HYDRATION 2-1

2 HYDRATION

2.1 Introduction

Hydration is defined as the chemical absorption of water into a substance, a process by which heat
is generated (hydration heat). The setting of concrete (which can be considered as a transition from
liquid to solid phase) is the most relevant example for the hydration process in the engineering
world.
The effects of the hydration process can be separated into different physical parts, where the thermal
and mechanical parts are the most relevant. The implementation of hydration models in FLAC 3D
follows this separation, as the hydration heat generation and heat transfer are dealt with in thermal
models, material hardening and strength development are implemented as constitutive models of
mechanical behavior. The hydration model is based on a procedure that considers empirical rules,
theoretical considerations and practical experiences (Onken and Rostsy 1995).
The coupling parameter between the different parts of the constitutive model are the hydration grade
and the concrete age te [s]. The hydration grade is defined as the ratio between the accumulated
hydration heat up to the current time Q [J/m3 ] and the ultimate hydration heat generated until total
completion Qmax [J/m3 ].

Q(te )
(te ) = (2.1)
Qmax

The time-dependent evolution of the hydration grade depends on the temperature of the concrete.
Thus, the heat flow above the boundaries and the heat conductivity within the hydrating material
are of specific importance for this evolution as well as the initial temperature. Lower temperatures
lead to a longer process with lower hydration heat generation, whereas higher temperatures lead
to a shorter process with higher hydration heat generation. An equivalent age te is introduced as a
measure of time that is independent of these real-time effects.

 
 t EA 1
R TRef T1
te = e d (2.2)
0

where R is the universal gas constant (8.314 J/K/mol), and EA is the activation energy [J/mol]. The
description of the thermal behavior is related to the reference temperature TRef [K], which is set to
20 C in Onken and Rostsy (1995); T is the materials temperature [K].
A thermal hydration model, th hydration, is implemented in FLAC 3D. For simulating a hydra-
tion process, a mechanical constitutive model that can adjust the mechanical material properties
corresponding to the hydration grade (or equivalent concrete age) is required.
During the hydration process, the values of elastic material parameters can vary over several orders
of magnitude. Accordingly, the gridpoint masses have to be adjusted for numerical stability, in

FLAC 3D Version 4.0

2-2 Thermal Analysis

small-strain mode as well as large-strain mode. The frequency of the update can be set by the user
with the SET geom rep value command.

FLAC 3D Version 4.0

HYDRATION 2-3

2.2 Thermal Hydration Model (th hydration)

It is primarily the hydration of the binding material (cement, in the case of concrete), that leads to
the hydration process.

Q(te ) = CQCe (te ) Qmax = CQCe,max (2.3)

where C is the content of binding material [kg/m3 ], QCe is the amount of hydration heat from
binding material produced up to the current time [J/kg], QCe,max is the ultimate hydration heat
generated until total completion [J/kg].
With respect to Eqs. (2.1) and (2.3), the current heat release is

e ) = CQ
q(te ) = Q(t Ce,max (2.4)

Without any restriction, the hydration process would run over the full range of the hydration grade,
which is = [0, 1]. The heat release is always active, and at the end, a heat in the amount of Qmax
(see Eq. (2.1)) has been generated. However, this process can be restricted by physical or numerical
reasons. Therefore, two independent limitations are available, one depending on temperature and
the other on hydration grade:

 
q(te ) T Tmax,q q(te ) max,q
q(te ) = or q(te ) = (2.5)
0 T Tmax,q 0 max,q

The behavior of hydration grade, depending on equivalent age, Eq. (2.2), is an exponential descrip-
tion with the two Jonasson material parameters, b [] and t1 [1/s]:

  b
1n 1+ tte
(te ) = e 1 (2.6)

The equivalent age is given in Eq. (2.2). As shown in Eq. (2.2), the only material parameter of
equivalent age is the activation energy.
Two versions of activation energy are currently available: a generalized model and a German
model. The generalized version for the activation energy as a function of the temperature, EA is
implemented as

EA,1 EA,2
EA (T ) = EA,2 + T T0,EA
(2.7)
1+e dTEA

FLAC 3D Version 4.0

2-4 Thermal Analysis

This defines the activation energy between the two limits, EA,1 [J/mol] and EA,2 [J/mol]. T0,EA
[K] gives the temperature where the difference between the limits is divided in half, and dTEA [K]
characterizes the S-shape of the function.
For the German concrete model (Onken and Rostsy 1995), values for the activation energy are
given in two ranges of temperature:


EA,1 + dEA,T (T0,EA T ) T T0,EA
EA (T ) = (2.8)
EA,1 T > T0,EA

Typical parameters are EA,1 = 30 kJ/mol, dEA,T = 1.47 kJ/mol/K and T0,EA = 293 K.
Both versions of the activation energy are implemented, and can be selected by the parameter iLaw
(0: German concrete; 1: general model).
The heat transfer is assumed to be isotropic, with the following functions of specific heat, cp , and
thermal conductivity, :

cp = cp,1 (1 + dcp, ) (1 + dcp,T T ) (2.9)

= 1 (1 + d ) (1 + dT T ) (2.10)

and constant coefficient of thermal expansion.

