Predicting thermal performance of a mass concrete foundation – A field monitoring case study

Predicting thermal performance of a mass concrete foundation – A field monitoring case study

Case Studies in Construction Materials 11 (2019) e00289 Contents lists available at ScienceDirect Case Studies in Construction Materials journal hom...

3MB Sizes 0 Downloads 2 Views

Case Studies in Construction Materials 11 (2019) e00289

Contents lists available at ScienceDirect

Case Studies in Construction Materials journal homepage: www.elsevier.com/locate/cscm

Predicting thermal performance of a mass concrete foundation – A field monitoring case study Yogiraj Sargama , Mahmoud Faytarounia , Kyle Ridingb , Kejin Wanga,* , Charles Jahrena , Jay Shena a b

Iowa State University, Department of Civil, Construction, and Environmental Engineering, 813 Bissell Road, Ames, IA, 50011, USA University of Florida, Engineering School of Sustainable Infrastructure and Environment, 265G Weil Hall, Gainesville, FL, 116580, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 9 July 2019 Received in revised form 26 September 2019 Accepted 8 October 2019

High-temperature differentials in a mass concrete structure pose great risks of temperature-induced stresses and cracking. Prior knowledge of temperature development within such a structure is essential. In this context, this paper presents a case study in which the construction of a mass concrete bridge foundation in Iowa, USA was investigated and a computer program, ConcreteWorks (CW), was used to predict its overall thermal performance with an aim to prevent thermal cracking. The properties of mass concrete mixes, required as CW inputs, were measured through isothermal and semi-adiabatic calorimetry tests. The temperature development profile, temperature differential, maturity, and compressive strength of the mixes were predicted and compared with those measured through the real-time monitoring of the bridge foundation. It was observed that CW predictions match well with their corresponding measured values. Three locations, core, top, and the face nearest to the core of the foundation, were found to be critical points for high temperature differentials. A sensitivity analysis, analyzing the effects of various mass concrete parameters, is also presented. The results provided clear insights into the temperature development of concrete with complex material compositions and environmental conditions. CW is a useful tool in developing thermal control plan for mass concrete projects. © 2019 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Case study Mass concrete Bridge foundation Thermal cracking ConcreteWorks Sensitivity analysis

1. Introduction ACI 116R defines mass concrete as “any volume of concrete with dimensions large enough to require that measures be taken to cope with generation of heat from the hydration of cement and attendant volume change to minimize cracking” [1]. Generally, structural members with a least dimension greater than 1.22 m (4 ft.) fall into this category. The early-age temperature development in mass concrete structures has a significant impact on their durability. A high temperature differential in such structures can result in large temperature-induced stresses that can cause cracking, especially at early ages [2–5]. The high temperature differential is primarily caused by a large amount of heat generated, due to hydration of cementitious materials, in the core of the structure that is dissipated at a very slow rate or is not dissipated locally, representing a true adiabatic condition [6–9]. To minimize the risk of cracking, various preventive measures are sometimes

* Corresponding author. E-mail addresses: [email protected] (Y. Sargam), [email protected] (M. Faytarouni), [email protected]fl.edu (K. Riding), [email protected] (K. Wang), [email protected] (C. Jahren), [email protected] (J. Shen). https://doi.org/10.1016/j.cscm.2019.e00289 2214-5095/© 2019 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

2

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

taken that includes (but not limited to) the use of supplementary cementitious materials (SCMs) to replace parts of cement, precooling aggregates and water before concrete mixing, the use of icy water or liquid nitrogen, cooling pipes, insulation blankets and others [10–13]. For example, the research has shown that the replacement of parts of cement with SCMs such as fly ash, slag, silica fume, etc., can effectively reduce the heat of hydration [6–9]. However, insulation blankets or cooling pipes might need to be installed even in the case of low heat concrete mixes to control temperature based on the weather conditions. Therefore, often a combination of various methods of temperature control is necessary and is employed in relevant projects. Nowadays, a very common procedure in mass concrete projects is to develop a thermal control plan (TCP) before the placement of concrete. Many studies have been carried out and a few finite element-based analysis computer programs (e.g., 4C Temp&Stress and ConcreteWorks) have been developed that help in predicting this temperature development. ConcreteWorks (CW) has specific capabilities to predict the early age thermal development and cracking potential of mass concrete and can assist in devising a TCP [14]. It contains modules for several structural concrete applications, including bridge deck types, precast concrete beams, and concrete pavements [15]. CW input data include: (a) concrete material properties (cementitious properties, mix proportions, etc.); (b) structural parameters (shape, dimension, subgrade condition etc.); (c) construction parameters (concrete placement temperature, casting rate, curing/insulation methods, formwork removal time etc.); and (d) environmental parameters (ambient temperature variation, relative humidity, wind speed etc.). The prediction of maximum absolute temperature, maximum temperature differential, maturity and compressive strength development with time, and cracking potential are furnished as outputs. The temperature development profile at any specified point in a mass concrete structure can also be obtained for analysis. Researchers have suggested the mix designs, construction technologies [16–21] and ways to reduce heat, the probability of cracking, and cost of various mass concrete projects [22–27]. These suggestions are based on experiences and observations from individual mass concrete projects, comprising mostly of dams and mat foundations. A few case studies could be found in the literature related to mass concrete construction. For example, Luther et al. [28] presented 30 case studies about various mass concrete projects in North America that included dams, mat foundations, bridge pier foundations and stems, reservoir foundations, and caissons. These case studies focused mainly on the concrete mixes (containing slag cement) used and the resulting temperature development. The details about the thermal control measures, adopted in the projects, were not discussed. Dilek [29] discussed the planning aspect of a critical mass concrete placement. A complete investigation of the construction process, with a focus on the adopted insulation regime, was presented. However, extensive validation of the predicted data was not discussed. To summarize, although a good amount of literature is available that discusses various aspects of a mass concrete project, studies related to aids available for developing a TCP and the validation of its application are scarce. This paper presents a case study where CW was used for prediction of temperature development in a mass concrete bridge foundation and it was validated with actual/measured data. A rectangular footing of a bridge pier in Iowa, USA was selected for the investigation. The properties of concrete and its constituent materials were determined through various lab and field tests. The measured material properties, foundation temperature development profiles, and the results of the thermal analysis performed using CW are presented along with a brief sensitivity analysis on the effect of various mass concrete parameters on temperature development. The observations from this case study will reinforce the importance of a TCP in a mass concrete project and can also provide experimental validation for the use of CW for future similar projects. 2. Investigation of a mass concrete foundation Real-time monitoring of the construction and heat development of a bridge foundation was carried out. A rectangular footing of a bridge pier, which was a mass concrete member as per ACI-207 [6], was investigated. The project chosen for the investigation was I-35 NB to US 30 WB (Ramp H) bridge in Ames, Iowa, USA. This was a 7-span continuous welded steel girder bridge constructed on 6 concrete piers with a total length of 515 m (1690 ft.) and a width of 11 m (36 ft.). Fig. 1(a) shows a cross-section of the bridge along with all the piers and the location map of the bridge is shown in Fig. 1(b). The rectangular footing of pier 4 was selected for this investigation as it was the largest amongst all footings with the dimensions of 10.06 m x 8.23 m x 2.13 m (33 ft. x 27 ft. x 7 ft.). The depth of the footing (2.13 m) was the critical dimension that qualified it as a mass concrete member as per Iowa DOT specification [30] and, therefore, its construction and early-age temperature profile were analyzed. The temperature development in a mass concrete structure is dependent on a range of factors such as the subsurface profile, boundary and environmental conditions, concrete mix proportion, and cooling method. For this reason, the construction of pier 4 footing was studied in three stages – before, during, and after the placement of concrete. Some of the important information is presented and discussed in the sections below. 2.1. Subsurface profile The bridge site was located in an area of Iowa that has been formed by extensive Wisconsin age glacial activity. During the initial stage of project finalization, a soil investigation at the job site was carried out by HDR Inc. The primary geologic strata encountered included topsoil, existing fill soils, cohesive alluvium, alluvial sand, glacial till, and bedrock. Topsoil depths ranged from 127 to 203 mm (5–8 inches) along the project alignment. Silty and fine sand was observed up to a depth of

