Full-scale two-story steel frame building under near-field explosions

Abstract

Explosions may have severe consequences on the integrity of structural or non-structural elements of a building. Being considered events with a low probability of occurrence, they are not considered directly in the design, except in certain special situations (accidental design situations). For far-field explosions, blast parameters and effects can be measured and modeled with relatively good precision. On the other hand, near-field explosions are more complex and more difficult to predict (intensity and distribution of the pressure, pressure-structure interaction, effects on materials). In addition, there is a limited amount of experimental data for near-field explosions, which makes the validation of the calculations (analytical, numerical) even more difficult. In this study, a full-scale steel frame building was subjected to a series of near-field explosions until a complete column removal. However, because of the limited amount of gravity loads, no progressive collapse was initiated. A numerical model was validated by comparing both the local strains in the most affected steel elements and the global deflections of the structure. Numerical modeling was done with Extreme Loading for Structures, ELS.

Full text

PROTECT 2019
Whistler, BC, Canada
September 16-17, 2019
I. Marginean, PhD.
Str. Ioan Curea nr.1, Timișoara, Romania, 300223
Email: ioan.marginean@upt.ro
Tel: +4 0256-403925
Full-scale two-story steel frame building under near-field
explosions
F. Dinu1,2, I. Marginean1, D. Dubina1,2, A. Kovacs3, E. Ghicioi3, R. Laszlo3, A. Khalil4, E. De Iuliis4
1 CMMC Department, Politehnica University Timisoara, Romania
2 CCTFA Research Centre, Romanian Academy, Timisoara Branch, Romania
2 National Institute for Research and Development in Mine Safety and Protection to Explosion, Petroșani, Romania
3 Applied Science International, LLC, Durham, NC 27704, USA
ABSTRACT
Explosions may have severe consequences on the integrity of structural or non-structural
elements of a building. Being considered events with a low probability of occurrence, they are
not considered directly in the design, except in certain special situations (accidental design
situations). For far-field explosions, blast parameters and effects can be measured and modeled
with relatively good precision. On the other hand, near-field explosions are more complex and
more difficult to predict (intensity and distribution of the pressure, pressure-structure interaction,
effects on materials). In addition, there is a limited amount of experimental data for near-field
explosions, which makes the validation of the calculations (analytical, numerical) even more
difficult. In this study, a full-scale steel frame building was subjected to a series of near-field
explosions until a complete column removal. However, because of the limited amount of gravity
loads, no progressive collapse was initiated. A numerical model was validated by comparing
both the local strains in the most affected steel elements and the global deflections of the
structure. Numerical modeling was done with Extreme Loading for Structures, ELS.
Keywords: robustness, progressive collapse, blast wave, overpressure, moment resisting frame.

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
1. INTRODUCTION
Explosions may have severe consequences on the integrity of structural or non-structural
elements of a building. The detonation of explosive devices in the proximity of the structural
elements can cause major damage, which in turn can propagate and may cause progressive
collapse. Progressive collapse may cause more injuries (or fatalities) than blast overpressure and
flying debris. However, progressive collapse can be prevented first by reducing exposure
(increasing safety distance), and second by increasing robustness. The ability of a structure to
resist such extreme actions without being affected at a disproportionately high level is also
required by the design rules in force (e.g. EN 1991-1-7[1]). Awareness of these risks requires
appropriate measures in the design and execution of building structural systems ([2],[3]). Direct
evaluation of the effects of an explosion is more complex than for other types of actions. There is
complexity both in terms of action modeling (e.g. maximum pressure intensity and distribution
over elements) and in terms of effects on the materials or elements (e.g. the effect of high
loading rate on the mechanical characteristics or explosion-structure interaction). The complexity
of the behavior gets even higher as the stand-off distance reduces and the explosion occurs in the
proximity of the structural elements, i.e. near-field explosion. In such situations, numerical
models calibrated against experimental data can be used to obtain more accurate results
([4][5][6][7][8][9]). A more convenient approach is the Alternate load path method (APM),
where for simplicity it is assumed that one column is lost (for example due to explosion) then the
capacity for carrying the redistributed loads is checked ([10][11][12][13][14][15]. However, it is
not yet well established if APM is representative of all types of explosive threats ([16][17]).
The paper presents the results of recent research carried out in the FRAMEBLAST project
(2017-2018) on the safety of building structures under extreme actions. A two-bay, two-span,
and two-story steel frame building was tested for different blast loading conditions to evaluate
the consequences of near field explosions on the structural elements. The experimental data were
combined with the numerical modeling to investigate the residual capacity of steel columns and
the potential for progressive collapse resulting from such extreme loading. Numerical modeling
was done with Extreme Loading for Structures ELS [18].
2. Description of the experimental model and blast testing
The steel frame building has two bays, two spans, and two stories (Figure 1.a). The bays and
spans measure 4.5 m and 3.0 m, respectively, while stories are 2.5 m high each. The structural
system is made of moment resisting frames on the x-direction (transversal direction), while on
the y-direction (longitudinal direction) concentrically braces are introduced in each frame. The
secondary beams are spaced at 1.5 m intervals. The extended end-plate bolted beam-to-column
connections at the moment resisting frames are designed as fully rigid and fully restrained
connections using M24 gr.10.9 bolts on a 16 mm thick end plate. Secondary beam-to-column
connections and secondary beam-to-main beam connections are pinned. The column bases are
welded to steel plates bolted to reinforced concrete girders, that constitute the foundations of the
structure. These connections are fully rigid and restrained. The design of the structure was done

