Material Design and Structural Configuration of Link Slabs ... · 1/26/2018 · Introduction to...

1

Research Seminar

Behrouz Shafei, Ph.D., P.E.

Assistant Professor

Michael Dopko

Graduate Research Assistant

Shahin Hajilar, Ph.D.

Post Doctoral Research Associate

Material Design and Structural Configuration of

Link Slabs for ABC Applications

Institute for Transportation (InTrans)

Department of Civil, Construction and Environmental Engineering

2

Project Details

Material Design and Structural Configuration of Link

Slabs for ABC Applications

Sponsor: Accelerated Bridge Construction University

Transportation Center

Evaluation of the Use of Link Slabs in Bridge

Projects

Sponsor: Iowa Department of Transportation

Principal Investigators:


Brent Phares, Ph.D., P.E.

Peter Taylor, Ph.D., P.E.

3

Outline of Presentation

Introduction

Design of Materials for Link Slabs

Conventional Fiber Reinforced Concrete

Flexible Cementitious Materials

Application of Link Slabs in Bridges

Case-Study Bridge in Iowa

Finite-Element Simulations and Results

Conclusions

Future Work

Acknowledgements

4


Introduction







Conclusions

Future Work

Acknowledgements

5

Problem Statement

Despite major advances in the design and construction

of Prefabricated Bridge Elements and Systems (PBES)

for ABC applications, bridge joints are still in need of

improvement and further investigation.

Expansion joints play a critical role in accommodating

unrestrained deformations of adjacent spans due to

thermal expansion and traffic loads.

The existing bridge joints, however, deteriorate rapidly

and require major maintenance efforts.

6

Problem Statement

One Promising Solution: Link Slab

Eliminates the opening between two adjacent spans

Minimizes the penetration of aggressive agents

Improves the long-term durability of bridge components

Research Needs

Materials

Design approaches

Detailing requirements

ABC applications

7

Research Objectives

The objectives of this research project include:

To identify the materials of choice for link slabs based on

the long-term durability and performance requirements

To eliminate the potential modes of damage and failure

(e.g., crack formation and debonding mechanisms)

To introduce a robust structural design for link slabs,

particularly for implementation in ABC projects

To develop the supporting design guidelines and

construction practices

8

Introduction to Link Slabs

Link slabs are intended to fill the free space provided by

intermediate expansion joints.

Due to the absence of free space for movement and

depending on the stiffness of the link slab:

Additional stresses can be developed in a bridge’s

superstructure and substructure under thermal and live

loads.

These stresses must be considered in design to avoid

damage to the bridge.

Link slab design with low stiffness and high ductility

would minimize these effects.

9

Introduction to Link Slabs

Girder end rotations from live load cause negative

flexure in the link slab, causing tensile stress on the

exposed deck surface.

Thermal expansion and contraction of the superstructure

adds to the stresses developed in the link slab.

10

Full-Depth Link Slabs

Full-depth link slabs

Debonded zone ≈ 5% of adjacent span length

Avoid stress concentrations

Transition Zone ≈ 2.5% of adjacent span length

Include shear stud connections for anchorage

Continuous top and bottom link slab reinforcement

From: Lepech & Li (2005)

11

Full-Depth Link Slabs

Full-depth link slabs

Full-depth link slabs using conventional reinforced

concrete (RC) material has been explored (e.g.,

Caner & Zia (1998), Okeil & ElSafty (2005),

Charuchaimontri et al. (2008)). The application of

conventional RC link slabs is limited by the inherent

material properties of concrete.