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

3

Fig. 1. US30-I35 Bridge in Iowa, USA.

3.35 m (11 ft.). The properties of the soil were measured using a Shelby tube cored up to a depth of 0.60 m (2 ft.). The measured values are presented in Table 1. Bedrock was encountered at depths ranging from 10.3 to 25.3 m (34–83 feet). The bedrock units appeared to include siltstone, sandstone, and shale based on examination of split-barrel samples, with varying degrees of weathering. 2.2. Footing support, subbase, and formwork The load transfer mechanism for the footing was a total of 30 HP 14  117 steel bearing H-piles driven 16.7 m (55 ft.) below the ground surface and the steel reinforcement footing cage was placed over these piles. Based on the outcomes of the soil investigation, a layer of crushed limestone aggregate was placed on the subbase to provide a firm and dry casting surface. Fig. 2 (a) shows the driven H-piles and the crushed rock casting surface. In central Iowa, typically wood and steel formwork materials are used to form footing placements. The choice of formwork material depends on the nature of construction as well as the availability and cost of the material. Wood formwork was used in the construction of the footing. This formwork consisted of plywood attached to galvanized cold-rolled steel supporting members with nails. These, in turn, were supported by vertical cold-rolled steel members of the longer cross-section.

Table 1 Properties of soil. Soil property

Measured value

Dry density Cohesion Angle of internal friction Modulus

19 kN/m3 20 kPa 30 15 MPa

4

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

Fig. 2. Pier 4 footing and insulation.

2.3. Insulation method The footing was constructed in the month of June when the average daily ambient temperature was 26  C (79  F). The temperature at the outer faces of footing was expected to be close to the ambient temperature, while that in the core was expected to be higher due to cement hydration thereby causing a substantial temperature differential. In order to keep the differential within the acceptable limits, the footing was insulated 10 h after the placement of concrete. The exterior of the wood formwork and the top face of the footing were wrapped with a 50 mm (2-inch) thick black insulation blanket with a specified R-value rating of 5 [Fig. 2(b)]. The R-value is equivalent to thickness divided by the thermal conductivity of the insulation material. A higher R-value means the higher ability of the insulation barrier to resist conductive heat transfer. For example, an R-value of 1 means low insulation ability of the barrier whereas an R-value of 5 means a high insulation ability. Real-time monitoring of concrete temperature and weather conditions led to the removal of the formwork on the 5th day after the placement of concrete. However, the footing was kept covered with insulation blankets up until the 10th day. 3. Experimental program 3.1. Installation of temperature sensors To monitor the temperature development of the concrete placement in footing, the temperature data were recorded in one-hour intervals and were monitored remotely as well at 4 -hs intervals for a period of 10 days from the day of the concrete placement. A total of 7 sets of temperature sensors were installed. Each set included a primary and a backup temperature sensor. After installation, the exact location of the sensors was measured as shown in Error! Reference source not found. (S1S7). The locations are also explained as follows: S1 – at the center of the concrete footing, installed 0.84 m (2.75 ft.) below the top surface S2 – in the middle of the length and width, near the top and lateral surfaces of the footing, installed at 0.15 m (6 in.) below the top surface S3 – in the middle of the length and height, installed at 0.07 m (3 in.) inside the long lateral surface, 0.88 m (2.9 ft.) from the top surface and 0.76 m (2.5 ft.) from the bottom surface S4 – installed outside the formwork, to measure the ambient air temperature S5 – 0.33 m (1.08 ft.) below S2, to monitor the temperature change along the vertical direction S6 – installed at the middle of the height and width, 0.07 (3 in.) inside the short lateral surface, 0.91 m (3 ft.) from the top surface and 0.76 m (2.5 ft.) from the bottom surface, to monitor the temperature in another cross-section, S7 – installed in the center, 0.30 m (1 ft.) above the bottom surface of the footing, to investigate the effect of the subgrade temperature 3.2. Materials and mix proportion Ready-mixed concrete was used for the placement of footing. The concrete mix proportion is shown in Table 2. Materials used in the concrete were ASTM C150 type I/II cement, class C fly ash, 25 mm (1-inch) nominal maximum size limestone coarse aggregate, and river sand as fine aggregate. Total cementitious content of the mix was replaced with 20% of fly ash (by weight). The raw materials used in the concrete were collected and used in the laboratory to perform tests for their properties. The chemical composition of cementitious materials is presented in Table 3. The composition of type I/II cement meets the requirements of both type I and type II cement as per ASTM C150. Coarse and fine aggregates were tested for their properties such as specific gravity, absorption, dry – rodded unit weight (DRUW),