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
considering the seismic design condition, combining the permanent actions (dead load D = 5
kN/m2), the variable actions (live load L = 4 kN/m2) and the seismic action (low seismicity,
horizontal acceleration = 0.10 g). Horizontal and vertical tying requirements for accidental
design situation were also verified using EN 1991-1-7 [1] provisions. The design resulted in
HEB260 section for columns, IPE270 section for main beams, IPE200 section for secondary
beams between columns, and IPE180 section for intermediate secondary beams. Note that
structural steel in beams, columns, and plates is S275 (yield strength of 275 N/mm2) and bolts
are class 10.9 (ultimate strength of 1000 N/mm2). Eight sensors have been used for pressure
measurements at four different locations near the structure (see also Figure 1.b):
- 1st location: 2.5 m from the middle perimeter column C2, and collinear with the explosive
charge;
- 2nd location: in front of the corner column C1 and in line with the explosive charge;
- 3rd location: 4.5 m away from the 2nd location and in line with the explosive charge;
- 4th location: 4.5 m away from the 3rd location and in line with the explosive charge.
At each location, a pair of two sensors, one normal and one aligned to the front frame, were
installed. 27 strain gauges were arranged on the structural elements to measure the history of
strains in the elements, i.e. columns (webs, flanges), beams (webs, flanges) and the end plates of
the beam-column joint (Figure 1.c). The strain gauges were used to measure specific
deformations in elements, and to determine the distribution of the stresses in front frame beams.
A total station was also used to measure global deflections in 20 different locations. 14 tracking
marks were tagged on the front frame (R1 to R14), and six on the left side frame (L1 to L6). Two
high-speed cameras were used to record and analyze the blasting events (see Figure 5). Before
testing, gravity loads with an equivalent load of 7.5 kN / m2 were placed on the floors. Note that
the loads were added only on the first bay. During the loading process, strains and deflections
were measured in the points indicated in the previous section. With the structure loaded, eight
blast tests were performed on the structure. The details are given in Table 1.
Table 1. Blast testing, with mass and position for each charge
Test name
Charge mass [g]
Distance, D [mm]
Height, H [mm]
E1
286
500
1750
E2
572
500
1750
E3
1144
500
1750
E4
2288
500
1750
E5
2288
200
1750
E6
2574
200
1750
E7
2574
200
1100
E8
*
*
*
Note:
- Distance D is measured from the front face of the central perimeter column C2
- Height H is measured from the column base plate
- Details about E8 are not provided for security reason

