Laboratory salt spray testing has been used for many years to accelerate natural atmospheric exposures with limited success. The corrosion of silver represents a stark example of the failings of such testing. In virtually any field environment, silver corrodes measurably in a month, but after 4 months in a standard salt spray test, no corrosion of silver is observed. The present work is concerned with the modification of the conventional salt spray corrosion test and its correlation to field measurements of silver. Two subsystems were designed and constructed to modify a conventional salt spray chamber. One subsystem consisted of ultraviolet A (UVA) lamps mounted on a movable rack, which allows them to be positioned to achieve the desired UVA light intensity. The other subsystem consisted of an ozone system based on a commercial ozone generator and a distribution manifold that allowed a uniform concentration of ozone to be maintained throughout the chamber. As constructed, the system can produce 50 W/m2 of UVA light and ozone concentrations up to 23.2 ppm. The system was then used to modify the protocol of ASTM B117. The results from the modified B117 test are correlated with the corrosion behavior and performance observed in field exposures. The corrosion products formed during the modified B117 exposure are the same as those observed after field exposures. Silver chloride (AgCl) was the main dominant corrosion product. Non-uniform corrosion occurred, and metallic silver grains were observed on the surface. The large acceleration factors obtained demonstrate that both ozone and UVA light can be used to replicate the type of corrosion found on silver at a wide range of geolocations. The extent of the acceleration can be controlled by the ozone concentration and the intensity of the UVA light.

Figures - uploaded by R. G. Kelly

Author content

All figure content in this area was uploaded by R. G. Kelly

Content may be subject to copyright.

ResearchGate Logo

Discover the world's research

  • 20+ million members
  • 135+ million publications
  • 700k+ research projects

Join for free

CORROSION ENGINEERING SECTION

036001-1

CORROSION—Vol. 68, No. 3

Submitted for publication July 15, 2011; in revised form, October

18, 2011.

‡ Corresponding author. E-mail: rgk6y@virginia.edu.

* Department of Materials Science and Engineering, University of

Virginia, Charlottesville, VA 22904.

** School of Materials Science and Engineering, Shenyang Jianzhu

University, Shenyang 110168, China.

Modification of ASTM B117 Salt Spray Corrosion

Test and Its Correlation to Field Measurements

of Silver Corrosion

Y. Wan,*,** E.N. Macha,* and R.G. Kelly‡,*

ABSTRACT

Laboratory salt spray testing has been used for many years

to accelerate natural atmospheric exposures with limited suc-

cess. The corrosion of silver represents a stark example of the

failings of such testing. In virtually any field environment,

silver corrodes measurably in a month, but after 4 months in

a standard salt spray test, no corrosion of silver is observed.

The present work is concerned with the modification of the

conventional salt spray corrosion test and its correlation to

field measurements of silver. Two subsystems were designed

and constructed to modify a conventional salt spray chamber.

One subsystem consisted of ultraviolet A (UVA) lamps mounted

on a movable rack, which allows them to be positioned to

achieve the desired UVA light intensity. The other subsystem

consisted of an ozone system based on a commercial ozone

generator and a distribution manifold that allowed a uniform

concentration of ozone to be maintained throughout the cham-

ber. As constructed, the system can produce 50 W/m2 of UVA

light and ozone concentrations up to 23.2 ppm. The system

was then used to modify the protocol of ASTM B117. The

results from the modified B117 test are correlated with the

corrosion behavior and performance observed in field expo-

sures. The corrosion products formed during the modified

B117 exposure are the same as those observed after field

exposures. Silver chloride (AgCl) was the main dominant

corrosion product. Non-uniform corrosion occurred, and metal-

lic silver grains were observed on the surface. The large accel-

eration factors obtained demonstrate that both ozone and UVA

light can be used to replicate the type of corrosion found on

silver at a wide range of geolocations. The extent of the accel-

eration can be controlled by the ozone concentration and the

intensity of the UVA light.

KEY WORDS: accelerated testing, atmospheric corrosion, cou-

lometric reduction, environmental parameter, salt spray test-

ing, silver

INTRODUCTION

Conventional accelerated corrosion tests (e.g., ASTM

B1171 ) are often used to accelerate the natural marine

atmospheric corrosion of materials for material selec-

tion or quality control. However, results from ASTM

B117 testing often do not correlate well with field

exposures,2-8 requiring caution in using the results for

prediction, as pointed out in the standard.1 A prime

example of the shortcoming of this testing is the cor-

rosion of silver. When silver is exposed to actual atmo-

spheric environments, it forms a silver chloride (AgCl)

film with a thickness that depends on the corrosiv-

ity of the exposure site and exposure time, with mea-

surable amounts forming in less than 30 days.2,9 In

contrast, when silver is exposed to the ASTM B117

standard test protocol, no AgCl forms even after four

months of continuous testing.2 Clearly, there must be

some atmospheric factors that affect corrosion of sil-

ver and are not present in the ASTM B117 testing.

This silver paradox implies that natural atmospheres

are more oxidizing than the atmosphere created in

standard salt spray testing. Note that silver is a mate-

rial of interest because it is being used increasingly

as a means of rapidly measuring the corrosiveness of

ISSN 0012-9312 (print), 1938-159X (online)

12/000055/$5.00+$0.50/0 © 2012, NACE International

CORROSION ENGINEERING SECTION

036001-2 CORROSION—MARCH 2012

natural environments, and the corrosion rate of silver

is used as an input to a predictive model of steel cor-

rosion.9

Atmospheric corrosion of silver is known to be

affected by a number of environmental variables, such

as temperature, UV light, and pollutant gases.2,10-11

Although silver is thermodynamically unstable rela-

tive to silver oxide under ambient atmospheric condi-

tions, the kinetics of silver oxide formation under

these conditions is negligible.12 However, silver has

been shown to corrode in the presence of 5 mol%

ozone (O3 ) in O2 at 300 K and ambient pressure.13-14

UVA light (295 nm to 365 nm) splits ozone to form

atomic oxygen, which is a highly reactive species15-16

that attacks silver. Recent work2,10 also has shown

that the more oxidizing character of natural environ-

ments relative to the B117 can be reproduced in the

laboratory testing via the introduction of UVA light

and ozone to the test environment. The results showed

that UVA light and ozone have important accelerating

effects on the corrosion of silver. These studies were

performed in custom-made exposure chambers.

The goal of this study was to extend the previous

work2,10 by modifying a commercial accelerated cor-

rosion test chamber to allow for the controlled intro-

duction of both ozone and UVA light. The modified

chamber was then characterized in terms of the range

of ozone and UVA light illumination that can be con-

trolled. A series of tests were performed to evaluate

the relative effects of ozone and UVA light on the cor-

rosion of silver, and comparisons were made to data

from field exposures at several geolocations that cov-

ered a wide range of corrosivity. Finally, a modifica-

tion to the B117 test protocol is recommended that

provides reasonable acceleration factors for silver for a

wide range of geolocation corrosivities.

