4 IUNCLASSIFIED

AD_ 401863

ARMED SERVICES TECHNICAL INFORMWAON AGENCYARLINGTON HALL STATIONARLINGM 12, VIRGINIA

UNCLASSIFIED

NOTICE: 14hen goverment or other dzravings, speci-fications or other data are used for any purposeother than in connection with a definitely relatedgoverment procurement operation, the U. S.Government thereby incurs no responsibility, nor anyobligation whatsoever; and the fact that the Govern-ment may have forimilated, furnished, or in any waysupplied the said drawings, specifications, or otherdata is not to be regarded by implication or other-wise as in any manner licensing the holder or anyother person or corporation, or conveying any rightsor permission to manufacture, use or sell anypatented invention that may in any way be relatedthereto.

I

*I teISC',] pFINAL TEmCHtNICAL, RElPORT"

PRELIMINARY STUDY INTO THE PRINCIPLES OF*, CONTINUOUS TONE ELECTROPHOTOGRAPHY

- June 15, 1962-December 15, 1962

5 12- APR i••3

I'rVA~rML.NT OF UMlNSLt

IContract Mo. DA-44-009-ENG-5100

Research and Development Precurement OfficeU. S. Army Engineers

Research and Development Laboratories

t ~ IrFort Belvoir, Virginia

ca

Americea Zinc, Lead end Smelting Cempany1515 Peel Brawn Building, St. Leais 1, Misseuri

I

T1R VIEWS CONTA!- - ' -- r,.iN RoPUsNIONLY THE V'EW'S C'? -HL A.~k I MAND HAVE NOT UE A V7ROD U MDEPARTMENT O1 THE. APBMY9

of this ropori. LZ ASTIA is limited*.

Preliminary Study into the Principles ofContinuous Tone Electrophotography

FINAL TECHNICAL REPORTAmerican Zinc, Lead and Smelting Company

Contract Nr DA-44-009-ENG-5100

Number of Copies Received - 35

COPY Distribution

1. Chief, Graphics Division, GIMRADA

2.-4. Chief, Reproduction Branch, GIMRADA

5. Directorate, GIMRADA

6.-7. Chief, Research & Analysis Division, GIMRADA

8. Air Force Development Field RepresentativeUSAERDL

9. Canadian Liaison Officer, USAERDL

10. Chief of EngineersATTN: ENGTE-MDepartment of the ArmyWashington 25, D. C.

11.- Chief of EngineersATTN: ENGTE-TDepartment of the ArmyWashington 25, D. C.

12. Chief of Research and DevelopmentDepartment of the ArmyWashington 25, D. C.

13. Headquarters USAFATTN: AFCIN 3C2bWashington 25, D. C.

14. Covmanding OfficerU. S. Army Intelligence Material Development AgencyFort Holabird, Baltimore 9, Maryland

15. Commanding OfficerArmy Map ServiceATTN: Code 6001Washington 25, D. C.

Cotv Distributi,

16. Commanding OfficerArmy Map ServiceATTN: Code 3401Washington 25, D. C.

17. MCLAEBAir Force Liaison OfficeU. S. Army Engir-:,r GIMRADAWright-Patterson AFB, Ohio

18. PresidentU. S. Army Armor BoaruATTN: Chief, Topo BranchFort Knox, KentL.'

19. Commanding OfficerU. S. Army Electro-ic Res & Dev LaboratoryATTN: SIGRA/SL-bFort Monmouth, New Jersey

20. Commanding GenerqlU. S. Army Electronic Proving GroundFort Huachuca, Arizona

21. The Hydrograi"U. S. Navy Oceanographic OfficeWashington 25, D. C.

22. Commanding OfficerU. S. Naval Photographic Interpretation CenterATTN: Evaluatit,, Department4301 Suitland RoadSuitland, Maryland

23. CommanderRome Air Development CenterATTN: RCWICGriffiss Air For-, Ease, New York

24. CommanderU. S. Air Force Aeronautical Chart & Information

CenterATTN: ACDEL-72nd and Arsenal StreetsSt. Louis 17, Missouri

2

p Distribution

25.-34. ASTIAArlington Hall StationArlington 12, Virginia

35. File Copy

Further information concerning this project may be obtained fromMr. Stephen W. Gibson, Acting Chief, Graphics Division, U. S. ArmyEngineer Geodesy, IntelligeL.,e and Mapping Resenrch and DevelopmentAgency, Fort Belvoir, Virginia, phone EDgewater 9-5500, extension 62255.

3

I 7th!I - ---- I- -l '

I IhoI

I z

KU oII. I d 3

jB I BIi '" I+0 0

'dO

I AfloI 1,1 IlvJ~

6'+ + '+.+ 6

I lif :

t7

II F 1z~

IIU * 0

00 0, z i .0 -0 -. N G

t; -46 Or

j-.a stilljo-d

148121Iij fl314

I

Erratum

Page 21, Item 9 should read 7.5, 15, 40, 60 and 100-wattimps and reflector.

Page 48, paragraph 4, lihe 3 should read Bruning No, 32-155Copytron paper.

Page 57, paragraph 2, line 2, last word should be desensiti-sation.

f

I

a

I

U

Ia

Va

I

!

r

FINAL TECHNICAL REPORT

PRELIMINARY STUDY INTO THE PRINCIPLES OFCONTINUOUS TONE ELECTROPHOTOGRAPHY

June 15, 1962-December 15, 1962

Contract No. DA-44-009-ENG-5100

Research and Development Procurement OfficeU. S. Army Engineers

Research and Development LaboratoriesFort Belvoir, Virginia

* -American Zinc Lead and Smelting Company1515 Paul Brown Building, St. Louis 1, Missouri

I

PREFACE

Five scientists and technicians were directly involved inthe work reported here. Messrs. G. E, Mason and HowardArmbruster were principally responsible for evaluation ofthe resins at the electrophotographic laboratory of theAmerican Zinc Oxide Company in Columbus, Ohio. They alsoperformed the bulk of the work on the study of the effectof additives to the coating slurry.

Messrs. Otto C. Klein, R. L. Bergman and G. P. Bollwerkworked in the electrophotographic testing laboratory inEast St. Louis, located at the Fairmont City Plant of theAmerican Zinc Company of Illinois. This facility isequipped in the same fashion as the Columbus laboratory,and the two groups frequently cross-checked results toinsure accuracy of data.

Assisting this group and ready for immediate help whencalled upon were Mr. K. A. Phillips, Director of Researchand Development of the American Zinc, Lead and SmeltingCompany, and Mr. J. H. Calbeck, recently retired fromfull time duties as Manager of Research for the AmericanZinc Oxide Company, a subsidiary.

The commercial plant for the manufacture of electro photo-graphic grade zinc oxide is located at Hillsboro, Illinois,a distance of about 60 miles from East St. Louis. At thisplant we had the direct assistance and cooperation of Mr.H. R. Wampler, Plant Manager, and Mr. 0. J. Hassel, PlantSuperintendent. These men cooperated to the extent thattheir entire commercial facility and their collectiveyears of manufacturing experience were at our disposalwhenever needed. They assisted in the selection of dif-ferent particle size oxides and in the design of a pilotfurnace to produce doped zinc oxide.

On May 29, 1962 a major conference of all interested per-sonnel was called by Mr. G. L. Spencer, Jr., Vice Presi-dent, American Zinc, Lead and Smelting Company. At thistime, he advised everyone of the major objectives and as-signed areas of responsibility. The actual contract per-iod was from June 15, 1962 to December 15, 1962, but work

was not terminated as of that date. Some of the data pre-sented here was gathered as late as December 27.

The author is grateful to Mr. Frederick C. Myers, Chief,Reproduction Branch, Graphics Division, U. S. Army Engi-neers, Geodesy, Intelligence and Mapping Research andDevelopment Agency, Fort Belvoir, Virginia. Through hisencouragement he contributed to making the project a suc-cess, and we extend our sincere appreciation for hishelp and guidance.

!

TABLE OF CONTENTS

Paxe No.Su2mmary*.... ** as * .0 00 . a 1

Introduction . . . . . . . . . . . . . . . . . 2-3

Investigation

IA. Raw Materials . . . . . . . . . . . . 4-11

B. Description of Equipment * e e e e s * 11-22

C. General Operating Procedure . . . . . 22-54

I Discussion . . . . . e e • • • • • 55-59

Conclusions sio. . . . . . 60

Reom.endation. . . . .. . . . .... 61

IIII

I

!

-1-

SUMMARY

This report describes all of the equipment necessary forthe successful production and evaluation of electrophoto-graphic prints.

The two basic ingredients of the electrophotographic coat-ing, namely, zinc oxide and resinous binder, are examinedand the effect of different types of binders are evaluated.

Using a Pliolite binder, the effect of change in particlesize of the zinc oxide is studied. Change in particlesize of the zinc oxide is used as a variable to show thathigher charge acceptance is possible in an unsensitized

r system when small particle size zinc oxide is used. Part-icle size is also examined from the standpoint of its ef-fect on spectral response of the coating, and spectrogramsare presented showing that zinc oxide absorbs strongly inthe ultra-violet and that changing the particle size doesnot shift this absorption curve.