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

5

Table 2 Mass concrete mix proportion. Material

Quantity kg/m3 (lb./yd3)

Cement (Type I/II) Fly Ash (Class C) Fine Agg. Coarse Agg. Water w/b

281 (474) 71 (119) 890 (1500) 900 (1517) 151 (255) 0.43

Table 3 Chemical composition of cementitious. Oxides

Type I/II cement(%)

Class C fly ash (%)

SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2O Others LOI

20.44 5.11 3.27 60.95 3.59 3.03 0.18 0.61 1.52 1.96

33.76 15.23 6.30 31.17 4.98 2.25 1.35 0.60 4.93 0.57

and fineness modulus. The specific gravity, absorption, and DRUW of coarse aggregate were measured to be 2.68, 0.82%, and 1608.90 kg/m3 (100.44 lbs./ft.3), respectively. The specific gravity, absorption, and fineness modulus of fine aggregate were measured to be 2.65, 0.98%, and 2.85, respectively. 3.3. Test methods Tests for measuring fresh properties of concrete were performed as per relevant ASTM standard test procedures. Slump (ASTM C143), air content (ASTM C173), and unit weight (ASTM C138) of the concrete mix were measured at the construction site during placement of concrete. Cylindrical specimens of dimensions 100 mm x 200 mm (4 in.  8 in.) were cast and cured in site conditions for various hardened properties tests such as maturity (ASTM C1074) and compressive strength (ASTM C39). Nurse-Saul maturity method was employed in this study to monitor the development of compressive strength of concrete based on its temperature history. This method is based on the assumption that samples of a given concrete mixture attain equal strengths if they attain an equal value of maturity index. As per this method, maturity index was calculated in terms of temperature-time factor (TTF). The compressive strength-maturity relationship was then developed by performing a regression analysis to determine a best-fit equation to the measured data. The best-fit equation was of the form given in Eqn. (1). S=a + b log (M)

(1) 

Where S is the compressive strength (in MPa), M is the TTF (in C–hours), and ‘a’ and ‘b’ are coefficients. Isothermal and semi-adiabatic calorimetry tests were performed to analyze the heat development due to the hydration of cementitious materials. In this study, an eight-channel PTC isothermal calorimeter was used to measure the heat generation of cement pastes following ASTM C1702. The semi-adiabatic calorimetry test method is explained in the following section. 3.3.1. Semi-adiabatic calorimetry The characterization of the temperature rise in a mass concrete structure requires an estimate of the adiabatic temperature rise of the concrete mixture [32]. Adiabatic calorimetry requires a process in which no heat is gained or lost to the system’s surroundings. However, due to its high set-up cost and the requirement of large sample size, it is less practical than a semi-adiabatic calorimeter. Therefore, in general, semi-adiabatic calorimetry is performed in which the heat loss is also measured and the measured temperature values of the concrete are corrected to account for this loss. Even though the semi-adiabatic calorimetry method is a common test, there is no standard test method for this. This study followed

6

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

the guidelines outlined by Poole et al. based on which a semi-adiabatic calorimeter was developed in the laboratory [32]. A schematic diagram showing the details of the calorimeter is shown in Fig. 4. A 0.61 m by 0.86 m (24 in. by 34 in.) cylindrical drum was used for building the semi-adiabatic calorimeter. Aeromarine insulation foam was poured inside the drum which solidifies and acts as the insulation. Top insulation lid was also prepared using the same insulation foam. For installing the 0.18 m x 0.36 m (7 in.  14 in.) steel chamber in the middle of the drum, galvanized steel sheet was used. Fresh concrete was collected from the construction site (into a 0.15 m x 0.30 m [6 in.  12 in.] cylinder) and was immediately transported to the laboratory (to minimize the heat loss) for the semi-adiabatic calorimetry test. The temperature was measured using Type T thermocouples (TC) at three locations – one at the center of the concrete specimen (MID), one at the surface of the steel chamber (EXT1) and one at an inch away from the chamber surface in the insulation (EXT2). EXT1 and EXT2 TCs were installed to measure the heat loss through the calorimeter. MID TC was placed 6 in. into the center of the fresh concrete specimen. A plug-in for this thermocouple was installed at the edge of the steel chamber opening. For connecting thermocouple wires to the data logger, a hole was drilled in the middle of the drum surface through which the wires were taken out. The test set up was kept in a closed room where temperature variations were limited. The temperature data were recorded using Pico Technology USB TC-08 data logger for 160 h at 15-minute intervals. The calorimeter was calibrated before the test to determine the calibration factors. The values of the calibration factors 1 and 2 were 0.0197 and 0.3970 W/ C, respectively. 4. Results and discussions 4.1. Material properties 4.1.1. General properties of concrete mix The ready-mixed concrete at the construction site was tested for its fresh properties following relevant ASTM standards as mentioned earlier. The values of slump, air content, unit weight, and temperature were measured to be 70 mm (2.75 in.), 7.50%, 2388.67 kg/m3 (149.12 lbs./ft.3), and 17.55  C (63.6  F), respectively. The compressive strength development was monitored by the Nurse-Saul maturity method as shown in Fig. 5(a). Compressive strength was tested according to ASTM C39 [31], and the values are plotted as a function of the temperature-time factor (TTF) in Fig. 5(b). The values of the coefficients ‘a’ and ‘b’ of the best-fit linear equation [Eq. (1)] were calculated to be -34.26 and 7.51, respectively. 4.1.2. Activation energy It is well-known in cement chemistry that, the hydration of cementitious materials is also sensitive to temperature like other chemical reactions. The apparent activation energy is a useful measure of the early-age temperature sensitivity of a concrete mixture and various calculation methods have been proposed for its calculation. A single linear approximation of reaction rate was used in this study for activation energy calculation [33]. The paste (cement + fly ash) samples were tested in the isothermal calorimeter at four constant temperatures: 10, 20, 30, and 40  C. The reaction rate was calculated using a single, best-fit least-squares line of the linear, acceleration phase of the isothermal rate of heat curve. The slope of the best-fit line determined the reaction rate k at a particular temperature. ln (k) was then plotted versus the inverse of absolute temperature to determine Ea. The plot obtained in this study is shown in Fig. 6. The Ea of this mix was calculated to be 34,173 J/mol. For similar cement and fly ash blends, Ea values lie in the range of 30,000–44,000 J/mol as found in the literature [34–36]. The Ea determined in this study also lies in this range. The calculated Ea was used as an input in CW for predictions of the temperature profile of mass concrete foundation. 4.1.3. Hydration curve parameters Since the heat loss in a real mass concrete structure is comparatively slower than that measured in semi-adiabatic calorimetry, the temperature development under these conditions needed to be known. The apparent activation energy (Ea, calculated from isothermal calorimetry), total heat of hydration (Hu, estimated from the chemical composition of cementitious materials), and the measured data from semi-adiabatic calorimetry were used to calculate a theoretical adiabatic hydration curve following the procedure outlined by Poole et al. [32]. The curve was modelled using the threeparameter exponential function as shown in Eq. (2). t b

