. . . . , , . 1-2 2007 Tech. Chron. Sci. J.

TCG, I, No 1-2 21

.. ...

. . . .
... , ...

( -
(, ) ),
, , -
(..) , , - .
. , ,

-
,
, ,
, ,
, -
. -
/ , - [1] (. 1). -
(-) ,
, . . ,
, ,
,
,
..,
E-,
.
. -
. ,
.

1.
,

(..), . ,
80, -
-
. , . -
,

, , . -
1985. 70 , , -
pilotis,
. ,
, -
, - [2, 3].

. , -
, ,
, ,
/
: 4.12.2006 : 15.3.2007
 22 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2

. ) ,
(-),
u r ,o 0.25 rs 1 X 2 X

(-), ,
/
, 0.5 cr
0 .5 (1.1)



/ ,
Db
, cr




, 20.

(pitting corrosion).


-
20.


su
, cor
, -
-


-

,
,
su,



,

)
(pitting
corrosion).
Coronelli and
Gambarova [7] (. -
(1.2)),

,
)

pit
,

su
cor
,

)

)


,
su,
0.1

[1,4,5].
0.5.
Coronelli and Gambarova [7] (. (1.2)), pit


, [1,4,5].
-,

= pitmax
pit .

, -,
, - 0.1 0.5.
a pit

,
,
,



cor
su max
sy
su sy 1
r pit
, r pit
pit =
1 (1.2).
- pit . a max
pit

[3,6].

[3,6]. -

,

, 10%
cor
su
sy su sy 1 r pit , r pit
a pit
1 (1.2).
, , 2. a max
pit
.
10% .
Ag = ..
As/st = .
Rcr As2.cor
/stcor =
.
X C b,Ad ==
.. .
c / Cg c = .
Db A = .
Db s/st = .
Ascor/stcor =
Ec /s /f = ,
.
c .
b,
fadh d = .
= .
fb c / C = .
c = .
D
cor
fb b/ enh
=
= .

1:1:





- Ec /s /f =

,
-.

.
. .
Figure1:1:Rust buildup
fc / fct =
Figure builduparound
aroundthe
thesteel barbar
steel andand
definition of corro-
definition of fadh = .
sion indices. .
corrosion indices. fb
fs,cr ==
.
.
f /
cor enh
=
fy,s b/ y,st =



1.1
. -.
1.1 fy,stfcres/ fct = =


.


, fs,cr .
= .



hrfy,s / y,st = = .


. ,
Icr/gr =
.

- fy,stres .
=
,.
,


Lb,st =
.
.
, =Db/Db ,
, (.1).
crh ==
.
.
r

, Vr,


, LsI / == .

cr gr
,

=Db/Db (.
1).
,
, VVs, r
, Mcr /y = .
.
Vr =
rs
V NbL/cb = =

..
, s.

, V , rs,
,s 2

P b,st = =
. ...
4, Rcr cr = .

V
= rs Vs.
,
, rs, - L = .

r s s = .

2 4,.

Mcr /y = .
, Vr
. ur,o =
, Nb /cb = .
, ,
, .
P
Vanch = =


...

,
Vr , ur,o,
- Rcr = .
( cr u , .
, , r,o
s
Viflex = =

.


( )
, - ur,o =
cr .
u r ,o 0.25 rs 1 X 2 X Vr = .
0.5 cr 0.5 (1.1) Vshear/c/s = ,
Db cr
.
 , Vs. .
,

[8,9]

. ,

. . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No
1-2 23
,

. .
V
Ief I
Vanch = Mu cr
I
. My gr

Viflex = Mcr
.
Ls
Vr = . Mcr
Vshear/c/s = ,
. cr y
u
wcr = . My 1 w
X = . stcor c
Vs = . Lb ,
2:
2:


.
.
eff = . Figure 2: Column response to lateral sway.

Figure 2: Column response to lateral sway.
y,flex/y,slip = .
pit = .
rs = . 3.1
cr = . , VSHEARRES
su cor
= .
sy/su = .
y /pl = . , Vshear, -
= . , Vc, -
= . , Vs. ,
y/u/pl = ,
. [8,9] -
sv / fv = -. -
lat,f = -. . -
n = ,
. ,
shr = . -
y/u/pl = , Moersch ( ,
Moersch ,
( ,
,
Vs),
. Vs),



,
,
sv =
-.





3. / fv
Vc
lat,f = -. Vc ,,
.
n
..
= .






. (.
(. 3),
3),


shr=
. ( ), ( ),

Moehle .. [9] :
.., Moehle .. [9] :
, Vshear ( ) V s Vc
3. V shear ( ) V s Vc (3.2)
. V s k st Ast d / s f y ,st
..
V s k st Ast d / s f y ,st (3.2)
(.
2)

Vc Rd K ( 1.2 40 sl ) 0.15 P / Ag bd
Vc Rd K ( 1.2 40 sl ) 0.15 P / Ag bd


1.15 0.075 , 0.7 1

.
.., 1.15 0.075 , 0.7 1
,


-

. ,
(3.1),
Vshear

: Viflex = Mu /Ls
.

, Vshear

, Vanch
(. 2) : VRd
/
.
V = M / Ls : Vsd

iflex u,


. (3.1), Vshear -
,

,
Vanch


(L

s
).