The effect of various additives to the coating slurry arediscussed and their usefulness in improving tonal rangeis evaluated.I

!II

i

1

"-2-

INTRLDUCTION

Electrostatic printing is no longer an unknown term in thegraphic arts field. The production of readable officecopy has already been successfully achieved by the AmericanPhotocopy Equipment Co., Evanston, Illinois, the CharlesBruning Co., Mount Prospect, Illinois, and the MicrostaticsDivision of the Smith-Corona Marchant Co., Skokie, Illinois.These three companies have successful, operating machinesand paper on the market, and we can expect a flood of com-petitors to announce their machines during the next twelvef months.

All efforts by the above "Big Three" have been directedtoward the design of a machine and a paper which will re-

jproduce line copy. As a result, the resin manufacturerand the producer of zinc oxide have been called upon forsamples of their product which, when mixed together wouldyield a black and white copy with the fewest possible shadesof gray.

The problems of the office copy group can be fully appre-I ciated only by one who has worked in this field. Because

it was and is a multi-million dollar market all effortshave been directed toward the design and sale of a suc-cessful product, and the basic research which would givethe answer to the many unknowns along the way simply hasbeen neglected or is locked up in the company file marked"look at this when we have more time".

The U. S. Army Engineers Research and Development Labora-

tories, Fort Belvoir, Virginia, as the contracting agency,r requested the American Zinc, Lead and Smelting Company toapproach the problem of electrostatics from a less hurriedand more theoretical standpoint. The investigation was tocover the effect of several major parameters upon the pro-duction of continuous tone prints. All efforts up to thispoint had been directed toward maximizing contrast. Nowthe problem shifted to black-and-white plus all the shadesof gray possible between white and black.

One military need for electrostatics is in map printingand in the supplying of tactical maps on very short notice.The ability to reproduce as many shades of gray as possible

I

I-3-

is most valuable when time is not available for the pro-duction of multi-color maps.

The usefulness of electrostatic printing for map reprodue-tion using miniature film transparencies has been shownwith the RCA Electrostatic Printer and more recently bythe Harris.lntertype machine.

Our part of this total problem then was to study the fea-sibility of extending the tonal range of the electropho-

-tographic coating, using any and all meang - our disposal.This was to be literally a "stab in the dark" to see ifit is possible to approach the tonal scale acAieved withconventional bromide photography.

The problem quickly broke down into two parts when it wasfound that the resin binder used determined to a great ex-tent the number of shades of gray possible. We thereforeset our Columbus laboratory to work investigating allavailable resins while the other investigators busiedthemselves with lindin the answers to the effects ofparticle sie of the nine oxide, of pi pnnt to resinratio, of different wavelengths of light, and of otherpertinent variables.

!II1

I

A. Raw Mtlerials

The zinc oxide used in this investigation was manufacturedb the American Zinc Load and Smelting Company at theirUllsboro Illinois facility. This plant has built a

French oxide furnace specifically designed for the cormer-cial manufacture of photoconductive zinc oxide, and thesamples described below were made with this furnace.

Table 1

Standard Test Data on Zinc Oxideosod in IUDL DA-4J-009-]NG-5100

Grade ZZZ-771 zzz-661 ZZz-661 ZZZ-661 ZZZ-661Lot No. 2-7863 2-5609 2-5686 2-5668 2-5678% Pb .0015 .0015 .0019 .0018 .0015SCd .0020 .0020 .0020 .0015 .0016

Percent +325 .0245 .0090 .0026 .0036 .0040Brightness 86.25 88.25 88.00 88.00 88.00Grit #31 #11Gardner oil 21.0 17.0 17.0 17.Rub-out oil 11.9 11.7 10.3 10.3 10.3Specif•c surface

(at/am) 7.91 4.31 3.43 3.15 2.64Specific surface

diameter (}) 0.136 0.250 0.314 0.342 0.408

All of the items on Table 1 are self-explanatory with thepossible exception of the last two rows. The second to

st item in the table Is identified as specific surface(m2/p). Our routine measurement of specific surface (insquare meters per gram) is carried out by dissolving afixed weight of sine oxide in a fixed quantity of 10 Ntartaric acid in a calorimeter. A plot of time versustemperature gives an accurate measure of the rate of so-lution of the sine oxide and its attendant liberation ofheat. The procedure is standardized against nitrogenand helium absorption techniques and further chocked bystatistical cotuts on prints made with the electron mi-scope.

I

II -5-

If we assume the sine oxide particles to be spheres, itis possible to calculate a particle size figure whichwe show as the last item in Table 1 as specific surfacediameter (ja). We know, of course, that French oxideconsists n6t of spheres but of needles and platelets,but the assumption is necessary for the calculation.The "specific surface diameter" then refers to a sub-micron size which is the calculated average diameter ofthe zinc oxide particles.

Figures 1 through 5 show electron microscope enlargements(20,O00X) of the five sine oxides used in this project.

The binders used in this work were secured from a numberof manufacturers. In most cases, the manufacturer wasadvised of the part his binder was to play in the program,and a number of samples were submitted. We present belowthe final list of binders chosen for tha work, along with[ the name of the manufacturer.

Table 2

F Resins Used in ERDL DA-4,-009-ENG-5100

Producert sType Code Producer

Pliolit,. S5-B Goodyear Chem. Co.Plioli•,e S-7 Goodyear Chem. Co.Silicone SR-82 General Electric Sil.Div.Silicone SR-111 General Electric Sil.Div.Silicone 840 Dow Chemical Co.Acryli. B-82 Rohm and HaasVinyl Butryal Resin XYHL Union Carbide PlasticsPiccoflex-Styrene 120 Pennsylvania Ind. Chem.75% Piccoflex) 1120 Pennsylvania Ind. Chem.25% Acrylic B-82 Rohm and Haas95% Styrene) A-75 Pennsylvania Ind. Chem.

5% Acrylic) B-82 Rohm and HaasDeloto 73-35A DeSoto Chem. CoatingsDeSoto 73-35B DeSoto Chem. CoatingsDeSoto 73-35C DeSoto Chem. CoatingsDeSoto 73-35D DeSoto Chem. CoatingsDeSoto 75-02 DeSoto Chem. Coatings

6

IIlk

•'ili•- . ,+ID " lt+• 1. Of/

S. k "•I I. • .• _ .+. -m •," _. - pg-' .. , - _ _•

0 o

*.m ... -, ; +• I • .,9• ,r,414

46.U

" '; ,.g,--. . " . " • tS ''- . """or

* Cj

O4

4k4

"w4

AS.

#MOO

FA

*A

~ .4 *

Sc

IfD

40P

dV '

1

4

VP 0� * , 0

a #44, 4 9

* �e.*. %

V

U'6

V.

9,4F.

4� V

4

"U.

-11-

Table 2 (Cont'd)

Resins Used in ERDL:DA-52-009-ENG-5100

Producer'sType Code Producer

DeSoto 75-648 DeSoto Chemical CoatingsAroplaz-Alkyd 6006X50 Archer Daniels, MidlandLustrasol-Alkyd 13-075 Reichoid Chemicals, Inc.Styrenated-Alkyd 13-030 Reichold Chemicals, Inc.Everflex G Dewey and Almyý5% Parez ) 613 American Cyanamid75% Everflex) G Dewey and Almy25% Parez) 613 American Cyanamid75% Gelva) S-55 Shawinigan Resins Corp.Parez 613 American CyanamidWater Soluble Alkyd T.S1.0-619 Anoco Chemicals Co.Dow 636 Dow Chemical Co.

B. Description of Equipment

1B. Humidity Chamber

The high cost of building temperature-and humidihy-controlled rooms has deterred many investi~tors from mak-ing electrophotographic electrical measurements understandard conditions. We found it expedient to build asmall chamber and to conduct all measurements in thischamber while keeping our instruments outside the cabinet.

Figure 6 shows the general construction details using twocommercial standard-size aluminum storm windows for thefront (entry) side of the cabinet. In Figure 7 is showna schematic diagram of the control system and of the re-circulated air.

All of the major items necessary to build this cabinetare given in the following list, and Figure 8 shows a pho-tograph of the cabinet in use. The two front windowshave been removed to show access to the working area.

12

I JI

ii -i -q

_ _ --- U!-90

1~ 13

ILI AIR

r -~ JVCAW VOLTAGI I r --il

Igo

1 J'ENJRSGUR ? t j~O~I

IA

-15-

List of Materials

1. Bantam Duct Hose, 4", from J.P. Bushnell Co.,3436 Lindell Boulevard, St. Louis, Mo.

2. Humidity sensing elements, Cat. No. 4-4823(see No. 4).

3. Converter to sense change in resistance of sen-sing elements and convert this to a d.c. signal, Cat.No. 15-6320, with 4-5014 cable.

4. Mvltiple mounting for sensing elements, Cat.No. 4-5J' Items 2,3, and 4 from Hygrodynamics, Inc.,949 Selin Road, Silver Springs, Maryland.

5. Room humidifier, Sears Roebuck & Co., Model716A.

6. Room dehumidifier, Oasis Model SD-32.

7. Leeds and Northrup, Cat. No. 3-930-010-044-6-004-1-38, Model S, Speedomax-H indicating strip chartrecorder controller with integral D.A.T. control unit.

8. Plexiglas and aluminum cabinet with standardhousehold aluminum windows. Dimensions 18?? deep, 30"high, 48" wide, built by local cabinet maker. Glassin windows replaced by 1/8" plexiglas.