a ðtÞ ¼  au   e ½ t 

ð2Þ

Where, α (t) is the degree of hydration at age t, and αu, β, and t are hydration curve parameters. The measured and calculated semi-adiabatic and calculated adiabatic curves are shown in Fig. 7. The values of curve-fit parameters (αu, β, and t) are presented in Table 4. αu ( = 0.748) is the ultimate degree of hydration (DOH), β ( = 0.840) is the hydration shape parameter, and Table 4 Estimated total heat, activation energy, and hydration curve parameters. Mixture

Hu, J/Kg

Ea, J/mol

αu

β

t, hours

R2

Concrete

472,296

34,173

0.748

0.840

20.006

0.994

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

7

Fig. 3. Location of temperature sensors installed in the footing (not in scale).

t ( = 20 h) is the hydration time parameter. A larger αu indicates a higher magnitude of ultimate DOH, larger β indicates a higher hydration rate at the linear portion of hydration curve, and a larger t means a larger delay of hydration [37]. These parameters of the concrete mix were then used as inputs in CW for temperature prediction for the mass concrete foundation. 4.2. Results from foundation investigation As discussed in section 3.1 and shown in Fig. 3, sensors were installed in the mass concrete foundation to monitor the temperature development after the placement of concrete. The monitoring was done until the insulation was removed. The measured values of temperature by all the sensors are plotted in Fig. 8. The differential temperatures along with the Iowa DOT specified limits in early-age mass concrete are shown in Fig. 9. According to the Iowa DOT specification [30], the following limits apply to mass concrete placements:  The concrete temperature at the time of placement shall be between 5  C and 21  C (40  F and 70  F).  The maximum concrete temperature shall not exceed 71  C (160  F).  The differential temperature between the core and a point 2 to 4 in. inside the surface along the shortest line from the core to the nearest surface or top surface of the element shall not exceed the limits of Table 5.  Thermal control of each mass placement shall be maintained until the temperature of the interior is within 10  C (50  F) of the average outside air temperature (determined by averaging the daily high and low temperatures over the preceding seven calendar days). Analyzing the measured data and the plots presented in Figs. 8,9, the following observations, can be made: 

Fig. 4. Schematic diagram of the semi-adiabatic calorimeter.

8

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

Fig. 5. (a) Compressive strength development; and (b) strength-maturity relationship.

Fig. 6. Arrhenius plot for activation energy calculation.

Fig. 7. Semi-adiabatic and true adiabatic curves.

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

Fig. 8. Temperature measured by installed sensors.

Fig. 9. Measured differential temperature in footing.

Table 5 Iowa DOT maximum temperature differential limits. Time (hours)

Maximum temperature differential,  C ( F)

0-24 24-48 48-72 >72

11 (20) 17 (30) 22 (40) 28 (50)

9

10

 

 

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

The concrete temperature at the time of placement was 19  C (66.2  F) which was within the Iowa DOT specified limits of 4.45  C and 21.11  C (40  F and 70  F). It can also be observed that during the initial 5–10 hours, the temperature recorded by the sensors fluctuates. Since the concrete was poured starting from the center of the footing and then moving towards the edges, it might have resulted in a higher value of temperature recorded by the sensor at the edge (e.g., S3) than that recorded by the core sensor (S1). Also, the sensors take some time to stabilize after concrete is poured, which explains the initial fluctuations observed almost in all the temperature recordings. The maximum concrete temperature of 65  C (149  F) was recorded at the core (sensor S1) that occurred after 40.35 h of concrete placement (Fig. 8). The temperature differential (TD) between the core (S1) and the short (S3) as well as the long side (S6) sensors (Fig. 9) was found to be well within the specified limits whereas that between the core and top sensors was observed to cross the specified limit only for a short duration from 44 to 48 h. Barring this, TD was well within the specified limits. The maximum value of TD was observed to be 22  C (39.6  F) that occurred around 73.5 h after the placement of concrete. The maximum value of the temperature recorded at the top of the footing (S2) was 48.88  C (120  F) while that at the bottom of the footing (S7) was 57  C (134.6  F) [Fig. 8(b)]. No significant difference was observed in the recorded temperatures in the centers of the short and long side faces (S3 and S6).

