Airblast and Wind-load Safety Films: A Case Study in All-Hazard Building Codes

ASC Proceedings of the 41st Annual Conference
University of Cincinnati - Cincinnati, Ohio
April 6 - 9, 2005


Airblast and Wind-load Safety Films: A Case Study in All-Hazard Building Codes

Kevin Grosskopf
University of Florida
Gainesville, FL

As the most vulnerable component in the building envelope, fragmentation failure of glazing systems are among the leading causes of injury and death during terrorist and storm events. As a result, safety films have gained popularity as a cost-effective alternative to high-strength or laminated glass, especially for building retrofits. Various combinations of wet glazed films and mechanical attachment systems, including those satisfying ASTM F1642 for glazing systems subject to airblast loadings, have been shown to meet windload and debris impact standards, including ASTM E1996 standards adopted by the International Building Code (IBC). Defining these and other synergies that may exist between human-caused hazards and more traditional natural hazards such as wind and seismic loading may be key in introducing anti-terrorism provisions to the commercial building industry. Such an “all-hazards” approach may compliment efforts underway to introduce meaningful terrorism resistant building standards into the IBC and minimize potential code duplication.
Key Words: Airblast loading, ASTM E1996, ASTM F1233, ASTM F1642, cyclic wind loading, glazing systems, impact resistance, International Building Code (IBC), uniform wind loading


Introduction

The events of 911 showed that counter-terrorism efforts to intervene and disrupt terrorist activities could not absolve the threat of terrorism alone, nor could existing emergency management resources be relied upon to respond and recover from incidents involving weapons of mass destruction, disruption and effect. As a result, security planners have embraced anti-terrorism measures to create a human environment that is difficult to attack, resilient to the consequences of such incidents, and protective of its populations and assets. The International Code however, used in 44 states, has yet to adopt antiterrorism standards (ICC, 2004).

In 1992, Hurricane Andrew caused more than 30 billion USD in property damage, ranking it as the most costly natural disaster in U.S. history. Unlike 911, the building and surety industries were quick to respond with new standards, codes and policy incentives for improving structural performance and survivability. Although a record hurricane season in 2004 left more than 170 dead and 7.9 million without power, combined damage totals from four major hurricanes were less than 20 billion USD (DOE, 2004).

The following paper presents research into airblast and wind load safety films to highlight the synergies that exist between human-caused and natural hazard test standards and the potential for developing all-hazard building codes.


Literature Review

In spite of the disproportionate attention given to chemical and biological threats, terrorist cells are most likely to use explosives for their ease of low-profile manufacture and delivery. Detonation of high order explosive produces a shock in air, which takes the form of a rapidly expanding pressure wave. The blast wave expands outward until its energy is diffused by nearby objects and the surrounding atmosphere. The expanding wave is referred to as the positive phase or reflected pressure load. A negative phase effect is experienced following propagation of the initial shock wave, when the atmosphere collapses into the vacuum created by the positive phase. Objects proximate to the blast are subjected to rapid rate loading from the outward movement of the pressure wave and an immediate suction effect in the reverse direction. The negative phase may coincide with the elastic rebound properties of the object, increasing the net loading effect on the object. The force of the blast wave in both phases is relative to the proximity of the object from the detonation. The basic formula used for determining the positive phase load is given below with corresponding structural effects (Table 1).

D = KW^1/3

where;

D is the distance in feet (0.305m) from the point of detonation
W is the weight of the explosive in pounds (0.454kg) of TNT
K is a numerical conversion factor, which relates the pressure in pounds per square inch (6.895kPa) to the distance (Table 1)

Table 1. K-factors and corresponding blast effects (Barstow, 1997).
K Factor	Blast Overpressure, psi		Object	Damage
45.0	1.0	(6.9 kPa)
30.0	2.0	(13.8 kPa)	Glazing system	Shattering, fragmentation of glass
20.0	3.0	(20.7 kPa)	Pre-engineered building	Buckling of pre-engineered building skins
18.0	4.0	(27.6 kPa)	Wood frame building	Studs and sheathing cracked
15.0	5.0	(34.5 kPa)		Collapse
10.0	10.0	(69.0 kPa)	CMU in-fill wall	Collapse
7.0	20.0	(137.9 kPa)	Reinforce concrete wall	Reinforcement steel exposed
5.0	40.0	(275.8 kPa)	Steel frame structures	Collapse
4.0	60.0	(413.7 kPa)
3.0	200.0	(1,379.0 kPa)


