-
ARTIFICIAL GROUND FREEZING FOR EAST SIDE ACCESS NORTHERN
BOULEVARD CROSSING FROZEN SOIL ARCH QUEENS, NEW YORK
Adam Curry, P.E., Moretrench American Corporation, Rockaway, NJ,
USA
Gregory Ziegler, V.P., P.E., Moretrench American Corporation,
Rockaway, NJ, USA
Abstract: The CQ039 Northern Boulevard Crossing Contract was one
part of the larger MTA Capital
Constructions East Side Access Project connecting the Long
Island Railroad with NYCs Grand
Central Station. This contract required the construction of a
SEM tunnel between two 85-ft deep
access shafts. Difficulties included working as deep as 55 feet
below the groundwater table while
mining through variable subsurface conditions with preclusions
to dewatering and settlement. In
addition, the tunneling would occur directly beneath an active
subway line, an elevated rail line (both
supported on piles), and Northern Boulevard, a main artery
through Queens, New York. In order to
create both a groundwater cutoff and a temporary excavation
support during the tunneling
operations, the tunneling contractor selected an artificially
frozen soil arch, installed horizontally from
the two access shafts.
In addition to a discussion of the overall approach, this paper
includes a detailed description of the
drilling and surveying procedures used for the installation of
the horizontal freeze pipes. It also
summarizes the results of the various tests performed at the
site to verify the performance of the
freeze system and the data collected while monitoring the freeze
formation. The paper closes with a
synopsis of the results and an update on the project.
I INTRODUCTION
The New York City Metropolitan Transit Authority awarded the
CQ-39 contract to Schiavone Kiewit
JV to construct a new tunnel structure below Northern Boulevard
connecting the Queens open-cut
area to the existing bellmouth north of Northern Boulevard. The
CQ-39 contract was part of the
larger East Side Access Project. The goal of the project is to
connect the Long Island Railroad
(LIRR) lines in Queens to a new LIRR terminal beneath Grand
Central Terminal in Manhattan. The
main intent is to increase the LIRRs capacity and shorten travel
time for Long Island commuters
traveling to the east side of Manhattan.
The proposed tunnel utilized Sequential Excavation Method (SEM)
tunneling methods with a frozen
arch to act as temporary structural support and groundwater
cutoff. The SEM method was developed
in the 1950s for work in consolidated formations and has been
advanced and adapted to become
suitable for an array of soft ground conditions. As its name
implies, SEM construction utilizes
excavations in short incremental lengths with benches sequenced
to maintain weight against the
face and reduce the span of the crown. Initial support is
provided with a spray-on shotcrete lining,
typically reinforced with steel arches applied immediately upon
excavation. Generally completed by
hand, there is no mobilization and setup of a Tunnel Boring
Machine (TBM).
-
Figure 1: SEM Excavation East Side Access Northern Boulevard
Crossing
Artificial ground freezing is the conversion of in situ pore
water into ice through the circulation of cold
liquid through a system of pipes installed in the ground to
impart compressive strength and
impermeability to a soil. The frozen in situ pore water acts as
a bonding agent creating a frozen soil
mass with noticeably improved compressive strength and decreased
permeability compared to the
surrounding unfrozen soils (Powers, Corwin, Schmall, Kaeck,
2007).
A typical ground freezing system for a shaft consists of a
series of freeze pipes installed along the
perimeter of the proposed excavation, extending into the
subsurface strata. Within each of the freeze
pipes, a smaller diameter feed pipe is installed permitting the
downward circulation of the cooling
liquid which then flows back to the surface through the annulus
of the larger pipe. Calcium chloride
(brine) is chilled by a series of electrically powered
refrigeration plants. As the circulation of the brine
progresses, cylindrical columns of frozen soil grow around each
pipe. The diameter of the columns
increases with time, forming a virtually water-tight impermeable
barrier.
-
II DESIGN OF GROUND FREEZING SYSTEM
The design process was a joint effort between the Owners
Engineers and the Construction Team.
The design process included several independent yet integrated
components summarized below:
Thermal analysis to determine the required freeze pipe spacing
and temperature
profile.
Structural analysis to ensure the frozen mass will have
sufficient strength to act as
temporary structural support during excavation prior to
shotcrete.
The geology in the general area in sequence was Fill (Stratum
1), Organic Deposits (Stratum 8),
Mixed Glacial Deposits (Stratum 2, 3, and 4), Glacial Till
(Stratum 5), Decomposed Rock (Stratum 6)
and Bedrock (Stratum 7). Figure 2 provides a geologic
profile.
Figure 2: Geologic Profile
The predominant soil unit in the frozen arch was Stratum 4.
Other partial components were Stratum
3 at higher elevations, in parts of the crown; and Stratum 5
closer to the base of the arch. Stratum 6
consisted of decomposed or weathered rock. The base of the arch
was keyed into rock (Stratum 7).