4.3. Thermal analysis using ConcreteWorks ConcreteWorks computer program was used for thermal analysis of the rectangular footing. In this analysis, the measured properties of the concrete mix, as presented in section 4.1, were used as inputs. For all other inputs, CW default values were used. Baseline values of all the inputs, used for the analysis of footing, are presented in Table 6. Usually, in rectangular footings that qualify as mass concrete, only three locations are critical from the perspective of thermal cracking. The maximum temperature is typically recorded at the center of the volume of a mass concrete member, while the minimum temperature occurs at a point a few inches into the concrete surface. The minimum temperature can occur at either of the top surface or at one of the face/edge/openings of the volume. Therefore, the temperature profile at three locations of the footing (core, top, and center of the short side) was analyzed using CW and compared with the measured values as shown in Fig. 10. Following observations can be made:  CW predicted Tmax to be 65.1  C (149.16  F) at the core of the footing, which was only slightly more than the measured value of 65  C (149  F). The complete temperature development profile at the core predicted by CW simulates the measured profile really well as shown in Fig. 10(a). CW prediction of the time to reach Tmax (52 h) was also close to that of the

Table 6 ConcreteWorks inputs for thermal analysis of bridge footing. Parameter Member Inputs Project location Unit system Analysis duration Concrete placement time Material Properties Cement content C Fly ash Water content Coarse agg. Content

Value Ames, IA, US Metric (English) 14 8

Units

days AM

See Table 2

Fine agg. content Air content Chemical admixture Environmental Inputs Concrete fresh temp Blanket R-value

19 (66.2) 0.088 (0.5)

Form type Soil temperature Footing subbase Side cure method

Wood 26.67 (80) Limestone Black plastic



C ( F) m2-K/W (hr-ft2- F/ BTU) 

C ( F)

Parameter

Value

Units

Shape choice Member width

Rectangular Footing 8.23 (27)

m (ft.)

Member length Member depth

10.06 (33) 2.13 (7)

m (ft.) m (ft.)

Cement type Cement chemistry values Hydration parameter values Thermal conductivity

I/II Measured (from Table 3) αu = 0.748; β = 0.840; t = 20 2.7 (1.56)

CTE Coarse agg. type Fine agg. type

8.28 (4.6) Limestone Siliceous river sand

Ave. daily max temp. Ave. daily min temp.

27.5 (81.5) 14.45 (58)



Ave. daily max solar radiation Ave. daily max wind speed Ave. daily max RH Ave. daily min RH

731.1 24.1 95.1 45

W/m2 m/s % %

W/m-K (BTU/hr-ft F) 106/ C ( F)



C ( F) C ( F)

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

11

Fig. 10. Measured and CW predicted temperature profiles at (a) core; (b) top; and (c) short side of the footing.

field-measured value (50 h). If the approximate time to reach Tmax is known in addition to the value of Tmax, an appropriate TCP can be developed in advance to control thermal cracking.  The Tmax prediction at the top of the footing by CW was 49.55  C (121.2  F) against a measured value of 49  C (120.2  F). Also, CW simulation of the complete temperature development profile at the top was quite close to the measured profile except for an underestimation after around 110 h [Fig. 10(b)]. CW models the temperature profile of a rectangular footing in four potential construction stages - (1) before the blanket or cure method is applied to the top surface; (2) when the blanket or cure method is applied to the top surface; (3) after the removal of formwork and cure method; and (4) the time period when a cure method on the top and sides is used after the forms and initial top surface curing methods are removed. In this study, the formwork in the field was removed after around 110 h while the top and sides of the footing were still covered with the blankets until around 240 h. This difference in the field construction method was not considered in the current version of CW which might be the possible reason for the underestimation of the top temperature profile after 110 h.  CW consistently underestimated the temperature profile at the short side [Fig. 10(c)]. Large temperature differences in a mass concrete member can be very detrimental from the perspective of thermal cracking/shock. The temperature difference causes a volume change due to expansion/contraction when the member is restrained by adjacent parts of the mass foundation which might result in cracking [9]. Therefore, most of the specifications (including Iowa DOT) restrict DTmax that a mass concrete member can experience during early age; various preventive measures are employed in this age to meet the specifications and to prevent cracking.

12

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

It is very helpful in developing a preventive TCP if the TD between the critical points within a mass concrete member can be predicted in advance. CW predicts Tmax as one of the outputs. The differential between two points could also be predicted and analyzed. In this study, the differentials between the core and other critical locations (top and short side of the rectangular footing) were predicted using CW and compared with the measured values. The differential temperature charts are shown in Fig. 11. The time series for maximum TD specified by Iowa DOT is also plotted in these charts. The observations from Fig. 11 are as follows:  CW prediction of the core and top TD profile simulated the measured profile for the first 110 h, as shown in Fig. 11 (a). Since this is the most critical TD, its prediction close to the actual value, (especially in the initial 110 h), can be very helpful in preparing the TCP. CW, however, overestimated the core and top TD after around 110 h of the placement of concrete the reason of which might be the underestimation of top temperature profile as explained earlier.  The TD profile between the core and the center of the face at the shortest distance from the core (short side) was overestimated by CW for the entire duration of analysis, as can be seen in Fig. 11 (b). On average, CW overestimated the TD between the core and short side by 5.27  C (9.5  F).

4.4. ConcreteWorks sensitivity analysis A brief sensitivity analysis was performed, using ConcreteWorks, to investigate the effects of various parameters on the temperature development in mass concrete. Critical variables in three major groups (mix proportion, concrete material properties, and construction parameters) were analyzed for their effects on the maximum temperature and temperature differential in the rectangular footing. The analyzed inputs, baseline values, and their input and output ranges (and trends) are shown in Table 7. The baseline values were kept the same as those used in the thermal analysis presented in section 4.3. The trends observed in maximum temperature (Tmax) and maximum TD (DTmax) corresponding to the changes in input types are plotted as bar charts in Fig. 12 and are discussed in following sub-sections. 4.4.1. Mix proportion parameters The mix proportion parameters evaluated were cement content, class C fly ash, class F fly ash, slag, and silica fume. It can be observed from Fig. 12(a) that increasing cement content from 231 kg/m3 (414 lbs. /yd3) to 435 kg/m3 (734 lbs. /yd3) increased Tmax as well as DTmax. This can be attributed to the increase in heat of hydration with an increase in cement content in the concrete mix. It can also be noted that Tmax and DTmax values obtained up to a cement content of 381 kg/m3 (594 lbs. /yd3) were still within the specified limit of 71  C (160  F) and 28  C (50  F), respectively. However, as the cement content was

Fig. 11. CW predicted and measured temperature differentials.