EXPERIMENTAL PROCEDURES

Salt Spray Chamber

A standard cyclic corrosion test chamber (Q-Lab

Corporation model CCT 1100,† Ohio) amenable for

conducting ASTM B117 testing, was used in this

work. A continuous salt spray is created by the

atomization of 5 wt% sodium chloride (NaCl) solu-

tion through a nozzle by the pumping compressed air

and solution. Compressed air is humidified by pas-

sage through a bubble tower on its way to the nozzle.

Freshly prepared salt solution is placed in the reser-

voir as specified by ASTM B117.1 The modification of

the cyclic corrosion test chamber, referred to hereafter

as the MB117, involved the design and construction

of two subsystems, as shown in Figure 1. One subsys-

tem was designed and constructed to produce a range

of UVA light intensity with the desired spectrum in the

chamber. The other subsystem was designed and con-

structed to generate ozone and distribute it uniformly

throughout the chamber.

Ultraviolet A Light

The UVA lamps (Q-Lab Corporation model UVA-

340† ) were chosen to simulate sunlight in the critical

short-wave UV region from 365 nm down to the solar

cutoff of 295 nm. The UVA lamps were used in pairs.

Two pairs of lamps were used, referred to as "Pair a"

and "Pair b." They are mounted on a movable rack

(shown in Figure 2), which allows them to be trans-

lated along both the y and z directions to create a

† Trade name.

FIGURE 1. Schematic view of the MB117 chamber.

CORROSION ENGINEERING SECTION

036001-3

CORROSION—Vol. 68, No. 3

desired UVA light intensity at a given specimen loca-

tion. The removable rack is made of structural fiber-

glass. Reflectors were included in the lamps to direct

as much of the UVA light toward the samples as pos-

sible. The maximum UVA intensity is controlled by the

total number and positions of UVA lamps being used.

A spectrometer (Ocean Optics Incorporation model

JAZ,† Florida) was used to measure the spectrum of

the UVA light, and a 1918-c Newport power meter

(Newport Corporation model 1918-c,† California) was

used to measure the UVA light intensity.

Ozone

A commercial ozone generator (Jelight Company,

Inc., model 1000, California) was used to produce

ozone on-line from clean air, which had been created

by passing compressed air through a silica gel desic-

cant and an activated carbon cartridge. The generator

used a 12 in (304.8 mm) by 2 in (50.8 mm) cell con-

taining an ozone-producing double-bore lamp to cre-

ate ozone from high-intensity UV light. The ozone was

introduced into the chamber via a custom-made baf-

fle to ensure a uniform spatial distribution. The con-

centration of ozone was recorded with a dual-cell, UV

photometric gas analyzer generally used for monitor-

ing ambient air. The ozone analyzer has a precision of

1.0 ppb and can measure ozone concentrations in the

air from 0.05 ppm to 200 ppm.

Sample Exposure in the Modified B117 Chamber

Silver samples of 75 by 15 by 0.25 mm and

99.95% purity were obtained from Lucas-Milhaupt,

Inc. (Cudahy, Wisconsin) and were wet-abraded with

600 grit silicon carbide (SiC) grinding paper. The sam-

ples were then cleaned with methanol and deionized

water and air-dried before they were put in the loca-

tions of the MB117 with a desired UVA light intensity

and a desired ozone concentration for different times.

Figure 2 shows the locations chosen to character-

ize the distribution of UVA light intensity within the

MB117 chamber. Eighteen locations in two horizon-

tal planes (Plane A and Plane B) were chosen. Plane

A was in the middle of the chamber, whereas Plane

B was at the level of the diffuser, which was near

the bottom of the chamber. Three triplicate samples,

exposed at one time, were used. The surface of the

samples were angled 20° from the vertical and paral-

lel to the principal direction of flow of spray through

the chamber. The MB117 test was a continuous expo-

sure to a 5 mass% salt spray (pH was 6.8) at 35°C, as

described in the ASTM B117 protocol.

Field Exposures

Field exposures of silver were conducted by

Abbott9 and Sugamoto, et al.,17 at six geolocations:

Coconut Island, Hawaii; Daytona Beach, Florida;

Whidbey Island, Washington; West Jefferson, Ohio;

Lyon Arboretum, Hawaii; Trenton, New Jersey. These

sites provided a wide range of corrosivities and loca-

tion types. Coconut Island and Daytona Beach are

marine atmospheres; Whidbey Island is a rural

marine atmosphere; West Jefferson is a mild urban

atmosphere; Lyon Arboretum is a rainforest atmo-

sphere; Trenton is an urban atmosphere.

Specimens (75 by 12 by 0.5 mm and 99.9% pure)

exposed to field sites were mounted onto plastic test

cards in slots, which could accept at least four plastic

test cards. The slots were kept in a card cage, which

consisted of an open plastic frame. The package was

then placed in position to provide free and natural air-

flow around the samples. The entire process of sample

placement and removal can be done within minutes.

This need for minimal manpower and/or reporting

requirements was also found to be critical to imple-

mentation.9

Analytic Techniques

After the exposure, corrosion products of silver

formed during the field exposure and the MB117

exposure were analyzed with coulometric reduction,

which was performed in a three-electrode cell. An area

of 1 cm2 of the exposed silver sample served as the

working electrode, a platinum-plated niobium mesh

served as the counter electrode, and a saturated mer-

cury/mercurous sulfate (MSE, Hg/Hg2 SO4 ) reference

electrode was used as the reference electrode. Before

introduction into the cell, the 0.1 M sodium sulfate

(Na2 SO4 , pH = 10) reduction electrolyte was deaer-

ated2,10 for at least 1 h by bubbling nitrogen. A con-

stant cathodic current density of –0.1 mA/cm2 was

applied immediately to the silver sample by a poten-

tiostat after addition of the solution to the cell. The

voltage was monitored until the voltage dropped to

approximately –1.7 VMSE , at which hydrogen was

evolved. Two values are extracted from each reduction

curve:18-19 the reduction potential and the reduction

charge. The reduction potentials serve to identify the

FIGURE 2. Selected physical locations (1 through 18) in the MB117

used for the measurement of the UVA light intensity and the ozone

concentration. Pair a and Pair b are two pairs of UVA lamps and can

be moved along y and z directions in the racks (red dashed lines).

CORROSION ENGINEERING SECTION

036001-4 CORROSION—MARCH 2012

chemical composition of the corrosion products and

were determined as the potential of the curve at the

midpoint of an invariant portion of the reduction

curve. The total reduction charge passed at one

potential represents the amount of that compound

present on the sample. The thickness of the corrosion

product layer of silver can be calculated from the

reduction charge, assuming that the corrosion prod-

uct film forms with theoretical density. The reduction

charges (per unit area, the same below) from field and

the modified B117 represented the means of triplicate

samples, standard deviations ranging from 1% to

8.5% (not shown).