(3.1)
/ Viflex = M
u /
Ls -

2

,



, ,
.
(Ls
3:3:

VshearV
shear

V = min {V ).
, Vanch
, Viflex} (3.1) (3.1) .
.
,
u,lim shear
Figure 3: Shear degradation owing to displacement ductility.
Figure 3: Shear degradation owing to displacement ductility.
.
40s1 . (3.2)
Vu,lim = min {Vshear, Vanch, Viflex} (3.1)

(s1 <2%),
3.1 , Vshearres
,
, Vshear, , Ag
 24 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2

40s1 . (3.2) , -,
-
(s1 <2%), (. -
, - =1). -
, Ag Vs=nstAstfy,st, nst
P .
Vs, s fy,st 45o. , , s
, d , d nst=0
st kst Vs=0,
1 -
(1 ). - [10].
, ,
4
- 3.2 , Vanchres
. Vs
,


, Astcor=
Ast(1-X)2.
-
(

Ast = Ast(1-X)
cor
2 ,
.

, ) ,

,
n,

,
, fb=(2/)
n
+fadh,
, Db,

w cr
= D b
( X
rs
-1) , [7], ,
f b
, , fadh
(fadh
1MPa).


,
wcr = Db (rs
-1) X , [7], -


(


,
, ,
(. 5),)
,
, n,nD
b
-

,
(.
4): ,

, : Dbi),

1 w 1

Db (. 4):
, fadh (fadh 1MPa).
, c = fct c,
-
stcor cr rs 1 X fct= 0.5(f c)
0.5

(. 5),

(3.3) 1 wLcrb ,st 1 LDbb,st
cor
st rs 1 X (3.3)

nD b



:
Lb ,st Lb ,st
i) (1
2 ,

,
c = fctc,
fct = 0.5(fc )0.5


ii)
),

)

(1
2
Pst,y ) , st = ,
st fy,st /(sNb),
0.33

fy,st
), st s
ii) - b
wcr
fy,stres
iii) ,
Pst,x , st = st fy,st /(sNb), 0.33
shr=3fct [11]. ,
st s b


,

sy
stcor iii) ,

2
c shr=3f aAct st[11].
f y ,st,

4:4:))

) )
f f
b

ct shr
-



.
. Db sN b Db
Figure , (3.5)
Figure4:4:a)a)Stretching
Stretchingdue
duetotosplitting
splittingcracking
crackingalong
alongthethe
main
main Ast f y ,st
bars and b) effect of corrosion on yielding of hoops. c
bars and b) effect of corrosion on yielding of hoops. f b 0.29 2 c f c 0.86 f c 0st.19y ,st aA f
fb
D f ct shr sN D (3.5)



- bDb sN b Db b b

(.. , Ast f
c y ,st
(..


=3),=3),
L L

b,st



29 f c
f bn, 0. 0.86 f c 0. 19
b,st
(. Db 5,) sN b Db
.
. -,
-, , -
,

fy,st, fy,st
,
- , ,
, . (1.1)
-

, fy,st,
res
, fy,stres
,
. (3.4).
.
n,
(3.4).
,
,
- (. 5,)
cor cor res .

V cor = k A
cor
f Vress d = / s.kst Ast fy,st d / s. , , . (1.1)
s st st y,st Tastani and Pantazopoulou [12]

sEs stcor
fy f,sty ,st E stcor 0 0 stcor
stcor sy sy ,
f yres
f,styres
(3.4)


, -
,st corcor (3.4)
0, 0, st st sy sy . Tastani
2( X )
and Pantazopoulou [12]
f bcor






,
, -, (3.6)
C R
c C Rcr Astcor f yred,st

cr
f ct c
D shr
(. b c Db N b s

=1).
(3.6), Rcr
. . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2 25

) fb n ) )
n
rmax
nDbLb

n(X) smmax smres

Lb fct c Lb
Ast f st , y Xu X
sN b ur,olim ur, Xshr
5: ) , ) ) -
.
Figure 5:a) Representation of frictional model for bond, b) response of concrete cover to radial displacement and c) coefficient of friction
as a function of corrosion penetration.

, 3.3 , Viflexres
2( X )
f bcor (3.6)

.. ( . -
C R C c Rcr Astcor f yred
c cr
f ct shr
,st )
D c Db N b s , sicor = Asi (1-X)2.
b

(3.6), Rcr (. (1.2)),
(. 1), - , res, ,
(. (1.1)) Viflexres, -
Rcr=ur,oCc/[ur,o+cr(Cc - 0.5Db)], Cc=c+0.5Db. - , i ,
, (),
h h
(. 5):
max
M res f ci Ac ,i 2 yi f si Asicor 2 y si (3.8)

f ci Ac ,i f sicor Asicor
(rmax M res
res
, smmax ). Viflex , P res
Ls
shr 0.001 max 20%,
,
cu=0.004,
. . fci, fsicor
, smres, Aci, Asicor -
i- .
( r(Xu=hr/0.5Db)=smres hr
). - -
[13,14,15] : rmax c=0.002 ()
0.9-1.4, smmax 0.15-0.25 smres 0.05. cu=0.004.