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

13

Table 7 ConcreteWorks sensitivity analysis. Range of Outputs,  C ( F) Group

Mix proportion inputs

Material Properties inputs

Input range

Unit

Tmax [trend]

DTmax [trend]

Cement content 281 (474)

231,281,331,381,435 (414,474,534,594,734)

kg/m3 (lb./yd3)

C fly ash replacement F fly ash replacement Slag replacement Silica fume replacement Cement type

20

0,20,30,40

%

0

0,20,30,40

%

0

0,30,50,70

%

0

0, 3, 5, 8

%

I/II

I, II, I/II, V

Type

53-72 (127162) [→] 57-53 (135127) [ ] 57-46 (135116) [ ] 57-59 (135137) [→] 57-54 (135129) [ ] 58-48 (137118) [ ]

12-37 (2267) [→] 14-12 (2621) [ ] 14-09 (2617) [ ] 14-13 (2625) [ ] 14-12 (2623) [ ] 15-10 (2618) [ ]

2.75 (1.59) 837 (0.2)

1.71,2.23,2.75,3.27 (0.99,1.29,1.59,1.89) 753,795,837,920 (0.18,0.19,0.20,0.22)

19 (66)

11,192,737 (51,668,196)

W/m-K (BTU/ hr.-ft-F) J/kg-K (BTU/ lb.-F)  C ( F)

Steel

Steel, wood, insulated steel

Type

0.08,0.14,0.19,0.25 (0.51,0.81,1.11,1.41)

m2-K/W (hr.ft2-F/BTU) method

57-54 (134130) [ ] 59-52 (138126) [ ] 48-74 (114164) [→] 56-56 (132132) [–] 56-57 (132134) [→] 56-56 (132132) [–] 56-57 (132134) [→]

16-12 (2922) [ ] 15-12 (2721) [ ] 11-21 (1937) [→] 13-13 (2424) [–] 13-14 (2425) [→] 13-13 (2424)[–] 05-15 (0924) [→]

Input type

Concrete k Combined aggregate Cp Placement temperature Formwork type Construction inputs

Baseline values

Blanket R-value 0.088 (0.51) Curing method Black plastic Type of subbase Limestone

White curing compound, black plastic, wet curing blanket, white or clear plastic Limestone, topsoil, concrete, sand

type

increased to 435 kg/m3 (734 lbs. /yd3), Tmax and DTmax increased up to 72  C (162  F) and 37  C (67  F), respectively thereby exceeding the temperature limits. Replacing cement with SCMs is an effective way of reducing heat and this is confirmed with the trends shown in Fig. 12(a). Increasing the replacement percentage of C fly ash, F fly ash, and silica fume reduced Tmax and DTmax. A similar reducing trend was also observed in the case of slag replacement from 0% to 50%. However, an increase in both Tmax and DTmax can be seen from 50% to 70%. This is contrary to earlier belief [5,38,39], but some of the recent experimental adiabatic studies on concrete mixes containing slag confirm this observation [40]. The effect of pozzolanic as well as the latent hydraulic activity of slag on the cement hydration might be the reasons for an increase in the generated heat. This needs to be investigated further. However, the observations from this sensitivity analysis can be used to optimize the concrete mix proportion for minimum heat generation. 4.4.2. Material properties The material properties parameters evaluated were cement type, concrete thermal conductivity (k), and combined aggregate specific heat capacity (Cp) [Fig. 12(b)]. The order of Tmax and DTmax for cement types were I > I/II > II > V. This is expected as the heat generation due to cement hydration depends predominantly on the C3A content of cement. The typical C3A contents of these cement types reduce in that order (I > I/II > II > V) thereby reducing Tmax and DTmax. Increasing k and Cp reduced Tmax and DTmax. Thermal conductivity is defined as the rate of heat conduction, and as it increases, the heat generated in the core of a mass concrete member is dissipated at a faster rate, thereby reducing Tmax and DTmax [41]. 4.4.3. Construction parameters The parameters related to mass concrete construction can also affect the temperature development. Placement temperature, formwork type, insulation blanket R-value, curing method, and subbase type were construction parameters evaluated in this analysis [Fig. 12(c)]. Mass concrete placement temperature was observed to have a great impact on Tmax and DTmax. An 18  C (30  F) increase in placement temperature caused a corresponding 18  C (30  F) and 7  C (11.45  F) increase in Tmax and DTmax, respectively. Placement temperature of 37  C (96  F), which is frequent in construction during summer, resulted in Tmax of 74  C (164  F) that is out of the specified limit of 71  C (160  F). Various formwork types (steel, wood, and insulated steel) and curing methods (curing compound, blanket, and white/black plastic) were evaluated separately but no change was observed in the temperature development. However, their combinations along with changes in other properties might result in an increase/decrease of Tmax and DTmax. A greater R-value represents better insulation power of the material, as discussed in section 2.3. However, increasing R-value of the insulation blanket from 0.08 to 0.25 m2-K/W (0.51 to 1.41 h-ft2-F/BTU) had minimal impact on Tmax and DTmax with increments of 0.86  C and 0.45  C (1.15  F and 0.63  F), respectively. Type of subbase, on the other hand, was found to affect heat dissipation considerably. Changing

14

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

Fig. 12. Effect of mass concrete parameters on maximum temperature (Tmax) and temperature differential (DTmax) [see Table 7 for the ranges of input values].