The images of the samples were characterized by

scanning electron microscopy (SEM) equipped with

energy-dispersive x-ray spectroscopy (EDS) modes.

X-ray diffraction (XRD) measurements were carried

out with an x-ray diffractometer using a focused and

monochromatized Cu Kα source (λ = 1.540598 Å). The

data were collected with a position-sensitive detec-

tor in a 2θ range of 10° to 110° with a resolution of

0.0083556°.

RESULTS

Spectrum and Intensity of the Ultraviolet A Light

The spectrum of the UVA lamp, with peak emis-

sion at 340 nm, was recorded and shown in Figure 3

when the lamps were positioned in the topmost plane

and Pair a and Pair b lamps (Figure 2) were in the for-

ward most and backward most positions, respectively,

in the MB117 chamber. Figure 3 shows the main radi-

ation of the lamps to have wavelengths in the region

from 300 nm to 400 nm, which is appropriate for UVA

light. This spectrum is similar to that of natural sun-

light in the UVA light region.20

The UVA light intensities in the locations of Fig-

ure 2 are shown in Table 1, which demonstrates that

the intensities were affected by the quantities of the

lamps and the distance between the location of the

sample and the UVA lamps. The more lamps that were

present, the stronger the UVA light intensity, and the

closer the specimen location was to the lamps, the

stronger the UVA light intensity at the location. The

UVA light intensity was the lowest at 0.5 W/m2 when

there was only one pair of lamps, and the lamps were

mounted in the opposite, topmost edges of the cham-

ber from the samples. The highest UVA intensity,

50 W/m2 , was present just below the lamps when

both pairs of lamps were next to one another in the

middle vertical plane (Plane C) of the chamber. The

UVA light intensities in the MB117 can be 100X

higher than those of UV light at the studied natural

locations, in which the intensities of UV light ranged

from 0 to 0.46 W/m2 ,21 which are shown in Table 2.

Concentration and Uniformity of Ozone

To measure the distribution of ozone in the

MB117, the ozone concentrations in the five locations

of the MB117 was recorded. The concentration distri-

bution of ozone in the MB117 is plotted for the differ-

ent chamber locations (Figure 2) of MB117 in Figure

4(a). The ozone concentrations increased from 0 to

about 14 ppm, and remained stable after the ozone

generator had been running approximately 1.2 h

when the flow of cleaned air was 2 SLPM (standard

liter per minute). The average concentration of ozone

and the standard deviation between locations in the

MB117 under these conditions were 14.17 ppm and

0.16 ppm, respectively. Standard deviations over time

for locations 18, 16, 14, 12, and 10 were 0.05 ppm,

0.01 ppm, 0.04 ppm, 0.02 ppm, and 0.06 ppm,

respectively. Results show that ozone can be distrib-

uted uniformly throughout the chamber and the con-

centrations maintained to a tight tolerance. At the

exposure sites, the average monthly ozone concentra-

tion ranged from 0 to 591 ppb,21 while the annual

average concentration of ozone ranged from 27 ppb to

FIGURE 3. Spectrum of the UVA lamp used in the MB117.

TABLE 1

Intensity of UVA Light Irradiated in the Locations of Figure 2, W/m2

Pair a and Pair b are in

Pair a is in Pair b is in Plane C and Close to the Pair a is at the Front Edge and

the Front Edge the Back Edge Locations 11, 14, and 17 Pair b is at the Back Edge

Location 1, 4, and 7 3, 6, and 9 11, 14, and 17 10, 12, 16, and 18 2, 5, and 8

Intensity 0.5 0.5 50 33 1.5

CORROSION ENGINEERING SECTION

036001-5

CORROSION—Vol. 68, No. 3

70 ppb.22-23 Therefore, the ozone concentration in the

MB117 could be 40X and 330X higher than the maxi-

mum average monthly and annual average ozone con-

centrations at the studied locations, respectively.

Ozone concentrations in the MB117 were

adjusted by adjusting the flow rate of compressed,

purified air. The maximum concentration of ozone

that could be stably maintained with the present con-

figuration was 23.2 ppm (Figure 2) after the generator

had been running for approximately 4 h with a flow

rate of cleaned air of 10 SLPM, as shown in Figure

4(b). The standard deviation over time for the maxi-

mum concentration of ozone was 0.2 ppm, obtained

by a series of samplings from 4.7 h to 23.6 h of expo-

sure. Note that much higher ozone concentrations

could be achieved if compressed oxygen were used to

produce ozone rather than air, but this change would

complicate the experimental arrangement, especially

for long-term exposures.

Corrosion of Silver in the Fields

and in the Modified B117 Chamber

To investigate the correlation between results

from MB117 and the field exposures, the corroded

silver surfaces from each were analyzed with the goal

of identifying the corrosion products of silver. Because

of the limitations of individual techniques, a suite of

complementary analytical techniques were used. Fig-

ure 5 shows the XRD results from samples exposed

to the MB117 testing or a six-month exposure at

Coconut Island. The series of peaks when 2θ was

approximately 27.8°, 32.2°, 46.2°, 54.8°, 57.5°, and

85.7° belonged to AgCl. The peaks when 2θ was about

38.1°, 44.3°, 64.4°, and 81.6° belonged to metallic

Ag. The peaks of silver(I) oxide (Ag2 O) are known to

be at 32.2°, 38.1°, and 54.8°, which overlap peaks for

AgCl and metallic Ag. However, the XRD patterns for

samples from the field and the MB117 chamber were

FIGURE 4. Concentration of ozone in the MB117 chamber when the

flow of cleaned compressed air was (a) 2 SLPM with measurements

at several locations and (b) 10 SLPM for location 8. Physical locations

are shown in Figure 2.

TABLE 2

Parameters in the Field Sites and Magnifying Capability of the Parameters of the MB117

Coconut Daytona Whidbey West Lyon

Location Island Beach Island Jefferson Arboretum Trenton

Type of environment Marine Marine Rural marine Mild urban Rainforest Urban

Average UVA intensity, 0.4620 0.18 0.07 0.11 0.3320

W/m2

Maximum magnification 109 278 714 455 152

of UVA intensity in the

MB117 to field

Average annual O3 52 70 56 70 4322 27

concentration, ppb21

Maximum magnification 446 331 414 331 540 860

of ozone in the MB117

to field sites

(b)

(a)

CORROSION ENGINEERING SECTION

036001-6 CORROSION—MARCH 2012

the same, although the relative intensities of the

peaks of the compounds on the surfaces of silver

were different.

