Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

ESD Corrosion A-Texas Water 2003 - ODOROUS GASES CAUSE COMPUTER by hjkuiw354


									                                                                  Proceedings of Texas Water 2003
                                                                               Corpus Christie, TX

                                       W. Brad M. Stanley
                                      Christopher O. Muller
                             Purafil Environmental Systems Division
                                        2654 Weaver Way
                                       Doraville, GA, USA


Corrosion of electronic equipment is a fact of life to chemical plants, pulp and paper mills, and
many other industries including wastewater treatment plants. The gases that cause corrosion-
related problems in these facilities – hydrogen sulfide, chlorine, nitrogen oxides, ammonia, etc. -
are the same gases that cause odor control problems at wastewater plants. Operators at
wastewater treatment plants are beginning to realize that the computerized control equipment
used in their facilities are subject to the same corrosive attack that leads to plant shut down in
other process industries.

Wastewater plant personnel responsible for operation and maintenance need to know if their
electronic control equipment is under attack from the corrosive, odorous gases. This question
can be answered by reactivity monitoring. There are several tools that can be used for
environmental reactivity monitoring: continuous corrosion monitors which provide data to the
computer control system, self contained monitors that can provide readouts of the environmental
condition and/or be linked to the control system, and reactivity coupons which can be used as
spot checks. These tools can correlate the corrosiveness of an environment to computer control
equipment reliability by published standard guidelines (Instrument Society of America’s
standard Environmental Conditions for Process Measurement and Control Systems: Airborne
Contaminants). These guidelines provide information as to the levels that will ensure reliable
operation. The standard levels and their corresponding equipment reliability predictions will be
presented here.

Once it has been determined that a corrosion problem exists, one must take steps to control it. If
source removal and/or dilution control are not options, one must consider removal of the
offending contaminants. Dry chemical scrubbing has been the method of choice for years in
corrosion control applications such as these. Several types of dry scrubbing systems are
available and their benefits as well as shortcomings are discussed here. The main factors
considered being the permanency of the removal obtained, capacity for specific corrosive gases,
and variations caused by different application issues. Many times applications have a need for a
multiple media approach where specific properties of each media are used to give a better overall


Most waste water treatment plants are familiar with the corrosion of pipes, Figure 1 A, and
supports, Figure 1 B, due to the presence of sulfides, chlorides, or other corrosive gases in the air
as well as in the process streams. Microelectronic devices are generally as susceptible to
corrosion as these materials, Figure 1 C (White, 1987). However, the fact that these same

                                                                  Proceedings of Texas Water 2003
                                                                               Corpus Christie, TX
environments can damage computerized control equipment at a waste water treatment plant has
not been considered to any great degree. Recently, one waste water treatment plant realized that
circuit boards of their control system were being replaced quite frequently (multiple times per
year). The hardware suppliers offer no warranty in the waste water industry, because they know
the environment will not allow the equipment to have a long life. Therefore, the waste water
treatment plant has to keep replacing circuit boards, but the root cause still exists and attacks the
old as well as new boards placed in the contaminated control room.

               (A)                                (B)                               (C)

Figure 1 – Corrosion of (A) Piping, (B) a Coated Handrail, and (B) a Printed Circuit
Board (NASA, 2001; Purafil, 2001).

In industries that have used distributed control systems, computerized process controls, and
electronic sensing/controlling equipment for years (paper mills, petroleum refineries, etc.), the
fact is well documented that the process control rooms must be protected from the corrosive
gases in the air. Results of corrosion in these industries begin as untraceable alarms that come
and go as corrosion products form circuit paths and clear themselves as current flows through
them. The final result of such corrosive attack is circuit board failure (Gavrilovic, 1987). This is
somewhat costly from the side of equipment replacement, but is even worse on the side of
production loss. In moderate sized pulp and paper processes, production losses can cost $10,000
per hour of downtime. Waste water process controls are subject to attack from the same gases
and the same component failures. The cost of process control failure may not be $10,000 per
hour of lost product, but there can be substantial amounts of waste to reprocess costing the plant
both time and money.