It has been found that a single, full depth crack is

likely to form in the middle of the link slab, which

requires frequent sealing and maintenance, defeating

the intended purpose of a link slab

12

Partial-Depth Link Slabs

Partial-depth link slabs

A thin slab, which is less intrusive to global structure

Application for rehabilitation projects

Design of link slab materials with the following objectives:

Reduce the required length

Minimize stress concentrations

Avoid compression buckling

ABC Applications

From: Larusson (2013)

13

Full-Depth vs. Partial Depth

Full-Depth Link Slab Partial-Depth Link Slab

Cast-in-place or Pre-cast

• Early strength for cast-in-place

• Connection details for pre-cast

• ABC applications

Cast-in-place or Pre-cast

• Early strength for cast-in-place

• Connection details for pre-cast

• ABC applications

• Continuity moment may introduce

additional stresses to other bridge

elements

• Continuity moment is minimized

• Less intrusive to the structure

Main structural component

• It must resist the transverse

moment between the girders

Structural component

• Flexible with minimum micro-cracks

to eliminate ingress

Design reinforcement based on link

slab design moment

Design from material strain capacity

with high ductility and low stiffness

14


Introduction







Conclusions

Future Work

Acknowledgements

15

Link Slab Materials

Desirable Link Slab Properties, especially for

ABC Applications

High early strength

Limited crack widths (below AASHTO limits)

Durable (i.e., long-term performance and corrosion

resistance)

Widely available and contractor-friendly

Adequate workability

Simple concrete mixing regime that is easily

reproducible

Research Question: Which market available synthetic

fiber(s) can provide the highest post crack toughness in FRC

without being detrimental to workability?

16

Fiber-Reinforced Concrete

Investigation 1: Which type(s) of synthetic fibers can provide

the best ductility and toughness to a conventional fiber-

reinforced concrete (FRC) mix when subject to tensile stresses.

High Strength Polyethylene (HSPE)

Polyvinyl Alcohol (PVA)

Basalt (BSLT)

Polypropylene (PP)

Cement Fly Ash Coarse

Agg.

Fine

Agg.

Water Water

Reducer

Fiber

(% Vol.)

593 254 1548 1266 322 Varied 0.5%-

1.5%

Paste + water reducer added for each mix to maintain workability

Base Mix Design

17

Basics of Conventional FRC

From Marković (2006)

FRC refers to a concrete mix that also contains

coarse aggregate.

When the coarse aggregate is removed from the system, the above

mechanism is invalid due to the increased homogeneity of the matrix

and corresponding change in fracture energy / crack development

18

Fiber Types and Parameters

Basalt

High Strength

Polyethylene PolypropylenePolyvinyl Alcohol

HSPE PVA BSLT PP

S.G. 0.98 1.30 2.10 0.91

Length (in) 0.75 0.75 1.70 1.5

Tensile Strength

(ksi)377 145 157 83-96

Modulus (ksi) 11460 3920 6380 690

19

Conventional FRC

3 Volume Percentages (0.5%, 1.0% and 1.5%)

Tested for each fiber type (except HSPE due to clumping during

mixing at lower volumes)

ASTM C1609 – flexural performance of fiber

reinforced concrete

Utilized to compare fibers to select a “best” fiber type and volume

for toughness and residual strength

Flexural performance is a good indicator of tensile performance

Vibrating Kelly Ball (VKelly) test used to monitor

fresh properties

Contractor-friendly material is the goal.

The material must be placed without complications in congested

reinforcement, if needed.

20

Conventional FRC

VKelly Test – 2 Parameters

VKelly Slump

Measures static yield stress of the

concrete

VKelly Index

Measures the mixture’s response to

vibration for consolidation

Kelly Ball Concrete

Vibrator

0

1

2

3

4

5

6

0 1 2 3 4 5 6 7

Pen

etra

tio

n D

epth

(in

)

Vibration duration, √s

Vkelly Index

Slope = Index

21

Test Results and Observations

Fresh PropertiesVKelly

Mix ID % Paste Added HRWR (oz/cwt) Slump Index

PP0.5 0.0 0.6 1.5 0.47

PP1.0 0.5 1.8 1.3 0.33

PP1.5 2.0 3.1 1.0 0.13

PVA0.5 2.4 2.9 2.6 N/A

PVA1.0 6.7 2.8 1.3 0.25

PVA1.5 11.5 6.3 2.0 0.19

BSLT0.5 0.0 1.0 2.3 0.55

BSLT1.0 0.0 2.6 1.9 0.43

BSLT1.5 0.0 2.6 1.3 0.30

HSPE0.5 2.8 2.1 0.7 0.28

HSPE1.0 4.0 5.5 1.7 0.33

HSPE1.5

HSPE was not mixed at 1.5% volume due to mixing complications at 1.0%

volume. HSPE fibers tend to stick to each other during the mixing process

forming clumps in the mix making them impractical to mix at high volumes

and non-contractor friendly.