Figure 6 shows SEM images and EDS results for

silver exposed in the MB117 as well as for a sample

exposed at Coconut Island, representative of all the

field exposure sites examined. The EDS indicates that

AgCl was the main corrosion product. The images

clearly show Ag grains on the surfaces and the non-

uniform corrosion that occurred in both the MB117

chamber and the field exposure.

For all silver samples, corrosion product accu-

mulation was assessed via coulometric reduction.

Coulometric reduction is an electrolytic method that

measures the potential as a function of reduction time

and is used to identify both the compound and deter-

mine the amount present. By measuring these param-

eters for known compounds, the potential plateau

position can be used to identify the compound and

the time during which the plateau is maintained can

be related to the amount of corrosion product via

Faraday's Second Law.

Figure 7 shows the coulometric reduction curves

for silver samples after exposure in the MB117 with

23 ppm ozone and 50 W/m2 UV light for 15 days and

Daytona Beach for 3 months. The reduction poten-

tials at approximately –0.18 VMSE and –0.3 VMSE belong

to the reduction potentials of Ag2 O and AgCl, respec-

tively, as demonstrated by the measurement of the

standards.

Reduction charges were calculated from the

reduction time by multiplying by the current density,

which was constant (–0.1 mA/cm2 ) in this work. The

reduction time was defined as from the beginning to

the midpoint between the adjacent two plateaus. For

example, for the reduction curve from the MB117

exposure (Figure 7[a]), the reduction time for Ag2 O is

from 0 to the time marked by the first blue dashed

line (1,290 s), whereas the reduction time for AgCl is

the time between the two blue dashed lines (12,350 s).

The main compound was found to be AgCl on the

sample from the Daytona Beach site as well as the

sample from the MB117 test. Although small reduc-

tion charges for Ag2 S of silver were observed for some

outdoor sites, for example, at West Jefferson, we only

discuss the reduction charge of AgCl of silver in the

present study.

Figure 8 depicts the reduction charges for the

AgCl formed on silver after exposure in the MB117

chamber with varying ozone concentrations and UVA

light levels for different times. It shows that silver cor-

roded in all cases of the MB117, in dramatic con-

trast to the lack of corrosion after four months with

the ASTM B117 method.2 Higher corrosion rates are

observed with higher ozone concentrations. At a con-

stant ozone concentration, higher UVA light intensity

increased the rate of attack. The ozone provides suffi-

cient oxidizing ability (above that of molecular oxygen)

to corrode silver rapidly. In the presence of UVA light,

additional oxidants (such as atomic oxygen15-16 ) are

created, increasing the rate of attack. In the presence

of Cl– , AgCl is formed. In other work,2,10 it has been

shown that in the absence of chloride, silver oxide

is formed in the presence of ozone, with higher rates

of oxide formation observed when UVA light is also

present.

The data in Figure 8 were fit to linear expres-

sions. The slope of the linear fit represents the cor-

rosion rate of silver in that environment. All of the

regression coefficients are larger than 0.975. Corro-

sion amounts for AgCl of silver increase linearly with

exposure time. This constant rate of corrosion has

been observed for field exposures9 and in the labo-

ratory for charge densities of up to 2 C/cm2 , corre-

sponding to a AgCl thickness of 5.3 µm, assuming

theoretical density (5.56 g/cm3 ).18,24

FIGURE 5. XRD pattern of silver after exposure at (a) Coconut Island

for 6 months and (b) the MB117 with 23 ppm ozone for 3 days.

(a)

(b)

CORROSION ENGINEERING SECTION

036001-7

CORROSION—Vol. 68, No. 3

DISCUSSION

The objective of this work was to extend the work

on the atmospheric corrosion of silver from custom-

made chambers2.10 to commercially available salt

spray test chambers. By doing so, the so-called "silver

paradox" has been resolved. The corrosion of silver

observed in virtually any natural atmosphere, but not

observed using standard salt spray testing protocols

(i.e., ASTM B117), can be reproduced in a commercial

test chamber if the chamber and test protocol are

appropriately modified. The modifications involve the

creation of atmospheres that are as oxidizing (or more

oxidizing if acceleration of corrosion is of interest) as

natural atmospheres. Natural atmospheres have

many oxidants. Standard salt spray test chambers

use laboratory air, which is generally highly filtered,

and the container precludes the presence of UV light.

The filtration of the air removes those natural oxi-

dants and the absence of UV light prevents the cre-

ation of additional oxidants, making ASTM B117

protocols far less aggressive in terms of oxidation

potential than natural atmospheres. Silver corrosion

is driven primarily by the oxidizing potential; there-

fore, standard protocols without oxidizers stronger

than molecular oxygen are going to be ineffective in

causing silver corrosion. It is interesting to note that

in some instances, small amounts of silver sulfide

(Ag2 S) can be observed on silver exposed for many

months in standard salt spray testing; it is a measure

of the cleanliness of the laboratory air with respect to

sulfur compounds, because silver rapidly reacts with

such species.

It is clear from Table 2 that the MB117 protocol

can provide UVA light intensities more than 100X

and ozone concentration more than 330X than those

experienced in a range of geolocations, creating more

oxidizing atmospheres that can accelerate the corro-

sion of silver. Inspection of Figure 8 shows that the

effects of up to 33 W/m2 of UVA light were modest

relative to the increasing ozone concentration in the

MB117. The addition of 1.5 W/m2 to 7 ppm ozone in-

FIGURE 6. (a) Image and (b and c) element distributions of different areas on the surfaces of the samples after exposure

at Coconut Island, for 6 months. (d) Image and (e) element distributions of the areas on the surfaces of the samples after

exposure in the MB117 with 23 ppm O3 and UV light for 3 days.

CORROSION ENGINEERING SECTION

036001-8 CORROSION—MARCH 2012

creased the rate by about 20%, and a further increase

to 33 W/m2 represented a 60% increase in rate rela-

tive to the ozone alone. Inclusion of UVA light may be

indicated for testing situations in which higher ozone

concentrations are undesirable for other reasons,

including ozone production limitations or control is-

sues, or personnel safety.

Based on the results from the coulometric reduc-

tion, XRD, SEM images, and EDS of the field and the

MB117 exposures, the corrosion products and dam-

age morphology produced on silver in the presence

of NaCl was similar for the field and the MB117 test-

ing exposures. The main corrosion product was AgCl,

non-uniform corrosion occurred, and silver grains

were found on the surfaces of silver after exposures.

