1

Effect of Component

Properties and Orientation on

Corrugated Container

EnduranceRoman E. Popil, Mike Schaepe

Institute of Paper Science and Technology

Atlanta Georgia

roman.popil@ipst.gatech.edu

Barry Hojjatie, Valdosta State University, GA. USA

2

Scientific objective: develop an

understanding of effects of physical

properties on box lifetime

• what factors govern the lifetime of stacked boxes

• how can we quantify the effects to predict performance ?

3

• Boxes vertically

stacked

• Experience creep,

i.e. progressive

vertical displacement

• Lifetime of boxes is

shortened by 5X

under cyclic humidity

conditions…

• This is…

“Accelerated creep”

This bottom box

has the classic

panel buckling

structure appears

ready to collapse

in about 30

minutes from this

time of snapshot

This guy won’t be

smiling too much

longer

4

Hygroexpansion of board

components changes how the box

handles load

Outside

(double face)

of the box

absorbs

moisture and

expands

largely in the

vertical

direction

Inside of the box

(single facing

linerboard)

remains the same

dimensions until it

equilibrates to the

same moisture

level

Humid outside Dry inside

5

Hygroexpansion increases the

stress

High

Humidity

outside

Low

humidity

inside

Load from stacked

boxes

Double face linerboard experiences

additional stress from its expansion from

absorbed moisture

Box wall side

view

6

50%

80%

24 hrs

time

doubleface medium

singleface

equilibrium Absorption phase

RH

Desorption phase

As outside humidity

changes, the outside

liner of the box

experiences a different

vertical stress than the

inside of the box which

is at a different moisture

level – so the load is not

evenly distributed

through the cross

section of the box wall

Phenomenon discussed

and modeled

numerically by Coffin

and Habeger JPPS

2001Pictorial representation of a sorption stress

gradient model for accelerated creep

The situation reverses when the

humidity outside decreases

7

Paper fails when combined

stresses exceed paper strength

A buckling pattern forms, the buckles eventually join to form a

crease which leads the board to fold and the box to collapse

ECT creep BCT creep

8

Experimental design

• Hypothesis: if we make the corrugated board

more uniform by having a heavier fluted

medium, stresses are more equalized get

more lifetime !!

• So, at IPST we made a series of corrugated

boards all using the same linerboard facings but

with different weights of medium ranging from

14# to 42#

9

Experimental study to determine effects

of board properties on BCT cyclic

humidity lifetime• Varied the fluted medium

basis weight from 68 to

205 gsm

• Linerboard basis weight

is constant at 205 gsm

• Boards and boxes made

at IPST Pilot Plant

• Box dimensions: 36 x 20

x 20 cm

• Selected flute size “A”

• MD shear rigidity

measured by IPST

torsion pendulum ranged

from 2.9 – 8.6 kN/m

• Static loads set for each

type of box to be at

22.5% of BCT failure load

10

Preparation of various test boxes

RSC boxes were manually

prepared from the IPST made

boards

Medium of various basis weight

were combined with 42#

linerboard using the IPST pilot

corrugator

11

Accelerated creep experiments

• Expose either corrugated board or boxes

to humidity varying from 50 to 80% RH,

keep temperature constant

• For board samples, vary humidity using a

ten hour period

• For boxes, vary the humidity from 50 to

80% RH in a 24 hour period (mimics

weather conditions od southeast US)

12

Basic experimental set-up

Vertical load @

~20% of failure load

Displacement

sensor

Data acquisition

using LabView™

records vertical

displacement with

time up to several

months

Test

specimen

Temperature maintained at 23 deg C, humidity varied from 50 to 80% RH

DAQ

13

Creep measuring equipment at IPST

For Test Boxes, inside a

programmable humidity walk-in

chamber:

BCT creep: 8 stations, 50 – 80%RH

24 hr cycle - used to obtain multiple

regression models for lifetime

prediction.

For board samples :

ECT creep, 12 stations, 50 –

80% RH 5 hour cycle - applied

to the A flute varying medium

basis weight set to compare with

BCT creep – also used for client

contract work

14

ECT creep stationsECT specimens epoxied onto horizontal edge mounts, vertical

load applied via transferring linear rail bearings and lead shot

dead weight , Hall probes measure vertical displacement

Suspended canisters filled with lead shot to 20%

ECT failure load

15

Why a ten hour period for boards ?Corrugated board humidity equilibration time test.