22


Clumping - PVA

23


Clumping - HSPEFresh Hardened

24


1.0% Basalt 1.0% Polypropylene

Good dispersion was achieved with Basalt and PP mixtures.

25

Flexural Performance

ASTM C1609 – Flexural Performance of FRC

4”x4”x14” beam specimens

3rd point bending

Displacement controlled

Load vs. average mid-span deflection

Performance parameters include:

First Peak Strength (Modulus of Rupture)

Residual Strengths at L/600 and L/150

deflection

Toughness up to L/150 deflection

Equivalent Flexural Strength Ratio

26

Link Slab Material – Conventional FRC

0

1000

2000

3000

4000

5000

6000

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

Lo

ad

(lb

)

Deflection (in)

HSPE

PP

BSLT

PVA

Load Deflection Curve for 1.0% Fiber

First Peak Strength – Flexural strength prior to crack formation

Toughness – Area under the load vs. mid-span deflection

curve, up to L/150 deflection. This indicates the ability of the

fibers to carry load after cracking.

27

Link Slab Material – FRC Results

500

550

600

650

700

750

0.5% 1.0% 1.5%

Stre

ngt

h (

psi

)

Fiber Content by Volume

First Peak Strength

HSPE PVA PP BSLT

0

50

100

150

200

250

300

350

0.5% 1.0% 1.5%

Tou

ghn

ess

(lb

-in

)

Fiber Content by Volume

Toughness (T_150)

HSPE PVA PP BSLT

First peak strength was not significantly affected by fiber type or volume

PP and Basalt showed increasing trend with volume, HSPE and PVA did not.

Toughness increased with increasing volume for all fiber types

Basalt > HSPE > PP > PVA for toughness

28

Observations and Findings

After mixing 4 types of concrete fiber at varying volumes:

Rank Mixing Characteristics Toughness Characteristics Cost (1 = highest)

1 Basalt Basalt Basalt

2 Polypropylene High Strength Polyethylene Polyvinyl Alcohol

3 Polyvinyl Alcohol Polypropylene Polypropylene

4 High Strength Polyethylene Polyvinyl Alcohol High Strength

Polyethylene

PVA and HSPE both had dispersion issues, but of a different

nature.

HSPE showed promising strength results, but is very challenging

to work with during mixing.

Basalt Minibars performed the best in all aspects.

29

Early Strength and Shrinkage

ABC cast in place concrete requires high early-age

strength, 3000 psi in 24-hours minimum was deemed

suitable

Accelerating Admixture (ACC)

Carbon micro-fiber

High cementitious materials content and expected

high reinforcement ratio and resulting shrinkage

cracking potential

Shrinkage reducing admixture (SRA)

Carbon micro-fiber

4 carbon fiber doses + ACC and SRA combinations

30

Benefits of Microfibers

Control micro-crack propagation during early stages

of crack growth

Increase composite strength, especially for tensile

properties

Control or prevent plastic shrinkage cracks

Fiber Synergy

It can improve the post crack performance of macro-fibers in

hybrid fiber systems

Microfibers hold micro-cracks tight, increasing the reinforcing

potential of macro fibers

Benefits of both macro and micro fibers can be increased in

hybrid fiber systems

31

Carbon Microfibers

Relatively high modulus (242 GPa) and tensile strength

(4.14 GPa)

Very stable in the cement environment

High aspect ratio – 830 (L = 6mm, D = 0.0072mm)

Utilize high tensile strength and elastic modulus to

increase strength parameters

32

Early Strength and Shrinkage

ASTM C39 - 24-hour

compressive strength

ASTM C1581 - 28 day

restrained shrinkage ring test

Restrained Shrinkage Ring Setup

33

Results – Fresh Properties

0.3%0.1% 0.5%

34

Results – 24-Hour Compressive Strength

2000

2200

2400

2600

2800

3000

3200

3400

3600

3800

4000

0 0.1 0.3 0.5

24

-hr.

