Do Concrete Materials Specifications Address Real Performance

1
Do Concrete Materials Specifications Address Real
Performance?

• David A. Lange
• University of Illinois at Urbana-Champaign

2
How do you spec concrete?

• 1930
• 6 bag mix
• 1970
• fc 3500 psi, 5 in slump
• And add some air entrainer
• 2010 ?

3
Is concrete that simple? How simple are your
expectations?

• Are we worried only about strength?
• What about
• Long-term durability
• Crack-free surfaces
• Perfect consolidation in conjested forms
• These cause more concrete to be replaced than
  structural failure!

4
Seeking the Holy Grail

• Admixtures developed in 1970s open the door to
  lower w/c and high strength
• Feasible high strength concrete moved from 6000
  psi to 16,000 psi
• Feasible w/c moved from 0.50 to 0.30
• Everybody loves high strength!

5
But there are trade-offs

• Low w/c ? high autogenous shrinkage
• High paste content ? greater vol change
• High E ? high stress for given strain
• High strength ? more brittle
• greater problems with cracking!

6
For example Early slab cracks

• Early age pavement cracking is a persistent
  problem
• Runway at Willard Airport (7/21/98)
• Early cracking within 18 hrs and additional
  cracking at 3-8 days

7
Concrete IS complex

• Properties change with time
• Microstructure changes with time
• Volume changes with time
• Self imposed stresses occur
• Plus, you are placing it in the field under
  variable weather conditions
• There are a million ways to make concrete for
  your desired workability, early strength,
  long-term performance

8
Overview

• Volume stability
• Internal RH and drying shrinkage
• Restrained stress
• Case Airport slab curling
• Case SCC segregation

9
Volume stability
Volume Change
Thermal
Shrinkage
Creep
External Influences
Autogenous shrinkage
External drying shrinkage
Basic creep
Drying creep
Heat release from hydration
Chemical shrinkage
Cement hydration
10
Chemical shrinkage
Ref PCA, Design Control of Concrete Mixtures
11
Self-dessication
Autogenous shrinkage
solid
Jensen Hansen, 2001
water
air (water vapor)
12
Chemical shrinkage drives autogenous shrinkage
Note The knee pt took place at only a 4
Ref Barcelo, 2000
The diversion of chemical and autogenous
shrinkage defines set
13
Measuring autogenous shrinkage

• Sometimes the easiest solution is also the best

14
Autogenous shrinkage
15
Concern is primarily low w/c
0.50 w/c
Initial set locks in paste structure
Cement grains initially separated by water
Extra water remains in small pores even at a1
0.30 w/c
Autogenous shrinkage
Pore fluid pressure reduced as smaller pores are
emptied
Pores to 50 nm emptied
Increasing degree of hydration
16
Internal RH Internal Drying
17
Mechanism of shrinkage

• Shrinkage dominated by capillary surface tension
  mechanism
• As water leaves pore system, curved menisci
  develop, creating reduction in RH and vacuum
  (underpressure) within the pore fluid

18
Physical source of stress
We can quantify the stress using measured
internal RH using Kelvin Laplace equation
p vapor pressure ? pore fluid pressure R
universal gas constant T temperature in
kelvins v molar volume of water
19
Measuring internal RH
20
Reduced RH drives shrinkage
21
Modeling RH Stress

• Add a fitting parameter

NOTE The fitting parameter is associated with
creep in the nanostructure
22
Long term autogenous shrinkage
23
External drying stresses
24
RH as function of time depth
Specimen demolded at 1 d
Different depths from drying surface in 3x3
concrete prism exposed to 50 RH and 23o C
25
External restraint stress superposed
26
Time to fracture (under full restraint) related
to gradient severity
Failed at 7.9 days
Failed at 3.3 days
27
Shrinkage problems

• Uniform shrinkage
• cracking under restraint
• Shrinkage Gradients
• Tensile stresses on top surface
• Curling behavior of slabs, and cracking under
  wheel loading

28
Evidence of surface drying damage
Hwang Young 84 Bisshop 02
29
Restrained stresses
30
Applying restraint
31
Typical Restrained Test Data
32
A versatile test method

• Assess early cracking tendencies

33
Volume stability
Volume Change
Thermal
Shrinkage
Creep
External Influences
Autogenous shrinkage
External drying shrinkage
Basic creep
Drying creep
Heat release from hydration
Chemical shrinkage
Cement hydration
34
Now we are ready for structural modeling!

