Home » Welding Safety: Welding Fumes
Welding Safety: Welding Fumes
Welding Safety: Protecting Employees from the Dangerous Effects of Overexposure to Welding Fumes
Mike Harris, Ph.D. CIH
Overexposure to some kinds of welding fumes can cause brain damage or lung cancer. Manganese and hexavalent chromium are the two components of welding fume that are associated with these illnesses. It is reasonable to ask:
In addressing these questions, this article briefly addresses questions regarding:
Welding ordinary carbon steel with most common welding process releases manganese to the shop air. Overexposure to manganese has been shown to cause damage to the brain that produces a long list of ailments including [i]:
The peer-reviewed scientific literature contains many articles that describe the neurotoxic effects of manganese exposure and welder overexposures to manganese. Rather than recite those findings here, a partial list of these papers is provided at the end of this article [ii].
What if the carbon steel my employees are welding does not contain manganese? Where does the manganese come from? According to the American Welding Society, Shielded Metal Arc Welding (“SMAW” or “stick welding”) is the most common welding process currently in use [iii]. The manganese comes from the SMAW welding electrodes. In fact, most of the welding fume generated during SMAW work comes from the electrodes rather than the metal being welded [iv].
It is the flux coating on the SMAW electrodes for ferrous metals, rather than the filler metal itself, that contains manganese [v]. Look at a current and up-to-date Material Safety Data Sheet (MSDS) for ferrous metal SMAW electrodes and you’ll see manganese listed. There are technical reasons why the manganese is incorporated in the flux. These reasons focus primarily on reducing weld cracking and reducing atmospheric contamination of the weld. You have no doubt noticed the cloud of fume (or “smoke” [vi]) that comes from the SMAW process. This cloud of fume physically displaces the atmosphere so that oxygen and nitrogen in the atmosphere do not contaminate and weaken the weld metal.
Some of the manganese in the cloud of flux fume becomes incorporated in the weld metal. This manganese increases the strength of the steel when it is hot, and as it cools, helps to reduce weld cracking [vii].
Manganese in the cloud of flux fume also reacts with oxygen in the atmosphere that may mix with the cloud of fume. When the manganese in the cloud of fume reacts and combines with the oxygen, the oxygen is prevented from contaminating the molten metal. This chemical reaction, preventing the atmospheric oxygen from getting to the molten metal, thereby helps preserve weld quality.
You have also probably noticed that some of the cloud of flux fume also ends up in the welder’s breathing zone. Along with the rest of the components of the flux fume, the SMAW welder will inhale manganese. The scientific papers noted at the beginning of this article make it clear that SMAW welders working with ordinary carbon steel are likely to be overexposed to manganese during the course of their work.
What is meant by “overexposure” to manganese? There are several values, measured in terms of the milligrams of manganese per cubic meter of air, that have been offered to establish what constitutes an acceptable concentration of manganese in the welder’s breathing zone. Three of these are listed below:
OSHA has recently adopted a new Permissible Exposure Limit for hexavalent chromium or “CrVI.” This PEL is set forth in 29 CFR 1910.1026 as 5 micrograms per cubic meter of air (5µg/m3). Since there are 1000 micrograms in a milligram, this value may also be written as 0.005 mg/m3. 5µg/m3 is the same as 0.005 mg/m3 so don’t let the scientific short hand throw you if you see it presented differently in different articles.
When we consider that there are more than 28 million micrograms in an ounce [ix], and that one cubic meter is about 1.3 cubic yards [x] it becomes quite clear that this is a VERY low exposure limit! Why is it so low? And where does this hexavalent chromium come from? Will my employees be overexposed to hexavalent chromium just by handling a chromium-containing alloy such as “stainless steel?”
Answering the question about the low hexavalent chromium PEL; when dealing with a known human carcinogen, it is accepted practice to set the exposure limit at a very low level. Given that cancer in any form is generally considered to be a bad thing, and that the process of carcinogenesis (development of cancer) is not always well understood, it’s understandable that a low exposure limit is considered prudent. There has actually been some discussion in the scientific community regarding this PEL. Some claim that it is not low enough and that 1µg/m3 is a more appropriate value. Nonetheless, for those of us working in the
In responding to the second question about hexavalent chromium generated during welding processes we need only consider the high energy environment of the electric arc welding process. During SMAW or Gas Metal Arc Welding (GMAW) work, the end of the electrode or filler wire is melted off and deposited in the weld pool. As the weld metal being deposited during welding passes through the welding arc, electrons are stripped from some of the chromium atoms. If 6 electrons are stripped from a chromium atom, that atom becomes hexavalent chromium [xi] and this atom, being shy of a full deck of electrons, is appropriately referred to as an “ion.”
