United States National Risk Management
Environmental Protection Research Laboratory
Agency Research Triangle Park, NC 27711
Research and Development EPA/600/SR-96/147 February 1997
Project Summary
Active Soil Depressurization
(ASD) Demonstration in a Large
Building
Ashley D. Williamson, Bobby E. Pyle, Susan E. McDonough, and
Charles S. Fowler
The purpose of this research effort
was to develop building standards for
radon-resistant large buildings for the
Florida Radon Research Program
(FRRP) of the State of Florida. Funda-
mental and applied research studies of
various building components (floor bar-
riers, mitigation systems, ventilation
systems, fill materials) and performance
standards were conducted. Field evalu-
ation and validation of the draft stan-
dards for residential structures have
been carried out. The effectiveness of
passive barriers, the modification of the
heating, ventilation, and air condition-
ing (HVAC) system operations to ame-
liorate indoor radon concentration, and
the use of active subslab depressur-
ization (ASD) systems have been in-
vestigated in demonstration (new con-
struction) houses and research struc-
tures in Florida. Current emphasis is
on large scale buildings. This study
evaluated the feasibility of implement-
ing such radon-resistant construction
techniques (especially ASD) in new
large buildings in Florida. The tech-
niques developed in this study for
Florida large buildings focused prima-
rily on the demonstration of passive
barriers to radon entry and ASD sys-
tems as applied to large buildings. The
results of this study will enable exist-
ing subslab pressure field extension
(PFE) models to be expanded to in-
clude large slabs. The results of these
improved models will be used to de-
sign more effective ASD systems for
new buildings. In addition to an ASD
system, other radon-resistant construc-
tion techniques were designed into the
building, including the installation of
adequate subslab membranes, sealing
of all slab openings, and HVAC opera-
tion to prevent depressurization of the
building interior. Implementation and
installation of these features were moni-
tored as the building was being con-
structed. Indoor radon concentrations
and radon entry were monitored in the
finished structure with the HVAC sys-
tem on and the ASD system off, and
with the ASD systems activated in a
temporary mode. Results from this
study have demonstrated that with suf-
ficient attention to building design and
construction, significant radon entry
into a large building constructed on a
site of high radon potential can be pre-
vented. The effectiveness of the ASD
system as a radon mitigation technique
could not be realistically evaluated due
to a lack of radon in the building. How-
ever, the PFE measurements suggest
that the design is more than adequate
to meet its purpose of pressure rever-
sal between the building interior and
the subslab regions.
This Project Summary was developed
by EPA's National Risk Management
Research Laboratory's Air Pollution
Prevention and Control Division, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
The Florida Radon Research Program
(FRRP) was implemented by the State of
Florida in 1989 to provide radon research
related to the detection, control, and abate-
ment of radon in Florida buildings. The
purpose of this research effort was to de-
velop building standards for radon-resis-
-------
tant buildings. Fundamental and applied
research studies of various building com-
ponents (floor barriers, mitigation systems,
ventilation systems, fill materials) and per-
formance standards were conducted. Field
evaluation and validation of the draft stan-
dards for residential structures followed.
The effectiveness of passive barriers,
modification of the heating, ventilation, and
air-conditioning (HVAC) system operations
to ameliorate indoor radon concentration,
and the use of ASD systems have been
investigated in demonstration (new con-
struction) houses and research structures
in Florida. Current research includes large
scale buildings.
This study evaluated the feasibility of
implementing such radon-resistant con-
struction techniques (especially ASD) in a
new large building in Florida. The major
specific objectives of the demonstration
were:
• Work with designers/architects to de-
sign an ASD system for the building
that was also broadly applicable to
other similarly sized buildings in
Florida;
• Test and evaluate the ASD and bar-
rier system effectiveness in a large
building built on a high radon poten-
tial site; and
• Expand existing subslab PFE models
to include large slabs.
This work complements previous diag-
nostic and mitigation strategy development
work in the FRRP. In addition to the ASD
system, other radon-resistant construction
techniques were designed into the build-
ing, including the installation of adequate
subslab membranes, sealing all slab open-
ings, and HVAC operation to prevent de-
pressurization of the building interior.
Implementation and installation of these
features were monitored as the building
was being constructed. Indoor radon con-
centrations and radon entry were moni-
tored in the finished building with the HVAC
system on and the ASD system off, and
with the ASD systems activated in a tem-
porary mode.