topsoil to concrete subbase, Tmax increased only by approximately 1  C (1.8  F) while an increase of 5.3  C (9.5  F) occurred in DTmax. A similar change of subbase type from limestone to sand resulted in a 0.7  C (1.2  F) and 18.4  C (5.1  F) increase in Tmax and DTmax, respectively. 5. Conclusions In this study, ConcreteWorks (CW) computer program was used to perform a brief sensitivity analysis and to predict the temperature profile of a mass concrete structure. The predictions were validated for their accuracies with the results and observations from the investigation of a mass concrete bridge footing. Specific conclusions from this study are as follows:

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

15

(1) The apparent activation energy (Ea) and hydration curve parameters (αu, β, and t), obtained respectively from isothermal and semi-adiabatic calorimetry measurements, are two significant material properties for temperature development predictions in a mass concrete member. (2) The temperature differentials, between the core and the midpoints of the top surface and the surface nearest to the core of a rectangular mass concrete footing, are critical for thermal cracking. The temperature development at these three locations is necessary to be monitored. (3) The maximum temperature, maximum temperature differential, maturity, and compressive strength predicted by CW were very consistent with the measured data before the formwork removal (100 h). (4) Sensitivity analysis revealed a considerable impact of cement content, class C and F fly ash, concrete thermal properties, placement temperature, and type of subbase on temperature development in the footing. High cement content and high placement temperature resulted in concrete temperature exceeding the specified limits. Based on this information, CW can be used to optimize the mix proportion and other parameters of mass concrete to keep the temperature within the limits. CW program can be calibrated with the measured properties of locally available raw materials and produced concrete. The calibrated program can then reliably be used for the thermal analysis of a mass concrete structure. The analysis results can aid in the development of thermal control plan to control cracking. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments This study was a part of a research project “Evaluate, Modify, and Adapt the ConcreteWorks Software for Iowa's Use”, sponsored by the Iowa Highway Research Board (IHRB). The authors would like to acknowledge the sponsorship. References [1] ACI (American Concrete Institute), Cement and Concrete Terminology, ACI 116R, 2000. [2] Y. Ballim, A numerical model and associated calorimeter for predicting temperature profiles in mass concrete, Cem. Concr. Compos. 26 (2004) 695– 703, doi:http://dx.doi.org/10.1016/S0958-9465(03)00093-3. [3] M. Nili, A.M. Salehi, Assessing the effectiveness of pozzolans in massive high-strength concrete, Constr. Build. Mater. 24 (2010) 2108–2116, doi:http:// dx.doi.org/10.1016/j.conbuildmat.2010.04.049. [4] P. Choktaweekarn, S. Tangtermsirikul, Effect of aggregate type, casting, thickness and curing condition on restrained strain of mass concrete, Songklanakarin J. Sci. Technol. 32 (2010) 391–402. [5] B. Kolani, L. Buffo-Lacarrière, A. Sellier, G. Escadeillas, L. Boutillon, L. Linger, Hydration of slag-blended cements, Cem. Concr. Compos. 34 (2012) 1009– 1018, doi:http://dx.doi.org/10.1016/j.cemconcomp.2012.05.007. [6] ACI (American Concrete Institute), Guide to Mass Concrete, ACI, 2006 207-06. [7] K.A. Riding, J.L. Poole, A.K. Schindler, M.C.G. Juenger, K.J. Folliard, Statistical determination of cracking probability for mass concrete, J. Mater. Civ. Eng. 26 (2013)04014058, doi:http://dx.doi.org/10.1061/(ASCE)MT.1943-5533.0000947. [8] K.J. Folliard, M. Juenger, A. Schindler, K. Riding, J. Poole, L.F. Kallivokas, S. Slatnick, J. Whigham, J.L. Meadows, Prediction model for concrete behavior: final report, Texas Dep. Transp. (2002). [9] K.A. Riding, J.L. Poole, A.K. Schindler, M.C.G. Juenger, K.J. Folliard, Evaluation of temperature prediction methods for mass concrete members, ACI Mater. J. 103 (2006). [10] K.L. Scrivener, B. Lothenbach, N. De Belie, E. Gruyaert, J. Skibsted, R. Snellings, A. Vollpracht, TC 238-SCM: hydration and microstructure of concrete with SCMs State of the art on methods to determine degree of reaction of SCMs, Mater. Struct. Constr. 48 (2015) 835–862, doi:http://dx.doi.org/10.1617/ s11527-015-0527-4. [11] B. Lothenbach, K. Scrivener, R.D. Hooton, Supplementary cementitious materials, Cem. Concr. Res. 41 (2011) 1244–1256, doi:http://dx.doi.org/10.1016/ j.cemconres.2010.12.001. [12] M.C.G. Juenger, R. Siddique, Recent advances in understanding the role of supplementary cementitious materials in concrete, Cem. Concr. Res. 78 (2015) 71–80, doi:http://dx.doi.org/10.1016/j.cemconres.2015.03.018. [13] N. Shanahan, V. Tran, A. Zayed, Heat of hydration prediction for blended cements, J. Therm. Anal. Calorim. 128 (2016) 1279–1291, doi:http://dx.doi.org/ 10.1007/s10973-016-6059-5. [14] K.A. Riding, Early Age Concrete Thermal Stress Measurement and Modeling, The University of Texas at Austin, Austin, TX, 2007. [15] K. Folliard, A. Schindler, P. Pesek, ConcreteWorks V3 Training / User Manual ConcreteWorks Software (P2), (2017) . [16] G. Zhang, H. Li, C. Shi, Study on mass concrete slab construction technology of dalian world finance center, Appl. Mech. Mater. 137 (2012) 175–180, doi: http://dx.doi.org/10.4028/www.scientific.net/AMM.137.175. [17] H. Zhu, H. Song, Q. Kuang, Application of seamless construction technique to mass concrete structures, Huazhong Keji Daxue Xuebao (Ziran kexue ban), Journal Huazhong Univ. Sci. Technol. (Natural Sci. Ed. 32 (2004) 100–102. [18] S.U. Yin, Y. Huang, Y.H. Yuan, Research on mass concrete construction techonogy of raft foundation of hongyun building, Adv. Mater. Res. 690 (693) (2013) 709–713, doi:http://dx.doi.org/10.4028/www.scientific.net/AMR.690-693.709. [19] C. Lv, C. Qiao, Mass concrete construction technology of high-rise building foundation platform, 2011 Int. Conf. Electr. Technol. Civ. Eng. ICETCE 2011 Proc. (2011) 442–445, doi:http://dx.doi.org/10.1109/ICETCE.2011.5775152. [20] J. Zhu, Y. Bin Zhang, X.F. Du, Y. Jiang, Study on mass concrete construction technology of raft foundation in Yantai Ocean development center, Adv. Mater. Res. 919–921 (2014) 1421–1425, doi:http://dx.doi.org/10.4028/www.scientific.net/AMR.919-921.1421. [21] S. Zheng, Z. Sun, H. Zhu, Mass concrete construction technology of the three gorges project, Zhongguo Kexue Jishu Kexue/Scientia sin, Technol. 47 (2017) 796–804, doi:http://dx.doi.org/10.1360/N092017-00096.

