DWM-logo.gif (6532 bytes)
Volume 6   Issue 8               September  2005

Long Awaited Research
IGMA Nears Completion on Two Important Research
Projects Involving IG Sustainability and Argon Permeability
by Margaret Webb

Manufacturing gas-filled insulating glass units (IGU) involves replacing the air with heavy gases such as argon or krypton (or mixtures of these gases) to improve the unit’s thermal performance.

The thermal performance of gas-filled IG units depends primarily on the seal integrity and the ability of the sealant to retain the gas in the IG cavity. Assuming good workmanship, it becomes necessary to ensure that the IG assembly does not lose the replacement gas by natural diffusion. Some sealants used in the manufacturing of IG units may diffuse the heavy gases more than others and may affect the long-term thermal performance of the units.

In 2004, the IGMA technical services committee agreed to initiate a research project to develop a new test protocol for argon permeability through IG units. This was developed as a five-stage research project with each phase expanding on the previous one.

Gas Permeability Study
The main objective of this research project is to provide crucial technical information on the gas permeability rate through sealant membranes and edge seal of IG units, which may provide manufacturers with proper sealant selection for their gas-filled IG units.

The original abstract for the research project consists of five stages:

1. Evaluation of the permeability of sheet materials. The objective of Phase 1 is to develop technical information on the gas permeability rate through different sealant membranes for the purpose of using the data as an aid in predicting the gas loss rate from sealed IG units. The intent is to produce a data library of transmission rates for all available sealants on the market and fill gases and mixtures. The data could be used in selecting sealants for use in gas filled IG units or in the development of mathematical models to predict the gas loss from sealed IG units.

2. Evaluation of the gas permeability of edge seal assemblies. The objective of Phase 2 is to develop a test cell configuration to allow the determination of gas permeability of a section of an IG unit edge. Using a section of the unit edge will take into account the geometry and interactions of the components that make up the sealant/spacer system and will provide a better estimate of the gas permeation rate from the actual unit.

3. Evaluation of gas permeability of sealant systems in a strained condition both physically and environmentally. The gas permeability of sealant systems will change when strained or due to weathering. Knowing the change in the gas transmission rate of the sealants due to mechanical stress and environmental stress will aid in the prediction of the gas loss in an IG unit.

4. Development of a mathematical model to predict gas loss rate from an IG unit. Using the data generated from the facsimile samples for gas transmission, the intent of this phase was to develop a mathematical model using finite element techniques to predict, with reasonable accuracy, the gas loss from a sealed IG unit.

This phase has subsequently been withdrawn since Aspen Research has developed the IG Durability Simulation Design Tool under the DOE project DE-FC26-01NT41258, “An Insulating Glass Knowledge Base.” While the tool is substantially complete, it still needs to be validated.

This is under consideration by the Insulating Glass Industry Durability Advisory Group formerly called the DOE IG Durability Task Group.

5. Development of an improved test method for measuring the gas loss rate from a sealed IG unit. The objective of this final phase is to develop a laboratory test method to determine the gas loss rate from a standard test size IG unit that offers greater speed and reproducibility than the current tests. 

A working group to provide project management was formed to develop the specific objectives for each phase of the research project, examine current test methods, develop the requests for proposals for each phase and provide over-sight on the bidding process and review interim data results.

Current Status
Regarding phase one, the thermal performance of IG units is of primary interest to the fenestration industry. Gas diffusion through the edge seal assemblies is one of the factors influencing the long-term durability of fenestration products.

Unlike other fenestration industry organizations, IGMA’s sole focus is on IG units. However in recognition of the importance of this issue to the fenestration industry, IGMA invited interested industry partners to co-sponsor this multi-phase, multi-year research project. 

Industry co-partners can participate in the development of the project, share in the results, receive joint recognition in the publication of the results and have access to the data. Co-sponsorship consists of industry support and recognition of the project, some funding support for each phase of the project and participation on the technical services committee. Currently there are three organizations that participate on the IGMA technical services committee, the American Architectural Manufacturers Association (AAMA), the Canadian Window and Door Manufacturers Association (CWDMA) and the Window and Door Manufacturers Association (WDMA). In addition, the National Fenestration Rating Council, along with AAMA, participate as major sponsors of Phase 1 of this project and CWDMA and WDMA participate as minor sponsors. 

Phase 1 Procedure:
In this phase of the project, 184 single slabs with 1.5 and 3.0 mm in thickness are tested at three different temperatures; -29°C (-20°F), 20°C (70°F), 60°C (140°F) to correspond to the ASTM E 2190 standard. Sealants undergoing testing are:

• Silicone (one and two part);
• Polyurethane;
• Polysulfide;
• Hot melt (butyl sealant, butyl spacer);
• Reactive hot melt (silane functional, urethane functional, butyl functional);
• Polyisobutylene;
• PIB/polysulfide composite; and
• PIB/silicone composite.

At this year’s IGMA annual conference, Andre Piers from TNO provided a status report on the co-sponsored IGMA gas permeability research project. Presentations on the project’s interim results were also made at the WDMA Technical Conference and the AAMA national summer meetings. 

The samples have been tested to the following protocol:

1. The slab is placed in the unit on and over a filterpaper (acting like a spacer).
2. This is clambed in the cell (sping supported sintered = permeabel plate) and closed by primarily the own sealant material and secondary by a O-ring (greased).
3. The volume of the cell collecting the argon is known (collecting cell).
4. Collecting cell is flushed with helium (checked by sampling of the Micro-GC).
5. Argon cell is flushed continuously with argon (small overpressure).
6. After the air content is below detection limit the collecting cell is then closed.
7. After a certain collecting time, the built-in sample-loop of the Micro-GC records a representative sample (e.g. 10 ul).
8. This 10 ul is then injected to the column (molesieve 5A) for analysis and the results are recorded.