According to the lab data in Figure 8 and the field

data,9 the rate of formation of AgCl was constant

in the atmospheric environments studied. There-

fore, according to the specification of ASTM B117,1

comparisons between the field data and the MB117

results can be made by considering a common expo-

sure time. For the current purposes, an exposure

time of 30 days was used. The reduction charges for

each version of the MB117 (i.e., combinations of ozone

concentration and UVA light intensity shown in Fig-

ure 8) were scaled to 30 days based upon the lin-

ear regression equations. The acceleration factor is

defined here as the ratio of the reduction charge of

the MB117-exposed sample to the charge of the field-

exposed sample. Calculated acceleration factors for

the six geolocations for several different versions of

MB117 are shown in Figure 9. It can be seen that the

two marine sites are the most corrosive toward sil-

ver. It is known that the salt deposition in marine

sites, such as Coconut Island and Daytona Beach, is

higher than in other areas. The UVA level in Coconut

Island is much higher than that in the other sites. The

ozone level in Daytona Beach is higher than that in

the other sites as well. Recall that no corrosion of sil-

ver was observed after four months of the ASTM B117

protocol; a combination of NaCl, ozone, and/or UV

light is required.2,10

The data in Figure 9 show that the MB117 can be

tuned to produce a range of desired acceleration fac-

tors. Such tuning may be of interest since there is no

doubt a trade-off between acceleration and accuracy.

Although the MB117 results presented here for sil-

ver are consistent with regard to the type of corrosion

products formed and the morphology of the corro-

sion, it is not clear at what combinations of ozone and

UVA light that correlation fails. Using Daytona Beach

as an example, corrosion of silver could be acceler-

ated in MB117 by up to 20X by adding 23 ppm ozone

(a)

(b)

FIGURE 7. Coulometric reduction curves for the silver samples

after exposure to: (a) the MB117 with 23 ppm ozone and UV light

(33 W/m2 ) for 15 days and (b) Daytona Beach for 3 months. The

reduction electrolyte was 0.1 M Na2 SO4 (pH = 10) solution and, it was

deaerated 1 h in advance.

FIGURE 8. Reduction charge and fitting curves for AgCl produced on

silver exposed in the MB117 with different ozone concentrations and

different exposure times.

CORROSION ENGINEERING SECTION

036001-9

CORROSION—Vol. 68, No. 3

alone to the ASTM B117 protocol. In practical terms,

this acceleration factor would allow about 6 months

of exposure at Daytona Beach to be simulated in one

week in the laboratory. If lower ozone concentrations

are desired, a very similar acceleration factor can be

achieved using 7 ppm ozone and 33 W/m2 UVA light.

The application of the same MB117 conditions for

prediction of corrosion at less corrosive geolocations

results in much higher acceleration factors, as shown

in Figure 9.

The practical implications of these results for the

estimation of corrosiveness using silver coupons is

substantial. Short-term (30-day) exposures of silver

have been and are being used to estimate corrosive-

ness of geolocations9 as well as indoor locations such

as control rooms,25 but there has been no means to

reproduce that corrosiveness in the laboratory using

generally available equipment (i.e., commercially

available salt spray test chambers). The approach pre-

sented in this study provides such a means. The flexi-

bility of the method provides testers with the ability

to tailor their chamber to the degree of acceleration of

interest.

It must be noted that all of the work presented

here focused on the corrosion of silver. Although sil-

ver is used as a corrosivity monitor and is the basis

for corrosion rate prediction for steel, copper, and alu-

minum, the vast majority of interest in accelerated

testing is aimed at structural materials and coated

structural materials. The testing framework presented

here is currently being applied to such materials, but

the impact of UVA light and ozone remain to be deter-

mined, especially relative to effects of time-of-wetness

and salt loading.

CONCLUSIONS

v A standard salt spray corrosion test chamber was

modified to allow the introduction and control of

ozone and UVA light to produce an accelerated test

that recreates the corrosion of silver observed in

geolocations around the world, but heretofore not

reproduced in a commercial salt spray corrosion test.

The modified test chamber has one subsystem that

produces up to 23.2 ppm ozone from cleaned com-

pressed air and another subsystem that produces up

to 50 W/m2 of UVA light intensity.

v Results show that silver formed the same corrosion

product and had the same damage morphology in the

MB117 and field exposures. Ozone accelerates the

corrosion of silver during salt spray testing, whereas

silver did not corrode measurably using the ASTM

B117 salt spray conditions. The addition of UVA light

increases the corrosion rate when ozone is also pres-

ent. Acceleration factors, based on the comparison of

coulometric reduction results from field and MB117

exposures, show that the MB117 can accelerate the

atmospheric corrosion of silver substantially relative

to natural atmospheres, up to a factor of 20 even for

a severe marine environment. Using only ozone at

23 ppm, the corrosion rate of silver in the MB117

was about 20X that of a severe marine location

(Daytona Beach, Florida), allowing 6 months' worth of

corrosion to develop in 1 week. For locations of lower

corrosiveness (Trenton, New Jersey), this version of

the MB117 would correspond to acceleration factors

of more than 120, meaning that 1 week of exposure

in the MB117 would correspond to ca. 2 years of field

exposure.

v Although this work has focused on the corrosion of

silver, extension to other materials is envisaged.

ACKNOWLEDGMENTS

This work was supported by grants from D.

Dunmire, Director of Corrosion Policy and Oversight,

Office of the Secretary of Defense. Special thanks to

B. Abbott of Battelle Memorial Laboratory (Columbus,

Ohio) for placing and recovering samples in field sites,

and to L.H. Hihara of University of Hawaii for Hawai-

ian samples. Many thanks also to C.S. Long, Meteo-

rologist of National Oceanic and Atmospheric Admin-

istration, for providing the UV light intensities at field

sites.

REFERENCES

1. ASTM B117-03, "Standard Practice for Operating Salt Spray Ap-

paratus" (West Conshohocken, PA: ASTM International, 2003).

2. D. Liang, H.C. Allen, G.S. Frankel, Z.Y. Chen, R.G. Kelly, Y. Wu,

B.E. Wyslouzil, J. Electrochem. Soc. 157, 4 (2010): p. C146-C156.

3. F. Deflorian, S. Rossi, L. Fedrizzi, C. Zanella, Prog. Org. Coat. 59

(2007): p. 244-250.

FIGURE 9. Acceleration factors (AF) for different versions of the

MB117 test. These AF were calculated based on the reduction

charges reported by Abbott9 for 1 month exposures and assuming

linear kinetics for silver corrosion.

CORROSION ENGINEERING SECTION

036001-10 CORROSION—MARCH 2012

4. R.L. Howard, S.B. Lyon, J.D. Scantlebury, Prog. Org. Coat. 37

(1999): p. 91-98.

5. K.N. Wernstahl, Polym. Degrad. Stab. 54 (1996): p. 57-65.

6. K.R. Baldwin, C.J.E. Smith, Aircraft Eng. Aero. Tech. 71, 3 (1999):

p. 239-244.

7. G.D. Davis, L.A. Krebs, C.M. Dacres, J. Coat. Technol. 74, 935

(2002): p. 69-74.

8. SAE J2334, "Laboratory Cyclic Corrosion Test" (Warrendale, PA:

SAE International, 2003).

