AD

TECHNICAL REPORT ECOM-2517

A STUDY OF FACTORS AFFECTING THE CORROSION RATE OF

MAGNESIUM IN VARIOUS ELECTROLYTES UNDER STATIC CONDITIONS

By

CARL A. NORDELL

VHARD COPY $. /,-6o

[-I CROFICHE

-AUGUST 1964

U. S. ARMY ELECTRONICS LABORATORIESU. S. ARMY ELECTRONICS COMMAND

Fort Monmouth, New Jersey

DDC AVAILABILITY NOTICE

QUALIFIED REQUESTERS MAY OBTAIN COPIES OF THIS REPORTFROM DDC.

THIS REPORT HASBEEN RELEASED TO THE OFFICE OF TECHNICAL

SERVICES, U. S. DEPARTMENT OF COMMERCE, WASHINGTON, D. C.20230, FOR SALE TO THE GENERAL PUBLIC.

The findings in this reportare not to be construed-as anofficial Department of the Armyposition, unless so-designatedby other authorized documents.

,(Destroy -this report when it is no -longer needed.

Do-not return it to the originator.)

TECHNICAL REPORT ECOM-2517A STUDY OF FACTORS AFFECTING THE CORROSION PATE OF

MAGNEIIt IN VARIOUS ELMTROLYT! UNDER STATIC CONDITIONS

by

Carl A. Nordell

Power Sources DivisionElectronic Components Department

DA Task No. lAO 13001 A 039-19

Abstract

The corrosion rates of magnesium and magnesium alloys in wagnesiumsalt solutions of chloride, bromideand perchlorate were studied bymeasuring the gaseous reaction produc as a function of timez Reliableapparatus was developed to accurately determine the- volumes of gasgenerated. Protective films were applied to the surface. of the magnesiummetal to suppress excessive chemical reactivity at the metal-electrolyteiuterface , thereby reducing static corrosion rates. Cramolin, one ofthe films tested, reduced the gas evolution of an AZ21 Mg-2N MgBr 2 systemfrom Wo cc/800 hrs to 0.1 ce/800 hrs.

U. S. ARMY ELECTRONICS LABORATORIESU. S. ARMY ELECTRONICS COIAWND

FORT I40IOUTH, N. J.

i

CONTENTS

ABSTRACT ±

INTRODUCTION I

EXPERIMENTAL PROCEWRE 1

a. Test Equipment 1b. Preparation of Sample 2c. Accuracy 2

DISCUSSION 2

CONCLUSIONS 4

Figures

1. Gas Collecting Apparatus 52. Microvolumetric Gas Collecting Apparatus 63. Electrolyte Concentration vs pH 74. Gas Evolution vs i of Several Electrolytes 85. Gas Erolution vs Elect-olyte Concentration 96. Gassing Rates of AZ-21 Magnesium in Four Normal Magnesium 10

Bromide with Protective Coatings7. Btfee of Dilute Cramolin on the Gas Evolution for AZ-21 Ui

Magrnesium in MgBr 2 and Mg(CiO ) Electrolytes8. Eff6ct of Benzene to Cramolin iatlo on Reaction Rate of 12

AZ-21 Alloy with Mgr 2 Electrolyte

Table

I. Gas Evolution in Milliliters per Square Inch of Magnesium 13During 400 Minutes Stand in Various Electrolytes

iii

A STUD! OF FACTORS AFFECTING THE CORROSION RATE OFMAGNESIUM IN VARIOUS ELECTROLYTES UNDER STATIC CONDITIONS

INTRODUCTION

In a magnesium-manganese dioxide primary cell, water in theelectrolyte reacts with the magnesium in accordance with the reaction

Mg + 2H20 ~~ Mg(OH) 2 + H2

to form a protective film of magnesium hydroxide at the anode. Althoughthis hydroxide film enhances the storage life of magnesium dry cells, itinduces undesireable transient polarization (delayed-action) when ananodic current is applied to the magnesium anode.

This study is based on the postulation that the application ofmicro-thin ionic conductive films to the magnesium anode will produce anoverall advantageous effect on the shelf life and delayed action of themagnesium dry cell. It was assumed that the protective fiLlms wouldprovide the effective separation of active materials so essential toprolonged shelf life, and would inhibit the formation of magnesiumhydroxide on the anode, thereby, reducing initial transient polarization.

The objective of this program was to study the chemical reactivityrate between magnesium and 'various electrolytes on stand and to investi-gate means of inhibiting thic reaction,

The method of investigation in the study involved static tests inwhich a magnesium alloy of known composition was placed in an enviromentof electrolyte. The gas generated by the resulting chemical reaction wascollected and measured as a function of time. The reaction

Mg + 2H20 -* Mg(OH)2 +

indicates that for every mole of Mg(OH)2 that is formed one mole of H2is evolved. It then follows that if the gaseous reaction product iscollected and measured for a specific period of time, the amount ofMg(OH) 2 formed during that period can readily be calculated.

a. Test Equipment

Large volumes of gas were collected in accordance with the dis-placement of liquid under glass technique. A 50 milliliter burette wasinverted in a 100 milliliter beaker of electrolyte. The burette wasthen filled with electrolyte by the evacutwtion of air through the stopcock. To prevent a change in electrolyte concentration due to evapor-ation, paraffin oil was used to cover the surface of the electrolyteremaining in the beaker (Figure 1).

