International Commission on Large Dams (ICOLD) - The Physical Properties of Hardened Conventional concrete in Dams

Подождите немного. Документ загружается.

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Section 8 (Frost resistance)

As submitted for ICOLD review, march 2008 Section 8-18

8.7 REFERENCES

[8.1] Fagerlund, G., RILEM, “The critical degree of saturation method of assessing the

freeze/thaw resistance of concrete”, Materials and constructions, p. 217-229, July-

August 1977.

[8.2] Neville, A.M., “Properties of Concrete”, Sir Isaac Pitman & Sons LTD., p. 350,

1963.

[8.3] Berra M., “Manuale – Diagnosi e scelta dei materiali da ripristino per le strutture in

calcestruzzo degradate dall’azione dei cicli di gelo e disgelo”, Enel – Polo Idraulico e

Strutturale Internal Report n° 5944, 1999. (in Italian).

[8.4] Cordon, W. A. “Freezing and Thawing of Concrete Mechanisms and Control ”, ACI

Monograph n° 3, pag. 28, 1966.

[8.5] Metha, P. K. “Concrete structure, properties and materials”, Prentice-Hall, Inc.,

Englewood Cliffs, New Jersey 070632, 1986.

[8.6] CEB (Comité Euro-International du Béton), “Draft CEB-Guide to durable concrete

structures”, Bulletin d’information n° 166, 1985.

[8.7] CEB (Comité Euro-International du Béton), “Durability of concrete structures-

State-of-the-art report”, Bulletin d’information n° 148, 1982.

[8.8] Pigeon, M., Prévost, J. And Simard, J.M., “Freeze-Tauw durability versus freezing

rate”, Aci Journal, p. 684-690, September-October 1985.

[8.9] Stark D.C.,”Effect of Vibration on the Air-Void System and Freeze-Thaw Durability

of Concrete”, Portland Cement Association (RD092.01T) 1886.

[8.10] Tognon, G. and Cangiano, S., “Sulla gelività del calcestruzzo”, Manutenzione,

riparazione e durabilità delle strutture in cemento armato – Collana di Ingegneria

Strutturale n°6 – International Centre for Mechanical Sciences CISM, Udine, p.66-80,

1986. (In Italian).

[8.11] American Concrete Institute, ACI Building Code 318,

[8.12] Hess R. E., "Künstliche Luftporen im Beton" [Artificial air entrained in Concrete],

Gazettenverlag Zurich (Switzerland).

[8.13] U.S. Bureau of Reclamation, "Concrete Manual », 8

edition, 1975

[8.14] European Standard EN 206, “Concrete, Performance, Production, Placing and

Compacting Criteria”, 2002.

[8.15] Macinis, C., Lau, e.C., « Maximum Aggregate Size effect on frost resistance of

concrete, ACI Journal, pag. 144-149, 1968.

[8.16] Kennedy T.B.,”Investigations of pozzolanic and other materials to replace part of

the Portland cement in mass concrete dams”, ICOLD Proc. 6

Congress, New York,

1958, Q.23, R120.

[8.17] Ammann A.,”Research on the influence of pozzolan and other admixtures on the

frost-resistance of concrete”, ICOLD Proc. 6

Congress, New York, 1958, Q.23, R105.

[8.18] Gebler S.H., Klieger P.,”Effect of Fly Ash on the Durability of Air-Entrained

Concrete”, Portland Cement Association (RD090.01T), ACI Publication SP-91, 1986.

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Section 8 (Frost resistance)

As submitted for ICOLD review, march 2008 Section 8-19

[8.19] ACI 232-1R-94, “Use of Natural Pozzolans in Concrete”.

[8.20]. ACI 233R-03, “Slag Cement in Concrete and Mortar”.

[8.21] Kreuzer, H. Personal communication, 2001

[8.22] Verbeck, G., “Non destructive tests for deterioration of concrete specimens

subjected to frost action”, Fifth Congress on Large Dams, Q.19, R.70, paper n°3,

1955.

[8.23] Roulet, C.A., “Effet du gel”, Durabilitè du béton armé – Origin des dégats,

Prévention, Assainissement – SIA Documentation n°89, Lausanne, p.23-27, 1985.

[8.24] Malhotra, V.M., “Mechanical properties and freezing and thawing resistance of

non-air-entrained, air entrained, and air-entrained superplasticized concrete using

ASTM test C 666, Procedures A and B”, Cement, Concrete and Aggregates, p. 3-23,

Summer 1982.

[8.25] Setzer, M.J., RILEM, “RILEM 117-FDC: Freze-Thaw and deicing resistance of

concrete – Report”, Materials and constructions, p. 3-6, Supplement March 1997.