Corrosion can be defined as the “deterioration of a substance (usually a metal) because of a
reaction with its environments” (ISA, 1985). Contaminants that influence the rate of corrosion
can be in the form of gases, liquids, or solids. Generally speaking, most environmental controls
for process control equipment maintain the area free of particulate and liquid contaminants as
well as attempt to stabilize temperature and humidity. Therefore, the effects of these factors will
not be discussed in as great detail as the corrosive effects of gaseous contaminants.

Corrosion affects computerized control systems because they operate on the basis of electron
flow through conductive metals. There are the large circuits which allow controls to
communicate with sensors, valves, etc., and there are the microcircuits within the controls
themselves. It is the latter where the most damage can be done because of size and proximity of

                                                                 Proceedings of Texas Water 2003
                                                                              Corpus Christie, TX
the components. Computerized process control equipment generates heat while using air to cool
the devices and prevent thermal failures. Thus air is in contact with the microcircuitry allowing
the gases it contains, the temperature, and the humidity to affect the conductive metals.

The fact that electronics need stable environments, in terms of humidity and temperature, is
understood by most. Fluctuations in humidity and temperature can cause condensation on the
components and open the door for electrolytic corrosion as well as accelerate the reactions of
other components in the air. Effects of temperature and humidity are shown in Table I (ISA,
1985; Muller, 1998).

Several gases have been found to detrimentally effect equipment by reacting and catalyzing
reactions with base metals. These compounds can be broken into the following categories:
acidic gases, caustic gases, and strong oxidants. Descriptions of these compounds and their
effects are also found in the table below. The contaminants most well known for their corrosive
effects are chlorine and sulfur containing gases (typically acid gases). These can cause corrosion
at low parts per billion levels by attacking copper, tin, silver, aluminum, and iron alloys.

Table I – Factors Affecting Corrosive Aggressiveness of the Environment (ISA, 1985;
Muller, 1992; Muller, 1998)

       Factor                                          Description

 Acidic Gases           Chlorine (Cl2), chlorine dioxide (ClO2), hydrogen chloride (HCl),
                        hydrogen fluoride (HF), etc. can cause severe corrosion on electronic
                        components at very low concentrations (10 ppb). Hydrogen sulfide
                        (H2S) can attack metals at low ppb levels and its corrosive effects are
                        accelerated by humidity and chlorine compounds. Sulfur oxides (SOX)
                        can dissolve in water and form sulfurous and sulfuric acid to react with
                        metal conductors. Nitrogen oxides (NOX) can have catalytic effects on
                        sulfide and chloride corrosion as well as combine with moisture to form
                        nitric acid.
 Caustic Gases          Ammonia (NH3) and other caustic gases can corrode copper and copper
 Relative Humidity      Higher humidities as well as fluctuations in humidity greatly accelerate
                        the corrosive affects of other contaminants in the air.
 Strong Oxidants        Ozone (O3) and chlorinated gases (Cl2, ClO2, etc.) can attack the surface
                        of plastics and elastomers. They also catalyze sulfide corrosion.
 Temperature            Every 10ºC increase can cause destructive chemical reaction rates to
                        more than double. Fluctuations in temperature can cause condensation,
                        leading to electrolytic and chemical corrosion.

Types & Effects of Corrosion

There are several types of corrosion that can occur in electronic circuits (uniform, galvanic,
crevice, pitting, stress, cracking, pore, creep, fretting, electrolytic, and whisker growth). Three
types of corrosion found in analyzed circuit boards are described here. They are pore corrosion,