1

To accurately measure very small volumes of gas, a special glasscell was constructed. This apparatus consisted of a 14/35 taperedground glass joint that had been fitted at one end with a 1.0 cubiccentimeter pipette and sealed at the other end to form a reaction chamberhaving a volume of approximately 10.0 cubic centimeters. The pipettewas extended into the reaction chamber well below the joint so that thecapillary of the pipette would always be below the level of the electro-lyte. In operation, the gas generated displaces an equal volume ofelectrolyte from the reaction chamber. The electrolyte is forced intothe calibrated pipette allowing measurement of the gas volume. Adetailed sketch of this apparatus is shown in Figure 2.

b. Preparation of Sample

All magnesium samples used in this investigation had a surfacearea of approximately one square inch and were cleaned in trichloro-ethylene before testing.

c. Accuracy

All of the graduated glassware used was in accordance withspecifications of the National Bureau of Standards. The burettes usedwere graduated from 0 to 50 milliliters in divisions of 0.1 milliliters.The 1.0 cubic centimeter pipettes were graduated in subdivisions of0.01 cubic centimeters making it possible to read accurately to + 0.002of a cubic centimeter. However, when using gas measuring devices certainrelative factors must be considered, especially when small volume changesare being determined. For instance, changes in temperature andatmospheric pressure will result in a substantial deviation from truetest measurements. Accordingly, these experiments were conducted in aroom where the temperature was maintained at 210C + 0.50C. By plcingthe reaction chambers of the test cells in an ambient temperature waterbath, the 0.50 C temperature deviation was reduced to approximately+ 0.25°C. Since atmospheric pressure is somewhat more difficult tocontrol, a method of numerical compensation was devised, An additionalmicrovolumetric gassing cell, containing only the type and concentrationof electrolyte being studied, was positioned in the water bath. Anychange noted in the electrolyte level in this calibrated pipette wasattributed to a change in atmospheric pressure and all other pipettereadings were compensated accordingly.

DISCUSSION

The objective of this program was to study the effect of electrolyteand anode composition on the open circuit corrosion rate of magnesiumelectrodes and to investigate fatty acid and silicon films as corrosioninhibitors. The results of an experiment designed to evaluate thecorrosion rate of magnesium and magnesium alloys in aqueous saltsolutions of various concentration are shown in Table I. Increasingthe concentration of the electrolyte solutions resulted in increasedreactivity. Data in Table I also shows that the alloy ingredients havea marked effect on the static corrosion characteristics at all normality

2

levels of the electrolyte solutions tested. Generally, as the aluminumcontent of the magnesium alloy increases the reactivity decreases.

MagnesiumAlloy % Al

AZ-10 1 0.5AZ-21 2 1AZ-31 3 1

Aluminum is added to magnesium to reduce local action caused by thepresence of iron, the major impurity in magnesium. The addition of zincto the alloy mixture results in a more uniform surface dissolution ofthe magnesim alloy.

To determine the relationship between the pH of the test electro-lytes and the corrosion rate, pH values were plotted as a function ofboth electrolyte concentration and gas evolution, (Figures 3 & 4). Asthe electrolyte concentration increases all solutions become moreacidic. It can also be seen that the pH range of magnesium perchlorateis much greater than the other electrolytes; consequently, it is the mostacidic at high concentrations and the most basic at low concentrations.Of interest in Figure 4 is the fact that the gassing rates tend toequalize as the electrolytes approach neutrality (Pi 7). Figure 5illustrates that the chemical reactivity of magnesium perchlorate withmagnesium is lower than that of magnesium bromide and magnesiumchloride, and that gas evolution decreases as the electrolyte concentra-tion decreases. The Jata also emphasized the advantage of magnesiumalloys over ire magnesium.

Under an investigation of anode coatings the following materialswere tested:

Cramolin Silicone Varnich #997Lanolin SilicladRicilan "C" Oleic AcidSilicone SF-96 PolycoSilicone R-1O

Unless otherwise stated.all tests were conducted using a 4 N solutionof magnesium bromide as the electrolyte and AZ-2! -- gne ium all-o- It

can be seen from Figure 6 that all of the coatings did retard reactivityas evidenced by the decrease in gas evolution. Cramolin, a commercialelectric contact cleaner containing fatty acids, proved to be the mosteffective. The effectiveness of Cramolin in reducing the corrosionrate of magnesium and magnesium alloys in various ele trolytes isfurther demonstrated in Figure 7.