Materials and Methods
Site Selection
In 1992, Southern Research Institute
(SRI) assisted in ambient radon measure-
ments at a construction site in Polk County,
FL. Polk Power Partners, L.P., were pre-
paring to construct an industrial building
on the site located on Highway 555, ap-
proximately 10 miles* southwest of Bartow,
FL, for Arc Energy and Central Southwest
Services, Inc. The structure was to be-
come the Mulberry Cogeneration Facility's
(MCF's) Plant Services Building. The pro-
posed building was to be approximately
21,000 ft2 and would consist of a metal
shell on a 6 - 8 in. floating slab. The
subslab area was to have five drain lines
encased in concrete, below the slab run-
ning the entire length of the building, ef-
fectively separating the slab into five cells.
In addition, all other subslab pipes for
power, water, and other utilities were to
be installed encased in concrete.
Six EPERM (High Sensitivity, Standard
Chamber) canisters were deployed. They
were placed at the site at ground level, at
12 ft and at 27 ft. On subsequent visits to
the site, gamma measurements were
made at ground level and 12 ft where the
EPERMs were deployed. The results of
these gamma measurements resulted in
a consistent 40 jxR/hr level at both ground
level and 12 ft. Permeability measurements
were carried out, and some radon grab
samples were taken at a depth of 3 ft.
Typical of reclaimed land, the results var-
ied from 3,000 to 10,000 pCi/L. On the
basis of these preliminary measurements,
the MCF was identified as a candidate
building for the Large Building ASD Study.
The slabs for the Control Room and the
Data Processing Unit (DPU) Room were
poured at a recessed depth of 18 in.,
relative to the rest of the slabs.
A planning meeting to discuss the site
was held at EPA's National Risk Manage-
ment Research Laboratory, Research Tri-
angle Park, NC, on January 20, 1993,
with David Sanchez and A. B. Craig of
EPA, Susan McDonough of SRI, and Tom
Pugh of Florida A&M University (FAMU).
A project conference was held at Black &
Veatch's (B&V's) offices in Kansas City,
MO, the next day. As a result of this
meeting, the owner decided to incorpo-
rate both passive sealing modifications and
an ASD system into the building.
ASD Matting Plan
The proposed facility and plant services
building drawings were obtained from the
contractors. The plans were reviewed and
probable radon entry points were identi-
fied. A proposed design of an ASD sys-
tem was developed at the initial planning
meeting attended by EPA, FAMU and SRI
researchers. Several areas of concern
were identified with respect to various
subslab and building design features. Sub-
sequent to the planning meeting, a con-
* Metric equivalents lor nonmetrlc units used in this
Summary are: 1 ml = 1.6 km, 1 ft2 = 929 cm2,1 ft = 30.5
cm, 11n. = 2.54 cm, 1 mil = 25.4 |im, 1 pCi/L = 37 Bq/
m3, and 1 In WC = 249 Pa.
ference with B&V was attended by EPA,
FAMU, and SRI. The project goals and
objectives, and the overall intent/theory of
radon-resistant construction were con-
veyed to members of the B&V design
team. Drawings of the proposed ASD sys-
tem, developed by the research team, and
existing passive radon controls, developed
by the B&V team, were discussed. The
B&V team was then left to design, price
out, identify schedule conflicts, and obtain
owner approval to incorporate the ASD
system in the final building design. Sev-
eral iterations of the University of Florida
residential PFE model were run to ensure
that the system was capable of providing
the required air flows throughout the sys-
tem. The B&V design team submitted pre-
liminary design drawings to the research
team for comments on the system design.
The comments were relayed to the B&V
team and were ultimately included in final
system design.
The current draft Florida radon stan-
dard would allow for the ASD system to
consist of ventilation matting such as Enka-
Vent spaced on 20-ft centers. The recom-
mended system layout for this building
utilizes ventilation strips on 15-ft centers.
This is a result of several factors:
• The building utilizes a 30-ft bay spac-
ing that lends itself to the 15-ft mod-
ule;
• The performance of the mat in this
configuration had not been extensively
modeled, so a slightly conservative
approach was considered prudent;
and
• The air permeability characteristics of
the soil immediately beneath the slab
were not well established.
Risers R-5 and R-7 were omitted during
construction, and the pipe connections to the
matting were closed and sealed. These
changes were necessitated by construction
changes in the building after the slabs were
poured. The building has unique features,
most notably the encasement of all underslab
plumbing and conduit in reinforced concrete.
In virtually every case, the plumbing trenches
were backfilled with concrete to the elevation
of the bottom of the slab. The trenches were
usually more than 3 ft deep. They were slightly
over excavated, and 2 - 4 in. of concrete was
cast over the exposed bottom of the trench to
create a work surface. Piping and reinforcing
were installed, and forms were set to limit the
width of the concrete to approximately 20 in.