16

Y. Sargam et al. / Case Studies in Construction Materials 11 (2019) e00289

[22] M.R.H. Dunstan, P.D. Mitchell, Results of a thermocouple study in mass concrete in the Upper Tamar Dam, Proc. Inst. Civ. Eng. (London). Part 1 - Des. Constr. 60 (1976) 27–52, doi:http://dx.doi.org/10.1680/iicep.1976.3485. [23] C. Wang, Z. Zhou, Q. Hu, J. Ou, Construction control of mass concrete of Nanjing 3rd yangtze Bridge using FRP-packaged FBG sensors, Proc. SPIE - Int. Soc. Opt. Eng. 5855 PART II (2005) 1012–1015, doi:http://dx.doi.org/10.1117/12.623727. [24] E.A. Abdun-Nur, F.A. Anderson, H.L. Boggs, D.A. Bonikowsky, R.A. Bradshaw Jr, E.G.W. Bush, L.H. Diaz, T.P. Dolen, K.D. Hansen, G.R. Mass, A.T. McCarthy, J. E. Oliverson, R.F. Oury, J.M. Raphael, E.K. Schrader, S.B. Tatro, Effect of restraint, volume change, and reinforcement on cracking of mass concrete, ACI Mater. J. 87 (1990) 271–295. [25] A.M. Lawrence, M. Tia, M. Bergin, Considerations for handling of mass concrete: control of internal restraint, ACI Mater. J. 111 (2014) 3–11, doi:http://dx. doi.org/10.14359/51686403. [26] Z. Xu, Y. Ding, C. Zhai, Q. Li, Cracking and temperature control in mass concrete construction, Adv. Mater. Res. 446–449 (2012) 841–845, doi:http://dx. doi.org/10.4028/www.scientific.net/AMR.446-449.841. [27] J. Poole, K. Riding, R.A. Browne, A. Schindler, Temperature management of mass concrete structures, Concr. Constr. - World Concr. 51 (2006) 47–53. [28] M. Luther, P. Bohme, W. Wilson, Case studies-North American mass concrete projects featuring ASTM C989 slag cement, ACI Spec. Publ. Fall 2008 Conv., American Concrete Institute, Farmington Hills, MI, 2008, pp. 13–30. [29] U. Dilek, Planning and execution of a mass concrete placement utilizing insulation regimen, Am. Concr. Institute, ACI Spec. Publ. 2011-Octob (2011) 1– 12. [30] D.O.T. Iowa, Developmental Specifications for Mass Concrete -Control of Heat of Hydration, (2009) . [31] C.39 ASTM, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens, ASTM, 2016, doi:http://dx.doi.org/10.1520/C0039. [32] J. Poole, Hydration Study of Cementitious Materials Using Semi-adiabatic Calorimetry, The University of Texas at Austin, Austin, TX, 2007. [33] J.L. Poole, K.A. Riding, K.J. Folliard, M.C.G. Juenger, A.K. Schindler, Methods for calculating activation energy for Portland cement, ACI Mater. J. 104 (2007) 303. [34] B. Klemczak,  Maciej Batog, Heat of hydration of low clinker cements Part II-determination of apparent activation energy and validity of the equivalent age approach, J. Therm. Anal. Calorim. 123 (2019), doi:http://dx.doi.org/10.1007/s10973-015-4791-x (n.d.). [35] D.P. Bentz, Activation energies of high-volume fly ash ternary blends: hydration and setting, Cem. Concr. Compos. 53 (2014) 214–223, doi:http://dx.doi. org/10.1016/j.cemconcomp.2014.06.018. [36] L.D. ’aloia, G. Chanvillard, Determining the “‘apparent’” activation energy of concrete E a —numerical simulations of the heat of hydration of cement, Cem. Concr. Res. 32 (2002) 1277–1289. [37] Q. Xu, J. Hu, J.M. Ruiz, K. Wang, Z. Ge, Isothermal calorimetry tests and modeling of cement hydration parameters, Thermochim. Acta 499 (2010) 91–99, doi:http://dx.doi.org/10.1016/j.tca.2009.11.007. [38] G. De, Schutter, Hydration and temperature development of concrete made with blast-furnace slag cement, Cem. Concr. Res. 29 (1999) 143–149. [39] S.J. Barnett, M.N. Soutos, J.H. Bungey, S.G. Millard, The effect of ground granulated blast furnace slag on the strength development and adiabatic temperature rise of concrete mixes, in: R.K. Dhir, T.A. Harrison, M.D. Newlands (Eds.), Cem. Cobminations Durable Concr., Thomas Telford Publishing, London, Dundee, Scotland, 2005, pp. 165–172. [40] H. Moon, S. Ramanathan, P. Suraneni, C.S. Shon, C.J. Lee, C.W. Chung, Revisiting the effect of slag in reducing heat of hydration in concrete in comparison to other supplementary cementitious materials, Materials (Basel). 11 (2018), doi:http://dx.doi.org/10.3390/ma11101847. [41] Y. Sargam, K. Wang, J.E. Alleman, Effect of modern concrete materials on thermal conductivity, J. Mater. Civ. Eng. (2019) In Press.