                                                                Proceedings of Texas Water 2003
                                                                             Corpus Christie, TX
creep corrosion, and “whisker growth.” Pore corrosion typically occurs on electric contacts
coated with noble metals. The noble metals have pores which allow the initiation of galvanic
corrosion (corrosion caused by the contact of two dissimilar metals). If any corrosive gases
(chlorine, hydrogen sulfide, sulfur dioxide) are present they will promote this corrosion of the
underlying metal. The corrosion of the underlying metal will then migrate through the pore to
the contact surface and increase the resistance of the contact to electron flow (Henriksen, 1991).
Creep corrosion is the result of copper or silver corrosion growing (or creeping) over the noble
metal top coating. This corrosion forms a non-conductive film on the contact surface and
inhibits the flow of current (Henriksen, 1991). “Whisker growth” can be defined as the growth of
microscopic metal crystals out of a conductive metal (Boonzaier, 1986). This is caused by the
free migration of ionic sulfide compounds over the conductive surface. They can collect at
dendrite boundaries where nucleation takes place and crystals grow out of the surface of the
metal (Arnold, 1954). These “whiskers” may cause untraceable alarms and other random
occurrence in the control system by making paths for current to flow (Michniewich, 1985).

Circuit boards from various facilities have been analyzed by Purafil. The images of corroded
components below were taken from such circuit boards. Figure 2 A is an example of pore
corrosion and creep corrosion taking place on traces of a printed circuit board (PCB). Figure 2 B
is an image of creep corrosion on the edge connectors of a circuit board. Edge connectors are a
very susceptible component of the circuit board. The corrosion salts from the reaction grow
underneath or over the noble metal top layers and separate the edge connector from the circuit
with a high resistance coating (Svedung, 1984). Figure 2 C shows evidence of both pore
corrosion and creep corrosion taking place on the edge connectors of a circuit board. Finally,
circuitry showing “whisker growth” can be seen in Figure 2 D.

                      (A)                                              (B)

                       (C)                                       (D)
Figure 2 – Images of corrosion: (A) pore and creep corrosion PCB traces, (B) creep
corrosion on connectors, (C) pore and creep corrosion on connectors, (D) whisker growth
on circuitry (Purafil, 2001; Muller, 1999).

                                                                Proceedings of Texas Water 2003
                                                                             Corpus Christie, TX

The best way to determine the corrosive aggressiveness of an environment is to analyze the
reaction bi-products of commonly used electronic materials (Noon, 1987). Thus this method of
environmental evaluation is commonly called reactivity monitoring. This method allows the
synergistic effects of multiple gases to be evaluated. The results of such environmental
assessments are also related to the ISA Standard S71.04-1985 for equipment reliability.

Initial Methods

A first step in this technology was to use copper coupons, for which reactivity has been studied
in much detail. These coupons are carefully prepared so that they are clean of contaminants
before placed in the environment being examined. Silver coupons are also commonly used to aid
in the determination of contaminant species present (Noon, 1987; Muller, 1991). It has been
shown that the presence of chlorine compounds can be identified from silver coupons as well as
other effects not captured by the copper coupons (Muller, 1990). The most inclusive method is
to use copper, silver, and gold coupons as shown in Figure 3 A.

           (A)                                 (B)                                   (C)
Figure 3 – Corrosion Classification Coupons: (A) A set of gold, silver, and copper coupons
before being exposed, (B) A set of gold, silver, and copper coupons having been exposed to
a G1 environment, (C) A set of gold, silver, and copper coupons having been exposed to a
GX environment.

Corrosion coupons are left in the environment for approximately 30 days. During this time, the
gases in the atmosphere contact and react with the metals to form corrosion films. Two coupon
sets that have been exposed to environments are shown in Figure 3 B and C. Figure 3 B is a set
of coupons from a mild G1 environment. Figure 3 C is from a severe GX environment. The
effects of the corrosive gases on these metal coupons is obvious. The coupons from the G1
environment appear very clean and the metal types can be identified easily. The coupons from
the GX environment are covered with corrosion films to the point that the copper coupon is
black, the silver coupon is a dull gray, and the gold coupon is covered with pores to make it look
dark gray. How these corrosion films correlate to environmental classifications will be discussed
in the next section. However, if these gases are in contact with circuitry, it is obvious that the
metallic leads would be affected in the same way as these coupons. Appearance change would
be first, but several months of a GX environment would cause intermittent failures and

                                                                Proceedings of Texas Water 2003
                                                                             Corpus Christie, TX
eventually total circuit failure. The thickness of the films are measured in angstroms of film
growth by electrolytic reduction of the metal coupon. The results of analysis are as follows:
total angstroms of corrosion formed, angstroms of individual corrosion films, and corrosive
gases which may be present. Presence of the following types of chemical species can typically
be detected from the analysis (as highlighted in Table II): active sulfur compounds, inorganic
chlorine compounds, oxides of nitrogen, and oxidized forms of sulfur. The total angstroms of
corrosion can be correlated to environmental classifications as is described in the section below.