. , .
fbcor . (3.6), fcres,
Fbcor = Db Lb fbcor, - , fc, -
-
[7]:
( jd -
1 N D 1X
) , f cres f c , 1 cb b rs
(3.9) 1 0.1 1 co b

res Db Lb f bcor N b jd P( d 0.5 h )

Vanch (3.7) 1 , -
Ls
,

 26 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2

- g
, cb b . b
. ( , b=0.8
b=0.5 -
- ). , My
lat,stcor - ,
b, ,
( Poisson) , .

fcc fc 3.1
cor cor
lat ,st y lat ,st x (3.10)

cr (. 6),
2 i- , wi,
-
fcc fc 1.5 kstconf sv cor res
f y ,st
i- ,

kstconf
My

-
, cb , f f bcor b, ,
, ( cr
s ,cr
cr
b . wi Es ,cr f s , y
; f s (3.13)

, [16]), svcor Db E s
.

res
fy,st

lat,stcor(. (3.4)).

(3.13)(.
6),
-
i-
cr
/ 4,
, wi, cr
( Poisson) , fs,cr
(. 6).
3.4
,
fcc fc 3.1
cor
lat cor

,st y lat ,st x


,
i-

2 (3.10)
f f cor (3.14)
cr(E s / Eco )cr V bi ( d ;x )f/s ,cr

wi fs ,cr I cr f
s ,cr
conf cor s ,y (3.13)
res Es Db E s
fcc fc 1.5 kst
sv
f y ,st . x d

conf , - , i
kst (3.13)
(cr, . 6) , i- .. ( = i
, cr / 4, i
. cr),
V
Ls. 6).-
( , [16]), svcor (.

,
,

fis,cr i

fy,st
res ,
= wi / (d-x),
,

(. (3.4)). . i,flex = i i (. 6).
. 2, , y, i,flex
V x

3.4 -
Ls, y,flex,
i fs
, y,slip ,

-


.

i-1
cr/4 fb
Ls i fb
,
,
(cr, . 6) wi/2

,
y y , flex
y ,slip (3.11) cr fs,cr
i-
.1 y Db f y ,, d
2 s
y , flex
y L,
s , y ,slip
cor
L
s
3 8 f b .
,
. 2,
, , 6: ..
, yy, Figure 6: Displacement indices along a R.C. column.
[17] ,

, -
Ls, y,flex,
, 2 y,slip, 6: ..
1 1 M 6:
Figure
Displacement ,
indicesalong ,
a R.C. column.
y
My My 1 cr
b cr MC 90 [11] =D /(3.6 ) =As/Ac,eff,
E co I ef E
co I cr I crI gr M y ,

cr
fA ( E s /E
s ,cr i ( db
co ) V x ) /s,eff
I cr s,eff
(3.14)
c,eff
y y , Iflex
gr y ,slipP As.
(3.12)
M cr f ct x d
01.5 h 2 Ag y Db f y ,s (3.11) 3.6
2fb / fct,
, i
y , flex y L s , y ,slip Ls 2fbcor/f . ,
-
3 8 fb cor i-
ct
.. (i = i cr), V -

(3.12), My Mcr
L (
s. i
, y,
, Igr Icr , i = wi
/ (d-x), ).


[17] ,
, h
s n=L / ,
cr i,flex = i i

My

My 1 1 1 M
2 (. 6). ,
y b cr , cr,
Eco I ef Eco I cr I cr I gr My
(3.12)
MC 90 [11] cr =Db /(3.6s,eff)
I gr s,eff=As/Ac,eff, Ac,eff
P
M cr f ct
0 .5 h A
 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2 27

[12], .. ,
1
L ( L cr ) u y , flex y ,slip p , flex p ,slip (3.18)
cor cr s s (3.15) dx
y , flex
2Es
f
E V 2L y ,s L s Db y ,s
s s cr f bcor 3 8 f cor
b
E co I cr 3 cr Db ( d x )

(3.15)
corr
l p su sy b
corr
D 1.2 E h su sy
2


y,flex ( . (3.11)), - 4 f bcor

.. . (3.18)
. (3.12) (3.15) - -
b, - .. -
.. , ,
2 , fbcor
My I gr I cr 1 3 E
b cor co
M I I I y , flex 2 M (3.16) fb.
cr gr cr cr L s y
, (..
),
,
3.5 fbcor . (3.18).
( 3.4, 3.5) -
,
, u, , -
.. - -
, , /
. , [2].
7 , ,
: u = y + pl. -
(0)
(. (3.11) (3.15)) =6.8% - (B2cR). 7
, y = y / Ls. -
.. ,
[17], -

.
(. (1.2)) ,
,


.
,
(0)

1 =6.8%
-
p:
, flex u pl
= l py , + plpl. cor
su
sy

x






- (B2cR).



.
d
(3.17)
2 E
D(. 1.(3.11)b

(3.15))
h su
cor
sy 30

,
p ,slip 4y = y / Ls. cor
pl
fb 25
.. [17], 0
1. 2f y ,s
Db
(kN)

20 (X=0): fb=6MPa
l p Ls (. (1.2)) ,
41 f bcor 15

p , flex pl l p , su
cor cor
pl sy
su E h / sy
Es
1
dx
10


cor cor E /E
Dyb,s 1.2 Esuh su sy hsy s
5 2cR
(X=6.8%): fbcor=3MPa
p ,slip pl
4 cor 0
fb
0 5 10 15 20 25 30 35 40
1.2l p f
(3.17)

y ,s Db (mm)
l p Ls Eh
1 f bcor (
4 - 7: 7:


.