0

0.001

0.002

0.003

0.004

0.005

1000 6000 11000 16000 21000 26000

time (sec)

Co

mp

res

sio

n S

tra

in

26A

33A

18 723D

18A 726E

42A

16A

14A

20A

14A

18A 723D

20A

33A

unloaded corrugated board samples

conditioned at 50% RH, placed in ECT

creep apparatus, RH changed from 50

to 30% then back to 50% RH 23 deg. C.

each point 5 min.

16

BCT creep station detail

Static load transfer

through ball bearing

Combined lead bricks,

steel plates lead shot at

22.5% of BCT load inside

of acrylic box supported

by vertical sliding rails

17Test box

Static load

container

Ultrasonic proximity transducer

measures vertical displacement of load

container

BCT creep station detail (2)

18

-0.01

-0.005

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

1 202 403 604 805 1006 1207 1408 1609 1810 2011 2212 2413 2614 2815 3016 3217 3418 3619 3820 4021 4222

42A

20A

26A

18A 726E

18A723D

16A

14A

33A

creep raw data

Increasing

Compression

strain

Voltage signals

converted to

displacements

through a

calibration

Time

Top of each

cycle

corresponds

to low

humidity

Bottom of each

cycle

corresponds to

high humidity

19

Generalized creep strain curve

x1,y1

x2,y2

x3,y3x4,y4

failure

primary secondary tertiary

time

Compression

strain

20

Secondary creep predicts lifetime

2

43

43

12

12

xx

yy

xx

yy

slope

slope

Alifetime

Calculate slope from the

creep curve:

General relationship between lifetime and the creep slope:

21

ECT (Board) creep summary, 83

observations

Lifetime = 0.0241(strain/sec)-0.98

R2 = 0.92

0.E+00

1.E+05

2.E+05

3.E+05

4.E+05

5.E+05

6.E+05

7.E+05

8.E+05

9.E+05

1.E+06

1.E-08 3.E-08 5.E-08 7.E-08 9.E-08 1.E-07 1.E-07

Creep rate strain/sec

Lif

eti

me

(s

ec

)

33A

26A

20A

18A 1

16A 14A

42A 18A2

Points are averaged by box

types (medium basis

weight)

Error bars are 95%

confidence intervals

22

Lifetime = 1.47E-04x-1.33E+00

R2 = .70

0.00E+00

5.00E+05

1.00E+06

1.50E+06

2.00E+06

2.50E+06

3.00E+06

3.50E+06

4.00E+06

1.00E-08 2.00E-08 3.00E-08 4.00E-08 5.00E-08 6.00E-08 7.00E-08 8.00E-08

Creep (strain/sec)

Life

tim

e (

se

c)

33A

42A18A

723D

14A

20A

18A 726E

16A

26A

Grouping BCT creep data by medium basis weight

Much more scatter evident compared to ECT creep data – same

board set !

This box lasts 37 days on

the average

23

lifetime vs creep rate RSC BCT

Lifetime = 0.1451Creep-0.89

R2 = 0.71

0.00E+00

1.00E+06

2.00E+06

3.00E+06

4.00E+06

5.00E+06

6.00E+06

7.00E+06

0.00E+00 2.00E-08 4.00E-08 6.00E-08 8.00E-08 1.00E-07 1.20E-07 1.40E-07 1.60E-07

creep (strain/sec)

Lif

eti

me

(s

ec

)Same BCT lifetime data plotted by point

68

observations

24

Where does the variability come from ?

• Natural variation in paper properties:» Basis weight ~ 2%

» Strength ~ 5 – 7%

• Preparation of test pieces, boxes» Cutting of edges

» Mounting

» Handling: folding of box flaps, gluing of joints

» Application of load – sudden vs gradual

» Imprecision of applied load

• Systematic error ? » Electrical noise, sensor calibration drifts

» Variations in humidity control

Parallelism, squareness

25

Regression Analyses of creep data

• Attempt to quantify creep, lifetime with measured

properties – single variable:

R2

p R2

p

ect lifetime 0.59 0.03

R55 0.33 0.14 R55 0.71 0.01

ECT 0.31 0.15 ECT 0.69 0.01

BCT 0.19 0.27 SCTm 0.54 0.04

SCTm 0.50 0.05 BWm 0.44 0.07

BWm 0.39 0.1

ECT lifetime BCT lifetime

Note: a) BCT lifetime correlation with ECT lifetime

b) ECT lifetime correlation with shear rigidity R55

26

• Multiple linear regression fits of the data:

Regression Analyses

41017.73.21710.151.917.9 mm BWSCTECTslifetimeECT

51078.23.3231.6729.027.7 BCTBWSCTslifetimeBCT mm

R2 ~ -0.84, SCT of the medium

appears to dominate lifetime in this

data

MEAN ERROR ~ 1 DAY

MEAN ERROR

~ 3 DAYS

27

Another idea: rotate the components

MD of box

components

CD

hygroexpansivity

this way

is larger than in

MD

Paper is directionally oriented - not the best way for strength and

hygroexpansivity

So what if we were to make boxes with one or more of the components rotated

90 degrees ?

MDThis box stretches

and shrinks less

with changes in

humidity, so get a

smaller built-up

stresses

Load = mgF

28

This is not so easy to do:

So we have to turn the paper around 90 degrees, splice it back

into a roll and run that through the corrugator

Lateral corrugator concept (ex Schaepe, 2004 Tappi summit)

MD of

a roll

29

Manual preparation of rotated component test

boxesFull width rolls of linerboard and medium were obtained and CD strips were

spliced together to produce transverse oriented 13” width rolls to run

through the pilot corrugator – 205 gsm linerboard, 127 gsm medium

MD

MD SCT is 1.7X CD SCT

MD hygroexpansivity is 0.4X CD

30

2.7

2.9

3.1

3.3

3.5

3.7

3.9

4.1

Regular Lateral Linear Only double

face rotated

BC

T k

N

BCT increases with component rotation• 36 x 20 x 20 cm test boxes, 205 gsm liners, 127 gsm medium

• BCT increases, 14% for “partial”, 26% for “lateral”, 33% for “linear”

23 %

33 %

14 %

(Medium is

not rotated)

(All components

are rotated)

Obvious

differences here

31

Observation of creep behavior of

rotated component boxes

• 7 boxes data plotted in each case

• lateral corrugated lifetime is somewhat longer,

hygroexpansive strains are smaller as expected

• scatter in BCT lifetime data is typical

strain strain

time time

~ 3.5%

Regular Lateral

32

Effect of rotated components on lifetime

Linear

corrugated

boxes last

about 2

weeks in

cyclic

humidity at

22.5% of

BCT load

Creep and lifetime

2.00E-08

2.50E-08

3.00E-08

3.50E-08

4.00E-08

4.50E-08

5.00E-08

5.50E-08

regular lateral linear

Cre

ep

(str

ain

/sec)

1.00E+05

3.00E+05

5.00E+05

7.00E+05

9.00E+05

1.10E+06

1.30E+06

1.50E+06

1.70E+06

1.90E+06

Lif

eti

me (

sec)

creepratelifetime

!

33

Pilot coating/corrugating trial - sequence

Rod coating of 42#

linerboard at Spectra-kote

Linerboard rolls coated

with polymer or

clay/polymer coatings

Rolls sent through

IPST pilot single facer

combined with WAM

Single facings

manually combined

with double facing

14 x 8 x 8” test

boxes manually

made

Coated test boxes put

through performance tests,

e.g. lifetime in cyclic RH

34

Relationship between hygro strain and creep

• Urbanik 1995 Wood Fibre Sci. hygroexpansion drives

mechanosorptive creep in cyclic humidity

• reduce hygroexpansive strain either through component rotation

of low WVTR coatings, creep rate goes down, lifetime increase

0

2

4

6

8

10

12

14

0 1 2 3 4 5 6 7 8

Creep rate (1/s x 10- 8

)

Hyg

roe

xp

an

sio

n s

tra

in (

mm

/mm

x1

0-

3 )

RSC series II

HSC series III

regular

linear lateral

uncoated

barrier coated linerboards

Lifetime < 1.5 weeks

Lifetime > 3 months

35

Summary

• Lifetime and creep observed for ECT and BCT, ECT creep dependencies are more clear

• Effect of medium properties characterized: fluting SCT and basis weight for BCT lifetime

• Effects of component rotation on lifetime observed, linearly corrugated lifetime is 2x higher,

• Low WVTR coated boxes last months compared to days

• Generally equalization of properties through the board cross section increases lifetime