After the concrete was cast and the forms
stripped, the remainder of the trench was
backfilled with compacted earth. This had the
effect of compartmentalizing the slab into many
discrete zones. As a result, the ASD matting
2
-------
design might appear overly conservative, un-
less consideration is given to the need for
intersecting each of these areas with some
portion of a ventilation mat. Despite these
constraints, a reasonably simple and efficient
use of ventilation mat was achieved.
ASD Installation and
Construction Observations
Personnel from SRI and FAMU made
several visits to the job site throughout
the construction period. The purpose of
these visits was to monitor, but not affect,
the construction process. However, the
construction crews and their supervisors
frequently asked advice about specific ASD
installation issues, which was then freely
given. Before inclusion of SRI or FAMU
personnel into the project, and based on
advice taken from publicly available litera-
ture, the contractor had agreed to install
6-mil polyethylene beneath the entire
subslab area, including below and along
the inside surface of the perimeter grade
beams. Although this step exceeded cur-
rent recommendations, it is not believed
to have had a significant effect on the
effectiveness of the barrier, since the con-
dition of the polyethylene rapidly wors-
ened as the beams were cast, stripped,
backfilled, etc. During the installation of
the ventilation mat, it became apparent
that large amounts of sand would be
tracked onto the mat surface by workmen,
and steps were taken to have them step
over, rather than walk on, the mat. Some
consideration should probably be given to
requiring clear polyethylene whenever an
ASD system is roughed in, so that an
inspector can look for sand contamination
and blockage of the mat. This is true for
all sizes of structures, but is especially
important on large slabs where consider-
able time is required to install the mat and
cover it with polyethylene.
Another problem that presented itself
during the slab-preparation process was
the cutting of the vapor barrier when rein-
forcing steel was dropped or dragged
across it, especially if the point of contact
was at the ventilation mat. This problem
may be specific to this structure, but will
probably occur to some degree in all build-
ings. Careful inspection and repair will be
necessary to limit this.
The slab was cast in four sections be-
ginning with the lowest level (the Control
Room and the DPU Room). This resulted
in some sections of ventilation mat being
exposed to wear and tear, sand, and waste
concrete that was dragged over the forms
or otherwise spilled. We advised the con-
tractor to "sleeve" the mat for a distance
of approximately 6 - 7 ft at these points.
This was easily accomplished by cutting a
section of polyethylene approximately 6 x
6 ft and placing it at the edge of the pour
immediately before installation of the mat.
The polyethylene was then folded over
the top of the mat and secured with tape.
While this prevents the mat from commu-
nicating directly with the soil in this small
section, the effect on the performance of
the system is negligible. This practice did
seem to effectively prevent flow along the
mat at this point. It is reasonable to as-
sume that a fully blocked mat might still
exhibit some very slight flow, since the
pressure field will propagate to some ex-
tent through the surrounding soil. This may
be a point worth investigating in detail.
The slab had one final feature unique to
the experiences of this program; several
hundred twisted copper strand grounding
cables were installed that penetrated the
slab. Since there is considerable open
space between each intersecting stand of
wire (approximately 26% of the area of
the cable itself that was generally 1.5 in.
in diameter), these were potential entry
points that were not anticipated during our
earlier consultations. However, it is not
difficult to imagine a variety of potential
solutions to this rather unusual situation.
Despite having been heavily reinforced
and cured in compliance with the current
recommendations, there was some crack-
ing visible in the slab. The effects of these
cracks on radon entry are discussed else-
where in the report.
In conclusion, it was not particularly dif-
ficult for the workmen to properly install
the ASD system or to achieve a high
degree of effectiveness with the slab seal-
ing practices if they had been through a
short training program. There was some
learning required on their part, and this
advanced most quickly through onsite
demonstration rather than discussion or
reference to sketches.
Diagnostic Measurements
Pressure Field Extension
Pressure field extension was measured
by attaching an in-line radon mitigation
fan to one of the ASD risers to produce a
pressure field in the Enka-Vent mat sys-
tem with the other risers capped and
sealed. Subslab pressures were measured
using a micromanometer attached to test
holes drilled through the slab.
Slab Crack Characterization
The flow through the slab cracks and
the radon concentration infiltrating through
these cracks were analyzed using tech-
niques developed for the FRRP.