9. Relays, Cat. No. 75P-603, Type PR7AY DPST,115v., Allied Radio Co.

10. Harco Axial Fan, Cat. No. T-6802, Herbach andRademann, 1204 Arch St., Philadelphia 7, Pennsylvania.

11. Plenum boxes for humidifier and dehumidifierand 4", 90 degrees elbows from local sheet-metal shop.

Figure 8 shows an overall view of the cabinet with thewindows removed. In actual operations, the technicianperforms his manipulation-s through armholes cut intothe plexiglas windows. The dehumidifier is shown onthe lower right side below the cabinet and the humidifier

II

-16-

is on the left . The plenum boxes covering the inlets andoutlets are clearly shown. Also shown is a small circu-lating fan in the chamber to insure thorough mixture ofthe air.

Humidity control is maintained by the Leeds and Northrupcontroller. This instrument is built with a neutral, ahigh and a low contactor. The humidifier is connected tothe neutral and low points and the dehumidifier is con-nected between the neutral and high points. The desiredhumidity is set on the scale of the instrument and thehumidifier or the dehumidifier goes into operation untilthe desired set point is reached. Both humidifier anddehumidifier have fans so that the conditioned air isblown into the chamber through the 4" bantam duct hose,and the emergent air is carried back to the same unit.This constitutes a closed system which makes for remark-ably easy and uniform control. No temperature controlis used. The large radiating surfaces of plexiglas andaluminum provide sufficient heat transfer so that thecabinet stays near the ambiek.t roor temporature. Humid-ity is currently being controiled +1.0% between 20 and70% relative humidity at 80°F.

2B. Resistivity Measuring Equipment.

The measurement of surface resistivity of zinc oxide pre-sents a problem not encountered in a bulk measurement.Zinc oxide is extremely active as a sorption-desorptionagent and the surface resistivity reflects the ambientatmosphere. Tests in our chamber show a parallel seriesof curves in which the surface resistivity of zinc oxideexactly parallels the output from a set of Hygrodynamicshumidity sensors. For this reason this measurementmust be made in a carefully controlled atmosphere.

A ten-gram sample of the oxide is compressed to 8,000pounds in a standard Carver test cylinder and a silvercontact paint (General Cement Co.) is applied to make acenter contact of 8.0 mm diameter. A guard ring is nextapplied so as to give a spacing of 8.0 mm between thetwo contacts (see Figure 9). After a 30-minute drying

interval at 110OF to remove the organic solvent used in

II

1?

ZINC OXIDE RESISTIVITY SPECIMEN

710TALO0. ). Z 9.CENIt DO T a

-SPACE 8C

FIGURE 9

I 2ROs~CrJN3N *, irj

S URFACE

FOCU-51C LENS

'I,

OP::T/CA L S KS TEN FORTHYE 35TUD Y OF ZnO0

R75/,S TIV1 T Y

FIGURE IR

-18-

the silver paint, the compressed pill is held at the de-sired humidity for two hours and then positioned in theresistivity measuring jig. The weight-loaded, silver-tipped contacts are looered on the pill and the jig ispositioned in the dark box as shown in Figure 10.

The optical system consists of a 300-watt projector us-ing a tungsten filament lamp, a plano-convex lens, and aset of condensing lenses to insure a parallel beam to thefirst mirror. This mirror is mounted in the top of thedark box and is set at a 45 degree angle to reflect thebeam downward. Another set of condensers is used to col-limate the beam and a final lens is used to focus the beamon the lower mirror, which reflects the beam onto thesample. The optical system is shown in Figure 11.

Provision has also been made in this system for the in-sertion of interference filters in the light beam sothat we can study the photo-response of zinc oxide atvarious wavelengths.

A General Radio electrometer type 1230-A is the measur-ing instrument, and an Esterline Angus model AW, 0-1milliampere meter monitors the output of the meter. Theelectrical circuitry is arranged in such a way that thesample can be subjected to repeated 10-second bursts oflight in order to study the fatigue effect as well asusing light saturation to determine the total change inresistivity from light to dark environment. Photo-current can be read by electrical switching of the cir-cuits without disturbing the sample. The wiring diagramfor this operation is shown in Figure 12.

3B. Electrophotographic Equipment

Much of the equipment necessary for the successful gen-eration of a static charge on a coated sheet is conmmer-cially available from R. J. Paulus, Anken Chemical andFilm Corp., Newton, New Jersey (Items 1-5).

1. Double corona power unit (8000-volt D.C. powerunit plus and minus, with two meters for use with doublecorona discharge unit.) A 4000- to 6000-volt negativeD.C. power pack with no meter is available at lower cost.

I 19

II

I W,

I rrf

I U-

Ir

I

I

1

II

20

GUARD- POINT ARRANPUIENTFOR RESISTANCE MEASUREMIENT

B/I-I olFR

t ip&PUT

___4o~j..£

IP4PUT

0 R

lE

INPUT

DIAGRAM FOR THE STUDY OFFATIGUE, LIGHT-DARK RESISTIVITr

AND PHOTOCURRENT IN ZnO

FIGURE 12

I

-21-

2. Corona discharge unit (corona wand).

3. Special iron filings.

4. Alnico magnetic brush.

5. Low M.P. toner, Code 39-50, Philip W. Hunt Co.,Palisades Park, New Jersey.

6. Brass (or aluminum) plate 1/8" x 12" x 18".This plate is used as a support for the coated sheet dur-ing corona charging. A good earth ground on this plateis essential.

7. Two sets of ball bearing drawer slides are usedto mount the brass plate on a horizontal surface. Thisallows the plate to be moved left to right and front toback during corona charging with the wand held in a fixedposition.

8. Photo timer (Time-o-Lite or equivalent).

9. 20, 40, 60, 100-watt lamps and reflector.

10. Standard .boratory drying oven capable of main-taining 1000C (for fusing the finished prints).

11. Osteriser or Waring blender.

12. Static Probe. This probe is made by cementingtogether four sheets, 1/8" x 4" x 8", of lucite or poly-styrene. The sheets are cemented along their edges soas to form a rectangular box 4, square by-B" high withno top or bottom. Two holes are drilled on oppositesides of the box, 1/8" from the bottom edge, and a two-mil platinum wire is threaded into position. The in-terior of the box is lined with aluminum foil. Connec-tion between the platinum wire and the electrometer ismade by using Belden coaxial cable RG62U or equivalent.The aluminum foil interior in the probe is connectedto the cable shield and is carried back to instrumentground.

The probe is calibrated by placing it on a metal platemaintained at some fixed and known D.C. potential. Moving

I

-22-

the platinum wire closer to or farther away from the D.C.field above the plate allows a. direct reading calibrationon the General Radio electrometer when the latter is setto read voltage.

13. Wire applicator. This is a 1/4" steel or brassrod wrapped with 22 or 24 thousandths nichrome wire.Such an applicator is commonly used in the paint trade.

C.. General Operating Procedure

Two general methods are commonly employed in the produc-tion of laboratory-size electrophotographic coatingmixes. The Waring blender (or Osterizer) is rapid, usu-ally requiring only a five-minute mix. The ball mill isthe second method. This is slow, requiring up to threeor four hours but usually giving better dispersions thanthe high-speed mixer.

Our standard method of mixing and the one used to makeup the coatings for the prints displayed in this reportis as follows:

Resin Stock Solution

Dissolve 354 pm Pliolite S5-B resin (Goodyear)in 1646 gm certified reagent xylene (17.7% drysolids solution).

Weigh out 203 g ZnO, add 141 pm resin solution (8:1ratio dry solids ZnO:resin) and spatula mix in a poly-ethylene beaker with a stainless steel spatula. Pourinto a high speed blender and mix at high speed forfive minutes. Pour the coating slurry into a g~ass jarand allow to cool to room temperature. After cooling,add 15 ml xylene and then coat treated paper sheetsusing a No. 22 wire drawdown bar.

1C. Preparation of Paper for Coating

An A. B. Dick mimeograph bond 43-1210, 801/2" x 11",Sub 16 is used. This is treated as follows:

I

-23-

a. Dissolve 15 gi paraffin in 400 gmtoluene

b. Dissolve 6 pi ammonium acetate in200 Sm 95% ethyl alcohol

Pour solutions (a) znd (b) together and warm until milki-ness disappears.

The solution is placed in a photographic tray (glass) andkept warm while the mimeograph paper is passed throughthe solution (total immersion). The dipped paper is driedfor a few minutes in a 1000C oven and is then ready forcoating. The coated sheets are air dried and finallydried for 15 minutes in a 1000C oven. Before testing, thecoated sheets are conditioned in a humidity chamber at 50%R.H. at 780F for 16 hours.

The rest " the equipment used in the project is of a spe-cialized nature and will be discussed while we are describ-ing the various phases of the work.

2C. Measurement of Electrical Resistivity of Zinc Oxide

The equipment used in this work was previously describedon pages 16 and 18 under Resistivity Measuring Equipment.The purpose of this phase of the work was to point out thevital importance of keeping the entire electrophotographicsystem moisture-free. Since zinc oxide is an extremelyfine particulate powder it is susceptible to moisture ad-sorption if it is improperly handled or stored prior touse. This work points out the deleterious effect ofmoisture and applies to entry of moisture at any stageof the manufacture of coated paper.

The resistivity specimens made on the Carver Press (Fig-ures 9 and 10) were conditioned in the dark at the de-sired humidity for four hours before taking the darksurface resistivity measurement. All work was done inthe humidity cabinet described earlier.

