Proceedings of the International rilem conference Materials, Systems and Structures in Civil Engineering 2016

73

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

2.4 Durability requirements 

The preliminary concrete proportioning sets the water to binder ratio (w/b) around 0.35 for 

structural concretes in HZM project, and both fly-ash and slag are used in binder. After 

model-based design through (1), (2) for chloride ingress and (3) for carbonation, it is found 

that the thickness and quality of concrete cover is controlled by the chloride ingress process, 

and the final requirements from model-based approach are given in Table 1.  

 

Table 1: Model-based durability design results for concrete structures in HZM project (design 

life t

SL

=50, 120 years,  =1.3)[6].  

Element Exposure 

t

SL

 

(year) 

DLS 

x

d,

nom

 

(mm) 

Design 

value D

nssd

  

(10

-12

 m

2

/s) 

Control 

value D

nssm

  

(10

-12

 m

2

/s) 

28d 56d 28d 56d 

Box girder (exterior) 

Salt 

fog  120  (a)  45  3.0 2.0 6.0 4.0 

Splashing 120  (a)  80  3.0 2.0 6.0 4.0 

Box girder (interior) 

CO

2

 

120 

(a) 35  - - - - 

Salt 

fog  120  (a)  45  3.0 2.0 6.0 4.0 

Pier, Pylon (exterior) 

Salt 

fog  120  (a)  50  3.5 2.2 7.0 4.5 

Splashing 120  (a)  85  3.5 2.2 7.0 4.5 

Pier (interior) 

CO

2

 

120 

(a) 35  - - - - 

Salt 

fog  120  (a)  50  3.5 2.2 7.0 4.5 

Bearing platforms 

Splashing 120  (a)  85  3.5 2.2 7.0 4.5 

Immerged 120  (a)  65  3.5 2.2 7.0 4.5 

Bored 

hole 

pile 

Immerged 120  (a)  65  3.5 2.2 7.0 4.5 

RC facilities 

(artificial islands) 

Splashing  50  (a)  60  3.5 2.2 7.0 4.5 

CO

2

 

50 (a) 20  - - - - 

Tunnel tube 

(exterior) 

Splashing 120  (a)  80  3.5 2.2 7.0 4.5 

Tunnel tube (interior) 

Salt 

fog  120  (a)  50  3.5 2.2 7.0 4.5 

CO

2

 

120 

(a) 30  - - - - 

 

On the basis of the design results for chloride ingress, the values of required cover thickness 

and chloride diffusion coefficient are obtained, modulated and given in Table 1 for the main 

concrete elements. These results provide the basis of durability requirements for RC elements 

in HZM project. The rapid chloride migration (RCM) test conform to NT Build 492 [8] is 

used to measure the chloride diffusivity in construction phase. However, the design values 

from (1) cannot be used directly for the quality control by RCM method because the D

Cl

0

 in 

(1) is the apparent chloride diffusion coefficient regressed from in-situ exposure tests at a 

given concrete age while RCM measures a short term migration coefficient under an electrical 

field. The former is a non-steady state diffusion (NSSD) coefficient while the latter is a non-

steady state migration (NSSM) coefficient. Thus it is necessary to establish a relationship 

between the NSSD design value and the NSSM quality control value. To this aim, concrete 

materials in the Zhanjiang Exposure Station are reconstituted in laboratory, and the NSSM 

coefficients are measured by RCM method at 28d and 56d. Using the correlation between 

---

74

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

these two groups of coefficients, the design values are converted to the corresponding RCM 

values for quality control in Table 1. 

 

 

3.  Quality control in construction 

 

3.1 Control criteria 

The acceptance criteria are set up respectively for the compressive strength and the chloride 

diffusivity. For the compressive strength, both the average and the minimum values for a 

given specimen group are controlled. For a group containing 5 or more specimens, the 

acceptance criterion is expressed as, 

 

fcu

fcu

cu,k

cu,min

cu,k

0

, 

m

s

f

f

f

C

  

(4) 

 

Here,  m

fcu

,s

fcu 

are respectively the average value and standard deviation of compressive 

strength of specimens, f

cu,k

 and f

cu,min

 are respectively the characteristic value for required 

compressive strength (strength grade) and the minimum strength of specimens, 

0

 is the 

assumed standard deviation of compressive strength. The coefficient C depends on the 

number of specimens: C=0.7 (5-9), 0.9 (10-19), 1.0 (>19). For specimen groups containing 2-

4 specimens, the following criteria apply, 

 

fcu

cu,k

0

cu,min

cu,k

0

, 0.5

m

f

f

f

  

(5) 

 

The acceptance criteria for chloride diffusivity from RCM method are formulated on the 

similar basis. According to the technical document [9], the following requirements apply to 

specimen groups containing 5 or more specimens, 

 

 

Cl

Cl

nssm,k

Cl,max

nssm,k

Cl

, 

m

s

D

D

D

C

 (6) 

 

Here, m

Cl

, s

Cl

 are respectively the average value and standard deviation of chloride diffusivity 

of specimens, D

nssm,k

 and D

Cl,max

 are respectively the characteristic values for required RCM 

chloride diffusivity and the maximum value of diffusivity, 

Cl

 is the standard deviation of 

chloride diffusivity regressed from laboratory specimens with enough number. The 

coefficient C takes the same value as (4) in terms of specimen number. For specimen groups 

containing 2-4 specimens, the following criteria apply, 

 

Cl

nssm,k

Cl

Cl,max

nssm,k

Cl

, 0.5

m

D

D

D

   

(7) 

 

Note that the design results D

nssm

 in Table 1 should be used as the characteristics value D

nssm,k

 

in (6) and (7). 

 

3.2 Chloride diffusivity 

The in-situ data of chloride diffusivity of structural concretes were collected from the on-site 

laboratories for different concrete elements in different contract bids. The chloride diffusion 

---