Post-Construction Ventilation and
Radon Entry Characteristics
Ventilation and radon entry into the
building were monitored using two sys-
tems. For building ventilation, a tracer gas
technique was used. For radon measure-
ments, a Campbell Scientific, Inc. 21 XL
Micrologger with 40K RAM storage sys-
tem was used. In the 21 XL system in-
stalled in the MCF building, the following
parameters were measured:
1. Ap1 between the Electrical Equipment
Room and the DPU Room.
2. Ap between the DPU Room and out-
side the building.
3. Ap3 between the HVAC Mechanical
Room and the DPU Room.
4. Ap4 between the Reception Area and
the DPU Room.
5. Ap5 between the subslab area and
the DPU Room.
6. Temperature on the 21 XL Panel lo-
cated in the DPU Room.
7. Barometric pressure outside the build-
ing.
8. Radon levels under the slab using a
Pylon AB5 monitor operating in the
pumped mode.
9. Radon levels in the DPU Room us-
ing a Pylon AB5 with a passive ra-
don detector (PRD) cell.
10. Radon levels outside the building us-
ing a Pylon AB5 monitor operating in
the pumped mode.
11. Radon levels in the Electrical Equip-
ment Room using a Femto-Tech ion-
ization monitor.
12. Radon levels in the Control Room
using a Femto-Tech ionization moni-
tor.
13. Radon levels in the Water Treatment
Area using a Femto-Tech ionization
monitor.
14. Radon levels in the Reception Area
using a Femto-Tech ionization moni-
tor.
Data from each of these parameters
was averaged (or totalized for the radon
monitors) and stored every 30 min. This
datalogging system was installed in the
building on August 8, 1994, and was made
fully operational shortly thereafter.
Results
Site Characterizations
Core samples were taken at three loca-
tions prior to pouring the slab. In each
core, soil samples were recorded down
to a depth of about 60 in. with separate
samples every 4 - 6 in. The average soil
radium content for cores one, two, and
three were 5.9, 3.9, and 6.5 pCi/g, re-
spectively, with an average for all three
3
-------
cores of 5.44 pCi/g. The radium content
has a tendency to drop with depth, with
the higher levels within 10-20 in. Perme-
ability and soil gas radon were also mea-
sured prior to slab pouring. However, due
to problems with the permeameter, the
only reliable data were obtained in the
north end of the building. The average
soil gas radon concentration was approxi-
mately 9,905 pC'i/L with.no discernible pat-
tern under the north end of the slab.
Pressure Field Extension
Measurements
After the slab was poured, PFE was
measured in the southwest corner of the
slab. These measurements were under-
taken not only to obtain a measure of the
PFE under the slab, but also to determine
the effects that deleting riser R-4 would
have on the effective coverage of the ASD
system. The tests were carried out with
an in-line fan installed on riser R-1 and
with risers R-2, R-4, R-5, and R-6 capped.
Risers R-3 and R-7 were removed after
the slab was poured due to changes in
the building usage. The openings to the
matting were closed and sealed. The
subslab pressures averaged about -100
Pa (-0.4 in. WC) relative to atmospheric
and, as might be expected, were gener-
ally higher (more negative) in the vicinity
of the Enka-Vent matting. This subslab
pressure level should be more than ad-
equate for effective operation of the ASD
system.
Slab Crack Mapping and
Measurements
Three prominent slab cracks were ana-
lyzed on October 22, 1993. Based on
these visible cracks, a conservative esti-
mate of the total crack area is approxi-
mately 14 ft2 (1.3 m2) of a total slab area
of 21,000 ft2, a figure of less than 0.1% of
the total area. In each case the flow of
soil gas through the cracks (as well as
the concentration of radon in the gas) was
sufficiently low that it should not signifi-
cantly contribute to the indoor radon lev-
els at realistic pressure differences.
Post-Construction Ventilation
and Radon Entry
Radon Entry
The data from the continuous monitor-
ing system are plotted on a daily basis in
the Appendix of the full report. These data
plots cover the period 8/17/94 through 1/
06/95. Operation of the building can be
divided into three periods:
Period 1. From 9/08/94, when the data
logger was first installed, until 10/22/94
when the ASD system was temporarily
energized.
Period 2.'From 10/22/94, when the ASD
mitigation fan was installed and energized,
until 10/28/94.
Period 3. From the time the ASD fan
was turned off until 11/01/94 during that
time several building operation upsets were
encountered.
The radon levels for Period 1, while
fluctuating somewhat, were very low over-
all and indeed were barely above ambient
concentrations. These are shown in Table
1 where, with the exception of the subslab,
the radon levels during Period 1 averaged
less that 1 pCi/L.
