Effects of Deicerpnsassociation.org/wp-content/uploads/2011... · APWA Meeting Spring 2011 Slide 3 of 69 3 Jan Olek, [email protected] Prepared for the 2011 APWA North American Snow

2011 APWA North American Snow & Pacific Northwest Snowfighters Conference, April 10‐12, 2011, Spokane, WAJan Olek ([email protected]) and Jason Weiss ([email protected]) , Purdue University, West Lafayette, IN

1 of 12

APWA Meeting Spring 2011 Slide 1 of 69

Prepared for the 2011 APWA North American Snow & Pacific Northwest Snowfighters Conference

April 10-12, 2011, Spokane, WA

This will be a two-part presentationPart 1 (J. Olek ) ▪ will briefly outline major challenges in the area of

accelerated testing of the effects of deicers on the ti f t


properties of concrete▪ will review the recently completed laboratory study on

the effects of magnesium and calcium chlorides on the properties of concrete

Part 2 (J. Weiss ) ▪ will review the impact of fluid transport on pavement joint

deterioration and will discuss the issues related to applicability of accelerated tests to field conditions


Jan Olek, [email protected] for the 2011 APWA North American Snow & Pacific

Northwest Snowfighters ConferenceApril 10-12, 2011, Spokane, WA

Investigate the effects of de-icing/anti-icing

chemicals on durability of concrete

E l t th h i l i t ti f d i i


Evaluate the chemical interactions of de-icing

chemicals with concrete matrix and aggregates

Compare the effects of de-icing chemicals on plain

(Type I) and fly ash (FA20) concretes

25

30

35

40

% o

f sol

ids

NaCl CaCl2 MgCl2

W/D regime F/T regime


0

5

10

15

20

1 2 2 3 4 4 Cur. 1 3 4 4 5 Cur.

Con

cent

ratio

n,

Wang et al. [1] Darwin et al. [2] Lee et al. [3] Cody et al. [4] Kozikowski et al. [5] Cur. [Current Study]

0 10 20 30 40 50 60 70

Darwin et al [2]

Lee et al. [3]

Cody et al. [4]

Temperature, oC


0 20 40 60 80 100 120 140

Current study

Wang et al. [1]

Darwin et al. [2]

Time, hours

Drying time, hours (varies from 8 to 34 hours)Wetting time, hours (varies from 16 to 132 hours)Drying Temp. , oC (varies from 23oC to 60 oC)Wetting temp. oC (varies from 4 oC to 60 oC)


2 of 12

-40 -20 0 20 40 60 80

L t l [3]

Cody et al. [4]

Current study

Temperature, oC


0 20 40 60 80 100 120 140

Wang et al. [1]

Kozikowski et al. [5]

Lee et al. [3]

Time, hours

Thawing time, hours (varies from 8 to 132 hours)Freezing time, hours (varies from 12 to 24 hours)Thawing Temp. , oC (varies from 20 oC to 58 oC)Freezing temp. oC (varies from 0 oC to -20 oC)

Previous studies utilized wide ranges of concentrations of deicers, temperature and lengths of exposure cycles; some combinations resulted in unrealistic conditions


resulted in unrealistic conditions

No prior investigations as to how accelerated testing affects the wetting and drying (or freezing and thawing) processes and thus the interaction of deicer solutions with concrete

The durability of the two types of concrete (plain and fly ash) when exposed to:

Wetting and drying (W/D) cycles

Freeze thaw (F/T) cycles


Freeze-thaw (F/T) cycles

Periodic monitoring of mass loss/ length changes and resonance frequencies of specimens exposed to W/D and F/T cycles

Determination of compressive strength of companion specimens

Cement (Type I)

Fly ash (Class C)

Air entraining agents


Air entraining agents

Deicing solutions- NaCl, MgCl2, CaCl2, Ice

Ban®

Water reducing admixtures

Coarse aggregates (limestone/ dolomite)

Mix TYPE I Mix FA 20Cement (lbs/ cu. yd) 586 470Fly ash (lbs/ cu. yd) - 117Water (lbs/ cu. yd) 246 246Fi A t (lb / d) 1303 1303


Fine Aggregates (lbs/ cu. yd) 1303 1303Coarse Aggregates (lbs/ cu. yd) 1780 1780w/cm 0.42 0.42Air Entrainer (oz. /cwt) 1.10 1.5 Water reducing admixture (oz./cwt)

6.0 3.3

Slump (in) 2.5 3.0 Air content (percent) 6.8 5.8


210 W/D cycles in 25% MgCl2

168 W/D cycles in 28% CaCl2


3 of 12


Testing ceased after 168 W/D cycles due to extensive surface deterioration.