• All this work defines material models that
  capture
• Autogenous shrinkage
• Drying shrinkage
• Creep
• Thermal deformation
• Interdependence of creep shrinkage

35
Case Airfield slabs
36
Curling of Slab on Ground
37
NAPTF slab cracking
SLAB CURLING
P
HIGH STRESS
Material (I)
Material (II)
38
Finite Element Model
¼ modeling using symmetric boundary conditions
NAPTF single slab
1. 20-node solid elements for slab 2. Non-linear
springs for base contact
39
Loadings
Temperature
Internal RH
Number are sensor locations (Depth from top
surfaces of the slab)
40
Deformation
Deformation
Ground Contacts
Ground Contacted
Displacement in z-axis (Bottom View)
41
Stress Distribution
Maximum Principle Stress
What will happen when wheel loads are applied ?
1.61 MPa (234 psi)
Age 68 days
42
Lift-off Displacement
Clip Gauge Setup
Lift-off Displacement
43
Analysis of stresses
smax 77 psi
smax 472 psi
smax 558 psi
Curling Only
Curling Wheel loading
No Curling
44
Case Self Consolidating Concrete
45
Several issues

• Do SCC mixtures tend toward higher shrinkage?
• How will segregation influence stresses?

46
We can expect problems

• Typical SCC has lower aggregate content, higher
  FA/CA ratio, and lower w/cm ratio

FA/CA Ratio
47
Problems can arise
Typical Concrete Safe Zone ?
w/b, paste
0.41, 33
0.40, 32
0.39, 37
0.34, 34
0.33, 40
48
Role of paste content and w/c ratio
Typical Concrete Safe Zone ?
w/c, Paste
0.40, 32
0.41, 33
0.34, 34
0.39, 37
0.33, 40
49
Acceptance Criteria w/c ratio

• Tazawa et al found that 0.30 was an acceptable
  threshold
• In our study, 0.34 keeps total shrinkage at
  reasonable levels
• 0.42 eliminates autogenous shrinkage
• Application specific limits
• High Restraint 0.42
• Med Restraint 0.34
• Low Restraint w/c based on strength or cost

50
Acceptance Criteria Paste Content

• IDOT max cement factor is 7.05 cwt/yd3
• At 705 lb/yd3, 0.40 w/c 32 paste
• Below 32, SCC has questionable fresh properties
• Is 34 a reasonable compromise?
• Application specific limits
• High Restraint 25-30
• Med Restraint 30-35
• Low Restraint Based on cost

51
Segregation

• SCC may segregate during placement
• Static or Dynamic
• How does this impact hardened performance?

52
Consider static segregation

• Specimen 8 x 8 x 20 prism
• 8 equal layers
• Each layer assignedCA, E and esh

53
Experiment

• ? Cast vertically to produce a segregated cross
  section
• ? Laid flat to measure deflection caused by
  autogenous shrinkage of segregated layer

54
Results
Deflection (in)
Concrete Age (d)
55
Model validation

• Now run model under restrained conditions to
  assess STRESS
• Model confirms we have reasonable rules for
  segregation limits
• HVSI 0 or 1 is OK
• HVSI 2 or 3 is BAD

HVSI Rating
56
(No Transcript)
57
Back to Specifications

• What is the real performance we need to ensure?
• More that strength
• Spec writers need to assert more control
• Example IDOT -- SCC will have limits on
  segregation, min. aggregate content, min. w/c

58
Specing real performance

• How do you impose long-term requirements using
  short-term properties?
• How do you impose limitation on long term
  cracking when factors are so extensive, including
  environment and loadings beyond control of
  material supplier?

59
Performance vs. Prescription

• Can Performance Based Specs do the whole job?
• Prescriptions
• Min. and max w/c
• Min. aggregate content
• Aggregate gradation limits
• Performance requirements
• Max. drying shrinkage, maybe autogenous shrinkage
• Permeability (RCPT ?)

60
Last thoughts

• Times they are achanging
• We have higher expectations
• We have new tools, new knowledge
• We are ever pushing the boundaries of past
  experience