Experiments were conducted on methods and techniques of applyingCramolin films to magnesium. Benzene, a solvent for Cramolin, was usedas a diluent to control film thickness. Cramolin solutions were appliedto the magnesium test sample by brushing, dipping or soaking the anodematerial for varied periods of time. Magnesium samples that had been

3

immersed in solution for at least two weeks gave the best results.Further experimentation resolved that the soaking process could beaccelerated by using an ultrasonic vibration bath. When a sample wasremoved from the Cramolin solution, it was either air dried at roomtemperature or oven dried at temperatures ranging from 40 to 60PC.Both drying techniques were investigated through a time range of 1 to18 hours. Oven drying at 55 to 600c for periods between 8 and 18 hoursproved to be the most favorable. To optimize the ratio of benzene toCramolin required to form the most effective film, microvolumetricgassing tests were conducted on magnesium test samples in accordancewith the following schedule:

Benzene Cramolin

0 to 15 to 17 to 19 to 1

11 to 1

The results of this experiment are shown in Figure 8. Maximum filmprotection was attained using a solution of 7 parts of benzene to 1 partof Cramolin.

CONCLUSIONS

The magnitude of the corrosion reaction between magnesium andaqueous salt solutions is a function of the following:

Electrolyte concentrationElectrolyte compositionAlloy ingredientsInhibitors

Of all the inhibitor materials tested, Cramolin, when properlyapplied, most effectively reduces the corrosion of magnesium duringstand in aqueous salt solutions.

Cramolin films applied to the magnesium anode limit reactivity atthe magnesium-electrolyte interface under static test conditions byhindering the accessibility of the electrolyte to the metal surface.Additional study will be required to determine the effect of thispassivity on initial transient polarization when an anodic current isapplied to the magnesium anode.

14

9

3=

[II

GAS COLLECTING APPARATUS

FIG. I5

'1 III

II

I II

I Ii

I I

I IIIl III I

II I I!1 IIi I II I I

I I II I

I I II I I

II1"

II i

I S! ([U}

MICROVOLUHETRIC GAS COLLECTING APPARATUSFIG. 2

6

7-

6-z0 Mg Br 2

<'- 5-I-z-

z~.

0w r

2Jw

w I

5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0;)H.

ELECTROLYTE CONCENTRATION VS. pH

FIG. 3

T

108-

96 4N 6N

84-

mg BM2ul72 4N

24

FE60 -

MgCI2M1zI0p 48

0

0

24-

22

5.0 5.5 6.0 6.5 70 7.5 8.0 8.5 9.0pH.

GAS EVOLUTION IN MILLILITERS/IN 2 OF PUREMAGNESIUM DURING '100 MINUTES IMMERSION VS. PH OF

SEVERAL ELECTROLYTES

FIG. 4

8

6- AZ-31 AZ-10IUR

I-I

5 - PURE 2

4-

23o 2- Mg C12z

0 20 40 60 80 100 .120 140 160GAS EVOLUTION (MILLILITERS)z

0

Ir 7-AZ-31I_6- jAz-21 AZ-10 XPUREzI

W -0

W z2

0 040 60 80 1020140 160i-- GAS EVOLUT ION (MILLILITERS)

j AZ-31w 7 /-AZ-21// I PURE>_ 6- bz-10

n30

Z2Mg (Cl 04 )2 -

20 40 60 80 100 120 140 160GAS EVOLUTION (MILL-ILITERS-)

GAS-EVOLUTION DURI-NG 1100 MINUTE TEST VS.ELECTROLYTE CONCENTRAT ION

FIG. 59

I.0

0O.6 C

L0.2 sIICH L E LI C A~CID

0 I2 3 4 5 6

TIME (HOURS)

GASSING RATES OF AZ-21 MAGNESIUM IN FOURNORMAL MAGNESIUM BROMIDE WITH PROTECTIVE

COATINGS

FIG. 8

10

.0

9~ 01

LuJ

O ~Mg (C 10) WITH CRAMOLI N

TIM05

EFFECT F 7 LIT (HOURS) 60 70

HAGNESIUH M 8 IN _e THE GAS EVOLUTIO1 FO2 & AfQ(CIo0 2 ELECTROLYTES A-2

FIG. 7

ALL MAGNESIUM SAM4PLES SOAKED IN CRAM4OLIN SOLUTIONS FORFOUR WEEKS AND THEN OVEN DRIED *6O0C FOR 18 MRS.

2.0-

W 1.5-

0

w-j0> 0.5w

0100 200 300 400

TIME (HOURS)

-EFFECT OF BENZENE TO CRAMOLIN RATIOON REACTION RATE OF AZ-21 ALLOY WITH MgBr2

ELECTROLYTE

FIG. 8

12

Cl) 00 0 C4U 0 LoLe"V)p- 00 91 * a 0 0 0 0 S

'I, -10 C#_ _ _

AK C\I &D 0O0 C4 LLn )0 MJ'0 0: : 9 -: - 9- 1. 2 2 1

wO 04+

f.-J -- 0C -Ln

c-4-

(flw

0M -d

C')

WD w

IxJ

W C0 L 440 LW~e~ wr 4t c ;

I- clJ C4 C1 0 t

1.

IrI

ww

13j

- a9

I -III

d~

A4 A

It I1 t8 -p I

I 4I

k ~ ~f 7I~ ~ i102~ ~ 1i~i~~jJ~~iI~ii