Winds produced by weather systems such as tornadoes and hurricanes, cause unique loading conditions on building structures. Asymmetric building geometries cause air to move over and around the building envelope at different velocities causing pressure gradients. Air striking a building at a given speed is forced to move faster over and around aspects of the building having greater surface area. Fast moving air produces a low-pressure gradient with respect to the slower moving air, causing outward pressure, up-lift or suction on some orientations while simultaneously causing inward pressure on others. Projectiles penetrating the building envelope through vulnerable openings such as windows, doors and unreinforced masonry, cause rapid pressurization changes within the building. The synergistic effect of multiple dynamic loading conditions can result in failure of the building envelope well below its code rated wind speed. Figure 1 shows wind damage of structures (left and center) as the result of glazing systems failure during Hurricane Ivan, September 2004. An adjacent structure (right) having protected exterior openings with sheathing, shows only minor roofing damage. Figure 2 shows progressive failure of duplex units following common failure of coastal facing fenestration.



Figure 1.
Protected and unprotected glazing systems during Hurricane Ivan, Pensacola Beach, 2004 (Escambia County Sheriff’s Office, 2004).



Figure 2.
Common glazing systems failure during Hurricane Ivan, Pensacola Beach, 2004 (Escambia County Sheriff’s Office, 2004).


Test Methods and Results

Window safety film can be used to increase the failure strength of existing commercial glazing assemblies and reduce glass fragmentation. Combined with appropriate structural silicones for wet glazing or the use of mechanical perimeter anchoring, safety films may greatly improve the survivability of building fenestration during storm or blast events. Safety films can be laminated to the interior surface of the glazing system and mechanically attached to the frame to offer maximum levels of protection. Unlike non-safety film applications, safety films have proven most effective when attached to the frame of the window, distributing loads to framing members rather than the glazing perimeter alone. Most suitable anchoring systems incorporate smooth curves rather than sharp angles to prevent tearing of the film under rapid load. Since a combination of commercially available safety films (Table 2), adhesives and anchoring systems exist, and since retrofit does not involve the replacement of the existing glazing system, retrofit may be accomplished at cost significantly lower than replacing existing glazing with safety glass.

Table 2. Typical 8-mil security grade window film (Barker, 2002).
Physical Properties		Solar Properties
Film thickness	0.008 inch (0.203 mm)	Total solar energy
Structure	Multi-ply laminate	% Transmitted	78
Tensile strength	25,000psi (172,375 kPa)	% Reflected	10
Break strength	200psi (1,379 kPa)	% Absorbed	12
Adhesive type	Acrylic pressure sensitive	Visible light
Peel strength	6psi (47.37 kPa)	% Transmitted	84
		% Reflected	10
		“U” Factor
		Median	1.08
		Design	1.12
		% Ultraviolet transmitted	0-4.0
		Shading coefficient	0.93


Airblast Testing

The U.S. General Services Administration (GSA) has been active in the development of criteria related to glass fragment mitigation including establishment of design loads and required levels of protection since the early development of the GSA Draft Security Criteria. Based largely on ASTM F1642, GSA developed a “Standard Test Method for Glazing and Glazing Systems Subject to Airblast Loadings,” since updated in January 2003 to the GSA “Standard Test Method for Glazing and Window Systems Subject to Dynamic Overpressure Loadings”. Category C facilities, or those facilities under moderate threat levels, such as a GSA field office with <450 employees and <150,000 ft² (14,000 m2) of floor area, require window fragment protection from blast loads with a peak pressure of 4 psi (27.58 kPa). A performance condition “4” (Table 3) is permitted for Category C facilities. GSA specifications which do not comply with ASTM F1642 criteria include the number of test specimens (3 minimum) and the fail criteria. ASTM F1642 requires a failure rating for any penetration in the daylight opening. The GSA fail criteria is determined by what extent the glass is retained by the frame and how far glass fragments travel.

Table 3. GSA Performance Conditions for Window Systems Response (GSA, 2003).
Performance Condition	Description	Fragments Exterior to Structure	Fragments Interior to Structure
1	Glass not cracked, fully survived and/or fully retained by frame and no glass fragments either inside or outside structure.	None	None
2	Glass may be cracked but is retained by the frame.	Yes	No significant fragments. Dusting or very small fragments near sill or on floor acceptable.
3a	Glass failed and not fully retained by the frame.	Yes	Yes – land on floor no more than 40 inches from window.
3b	Glass failed and not fully retained by the frame.	Yes	Yes – land on floor no more than 10 ft from window.
4	Glass failed and not fully retained by the frame.	Yes	Yes – land on floor more than 10 ft from window and impact vertical surface located not more than 10 ft behind window and no higher than 2ft above floor level
5	Glass fails catastrophically.	Yes	Yes – land on floor more than 10 ft from window and impact vertical surface located not more than 10 ft behind window above a height of 2 ft.