The CrVI ion is not stable. Rather it is highly reactive and will often rapidly combine with atmospheric oxygen to form Cr2O3. In the case of SMAW work and Flux Core Arc Welding (FCAW) work, the alkaline metals in the flux will tend to stabilize the CrVI so that it remains in the negative 6 valence state long enough to reach the breathing zone of the welder [xii]. This means that the highly reactive CrVI ion can be inhaled and can attack the lung tissue.
Since most of the welding processes fume generated comes from the welding consumable (noted above), the filler metal employed during electric arc welding is the primary source of hexavalent chromium in welding fume. Consequently, when working to comply with the provisions of the OSHA Chromium VI standard (29 CFR 19101.1026) one would be well-advised to review the Material Safety Data Sheets (MSDSs) for the welding consumables to determine if chromium (that may be converted to hexavalent chromium during welding) is present in the consumables. One should also be aware that some carbon steels contain recycled metals that include chromium. Even though most of the welding fume comes from the electrodes/filler wire, some of the fume does come from the metal being welded. Consequently, there is a potential for hexavalent chromium in the welding fume from these steels. The MSDS for these steel may or may not include chromium because, when present, chromium may constitute only a fraction of a percent of the metal.
Here, it gets a little sticky. The OSHA Hazard Communication Standard (29 CFR 1910.1200 (d)(5)(ii) requires that the MSDS lists any carcinogen if it is present in the material at concentrations of 0.1% or more. So, one would think that the MSDS for a chromium-containing alloy would include hexavalent chromium, right? However, the chromium in the filler metal (or the base metal, for that matter) does not exist as CrVI. Chromium is generally, and appropriately, listed as chromium metal in a neutral valence state. Chromium in the neutral valence state is not listed as a carcinogen and need not be listed unless it is present at a level of 1% or more. So, how do you find out if the chromium in the metal is being ionized to the negative 6 valence state in concentrations high enough to present a health risk to your employees? YOU MEASURE IT!!
Measuring personal exposures to airborne contaminants is called “personal breathing zone monitoring” and is accomplished by using small battery-powered vacuum pumps (about the size of 2 or 3 packs of cigarettes), flexible tubing and a filter. The pumps are usually hung on the worker’s belt. A piece of flexible plastic tubing (usually “Tygon®”) connects the pump to a filter cassette about 1½” across. This filter cassette needs to be placed in the welder’s breathing zone, inside the welding helmet. The vacuum pump is turned on at the beginning of the shift and turned off at the end of the shift. This period of time is called the “sample period” or “monitoring period” and is measured in minutes. Flow through the filter cassette is calibrated via a specialized flow meter and is measured in liters per minute. Knowing the flow rate and the sample period, one multiplies the flow rate (liters per minute) by the sample period (minutes), the minutes cancel out and the sample volume (number of liters of air that flowed through the filter) is recorded. The filter is sent to an analytical laboratory for analysis. It is best practice to use a lab that has been accredited by the American Industrial Hygiene Association (AIHA). The lab is informed as to how much air has been drawn through the filter by recording that value on the Chain of Custody that goes along with the filter to the lab. The same technique (using a different filter and analytical method) is used to monitor for manganese exposures.
If there is a reasonable likelihood that workers will be exposed to CrVI, The OSHA CrVI standard requires the employer to perform initial monitoring that includes personal breathing zone air samples for:
The CrVI standard also notes that Historical and Objective data are acceptable but it may be difficult to prove or document that data not collected at your facility on your workers is really representative of what happens at your plant.
Sounds simple enough, doesn’t it? It’s also a fairly specialized task that is best performed by a trained Industrial Hygienist (or Occupational Health Scientist in the
What if the manganese or CrVI values measured by personal breathing zone monitoring are above the exposure limits for welding process safety? Do we just stop welding and go out of business? Do we ignore the monitoring results and hope no one gets sick? Do we just put people in respirators? Is there any other choice?