Table II – Metal Coupons for Reactivity Monitoring

 Common Metals Used                              Chemical Species Detected

     Copper & silver          active sulfur compounds, inorganic chlorides, oxidized forms of
                              sulfur, and oxidized forms of nitrogen

The Standard for Environmental Assessment

The Instrumentation, Systems, and Automation Society (previously Instrument Society of
America) has published guidelines for classification of corrosive environments. The standard
ISA-S71.04-1985 correlates the environment classifications to angstroms of corrosion developed
on sacrificial copper strips, as described above. Environments are separated into four severity
levels. Each severity level has an explanation which predicts the reliability of equipment in that
environment. Table III displays the factors of the classification scheme.

Table III – Classification of Reactive Environments (ISA, 1985)

 Severity   Environmental        Copper Reactivity
                                                              Environmental Reliability
  Level      Description             Level*

    G1            Mild                 < 300          Corrosion is not a factor

    G2          Moderate               < 1000         Corrosion may be a factor

    G3           Harsh                 < 2000         High probability of corrosive attack.
                                                      Environmental controls or specially
                                                      designed and packaged equipment should
                                                      be used.

   GX            Severe                > 2000         Only specially designed and packaged
                                                      equipment would be expected to survive
                                                      without environmental controls.
 * In angstroms per 30 days

                                                                 Proceedings of Texas Water 2003
                                                                              Corpus Christie, TX
The table can be read as follows: for the severity level G1 (Mild), the corrosion formed on a
copper coupon must be less than 300 angstroms. At the G1 level corrosion is not a factor in
determining equipment reliability. G1 is the best environmental classification and GX is the
worst. In a GX environment, only specially designed equipment would be expected to survive.
Environments classified as G1 or lower have been recommended for process control equipment
(Lofton, 1995; HP, 1983). Table IV describes circuit board failures and their corresponding
severity levels. Environments in the G1 range have been shown to cause intermittent edge card
failures in 1 to 3 years. In the G2 range, these intermittent edge card failures can begin in 6-8
months. In the GX range, corrosion has been seen to creep across insulation and PCB traces in 4
to 6 months (HP, 1983). These failures can range from false signals to board failure and thus
loss of process control during that time. Therefore, it is very important to keep the electronics of
a control area in a protected environment.

Table IV - Reported circuit board failures with corresponding copper corrosion rates (HP,

   SEVERITY                         FAILURE TYPE                            Time to
    LEVEL                                                                   Failure

                    Edge card failures                                   4-5 years
 G1 – Mild          Edge card intermittent failures                      1-3 years
 G2 – Moderate      Edge card intermittent failures                      6-8 months
 GX – Severe        Creep across insulation & PCB traces                 4-6 months

The ISA standard also provides gas concentrations that are believed to approximate the copper
reactivity levels. These are displayed in Table V. It is evident that the three most corrosive
gases from this list are chlorine (Cl2), hydrogen fluoride (HF), and hydrogen sulfide (H2S). The
standard ranks mercaptans and organic sulfides with Cl2 and H2S in corrosive potential. Thus,
the odorous compounds present at waste water treatment facilities can be some of the most
corrosive gases to computerized control equipment.

Table V – Approximated Gas Concentrations at the Severity Levels (ISA, 1985).

 Gas                      G1                 G2                    G3                  GX
                   Concentrations Concentrations              Concentrations      Concentrations
 Cl2                      <1                 <2                   < 10                > 10
 H2S                      <3                < 10                  < 50                > 50
 HF                       <1                 <2                   < 10                > 10
 NH3                    < 500             < 10,000              < 25,000            > 25,000
 NOx                     < 50               < 125                < 1250              > 1250
 O3                       <2                < 25                  < 100               > 100
 SO2, SO3                < 10               < 100                 < 300               > 300
*Note: All concentrations in ppb by volume.