168 W/D cycles in 28% CaCl2 210 W/D cycles in 25% MgCl2

Type IFA20


The use of fly ash helps reduce the rate of deterioration.

Type I

FA20


Light surface deterioration visible after 275 W/D cycles (64% longer than CaCl2). Testing continued until 350 W/D cycles (over 100% longer than CaCl2).

Specimens stored in calcium chloride solution exhibited significant mass loss due to severe


loss due to severe spalling of the edgesSpecimens stored in magnesium chloride solution exhibit small mass gain

Fly ash concrete specimens exhibited similar

fil i


mass profile in both solutions during their respective exposure periods.


4 of 12

Specimens exposed to calcium chloride solution exhibited early decrease in the RDME values

The RDM of specimens exposed to magnesium chloride solution started showing similar rate of


started showing similar rate of decrease in RDME but only after more than 200 W/D cycles

RDME values dropped to about 83% of the original value after about 154 cycles for specimens in CaCl2 solution and after about 350 cycles for specimens in MgCl2 solution

After the initial increase, the RDME values of FA 20 specimens exposed to calcium chloride solution started decreasing (after about 50 W/D cycles) and

h d 100% t th ti


reached 100% at the time the test was discontinued.RDME values for FA 20 specimens exposed to magnesium chloride solution remained at levels higher than 100% until the completion of the test (at about 350 cycles).

Compressive strength was measured on cylinders exposed to W/D regime after the completion of exposure in respective solutions.


The normalized compressive strength for specimens in different deicing solutions was calculated with respect to compressive strength of the specimens in deionized water.

350 cycles350 cycles 378

cycles

350 cycles168

cycles

60

80

100

120

ress

ive

stre

ngth

, %


Note: The duration of W/D cycles was 350 cycles for all specimens except for specimens in calcium chloride solution (168 cycles), and NaCl with Ice Ban® (378 cycles)

0

20

40

60

Deionized water NaCl CaCl2 MgCl2 NaCl+CaCl2 NaCl+ Ice Ban NaCl+ MgCl2

Nor

mal

ized

com

pr

Exposure solution

Type I Fly ash


extensive surface deterioration (similar to that observed during W/D exposure)



5 of 12

There is continuous increase in mass for specimens exposed to magnesium


to magnesium chloride solutionThe specimens in calcium chloride solution show continuous mass loss

FA 20 specimens in calcium chloride solution show more mass gain than


mass gain than corresponding specimens in magnesium chloride solution

Specimens show decreasing trend for RDME right after 36 cycles


Similar values of RDME after 250 cycles for Type I specimens in both solutions

The relative dynamic modulus of elasticity shows decreasing trend for FA 20


specimens in calcium chloride solution There is no such reduction for FA 20 specimens exposed to magnesium chloride solution for up to 300 cycles


In CaCl2 solution In MgCl2 solution

top surface (0 -0.5 cm) middle (~0.8-1.3 cm) bottom (~ 1.8-2.2cm)


No traces of Cl in bottom part of specimen

3CaO·Al2O3·CaCl2 · 10 H2O


6 of 12

top surface (0 -0.4 cm) middle (~0.8-1.3 cm) bottom (~ 1.8-2.2cm)


Small peak of Cl in bottom part of specimen3CaO·Al2O3·CaCl2 · 10 H2O

External signs of deterioration were more extensive in Type I specimens exposed to 168 W/D cycles in CaCl2 solution than those observed in specimens exposed to 210 W/D cycles in MgCl2 solution (despite 25% longer exposure). Initial signs of distress were observed in the MgCl2 exposed specimens only after 275 W/D cycles but even then the overall


specimens only after 275 W/D cycles but even then the overall damage was still much less than that observed in the CaCl2specimens after 168 W/D cycles (60% shorter exposure).Surface deterioration of Type I specimens exposed to the MgCl2 solution for 350 W/D cycles was observed to be comparable to the surface deterioration of the equivalent specimens in the CaCl2 for 168 W/D cycles. This indicates the specimens in the CaCl2 were deteriorating at the rate at least twice as fast as the specimens in the MgCl2 solution.

The RDME values of the specimens exposed to MgCl2stayed at the 100% level for almost 200 W/D cycles as opposed to about 30 cycles for specimens exposed to the CaCl2. This indicates that the CaCl2 solutions

l t th d t i ti t t i t l 3


accelerate the deterioration rate to approximately 3 times that of the MgCl2 solutions.

The fly ash concrete demonstrates better performance compared with the Type I specimens.