In March 2004, ABS Consulting conducted blast testing on twenty-seven (27) windows in accordance with GSA and ASTM F-1642 test protocols. Simulated blast loads were applied using a “shock tube”, a device that generates a sudden burst of compressed air that applies a blast pulse to a test specimen attached to the end of the tube. Test specimens consisted of monolithic and insulating annealed glass with daylight openings measuring 47 inches (1.19m) x 66 inches (1.68m). Frames consisted of extruded aluminum with glass secured by gasket or structural silicone. Following each test, glass fragments were collected and weighed by zone in the test enclosure. Fragments striking and embedding in a “witness panel” of foam board positioned vertically 10 feet (3.05m) from the test specimens were collected and documented. Frame deflections and performance of frame anchorage were recorded. Performance conditions for each test were assigned in accordance with GSA criteria (Table 4).

Table 4 Sample measured blast loads test specimens (Barker, 2004).
TestNo.	Specimen		Mass of Glass by Zone (g)		Penetrations Number/Depth (mm)		GSA Performance Condition
		3A		3B	4	5
1	¼” monolithic AG, no upgrade	2,239		26,399	-	2/16	5
14	¼” monolithic AG, 8-mil safety film, mechanical attachment on 4 sides	0		0	-	-	2
16	¼” insulating AG w/ ½” AS, 8-mil safety film, mechanical attachment on 2 sides	40		5	-	-	2
18	¼” insulating AG w/ ½” AS, 8-mil safety film, wet glazed 4 sides w/ struct silicone	4,990		4,990	-	-	3B

AG = Annealed Glass, AS = Air Space



Figure 3.
Test specimen 1 (Barker, 2004).



Figure 4.
Test specimen 14 (Barker, 2004).



Figure 5..
Test specimen 16 (Barker, 2004).



Figure 6.
Test specimen 18 (Barker, 2004).



Impact Resistance and Windload Cycle Testing

Perhaps a more eminent threat than terrorist use of explosives are wind loads and airborne debris from major weather events. In March of 2001, American Test Laboratories of South Florida (ATL) conducted ASTM E1996 “Specification for Performance of Exterior Windows, Curtain Walls, Doors, and Storm Shutters Impacted by Windborne Debris in Hurricanes” on 3/16 inch (4.76 mm) tempered sliding glass doors laminated on the inside surface with an 8-mil safety film. The daylight opening on the test specimens measured 45 inches (1.14 m) wide and 91 inches (2.31 m) in length. Doors were pocket glazed on extruded aluminum using a 3/16 inch (4.76 mm) vinyl gasket with a 0.522 inch (13.26 mm) bite. The perimeter was wet glazed using a structural silicone with a 0.340 inch (8.64 mm) overlap on the glass.

In accordance with ASTM E1996, large missile impact testing consisted of projecting a #2 Southern Yellow Pine 2 inch (50.8 mm) by 4 inch (101.6 mm) cross-sectional timber, approximately 47.5 inches (1.21 m) in length and 4.5 lbs (2.04 kg) in weight, at three test specimens, A, B and C. The projectile impacted each test specimen at 40.3 ft/sec (12.28 m/sec). None of the impacts penetrated the specimens and there was no separation of the glass from the glazing systems (Fig. 7). Following impact testing, cyclic wind load simulation (cycle) tests were then conducted on the test specimens. Specimens showed no resultant failure or duress after cycle tests and no separation of glass from the aluminum frame.



Figure 7.
ASTM E1996 center impact testing of 8-mil safety film and structural silicone (Henry and Mehner, 2001).



Discussion

GSA Security Criteria apply to new construction of general purpose office buildings and to major renovations where appropriate. GSA operates more than 8,000 buildings worldwide. Of these, more than 2,000 are federally owned. Best estimates indicate that there are over 35 million square feet of fenestration in this subset of GSA buildings alone (Smith, 2003). Although GSA represents an appreciable building stock and is perhaps the largest “user” of ASTM F-series standards, these and other F-series adopters represent a small fraction of the total, potentially vulnerable, building market. In 2003, 16.1 billion USD of non-residential federal construction was put in place, or 3.7% of the 432.8 billion USD domestic market. The total value for federal office and commercial construction for which GSA Security Criteria may in part apply was 4.4 billion USD in 2003, or 4.2% of the total 103.8 billion USD domestic market (U.S. Census, 2004).