Obviously, the first two are not really choices at all. And welding respirators are really not the first choice. In fact if we look at the OSHA Chromium VI standard (29 CFR 19101.1026 (f)) the federal standard says we should:
Again, using engineering controls in preference to respirators is best practice for manganese as well as other airborne fumes and contaminants.
Why not just go with respirator to begin with? Wouldn’t that be easier?
Not necessarily. To begin with, there are a lot of hoops to jump through when putting people into respirators. While the OSHA standards apply only to those of us working in the
As an aside, it is worth noting that supplied air respirators are often fitted with belt-mounted vortex coolers on the airline to the respirator. Vortex coolers provide a much more comfortable working environment in hot conditions. Vortex coolers consume more air than is necessary for respiration alone. Consequently, it may be more economical to use a breathing air compressor or an ordinary compressor with a breathing air panel than to use compressed breathing air cylinders when vortex coolers are used.
Engineering controls cost more to implement than respirators at the front end due to higher welding fume protection equipment purchase and installation costs. Also, like respirators, engineering controls require maintenance. In response to a need for better understanding of these maintenance requirements, the ACGIH has published a volume describing appropriate maintenance procedures for engineering controls [xiv].
However, engineering controls have a lot of advantages over respirators. One way to discuss engineering controls for welding fumes is address a few commonly asked questions:
Answering the last question first; No, engineering controls will not ALWAYS be the best solution. When work is conducted in small, restricted or confined spaces or areas, it may not be physically possible to implement an effective engineering control. In that case, best practice is to use engineering controls to lower exposures as much as possible and complete the employee exposure reduction using respiratory protection.
When discussion control of welding fumes, an Engineering Control is a ventilation system. Since the fume from welding is a very small particle (less than one micron in aerodynamic diameter) it is light and moves with the surrounding air. If we ventilate the worker’s breathing zone and remove contaminated air, we remove the welding fume along with it.
Presuming effective capture and removal of the fume, the best engineering controls require the least worker involvement. By comparison, use of respirators requires worker acceptance of the respirator as part of the suite of PPE required for the job and a lot of worker involvement every time the respirator is donned. In some cases, this requirement may not receive the attention it deserves and the respirators may not be used as intended. Engineering controls, while still requiring worker involvement, are less demanding in that respect and are judged more likely to be used as intended and are therefore preferred over respiratory protection.
It is important to be aware that engineering controls may not always completely preclude the need for respiratory protection. However, engineering controls will, in almost every instance, reduce the level of respiratory protection needed to control employee exposures to less than the TLV for manganese or PEL for hexavalent chromium. As an example, using engineering controls to lower manganese concentrations from 15 mg/m3 (75 times the TLV) to 1.5 mg/m3 (7.5 times the TLV) reduces the required level of respiratory protection from Half Face Supplied Air Respirator to Half Face Air Purifying Respirator. This reduction in respiratory protection requirements eliminates two drawbacks to supplied air respiratory protection:
In the simplest sense, opening the shop doors and using a big fan to blow air though the shop may be considered an engineering control. This technique of diluting the fume concentration by blowing a lot of air through the shop is called (not surprisingly) “dilution ventilation.” The nice thing about dilution ventilation is that, if it works (as measured by personal breathing zone monitoring), this is a simple solution and, in small scale applications, may be a relatively cheap solution. Of course, this practice may not be as effective as desired for welders in the back corners of the shop and working conditions may be much breezier than desired for workers nearer the fan. And, in winter the cold blast of air generated by the fan may cause the fan to become unplugged, “somehow.” Also, welders working nearer the fan may experience compromised weld quality due to excessive air movement displacing the shielding gases (in the case of GTAW, GMAW and gas-shielded FCAW) or shielding fume generated by welding flux (in the case of SMAW and FCAW).
From an exposure control point of view, the drawbacks to using a big shop fan (or fans) for welding fume protection include:
Many of the drawbacks to simple dilution ventilation may be successfully addressed by using Ambient Capture Air Cleaning. The term “Ambient Capture Air Cleaning” refers to installing an appropriate number of air cleaners above the welding areas, cleaning the ambient air, and returning the clean air to the shop. This technique provides the primary benefit of simple dilution ventilation (lower overall fume concentrations) without the need for bringing in a lot of cold air in the winter and hot air in the summer. However, the welding fume in the immediate area of the welder (specifically the breathing zone) may not be well diluted. In this respect, both simple dilution ventilation and ambient capture air cleaning are more effective at reducing exposures for other employees in the shop than they are for the welders themselves.