From generally established knowledge of the relevant strengths
of frozen soils, it was evident that
Stratum 3 and Stratum 5 (both sands/silts, with gravel) would
have greater frozen strength properties
to those in Stratum 4. Frozen soil strength testing was
therefore limited to Stratum 4.
-
In order to describe the stress and deformation behavior, the
following tests were performed by CDM
Consultants in Bochum Germany.
Testing of unfrozen and thawed samples
o Uniaxial compression tests
o Triaxial undrained, consolidated compression tests
Testing of frozen samples
o Uniaxial compression tests (T = -10 C, T = -15 C)
o Triaxial compression tests (T = -15 C)
o Uniaxial creep tests (T = -15 C)
The following table summarizes the results of the lab tests on
Stratum 4:
TemperatureUniaxial Compressive
Strength
Shear
Parameters
2 week 2 month 3 months 6 months 2 week 2 month 3 months 6
months 2 week 2 month 3 months 6 months
Stratum 4 -10 4.95 10.0 / 2.14 330 325 320 320 1.18 1.13 1.08
1.05 0.49 0.47 0.45 0.44
Stratum 4 -15 6.85 10.0 / 2.2 390 385 380 375 1.63 1.55 1.5 1.45
0.68 0.65 0.63 0.61
Layer
o/MN/m2MN/m2
Time dependent values
Time Dependent Frozen Soil Parameter
Youngs Modulus of Elasticity Compressive Strength (allowable)
cohesion intercept (allowable)
MN/m2 MN/m2
Short term values
(Lab Conditions)
oC MN/m2
Table 1: Results of Frozen Soil Tests
Based on the results of the soil investigation the following
criteria needed to be met to achieve an
acceptable frozen arch:
Arch minimum thickness of 6 feet
Arch temperature of -32 degrees Celsius at the freeze pipes and
-1 degrees Celsius at the
intrados and extrados
The thermal analysis on this project was conducted using the
finite element (FEM) heat transfer
program Temp/W, developed and distributed by GeoSlope of
Calgary, Alberta. Several models were
evaluated as part of the thermal analyses for this project. The
results of the analyses are
summarized below:
4.0 ft. nominal spacing yields closure in 16 days and a 6 ft.
thickness in 35 days.
-
5.0 ft. nominal spacing yields closure in 25 days and a 6 ft.
thickness in 46 days.
6.0 ft. nominal spacing yields closure in 34 days and a 6 ft.
thickness in 55 days.
7.0 ft. nominal spacing yields closure in 46 days and a 6 ft.
thickness in 64 days.
Based on the results it was determined that 43 freeze pipes on
approximately 3.5 foot spacing would
be the basis for the design.
III CONSTRUCTION
The initial ground freezing approach required the installation
of (81) horizontally drilled holes
approximately 120 feet in length from the newly constructed
Queens open-cut area to the existing
excavation located to the north of Northern Blvd. The
arrangement consisted of 43 freeze pipes, 3
temperature monitoring pipes, 22 compensation grout holes/heat
pipes, 3 draining wells, and 10 void
grout holes. In addition, 4 vertical freeze pipes were also
required. The purpose of the void grout,
compensation grout, and heat pipes were to mitigate heave or
settlement during the construction
process.
Combinations of track mounted geotechnical dual rotary drill
rigs and horizontal skid mounted coring
rigs were utilized to drill the holes. The contractor
anticipated encountering obstructions that included
boulders, jet grout, concrete filled steel pipe piles, slurry
wall reinforcement, and rock. In the event
that an obstruction was encountered and was not able to be
penetrated, the contractor had the
ability to telescope drill casing giving a higher chance of
success. As a contingency, the contractor
supplied a horizontally mounted sonic drill rig to penetrate the
obstructions in the event that
traditional rotary or coring methods were unsuccessful.
Horizontal boreholes penetrated the slurry wall in varying
depths from 20 to 55 feet below the water
table. Given the nature of the silty sands to flow under these
conditions it was critical for the drilling
to be performed utilizing a groundwater control device (GWCD).
The GWCD consisted of a steel
trumpet cored and mounted in the slurry wall. Attached to each
trumpet was a combination of
chambers, valves, and seals that allowed drilling to occur with
minimal to no soil loss.
-
Figure 3: Groundwater Control Device
It was important to accurately survey the borehole upon
completion to ensure that adequate pipe to
pipe spacing was achieved. The contractor chose to utilize a
Reflex Gyroscope to perform the
borehole surveys. Several factors needed to be considered when
choosing the correct instrument. In
addition to accuracy, the gyroscope needed to be able to survey
on a flat horizontal plane as well as
inside a steel casing. These criteria eliminated standard
inclinometers and magnetic survey
instruments. The reflex gyro utilized a digital micro gyro which
consisted of a silicon microchip and
an integrated circuit assembled inside a nonmagnetic package.
This combination allowed the
instrument to provide both azimuth and dip on any plane and in
any environment. The probe
transmitted the data to a field PC utilizing Bluetooth
technology. Once processed the output could be
plotted in AutoCad for easy interpretation.