[8.26] Giulietti, P., Tognon, G.P. “Il metodo del grado di saturazione critico per la prova

della resistenza al gelo del calcestruzzo, a confronto con il metodo UNI 7087-72: prime

considerazioni sulla base di risultati sperimentali”, AITEC Conference – La durabilità

delle opere in calcestruzzo, Padova, p. 231-238, 8-9 October 1987 (in Italian).

[8.27] Fagerlund, G., RILEM, “The critical degree of saturation method of assessing the

freeze/thaw resistance of concrete”, Materials and constructions, p. 217-229, July-

August 1977.

[8.28] Orchard, D.F., “Concrete Technology Vol.1 Properties of Materials”, Applied

Science, London , 1979.

[8.29] American Concrete Institute, “Guide to Durable Concrete – ACI 201.2R-77”, ACI

Manual of Concrete Practice, Part 1, Detroit, MI, USA, 1991.

[8.30] Bitterli K. et al., “Zusammenhänge zwischen Betonporosität und

Frostbeständigkeit” (Relation between concrete porosity and frost resistance),

Schweizer Ingehieur & Architekt, 25/85, 628-636.

[8.31] Hulshizer, A.J., “Air-Entraining Control or Consequences”, Concrete International,

p. 38-40, July 1997.

[8.32] Concrete Society, “Analysis of Hardened concrete”, Concrete Society Technical

Report n° 32, London, 1989.

[8.33] Lee, C.R., “Surface temperatures in concrete dams subjected to frost”, Fifth

ICOLD Congress on Large Dams, Q.19, R.70, paper n°6, 1955.

[8.34] Kokubu, M., Kobayashi, M., “Long term observations on frost damage to da

concrete using large test blocks”, 20

ICOLD Congress on Large Dams, Q.78, R.61,

Beijing, 2000.

[8.35] Ihii, K., Egawa, K., Tsutsumi. T., Noguchi, H., “Prediction of concrete

deterioration by cyclic freezing and thawing”, Concrete library of JSCE, n° 31, June

1998.

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 1

A FRACTURE ENERGY: APPLICATION FOR DAMS

A FRACTURE ENERGY: APPLICATION FOR DAMS............................. 1

A.1 GENERAL ...........................................................................................................1

A.2 FRACTURE PARAMETERS ...............................................................................2

A.3 DETERMINATION OF FRACTURE PARAMETERS...........................................4

A.4 FACTORS INFLUENCING THE FRACTURE ENERGY .....................................6

A.5 APPLICABILITY OF LINEAR ELASTIC FRACTURE MECHANICS (LEFM) ...10

A.6 REFERENCES ..................................................................................................11

A.1 GENERAL

Surface cracks are among the most common flaws in many practical civil engineering

structures. In concrete dams, because of the joints generated by a stepwise successive

construction of layered concrete blocks, and because of thermal operating gradients,

cracks could have relevant dimensions from the beginning of their life [A.1]. Also, while

cracks in dams occur after the casting of concrete, already existing continuous or

discontinuous cracks in the rock may widen even more or new fractures may form.

Therefore, assuming that no existing dam is completely without cracks, the question

whether the safety of such dams is guaranteed appears justified. Compounding the

problem of overall dam deterioration even further, is that existing dams are also being

subjected to potentially higher loads due to an improved understanding of flood

statistics and earthquake loading. In response to the challenge of evaluating older

dams under higher flood levels, research was initiated world-wide to develop new

computational techniques. Those methods were initially based on the finite element

method but followed the conventional linear elastic approach.

More recently the concept of fracture mechanics of concrete has been investigated,

analytical models developed, experimental tests performed and material properties

studied [A.2] [A.3] [A.4] [A.5] [A.6] [A.7] [A.8] [A.9] [A.10] [A.11] [A.12] [A.13]. The US

Bureau of Reclamation, the Electric Power Research Institute, the University of Vienna

and the US Army Corps of Engineers have, among others, all recognised the potential

application of fracture mechanics to the analysis of cracked concrete dams [A.14]

[A.15] [A.16] [A.17] [A.18]. The latter, through an Engineering Technical Letter (1991)

[A.19], now requires a fracture mechanics investigation prior to rehabilitation of cracked

massive concrete structures. Moreover, it should be recognised that dams, as main

examples of unreinforced concrete structures, are indeed prime candidates for the

application of fracture mechanics, be it linear or nonlinear.

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 2

In the early days fracture mechanics of concrete borrowed extensively from the Linear

Elastic Fracture Mechanics (LEFM) of metal [A.20]. However, it was soon discovered

that, in concrete structures, LEFM may not necessarily be applicable. The reason for

this deviation for concrete behaviour is the development of a micro-cracked zone at the

crack tip (Fracture Process Zone or FPZ – Fig. A.1 a) where the material undergoes

strain-softening, thus where the stress normal to the crack plane decreases with

increasing strain.