Co

mp

ress

ive

Str

en

gth

(p

si)

Carbon Fiber Content (% by Volume) and Mix ID

24-hr. Compressive Strength vs. Carbon % by Volume with Shrinkage Reducing / Accelerating Admixture

SRA + 0 ACC 0 SRA + 0 ACC SRA + ACC 0 SRA + ACC

0.0S 0.0AS 0.0A0.0 0.1 0.1A0.1AS0.1S 0.3S 0.3A0.3AS0.3 0.5AS0.5 0.5A0.5S

35

Results – Restrained Shrinkage Test

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

0.0% Carbon Content

0 SRA + 0 ACC 0 SRA + ACC SRA + 0 ACC SRA + ACCa)

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

0.1% Carbon Content

0 SRA + 0 ACC 0 SRA + ACC SRA + 0 ACC SRA + ACCb)

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

0.3% Carbon Content

0 SRA + 0 ACC 0 SRA + ACC SRA + 0 ACC SRA + ACCc)

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

0.5% Carbon Content

0 SRA + 0 ACC 0 SRA + ACC SRA + 0 ACC SRA + ACCd)

36


-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

0 SRA + 0 ACC

0.0% Carbon 0.1% Carbon 0.3% Carbon 0.5% Carbona)

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

0 SRA + ACC

0.0% Carbon 0.1% Carbon 0.3% Carbon 0.5% Carbonb)

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

SRA + 0 ACC

0.0% Carbon 0.1% Carbon 0.3% Carbon 0.5% Carbonc)

-160

-140

-120

-100

-80

-60

-40

-20

0

20

0 5 10 15 20 25 30

Str

ain

x1

0-6

Specimen Age (days)

SRA + ACC

0.0% Carbon 0.1% Carbon 0.3% Carbon 0.5% Carbond)

37


Crack Close-up

Cracked

Not Cracked

38

Findings & Conclusions

More than one way to achieve early strength for ABC

and shrinkage crack control

Binary Admixtures

Micro-carbon fiber

The use of SRA and ACC would be more contractor

friendly since the mixing of carbon fibers takes

special attention

39

Findings & Conclusions

Basalt macro fibers provide high toughness and post

crack performance without being detrimental to

workability

SRA and ACC work well together and can be utilized to

achieve an effective link slab material, especially for

ABC applications

• Carbon fiber increases 24 hr. compressive strength with

increasing volume – good for ABC cast-in-place, but

– It requires high superplasticizer dose to maintain workability

– Not contractor friendly

– Relatively high cost

40


Introduction







Conclusions

Future Work

Acknowledgements

41

Case-Study Bridge

Full Scale FE Model of 9 Span Bridge near Vinton, Iowa

Geometry

Continuous supports at piers # 1, 2, 4, 5, 7, & 8

Bearing pad supports at both abutments and piers # 3 & 6

Investigate replacement of expansion joints at piers # 3 & 6 with link

slabs

South North

42

Finite Element Model

Numerical Simulation Details

Replace expansion joints over piers 3 & 6 with link slabs and

compare with no link slab case

Consider 2 idealized support conditions (Roller or Pinned) under

expansion joints and link slabs (both abutments and piers # 3 & 6)

• In-situ bearing pads would act somewhere between these two

idealized conditions

• Lateral stiffness of bearing pads increases with age

Roller support lower bound for bearing pad stiffness

Pinned support upper bound for bearing pad stiffness

Results focus on deck/girder displacement and pier reactions

43


Roller Support Model »