9. W.H. Abbott, "A Decade of Corrosion Monitoring in the World's

Military Operating Environments—A Summary of Results"

(Columbus, OH: Battelle Columbus Operations, 2008), pp. 2-22,

61 and appendix F; https://www.corrdefense.org/Academia

Government and Industry/A Decade of Corrosion Monitoring.pdf.

10. Z.Y. Chen, D. Liang, G. Ma, G.S. Frankel, H.C. Allen, R.G. Kelly,

Corros. Eng. Sci. Technol. 45, 2 (2010): p. 169-180.

11. D.W. Rice, P. Peterson, E.B.P. Phipps, R.J. Cappelll, R.

Tremoureux, J. Electrochem. Soc. (1981): p. 275-284.

12. T.E. Graedel, J. Electrochem. Soc. 139 (1992): p. 1963-1970.

13. G.I.N. Waterhouse, G.A. Bowmaker, J.B. Metson, Appl. Surf. Sci.

183 (2001): p. 191-204.

14. R.O. Suzuki, T. Ogawa, K. Ono, J. Am. Ceram. Soc. 82 (1999): p.

2033-2038.

15. B.J. Finlayson-Pitts, J.N. Pitts, Chemistry of the Upper and Lower

Atmosphere (San Diego, CA: Academic Press, 2000), p. 86-94.

16. S. Chapman, Rep. Prog. Phys. 9 (1942): p. 92-100.

17. R. Sugamoto, G.A. Hawthorn, L.H. Hihara, "Comparison of Atmo-

spheric Chloride Deposition Measurement Techniques in a Multi-

tude of Microclimates," 2011 DOD Corrosion Conference (Houston,

TX: NACE International, 2011).

18. ASTM B825-02, "Standard Test Method for Coulometric Reduc-

tion of Surface Films on Metallic Test Samples" (West Con-

shohocken, PA: ASTM International, 2002).

19. S.J. Krumbein, B. Newell, V. Pascucci, J. Test. Eval. 17 (1989): p.

357-367.

20. D.D. Moyal, A.M. Fourtanier, J. Am. Acad. Dermatol. 58, 5 (2008):

p. 5149.

21. JMA & NASA-WFF, World Ozone and Ultraviolet Radiation Data

Centre (WOUDC) [Data], http://www.woudc.org (2011).

22. AIRNOW Tech., http://www.airnowtech.org (2011).

23. J.M. Harris, S.J. Oltmas, E.J. Dlugokencky, P.C. Novelli, Geophys.

Res. Lett. 25 (1998): p. 1895-1898.

24. E.B. Neiser, "Atmospheric Corrosion of Silver and its Relation

to Accelerated Corrosion Testing" (Master thesis, University of

Virginia, 2010).

25. W.B.A. Sharp, L. Falat, Krasowski, Tappi J. 72, 10 (1989).

... Dengan beberapa modifikasi yang tidak ada pada persyaratan ASTM B117, mesin uji salt spray dapat digunakan untuk material perak. Modifikasi yang dilakukan adalah penambahan ozone dan sinar UVA yang berperan dalam pengkorosian perak (Wan, Macha, and Kelly 2012). ...

... Penggunaan jenis termokontrol sangat berpengaruh pada temperatur yang dihasikan pada alat uji.Pratama dan Sulistijono melakukan pengujian kabut garam pada temperatur yang dijaga sebesar 35°C selama 48 jam(Pratama and Sulistijono 2013). Begitu juga pengujian yang dilakukan oleh Wan dkk mengacu pada ASTM B117 dengan temperature chamber pengkabutan dijaga pada 35°C(Wan, Macha, and Kelly 2012).Dimensi dari chamber uji pengkabutan ini sekitar 0,88 m 3 , diperlukan energi yang cukup untuk memanaskan chamber tersebut. Konsumsi listrik alat uji ini diukur menggunakan pengukur daya yang ada pada panel alat uji. ...

  • ST. Hendri Siswanto
  • Sina Jamilah Sina Jamilah

Salah satu pengujian untuk produk logam adalah pengujian terhadap ketahan korosi. Korosi dapat menyebabkan kerugian yang berbahaya dan mahal jika tidak diantisipasi. Oleh karena itu, pengujian ini sangat penting untuk mengetahui kualitas material pada produk-produk tertentu. Penelitian ini bertujuan untuk membuat mesin uji kabut garam dengan spesifikasi yang sesuai dengan standar JIS Z 2371 dan ASTM B117-11. Penelitian ini dibatasi pada pengujian fungsi mesin uji kabut garam tanpa menggunakan spesimen uji. Metode yang digunakan dalam perancangan ini adalah hasil studi literatur untuk membuat alat uji yang sesuai standar. Penggunaan rangkaian relay, kontrol waktu dan temperatur sebagai sistem pengendali, menjadikan mesin ini dapat bekerja secara otomatis sesuai dengan pengaturan mesin yang telah ditentukan berdasarkan standar. Setelah proses pembuatan selesai, alat uji ini diuji untuk mengetahui tingkat keberhasilan dan daya tahannya. Jumlah kabut garam yang dihasilkan selama satu jam adalah 1,2 – 1,8 ml. Tempertur yang dicapai pada chamber pengkabutan adalah 36-37 ºC, sedangkan temperatur di chamber saturator nilainya antara 47-49 ºC. Uji performa selama 12 jam menujukkan bahwa kondisi alat uji dalam keadaan stabil dan tidak ada kerusakan. Dari hasil pengujian tersebut, maka alat uji dapat bekerja sesuai dengan persyaratan standar dan dengan daya tahan yang baik.

... Silver is sensitive to chloride (Cl − ) and silver chloride will be formed as a result of the reaction [10] [11] [12] [15] [16] [17]. Also this does not agree with results revealed that silver chloride compound was not identified on surface film of silver coupons after the exposure in an ASTM B117 salt spray chamber [18] and this compatibility with previous studies mentioned that silver does not react directly with chlorine gas and the presence of silver chloride as corrosion product due to burial in a chloride rich environment [19] [20]. ...

... few studies were presented on the silver deterioration tests inside climate chamber, these studies used two types of the deterioration factors: high relative humidity [10][29] and gaseous pollutants in presence relative humidity[2] [25][29]. Most studies of silver deterioration were used corrosive solutions as deterioration factors, such as BaS 5 g/l solution for 24 Hours[18] and Na 2 S were used as the tarnishing solution[30] [31]. Acetic acid solutions were used as simulation of emissions vapors in wooden cabinets and CuCl 2 50 g/l for 20 min, and NaCl[29] were used of AgCl silver patina[19]. ...

... 15 To accurately mimic real field environments, various environmental factors have been considered in laboratory acceleration tests, such as wet/dry cycling, 8,10 thermal cycling, [16][17] UV, 9 and ozone. [18][19][20][21] However, real environments are very complex and can change enormously with time and location. Of importance are the deposition of aerosols and particulates, wet/dry cycling corresponding to diurnal variations, weather-driven relative humidity changes, and episodic precipitation. ...