The nonlinear behaviour in the FPZ could be expressed by a softening relationship

between the normal stress across the crack σ

versus the crack opening displacement

w, as was proposed by Hillerborg and co-workers [A.21]. The area under the σ

- w

curve represents the energy required to create one unit area of crack surface and is

defined as fracture energy of material G



(A.1)

where w

is the critical crack opening displacement beyond which no tensile stresses

can be transmitted across the crack.

The experimental determination of the specific fracture energy of concrete, G

, and the

accompanying shape of the softening diagram has lately been a topic of much research

[A.3] [A.22] [A.23].

A.2 FRACTURE PARAMETERS

The most important input parameter for nonlinear fracture mechanics calculations of

concrete is undoubtedly the stress-crack opening relationship (Fig. A.1 a) [A.24]; the

fracture energy is a derivative parameter of this relation (Eq. A.1). Different softening

laws for cracked concrete and a wide range of values for the specific fracture energy

have been experimentally determined [A.25] [A.26] [A.27] [A.28] [A.29] [A.30] [A.31]. A

simple law that provides a good approximation of the material behaviour is bilinear

softening law, in which the first steeper branch represents the large scale debonding, or

fracture, of aggregates, and the second branch represents the frictional pull-out of

aggregates which characterises the behaviour of larger cracks (Fig. A.1 c). The break

point coordinates of this law can be expressed as:

ct1

= c

= =

−

where f

is the concrete tensile strength. By assuming this law and given c

and t

concrete softening depends exclusively on the tensile strength and on the fracture

energy. Various researchers have proposed different values for coefficients t

and c

Wittmann and co-workers [A.28] proposed t

= 0.25 and c

= 0.75 (and thus c

= 5.0),

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 3

Brühwiler [A.19] proposed t

= 0.4 and c

= 0.8 (and thus c

= 3.0), while Petersson

proposed t

= 0.3 and c

= 0.8 (and thus c

= 3.6) [A.26].

Fig. A.1 - Stress-crack opening displacement curve and Fracture Process Zone (a);

linear (b) and bilinear (c) softening law.

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 4

Concrete dams are characterised by the presence of lift joints. When cracks propagate

along these joints, a bilinear softening law has no physical justification since there can

be no crack bridging by aggregates and a linear softening law (Fig. A.1 b) should be

adopted [A.32]:

= f

int

(1-

) = f

int

(1- f

int

) (A.2)

where f

int

is the tensile strength of the interface.

From elastic and fracture parameters, a magnitude having a dimension of length may

be defined under the name of characteristic length l

[A.21]:

l EG f

ch F

= /

(A.3)

where E is the Young’s modulus. The characteristic length characterises the brittleness

of the material: the shorter l

is, the more brittle material is [A.26].

A.3 DETERMINATION OF FRACTURE PARAMETERS

The uniaxial tensile test is probably the most fundamental fracture test; in fact when

performed under deformation control, this test directly yields a stress-deformation

relationship that includes all relevant fracture mechanics parameters (Fig. A.1 a). The

observed

-w curve could be considered as a material property if a uniform

deformation distribution over the fracture zone occurs during the entire test.

Unfortunately, this condition is not fulfilled in most of the reported tests, because of the

stress concentration around the notch or because of the rotation of the loading platens

[A.33] [A.34]; however, the obtained relationship can be considered a good

approximation of the actual material behaviour.

If the shape of the softening law is known (i.e. parameters w

and s

of a bilinear

softening law are known) as well as the tensile strength of the material, fracture energy

completely defines the behaviour of concrete in the fracture process zone. The tensile

strength could be determined by means of splitting or direct tension tests while fracture

energy could be determined by a bending test; the latter, because of its simplicity, has

been recommended by RILEM TC-50 FMC (work-of-fracture method) [A.35] [A.36]. In

this method, notched beams are loaded under three-point bending: the fracture energy

is assumed as the energy supplied to concrete specimen divided by the fracture area

(projected on a plane perpendicular to the tensile stress direction).

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 5

Fig. A.2 - Wedge splitting Specimen Shape (a); Alternative Shapes (b, c, d); Loading

devices (e, f,); Forces acting on the wedge (g) [A.10]; biaxial test equipment (h) [A.40].

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 6

Dam concrete, which is characterised by large aggregates, can be meaningfully tested

to determine the fracture energy only if large specimens are used. This is not easily

performable in the laboratory by using bending tests that would require a large space,

heavy specimens and a huge testing machines. As an alternative method, a wedge-

splitting test can be performed [A.37] [A.38] [A.39] (Fig. A.2 a, b, c); in this a simple

test, similar to the compact tension test used for metals, G

determination is based on

the same principle of the work-of-fracture method. This test is characterised by a

negligible self-weight and a larger ratio between the fracture area and the specimen

volume with respect to the beam tests; also, it can be easily cast at the construction site

and can also be obtained from drilled cores from existing structures (Fig. A.2 b, c, d).