The benefits of using ambient capture air cleaning rather than using a big shop fan (or fans) for protection against the risks of welding fumes include:
Several air cleaning technologies are available. The most common are:
Recirculation reduces annual utility costs for heating and/or cooling the shop air. The heating and cooling costs associated with just “blowing it out the door” can be calculated using these two links:
For example, using the “heatloss” spreadsheet, the heating costs associated with blowing 2,000 cfm out of the shop for a 60 hour work week (who works 40 hours anymore?) when a gas-fired heat exchanger is used to heat the shop to 74 degrees in
Air cleaners must be maintained to perform properly, just like the home vacuum cleaner. If the bag in the vacuum cleaner is allowed to stay dirty, the vacuum cleaner will not perform up to expectations. The same applies to an air cleaner. The maintenance schedule for an air cleaner in a welding application is driven by the amount of fume that is captured. The amount of fume per unit time (e.g., per week) is a function of:
Typical maintenance requirements are briefly outlined below:
Local exhaust ventilation (LEV) has an obvious advantage over dilution ventilation and/or ambient air cleaning: The welding fume is captured BEFORE it passes by the welder’s breathing zone. For this reason, the American Welding Society (AWS) recommends LEV as the preferred means of collecting welding fume for the work environment [xv]. In “Industrial Ventilation, A Manual of Recommended Practice” published by the ACGIH, LEV is also described as the preferred method for capturing welding fumes in the workplace [xvi].
The term “local exhaust ventilation” (or LEV) as used here refers to equipment sometimes called “elephant trunks.” The exhaust from the LEV is typically exhausted to the outside without regard to air filtration before leaving the shop. Before installing this engineering control, it may be prudent to check with local environmental officials to verify that implementing this kind of exhaust system will not result in the facility being identified as a “point source emitter.”
Local exhaust ventilation should be applied with care as excessive airflow velocities can create the same problems noted for excessive airflow velocities in the discussion on using big shop fans, above. The AWS offers the following guidance for application of ventilation without compromising weld quality:
“Ventilation should not produce more than approximately 100 feet per minute (0.5 meters per second) air velocity at the work (welding or cutting) zone. This is to prevent disturbance of the arc or flame. It should be recognized that approximately 100 feet per minute (0.5 meters per second) air velocity is a recommended maximum value for quality control purposes in welding or cutting.It is not intended to imply adequacy in contaminant control for worker protection [xvii]”
Placing the exhaust system intake too close to the weld can create airflow velocities in excess of 100 fpm (100 fpm is equivalent to 1.2 miles per hour). Particular care may be required when welding alloys that are more easily compromised by atmospheric contamination (e.g., stainless steel). Air flow velocity meters can be rented from suppliers to be found online and used to verify air flow rates at the weld pool when LEV is used to control welder exposures to components of welding fume.
Source Capture Air Cleaning essentially uses the same sort of elephant truck as LEV and carries the same technical advantages as LEV with the added benefit of cleaning the air. Cleaning the air allows it to be re-circulated into the work environment. The economic benefits described for ambient capture air cleaning compared to simple dilution ventilation also apply to Source Capture Air Cleaning compared to simple LEV.
For comparison, we’ve provided below a brief evaluation comparing local exhaust ventilation to welding fume control from 5 welding stations.
Let’s assume the 5 welding stations are in use 40 hours per week each. It is generally accepted that for an 8” diameter exhaust hose or arm, a minimum of 1000 cfm is recommended. If we use the 1000 cfm per station times 5 stations, we will require a total of either 5000 cfm of exhaust or air filtration. Again, using the heat loss calculator and assuming we are located in Chicago and have a desired indoor temperature of 70 degrees F heated by direct fired gas, the cost of the heat loss for ventilation is $4784 (q = 800, C= .01664, dg= 7468). A disposable media type system like our model M73 with all necessary filters, motor, blower and electronics can be purchased for around $6000. The cost of the ductwork would be necessary for local exhaust ventilation or filtration, so it can be ignored as it would not be different between the two options (with the exception that the exhaust ventilation would actually require more ductwork since it needs to be ducted to the outside). The return on investment (ROI) in this instance is a little over 1 year using only the energy savings due to heat loss. Additional considerations would need to be made as this example does not even take into consideration the cost of the exhaust fan in comparison, but there is obviously an additional cost for changing the filters of the air filtration system as well. There may also be the need for additional makeup air equipment due to the large negative pressure that an exhaust fan of this size would put on the building.
