8/3/2019 728174

1/51

A THERMOPLASTIC RANSPARENTADHESIVEFOR BONDING POLYCARBONATEO GLASSJuly 1971 ,

GEORGE L. BALL Ill, PHILIP H. WILKEN,CHARLES 3. NORTH and IVAL 0. SALYERMonsanto Research CorporationDayton LaboratoryDayton, Ohio 45407

Final Report - Contract DAAG46-71-C-0007

Approved for public release; distribution unlimited.

ARMY MATERIALS AND MECHANICS RESEARCH CENTERWatertown, Massachusetts 02172

8/3/2019 728174

2/51

The findings in this report are not to be construed aaan official Department of the Army position, unless sodesignated by other authorized documents.Mention of any trade names or manufacturers in this reportshall not be construed as advertising nor as an officialindorsement or approval of such products or companier bythe United States Government.

DISPOSITION INSTRUCTIONSDeatro tbia6 0 not report wbea it is no longerreturn it to tbe originator, aeedad.

8/3/2019 728174

3/51

AMMRC CR 71- lo

A THERMOPLASTIC TRANSPARENT ADHESIVEFOR BONDING POLYCARBONATE TO GLASS

July 1971

George L.Charles J. Ball III, Philip H. Wilken,North and Ival 0. SalyerMonsanto Research CorporationDayton LaboratoryDayton, Ohio 45407

Final Report - ContractD/A Project IF162205AA52 DAAG46-71-C-0007AMMRC Code 512E.ll.A52Aircraft Survivability

Approved for public release; distribution unlimited

Prepared for

ARMY MATERIALS AND MECHANICS RESEARCH CENTERWatertown, Massachusetts 02172

8/3/2019 728174

4/51

ABSTRACTThe extent of the usefulness of glass and polycarbonate in armoredwindows depends both on the properties of these two materials, aswell as the availability of transparent adhesives for bonding thepolycarbonate to glass and to itself. Until now, no completelysuitable transparent adhesive of the thermoplastic type has beenavailable. Accordingly, ethylene terpolymers varying in molecularweight, hydroxyl content, and degree of partial crosslinking wereinvestigated. The primary requisites for the adhesive were that itbe thermoplastic, optically clear, hydrolytically stable, andmaintain the integrity of a glass-polycarbonate laminate whenexposed to either thermal cycling from -65 up to +165OF or ballisticimpact. An ethylene terpolymer material identified as ETA #138150was made available which more than fulfilled the performance require-ments.

iii

8/3/2019 728174

5/51

TABLE OF CONTENTS

1. INTRODUCTION2. SUMMARY3m EXPERIMENTALDISCUSSION

3.1

3.2

3.3

Abstract

PREPARATIONAND PHYSICAL CHARACTERIZATIONOF THE BULK ADHESIVES3.1.1 Synthesis and Analytical Character-ization3.1.2 Thermal Analysis3.1.3 Mechanical Results3.1.4 Thermal Cyclic AnalysisDEVELOPMENT ND CHARACTERIZATIONOF THERMALFORM STABILITY3.2.2 Physical Characterization302.3 Thermal Cyclic Exposure and LaminateIntegrity3.2.4 Ballistic AdhesionLAMINATING TECHNIQUESAND CONDITIONS3.3.1 Laminating Technique - Screening3.3.2 Polymer Melt Flow Characteristics3*3*3 Optimized Laminating Process-Air Autoclave3.3.4 Secondary Laminating Process-Hydraulic Press

Cards

Page12777

103162121212727292938

Distribution ListDD FORM 1473

iV

8/3/2019 728174

6/51

LIST OF FIGURESFigure

1.

2.

3.

4.

5.

6.

8.

9.

10.11.