The drawbacks of this test are related to the friction at the wedge loading and can be

reduced by hardening and carefully polishing the wedge surfaces (Fig. A.2 g) [A.10].

A.4 FACTORS INFLUENCING THE FRACTURE ENERGY

Factors known to influence G

include specimen type and size, aggregate type,

maximum grain size, concrete composition and strength, concrete moisture and loading

rate [A.3], [A.22], [A.41]. However, it is worth mentioning that several phenomena

influencing the determination of fracture energy are not yet clear [A.42]; as a

consequence, with regard to the relationships between G

and the various parameters

found in the literature, only general trends should be compared [A.27].

Different authors found that G

increases with the maximum aggregate size: when

varying it from 2 to 38 mm, they found values of G

from 50 to 200 J/m

[A.26], [A.28]

[A.36]. Experimental results on concrete with 76 mm maximum aggregate size provided

values of the fracture energy up to 280 J/m

[A.43] [A.44] He, S.,. Plesha, M.E.,

Rowlands, R.E., Bazant Z.P., “Fracture energy tests of dam concrete with rate and size

effects”, Dam Engineering, 3(2), p. 139-159, 1992. Eds., Aedificatio Publisher, pp.

1937-1948, ISBN 3-931681-24-6, 1998.

[A.45] Trunk B., and Wittmann F.H. “Experimental investigation into the size

dependence of fracture mechanics parameters”, International Conference on Fracture

Mechanics of Concrete and Concrete Structures, FRAMCOS3, Mihashi and Rokugo.

Values even larger should be expected from actual dam concrete with maximum

aggregate size of 100-150mm, even though no results are available so far. Trunk and

Wittmann [A.45] performed wedge splitting tests on specimens ranging between 50 and

3200 mm made of cement paste, mortar, normal concrete and dam concrete.

Experimental results showed that the fracture energy increases with increasing

maximum aggregate and specimen size. For large specimens the fracture energy

reached a limit value for a given aggregate size. However, for the investigated size

range, no limit value of the fracture energy as function of the aggregate size was found.

In experimental tests on cold jointed specimens performed by Saouma and co-workers

[A.43], values of the fracture energy variable between 80 and 140 J/m

were obtained.

Whenever the fracture surface runs through the aggregate particles, resulting in a

smooth crack path [A.26] and in a higher tensile strength: this makes concrete a more

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 7

brittle material (smaller value of l

) than concrete with a tortuous crack path [A.29]

[A.41].

Small fracture energy values can have two reasons:

 Brittle material with high tensile strength but less post-peak softening (smooth

crack path),

 more ductile material with low bondage between paste and coarse aggregates

(low peak with tortuous crack path)

For increasing values of the concrete compressive strength, and thus for decreasing

values of the water-cement ratio, fracture energy increases up to a maximum value and

then decreases for very high strength concrete, which is characterised by brittle failure

[A.26], [A.29] [A.28] [A.31].

A similar unsteady development is noticed when concrete is tested under biaxial

loading, by which one stress component is tensile (σ

), the other compressive (σ

Such a state prevails at the upstream heel of many arch dams (Fig. A.3).

Fig. A.3 - Virtual biaxial cube embedded in an arch dam

Biaxial test results indicate that G

decreases within the elastic range of compressive

loading and that at the onset of microcracking G

starts to increase again prior to

total material disintegration, see Fig. 2.15 in Section 2. This unsteady development

ensues from restraints in the expansion of the FPZ for the decreasing G

-branch, and

from aggregate interlock for the rising branch. Moreover, it was found (as could be

expected) that aggregate interlock increases with increasing Maximum Size Aggregate

(MSA), as shown in Fig. A.4.

ICOLD Bulletin: The Physical Properties of Hardened Conventional Concrete in Dams

Appendix A (Fracture mechanics: application for dams)

As submitted for ICOLD review, march 2008 Appendix A- 8

Fig. A.4 - Distribution of fracture energy in Gf1 (attributed to microcracking) and Gf2

(attributed to aggregate interlock) for different stages of  loading and 3 MSA's, see

also Fig. 2.19 in Section 2.

Fig. A.5 is an example how sensitive fracture energy is to the quality of bonding

between large aggregates and paste. Given normal-weight aggregates with a higher

strength than the paste, breakage through the aggregates indicates a good bondage

within the concrete texture and thus a higher G

-value as compared to breakage

around the large aggregate sizes resulting in a much lower G

-value.

Such findings may be useful for formulating damage models and for assessing post-

earthquake stability of massive concrete structures.