).
cor

su sy E h / E s

- Figure
, Figure
.
7:Verification of the methodology for assessment of rotation
7: Verification of the methodology for assessment of
capacity.

y ,s su cor

sy E h / E s rotation capacity.

lp ,
Eh
(
28 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2

4. :
- ,

() . ,
3 -
, Mcor=MRd(2-X)X, MRd
- . , ,
.. ,
M cor
cor X ( 2 X ) Rd (4.1)
-, bd 2 f c

-
, ,
[18] eff =0.004,
[18]. , - ( -
, ),
, , , - .
(
) . , = x/d, : eff=0.004 -
-, cu=0.004
low=0.004(1+d/d)/ (0.004+0.004)=0.5(1+d/d) d
)
.., ) - . -
(bal 0.64),
, ,
. , .
- low
- ,
. ) cor (. (4.1)). -
- , - , -
low (
, eff),
,
[5]. , . (4.1).

, - .
,
, -
. , - -,
-
[19]. (, -
). Viflexenh
, , fcc,
4.1 , Viflexenh , cc,
,

-
f cc f c 1.5 k stconf sv f y ,st k conf
cor red
f fv E f eff (4.2)

( ) k stconf sv f y ,st k conf
cor red
f fv E f eff
.. - cc co 0.015
fc

[20]. - eff
- 0.004 0.5fu,d ,
fu,d
. , [18]. kfconf =1-(b2+h2)/(3Ag(1-s))
 Db c


(4.5) 0.94%(66.5

Astcor f yred
,st 2n f t f E f eff

Db N b s Db N b

. . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2 29

- (,
(b h 5.


) 8:
), fv = . , - Figure 8: Up
2ntf(b+h)/bh - ,
-
/


n tf f - -,
..,
. -

, 2 3,

, cc,u, ( cu=0.004 Vw,f
(. 8).

cc,u cu) : .
Ls=1500mm,

400400mm 1020
k cor f red k conf E 4.3
(D
b=20mm)
,
Vanch

enh
st sv y,st f fv f eff
cc,u cu 0.075 0.1 (4.3) c=40mm
fc (. (4.1))
8/100mm S400 (fy=400MPa

fu=500MPa). (3.6),
-

,

sv=1.04%.




fc = 20MPa, :



4.2 , Vshearenh

( kstconf=0.5)
fcc = 22MPa.
-



,

,



fb =7.10MPa

- ( rmax
, =0.9).

.

-,


, ,

Vw,f =153k
,
Lb=500mm.
,
( , -




, cu=0.004

-

Viflexenh
V iflex=159kN

[2].
-
,
u=2.7%.

.
(3.2)
) - V
shear=248kN,
,
Vanch =223kN (. (3.7)).
- . .
=5% pit=0.1 (apit,max=0.5).
. - (3.6), , fbenh, -

( . (4.5). 2rres(X=0.0
),
, . (3.2) -

nflong=2
- ,

Vshearenh ( q ) min( q old ),(q new ) Vshear res
Vw, f

(4.4) 1020
Vw, f lat , f b h , lat , f 2k vf nt f E f eff / b
8/100
nfanch=5 nfshear=2

1.15 0.075 , 0.7 1 nfanch=5.

).
1500

q , lat,f
,


, kfv -
(
kfv=1) eff
4.1.
(4.4)
-, 8: -.
- Figure 8: Upgrading of column through E-FRP jackets.
. . (4.4), , [12],
[22] , , eff,
qnew
3.5 0.002. , /
[21]. . ,
(4.4) . (4.5) -
, , . (3.6).
30 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2

, -

2 ( X ) C c Rcr C Rcr
f benh f ct c shr . -
Db c (4.5) :

ffu=3500MPa fu=0.015,
Ef=230GPa tf=0.13mm.
Astcor f yred 2 n t E cor=27kNm (. (4.1))
,st f f f eff
nflong=1.58 2
Db N b s D Nb 400mm -
b

. -
Viflexenh156kN.
,
5. =2.5. (4.4),

- Vw,f =153kN, nfshear=1.6 (
/ 2 - -
.., - -
, ). ,
(. 8). -
Ls=1500mm, (fbenh = fb). ,
400400mm 1020 (Db=20mm) ,
,
c=40mm 8/100mm - 2rres(X=0.05)/=0.3.
S400 (fy=400MPa fu=500MPa). , -
Vanchenh=212kN
sv=1.04%. fc = 20MPa, (95%Vanch). -
(-
kstconf=0.5) fcc = 22MPa. ,
, ,
fb =7.10MPa ( rmax=0.9). -, =11.3MPa.

, Lb=500mm. nfanch=5.2 ( 5 -
- ). -
cu=0.004 ,
Viflex=159kN u=2.7%. -
. (3.2) Vshear=248kN,

Vanch =223kN (. (3.7)). - [23].
=5% pit=0.1 (apit,max=0.5). - , (. (3.18)),
-
18% . (1.2). , u=2.16%.
: Viflexres
= 136kN (Viflexres / Viflex=14.5%), Vshearres=105kN (Vshearres
/ Vshear=57.7%), fbcor=2MPa Vanchres =66.5kN (Vanchres 6.
/Vanch =70.18%).
yres=0.94%, 0.94%(105kN/
136kN)=0.73%
0.94%(66.5kN/136kN)=0.46%
.
.., -
, . ,
-
, . .

. . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2 31

() Engineering, Vol. 130, No. (8, pp. 12141224.

8. Martin-Perez B. and Pantazopoulou S. (2001). Effect of bond,
. aggregate interlock and dowel action on the shear-strength degradation of
, r.c. Elsevier Engineering Structures, Vol. 23, pp. 214-227.
/ , 9. Moehle J., Elwood K. and Sezen H. (2002). Gravity load collapse
, of building frames during earthquakes. Proceedings of S. Uzumeri
Symposium on Behavior and design of concrete structures for seismic
. performance, ACI SP-197, pp. 215-238.
10. Syntzirma D. and Pantazopoulou S. (2006). Deformation capacity
, of r.c. members with brittle details under cyclic loads. ACI Special
, , Publication Cyclic Shear, edited by ACI committee 445 (shear and
torsion), in press.
. 11. Tastani S.P. and Pantazopoulou S.J (2005). Recovery f seismic
, , - resistance in corrosion-damaged reinforced concrete through FRP jacketing.
International Journal of Materials and Product Technology, Vol. 23,
/ - Nos. 3/4, pp. 389 -415.
12. Priestley M., Seible F. and Calvi M. (1996). Seismic design and
. retrofit of bridges. J. Wiley and Sons Inc., N.York.
13. ACI 318-02 (2002). Building code requirements for structural
concrete & commentary, American Concrete Institute, Detroit,
Michigan.
6. 14. Lundgren K. (2002). Modelling the effect of corrosion on bond in
r.c. Magazine of Concrete Research, Vol. 54, No. 3, pp. 165-173.
1. Pantazopoulou S.J. and Papoulia K.D. (2001). Modelling of cover- 15. (2000).
cracking due to reinforcement corrosion in RC structures. ASCE Journal , -, 2001.
of Engineering Mechanics, Vol. 127, No. 4, pp. 342-351. 16. CEB-FIP MC 90 (1993). Model Code 1990. Thomas Telford
2. . (2006). Pubs., London.
. , . 17. FIB Bulletin No. 24 (2003). Seismic assessment and retrofit
, , . 258 of r.c. buildings. Technical report prepared by T.G. 7.1, International
3. Bousias N., Trianatafillou T., Fardis M., Spathis L. and ORegan Federation for Structural Concrete (fib).
B. (2004). Fiber-reinforced polymer retrofitting of rectangular RC columns 18. ACI 440.2R-02 (2002). Guide for the design and construction of
with or without corrosion. ACI Structural Journal, Vol. 101, No. 4, pp. externally bonded FRP systems for strengthening concrete structures.
512-520. American Concrete Institute, Detroit, Michigan.
4. Pantazopoulou S.J., Bonacci J.F., Sheikh S., Thomas M.D.A., and 19. . . (2006).
Hearn N. (2001). Repair of corrosion-damaged columns with FRP wraps.
ASCE J. of Composites for Construction, Vol. 5, No. 1, pp. 3-11. (). 15o , .
5. Tastani S. and Pantazopoulou S. (2004). Experimental evaluation of 20. (2000). . -
FRP jackets in upgrading r.c. corroded columns with substandard detailing. , 2001.
Elsevier Engineering Structures, Vol. 26, No. 6, pp. 817-829. 21. .. .. (2006).
6. Lee. H., Kage T., Noguchi T. and Tomosawa F. (2003). An - : .
experimental study on the retrofitting effects of reinforced concrete 15o , .
columns damaged by rebar corrosion strengthened with carbon fiber sheets. 22. Tastani S. and Pantazopoulou J. Fiber reinforced polymers in
Cement and Concrete Research, Vol. 33, No. 4, pp.563570. seismic upgrading of existing reinforced concrete structures. Proceedings
7. Coronelli D. and Gambarova P.G. (2004). Structural assessment 8th U.S. National Conference on Earthquake Engineering (8NCEE),
of corroded r.c. beams: modeling guidelines. ASCE J. of Structural San Francisco, California.