Jason Weiss, [email protected] for the 2011 APWA North American Snow & Pacific

Northwest Snowfighters ConferenceApril 10-12, 2011, Spokane, WA

Motivation for Research

Damage is seen in the transverse and longitudinal joints in many pavements.


Some have related this entirely to de-icers.

Is this fair or are there other factors?

Pavement Joint Deterioration

• Deterioration of pavement joints has caused many to question the cause of this distress

• Many have sought to explain with:– Poor Air Entrainment


– Poor Air Entrainment– Poor Construction– Design Changes– New Deicing Practices– New Deicing Materials

• However closer examination is needed


7 of 12

Joint Details

• Joints are placed in concrete pavements to control shrinkage and thermal cracking

• The joint is designed to crack thereby relieving stress in the pavement


relieving stress in the pavement• Typically a single cut is placed to 1/3 of the

depth of the pavement at early age • After cracking a second cut is made • To keep water and incompressibles out of

the joint sealant is placed in the joint

An Ideal Longitudinal Joint

Water-Salt Solution2nd Cut


Weiss and Nantung 2005

Joint Sealant

Backer Rod

Initial Cut

Crack

Less ideal case – Damaged seal




Joint Sealant(Failed)

Backer Rod

Initial Cut

Crack

The least ideal case – No Crack




High Degree of Saturation Zone

Initial Cut

Damaged Retention

Hypothesis for Saturation and Damage in Joints

• Fluid entering joint is trapped• Saw cut joints are

not opening • As a result, fluid


As a result, fluid can not drain out to drainable sub-base under the PCCP

• Silicone sealant and backer rod limit drying of trapped water Weiss and Nantung 2005


Why be concerned about fluid ingress

PoorD bilit

Good Durability

to F

rost

(%) 100

80

60Poor

D bilit

Good Durability

to F

rost

(%) 100

80

60


Durability

Degree of Saturation (%)60 65 70 75 80 85 90 95 100

Res

ista

nce

t

40

20

0

After CEB 1957

Durability


Res

ista

nce

t

40

20

0

After CEB 1957

There is a critical saturation that makes concrete

susceptible to repeated F-T


8 of 12

Influence of Paste Porosity

6.0

7.0

8.0

orpt

ion

/mm

2 )

50%RH65%RH80%RHStandard Cond.

• High w/c mixes absorb more water faster than low w/c mixes

12.0

14.0

16.0

tion

5 )

50%RH65%RH80%RHStandard Cond.


0.30 0.35 0.40 0.45 0.50 0.55water/cement

0.0

1.0

2.0

3.0

4.0

5.0

Cum

ulat

ed A

bso

at 8

day

s (m

m3 /

0.30 0.35 0.40 0.45 0.50 0.55water/cement

0.0

2.0

4.0

6.0

8.0

10.0In

itial

Abs

orpt

(10-3

mm

/s0.

Note: Standard conditioning from saturation

Degree Of Saturation and Air Content

• Once saturated all concrete fails

• Entrained air alters time to DOS but not

0.20

0.40

0.60

0.80

1.00

E/E 0

M-6N=0N=1N=2N=3N=4


the occurrence

Li et al. to be published 2010

75 80 85 90 95 100Degree of Saturation (%)

0.00

N=5N=6

Times to Reach 92% DOS

• 50% Relative Humidity – 0.4 years 4% Equivalent Concrete Air– 6.1 years 6.5% Equivalent Concrete Air– 5.8 years 9% Equivalent Concrete Air

• 65% Relative Humidity

Li et al. to be published 2010


65% Relative Humidity – 0.5 years 4% Eq. Air– 2.6 years 6.5% Eq. Air– 9.6 years 9% Eq. Air

• 80% Relative Humidity – 0.3 years 4% Eq. Air– 5.9 years 6.5% Eq. Air– 6.3years 9% Eq. Air

The Drive for Fast Results

• We can not wait 10 to 20 years to see if a pavement may be susceptible to damageA lt l t d


• As a result accelerated tests are developed to rapidly provided feedback on how damage develops and how this is related to the mixture

Do Accelerated TestsAccurately Represent this Issue

• Do they provide an indication of the real life damage that occurs ?

• Do they examine actual mechanisms or do they create alternate damage ?


• Are there shortcomings ?• What time scaling functions should be used ?• Can they mislead ?• Well intentioned but may not be thought through• Will not touch on possible reactions today

Outline

• Concrete Pavement Joints – A Review of Common Construction Practices

• A Discussion of Partial Saturation and Why it is Important for this Problem


it is Important for this Problem• Accelerated Tests and Some Issues that

Arise that Should be Remembered• These Test Details are Crucial when One

Interprets Tests Reported by Different Research Programs


9 of 12

Important Issue #1

• Accelerated testing can change the mechanics of the problem

• Accelerated test can trigger something different than what is seen in field conditions

• When this happens ‘scaling results’ in terms of


When this happens scaling results in terms of time is very difficult

• Accelerated testing should not just tell if something happens but when.