Glass-polycarbonate Laminate Bonded with theOptimized Transparent Ethylene TerpolymerAdhesive 138150Photographs of the Ethylene Terpolymer Sheet(138150) Delivered to the U. S. ArmyMolecular Weight Distribution Curves for theTwo Primary Candidate Ethylene TerpolymerSystemsModulus-Temperature Characteristics of theSix Candidate Ethylene TerpolymersTemperature-Elongation (Zero Tensile Strength)Characteristics of the Six Candidate EthyleneTerpolymersThe Effect of Partial Crosslinking on theModulus-Temperature Characteristics of theEthylene TerpolymerThe Effect of Partial Crosslinking onElongation-Temperature Characteristics ofthe Ethylene TerpolymerOptical Transmission Characteristics of theOptimized Ethylene Terpolymer Adhesive 1381.50Photographs of Transparent Laminates FollowingBallistic Impact?ressure-Vacuum Autoclave for the Preparationof TYansparent Laminates (schematic).Melt-flow Characteristics (30 mil channel) ofthe Optimized Transparent Ethylene TerpolymerAdhesive 138150

Page4

6

4

11

12

19

20

23

26

28

V

8/3/2019 728174

7/51

LIST OF FIGURESFigure

12. Cleaning and Drying of Glass and PolycarbonatePlates

13. Lay-up and Packaging of Laminates14. Bagging of the Lay-up15. Closure and Sealing of the Vacuum Bag16. Closure and Sealing of the Autoclave17. Removal and Inspection of the Fabricated Part18. Typical Laminating Cycle for the Optimized

Ethylene Terpolymer Adhesive 138150

Page32

333435363739

vi

8/3/2019 728174

8/51

LIST OF TABLESTable Page

1 Physical Characteristics of the Optimized Transparent 3Ethylene Terpolymer Adhesive 138150

2 Physical Characteristics of the Candidate Ethylene 8Terpolymer Adhesives3 Performance of Candidate Ethylene Terpolymer 15Adhesives Subjected to Thermal Cycling4 Analytical and Thermal Characteristics of the Partial- 18ly Crosslinked Transparent Ethylene Terpolymer Adhesive138150

Physical Characteristics of the Partially CrosslinkedTransparent Ethylene Terpolymer Adhesive 13815022

Characteristics and Results of Transparent Laminates 25Bonded with the Ethylene Terpolymer Adhesives Follow-ing Ballistic Impact

vii

8/3/2019 728174

9/51

8/3/2019 728174

10/51

1. INTRODUCTIONMajor improvements in glasses and polycarbonate plastics have occurred recent years which make them most suitable for ballistic applications.Polyvinylbutyral, a proven adhesive through use in transparent safetyshields, also continues to provide an effective means for bondinglayers of glass of various types together. Polyvinylbutyral, however,has not been prepared in a completely suitable form for use in atransparent polycarbonate laminate.A program was designed to define, on a limited scale, the best one ofsix ethylene terpolymer adhesives for bonding polycarbonate to glass.This adhesive was to be used in laminates which could be subjected toballistic impact. However, the purpose of the adhesive was only tomaintain the integrity of the laminate, and not necessarily to pro-vide any particular impact characteristic.The program included the synthesis of six compositions, in a narrow,well-defined region that was bracketed in previous experimental work(ref. 1). These six basic compositions were to be characterized andimproved thermal form stability provided to the best terpolymerthrough a partial crosslinking. It was intended, however, that thematerial remain a thermoplastic.A major consideration for the adhesive was that it maintain theintegrity of a glass-polycarbonate laminate over the temperaturerange from -65OF up to t165F. Finally, a set of laminating condi-tions for the best ethylene terpolymer adhesive was to be establishedand thirty sheets of this material delivered to the U. S. Army.

Ref. 1. G. L. Ball III and I. 0. Salyer, "Development of a TransparentAdhesive Compatible with Polycarbonate for Use in BallisticShields," Technical Report AFML-TR-70-144, June 1970.