Connection Boundary

Conditions

South

North

44


Connection Boundary

Conditions

Pinned Support Model »

South

North

45


Numerical Simulation

Finite Element Models

4 bridge models created/analyzed

Roller without link slab (RwoLS)

Roller with link slab (RwLS)

Pinned without link slab (PwoLS)

Pinned with link slab (PwLS)

Analysis software package ABAQUS CAE utilized

Link slabs were modeled by filling 6” expansion gap with

concrete material analogous to the concrete deck material

3x smaller mesh used for piers to increase accuracy

Expansion gap filled with

concrete for models with link

slabs

46

Finite Element Simulations

Numerical Simulations

Loading Conditions

• All 4 models analyzed for 3 loading conditions separately

Dead load (DL)

Positive temperature load (+TL)

Negative temperature load (-TL)

• Total stresses, moments and forces were calculated by

combining either (DL+TL) or (DL-TL)

• ±TL was taken as 70 ºF to produce conservative results

• DL was calculated as the self weight of the bridge material

47


Results

+TL and –TL produced results of the same magnitude for all

4 different bridge models

Due to symmetry, results are analyzed for half of the bridge

Exaggerated Deformed

Shape of the bridge under

a) +TL and b) -TL

48


With Link Slab

Lateral Girder / Deck displacements were zero

Scale Factor: 300

Roller Connections Pinned Connections

Scale Factor: 200

49

Roller Pinned Roller No Link SlabPinned Pinned

372.7 139.4

504.3

372.3 168.4

531.6

0.2 124.1

124.3

373.2 177.6

513.0

Roller Pinned Roller with Link SlabPinned Pinned

2491.4 231.0

2532.8

1768.6 252.8

1839.9

2.5 128.1

125.7

352.9 171.8

494.9

Pinned Pinned Pinned No Link SlabPinned Pinned

200.0 189.9

287.9

554.0 251.8

516.9

315.4 156.8

419.5

262.0 186.9

314.7

Pinned Pinned Pinned with Link SlabPinned Pinned

106.3 154.8

261.0

69.8 153.7

218.5

6.2 154.8

148.6

20.4 151.0

171.4

ΔT = +70 ºFConnection Type

Thermal Stress (psi) Stress due to D.L. (psi)

Total Stress (psi)

S. Abut.

# 1 # 2 # 3 # 4

50

Roller Pinned Roller No Link SlabPinned Pinned

372.7 139.4

510.1

372.3 168.4

540.7

0.2 124.1

123.9

373.2 177.6

550.8

Roller Pinned Roller with Link SlabPinned Pinned

2491.4 231.0

2720.6

1768.6 252.8

2020.4

2.5 128.1

130.6

352.9 171.8

522.3

Pinned Pinned Pinned No Link SlabPinned Pinned

200.0 189.9

307.1

554.0 251.8

805.8

315.4 156.8

323.0

262.0 186.9

448.9

Pinned Pinned Pinned with Link SlabPinned Pinned

106.3 154.8

141.2

69.8 153.7

219.8

6.2 154.8

161.0

20.4 151.0

160.7

ΔT = -70 ºFConnection Type

Thermal Stress (psi) Stress due to D.L. (psi)

Total Stress (psi)

S. Abut.

# 1 # 2 # 3 # 4

51


Results: Substructure Reactions with Link Slabs

For the RwLS model a build up of stresses, moments and forces

was discovered under temperature loading in the outer piers

The PwLS model had negligible stress, moment or force build up

in all piers

0.0

500.0

1000.0

1500.0

2000.0

2500.0

3000.0

Pier #1 Pier #2 Pier #3 Pier #4

Str

ess

(psi

)

RwoLS PwoLS RwLS PwLS

0.0

500.0

1000.0

1500.0

2000.0

2500.0

3000.0


Str

ess

(p

si)

TL DL

Pier Stresses

52


Results: Substructure

Dead loads produced negligible reactions for all cases

Temperature loads produced substantial reactions in outer piers for

RwLS case

-10000.0

-8000.0

-6000.0

-4000.0

-2000.0

0.0

2000.0

4000.0


Mo

men

t (k

ip-f

t)


-350.0

-300.0

-250.0

-200.0

-150.0

-100.0

-50.0

0.0

50.0

100.0


Fo

rce

(kip

s)


Bending Moment from TL Shear Force from TL

53

Conclusions

Conclusions:

Displacements were relatively small at the expansion gap –

worst case was 1.7” of movement.