During Period 2 an in-line radon fan
was connected to riser R-5 to temporarily
activate the ASD system. These results
are also shown in Table 1. No significant
reduction in the low radon concentrations
was observed; in fact, at first inspection it
would appear that the ASD operation
raised the radon levels in the occupied
areas of the building. However, closer in-
spection of the data shows that the ambi-
ent radon levels increased as well over
this time period from 0.25 pCi/L in Period
1 to 0.64 pCi/L in Period 2. The increases
inside the building followed the ambient
levels during this same time period. In
order to better understand the results of
the ASD operation, the values in Table 1
were adjusted by the changes in the am-
bient radon levels. The resulting changes
in average indoor radon, less than 0.5
pCi/L, are probably not experimentally sig-
nificant. In short, the base levels of radon
in the building appear to be so low that no
appreciable change results from activat-
ing the ASD system. In contrast, the
subslab radon levels that averaged about
13,000 pCi/L during Period 1 were
noticably decreased during and shortly af-
ter ASD activation. It is seen in Table 1
Table 1. Average Radon Levels Measured at the Mulberry Cogeneration Facility (in pCi/L)
Condition Elect. Water
of the ASD DPU Ambient Equpt. Control Treat. Office/
System Subslab Room Pumped Room Room Room Recpt.
ASD Off
9/08 to
10/22/94
13043
(13043)
0.73
(0.21)
0.52
0.70
(0.18)
1.09
(0.57)
0.80
(0.28)
0.72
(0.20)
ASD On
10/22 to
10/28/94
9772
1.14
(0.34)
0.80
1.50
(0.70)
1.60
(0.81)
1.20
(0.40)
0.96
(0.17)
ASD Off
10/28 to
11/01/94
9486
0.71
(")
1.09
0.74
(")
1.12
(0.04)
1.01
(")
0.70
n
Notes: Numbers in parenthesis are after subtracting the ambient radon background levels. (") indicates that the radon levels were below the lower
limit of the radon monitor.
4
-------
that activation of the ASD system dropped
the subslab radon levels by approximately
25%.
Except for the reception area, the occu-
pied zones were pressurized relative to
the outdoors. The time traces of the pres-
sure measurements are shown in more
detail in the Appendix of the full report.
These plots show a consistent pressuriza-
tion of most zones with significant fluctua-
tions that appear to be due to changes in
status of some building system that was
not specifically monitored, such as venti-
lation fans in the bay areas. It was ob-
served that the door from the Mechanical
Room to the outdoors was frequently ajar,
resulting in large swings in several pres-
sures at unpredictable times. The Office/
Reception areas fluctuated in pressure
relative to the outdoors, although on the
average these areas were depressurized.
Also, the pressure under the slab without
the ASD system running is quite positive.
This is most likely due to air from the
building interior being forced down under
the slab through the construction joints
and floating edge cracks.
Conclusions and
Recommendations
In view of the low radon levels inside
the building and the fact that the site has
a high radon potential, it appears that
either the radon barrier system installed
under the slab, the overall pressurization
of the HVAC systems, or some combina-
tion of the two, is very effective in pre-
venting radon entry into the building. In
either case, results from this study have
demonstrated that, with sufficient atten-
tion to building design and construction,
significant radon entry into a large build-
ing constructed on a site of high radon
potential can be prevented.
Due to the continuous occupancy of the
building, starting before completion of con-
struction, the planned passive (HVAC off)
experiments were not completed. These
experiments would have better defined the
relative effects of both the barriers and
the HVAC operation. However, since at
least part of the structure is apparently
depressurized with respect to the outdoors,
we assume that the primary mitigation
structural element was the slab/barrier in-
tegrity rather than the HVAC system. Fur-
ther measurements during an unoccupied
period without HVAC operation would have
been highly desirable.
The effectiveness of the ASD system
as a radon mitigation technique could not
be realistically evaluated due to a lack of
radon in the building. However, the PFE
measurements suggest that the design is
more than adequate to meet its purpose
of pressure reversal between the building
interior and the subslab regions.
5
-------
Ashley D. Williamson, Bobby E. Pyle, Susan E. McDonough, and Charles S. Fowler
are with Southern Research Institute, Birmingham, AL 35255.
Marc Y. Menetrez is the EPA Project Officer (see below).
The complete report, entitled "Active Soil Depressurization (ASD) Demonstration in
a Large Building," (Order No. PB97-133805; Cost: $28.00, subject to change) will
be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air Pollution Prevention and Control Division
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
EPA/600/SR-96/147
------- |