8/3/2019 728174

11/51

2. SUMMARYA study was conducted on six ethylene terpolymers encompassing twomolecular weight levels and three hydroxyl content levels. Eachof these terpolymers was characterized physically. All were foundto be good transparent adhesives, but their relative merit wasestablished by their ability toward maintaining the integrity of aglass-polycarbonate laminate at temperatures downto -65F. In nocase did the adhesive"fai1, however, the compositions with thehigher hydroxyl contents caused failure of soda-lime glass in thelaminate. This was due to the inability of the adhesive to transferthe severe strain caused by the different degrees of contraction ofthe glass and polycarbonate,The best unmodified ethylene terpolymer passed the low temperaturerequirements. It had a molecular weight of 105,000 and a hydroxylcontent of 2.1%. This transparent adhesive would be usable in aglass-polycarbotiate laminate over the temperature range from -65Fup to 145'F.Increased thermal form (dimensional) stability was provided to theethylene terpolymer adhesive through a partial crosslinking. Thisprovided an adhesive usable in a laminate from -65F up to 1650~,and the thermoplastic character was maintained. The physicalcharacteristics of this optimized t::ansparent ethylene terpolymeradhesive (#138150) are summarized in Table 1.The characteristics listed in Table 1 do not fully describe theexcellent broad utility of this transparent adhesive. A moredetailed description of the various parameters investigated, includ-ing the effects of time, temperature, elongation and wavelength, isgiven in graphic form in the experimental discussion.The ability of the adhesive to maintain the integrity of glass-polycarbonate laminates from -65OF up to t165F was confirmedthrough repeated thermal cycling. A typical exposed specimen isshown in Figure 1 (a). The utility of the adhesive in maintainingthe integrity of a glass-polycarbonate laminate upon impact with acaliber 0.222 projectile, at room temperature, was also demon-strated. An impacted laminate is shown in Figure 1 (b). Eventhough the glass fractured and spalled, no adhesive or cohesivefailure of the ethylene terpolymer occurred.Sheets measuring 8 in. x 8 in. of the optimized transparent ethyleneterpolymer adhesive 138150 were prepared in thicknesses from 10 to30 mils. Thirty of these, shown typically in Figure 2, were deliveredto the U. S. Army, AMMRC, Watertown, Mass.

8/3/2019 728174

12/51

Table 1PHYSICAL CHARACTERISTICS OF THE OPTIMIZED TRANSPARENT

ETHYLENE TERPOLYMER ADHESIVE 138150Property

Hydroxyl Content, %Inherent ViscosityMelt Index, g/10 min.LaminatingTemperature, OFPressure, psiThermophysical ProfileT45,000, OF

T675, OFSR, F"Resistance to Heat, OFZero Tensile Strength, OFLaminate Adhesive Utility TemperatureUpper, OFLower, OFTensile D4126Strength, psi D412Modulus, psi D412Elongation, % D412StrengthImpact (Tensile), ft*lb/in.sqTear, lb/in. thicknessAdhesive, psiBallistic AdhesionDensity, g/ccColor (Laminate)Transmission, % 450 mv550 mFr600 mu

ASTMMethod

D2857D12381

D10432D1043D1043D1043

- 3Dl6_37;

D792-10

.2.10.80.7

23015-28

3765275145

+165-6525005oo1400

600160480Good0.98

89-9090

Method A, condition c [150c, 1620 g, 411. 0.316/0.06312Indicates temperature at modulus shown, stifflex range (SR) iSdifference between two moduli3Contlnuous In air without dlscoloratlon*Indicates load bearing characteristics only at break [not 2%)Based on thermal cyclic exposure of glass-glass andglass-polycarbonate laminatesbSpec1mer-i C, 20 in./mln. crosshead, modulus at 10% elongation(39 mil thick)Specimen was type C, ASTM 9412LO.O3 In. thicknessq0.005 in.gluellne thickness, 0.5 in. overlap, bonded to aluminum,failures 100% adhesive, crosshead speed 0.05 ln./mln.lODetermined on PEE Model 450 spectrophotometer, 0.04 in. thickbetween 2 pieces of quartz 0.12 in. thick. Reference air.

3

8/3/2019 728174

13/51

SAMPLE NO. DATE

l-

a, OPTICAL TEST OPTICAL TEST; OPTICAL TEST

2OPTICAL TEST l-u 4

; OiTlCALTESTOPTKilL:1 TTKAL TEST OrlTJCAL TEST OPTICAL T: &TICAL TEST OPTICAL TOk PTICAL TEST OPTICAL TES .d..&.ln-OPTICAF; TEST- - -_ -.--OPTICAL TiiSMONSANTO MONSANTO---(a) Following thermal cycling from-65 to ti65OFSAMPLE NO. DATE

MONSANTO MONSANTO(b) Following impact with caliber0.222 Projectile at 1000 ft/sec

Figure 1. Glass-polycarbonate LaminateBonded with the Optimized TransparentEthylene Terpolymer Adhesive 138150

4

8/3/2019 728174

14/51

Figure 2 . Photographs of the Ethylene Terpolymer :(138150) Delivered to the U. S. Army 5heet