The images below show Source Capture Air Cleaning in action.
Air Cleaner above is model AQE 4000, a welding fume filtration system with mechanical source capture arms on a welding and grinding station.
Photo above of source capture from a welding station using model AQE 2000 portable cartridge collector from Air Quality Engineering, Inc.
The GMAW and FCAW welding processes are amenable to installation or integration of source capture equipment as a part of the welding “gun.” These guns are generally referred to as “fume guns.” The source capture intake is either built into the fume gun during manufacture or added on the gun as an accessory. In either case, the fume intake is adjacent to the gas cup and is connected to a small diameter hose (often approximately 1½” to 2”in diameter) that is parallel to the power cable and wire feed tube. The other end of the hose is connected to a specialized blower assembly that is capable of moving air through the small diameter duct or hose. This blower must be capable of pulling an adequate amount of air at up to 60” of static pressure. Static pressure is a measure of how had the blower has to work to overcome the airflow restriction in a ventilation system. By way of comparison, LEV and source capture air cleaners typically include blowers designed to work at less than about 6” of static pressure. As you can see there is a big difference between the requirements for a fume gun and an LEV or source capture air cleaner. A simple shop vacuum will not provide this level of performance necessary for a fume gun nor will it provide the level of filtration necessary for recirculation of the air to the shop.
While fume guns are useful, they are less easily manipulated than ordinary GMAW or FCAW guns due to the fairly rigid hose. Why is the hose more rigid than is optimal for maneuverability? The hose must be strong enough to resist collapse from the high vacuum necessary to move air through its’ small diameter and this strength makes them less flexible that one might desire. Also, one should note that fume guns work reasonably well when welding in the flat position and overhead position but are less effective when welding in the vertical and horizontal positions [xviii]
Due to their toxic effects, both manganese and hexavalent chromium in welding fume have been the subject of increased attention. Employees welding ferrous metals via common welding processes may be exposed above the TLV for manganese. Employees welding chromium-containing alloys may be exposed above the PEL for hexavalent chromium (CrVI).
Engineering controls are the preferred method for lowering employee exposures to the contaminants in welding fume. However, respiratory protection may be required n addition to engineering controls to decrease employee exposures to less than the TLV for manganese or the PEL for CrVI.
Welding respirators require a program of respirator and cartridge selection, medical qualification, training, fit-testing and respirator maintenance and storage. Best practice and, in the
Source capture or local exhaust ventilation is likely to be more effective at reducing welder exposures than dilution ventilation or ambient capture air cleaning. Any engineering control requires periodic maintenance. When using air cleaners instead of fume exhaust, filtration maintenance costs may be more than compensated for by utility cost savings associated with keeping heated or conditioned air in the building rather than blowing those heating and cooling dollars out of the building.
Mike Harris received an earned research Doctorate from
In 2002 Dr. Harris was named "Friend of the American Industrial Hygiene Association Press” and in 2003, his book, "Welding Health and Safety: A Field Guide for the OEHS Professional" was selected as “Critic’s Choice” for best non-committee AIHA publication.
He is lead author of Chapter 42 "Confined Spaces in "The Occupational Environment - Its Evaluation and Control" and Editor of "Essential Resources in Industrial Hygiene, A Compendium of Current Practices, Standards and Guidelines.”
Mike is also co-author of "Field Guidelines for Temporary Ventilation of Confined Spaces with an Emphasis on Hotwork" with Stephanie Carter and Lindsay Booher.
His welding experience includes:
Mike can reached via Air Quality Engineering or at the address below:
Mike Harris Ph.D., CIH
office: (225) 387-2847
[i] “Documentation of the Threshold Limit Values and Biological Exposure Indices – Manganese” American Conference of Governmental Industrial Hygienists,
[ii] Bowler, R. M., (2003): Decrements in verbal learning, working memory, cognitive flexibility, visuomotor processing speed and motor efficiency among welders when compared to a control group of non-manganese-exposed workers.
