @see Table 7). >1 31 18 12 'Po :> 0 40 IY. 30- 17 12 9 20 10- 5 .4 3 D, B,C A D, C, k (PDE) S 6 7 8 9 10 0 4 0 50 3 2 3 4 5 2 0 Senaration T)Istcinc,-e. ft. The estiinated re.covory tiiiies after exposure to the smaller photo- flasli sources Lire sho,,vn in Figure 16. It is not ariticipatc-.-d that eye damage could be if'Lectcd by those charges even at short separation distances. Again the "I%." and 'IC" iii@l>.tures are expected 9to be the most efficient. 29 --MINN p ------ Target Display S and SDI"-' refer to photoflash Luminance (ML) cartridge design. 2 .5 1 .1 A,B,C and D refer to pyrotechnic mixture used (see Table 1). '6 5 3 2 80 - Expected Eye Damage 2 70 tn 60 ci 50 0 31 18 12 40 30 17 -12 9- 20 10 ....... 5 4 3 Ss 0 I I ---- I 0 2 3 4 5 2 6 7 Separation Distance, ft. 6. CONCJITSIONS Sixteen pyroteclinic fl,.isli mixtures fragmenting container combinations were tested'. It was shown th:it sigiiificant fl,--isli blindness effects can be expected to result from the exl)osure to these flashes, all o'L wliich occur within 50 mscc. These eff2ects can result by exposure to these charges at distances within a raiige of 50 feet depending on the pyrotechnic mixture and qtiantity, container design, and method of initiation. 6. 1 CARTRIDGE DESIGN The series S and SDE charges (i.e., the 2.7 in. photoflash cartridge) produced the most intense light and are expected to be effectiv2e at distances as far as 50 feet. The external explosive burster attached to the outside of the "S" cartridge was expected to increase the light intensity by compacting the mix before ignition., However, for the larger cartridge this does not appear to have been successful. For the smaller 0.83 in. photoflash cart- ridge the expected trend resulted. The effective compaction by this. i2mplod- ing mechanism probably increases with decreasing cross-sectional area. 6.2 PHOTOFIASH MIXTURE The type "A" and "C" mixtures in all cases generated the most light output. In some cases the "C" mixture produced light pulses of longer duratioli as previously anticipated. . The "B" and "D" mixtures were not as e2ffective. In fact the performance of the "D" mixture was relatively poor. 6.3 DATA INTERPRETATIONS In order to estimate. the flash blindness effects, correlations between reported data had to be made. The results of the analyses appear to be consisteiit with expectation, namely that recovery time is dependent not only on the flash energy but also o2n the luminance of objects which are visually sought during the recovery period. Also the relatively insensi- tive change of recovery time at flash energies which produce 90 to 100 percent pigment bleach was shown in this analysis. It is useful to note that the large light sources used in our experiments (i.e. , the S and SDE series) , prodticed more irtei-ise illumination than the source employed by the Vertex Corporation. Correspondingly, longer incapacitation times are predicted for the S and SDE photoflash units as cornpared with the G.&. 50 photoflash used in the Vertex studies. In additioti, on the basis of our independent experimental data, we could predict the shorter incapacitation times reported by Vertex6 for their weaker light source. 31 REFER,ENCES 2. Engi icerinq Design Hrincli)ool, series "Part One Theory and. Application" Al,,IC; pampILt AIVCP 706-185 (Apr. 19 67). 3 . M. M. Zaret and G. M. Grosaf, "Visual and Retinal Effects of 2 Exposure to High Intensity Light Sources" in AC.,AT-ZD Conference Proceedings No. 11 Loss of Vision From Hiqh IiitensitY Light J. - NATO (Mar. 1966). 4. J. L.- BroNvn, "Experimental Investigations of Flash Blindness", Human Factors (Oct. 1964) p2. 503. 5. D. W. DeMott and T. P. Davis, "Irradiance Thresholds for Chlorioretinal Lesions", A.M.A. Arch. Ophthalmol. 62, 653 (1959). 6. J. F. Parker, "Visual Impairment From Exposure To High Intensity Light Sources", Bio Technology Inc. Report 63-2, Contract Nonr- 4022(00), (May 1963). 7. E.2 Borges, "A Review of the Visual Process - A Literature Survey", Technology Inc. (1964). B. R. Hubbard and A. Kropf, "Molecular Isomers In Vision',., Scientific America qan. 1967) p. 64. 9. N. Turro, "Molecular Photochemistry," Banjamin, New York (1967). lo. W. S. H. Rushton, "Dark-Adaptation and the Regeneration of Rhodo.psin2" J. Physiol. 