International Building Code (IBC)

In 2003, three U.S. model building codes, the National Building Code (NBC), the Uniform Building Code (UBC) and the Standard Building Code (SBC) were replaced by the 2003 International Building Code (IBC). According to the IBC sanctioning body, the International Code Council (ICC), 44 states use the I-Codes at either state or local level. Although many code jurisdictions will continue to use previously adopted versions of the NBC, UBC and SBC, the IBC will increasingly represent the national consensus building standard. Unfortunately, a complete search of all ASTM F-series standards was cross referenced to a list of IBC adopted standards, and none have yet to be adopted (Hugo, 2004). However, the ICC is in the process of forming an Ad Hoc Committee on Terrorism Resistant Buildings which will explore the means and methods available to introduce security systems and equipment into future generations of the IBC (ICC, 2004).

All Hazards Code Logic

Perhaps the greatest challenge confronting the code adoption of antiterrorism building standards is how the code will be administered. Traditional options range from an elective code to standard mandatory provisions. Since mandating antiterrorism standards will likely prove difficult, an “all-hazards” approach that integrates antiterrorism requirements into the framework of existing code provisions for traditional and natural hazards may be a viable alternative. Using the protective window safety film testing as a case study, it has been shown that product satisfaction of an E-series standard, ASTM E1996, may also satisfy an F-series standard, ASTM F1642, and possibly other security systems and equipment standards such as ASTM F1233, Standard Test Method for Security Glazing Materials and Systems. An all-hazards approach to ASTM standards development addresses the issue of code synergy and establishes a realistic strategy for possible adoption into the IBC. A similar strategy was used to implement new terrorism-related fire codes and standards for high rise buildings in the months following 911. These codes were adopted under an all-hazards approach without specifically referencing specific acts of terrorism. Integrating natural and human-caused hazards within the I-Codes will not be without challenges as many existing standards are only applicable to certain geographic regions, occupancy groups and exposure categories. However, many of the same characteristics that make buildings vulnerable to natural hazards, also make these buildings vulnerable to terrorism.


References

Barker, Darrell. (2002). Test Program for Wet Glazed Window Safety Film. Summary Report. ABS Consulting, Inc. San Antonio, August 2002.

Barker, Darrell. (2004). Window Safety Film Test Program. Final Report, ABS Consulting Inc. San Antonio, May 2004.

Barstow, Brian. (1997). JIATF East Project Blast Analysis. U.S. Navy EODB/Army TM/USAF TO 60A-1-1-4. Naval Facilities Engineering Command. Norfolk, February 1997.

Bowman, Dave. (2004). Interview. Standards Development, International Code Council. Falls Church, July 2004.

Grendon Design Agency, Ltd. (2001). Robustness Testing of Inclined Glazing Using The FrameGard Anchoring System and Madico 8mil Multi-ply Safety Film. Test Report GDA/1089. Northants, U.K., January 2001.

Hattem, Henry and William R. Mehner. (2001). ATL Report #98-0213.05. ASTM E1996 Testing. American Test Laboratory of South Florida. Miami, March 2001.

Hugo, Joseph. (2004). Interview. ASTM Committee F12 on Security Systems and Equipment, West Conshohocken, July 2004.

International Code Council. (2004). Ad Hoc Committee on Terrorism Resistant Buildings. URL http://www.iccsafe.org/cs/cc/calls.html#adhoc, Falls Church, VA, July 2004.

International Code Council. (2004). International Code Adoptions. URL http://www.iccsafe.org/government/adoption.html, Falls Church, July 2004.

Office of Energy Assurance. (2004). Hurricane Situation Report. U.S. Department of Energy, http://www.ea.doe.gov/pdfs/hurrcharley_sitrept_081504_1500.pdf. Washington D.C., October 2004.

Smith, Joseph L. (2003). Anti-Terrorism: Criteria, Tools & Technology. Applied Research Associates, Inc. Washington, D.C., February 2003.

U.S. Census Bureau. (2004). 2003 Annual Value of Construction Put in Place. Construction Spending, Manufacturing, Mining and Construction Statistics. URL http://www.census.gov/const/www/c30index.html. Washington D.C., May 2004.

U.S. General Services Administration. (2003). U.S. General Services Administration Standard Test Method for Glazing and Window Systems Subject to Dynamic Overpressure Loadings. January 2003.