Testing has now been completed on all slab samples. The full report was presented and reviewed at the IGMA Technical Services Committee meeting in August 2005.

IGMA Field Correlation Study
This is the continuation and final milestone of the original SIGMA/HUD co-funded study that was undertaken in 1980 to determine the correlation of field failures to ASTM E 773 Class C vs. Class CB vs. Class CBA.

The original parameters encompassed 140 buildings in 40 cities containing a total of 40,000 IGUs of which two-thirds were field-glazed commercial units and the remaining one third were residential window units—2,400 IG units were selected from the total sample. These original IGUs were aluminum or anodized aluminum spacer with single seal polysulfide and hot-met butyl and dual-seal polyisobutylene with a polysulfide, silicone or hot melt butyl secondary sealant.

Over the course of the first ten years, the units were visually inspected eight times. Almost all 2,400 units faced south or southwest. 

New technology profiles were added at the ten year milestone, which consisted of polyurethane and Swiggle® sealants, warm-edge technology spacer designs, bent corners, low-E glass, internal grid systems and argon gas fill. This study focused on the CBA units only.

In general, the results from the ten and 15-year data mark showed little variation due to climatic conditions. There were no failures in 80 percent of the buildings and 60 percent of failures were accelerated by glazing systems that held water at or near the edge sealant.

Now at the 25 and 15-year mark, to date 1,150 units in total have been inspected in Atlanta; Phoenix; Cleveland, Minneapolis and Tampa. The CBA units inspected to date have a 4.8-percent failure rate at the 25-year mark and the newer technologies added at year 10 have a 1.3 percent failure rate at the 15-year mark. This study will be completed by the end of 2005, and a full report will be presented during the IGMA annual conference next year.

Generalizations Drawn from the Field Correlation Study
Performance varied little from hot to cold climates; wet to dry climates or sea level to mountain exposures, and 80 percent of the buildings had no failures. Of the failures recorded at least 60 percent of these were accelerated by glazing systems that held water at or near the edge sealant.

Two buildings in Dallas had 32 of the 69 Class C failures. These buildings used structural gaskets that were not weeped. The same unit manufacturer had 60 units without failures in three other buildings in Dallas. A complex of small buildings in one site in Santa Rosa, Calif., had 14 of the 26 Class CBA failures. There were no sill weeps causing water to be collected and held at or near the edge seal on the bottom of the unit.

The results demonstrated clearly that Class CBA units had significantly fewer failures than Class C and Class CB. The study was one of the primary sources for eliminating the CBA testing levels in the development of the ASTM E 2188, E 2189 and E 2190 harmonized standard between Canada and the United States.

Unit Longevity Conclusions on CBA Certified IG Units
Class CBA had a south-facing failure rate of 26 of 1,049 units or 2.5 percent in ten years. The failure rate after 15 years was 4.1 percent. Removing the complex in Santa Rosa, which had 14 failures in 44 CBA certified units clearly attributable to improper glazing, the Class CBA south-facing failure rate becomes 12 of 1,005 units, or 1.2 percent in ten years. Eliminating this same complex, the failure rate becomes 2.9 percent after 15 years.

IGMA Recommended Key Elements to Long-Term Unit Performance:
A. Unit construction should pass CBA level in the ASTM E 773/E 774 program. All IGMA members were required to certify to the CBA level and now to the ASTM E 2190 standard, which only tests to this level.

B. Certified units should be glazed with systems that keep water ingress at a minimum and weep water that gets in the glazing system away from the edge sealant. IGMA unit manufacturers should follow the glazing recommendations in the IGMA technical publication, TM-3000-90(04), North American Glazing Guidelines For Sealed Insulating Glass Units For Commercial and Residential Use.

C. IGMA unit manufacturers should reference IGMA TM-3000-90(04), North American Glazing Guidelines For Sealed Insulating Glass Units For Commercial and Residential Use to their glazing contractors.

D. IGMA unit manufacturers are encouraged to use the IGMA Sealant Manufacturers Minimum Sealant Dimensions and Place-ment Survey (TB-1201-89(04) document to maintain edge-seal quality and ASTM E 2190 test performance.

E. IGMA unit manufacturers should establish and maintain an in-house Quality Control - Quality Assurance Program equivalent to IGMA TM-4000-02, Insulating Glass Manufacturing Quality Procedure Manual.

Copies of the above publications can be ordered online at www.igmaonline.org/publications.
Once complete, IGMA will be preparing a technical bulletin on the complete study and plans are currently underway to initiate a new field correlation study on IG units certified to the ASTM E 2190 standard. Detailed information on this study at the 10 and 15 year mark are available from IGMA.

Further information on IGMA research programs or other activities can be obtained by contacting the IGMA office or through the IGMA website at www.igmaonline.org

The results compiled under this study were verified and validated by statistician Dr. Joseph P. Schwieterman, associate professor of Public Services Management at DePaul University. Dr. Schwieterman is also director of the school’s Chaddick Institute for Metropolitan Development in Chicago, which studies land-use, transportation and infrastructure. He has a Ph.D. in public policy studies from the University of Chicago and is a specialist in research methods, economic development and urban analysis. 

Margaret Webb is the executive director for the Insulating Glass Manufacturers Alliance. She may be reached at mwebb@igmaonline.org.


DWM
© Copyright Key Communications Inc. All rights reserved. No reproduction of any type without expressed written permission.