156, 166 (1961). ii. R. S. Weale, "Vision and Fundus Reflectiometry: A Review", Photo- chem and Photobio. 4, 67 (1965). 12. H. J. A. Dartnal, The Vistial Pigment Methuen and Co., Londori: (1957). 13. R. D. Metcalf and R. E. Horn, "Visual Recovery Times From High 2 Intensity Flashes of Light", WAD(@; Tech. Report 58-232 (Oct. 1958). 14. J. H. Hill and G. T. Chisum, "Flash Blindness Protection," Aero- space Med. 13 _,958 (1962). is. R. E. Tiller, R.' W. Garrett and J. H. Cronander, "Selectivity Induced Degradation of Scotopic Vision", Vertex Corp. Repo,-t, ARPA 11--lontract DAAI,'02-69-C-0338 (June 1969). In order to clarify the question of potential zippliczitiois of b,-ight light sources, a series of simple scenarios have been developed, wliich illustrate possible suitable situ,.-itions. The use of a detonating pyrotechnic'permits the generation of casing frac-nion's as well as intense l2ight, if the material is enclosed in a metal casing. On the other hand, packing the pyrotechnic into a non- metallic (e.g. , cardboard) casing, essentially eliminates any significant fragment hazard. These two modes of operation find separate regimes of possible application. PERIhiETER D@T-"FENSE 2 Given a situation in which a village or a group of men wish to provide a very distinct indication of an attempt aL. perimeter penetration, by the enemy, and in addition wish to either inflict temporary optical incapacitation alone, permanent optical incapacitation alone, or fragment damage in addi-tion to the optical incapacitation, these pyrotechn,.c light 2 sources can play a useful role. Thiis, cased in metal and triggered by sensors (or trip wires)' within the effective fragment range, they provide direct fragment damage cap-abilityv!ith a good possibility of severe permanent optical impairment at such relatively short ranges. Triggered by sensors deployed outside the effective f2ragment range, the effects would be primarily tdmporary optical incapacitation and disorientation with a low probability of fragment damage. In specific situations, calling for no fragmentation effects, such as one in which friendly personnel may inadvertantly trigger the charge, the sensors can be deployed far enough away to assure only temporary op2tical incap-mcitation and fragmentation can be completely eliminated with a cardboard casing for the pyrotechnic, II. VEHICLE PROTECTION AGAINST KIDNAP ATTEMPT Given the premise that abductors (e.g. , of South American diplo- matic representatives) do not wish to kill the hostage during the kidnap attempt, a system for providing even 5 - 10 seconds of2 optical incapaci- tation in a 3600 field around the car in which the hostage is driving, provides an opportunity for escape, while the abductors are optically dis- oriented. This system would be more effective at night than in the day- time. The light source could be either pyrotethnic or electric discharge. It could be made safe against accidental discharge causing permane7nt damage to innocent bystanders. 33 DTSRIJFR!ON C).r CONVOY IIY CAU,@)11-.T'--.. Ol'TICliL liNTC-,A.'-'AC!Ti'@TIOI\l OF I,T--'AD DIIIARLR- The scciiario licrc Is relatively simple In that 2the Icad driver can be optically lncap,-icit,-ttcd as lie's rotin,lling a turn, or causcd to block the road by his inability to see it for a sufficient time to cause a wreck. IV. ESCAPE FROLI AN ENCLOSUPE WITI-I NO PB-RMANENT DAIvIA'-IF, TO OTIlrRS 2 In some situations, where the presence of innocent bystanders, e.g. women and children prevents the use of more damaging techniques, the use of temporary optical incapacitation is of potential interest. IIIT V. PRELIMINARY TO__ AI;DIVIDUAL CAPTU,t)%E Whe2re a single individual is to be captured alive, the use of optical incapacitation may provide useful assistance. Thus, a bright light source generated n ear him by impact functioning of a device fired from a shotgun can provide sufficient temporary optical incapacitation to permit other capture techniques to be em@l:oyed more reliably. 2 VI. STJMMP.RY While these scenarios do not provide a complete list of pote tial applications, they should be useful in examining the value of a s ;tem which combines the capability for fragmentation damage, severe pernianent optical incapacitation and transient optical incapacitation with.the choice fairly easy to control. 1 34