An examination of the curves shown in Figure 13 show theeffect of humidity on the electrical resistivity of zincoxide. All samples show a sharp decrease in resistivityas the humidity increases (see Figure 13). Furthermore,

41. r 44 R

4~~~f +4 .,-

1-

... .. . ...( .* L ' . .

4 41 44 ITT

~ tLt

4 f-

1-44INt

Hi -;4 M 2 2

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

-7 -ii4

f. K

z ,4 1 i.. ... .. .

ffj7 ...... .....

It,

!1 tPercentI Rlt Ve ! Humidit at..

I

-25-

the smaller particle sizes are the most susceptible tochange in humidity. This ia extremely important becauseour measurements show that a1 of the samples testedfall to 1 g 10° ohm-cm resi•livity under illumination.The 1 x 100 ohm-cm figure also holds true over the en-tire humidity range in which we worked from 30-70% R.H.at 76 0F.

Maximum photoconductive response can be achieved onlywhen the resistivity of the zinc oxide shows a sharp de-crease under illumination. If an oxide such as the 0.136micron sample is used at, for example, 55% R.H. tke re-sistivity under illumination can drop from 1 x 10 to1 x 100 (the lower limiting value under tungsten illu-mination at the light intensity level used in this ex-periment). A pigment having a particle diameter of 314.342 or .408 microns under the same gonditions i55% R.H.iwill drop in resistivity from 2 x 10Y to 1 x 100 . Sincedrop in resistivity is a direct Ohm's Law function ofconductivity, we can expect that higher humidities caus-ing higher conductivity in the pigment will result inlower charge acceptance, and this we find to be true aswill be shown later in this report.

It would appear therefore, that the best continuous toneprints made will be those in which the pigment-bindersystem has been freed of all traces of moisture.

Resistivity measurements of this type are useful becauseunless the bulk and surface resistivity of the zincoxide is within certain limits it cannot function as aphotoconductor. That is the resistivity must be highenough so that current flow between adjacent particlesis impeded sufficiently to enable the ZnO surface totake on and hold a static charge.

Recordings of the resistivity measurements under illu-mination followed by darkness confirms the work ofMo0Iwo (Seitz, "Solid State Physics", Academic Press,N.Y., 1959, Vol. 8, p. 209) and others, and shows thatthe "recovery" of the zinc oxide is quite rapid. Thecur•e showing the increase in resistivity of the

I

-26-

specimen after the light is extinguished, shows an ex-tremely rapid recovery. The time interval necessary forrecovery is less than one minute when using a tungstenfilament lamp and at resistivity values of the order in-dicated previously.

_3C. seageen to0 Electr~icp. parameterg of Ele gtro-phgtographic coatings [under yratten OA Safelight Con=

Using the equipment described previously we next proceedto the el.ectrical measurements of the coated surfaces.

The coated sheet which has been conditioned at the de-sired humidity (described previously under heading Cl)is placed, coated side up, on the metal plate under thmcorona wand. It is most convenient to suspend the waniat an adjustabln distance over the metal p ate. Fornormal electrostatic charging of paper using 2-mil plati-num wires in the wand and working in a room at 50% R.H.at 780F, this distance is about one-half inch. The high-voltage generator is turned on and the electrostaticcharging is monitored by using a 0-100 micrfampere meterin the high-voltage line between the generator and thecorona wand. We have found that results are much morereproducible when we maintain the same current flow onsuccessive tests.

The metal plate (thoroughly grounded) is moved left toright and front to back on the drawer slides on which itis mounted. This movement continues throughout thecharging cycle in order to assist in levelling out theelectrostatic charge on the coated surface. Tht micro-ampere teer is monitored and the current is held at50-60 microamperes during the charging cycle, which mayvary from 10 seconds to one minute at the discretion ofthe operator.

It will be noted that when a te!st is made on a conduc-tive-base sheet it is necessary to move the corona wanda greater distance than 1/2 inch from the ground platein order to maintain 50-60 microamperes of current.Similarly, when a test is made on a coating which hasbeen applied to aluminum foil, it is necessary to

I

-27-

increase the wand-to-plate distance to nearly one inchin order to maintain a 50-60 microampere current.

After charging, the General Radio electrometer andEsterline Angus recorder, described under B2, are used.The plastic 4" x 41" x 8" electrostatic probe is placedon the charged sheet and the electrostatic voltage isnoted. After a standard time interval of about 10-20seconds, during which time the loss of charge in thedark is noted (dark decay or half-life), a light isturned on and the static field collapses. The recorderproduces a record of this entire procedure and from thechart it is possible to see the maximum charge whichthe coated surface would accept, the rate at which thestatic charge leaks off, the speed of discharge underillumination and the residual voltage left after illu-mination.

At this point it is customary to take a second coatedsheet which has had the same coating and humidity stor-age as the one used above and to make an electrophoto-graphic print. The electrical charging cycle is thesame as described. When it is complete, an image isprojected on the electrified surface (or made by contactprinting). The exposure must be controlled since over-or under-exposure causes the same final effect as whenworking with silver paper (except of course that the ef-fects are opposite - over-exposure in conventionalelectrophotography causing a weak image).

After exposure the electrical image is made visible bydevelopment. it is convenient to make the developerusing a l0-ml glass graduate. A volumetric measure of10 ml of iron filing and 1-2 ml of Hunt 39-50 toner isused. This is thoroughly mixed and applied to theelectrical image with a bar magnet (magnetic brush).The areas which retain a negative charge will attractthe positively charged dye particles and the imagebecomes visible. The image is made permanent by plac-ing the picture in an oven at 1000C for a few minutes.The Hunt toner contains a low melting resin, whichfuses and fixes the image to the paper.

t

Insert on Page 28 at end of Section 4C and before Section 5C.

4C-a.

Dielectric Constants of VariousResins Used in Electrovhotogr•ahy

FrequencL(Cycles r Second) at WhichRtirnnatfon wag Mad_

Acrylic B-82 4.7 4-.1 3.3 2.9Pliolite S5-B - 2.5 - - 2.5Pliolite S-7 - 2.9 - -Vinyl Butral 3.61 3.58 3.33 -

Measurement of the dielectric constant of the resins used inelectrophotography was considered because we felt that thedielectric constant would in large measure, determine theability of the resin to accept and to hold a blanket nega-tive electrostatic charge. While this was found to be truein a general sense, it was also found that other variablesplayed a much more important part (oxide to N.V. resin ratio,type of solvent used, particle sise of the ZnO) and we there-fore concluded that the use of the dielectric constant wasimportant only in a general sense in thav when trying to de-velop absolute maximum charge acceptance it is best to use aresin with the highest possible dielectric constant.

I i

-28-

4C. Effect of Resin Binder on Continuous Tone Behaviorof E ectrolphotographic Surfaces.

Out of a total of approximately fifty resins tested wehave selected the samples which showed the most promisein electrophotography and have tabulated the results ofthe tests in Table 3. The data is self-explanatory,and the tonal range shown in the right-hand column in-dicates the range of values obtained. That is, if a 13were to be seen in this column it would represent 13shades of gray from black-to-white.

As a further aid to future work and possible correlationof electrophotographic behavior and structure, we haveindicated in Figure 14 the repeating portion of thestructure of a number of the resins tested.

As an extension of this work we present in Table 4 aseries of tests in which we have varied the ZnO to non-volatile resin ratio over a wide range and using twodifferent resins. As can be seen from the tonal rangevalues this change in the solids ratio has little orno noticeable effect on the tonal range of the coating.

C. Effect of Particle Size of Zinc 0 de on Spectralespone as Determined bX CDanMe in Electrical Re-sistivity, ..

As raw material suppliers we have been in a favorableposition to see the strong demand for zinc oxide samplesof every conceivable particle size for experimental usein electrophotography. The general impression in thetrade is that particle size is very important. This hascertainly been found to be true with respect to totalcharge acceptance (saturation voltage). One of thecloudy areas not yet reported on has been the effectof particle size on the panchromatic behavior of theelectrophotographic layer.

Using the optical system described on Page 17 of thisreport, we prepared specimens using the electrophoto-graphic grade zinc oxide set aside for this project.Sheets of West Virginia Paper Co. stock were coated as

0

t. 14 0

m.~ 41 r-4 0 4 C YC uC CYM04% A

fb 4r4~ H .-4 HMr4i r,4 H r1 HM r4 rlr,4 r,4 r4r,

-~~~4 0~.*** *

M r4

soI 0 a 0 0;

'a v

14 ri30.8 a *.. 6 tGo 0 .1 ui.: a 1111k 8 C3 r

~0 90,01611 rC.)C.ii, ugo0 r1% C4 . .

-V ~ 14r 0 0

.8 0 8.8 0 0% 0 0% %W 0

%014%O

1 ~ .. A.10 11P 0

I30

1

1) DOSoto Resins 72-35A, 72-355, 72-35C. 72-35D, and 75-02.

5) Pol0vinyl Acetate.

C 2-- CH2 02"- C

X - 40:20 Note % -3' 60-80 Hole %

I Alkyl Utaer

Differeuces between resins are variation inolacular weight and ratio of X to Y.

2) DeSoto Resin 72-643. 6) Polyvinyl. lutyral (lutrar 3-72A).

I I 78 HoleR - Alkyl l9ter X 20 Hole I

Z -2 mole %B - Abkyl

7) Acryloid 3-82.

3) Pliolite S-5I & S7.