Bowler, R. M., Gysens, S., Diamond, E., Booty, A., Hartney, C. and Roels, H. A.: Neuropsycological sequelae of exposure to welding fumes in a group of occupationally exposed men. International Journal Hygiene Environmental Health. Vol 206, pp. 1 - 13, (2003)
Satya V. Chandra, Girja S. Shukla and R.S. Srivastava, "Manganese Exposure to Welders" Clinical Toxicology, Vol 18, No. 4, 1981 (See above table)
Rasmussen and Jenson (1987): Two welders with advanced stages of Mn poisoning with symptoms of Parkinsonism. “The Organic Psychosynsdrome in Manual Metal Arc Welders: A Possible Consequence of Manganese Toxicity.” Ugeskr Laer 149:3497 - 3498.
Racette, B.A., Tabbal, S. D.,
[iii] “Welding Handbook, Volume 2” Eighth Edition, pg 46, American Welding Society, 550 N. W. LeJeune Road, Miami, FL, 1995
[iv] “Fumes and Gases in the Welding Environment” American Welding Society,
[v] If your employees perform GMAW welding on some grades of aluminum, particularly 3000 series alloys, they are also at risk for overexposure to manganese
[vi] There is a technical difference between “fume” and “smoke.” “Fume” is a very fine particle that is produced when metal is vaporized and cooled. During the welding process, most of the metal is heated to a liquid state and becomes molten. This is the weld pool which is manipulated by the welder. Some of the molten metal becomes so hot that it vaporizes. However, the vaporized metal cools almost instantly into the very fine particulate that scientists call “fume.” In common usage, we recognize that many people incorrectly call vapors (like the vapors that come out of the gas tank when filling up the family car) fumes. For reasons that are discussed in “Selecting an Air Cleaner for Welding Fumes” the difference between “fumes” and “vapors” is more important than just scientific vocabulary.
[vii] “Welding Handbook, Volume 4” Eighth Edition, pg 10, American Welding Society, 550 N. W. LeJeune Road, Miami, FL, 1995
[viii] Documentation of the Threshold Limit Values and Biological Exposure Indices”, Manganese, American Conference of Governmental Industrial Hygienists,
[ix] The actual number is 28,349,523.1 micrograms per ounce.
[x] The actual number is 1.30795062 cubic yards per cubic meter.
[xi] The term “hexavalent” for a valence state of 6, is analogous to “hexagon,” describing a 6-sided figure. The term “valence” is used to describe the presence of excess electrons or the deficit of adequate electrons relative to the number of protons in the nucleus of an atom.
[xii] “Reducing Exposure to Hexavalent Chromium in Welding Fumes” Susan R. Fiore, Welding Journal, American Welding Society, 550 N. W. LeJune Road, Miami, FL.
August 2006 pp 38 – 42.
[xiii] The actual Protection Factor, or measured Fit Factor when tested via Quantitative Fit Test, for HFAPR is generally better than 100. Some wearers may achieve Fit Factors of 1000 or more. However, the respirator testing environment is often a medical facility and the wearer is concentrating on getting a good fit. In actual use, the fit of the respirator may be affected by beard growth over the course of the day, facial cleanliness, and care in donning the respirator.
[xiv] . “Industrial Ventilation, A Manual of Recommended Practice for Operation and Maintenance.” American Conference of Governmental Industrial Hygienists,
[xv] ANSI Z49.1 – 2005, “Safety in Welding, Cutting and Allied Processes,” Section E5.4 pg 12. American Welding Society,
[xvi] “Industrial Ventilation, A Manual of Recommended Practice for Design, 26th Edition” pg 13-153. American Conference of Governmental Industrial Hygienists,
[xvii] ANSI Z49.1 – 2005, “Safety in Welding, Cutting and Allied Processes,” Section E5.4 pg 12. American Welding Society,
[xviii] A useful discussion and summary of the use of “fume guns” is provided in “Development of Lightweight Fume Extraction Guns” Naval Surface Warfare Center CD Code 2230-Design Integration tools, Bldg 192 Room 128,