FLAC 3D Version 4.0

HYDRATION 2-5

2.3 A Modified Drucker-Prager Model for Hydration (m hyd dp)

The mechanical aspects of hydration in FLAC 3D are handled by a modified Drucker-Prager consti-
tutive model where elastic and strength properties depend on the hydration grade, (Hinze 1987).
Due to a dormant phase, the evolution of strength and stiffness starts with some delay. This is taken
into account by the minimum degree of hydration, 0 . This value marks the transition between
the suspension and solid-state behavior. Beyond 0 , strength and stiffness do not always depend
linearly on hydration grade. Thus, a relationship is introduced, based on the idea of a multiplicative
split of the final values of material properties and the degree of hydration, including the minimum
degree of hydration according to the power law in Eq. (2.11):

 n
0
f unc() = (2.11)
1 0

With this formulation, the actual Youngs modulus, E, during the hydration process is

 a
0
E() = Ecte (2.12)
1 0

where Ecte is the Youngs modulus after complete hydration [Pa], and a is the power exponent.
The actual uniaxial compressive strength c and the uniaxial strength t also depend on the function
in Eq. (2.11).

 3/2
fcte 0
c () = 0, 85 (2.13)
c 1 0


0
t () = fcte (2.14)
1 0

where fcte is the uniaxial strength [Pa] after total completion of the hydration process, and c is a
material parameter [Pa].
The yield criterion in the Drucker-Prager constitutive model (see Section 2.5.1 in Theory and
Background) is

0= +q k (2.15)

where q and k are material parameters, and and are stress invariants. q and k can be derived
from the actual uniaxial compressive and tensile strengths, c and t .

FLAC 3D Version 4.0

2-6 Thermal Analysis

3 (c t )
q= (2.16)
c + t

2 c t
k= (2.17)
3 (c + t )

FLAC 3D Version 4.0

HYDRATION 2-7

2.4 FLAC 3D Commands

All commands have the same structure as those in the standard version of FLAC 3D. No new
commands are required, but additional keywords are used with existing commands. The new
keywords for each command are described as follows:
MODEL keyword <range . . . >
This command associates a constitutive model with an area of the grid corresponding
to a range of zones. See Section 1.1.3 in the Command Reference for an explanation
of the range phrase.
During the calculation, zones will behave according to a model corresponding to
one of the following keywords:
th hydration thermal module of hydration model
m hyd dp mechanical module of hydration model, derived from Drucker-Prager
PROPERTY keyword value <range . . . >
This command assigns properties for models identified by the MODEL command.
Thermal Model MODEL th hydration
(1) q cemax maximum hydration heat from binding material, QCe,max
(2) cement binding material, C
(3) b const Jonasson parameter, b
(4) t1 const Jonasson parameter, t1
(5) gas const universal gas constant, R
(6) T 0Ref reference temperature, TRef (default = 293K)
(7) E A1 material parameter of activation energy, EA,1
(8) E A2 material parameter of activation energy, EA,2
(9) T 0EA material parameter of activation energy, T0,EA
(10) dT EA material parameter of activation energy, dTEA
(11) dE AT material parameter of activation energy, dEAT
(12) a maxq maximum hydration grade for heat release, max,q
(13) T maxq maximum temperature for heat release, Tmax,q
(14) C p1 material parameter of specific heat, cp,1
(15) dC pa material parameter of specific heat, dcp,
(16) dC pT material parameter of specific heat, dcp,T
(17) lambda 1 material parameter of conductivity, 1
(18) dlambda a material parameter of conductivity, d