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
a) b)
c)
Figure 1. Overview of the steel frame structure: a) 3D geometry; b) photo before testing with the position of the
pressure sensors; c) detailed views with the strain gauges
3. Experimental results
Tests E1 and E2 (explosive charges less than 560 g) did not produce any plastic deformations in
the steel members (column, beams). However, following the E3 test, deformations remained at
the level of the column web at the level of the explosive charge and also in the column’s flanges,
see Figure 2.
a) b) c)
Figure 2. Central perimeter (view from inside) for test E3: a) view with the position of the strain gauges; b) history
of strains at locations T1-T4 (left flange); c) history of strains at locations T13-T15 (web)
Sensor 1
Sensor
3
Sensor
4
Sensor
2
Explosive charge
-0.002
-0.001
0
0.001
0.002
0.003
0.004
0 0.01 0.02 0.03 0.04 0.05
strain,
Time, s
T13
T15
-0.002
0
0.002
0.004
0.006
0.008
0.01
0.012
0 0.01 0.02 0.03 0.04 0.05
strain
Time, s
T1
T2
T3
T4

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
Figure 3 shows the plastic deformations and fracture of the web produced in the column
following tests E4, E5, and E6. Figure 4 plots the variation of the overpressure in the shock wave
and the maximum values recorded at each position of the pressure sensors for the test E6. The
difference between the maximum overpressure at locations 1 and 2 is 61% (center column vs
corner column), while the difference between sensors 3 and 4 is much less (22%).
a) b) c)
Figure 3. Views of central column after tests E4 (a), E5 (b) and E6 (c)
Figure 4. Pressure during test E6: details of the pressure ramp and maximum values highlighted at each location
After the test E8, the central column C2 was completely removed, see Figure 5. The adjacent
beams from the 1st and 2nd floors (C1-2, C2-3) also underwent residual deformations relative to
the initial stage (before the loss of the column), see Figure 6.
Figure 7 presents comparatively the wave propagation around the structure for tests E6 and E8. It
can be noticed that in case of the test E8, the incident shock is largely reflected at the contact
with the column, which also leads to an increase in the pressures recorded at pressure sensor
locations 1 and 2. For E6, the clear space to the column allows the propagation of the incident
wave inside the building and thus the reflected pressure is reduced.
1631.2012
630.2429
270.0078 210.7333
-500
0
500
1000
1500
2000
-2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Preassure, mbar
Time, ms
sensor 1
sensor 2
sensor 3
sensor 4

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
Figure 5. Central column failure mode after E8 test
Figure 6. Vertical displacements of perimeter beams from the first floor (C1-2 and C2-3) after experimental tests
a)
b)
Figure 7. Shock wave propagation: a) test E6; b) test E8
-20.0
-15.0
-10.0
-5.0
0.0
0 2.25 4.5 6.75 9
vertical displacement, mm
position on the beam span, m
initial gravity loading
E3
E4
E5
E6
E8

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
4. Numerical modelling of near-field explosion
4.1. Model calibration
The numerical analyses were performed using Extreme Loading for Structures (ELS) software,
which employs a nonlinear solver based on the applied element method [19]. The experimental
data obtained from the blast tests were used to calibrate the numerical model, see Figure 8. The
3D geometrical model of the specimen was constructed as an assembly of small (discrete)
elements, connected by springs which are generated at contact points distributed around the
element’s mutual surfaces, paired as one normal and respectively two perpendicular shear
springs. These springs can be removed when strain values reach the separation strain or can be
generated when contact occurs between elements, thus resulting in the modeling element
separation and collision. The material models were defined by their main characteristics, i.e.
elastic properties, yield strength, ultimate strength, maximum allowable elongation, and
separation elongation. Structural steel S275 was assigned for all steel elements (beams, columns,
plates) and class 10.9 bolts were used for connections.
Figure 8. Geometry of 3D numerical model and the representation of the blast in ELS
The experimental data obtained in the testing program were used to calibrate the numerical
model. Based on the pressure values measured at the four points (see Figure 1.b), pressure curves
were obtained for each explosive charge. Figure 9 shows the curve pressure vs. distance for an
explosive charge of 2574 g. In Figure 10 is shown the distribution of the peak pressures on the
central column in the area adjacent to the explosion for a load of 2574 g TNT positioned at 0.2 m
(test E6).
Figure 11 shows the residual vertical displacements in the two perimeter beams from the first
floor (C1-2, C2-3) for the test E8, which was selected because it was the only test leading to
complete column removal. It can be noticed that the values obtained in the numerical analysis are
very close to the experimental ones.