                                                                 Proceedings of Texas Water 2003
                                                                              Corpus Christie, TX
Technology Developments in Reactivity Monitoring

Using metal strips that are placed in a corrosive environment can be very useful for initial
classification of an environment or for verification of a protected environment. They also allow
detection of what types of contaminants are present. However, the results produced are only
indicative of the time for which the metal strips were exposed to the environment. This leads to
the need of a real time monitor that can notify operators of corrosive changes in the environment

Quartz crystal microbalance (QCM) technology allows the corrosive aggressiveness of an
environment to be monitored in real time. The visual comparison of corrosion on copper and
silver coupons to corrosion on copper and silver plated QCMs can be seen in Figure 4. The
quartz crystal oscillates at a certain frequency when no corrosion is present. When contaminants
in the air react with the metal plating on the quartz crystal, the frequency of oscillation changes
and this change in frequency is transformed to angstroms of corrosion through patented

                                Figure 4 – Coupons and QCMs

The classification of the ISA standard is based on copper coupon exposure over 30 days. Using
the QCM, the classification scheme can be evaluated over a 24 hour period instead of over a 30
day period. This allows detection of events that release corrosive gases into the monitored space.
An event can be a shift in wind direction, a release of corrosive gases from part of the process, or
propping open of control room doors. Using the QCM technology allows data to be logged in
the process control system so cyclic events can be found and a solution formed. Table VI
correlates the 30 day corrosion standard with 24 hour measurements. A G1 environment is
defined by less than 300 angstroms of corrosion in 30 days on a copper coupon. If this is
normalized to a 24 hour period, it would be less than 10 angstroms in 24 hours. By logging the
data in a control system, alarms or alerts can be triggered when readings above the desired level
occur to prompt further evaluation of the event or environmental controls.

                                                               Proceedings of Texas Water 2003
                                                                            Corpus Christie, TX
Table VI – ISA Severity Levels Correlated to 24 Hour QCM Readings

   Environmental Severity         ISA-Standard S71.04-1985             QCM Correlation
       G1 – Mild                       < 300 Å / 30 days                < 10 Å / 24 hours
        G2 – Moderate                  < 1000 Å / 30 days               < 33 Å / 24 hours
        G3 – Harsh                     < 2000 Å / 30 days               < 66 Å / 24 hours
        GX – Severe                    > 2000 Å / 30 days               > 66 Å / 24 hours


The most common method to protect electronic control equipment from corrosive gases is
through dry-scrubbing. This method has been used for years in other process industries
(chemical, petrochemical, pulp and paper, semiconductor, power production). Typically, areas
housing the distributed control system (DCS), computer controls, or other system electronics are
pressurized with clean air so contaminant infiltration cannot occur. The air is put through a
system employing the adsorptive and chemisorptive media which removes and reacts with
corrosive gases to give a protected environment for operation of the computerized controls.

For areas with high concentrations of corrosive gases (> 0.5 ppm), bulk fill units are most often
employed. These would contain beds of dry-scrubbing media one to three feet in depth to
remove contaminants with high efficiency. These are used to pressurize the control room to 0.05
to 0.1 inches of water and keep it free of gaseous contaminants. Then a recirculation unit is
many times put inside the control room to remove any internally generated contaminants brought
in by personnel leaving and entering the room and other means. One such setup is shown in the
Figure 5.

                          Recirculation unit

 Air lock entry                                      A/C            Pressurization unit

                                                                                  Outside air

Figure 5 – Diagram showing pressurization and recirculation for a computer control room

Types of Dry-Scrubbing Media

There are different types of dry-scrubbing media used to remove corrosive gases from the air.
Reactive gases (hydrogen sulfide, sulfur dioxide, hydrogen chloride, chlorine, ammonia) are
removed via chemisorption on engineered media, which reacts with the contaminants and
permanently holds them on the media. Higher molecular weight gases are removed via
adsorption on media with high surface area to mass ratios.