FLAC 3D Version 4.0

2-8 Thermal Analysis

(19) dlambda T material parameter of conductivity, dT

(20) thexp thermal expansion coefficient
(21) iLaw law for activation energy:
iLaw = 0 German concrete (Eq. (2.8)) default
iLaw = 1 generalized form (Eq. (2.9))
read only (these properties change during the hydration process)
(22) conductivity conductivity,
(23) spec heat specific heat, cp
(24) alpha hydration grade,
(25) e activate activation energy, EA
(26) t concrete effective concrete age, te
(27) t therm real time period since activation of th hydration
(28) heat hydration heat rate, q
Mechanical Model MODEL m hyd dp
(1) cte bulk bulk modulus for = 1
(2) cte shear shear modulus for = 1
(3) cte tension tensile strength for = 1, fcte
(4) cte young Youngs modulus for = 1, Ecte
(5) cte alpha minimum hydration grade, 0
(6) a const material parameter, a
(7) c const material parameter, c
(8) dAlpha min minimum difference of ( 0 )min
(9) poisson Poissons ratio,
read only (these properties change during the hydration process)
(10) young Youngs modulus, E
(11) bulk bulk modulus, K
(12) shear shear modulus, G
(13) tension tensile strength limit, t
(14) compression compressive strength limit, C
(15) qvol Drucker-Prager material parameter, q
(16) kshear Drucker-Prager material parameter, k
(17) qdil Drucker-Prager material parameter, q

FLAC 3D Version 4.0

HYDRATION 2-9

2.5 Example Problems

2.5.1 Concrete Inclusion in an Elastic Medium

This example consists of a concrete inclusion inside an elastic and thermal isotropic material. The
model has a size of 10 m 1 m 10 m, and is fixed in the y-direction (Figure 2.1). Roller
boundary conditions are applied at the model boundaries. The material properties are listed in
Table 2.1 (elastic frame) and Table 2.2 (concrete inclusion). The model was run for a thermal time
of 2 days.
Figures 2.2 to 2.6 show the evolution of various parameters during the hydration process: gridpoint
temperatures at five points (Figure 2.2); hydration grade (Figure 2.3); tensile and compressive
strength (Figure 2.4); elastic parameters (Figure 2.5); and the generated hydration heat (Figure 2.6).

Figure 2.1 FLAC 3D model for the concrete inclusion test

Table 2.1 Material properties for the elastic frame

Bulk modulus (K) 1000 MPa
Shear modulus (G) 700 MPa
Specific heat (CP ) 0.2 J/kg/K
Thermal conductivity (k) 20.0 W/m/K
Linear thermal expansion coefficient (t ) 104 C1

FLAC 3D Version 4.0

2 - 10 Thermal Analysis

Table 2.2 Material properties for the concrete inclusion

Maximum amount of generated heat (Qmax Ce ) 105 J/kg
Cement concentration (C) 330 kg/m3
Material parameter (b) 1.114
Material parameter (t1) 7.2 104 [s]
Universal gas constant (R) 8.314 J/mol
Activation energy (EA,1 ) 33.5 J/mol
(dEA,T ) 1.47 kJ/mol/K
Specific heat (CP,1 ) 0.2 J/kg/K
Thermal conductivity (1 ) 2.0 W/m/K
Linear thermal expansion coefficient (t ) 105 C1
Specific parameter for cement (0 ) 0.20
Youngs modulus after complete hydration (Ecte ) 1000 MPa
Material parameter (c) 0.4
Material parameter (a) 0.6
Minimum value for ( 0 ) 104

Figure 2.2 Evolution of gridpoint temperatures for the concrete inclusion

test as a function of the concrete age

FLAC 3D Version 4.0

HYDRATION 2 - 11

Figure 2.3 Evolution of the hydration grade for the center zone of the con-
crete inclusion as a function of the concrete age

Figure 2.4 Evolution of the tensile and compressive strength for the center
zone of the concrete inclusion as a function of the concrete age

FLAC 3D Version 4.0

2 - 12 Thermal Analysis

Figure 2.5 Evolution of the elastic parameters for the center zone of the
concrete inclusion as a function of the concrete age

Figure 2.6 Evolution of the hydration heat for the center zone of the concrete
inclusion as a function of the concrete age

FLAC 3D Version 4.0

HYDRATION 2 - 13

Example 2.1 HYDRAT-EXAMPLE-1.F3DAT

;--------------------------------------------------------------
; Concrete inclusion in an elastic, thermal isotropic medium
;--------------------------------------------------------------

new
title
Concrete inclusion in an elastic, thermal isotropic medium

config thermal
;
gen zone brick size 11 1 11
group concrete range x=4,7 z=4,7

model elast th_iso

model m_hyd_dp th_hydration range group concrete
;
prop dens 2000 bulk 1e3 shear 0.7e3
prop cond 20.0 thex 1e-4 spec_heat 0.2

; Modified Drucker-Prager
prop cte_bulk 0.98e3 cte_shear 0.50e3 range group concrete
prop c_const 0.4 a_const 0.6 range group concrete
prop cte_alpha 0.20 cte_tension 2.0 range group concrete
prop dalpha_min 1e-4 range group concrete