2 -c - cu-cu, C. 2 -cN CN321

C13

Pliolite S-53 . I - 67 Mole I

- 33 Hole

Pliolite S7 - - 538 Mol I

Y - 42 aole

4) G•neral •lacri a-0r . FIGURE 14-,

3. o" -•,r- " STRUCTURE OF VARIOUS£1o¶ - I o RESINS BEING EXAMINED

CoI3 &3 C,3 FOR USE INT exact ratio of these, unite is not k ELECT ROPHOTOGRAPHY

1~ 31

4i

(f40

I A~ r- t- C- c- r. - .- u A Nu

II Z1 lAAO IU

r-8r

Ir~

00 0 0 0 c0cr. **..4

34 .* **..

cc -V~U % I -

ok N N Ný00 ?I

II I a aO

-32-

described on Page 22 in this report, and silver contactswere painted on the coated sheets, The sheets were con-ditioned for 16 hours at 50% R.H. at 78*F and were thenplaced in the resistivity measuring equipment.

The optical system consists principally of a tungstenlamp and an interference filter (Bausch and Lomb No.42-47-55-01, set of 5). Light intensity at the measur-ing surface is standardized, using a 3/8" diameter, 8-

junction bismuth-silver Eppley thermopile, and the tung-sten filament bulb is moved away from the condensinglens of the projector until an illumination level of0.00036 watts/cm2 is reached. This same level of illu-rmination in watts per square centimeter is then main-tained using the 450, 500, 550, 600 and 650 nm inter-ference filters by moving the bulb closer to theprojector condenser lens.

The electrical surface resistivity of the zinc oxide-resin surface was measured first in the dark and thenunder standard illumination. The results are shown inTable 5.

6C. Effect of Particle Size of Zinc Oxide on SpectralResponse as Determined Using a Grating 'pectrogravh.

The five zinc oxide samples set aside for this projectwere mixed in our standard coating formula preparatoryto coating.

A Central Scientific Co. Cat. No. 87102 grating spec-trograph is used to make the spectrograms. The filmholder and slotted masks are removed and a piece ofclear film is taped over the film holder opening. Amercury lamp (for example, Cenco Cat. No. 87268 high

essure mercury arc light source) is used as a cali-grating light source and the prominent mercury linesat 5790, 5460, 4358, 4046 and 3650 A.U. are marked onthe film. This film is then inscribed at these pointswith the proper wavelength values and the calibratedfilm is then permanently taped in place on the spectro-graph. A light source such as Cenco 86605-1 opticalbench illuminatbr is satisfactory and it is well touse a light shielA such as Cenco 87118.

*1 -33-

co4 00 00&N*4 ON f-4 Fi-1 N 4 N 4 r

' I c'a xx 4 mm r N

Al00 00 00 0 0 00r-. .I-4 0-1 r4 r-4 2 i-4 e-4r4 r4r

C) 00 0A ~e0 0u W%

0 e-4 .l -4 N- rq

01 00 .00 0 o00 t 2211 r- - H -r-4 N.1 r. S.

N r- 4 r4l-I H 0 0~

CCU* 0 00 00 00 00 00 44U UN

H r-4 r4 v-4.-4r4 ~ M HP- 94 r4-I A.44

W% Q 0U um~ 00% UN~ H~q4.94 Por-4 ýM ýl t-: %0

t 0 9 4 *4 0 4- -CI A A... 1 r4IPr

H9 A V54 d 4

0 a

I~ OM V4.- %4-4 N4v Cf

t4i

I,

The sensitometric measurement is made by first corona-charging a coated sheet. The charged sheet is placedagainst the calibrated film on the spectrograph and theshutter is opened. An exposure time of 30 seconds isusually sufficient to produce a good density print ondevelopment after the exposure. The calibrated filmprints the wavelength scale directly on the sensitizedsheet, thus making the determination extremely rapid.

A Cenco 87123 logarithmic sector disc can be added tothe optical system to make the determinations quantita-tive.

Figure 15 shows the sensitometric sheets prepared by us-ing ziznc oxide samples ranging from 0.134 to 0.408 micronaveraCe particle diameter.

Evidence presented both by electrical change in resist-ance and by direct photographs on the spectrograph showquite clearly that, regardless of the particle size ofthe zinc oxide (within the limits here chosen), the sen-sitivity of the undyed coating does not respond to achange in particle size of the zinc oxide.

C. Effect of Paricle Size of Zinc Oxide on TotalC are Acceptance.

In the previous section we demonstrated the lack of de-pendence of spectral response on particle size of thezinc oxide in the binder coat. We have pursued thisfacet of the work to its ultimate conclusion and now pre-sent data which indicates that particle size does indeedplay an important role when we attempt to increase or de-crease the electrostatic field on the coated sheet. Fur-thermore, the speed of electrical discharge appears to bea function of particle size in an unsensitized ZnO-bindersystem. Table 6 shows the numerical results of measuringa family of curves obtained by electrometer tests on var-ious particle size zinc oxide-binder coating m!xes.Charge acceptance is highest in the case of the smallerparticle size oxide and charge acceptance decreases asthe particle size increases. Figure 16 shows a singleelectrometer tracing and the points usually selectedfor evaluation, namely, (1) total charge acceptance,

31

61gj 0

4.e- (5,-4

0(06)54

ooMa 0

w 4 )4 )

00

0 U'004

S0 ON:O3m 0

00.'

4-G) .4 so4)r4 "q

coo0 0 r"

ogo

D

0Cf)

0 z

g .rI

(54.

4)0 00t4)

c; .4

36

lIecfrQnieter drift 2v

probe lift~d fromn surfaice

to rear-h ýestduai 6.5 ~

-55

14y-0 *,

a,,5 s4c

a) gt onda rkwde W1

I *14-e

C~ape a~n

I

-37-.

(2) charge level at the end of the dark-decay (calledhalf-life by some investigators)interval (3) voltage1.25 seconds after illumination begins, 14) voltage 2.50seconds after Illumination begins, (5) residual voltage(when voltage drop approaches a linear function), and16) number of seconds to reach residual voltage. Table

shows how this data is tabulated for evaluation pur-poses. The data is "neat" in the respect that it leavesno doubt regarding (1) total charge acceptance is high-est in the case of the smaller particle size oxides, (2)residual voltage is also the highest in the case of thesmall particle size oxides, and (3) the time to reachresidual voltage is the longest in the case of the smallparticle size oxides. A photographic reproduction of afamily of curves is shown in Figure 17.

The work was repeated at two humidities (Table 6) andthe results are exactly as expected.

1. Charge acceptance is highest in the case of theI lower humidity.

2. Speed of discharge is slowest in the lower hu-midity test coating.

3. Residual voltage is highest in the case of thelower humidity test coating.

I 'U

I 0 -

CLLL -

biN 4)

IL

ft'9-

0 ~rI00 r0 10 %%

0 ou FA v-4 P-4 0- 1 -;

oH 0

:9i

00 0 00 0000 00 !

0 0N0 C % N% 8CVC (v N 0 P N 1

0443 00 00 O401 U 0 :95 % om!o & 0.c

NO 0 V544 ,r ^ DG ,8P

1 0 000

0 00000 00 Ow0 %-M ~~~~4~~

0"oi z 4 Y 2

4C3' ow5-P1* 0c131~ ~ SO4455

NcV%

womee 400000 *

Insert on Page 40 at end of Section 7C and before Section 8C.

7C-a. Effect of Zing Oxide Particle Sizeoon Qielectric Break-down Voltages of Coated Sheets at various R dlties.

Some of the coated sheets prepared for charge acceptance testsand made as described on Page 22 were used in this work. Wecoated both papr and aluminum foil using a wire drawdown barand coating weight of twenty pounds. All voltages were deter-mined using the ASTM Standard Dielectric Strength Procedureand a 10-second approach to actual breakdown. It should benoted that in this test, a D.C. potential was applied, whereasthe regular ASTM test utilizes 60 cycle A.C.

An analysis of the breakdown voltages showed the following:

1. Particle size of the zinc oxide used in the coatinghas very little effect on electrical breakdown of the coating.This was found to be true both on paper and aluminum foil sub-strates.

2. Breakdown voltages varied from 1750 volts D.C. at 30%R.H. to 600 volts D.C. at 70% R.H. using the same pigment(0.136 micron size). Intermediate humidities have intermed-iate breakdown voltages pretty much as expected.

From this we conclude that particle size has little effect onelectrical breakdown and that the breakdown voltage is muchmore dependent on the type of resin used and on the totalmoisture present in the system when the test is made.

I

-40-

It would appear that the particle sis3 of the zinc oxideused in an electrophotographic coating would be dictatedby the ultimate speed desired in the coating. Whilelarger particle size oxides do not accept as high acharge as the smaller particle size oxides, they dis-charge more rapidly and in effect produce a "faster"(higher emulsion speed) coating. This is true, ofcourse, only in an unsensitized system.

Dye sensitization as currently practiced in the indus-try today changes the particle-size picture. Since theorganic dye is adsorbed on the surface, the smaller-particle oxides present a larger specific surface withresultant higher dye-adsorbing capacity. When the cor-rect sensitizer dye is chosen, the junction between zincoxide and binder becomes more conducting in the ZnO -- )ýbinder direction under illumination and the smaller-sizeoxide then shows a significant increase in speed over thelarger-size sinc oxide. This increased conduction of thedye-sensitized sinc oxide is used to further advantagebecause lower total charge acceptance figures are ob-tained on a sensitized oxide. This also contributes to-ward the faster coating.