8/3/2019 728174

15/51

3. EXPERIMENTAL DISCUSSION3.1 PREPARATION AND PHYSICAL CHARACTERIZATION OF

THE BULK ADHESIVESThe synthesis and characterization of the bulk adhesives involvedthe preparation of batches of material in minimum quantities of200 g9 thus their character could be determined analytically andthen physically through thermal, mechanical, and dynamic thermalconditions, It was the purpose of this characterization to deter-mine the best of six candidate systems to be used in providing an 'improved interlayer adhesive. The exact nature of the ethyleneterpolymer adhesive is proprietary, however, the candidate materialswere based on compositions of two different molecular weights, eachof which was modified to three hydroxyl content levels.3.1.1 Synthesis and Analytical CharacterizationIt was first necessary to establish the molecular weights to assurethat the materials provided a range of suitable spread. The weightaverage and number average molecular weights were determined at zerohydroxyl content by gel permation chromatography (GPC). Theseresults, shown in Table 2 and Figure 3, indicate about 20% differ-ence in the weight average molecular weight. A third polymer oflower molecular weight is also shown which was subjectively elimin-ated due to its obvious poor strength and high degree of tackiness.It will be discussed no further.Each of the two systems was modified to adjust its hydroxyl contentover the range from 2.1% to 4.1% and the content was established.These results are shown in Table 2 indicating three levels.In order to validate the physical differences in the six candidatepolymers, and also to establish the relative melt-flow characteris-tics, the inherent (solution) viscosities, and melt indices weredetermined. These are also shown in Table 2. The inherent viscos-ities and melt indices are an indication of both the molecularweight and the hydroxyl content of the terpolymers.The melt indices were expected to vary with molecular weight and toincrease (exhibit increased flow) as molecular weight decreased.The almost order of magnitude change in melt index, however, wasnot anticipated for only a 20% change in molecular weight. Theseresults, however, were quite indicative of others which were tofollow. In particular, a high degree of tackiness, relativelylow strength, and severe handling problems were found with the lowermolecular weight product even at its highest hydroxyl content.

8/3/2019 728174

16/51

Table 2PHYSICAL CHARACTERISTICS OF CANDIDATE ETHYLENE-_TERP-OLYMER.ADH&SIVES

SAMPLE laID NumberHydroxyl Content, %Molecular Wt, wt av !l)Molecular Wt, no. av (1)Inherent Viscosity (2)Melt Index,&10 min. (')

Z3i30105,00030,0000,5649.5

Thermophysical Profile (4)

SR, F"Zero Tensile Strength, OF (5)Tensile (6,')Strength, psi 21020

Modulus, psi 21020Elongation, % 21020

Impact (Tensile) Strength (8)ft.lb/in2Tear Strength, lb/in.thicknessDensity, g/cc (9)Color (qualitative)

lb:3:4g105,000

3yg7:5

lc;3"142.105,000

3:%:6:4-30

5585

144

?a c1375142.189,60023,4000,44066.0

2bi3Z5""89,60023,400013495465

2ck3L508g ,60023,4000,247310

-35 -3015 4050 70

119 138

-47 -33 -33-4 5 1843 38 5191 99 120

115016501700290031002700

700100012508%740385

215022002500140017501250

;;z600

g90130230

105021003200340

80 400180 550200 750330360340

160019002000400

1601802502700>2goo>IJooo

560

320430520185019002000

220 655110 190 220 10 40 70

0.98 0.98 0.98 0.99 0.99 0699good good good good good good(1)Measured at zero hydroxylcontent by Gel permeation chromatography (QPC). See Figure 3(2)ASTM D2857(3)ASTM D1238 Method "A" - 150C - ASTM 4/l orifice ,083" dia x ,316" long(4)ASTM D1043(5)ASTM Dl637(6)ASTM D412, Specimen C - strength is maximum, not necessarily load to break, Modulus isinitial at slO% elongation(7)The numbers 2, 10 and 20 refer to test speed in in./min,(a)ASTM 1822 with Specimen C (ASTM D412), Test velocity 11 ft/sec, Test 30 - 60 ft lb range(9)ASTM 792

8/3/2019 728174

17/51

000 u000 a43 o\oo c!JE GFY hoc.ocu p

.r-__ . ,_I_I ~-.. -,..._..-:--. . ,-