                                                                Proceedings of Texas Water 2003
                                                                             Corpus Christie, TX
Engineered media is manufactured through a dry feed process of specific base components and
agglomerated by using reactive liquid impregnants. Adsorptive media is most commonly made
from coconut shell or bituminous carbon which has been permeated with a system of pores.
Images of such media are shown in Figure 6. Figure 6 A is engineered media impregnated with
potassium permanganate. It has the ability to react with a wide range of gases, from organics to
active sulfur compounds and chloride compounds. Figure 6 B is engineered media impregnated
with potassium hydroxide. It is able to realize high capacities for acidic gases including chlorine
and sulfur compounds. Figure 6 C is bituminous activated carbon. It has good effectiveness
against hydrocarbons, some organic sulfides, and high molecular weight volatile organic
compounds. The most effective gas-phase filtration will involve a combination of dry-scrubbing
media such as those shown here.

            (A)                                (B)                                 (C)
Figure 6 – Dry-scrubbing media: (A) Potassium permanganate engineered media, (B)
Potassium hydroxide engineered media, and (C) Bituminous activated carbon.


One of the many cases where this advice has been followed was in Cairo, Egypt. A telephone
switching station was located approximately half a mile away from a sewage processing area.
This processing area released corrosive gases such as hydrogen sulfide (H2S) and sulfur dioxide
(SO2) into the atmosphere, which made its way into the switching station. When the air quality
of the switching station was assessed, a SO2 concentration of 3 ppm was measured. This
concentration of SO2 corresponds to a GX environment and was far in excess of the electronic
equipment manufacturer’s specification, 5 ppb. The response of the switching station supervisor
was to install a modular dry-scrubbing system to remove the corrosive gases. His action resulted
in the target concentration of 5 ppb being met and the vital electronic equipment inside the
switching station being protected from corrosive attack.

There have been other cases in the United States where QCM sensors are lasting less than one
month at wastewater facilities. A QCM sensor can read corrosion up to 4,000 angstroms. This
would indicate that the environment is GX - Severe. In this range, failures can happen in 4 to 8
months (edge card failures, creep across traces, etc.). The respective plant indicated that this
data has been very useful in getting changes made to the HVAC system in that control room.
Thus, protective action can be spurred on by such findings as these.

                                                               Proceedings of Texas Water 2003
                                                                            Corpus Christie, TX

The types and concentrations of gases present at waste water treatment plants are able to cause
severe corrosive damage to the electronic control equipment. The most corrosively damaging
gases are the prevalent ones present at such plants (hydrogen sulfide, chlorine compounds
mercaptans). These gases are able to cause corrosion at low parts per billion concentrations.
Effects of such gases have been described as pore corrosion, creep corrosion, and whisker
growth. All of which cause circuit failures either by making random circuit paths or by forming
non-conductive layers.

Assessment and monitoring of an environment can be carried out through reactivity monitoring.
The use of copper, silver, and gold metal coupons or copper and silver plated quartz crystal
microbalances, can determine the corrosive potential of an environment. They are directly
related to the ISA Standard S71.04-1985 which describes the severity levels G1-Mild, G2-
Moderate, G3-Harsh, and GX-Severe. It has been recommended by computer manufacturers, as
well as others, that the computer equipment be kept in or below the G1 level. Correlation of
failure reports and copper corrosion rates have indicated that the frequency of failures quickly
increases above this level.

Protection of computerized and electronic control rooms has been performed for years in other
process industries (pulp and paper, petrochemical, power production) by using dry-scrubbing
technology. Such systems are able to perform gas-phase filtration of the air coming into a
control room as well as positively pressurize the control room to prevent infiltration of
contaminants. Commonly recirculation units are also placed in the control room to deal with
internally generated contaminants. Dry-scrubbing systems using multiple dry-scrubbing media
allows for the most effective filtration of contaminants. Field experience has shown that gaseous
levels at waste water treatment plants are in the range which can cause severe damage to
electronic components. It has also shown that corrosive contaminants can be monitored as well
as removed by the technologies presented here. Environmental assessment and protection of
computer control rooms, as well as off-site controls, is key to maintaining a cost effective and
properly functioning plant.