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
Figure 12 shows the strain vs time at two locations on the web of the central column C2, i.e. T13
and T15. At location T15 (top side of the column), both maximum and residual strains from
numerical analysis are lower than the experimental ones. These differences are most probably
attributed to the underestimation of the pressure at the point of interest. At location T13 (bottom
side of the column), the experimental and numerical values are much closer.
Figure 9. Pressure vs distance for an explosive charge of 2574 g TNT
Figure 10. Pressure distribution in the areas adjacent to the detonation point, test E6
Figure 11. Vertical displacements in the first story perimeter beam, test E8, experimental vs. numerical results
-25.0
-20.0
-15.0
-10.0
-5.0
0.0
0 2.25 4.5 6.75 9
vertical displacement, mm
position on the beam span, m
E8

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
Figure 12. Strain history in the column web from test E6: a) location of strain gauges T13 and T15; b) experimental
vs. numerical strain curve
Fig. 20 shows the evolution of normal strains in the column C2 during the test E6. It may be seen
that the strains are larger on the edges of the web (close to the fillet between web and flange)
where the plate is stiffer. Also, at same distance from the point of detonation, the strains are
larger at the top side of the column, also due to larger stiffness of the plate at the contact with the
continuity plate of the beam-to-column joint.
Figure 13. Evolution in time of normal strains in the web column C2, test E6
Figure 14 shows the evolution in time of the initiation and propagation of the rupture in the
central column C2 and the state of damage after the test E6 (2574 g TNT at 0.2 m from the
column web) both in the test and in the numerical simulation. It can be seen that numerical
analysis shows compliance with the experiment results and therefore can provide further details
on the development of failure and the residual capacity of the element, respectively of the
structure.
-2000
0
2000
4000
6000
8000
10000
0 0.0005 0.001 0.0015 0.002 0.0025 0.003
strain, μm/m
Time, s
T15 num
T15 exp
T13 num
T13 exp
T15
T13

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
a) b)
Figure 14. Comparison between state of damage in the column (fracture and separation of the web), numerical vs
experimental, test E6: a) initiation and development of fracture in the column web, numerical;
b) view after experimental test
5. Conclusions
Explosions may have severe consequences on the integrity of structural or non-structural
elements of a building and implicitly to the safety of the occupants. While for far-field
explosions, blast parameters and effects can be measured and modeled with relatively good
precision, near-field explosions are more complex and more difficult to predict (intensity and
distribution of the pressure, pressure-structure interaction, effects on materials).
For this reason, a two-bay, two-span, and two-story steel frame building model was tested for
different blast loading conditions to evaluate the consequences of near-field explosions on the
structural elements. The results of the blast tests showed that the interaction between the shock
wave and the structure may result in a significant increase in the maximum pressure and
implicitly in the level of deformations in the structure. Also, increasing the safety distance is the
most effective measure of damage reduction in the structure.
The specific instrumentation (pressure, strains, video) provided extensive data that allowed to
calibrate the numerical models and to go deeper into the blast-structure interaction process and
sequences of failure. The data will be further used for the design of new blast tests where the
gravity loads will be increased, and the structure tested again after the damaged elements are
replaced by new ones. This will allow to develop a full view of the direct and indirect effects of a