8C. Effeet of Additives on the Continuous Tone Charac-teristi cs of an ElectrophotoAraphic Coatin .

The effects of a number of additives or combinations ofadditives on the continuous tone of electrophotographiccoatings are illustrated in Table 7. These additivesare only a few of many that have been tried, but sinceadditives fall in several different categories we as-sume that all or most additives chosen from a specificcategory would have similar effects on continuous tonecharacteristics. In general, additives for electropho-tographic coatings fall into the following categories:(1) additives to improve the charge acceptance of acoating usually oxidants or oxygen-containing mater-ials, (2) additives to increase the speed of the coat-ing, (3) additives which decrease the dark decay rate -usally waxes or high molecular weight aliphatic mater-ials, (4) additives which increase sensitivity of acoating to all colors of light - dyes or combinationof dyes, (5) additives which change the mechanical sur-face properties of a coating - for better adherence of

jp o

ddd~d a a 8 d ad8 add 88 Ad

u m r - r0 , u % o 0 00 c m 0 0 Y A A~ 'S 's ~ A0 'I - I C YJ pA

0* .4 4 .

V M. I'

TI, 0 0

ull0 o gA 0§

m% Go PE'E I

I - 110, ]1 S

A~M 3~. 2INEbbk bi11f I 4 o1d jI 1~

IN 000 000 0 0 0a 0 00 .0 0 *.~ I P.A ýA

I

-42-

toner or to produce adherence of the toner at lower volt-ages, and (6) additives to produce desirable rheologicalproperties, used for the most part in aqueous systems asthickening agents, viscosity stabilizers, pH controllers,etc.

The charge acceptance of French process zinc oxide, suchas AZO-ZZZ-661, either produced fresh or partially car-bonated when aged in the presence of moisture, can beincreased materially by the addition of small amounts ofbenzyl peroxide, 30% hydrogen peroxide potassium perman-ganate or other oxidants. When potassium permanganate in.1% and .2% additions, based on the weight of zinc oxide,was added to an aged oxide (Table 7) in a Pliolite S-5Bcoating, it was found that the tonal range was reducedfrom 7 to 4 or 5 by this addition. The tonal ranges weredetermined by using a No. 3 Kodak 21-step tablet and Hunt39-50 toner. In general, all our results to date indi-cate that addition agents of this type reduce the tonalrange with the presently available toners.

Small quantities of other additives, such as cationicagent Nopcogen 14L, p-dioxane solvent, ascorbic acid,etc., the selection of which depends on the solvent andother materials of the coating mix, can increase thespeed of the coating materially. These additives havebeen found to be particularly desirable when coatingspeed is essential with a minimum amount of dyes to ob-tain a whiter coating. In most cases where the chargeacceptance is not materially reduced, agents to increasecoating speed reduce the tonal range. The exception tothe rule is 14L. This agent decreases charge acceptancewhile increasing the tonal range.

Materials such as paraffin and hydrozo wax, which havehigh electrical resistance, can be incorporated in sol-vent systems at 1 to 4%, based on weight of zinc oxide,to decrease dark decay. However, it was found thatagents of this type (Table 7) have little or no effecton the continuous tone quality of the electrophotogra-phic coating.

Specific dyes or combinations of two or more dyes havebeen found to increase coating speed and to increase thesensitivity of the coating to all colors.

U

-43-

There is some evidence to indicate that the physical

characteristics of the coated surface play an importantrole in the development of an electrical image. Thus,a number of electrophotographic coatings have been pro-duced in which the mechanical surface properties havemade it impossible to produce quality prints. This isusually apparent in coatings which are extremely smooth;the toner that adhere originally is rapidly displaced onfurther brushing.

An example of a coating of this type is one produced withthe water-soluble alkyd, tri-metallic neopentyl glycol.It is possible with this material to develop an image andthen, by further work, to completely remove the image.(The experiment described here was done on a coating hav-ing good dark-decay characteristics and the image removedwas not due to loss of static charge by dark decay.)

With some acrylic resins the addition of paraffin to thecoating has changed the mechanical surface properties togive uniform deposition of the toner but this does notincrease the tonal range.

Rheological characteristics of coatings are controlledin solvent systems by the resinous binder while inaqueous-base systems agents such as Daxad 11 and 1,4butanediol are effective. In aqueous systems additivessuch as Daxad did not affect the continuous tone char-acteristics of the print. In solvent-based systems,high solids content, high pigment to resin ratios, andlow viscosity are the desired combination of properties.For example, the DeSoto resin 72-64B had a relativelylow viscosity at high non-volatile content and high pig-ment to resin ratio (750 centipoise viscosity at apigment-resin ratio of 12:1 at 80% non-volatile).

9C. Effect of Light Intensity on Tonal Quality.

The procedure used in testing light intensity was toprint a step tablet using different wattage lamp bulbsfor different time intervals. Obviously, it would notbe wise to use a single bulb and control the intensitywith a powerstat because of the wide color shift whichwould be encountered, with strong reds predominatingwhen the bulb was operated at low voltage.

Insert on Page 44 at end of Section 9C and before Section lOC.

9C-, Correlation Between Electrophotographic Charge Densityand the Resultant Print Imane Density.

It was our feeling at the beginning of this work that in orderto achieve a good dynamic range consisting of perhaps twelve tofourteen shades of gray it would be first necessary to developa high total charge acceptasce. This procedure would then al-low, for example, fourteen steps between white and black in 50volt increments. Work with a commercially available black toner(Hunt 39-50) showed that by our test method, black toner part-icles were capable of "seeing" 10 volt increments and could pe9r-haps be made to "see" even smaller increments by a more refinedtest technique. Armed with this data we then proceeded to testa number of toners and carriers submitted to us by the PhilipW. Hunt Company Palisades Park, New Jersey, hoping that abetter combination of developer and carrier could be found.

Our results showed that with Hunt Carr' - No. 1 and No. 2, andHunt 39-50 toner powder, almost no toner adherence was observedon a step tablet carrying voltages ranging from 450 volts to 10volts and made by exposure of a charged sheet through a nega-tive prepared expressly for this purpose. The experiment wasrepeated using a special "low charge toner" prepared by Huntand the results were the same. The Hunt Carrier No. 1 and No.2 (iron filings) carried no designation other than No. 1 andNo. 2.

Next, we used Belmont iron carrier and compared Hunt 39-50toner with Hunt low charge toner. On a step tablet carryingvoltages ranging from 500 volts to 30 volts, the best reflec-tion density which would develop was only 0.75. The two steptablets were nearly identical in appearance.

A final test was made using Hunt 39-50 toner and Cormack ironcarrier (now currently supplied by Anken Chemical Co.,Newton, New Jersey, and designated as #100-200 mesh iron fil-ings). On a step tablet carrying voltages ranging from 675to 10 volts we were able to distinguish eight distinct stepswith the reflectio density ranging up to 1.65. With thisoverall dynamic -ange it should be possible to develop manyintermediate shades of gray.

After the work on tonal scales had advanced to the point wherewe had standardised on the use of a Pliolite binder (ratherthan SR-82 as was used in the above work) we found that high-charge acceptance was not a necessary requirement to the de-velopment of good black print density. A stop tablet with

9C-a (continued)

voltages ranging from 150 volts to 10 volts and made withPliolite S5-B. or S-7 binder gave a black density and dynamicrange equal to the one produced with SR-82.We therefore conclude that the combination of Hunt 39-50 tonerand the iron filings from Anken Chemical Company, is the bestpossible combination which we can use at this time for an ex-panded dynamic range. These materials could certainly be im-proved but they represent counercially available products atthe present time.

We also conclude from this that high charge acceptance is nota primary requisite to the production of an extended grayscale. If the toner powder can "see" differences of only tenvolts, a fourteen step tablet could be made on a sheet havingonly 140-150 volts total charge acceptance.

I

-44-

We used standard formula of pliolite S5-B and the 0.342micron oxide. Light conditions were as follows:

Watt Rating Light Intensity asof Bulb Measured on G.E. ExposureUsed in Foot-CandleTime Foot Candle

Printing Meter (Seconds) Seconds

100 600 0.1 6060 300 0.2 6040 150 0.4 6015 50 1.2 60

7.5 25 2.4 60

The five step tablets were all equally exposed (withinallowable experimental error) and despite the slightextra red which one would expect from a 7.5-watt bulbversus a 100-watt bulb, the 7.5-watt light source gavethe same tonal scale as the 100-watt lamp, and residualvoltages after exposure through a step tablet were allwithin the limits of experimental error.

These residual voltages were measured by using a simplehand probe (as is used with a conventional VON). Apiece of insulating spaghetti is slipped over the metalend of the probe to keep the metal probe from making con-tact with the sheet under test. The probe can be cali-brated in the same way as the 4" x 4" x 8" probe, thatis, by using a metal plate held at a fixed D.C. potential.The insulation spaghetti over the metal end of the probeis lengthened or shortened until the electrometer readsthe actual voltage on the D.C. test plate. This is avery useful hand probe for the exploration of small areas.In this work we note no departure from the reciprocitylaw, and we can expect no improvement in quality by usingstronger or weaker light sources within the ranges triedin this experiment.

1OC. Effect of Alternate Charge-Discharge Cycles onTonal Qualty,.

The procedure was to charge a sheet tý- saturation volt-age and then to expose it under a step tablet at 60-footcandle seconds. Residual voltage was then measured with

!