• Some aspects of what people are seeing may occur 80 years in the future.

A Brief Discussion of Mechanisms of Fluid Transport

• Concrete is a porous material (wide range of sizes, pores like to be wet)

• Desorption (drying) – water leaves (slow)• Absorption (unsat.) – fluid ingress (fast)


p ( ) g ( )• Diffusion (sat.) – species gradient (slow)• Wick

Important Issue #2

• Accelerated testing can change the boundary conditions of the problem

• Accelerated test can trigger something different than what is seen in field conditions


different than what is seen in field conditions• This causes different driving forces than

what occurs in the problem

Accelerated Testing

• Some pushing for “fast” tests, but what do they represent

MicrocracksFluidVapor


80% RH

Oven Dry

FluidVapor

Actual RHIn Pavement

• RH in field concrete is very high

• Implication only a thin skin dries if any (absorption) while core

11/1/09 12/1/09 1/1/10 2/1/10 3/1/10 4/1/10 5/1/1075

80

85

90

95

100

Rel

ativ

e H

umid

ity (%

)

-10.0

0.0

10.0

20.0

30.0

Tem

pera

ture

(C)

12 mm RH25 mm RH50 mm RH92 mm RH

Covered Slab

Kompare et al. to be published 2010


(absorption) while core remains mostly saturated (diffusion)

• In joints little drying as sealant traps water

11/1/09 12/1/09 1/1/10 2/1/10 3/1/10 4/1/10 5/1/1075

80

85

90

95

100

Rel

ativ

e H

umid

ity (%

)

-10.0

0.0

10.0

20.0

30.0

Tem

pera

ture

(C)

12 mm RH

Submerged

Note: It is believed the slabs were not submerged after March

Absorption vs. Conditioning

• Samples stored at a low RH absorb more fluid [Castro et al.2010]

• Microcraking

8

10

12

on (m

m3 /m

m2 )

w/c = 0.45, Oven Dryw/c = 0.45, 50% RHw/c = 0.45, 65% RHw/c = 0.45, 80% RH

c)


• Microcraking opens paths

• Makes sense as more pore space is open however it is not the critical factor (index)

0 20 40 60 80 100Time (days)

0

2

4

6

Wat

er A

bsor

ptio


10 of 12

Important Issue #3

• Some tests are being used without a full appreciation of what parameters are important

• Here we will talk about a test that measures


Here we will talk about a test that measures rate….. While rate is one component it is really not the most important rather what we want is “how close are we to the end”

• While many are measuring the rate of absorption. Degree Of Saturation is what drives the Freeze-Thaw process

Good Durability (%) 100

Good Durability (%) 100

Degree of Saturation

• What is important is the volume of pores that are filled/the volume available (Degree of saturation)

• Freeze-thaw damage requires a critical degree ofsaturation (75 90%


PoorDurability

Good Durability


Res

ista

nce

to F

rost 80

60

40

20

0

After CEB 1957

PoorDurability

Good Durability


Res

ista

nce

to F

rost 80

60

40

20

0

After CEB 1957

There is a critical saturation that makes concrete

susceptible to repeated F-T

(75-90% from literature)

• Mixture dependent

Absorption vs Degree of Saturation

• Absorption less important than DOS saturation• Harder to saturate ‘wet concrete’ [Castro et al. 2010]

0.8

1.0

on

10

12

3 /mm

2 ) c)


0 20 40 60 80 100Time (days)

0.0

0.2

0.4

0.6

0.8

Deg

ree

of S

atur

atio


c)

0 20 40 60 80 100Time (days)

0

2

4

6

8

Wat

er A

bsor

ptio

n (m

m3


Important Issue #4

• Improper conclusions can be drawn if all the details are not considered

• Many studies are pulling field cores for example and performing absorption tests


example and performing absorption tests• Using tests data

to infer items aboutchanges in the materialstructure

• DangerousA

bsor

ptio

n

Square Root of Time

Pristine

Exposed to Salt

Important Issue #4

• The presence of deicing salts in the field samples impacts the absorption

• Many mistakenly identify this as damage or Pristine


this as damage or ‘pore filling’

• Care must be taken in analyzing field concrete exposed to deicing salt solutions