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
near field explosion and to provide the measures to increase the robustness of frame building
structures in case of extreme loading conditions.
ACKNOWLEDGMENTS
This work was supported by Ministry of Research and Innovation of Romania, CNCS-
UEFISCDI, application number MC 936, within PNCDI III.
REFERENCES
[1] CEN. (2006). Eurocode 1: Actions on structures - Part 1-7: Accidental actions In (Vol. EN
1991-1-7). Brussels: European Committee for Standardisation.
[2] ASCE. (2010). Minimum Design Loads for Buildings and Other Structures. In (Vol.
ASCE/SEI 7-10).
[3] DoD. (2016). Unified facilities criteria: Design of buildings to resist progressive collapse.
In (Vol. UFC 4-023-03-with Change 1). Washington (DC), US: United States Department
of Defense.
[4] Magallanes, J. M., Martinez, R., & Koenig, J. W. (2006). Experimental results of the AISC
full-scale column blast test. Rep. TR-06, 20.
[5] Mazurkiewicz, L., Malachowski, J., & Baranowski, P. (2015). Blast loading influence on
load carrying capacity of I-column. Engineering Structures, 104, 107-115.
doi:10.1016/j.engstruct.2015.09.025.
[6] Krishnappa, N., Bruneau, M., & Warn, G. P. (2014). Weak-axis behavior of wide flange
columns subjected to blast. Journal of Structural Engineering, 140(5), 04013108.
[7] Zhang, F., Wu, C., Zhao, X.-L., Xiang, H., Li, Z.-X., Fang, Q., . . . Packer, J. A. (2016).
Experimental study of CFDST columns infilled with UHPC under close-range blast
loading. International Journal of Impact Engineering, 93, 184-195.
doi:http://dx.doi.org/10.1016/j.ijimpeng.2016.01.011.
[8] Li, M., Zong, Z., Liu, L., & Lou, F. (2018). Experimental and numerical study on damage
mechanism of CFDST bridge columns subjected to contact explosion. Engineering
Structures, 159, 265-276. doi:https://doi.org/10.1016/j.engstruct.2018.01.006.
[9] Dinu, F., Marginean, I., Sigauan, A., Kovacs, A., Ghicioi, E., & Vasilescu, D. (2016).
Effects of close range blasts on steel frames. Experimental testing and numerical
validation. Paper presented at the The International Colloquium on Stability and Ductility
of Steel Structures.
[10] Astaneh-Asl, A., Madsen, E. A., Noble, C., Jung, R., McCallen, D. B., Hoehler, M. S., . . .
Hwa, R. (2001). Use of catenary cables to prevent progressive collapse of buildings.
Report No.: UCB/CEE-STEEL-2001/02.
[11] Sadek, F., El-Tawil, S., & Lew, H. (2008). Robustness of composite floor systems with
shear connections: Modeling, simulation, and evaluation. Journal of Structural
Engineering, 134(11), 1717-1725.

PROTECT 2019: 7th International Colloquium on Performance, Protection and Strengthening of Structures Under
Extreme Loading and Events, Whistler, BC, Canada, September 16-17, 2019
[12] Alashker, Y., El-Tawil, S., & Sadek, F. (2010). Progressive Collapse Resistance of Steel-
Concrete Composite Floors. Journal of Structural Engineering, 136(10), 1187-1196.
doi:10.1061/(ASCE)ST.1943-541X.0000230.
[13] Yang, B., & Tan, K. H. (2013). Experimental tests of different types of bolted steel beam–
column joints under a central-column-removal scenario. Engineering Structures, 54, 112-
130. doi:10.1016/j.engstruct.2013.03.037.
[14] Dinu, F., Marginean, I., Dubina, D., & Petran, I. (2016). Experimental testing and
numerical analysis of 3D steel frame system under column loss. Engineering Structures,
113, 59-70. doi:10.1016/j.engstruct.2016.01.022.
[15] Dinu, F., Marginean, I., & Dubina, D. (2017). Experimental testing and numerical
modelling of steel moment-frame connections under column loss. Engineering Structures,
151, 861-878. doi:https://doi.org/10.1016/j.engstruct.2017.08.068.
[16] McConnell, J. R., & Brown, H. J. E. S. (2011). Evaluation of progressive collapse alternate
load path analyses in designing for blast resistance of steel columns. 33(10), 2899-2909.
[17] Dinu, F., Marginean, I., Dubina, D., Khalil, A., & De Iuliis, E. (2018). Factors affecting
the response of steel columns to close-in detonations. Paper presented at the Proceedings of
the 12th International Conference on Advances in Steel-Concrete Composite Structures.
ASCCS 2018.
[18] ELS. (2017). Extreme loading for structures (Version 5.0). Durham, NC: Applied Science
International.
[19] Tagel-Din, H., & Meguro, K. (2000). Applied Element Method for simulation of nonlinear
materials: theory and application for RC structures. Structural Eng./Earthquake Eng.,
JSCE, 17(2).