... [10][11][12][13][14] The corrosion attack on coated and scribed metals has been compared visually after lab and field exposure to establish a correlation between them. [10][11][12][13][14]27 Corrosion products in the lab and field have also been analyzed by x-ray diffraction, [10][11][12]21,28 energy dispersive spectroscopy, [11][12] Raman spectroscopy, 10,28 and x-ray photon-electron spectroscopy. [29][30] Visualization and corrosion product analysis are comparisons that can generate support for a particular corrosion mechanism, but they do not provide sufficient information for accurate lifetime assessment. ...

A galvanic test panel design incorporating a painted and scribed Al alloy panel and uncoated through-hole noble fasteners has recently been utilized to compare effects of different surface pretreatments on galvanic attack in laboratory chambers. In this work, corrosion of galvanic panels composed of a coated AA2024-T3 panel and uncoated 316 stainless steel fasteners was quantified after exposure to ASTM B117 and a beach field site. Galvanic currents were continuously monitored between the panel and stainless steel fasteners in the field and in the laboratory chamber, exhibiting current transients and stable high value currents, respectively, associated with the two different environmental conditions. For all coating systems, exposure in the laboratory chamber resulted in larger galvanic current and greater extent of corrosion than what was generated by field exposure. However, the nature of the corrosion in the field was different, even though the galvanic current measured during field exposure was similar for the two coating systems. The galvanic current and analysis by optical profilometry provided quantification of the attack and allowed for the determination of acceleration factors to describe the influence of either galvanic coupling or an environment to accelerate coating degradation relative to the condition of no galvanic coupling or of another environment. An acceleration factor for the oxygen reduction reaction was also determined by comparing cathodic currents of a bare SS316 fastener in the different environments.

... The salt fog accumulated in both volumetric cylinders for at least 80 mm 3 , in which the first cylinder was kept near the salt spray nozzle and another cylinder was placed farthest from the sprayer. Visual inspection of specimens was then performed and compared with the microscopic examination results at 100 μm magnification on the 24, 120, 240, 360 and 480 hours [9]. ...

... Some AgCl on the daguerreotype surface may be residue from the sensitization step [equation (1)] (Ravines et al., 2016b). Similar to the exposure step, any residual AgCl trapped within the Ag lattice may undergo reduction from UV exposure [equation (2)] to form Ag nanoparticles (Wan et al., 2012). This has been noted to occur on daguerreotypes under natural and experimental conditions (Robinson & Vicenzi, 2015), 2AgðsÞ þ Cl 2 ðgÞ ! ...

Louis-Jacques-Mandé Daguerre introduced the first successful photographic process, the daguerreotype, in 1839. Tarnished regions on daguerreotypes supplied by the National Gallery of Canada were examined using scanning electron microscopy energy-dispersive X-ray spectroscopy and synchrotron-radiation analysis. Synchrotron X-ray fluorescence imaging visualized the distribution of sulfur and chlorine, two primary tarnish contributors, and showed that they were associated with the distribution of image particles on the surface. X-ray absorption near-edge structure spectroscopy determined the tarnish to be primarily composed of AgCl and Ag 2 S. Au 2 S, Au 2 SO 4 , HAuCl 4 and HgSO 4 were also observed to be minor contributors. Environmental contamination may be a source of these degradation compounds. Implications of these findings will be discussed.

... For example, carbonation is the major cause for the failure of concrete structure due to the erosion of carbon dioxide [15]. The corrosion of silver can be accelerated by the presence of ozone and ultraviolet [36]. Solar radiation, wind, and rain can influence the deposition rate of chloride and the time of wetness [4]. ...

... Research on Dynamic Potentiometric Polarization Test in NaCl Solution And With Immers Test To Determine Comparative Behaviour of AZ31B Magnesium Alloy Corrosion [14]. Research on Correlations That Occur On The Meassurement of Silver Corrosion Field And Modification of Salt Spray Salt Test ASTM B117 [15]. Examines The Modification of Accelerated Corrosion Space And Its Effect on The Corrosion of The Silver Atmosphere Occuring In The External Environment [16]. ...

... In Springer handbook of nanotechnology (pp. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] ...

This paper presents practical results of comparison of corrosion resistance in products with application of nano ceramic sealant, the results are compared with products that have surface treatments of electrolytic galvanization, KLT and organometallic. The ceramic nano sealant consists of nano metric colloidal silica. The tests carried out followed the procedures established in the international standard ASTM B117 which assigns test standards to evaluate the corrosion of products. To perform the tests, samples were developed with the application of the nano ceramic sealant and samples that only passed through the process of electroplating, KTL and organometallic electroplating. All samples were submitted to the tests established by ASTM B117. The obtained results, express superior performance of corrosion resistance of the products that went through the process with nano ceramic finish.

... In Springer handbook of nanotechnology (pp. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] ...

This paper presents practical results of comparison of corrosion resistance in products with application of nano ceramic sealant, the results are compared with products that have surface treatments of electrolytic galvanization, KLT and organometallic. The ceramic nano sealant consists of nano metric colloidal silica. The tests carried out followed the procedures established in the international standard ASTM B117 which assigns test standards to evaluate the corrosion of products. To perform the tests, samples were developed with the application of the nano ceramic sealant and samples that only passed through the process of electroplating, KTL and organometallic electroplating. All samples were submitted to the tests established by ASTM B117. The obtained results, express superior performance of corrosion resistance of the products that went through the process with nano ceramic finish.

Most solar thermal power generation systems use mirrors over a large area to concentrate solar energy onto the absorber tube. The optical properties of the mirrors directly affect the performance of the concentrating system. Corrosion attack was observed on these mirrors exposed to ambient which cause reduction in optical efficiency as well as working life of the mirrors. Stereo micrograph images and Energy-dispersive X-ray spectroscopy (EDS) results of the mirror samples confirmed loss of silver due to corrosion. Accelerated corrosion testing of few unexposed mirror samples confirmed that factors such as moisture permeability and the reaction of silver with chloride ions at the Cu-Ag interface play a key role in the degradation of the thin silver metal coating. Additional polymer coating on back surface of the mirror was found to be effective in resisting corrosion from ozone and UV radiation. This paper presents study of corrosion prevention of the thin silver layer of the mirror by different polymer protective coating given additionally on their back surfaces. To improve the life of mirror the thin metal coatings must have uniform thickness, be contaminant free, properly edge sealed and have protective coating applied on the back surface.