.. ,
, ..., 67100 , stastani@civil.duth.gr
. . ,
, ..., 67100 , pantaz@civil.duth.gr
32 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2
Extended summary
Seismic Assessment and Upgrading of corrosion-damaged
R.C. members through FRP jacketing
S.P. TASTANI
Ph.D., M.Sc. Civil Engineer D.U.Th.
Abstract
This paper explores the structural behavior of corroded reinforced
concrete (r.c.) members and the efficacy of repair schemes that
include FRP jacketing, both as a means of curbing the corrosion
process and for strength recovery. Corrosion of steel reinforcement
is a critical issue in the assessment of old construction, because its
harmful influence on structural stiffness, strength and ductility may
drastically limit the service life of the exposed construction. In as-
sessing the residual strength and deformation capacity of corroded
structures, several parameters are considered, such as reduction
in bar diameter owing to iron depletion and the consequent loss of
flexural strength, attenuation of rib height in the case of ribbed bars
and deterioration of friction between reinforcement and concrete,
both of which have detrimental effects on bond strength. The ex-
pansive tendency of corrosion products is the basic vehicle driving
the proposed repair/strengthening methods, which combine various
techniques with the use of FRP jackets. Simplified design expres-
sions obtained from first principles are proposed for assessment
and redesign applications that link FRP properties with attainable
strength, corrosion indices and target deformation capacity of the
structural member.
1. INTRODUCTION
Reinforced concrete structures built prior to the introduc-
tion of modern seismic requirements abound throughout the
developed world. The modern form of reinforcement detail-
ing became a required practice in the early 1980s, as a result
of a sweeping change in most International Design Codes.
Structures designed prior to that point conform to earlier
generations of design codes as these evolved from the early
1900s to the 1980s. By todays standards, typically these
are low-ductility systems often combining non-seismic de-
tails with compounded corrosion damage owing to long term
exposure to aggressive environments. Realistic management
of the seismic risk presented by this class of structures is a
pressing priority and an issue of great practical interest.
Assessing the implications of corrosion on strength and
deformation capacity of existing r.c. members, and develop-
Submitted: Dec. 4. 2006 Accepted: Mar. 15, 2007
S.J. PANTAZOPOULOU
Professor, Civil Engineering Dept., D.U.Th.
ment of seismic upgrading / rehabilitation methods with FRP
materials is the focus of the present work.
Corrosion products are a mix of ferrous and ferric oxides
which generally occupy a multiple volume of the depleted
parent metal; the exact volume of rust depends on its compo-
sition [1]. Accumulation of rust affects strength, deformation
capacity and the mode of failure in concrete through several
mechanisms. These include reduction of reinforcement area,
spalling of the concrete cover, break-down of bond mecha-
nism between concrete and reinforcement resulting in the re-
duced resistance of bar against slippage and embrittlement of
reinforcement. Bar section loss and cover spalling have di-
rect effects on all strength mechanisms (flexure, shear, bond)
and on the associated deformation capacities. A second level
implication is the change in the design mode of failure that
eventually controls seismic performance [2,3].
Implications of corrosion damage in terms of strength
are estimated from basic mechanics, given the extent of
metal depletion and strain embrittlement of affected steel.
Metal depletion is quantified through the depth of corrosion
penetration, X, i.e. the percent loss of bar diameter (figure
1). Given the volumetric relation between the deposited
rust and the consumed metal, and assuming that corrosion
is uniform over the bar and that rust may be deposited in
the radial cracks of the cover, the radial displacement ur,o
the rust imposes on the internal concrete boundary may be
obtained using geometry (eq. (1.1)). Embrittlement refers
to a loss in deformation capacity of the reinforcement and
it is of severe degree under pitting corrosion. The reduced
deformation capacity of reinforcement is determined by a
linear attenuation model proposed by Coronelli and Gam-
barova [7] (eq. (1.2)).
During the last two decades, intensive research has been
conducted into the rehabilitation of corrosion-damaged con-
crete structures with FRP jacketing. A motivating premise
was the chemical inertness of these materials to the usual
corrosion agents affecting conventional steel. Since the early
efforts it has been established that, apart from its efficacy as a
confining medium, FRP jacketing can also slow down mark-
. . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2
edly the chemical process of iron depletion under conditions
of continued post-repair exposure [1, 4, 5].
Laboratory experiments on severely corroded column
specimens under loads simulating earthquake effects have
illustrated that by proper detailing FRP repairs may success-
fully enhance ductility and deformation capacity, recover
strength and hysteretic damping properties of corrosion-
damaged reinforced concrete to levels comparable to the
uncorroded condition [3, 6].
2. ASSESSMENT OF CORRODED R.C.
MEMBERS
Seismic upgrading of a corroded r.c. element with FRP
jacketing aims to mitigate the occurrence of premature
modes of failure other than flexural, which is the least un-
desirable. With reference to a simple cantilever model that
represents half the span of a continuous member in a frame
structure under lateral sway (figure 2), forces used in rede-
sign must satisfy the qualitative relationship (3.1), where
Vshear is the nominal shear strength, Vanch the shear force
acting in the span when the anchorage/lap-splice reaches its
development capacity, Viflex = Mu/Ls is the seismic shear force
required to develop the ideal flexural strength of the member
and Ls is the shear span. Equation (3.1) is used in the assess-
ment of dependable capacity and also in dimensioning the
upgrading scheme.
2.1 Shear strength of corroded member
The nominal shear resistance (Vn) of r.c. members com-
prises concrete and web-steel contributions (Vc and Vs). Un-
der cyclic displacement reversals both terms degrade with
increasing intensity of imposed displacement ductility, ,
owing to the susceptibility of concretes strength to cracking
[8, 9]. Upon reversed seismic loading bi-diagonal cracks in
the web limit the strength of the compression struts that sus-
tain the Vs term, whereas aggregate interlock and dowel ac-
tion (the Vc term) break down quickly with increasing width
of cracks. A simple model proposed by Moehle et.al. [9] is
adopted to qualitatively describe the relationship between
the aforementioned variables while maintaining the familiar
code format for Vshear (eq. (3.2), figure 3).
Stirrups are the first to be affected by corrosion, owing to
their proximity to the exposed surface, whereas the section
loss is more dramatic for the transverse reinforcement as it
is usually fabricated from small-diameter bars. Thus, the Vs
term is lowered by the apparent reduction of the cross-sec-
tional area of stirrups. Also, opening of splitting cracks [7]
due to corrosion along the longitudinal reinforcement results
in stretching of the stirrup legs (figure 4), consuming part of
33
the available pre-yield deformation capacity of the metal (eq.
(3.3)). For this reason, the web contribution Vs in equation
(3.2) is calculated using the reduced rather than the nominal
yield strength (eq. (3.4)).
2.2 Bond of corroded anchorages
Bond degradation due to corrosion leads to a dramatic
loss of stiffness in the member even under serviceability
conditions. The frictional analog is adopted to model bond