Arnold, B.M. (1954, November). “A Hidden Cause of Failure in Electronic Equipment: Metal
       Whiskers.” Electrical Manufacturing. pp. 110-114.
Boonzaier, W.G. (1986, July). “Acid Gas Corrosion Protection.” SA Measurement & Control.
       pp. 5-9.
Gavrilovic, John. (1987). “Microchemical Identification of Corrosion Related Problems on
       Electronic Components.” Electronic Packaging and Corrosion in Microelectronics -
       Proceedings of ASM’s Third Conference on Electronic Packaging: Materials and
       Processes & Corrosion in Microelectronics. ASM International. pp. 63.
Henriksen, Jan, & Hienonen, R., & Imrell, T., & Leygraf, C., & Sjogren, L. (1991). Corrosion of
       Electronics: A Handbook based on Experiences from a Nordic Research Project.
       Sweden: Litohuset AB, Stockholm.
HP. (1983, February). “Corrosion Rate and Atmospheric Pollutant Level Guidelines for Hewlett-
       Packard Computer System Installations.” Cupertino, CA: Hewlett-Packard Co.

                                                              Proceedings of Texas Water 2003
                                                                           Corpus Christie, TX
ISA. (1985). Standard ISA-S71.04 - Environmental Conditions for Process Measurement and
        Control Systems: Airborne Contaminants. Research Triangle Park, NC: Instrument
        Society of America. pp. 12, 13, 10, 7, 9, 15, 11.
Lofton, Bob, & Fugler, Allen. (1995, July). “How to Protect Your Controls From the Elements.”
        Control. pp. 50-55.
Michniewich, James J. (1985, March). “Field Experience with Corrosion of Computer
        Equipment.” Corrosion 85: The International Corrosion Forum Devoted Exclusively to
        the Protection and Performance of Materials. Houston, TX: NACE.
Muller C.O. (1990, February). “Combination Corrosion Coupon Testing Needed for Today’s
        Control Equipment: Accurate Reliability Analyses for Electron/Electrical Devices in
        Pulp and Paper Mills Require Silver and Gold Coupons in Addition to Copper.” Pulp &
Muller, C. O. (1991, January). “Determination of Electrical/Electronic Equipment Reliability in
        Pulp and Paper Mills.” Proceedings of Canadian Pulp & Paper Association 77th Annual
        Meeting. Montreal, Canada: CPPA.
Muller, C. O. (1992, November/December). “The Neglected Element: Gaseous Corrosion.” The
        Computer Operations Manager.
Muller, Chris. (1998). “Humidity and Corrosion.” Purafil Literature [on-line]. Available: Accessed
Muller, Christopher O. (1999). “Control of Corrosive Gases to Avoid Electrical Equipment
        Failure.” PITA Annual Conference/ PAPEX ’99.
NASA. (2001). “Corrosion Fundamentals.” Kennedy Space Center Corrosion Technology
        Testbed [on-line]. Available:, Accessed 3/06/2002.
Noon, David W. (1987). “Corrosion and Reliability: Industrial Process Control Electronics.”
        Electronic Packaging and Corrosion in Microelectronics - Proceedings of ASM’s Third
        Conference on Electronic Packaging: Materials and Processes & Corrosion in
        Microelectronics. ASM International. pp. 50.
Purafil. (2001). Internally Performed Analysis of Corroded Circuit Boards.
Svedung, O., & Johannson, G. (1984). “Atmospheric Corrosion of Gold Coated Contact
        Materials Exposed to Humid Atmospheres Containing Low Concentrations of SO2/NO2
        and SO2/Cl2.” Proceedings of the 9th Scandinavian Corrosion Conference.
White, Henry S. (1987). “Corrosion Principles in Microelectronics.” Electronic Packaging and
        Corrosion in Microelectronics - Proceedings of ASM’s Third Conference on Electronic
        Packaging: Materials and Processes & Corrosion in Microelectronics. ASM
        International. pp. 33.


To top