Understanding, recreating in the lab, and predicting atmospheric corrosion would be of great assistance to anyone whose materials or structures are exposed to the atmosphere. A first step in the development of the accelerated aging tests which would provide this information must be an understanding of the corrosive agents present in a given environment and their effects on the exposed material. In this work silver is exposed both to field and laboratory conditions. Laboratory exposures controlled relative humidity, ambient atmospheric gas (i.e., air or nitrogen), and exposure to UV light. The presence of both air and UV illumination was necessary for corrosion for the limited exposure times here. A complex relationship between the amount of corrosion and relative humidity was found; increasing the relative humidity did not necessarily increase the amount of corrosion. Field exposures showed that silver chloride and silver sulfide are common corrosion products and silver chloride is the main product from coastal exposures. Coastal corrosion rates were found to be higher than inland corrosion rates.

  • Guy Davis Guy Davis
  • L.A. Krebs
  • C. M. Dacres

An in-situ corrosion sensor was used to obtain electrochemical impedance spectroscopy (EIS) measurements on coated panels under a variety of accelerated laboratory test conditions as well as ambient exposure at a Florida beach. Three studies are reported. The first compared the sensor (EIS) measurements taken in a salt fog chamber to those obtained using a clamp-on cell and the conventional remote electrode/immersion approach. For coatings with minimal edge defects, the two methods gave equivalent results. For coatings with edge defects, the sensor was able to detect the defects provided the surface was wet, as in the salt fog chamber. In contrast, the conventional approach was unable to detect defects unless they were within the confines of the clamp-on cell. In the second case, sensor measurements were used to compare coating degradation during salt fog, a cyclic corrosion test, humidity, and immersion to that occurring at a Florida beach. The cyclic corrosion test showed an excellent correlation with beach exposure while the salt fog and other test showed very little correlation, suggesting that the cyclic test is more valid for discriminating coating performance for seacoast exposure. The sensor also indicated that the test could be short-ened by up to 40% without significantly reducing the validity of the test. In the final example, a series of primers and appliqués were evaluated using the cyclic corrosion test. The sensor EIS results allowed a discrimination between the materials sets even though there was little or no visual difference between the specimens.

  • J. M. Harris
  • Samuel J Oltmans Samuel J Oltmans
  • E. J. Dlugokencky
  • T. Mefford

Measurements of CH4, CO, O3, and H2O vapor from Mauna Loa Observatory (MLO) are examined in conjunction with isentropic trajectories to investigate the cause of a maximum in tropospheric O3 consistently observed during spring. CO and O3 have been found to be positively correlated in pollution plumes containing O3 precursors downwind of industrialized regions. However, we report that during continental transport from Asia, O3 is not correlated with either CO or CH4, although CO and CH4 are strongly correlated. The relationship between CO and CH4 suggests common source regions. The lack of correlation between these species and O3 probably indicates an O3 source distinct from that of CO and CH4. While Asian pollution does not appear to be a strong candidate for causing the spring increase in O3, transport characteristics and H2O vapor measurements are consistent with both an upper-tropospheric/stratospheric contribution and an enhancement from transport across O3 gradients.

  • A Wolfenden
  • SJ Krumbein
  • B Newell
  • Vincent Pascucci Vincent Pascucci

This paper describes a standard procedure for using constant-current electrolytic reduction ('coulometric reduction') to determine the relative buildup of corrosion and tarnish films on control coupons from environmental tests, and discusses the types of results and correlations that may be expected. Examples of the applications of this proposed ASTM standard method will be presented for two types of environmental exposure: the mixed flowing gas test and the humid sulfur vapor ('flowers-of-sulfur') test.

The protective properties of organic coatings for outdoor applications are generally evaluated by means of accelerated laboratory tests, including electrochemical techniques. The coatings are stressed by different mechanical, chemical, thermal loads and the effects on the protective properties can be measured by using well established electrochemical techniques, like electrochemical impedance spectroscopy, electrochemical noise, etc. An open question is how these accelerated tests can be correlated with natural exposure in different environments.

  • K.R. Baldwin
  • C.J.E. Smith

Examines accelerated methods for the corrosion testing of materials, coatings and surface treatments used in the aerospace and defence industries. The drawbacks with some current accelerated corrosion tests are examined, particularly the problems experienced with neutral salt spray tests. Specific examples are given which identify the acute discrepancy between salt spray and marine exposure in the corrosion testing of metallic coatings on steels. Examines some recent advances in corrosion testing of aerospace materials, pre-treatments and organic coatings, and outlines some preliminary research conducted at DERA Farnborough in developing more accurate test methods.

The corrosion of Ag in an atmosphere of ozone and humidity with or without irradiation by ultraviolet (UV) light was investigated. A modified coulometric reduction technique was used, substituting sulphate solution for chloride solution, to prevent the spontaneous transformation of silver oxide corrosion product to chloride in the reduction solution. The presence of both ozone and UV radiation was required for fast corrosion of Ag to occur. The amount of corrosion product for a given exposure time increased with ozone concentration, whereas the relative humidity had little effect. An incubation time for the corrosion reaction was observed. The presence of both ozone and UV radiation were necessary for rapid corrosion because the photodissociation of ozone generates reactive atomic oxygen, which reacts with Ag rapidly to form Ag2O. The corrosion reaction on bare silver was minimally affected by the relative humidity in the environment, which is contrary to common atmospheric corrosion experience.

The possible use of ozone to produce higher oxides was analyzed thermodynamically. Because the activity of oxygen in ozone is ∼1018 at room temperature and ∼108 at 773 K, ozone may react with metals to yield higher-oxidation-state products than does an ambient pressure of pure oxygen. In agreement with this thermodynamic prediction, silver oxides (AgO and Ag2O) were synthesized experimentally by blowing a gas mixture of 6 vol% ozone and oxygen through a water-cooled lance. Neither Ag3O4 nor Ag2O3 was detected. Slow oxygen diffusion into a metal and/or reaction products and decomposition of unstable oxides that are formed successively are obstacles to practical applications of syntheses via ozone oxidation.

Various accelerated cabinet tests have been used for the evaluation of the cut-edge corrosion of coil-coated architectural cladding. These include the conventional ASTM B-117 method (5% continuous NaCl spray), the standard Prohesion test (0.35% (NH4)2SO4 + 0.05% NaCl wet/dry spray) as well as modified wet/dry spray tests using a relatively dilute artificial acid rain solution, shallow specimen incline angles with variations in the ratio of the wet and dry periods. Comparisons with outdoor exposure samples reveal that the B-117 test shows unrealistic corrosion morphology with the most realistic cut-edge corrosion given by the modified tests. Furthermore, with all types of wet/dry test specimen-to-specimen variation is considerably reduced compared with the continuous salt spray test. In the novel wet-dry tests the acceleration factor appears consistent with the number of wet and dry cycles however, the overall acceleration factor is limited and, hence, test times are extended. It is concluded that, over a 1000 h test time, the standard Prohesion test appears to show the best combination of realism and acceleration.