both before and after rehabilitation with FRPs. Bond
strength fb is proportional to 2n/ where is the frictional
coefficient and n the normal confining pressure over the
anchorage length (figure 5a), provided either by the cover
concrete (supported by hoop tension stresses in the cover),
by transverse reinforcement crossing the splitting crack path,
or by transverse compressive stress fields existing in the an-
chorage region. In this light, it is evident that rust accumula-
tion, being an expansive process, generates radial stress over
the lateral bar surface that competes with bond action for the
tensile resistance of the concrete cover for equilibrium in the
hoop direction.
Taking into account that corrosion negatively affects the
stirrup contribution and the coefficient of friction (by deple-
tion of ribs) the residual bond strength fbcor is given by equa-
tion (3.6). Values for the various frictional coefficients (for
ribbed or smooth bars) have been quantified indirectly from
tests published in the literature [12, 13, 14]. For intermediate
levels of rib depletion X a linear interpolation may be assumed
for the coefficient of friction as illustrated in figure 5c.
From the estimated bond strength of corroded bars (eq.
(3.6)), the development capacity of corroded anchorages
is calculated, mobilizing equilibrium along the anchorage
length, and from the corresponding flexural moment (evalu-
ated from the equilibrium of sectional forces) the force in the
shear span is resolved using statics (eq. (3.7).
2.3 Residual flexural strength
For load combinations below the balance point (i.e. when
axial load is less than 45% of the cross-sectional compres-
sion capacity), the residual flexural strength of corroded
members is controlled by the yield capacity of tension re-
inforcement. Residual capacity is calculated at a limiting
compressive concrete strain of 0.004, which corresponds to
cover spalling. The reduced cross-sectional area of longitu-
dinal reinforcing bars is expressed as a function of corrosion
penetration and the post-yielding steel stress fsi is evaluated
considering equation (1.2).
The concrete stress-strain curve follows a parabolic
relationship up to peak value (0.002) and remains constant
34 . . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2
thereafter until failure (0.004). The concrete strength of cov-
er layers is reduced from its nominal value due to transverse
splitting cracks (eq. (3.9)), while that of layers confined by
stirrups is increased owing to the exerted confining pressure,
which is also reduced due to corrosion (eq. (3.10)).
2.4 Flexural ductility and rotation capacity
The primary cause of degradation of flexural stiffness in
corroded members is the partial loss of bond; without bond
tension stiffening of concrete between successive cracks can-
not be mobilized, whereas crack widths become excessive.
The effect in the case of corrosion along the anchorage of
primary reinforcement is similar, for in that case pullout due
to slip may govern the total lateral drift. With reference to
figure 6, at the serviceability limit state member deflections
comprise contributions from distributed curvature along the
member length and lumped rotation at the face of the sup-
port due to pullout (eq. (3.11)). Yield curvature, y, may be
calculated at the cracked stage (where the applied moment is
less or equal to the yield moment) as a function of sectional
stiffness (eq. (3.12)).
Since slippage is facilitated with corrosion, the lumped
rotation occurring at crack locations is mostly responsible
for tip flexural deflection of the cantilever. Assuming uni-
form bond stress distribution over the mean cracking dis-
tance cr, and mobilizing force equilibrium along the 0.5cr
and integration of lumped rotation at each crack location
upon stabilized cracking, total flexural tip deflection may be
obtained (eq. (3.15)).
Rotation capacity (drift), u, of the corroded r.c. member
is obtained from flexural and pullout contributions. Each of
these contributions comprises elastic and plastic components
as u=y+pl.
These terms are calculated from first principles, consid-
ering the reduced deformation capacity of the reinforcement
due to embrittlement and the deteriorating bond strength
(eq. (3.18)). If equation (3.18) is used for the assessment
of the rotation capacity of an upgrading member with FRP,
whereas repair includes also replacement of deteriorated
concrete in the anchorage of the corroded bars, then the
enhanced bond strength rather than the reduced value due to
corrosion should be used. However, if the anchorage length
of corroded bars is inaccessible, so that no intervention may
be possible, then the reduced value of bond strength is used.
3. STRENGTH RECOVERY THROUGH
FRP JACKETING
From the assessment procedures described in the preced-
ing text, it is clear that all strength and deformation compo-
nents of a r.c. member entering equation (3.1) are affected by
corrosion. The objective of this section is to detail the contri-
bution of the FRP jacketing when this is used to upgrade the
strength and deformation capacity of affected members. As a
confining device, FRP jackets impart toughness and enhance
shear, anchorage and compression capacity of the encased
concrete member [18]. Externally bonded FRP strips may
also be applied as additional tension reinforcement for the
recovery of flexural strength, provided they are properly
anchored. However, apart from these favorable aspects of
FRP jacketing, the following additional attributes need be
considered in formulating the rehabilitation framework: i)
Jackets have no influence on lateral stiffness. ii) By effec-
tively reducing shear cracking in the plastic hinge regions,
all deformation is forced to occur within few flexural cracks
thereby promoting large strain demands in the embedded
longitudinal reinforcement. This may lead to bar fracture
unless the rehabilitation framework includes measures for
stiffening the affected structure. iii) Jackets are suscep-
tible to rupture at points of localized deformation demand
and hence, although they effectively postpone buckling of
compression reinforcement to higher levels of deformation,
they cannot altogether prevent it, particularly if stirrups have
wasted away due to corrosion [5]. For the reasons stated it
is concluded that FRP jacketing may be considered as a lo-
cal intervention for seismic upgrading, in that jackets may
successfully increase strength and deformation indices of an
individual corroded member without, however, controlling
global demands.
3.1 Recovery of flexural strength
Experimental studies have shown that an effective
method of adding longitudinal FRP reinforcement in flexural
members is by mounting the bars or laminates in surface-
cut grooves on the tension face. As with externally-bonded
laminates, near-surface mounted reinforcement also fails by
delamination near cracks or at the cut-off points; however,
this type of failure occurs at much higher levels of effective
strain, thereby enabling a more efficient use of the material.
For detailing applications, the required additional reinforce-
ment area to achieve flexural strength recovery is estimated
from the moment reduction owing to primary-reinforcement
section loss at the critical section (eq. (4.1)). The allowable
strain of externally mounted FRP reinforcement for design is
limited to eff = 0.004 so as to eliminate the risk of debonding
prior to development of flexural strength [18]. This places a
lower limit for the normalized depth of compression zone
when the concrete reaches the crushing strain 0.004. Also, an
upper limit for is the value associated with balanced failure
(bal 0.64) in order to ensure bar yielding prior to concrete
crushing failure. These two limits represent an allowable
range for after strengthening, based on which the required
additional reinforcement may be estimated.
. . . . , , . 1-2 2007 Tech. Chron. Sci. J. TCG, I, No 1-2 35