-
Figure 4: Surveying Using Gyroscope
Prior to any freeze pipe installation it was required by
specification that the contractor demonstrate
the chosen means and methods for settlement control. A trial
grout program was established which
consisted of a combination of void grouting and compensation
grouting. Void grouting was
conducted utilizing a non-cementitious grout injected through
the end of the drill casing in stages as
the casing was retracted. Compensation grout consisted of repeat
injections of both cementitious
and non-cementitious grout through Tube-A-Manchette (TAM) pipes.
The trial grout program was
considered successful once the contractor demonstrated the
ability to lift the subway structure as
measured by total stations with prisms inside the subway
structure.
In addition to the trial grout program, an additional eight (8)
void grout holes were completed prior to
the start of freeze pipe installation. It was believed based on
previous work in the area that voids
existed underneath the subway structure that would result in
uneven heave or settlement during the
construction work. The void grouting would reduce the number and
size of voids allowing for more
predictable settlement and heave. In total 19,100 gallons of
non-cementitious grout was injected
prior to the start of freeze pipe installation.
The first freeze pipe was installed utilizing high speed
rotation coring methods. Subsequent survey of
the pipe indicated a much higher deviation downward than would
be accepted based on the design
and that the soils near the top of the arch were less dense than
anticipated. It was determined to do
an additional 10 void grout holes to consolidate the soils prior
to continuing with the freeze pipe
installation. An additional 40,000 gallons of non-cementitious
grout was injected beneath the subway
box.
-
Figure 5: Freeze Pipe Installation High Speed Coring Rig
Freeze pipe installation was completed after approximately
twelve months including the time
required by the general contractor to excavate and brace a
portion of the bellmouth that was
previously unexcavated. In total 62 freeze pipes were installed
utilizing several different types of drill
rigs to overcome access and different strata.
-
Figure 6: Installing Freeze Pipes with Geotechnical Drill
Rig
In addition to freeze pipes, it was necessary to install several
temperature monitoring pipes.
Generally, these pipes are drilled and constructed similar to
freeze pipes but instead of circulating
brine they are equipped with RTD sensors, thermocouple wire, or
fiber optic wire. The temperature
probes were used to measure the change in the ground temperature
as the freezing progressed.
Typically, the locations of the temperature pipes were selected
based on the as-built surveys of the
pipes and installed in areas which had the greatest pipe
deviation so that the worst case scenario
could be monitored. Given the excavation sequence, the
monitoring pipes had to be chosen before
all of the freeze pipes were drilled. In total five temperature
monitoring pipes were installed.
The drill casing used to advance the borehole was left in place
and a 4 inch schedule 40 steel pipe
with welded joints was installed inside each casing. The pipe
was pressure tested to safeguard
against leaks and the annulus between the casing and the pipe
was grouted with a non-shrink grout.
The grout provided proper heat transfer between the circulating
brine and the surrounding soils.
All of the freeze pipes were connected to a common supply and
return header pipe which could
circulate the chilled brine back to the refrigeration plant at a
rate of approximately 20 gpm through
each freeze pipe. A single refrigeration plant with a rating of
312 tons of refrigeration was utilized.
The freeze plant used ammonia as a primary refrigerant to chill
the 28% calcium chloride solution to
temperatures ranging between -25 to -30 degrees Celsius.
Refrigeration was accomplished with two,
400 horesepower electrically-powered compressors contained
within the custom-made freeze plant
manufactured specifically for ground freezing operations.
-
The actual freezing took approximately 20 weeks to achieve the
closure. During that time, constant
data acquisition was maintained. The key components to the
freezing were temperatures and
groundwater levels. For this project, the pressure inside the
frozen arch was monitored with the use
of pressure transducers hooked up to wells installed within the
arch footprint. A gradual increase in
groundwater pressure inside the arch would indicate that the
frozen mass had achieved closure
resulting from the frozen soil mass growing inwards increasing
the pore pressure on the unfrozen
soils. Groundwater levels were also measured in a series of
vertical piezometers across the entire
site. Daily records of the ground temperatures recorded in the
temperature pipes provided data that
could be used to verify assumptions made in the initial thermal
model. The use of actual ground
temperatures in revised calibrated models aided in the
determination of the in-situ wall thickness and
temperature.
IV CONCLUSION
The General Contractor completed the final lining for the SEM
tunnel on February 25, 2013. The
freezing system operated as designed for approximately 455
days.
Figure 7: Completed Tunnel
This project represents the ability for artificial ground
freezing to act as both structural support and
groundwater cutoff in highly variable urban soils.
-
REFERENCES
Powers, J., Corwin, A., Schmall P., and Kaeck W., 2007.
Construction Dewatering and Groundwater
Control New Methods and Applications. John Wiley & Sons Inc,
United States, pp 491-500
Andersland, O.B., and Ladanyi, B. 2004. Frozen Ground
Engineering, Second Edition. John Wiley &
Sons, Inc., Hoboken, NJ.