-45-

the small hand probe described under 9C. After 70 se-conds for dark adaption the sheet was recharged and asecond exposure and measurement was taken. The proce-dure was then repeated a third time. The voltages ob-served seemed to be a little lower on the second cyclebut recovered to their full value on the third cycle.In other words, we observed no shrinkage or expansionof the gray scale by alternate charge-discIiarge cycles.The fact that we were using the fairly conductive binderS5-B (as contrasted with a silicone) accounts for thefact that we were able to, in a large extent, eliminatefatigue. High conducting resins relax (dark adapt) morerapidly. If we had been using a less conductive resinthe system would have had a tendency to reach satura-tion at a lower voltage.

llC. Effect of Doping Zinc Oxide During Manufacture onits Continuous Tone Properties.

A comprehensive survey of all types of zinc oxide whichare commercially available (both American and French pro-cess) shows that all zinc oxide is photoconductive. Thedegree of photocoin~ictivity, however, varies widely withthe past thermal history and origin of the zinc oxide.

American process oxide is manufacturedby burning zincvapor in the presence of CO and C02 formed by the re-actions

C + 02 - CO2C + C02 .- w 2C0CO + ZnO -) Zn + CO2Zn + CO2 -j ZnO + CO

Here the zinciferous material, such as roasted zinc sul-fide (converted by the prior roasting to crude ZnO), ismixed with coal or coke, and the mixture is laid on asteel hearth where it is ignited. The above reactionsrepresent only a part of the mechanism. The crude ZnOused, having had no prior refining other than mechani-cal concentration, carries portions of the associatedminerals such as copper, cadmium, magnesium, calciumand other impurities. All of the metals present show

!

-46-

a sufficient partial pressure at the boiling point of zincto cause their volatilization in part or in total. As aresult, American process zinc oxide contains numerous im-purities in variable amounts. The chemical analysis of atypical American process lead-free zinc oxide is shown inTable 8 below.

Table 8

Analysis of Typical AmericanProcess Zinc Oxide*

Zinc oxide 99.0% minimumZinc sulfate 0.20-0.35%Lead 0.01-0.03%Cadmium 0.05-0.10%Sulfur 0.02-0.014%Acidity (SO ) 0.02-0.10%Water soluble 0.20-0.40%Insoluble in HCU 0.05-0.20%Ignition loss (91 hr/760°C) 0.05-0.20%Loss at 110C 0.05-0.15%Moisture 0.•01-0•.6Chlorine 0.01-0.03%Copper 0.01-0.03%Iron 0.01-0.03%

*Mathewson, C.H., "Zinc", ACS Monograph, Reinhold Pub.Co., N.Y., 1959, p. 350.

General speaking, American process oxide shows very poorphotoresponse despite the fact that its bulk and surfaceresistivity does not differ greatly from the French pro-cess oxide.

French process oxide is made by simply burning metalliczinc. Pure zinc metal is manufactured by distillationor by electrolysis. This metal is remelted, raised tothe boiling point and allowed to burn under carefullycontrolled conditions. If the burning'is conducted ina deficiency of air the final product assumes a graycolor characteristic of its content of metallic (un-burned) zinc. If the burning is conducted in a large

I

chamber held at high temperature the zinc oxide parti-cles will grow to larger size. +hus by careful controlof the amount of air used and by the design of the burn-ing furnace, it is possible to produce a wide variety ofzinc oxide types with respect to particle size and shapeand to chemical, physical and electrical behavior.

Table 9

Analysis of Typical French. rocess Zinc Oxide

Lead .002%Cadmium .0003%Iron .0006%Copper 100003%Silver .00010%Chromium .0001%Tin .003%Calcium TraceMagnesium TraceSilicon TraceBoron TraceFluorine TraceArsenic Trace

Doping of the zinc oxide produced by either of the aboveprocesses can be done by moistening the oxide with asolution of the desired doping agent. For the dopingto be effective, however, the treated oxide should beheated to a temperature sufficiently high to allow mi-gration of the doping agent into the ZnOlattice. Forzinc oxide, the temperature at which ionic migration be-gins is also the temperature (about 450 0C) at which part-icle growth begins. Therefore, when the treated oxidehas been heated to a sufficiently high temperature tocause ionic migration, the particle size distributionhas been radically shifted due to the growth of the largeparticles at the expense of the fine particles. Electronmicroscope pictures show quite clearly that "pirating" ofthe small crystallites takes place, and after a heatingperiod of 2-3 hours at 5000C nearly all of the smallerparticles have disappeared.

I

-48-

Doping by this technique when using large single crystalsis, of course an entirely satisfactory procedure butdoping a particulate powder consisting of crystallitesranging from 0.02-0.6 micron diameter results in twoseparate simultaneous phenomena (doping and shift inparticle size distribution).

Being aware of this problem, we decided to do all of ourdoping by introduction of the foreign element directlyinto the crystal via thermal treatment during burning ofthe metallic zinc.

We have constructed a furnace in which a 15-liter quartzflask is heated by a gas flame inside a brick muffle.The top of the flask is closed with a quartz plate inwhich is drilled a 1.25-inch opening. The neck of theflask is surrounded by a heavy ceramic shield which canbe sealed to prevent entry of outside air. The top ofthe ceramic shield is connected to a fan and the exitfrom the fan is connected to a baghouse. When the zincmetal in the flask is heated, the vapors escape throughthe opening in the lid and burn inside the ceramicchamber. Air is admitted in an amount sufficient toburn the zinc vapor. The zinc oxide goes through thefan and is collected in the baghouse. Using this sys-tem we can produce about one kilo of oxide per hour.Usually after one hour of operation under normal condi-tions we add the doping element directly to the moltenzinc in the quartz flask and then continue to run untilenough sample is collected. Using this scheme we havebegun our doping experiments using antimony and arsenic.The antimony addition produced nothing of merit, butthe use of enough arsenic to produce 0.12% As in thezinc oxide gave a product which had a slightly ex-tended gray scale (see Figure 18).

Figure 18 is self-explanatory but it is worth while topoint out the short tonal range of a paper which is nowcommercially available, Bruning No. 32-155 Coptron paper.All efforts have been directed toward the production of a"short-scale" step tablet. Other commercially-availablegapers designed for office copy work are similar in theirbehavior.

1 '9

a4

Ic0

CD 0

LAl

0

0

I 0

00W

0 NL

0 C;0.

< 0

!

-50-

12c. Effect of Larger Particle Size on Tonal Scale

Because previous work indicated that arsenic doping wasbeneficial in the extension of the tonal range we exam-ined this pipent in greater detail. We normally makethree products from a run. The first product is undopedFrench oxide, and this is used as the control sample.Then, without interrupting the operation in any way weintroduce the doping element into the molten zinc pool.The run is then continued until sufficient doped productis produced. A third product is collected when we dis-assemble the furnace, and this represents the oxidewhich has settled out in the flues. An analysis ofthese three products showed the following:

AverageParticle Size

As S0Q3 it~ P j (Microns)

Undoped oxide .001- .046 .0034 .0028 0.280Doped oxide 0.16 .023 .100 .250 0.230Flue cleanings 0.12 .032 .066 .191 0.325

Prints made from the above oxides showed the flue clean-ings to give a slightly better tonal range than thedoped oxide. Since these two oxides are almost identicalin analysis we suspected that the larger particle sizemight be responsible for the better print. A number ofruns were then made in which larger particle size oxidewas produced. It was found that when we produced a largeparticle size we could get a less mottled print with anextended tonal range. This oxide was made using no ar-senic dope and containing only the normal amounts of leadand cadmium encountered in commercial French oxide. Aroutine analysis of the zinc oxide made on this experi-mental run resulted in the following data.

IS~-51-

Table 9

Chemical and Physical Test Dataon Large Particle Size ZnO

Spec. Part.Surf. Diam.

Sample 11b I_ Cd 'A S03. Le/) Al•

Normal product .0045 .0030 .084 3.48 0.309Flue cleanings .0045 .0038 .050 2.23 0.484Combustion chamber

cleanings .0065 .0053 .012 O.604 1.79

When the large particle size oxide was run through theroutine electrophotographic tests we found that its totalcharge acceptance was considerably below that which wehave heretofore considered necessary to produce a goodquality print. The measurements were as follows:

Total Dark After After Resid- Sec. toCharge Decay 1.25 2.50 ual Reach

Sample Accept. Volt. Second Second Volt. Residual

Normal prod. 140 130 55 40 20 8.0Flue clngs. 130 120 40 20 0 7.5Comb. chbr.cleanings 50 40 20 15 10 7.5

As can be seen from Figure 19, a fairly good tonal scalecan be achieved simply by increasing the particle sizeof the zinc oxide. A word of caution must be expressedhere. We speak of an increase in particle size as beingresponsible for the better quality print. This increasedsize may be only a side-effect. To produce the largersize we must operate the furnace at a higher temperature,and the air to zinc vapor ratio is increased. The ther-mal history of the oxide changes because the crystal-lites stay in the "growing" portion of the combustionchamber for a longer time and the collection system ishotter, resulting in longer storage of the oxide at ele-vated temperature. Any or all of these or additionalunlkown factors could be responsible for the betteroxide. We simply refer to it as large-particle zinc

52

Ililk'? ~~W

Al. 41ah'

A~tA

~ ,I*4~4

-W~

Att

4 ,

p4

h-. %Ik

%.t mA

FIGURE 19

-53-

oxide because this is one of the measurable parameters.Proper doping could possibly further enhance the advant-"age gained by the larger particle size.