Abs

orpt

ion

Square Root of Time

Pristine

PreviouslyExposed to Salt

Important Issue #4

• Deicing solutions reduce the rate of fluid absorption [Hall and Hoff, 2002]

0.8

1.0

1.2

/ (γ/η)

0.5 w

ater

NaCl theoretical


2002]

0 10 20 30 40Concentration (Mass %)

0.2

0.4

0.6

(γ/η

)0.5 so

lutio

n NaCl theoreticalNaCl measuredNaCl relative sorptivityCaCl2 theoreticalCaCl2 measuredCaCl2 relative sorptivityMgCl2 theoreticalMgCl2 measured

MgCl2 relative sorptivity

Spragg et al. 2010


11 of 12

Important Issue #5

• Moisture history influences the absorption properties of the concrete (Hystersis)

• Here – Drying followed by ASTM standard

8

10

mm

2 )

w/c = 0.40, 50%RH + Standard cond.w/c = 0.40, 65%RH + Standard cond.w/c = 0.40, 80%RH + Standard cond.


by ASTM standard • Saturation is suggested

for beginning test• Tests of field or oddly

conditioned samplescan be very suspect 0 2 4 6 8 10

Time (days)

0

2

4

6A

bsor

ptio

n (m

m3 /m w/c = 0.40, Pre-Sat. + Standard cond.

Important Issue #6

• Drying and wetting are different• Overly severe drying (low RH) unfairly treats

the salt solutions • The system with the salt never ‘dries’ while


ythe reference does (all depends on conc and boundary conditions)

• Salt solutions are in equilibrium at lower RH than water as such they evaporate less

• As mentioned field cases are wet the majority of the time in the joints

“Drying of Concrete with Deicers”

0.4

0.2

0.0

-0.2

f Wat

er L

oss

/ mm

2 of s

urfa

ce)

0 10 20 30 40Concentration (Mass %)

0

20

40

60

80

100

RH

(%)

NaClCaCl2MgCl2 (c)


Spragg et al. submitted 2010

1.0

0.8

0.6

Volu

me

of(m

m3 o

f wat

er

0 24 48 72 96 120 144 168 192 216

Time (h)

DI WaterMgCl2 0.9%MgCl2 30%CaCl2 0.96%CaCl2 32%NaCl 0.7%NaCl 23%

Illustration of Importance

• Wetting and drying illustration for water and deicing solution

et

With DeicersCritical DOS


• Exasterbated with excessive drying

Dry

W

e

Water

Important Observation #7

• The time scale differs between drying and wetting – Drying takes a much longer time

0.0

-0.2

Loss

f sur

face

)

8

10

on

f sur

face

)

1 Week of Drying


• Many strategies do not appear to scale

1.0

0.8

0.6

0.4

0.2

Volu

me

of W

ater

L(m

m3 o

f wat

er /

mm

2 of

0 24 48 72 96 120 144 168 192 216

Time (h)

DI WaterMgCl2 0.9%MgCl2 30%CaCl2 0.96%CaCl2 32%NaCl 0.7%NaCl 23%

0 2 4 6 8 10Time (days)

0

2

4

6

Wat

er A

bsor

ptio

(mm

3 of w

ater

/ m

m2 of

65% paste volume55% paste volume45% paste volume

1 Week of Drying

Hrs of wetting

Important Issue #7 - Viscosity

(a) Horizontal and (b) vertical ingress in cement paste



12 of 12

Viscosity Measurements

• Viscosity effects transport • Penetration as well

1

10

100

s)

SME-PS Blends, 25OCWater, 25OC

0 10 20 30 40

1

10

100


0 10 20 30 40Polystyrene Content (%)

0.0001

0.001

0.01

0.1

Visc

osity

(Pa

-

0.0001

0.001

0.01

0.1

Summary of Saturation in Joint

• Water entering joints, particularly longitudinal saw joint, is trapped

• Longitudinal saw cut joints not cracking or if cracking not opening


• As a result, water can not drain out to drainable sub-base under PCCP

• Silicone sealant and backer rod limit evaporation/drying of trapped water

• Higher w/c mixtures are more susceptible to damage than low w/c mixtures

Summary

• Tried to review issues in many tests• Mechanism (sorption, drying, diffusion)• Importance of boundary conditions

Ab ti t d f t ti


• Absorption rate versus degree of saturation• Errors when testing field samples (presence

of salts, moisture history) • Importance of drying versus wetting time

scales and boundary conditions• Must consider in evaluating data

Documents

Effects of Deicerpnsassociation.org/wp-content/uploads/2011... · APWA Meeting Spring 2011 Slide 3 of 69 3 Jan Olek, [email protected] Prepared for the 2011 APWA North American Snow