Link slabs may work well with laterally stiff bearing pad

supports that act most like ideal pinned connections.

Link slabs would not work well with bearing pads that have

low lateral stiffness, as a build up of stresses, moments and

forces would result in the outer piers between the link slab

and abutment.

Retrofit of link slabs to replace expansion joints in bridges

with laterally stiff bearing pads may be beneficial based on

the results of this study.

54


Introduction







Conclusions

Future Work

Acknowledgements

55

Future Work

Material and structural design are closely related for

link slab design.

Material tailoring is expected to provide a range of link slab

options with varying cost and effectiveness for different

scenarios.

The final link slab design must provide a simple option for

contractors for conventional and ABC applications.

As shown in the case-study bridge investigation, link slabs

can have adverse effects on certain parts of the structure,

if used not properly.

56

Future Work

Direct tensile tests with large dog-bone (4”x4” cross

section in active zone) specimens, including rebars

embedded in the center of the active zone

Observe cracking pattern at different tensile strains

Observe how fibers change the cracking pattern

Evaluate the reinforcement effect on the cracking pattern

Use of non-metallic reinforcement

Use of flexible cementitious materials

(e.g., UHPC and textile concrete)

Lab tests on full-scale link slabs to investigate the

performance of the entire system and provide structural

design guides and recommendations

57

Future Work

Incorporation of the best choices identified for link slab

materials in the developed finite element models

Further refine the finite element model details by

incorporating the bearing pads with their associated

properties

Investigation of change of demand on various structural

components for rehabilitation projects

Identification of the bridge design and configurations that

are most appropriate to include link slabs

58

Publications

Dopko, M., Najimi, M., Shafei, B., Taylor, P., and Phares, B. (2018)

Flexural performance evaluation of fiber reinforced concrete incorporating

multiple macro-synthetic fibers, Transportation Research Record: Journal

of the Transportation Research Board [In-Press].

Dopko, M., Shafei, B., Wang, X., Taylor, P., and Phares, B. (2018)

Assessment of strength and restrained shrinkage behaviors on micro-

carbon fiber reinforced concrete with binary chemical admixtures,

Transportation Research Board Annual Meeting, January 7-11,

Washington, DC.

Dopko, M., Shafei, B., Taylor, P., and Phares, B. (2017) Comparison of

multiple concrete fiber types and volumes for cast-in-place ABC link

slabs, National Accelerated Bridge Construction Conference, December

7-8, Miami, FL.

Hajilar, S., Dopko, M., Shafei, B., Phares, B., and Bierwagen, D. (2017)

Feasibility assessment of use of link slabs in a case study bridge in Iowa,

Transportation Research Board Annual Meeting, January 8-12,

Washington, DC.

59

Acknowledgement

The support of the Accelerated Bridge Construction

University Transportation Center (ABC UTC) and the Iowa

Department of Transportation (Iowa DOT) is gratefully

acknowledged. We are also thankful to Reforcetech,

Kuraray, Forta, William Barnet & Son, LLC for supplying

the fibers used for this project.

60

Thank you!

Questions?

Contact:


Iowa State University

Phone: 515-294-4058

Email: [email protected]

mailto:[email protected]

Material Design and Structural Configuration of Link Slabs ... · 1/26/2018 · Introduction to...

Documents

Transcript of Material Design and Structural Configuration of Link Slabs ... · 1/26/2018 · Introduction to...