Flexural resistance may also be influenced indirectly by
placement of FRP jackets, through the strength increase of
concrete in the compression zone owing to confinement.
Flexural strength also benefits by containment of the cracked
corrosion-contaminated cover that would otherwise have
spalled-off at ultimate. In calculating the enhanced flexural
strength of the encased member from first principles, the en-
hanced axial strength and deformation capacity of concrete
in the compression zone of the cross section is calculated
from equation (4.2).
3.2 Recovery of shear strength
In redesigning FRP-jacketed r.c. members with corroded
stirrups the objective is to recover the initial shear strength
and to secure sufficient displacement ductility that would ex-
ceed the design demands. Considering degradation of shear
strength with ductility demand, the repaired shear strength is
obtained by equation (4.4), where q is the behavior index of
the redesigned structure (assuming the equal displacement rule
this is taken to be equal to the displacement ductility demand
imposed by the design earthquake on the structure). The target
values of q and used in the redesign of the member should
not exceed the value of 3.5 that is currently recommended for
new designs [20]. Equation (4.4) can be used for determining
the required number of jacket layers to be used as shear rein-
forcement in the upgrading scheme. The proposed equation
is a lower bound expression where shear resistance is taken
to degrade with ductility demand, even in the presence of
jacketing, because of the ongoing degradation of contributing
mechanisms (bond, dowel action and aggregate interlock).
a jacket contribution to the confining sources. The average
bond strength after jacketing over the development length of
a bar or lap splice is calculated through equation (4.5).
4. CONCLUSIONS
Methods for assessment of the residual strength of cor-
roded r.c. members have been developed from first prin-
ciples, along with detailed guidelines for seismic upgrading
of such members with FRP jacketing. Corrosion is quanti-
fied by the depth of penetration (bar diameter loss) and by
consideration of ductility reduction due to embrittlement of
steel. Actions considered include flexural and shear strength,
development capacity of anchored corroded reinforcement,
residual stiffness and rotation capacity. The paper presents
detailing methods for upgrading all these indices of resis-
tance with externally bonded FRP layers and jackets. The
methods developed produce results that are consistent with
the limited experimental evidence that is currently available
in the literature on corroded r.c. members upgraded with
FRPs for seismic resistance. To be used in practice, further
calibration of the proposed procedures with experimental
evidence would be required.

3.3 Enhancement of anchorage / lap-splice strength

Equation (3.6) for assessment of the residual bond

strength recognizes two discrete mechanisms through which
corrosion causes bond degradation, namely loss of frictional
resistance and loss of confining pressure by the cover due
to cracking. Because of the inevitable flattening of the ribs,
the coefficient of friction cannot be recovered: however, re-
moval of cracked cover and replacement with healthy grout
in combination with FRP jacket has proven very effective
in rehabilitating the development capacity of corroded bar
anchorages. To be effective, FRP sheets are bonded transver-
sally to the bars axis so as to arrest propagation and open-
ing of splitting cracks. To quantify this function of the FRP
jacket the frictional concept for bond is extended to include
S. P. Tastani,
Ph., D, M.Sc. Civil Engineer, Demokritus University of Thrace, 67 100 Xanthi, stastani@civil.duth.gr
S. J. Pantazopoulou,
Professor, Engineering Department, Demokritus University of Thrace, 67 100 Xanthi, pantaz@civil.duth.gr