The direction of future work is certainly clearly pointedout in Figure 20. Here we have presented a graph in whichwe plotted the Kodak step tablet number versus the actualreflection density. Two step tablets were made similarto those shown in Figure 18. Using a Welch Densichrondensitometer with a direct reading optical density scaleand a reflection unit No. 3832A it is possible to readthe step tablets and determine the reflection density.As a closing portion on this report we point out the di-rection for future work by showing a comparison betweenthe results of silver photography and electrophotography.When the density values are compared we see that ourwhites are not pure enough and that the black densitiesbegin to fall off at about one-half the value realized inconventional silver photography.

Density UR20

1 .3 ...... ..

01.2

0.9

z51 0. 1 2 3 Se8Nme

-55-

Discussion

Electrical Resistivity of Zinc Oxide

The data presented in Figure 13 shows the deleteriouseffect of moisture on photoconductive response. As themoisture content of the zinc oxide increases its elec-trical resistivity decreases due to an enhanced elec-trical path between adjacent zinc oxide crystals. Themoisture need not necessarily originate in the zincoxide. We have seen numerous resins which containedmore than enough moisture to seriously injure photocon-ductive response. Moisture can also enter via additionagents, coating machinery or even if the coating mixesare made in a room where both the temperature and humid-ity have reached a high figure.

Zinc oxide is also susceptible to moisture during stor-age. If it is stored in an unprotected condition in ahumid atmosphere it will adsorb carbon dioxide andslowly begin a conversion to zinc carbonate. This doesnot cause any significant change in the electrical re-sistivity of the oxide (in the dark) but its photore-sponse diminishes, and the resistivity under illumina-tion is not as low as an uncarbonated sample. Oxidewhich has begun this conversion can be restored to itsoriginal condition by careful heating to decompose thecarbonate.

Another effect of moisture should be mentioned. Photo-conductive zinc oxide can be slurried in water, fil-tered, dried and reground. Oxide which has had thistreatment does not lose its photoconductivity; if itis thoroughly dried it is as active after the slurry-ing as before. If however tap water is used whichis heavy in mineral salts, It will be found that thecalcium and magnesium salts have been injurious to thephotoconductive response.

The sensitivity of zinc oxide to moisture changes withthe particle size as was shown in Figure 13. Very smallparticle size is extremely susceptible to moisture ad-sorption. Since photoconductive response dimishes withincrease in moisture in a given system we might be led

I

.1 -56-

to believe that the smaller particle size zinc oxidegave a shorter tonal scale. A careful check for mois-ture content will usually show the smaller particle sizezinc oxide to carry the highest moisture and this highermoisture is responsible for the shorter gray scale.

One final word on electrical resistivity of zinc oxide.The resistivity results reported in the literature forzinc oxide range from 1 x lO"L to nearly infinite re-sistance. After making many measurements with and with-out interference filters, we believe that all of the re-sults reported are correct. Almost any value obtainedcan be varied through several orders of magnitude by themeasuring technique. For example, a large single cry-stal (2.5 x 2.5 x 10 mm) produced accidently in ourHillsboro furnace combustion chamber was examined andfound to have a resistivity of about 50 ohm-cm. Thiscrystal was then carefully pulverized and reduced to asize which was roughly comparable to our larger parti-cle size photoconductive zinc oxide. The resistivitywas almost an exact match of the French oxide. We canonly conclude that the resistivity is in the grainboundaries between crystals.

Measurement of Electrical Parameters of Electrophoto-grapBic Coatinks.

The so-called "static" method of teseing a photoconduc-tive coating presents a number of advantages and disad-vantages as opposed to the "dynamic" test. Using aprobe and a recorder it is possible to develop a rou-tine technique which gives extremely reproducible re-sults (±15 volts on a 350-volt test). Also, the chartsdrawn by the recorder (Figure 16) lend themselves tocareful examination and study.

The obvious disadvantage of the "static" test is its in-ability to show the rate at which the charge is acceptedby the coating. A complete and thorough test shouldtherefore probably include both methods utilizing theattendant advantages of each method.

One of the interesting experiments, for example, whichwe have been able to perform with the static test andwhich does not readily lend itself to duplication with

-57-

the "dynamic" test involves the strange behavior of azinc oxide coated sheet which has been corona charged,exposed and then simply stored until no electricalcharge can be detected with our instruments. The ap-plication of a good developer causes the appearance ofan image where no measurable static charge was to befound. Some strange condition, possibly internal po-larization, causes the developer particles to adherein areas where we cannot detect an electrostatic field.

Pre-exposure to cause fatigue is a well recognised phe-nomena, but pro-exposure to cause complete densitiza-tion can also be performed and checked by the "static"method of measurement. If zinc oxide is mixed with ahigh resistivity resin and coated on a shoot and thissheet is charged to caturation (for example, 1000 volts)we would next, in the logical course of events, exposethe sheet to the image. If, however the sheet is givena very short exposure so that the voltage is decreasedto perhaps 900 volts before exposure to the image, itwil be found that no image will develop regardless ofthe exposure used to project the image. A very shortpre-exposure has caused some unknown phenomena to oc-cur in the coating so that it becomes insensitive tothe subsequent light which carries the image. We can-not explain this, but it is possible that the dipolesformed during corona charging are sensitive to lightand collapse during the pre-exposure. The currentwhich passes through the sheet during charging estab-lishes these dipoles (oxygen) and the static fieldnormally maintains this warped condition. When lightstrikes the surface the zinc oxide becomes sufficientlyconductive to allow current flow, and electron-holerecombination occurs. We are inclindd to be .Leve thattwo separate and distinct phenomena occur during expo-sure to light- (a) return of the dipole to its normalstate, and (b) collapse of the static field due to thelowered electrical resistivity of the zinc oxide.Only further work on this problem will furnish theanswer. The first line of attack would be to learnhow to detect the dipole in the presence of the staticfield.

II -58-

Effect of Resin Binder on Continuous Tone

Binders which are relatively non-conductive, such asalkyds and silicones, produce coatings which have thehighest charge acceptance. It is natural to expectthat a highly charged surface would be more effectivein holding toner than a surface with a low charge. Wehave not found this to be true. Experiments withcoated surfaces charged to high voltage have shown thatthe ability to attract and hold toner has very littlerelationship to total voltage on the surface. This canbe demonstrated very quickly by running a comparisonbetween an alkyd resin and Pliolite S5-B using the sameoxide. The alkyd will accept a charge of 600 volts ormore while the Pliolite will accept no more than per-haps 250 volts. Despite this, the Pliolite will showseven or eight steps of gray while the alkyd shows onlyfive or six at the most. We have no explanation. Wecan only suggest again that perhaps internal polariza-tion is responsible and that static charge is simply anaccompanying condition which has been established dur-ing the charging cycle. The developer then sees onlydifferences in polarization instead of differences inpotential.

Effect of DopinK Zinc Oxide

Prior work in this field has been done almost exclu-sively on single crystals of zinc oxide. Unfortunately,the results of this work are not generally applicable toa particulate powder. The electrical resistivity of asample of zinc oxide, both surface and bulk values, re-sides in the grain boundaries between crystals, and theeffect of any doping is very likely to be obscured bythe very high resistivity values of the zinc oxidepowder.

Numerous attempts have been made to produce a more "con-ductive" oxide by doping and/or by heat treatment. Itis true that the resistivity can be changed, but whetherthe oxide becomes "conductive" or not depends on the de-finition of the work "conductive".

!

-59-

Our doping experiments were inconclusive, and not enoughexperimental work was performed to really accumulatedata which would set a pattern. We hope to be able torun the pilot furnace on a routine basis in the nearfuture and thereby try enough doping agents to enableus to set a pattern for their behavior. We have learned,however, that any "dope" will have to be added in grossamounts and that our long-studied technique of dopinggermanium melts will be of little value in predicting theresults which might be expected from doping zinc oxide.

I

I -60.-

CONCLUSIONSAn analysis of the factual data leads to the followingconclusions:

1. In a zinc oxide-resin system, the strongestphotoconductive response will be attained when moistureis carefully excluded from the coating mix.

2. In an unsensitized system, photoconductive re-sponse is slightly more rapid when using larger particlesize zinc oxide.

3. The use of a smaller particle size zinc oxidealows the development of a higher electrostatic chargeon a coated sheet.

4. Less conductive resins, such as silicone inconjunction with small particle size zinc oxide give thehighest possible electrostatic charge on the coatedsheet.

5. Particle size of the zinc ox•le has no notice-able effect on spectral response within the limits spe-cified in the above work.

6. The development of a high electrostatic chargeon a coated surface is not a guarantee of an extendedtonal scale.

7. Shortest tonal ranges are observed with thealkyd resins and longest tonal ranges are attained whenusing Pliolite resins.

8. Experiments on doping of zinc oxide have beeninconclusive,but there is some evidence to indicate thatarsenic might be beneficial. This is clouded by the in-troduction of a second variable, namely, large particlesize.

Future work in this field rc.ful if the doping experimc, tr , wt o 1imore work should be carriJ rjnflcoating mix. From a theor.t c,,l :;tdnI.be helpful if a workable t iev,-y h,,,i i,plong the same lines as i., n. in plj-',w,sulfide. Kallmann at New ftnent of a theory bpsed upt.

.. ization. A •rn1,c "--.apply to zinc oxide