; Thermal hydration
prop cond 2.0 thex 1e-5 spec_heat=0.2 range group concrete
prop cement 330 b_const -1.114 t1_const 7.2e4 range group concrete
prop gas_const 8.314 T_0Ref=293 a_maxq=1.0 t_maxq=1e4 range group concrete
prop E_a1=33.5 dT_Ea=1.47 T_0EA=273 iLaw=0 C_p1=0.2 range group concrete
prop q_cemax=1e5 lambda_1=2.0 range group concrete

ini temp 300

fix temp 300 ra x -0.1 0.1
fix temp 300 ra x 10.9 11.1
fix temp 300 ra z 10.9 11.1
fix temp 300 ra z -0.1 0.1

fix x ra x -0.1 0.1

fix x ra x 10.9 11.1
fix y
fix z ra z -0.1 0.1
fix z ra z 10.9 11.1

FLAC 3D Version 4.0

2 - 14 Thermal Analysis

hist id 1 gp temp 1 0 5
hist id 2 gp temp 2 0 5
hist id 3 gp temp 3 0 5
hist id 4 gp temp 4 0 5
hist id 5 gp temp 5 0 5

def hist_setup
hzp = z_nearall(6.5,0.5,6.5)
end
@hist_setup

def hyd_hist
h_alpha = z_prop(hzp,alpha)
h_q = z_prop(hzp,heat)
h_te = z_prop(hzp,t_concrete)
h_tension = z_prop(hzp,tension)
h_comp = z_prop(hzp,compression)
h_young = z_prop(hzp,young)
h_bulk = z_prop(hzp,bulk)
h_shear = z_prop(hzp,shear)
end

hist id 6 @hyd_hist
hist id 7 @h_alpha
hist id 8 @h_q
hist id 9 @h_te
hist id 10 @h_tension
hist id 11 @h_comp
hist id 12 @h_young
hist id 13 @h_bulk
hist id 14 @h_shear

set geom_rep 1
hist nstep 100

solve age 172800

save hyd_exp1_2d.sav

return

FLAC 3D Version 4.0

HYDRATION 2 - 15

2.5.2 Concrete Wall on Elastic Base Plate

The numerical model (Figure 2.7) used here is based on a model that was used by Onken and
Rostsy (1995) to compare numerical results and temperature measurements. It consists of a new
concrete wall (10 1 40 elements) footed on a base plate of old concrete (10 1 5 elements).
Since the material parameters and boundary conditions are not available from Onken and Rostsy
(1995), the material parameters from Example 2.1 were used, as were typical temperature boundary
conditions.
For the concrete wall, the thermal hydration model for German concrete (th hydration) and the
modified Drucker-Prager with hydration (m hyd dp), the base plate is considered elastic and thermal
isotropic.
The model was under summer temperature conditions. The temperatures at the wall boundary
were fixed to air temperature; the temperatures at the base plate were fixed to the initial ground
temperature:
initial wall temperature 296 K
wall boundary temperature fixed at 293 K
initial base plate temperature 285 K
base plate boundary temperature fixed at 285 K
The numerical results are displayed in Figures 2.8 to 2.14: the temperature distribution inside the
model after 1 and 3 days (Figures 2.8 and 2.9); evolution of gridpoint temperatures (Figure 2.10);
evolution of the hydration grade (Figure 2.11); evolution of the strength parameters (Figure 2.12);
evolution of the elastic parameters (Figure 2.13); and evolution of the hydration heat (Figure 2.14).
The results are qualitatively in good agreement with the results given by Onken and Rostsy (1995).
(A more detailed comparison is not available due to limited access to material parameter and
boundary conditions.)

FLAC 3D Version 4.0

2 - 16 Thermal Analysis

Figure 2.7 Model of the concrete wall on an elastic base plate

FLAC 3D Version 4.0

HYDRATION 2 - 17

Figure 2.8 Temperature distribution after 1 day (concrete age)

Figure 2.9 Temperature distribution after 3 days (concrete age)

FLAC 3D Version 4.0

2 - 18 Thermal Analysis

Figure 2.10 Evolution of gridpoint temperatures for the concrete inclusion

test as a function of the concrete age

Figure 2.11 Evolution of the hydration grade as a function of the concrete

age

FLAC 3D Version 4.0

HYDRATION 2 - 19

Figure 2.12 Evolution of the tensile and compressive strengths as a function

of the concrete age

Figure 2.13 Evolution of the elastic parameters as a function of the concrete

age

FLAC 3D Version 4.0

2 - 20 Thermal Analysis

Figure 2.14 Evolution of the generated hydration heat as a function of the

concrete age

Example 2.2 HYDRAT-EXAMPLE-2.F3DAT

;--------------------------------------
; Concrete wall on elastic base plate
;--------------------------------------
new
title
Concrete wall on elastic base plate - summer temperature conditions

config thermal
;
gen zone brick p0=(-0.5,0.0,0.0) p1=(0.5,0.0,0.0) p2=(-0.5,1.0,0.0) &
p3=(-0.5,0.0,10.0) size = 10,5,100
gen zone brick p0=(-2.5,0.0,-1.0) p1=(2.5,0.0,-1.0) p2=(-2.5,1.0,-1.0) &
p3=(-2.5,0.0,0.0) size = 20,5,5
attach face range z -0.01 0.01
;
group concrete range z 0 100
group base_plate range z -100 0

model elast th_iso range group base_plate

model m_hyd_dp th_hydration range group concrete

FLAC 3D Version 4.0

HYDRATION 2 - 21

prop dens 2000 bulk 1e3 shear 0.7e3

prop cond 20. thex 1e-4 spec_heat 0.2
prop cond 2. thex 1e-5 range group concrete

; Modified Drucker-Prager
prop cte_bulk 0.98e3 cte_shear 0.50e3 range group concrete
prop c_const 0.4 a_const 0.6 range group concrete
prop cte_alpha 0.20 cte_tension 2.0 range group concrete
prop dalpha_min 1e-4 range group concrete

; Thermal hydration
prop cond 2.0 thex 1e-5 spec_heat=0.2 range group concrete
prop cement 330 b_const -1.114 t1_const 7.2e4 range group concrete
prop gas_const 8.314 T_0Ref=293 a_maxq=1.0 t_maxq=1e4 range group concrete
prop E_a1=33.5 dT_Ea=1.47 T_0EA=273 iLaw=0 C_p1=0.2 range group concrete
prop q_cemax=1e5 lambda_1=2.0 range group concrete

ini temp 293

fix temp 293 ra x -0.51,-0.49
fix temp 293 ra x 0.49, 0.51
fix temp 293 ra z 9.9,10.1

ini temp 285 range group base_plate

fix temp 285 ra z -1.01 -0.99
fix temp 285 ra x -10.0 -0.49 z -0.01 0.01
fix temp 285 ra x 0.49 10.0 z -0.01 0.01

fix x y z ra z -0.9 -1.1

fix x ra x -2.6 -2.4
fix x ra x 2.4 2.6

hist gp temp 0 0.5 1

hist gp temp 0 0.5 2
hist gp temp 0 0.5 3
hist gp temp 0 0.5 4
hist gp temp 0 0.5 5

def hist_setup
hzp = z_near(0,0,5)
end
@hist_setup

def hyd_hist
h_alpha = z_prop(hzp,alpha)
h_q = z_prop(hzp,heat)

FLAC 3D Version 4.0

2 - 22 Thermal Analysis

h_te = z_prop(hzp,t_concrete)
h_tension = z_prop(hzp,tension)
h_comp = z_prop(hzp,compression)
h_young = z_prop(hzp,young)
h_bulk = z_prop(hzp,bulk)
h_shear = z_prop(hzp,shear)
h_time = thtime
end

hist @hyd_hist
hist @h_alpha @h_q @h_te
hist @h_tension @h_comp
hist @h_young @h_bulk @h_shear
hist @h_time

set geom_rep = 10
solve age 86400
save hyd_exp2_1d.f3sav
solve age 259200
save hyd_exp2_3d.f3sav
return

FLAC 3D Version 4.0

HYDRATION 2 - 23

2.6 References

Engelhardt, J., et al. A constitutive law of salt concrete used for closure of an LILW-repository,
in Proceedings of ICEM 03: The 9th International Conference on Radioactive Waste Manage-
ment and Environmental Remediation (September 21 25, 2003, Examination School, Oxford,
England). icem03-4570.
Hinze, D. Zur Beurteilung des phsikalischen nicht-linearen Betonverhaltens bei mehrachsigem
Spannungszustand mit Hilfe differenzeiller Stoffgesetze unter Anwendung der FEM, Thesis,
Hochschule fr Architektur und Bauwesen, Weimar, 1987.
Onken, P., and F. Rostsy. Wirksame Betonzugfestigkeit im Bauwerk bei frh einsetzendem
Temperaturzwang, DAfStb Heft 449. Berlin: Beuth-Verlag, 1995.

FLAC 3D Version 4.0

2 - 24 Thermal Analysis

FLAC 3D Version 4.0