OVERWRAP COMPOSITE REPAIRS OF OFFSHORE RISERS AT TOPSIDE AND SPLASH - - PDF document

SLIDE 1

18TH INTERNATIONAL CONFERENCE ON COMPOSITE MATERIALS

1 Introduction This paper describes the work carried out to develop, test and field-demonstrate a polymer resin composite overwrap system capable for use in the repair and/or corrosion protection of pipelines and risers in the oil and gas industry. The system is designed to be applicable in dry conditions, e.g. for pipelines and sections of risers at and close to the platform; in wet conditions, e.g. risers at the splash zones; and in fully submerged underwater conditions, e.g. for subsea pipelines. It is also the intention of this research to develop a system that can be used in hot service conditions of up to approximately 100°C. Hence, based on ISO 24817†, to which the resin system is developed and qualified, it is necessary that a minimum glass transition temperature (Tg) of 130°C is achieved. 1.1 Overwrap repair Fibre reinforced polymer (FRP) composites have high specific strength and stiffness. Combined with their corrosion resistance, it makes them an attractive material choice for weight sensitive structures in environment of high atmospheric

• corrosivity. Bonded FRP repairs have been widely

used in retrofitting of civil concrete structures but their application to offshore steel structures is limited due to a lack of long-term performance database for life prediction [1,2]. Offshore platforms are essentially civil structures, albeit

• perating in a harsher environment. The important

physical and in-service properties of thermosetting polymer composites for civil application have been reported by Hollaway [3]. Engineered FRP or composite overwrap repair, which can be applied in-situ without hot work, can

† ISO/TS 24817:2006 Petroleum, petrochemical and natural gas

industries -- Composite repairs for pipework -- Qualification and design, installation, testing and inspection

fully or partially restore the pressure capacity of a corroded or ruptured pipe, with the added corrosion protection benefit [4]. Good adhesion between the composite repair and the corroded substrate is essential for effective load transfer between the pipe and the composite overwrap. It is also important that the adhesion is sufficiently good to prevent the ingress of water into the repair-substrate interface. This requires the cleaning and profiling of the substrate surface prior to bonding of the composite

• wrap. On onshore dry applications, abrasive

blasting is commonly used, but for underwater applications, other methods such as sub-aquabrade has been reported to be much more effective [5]. Sen et al. [6] reported a repair undertaken on partially submerged corrosion damaged concrete piles with both glass and carbon fibre reinforced composites, and hydraulic power tools were shown to be effective for removing marine growth and surface cleaning. Early thermosetting systems can only be applied on dry surface and is adversely affected by moisture and surface contaminants. Two chemical types generally used for pipeline repair are epoxy and polyurethane (PU) resins [7,8]. PUs have the advantage of curing at ambient temperatures via moisture activation, but the downside is that they are hydrolysed relatively easily in water and they have relatively low Tg values. In contrast, epoxy resins are more moisture stable and can be formulated to provide high Tg and environmental stability using aromatic amino compounds [9]. 2 Experimental – From Laboratory to Field In the current research, epoxy-amine systems were formulated and, in the initial stages, screened based on their (i) ability to cure underwater; (ii) pot life; (iii) Tg; and (iv) adhesion to steel. Once optimised, the resin formulation was blended in scaled-up batches of 20kg in a jacketed

OVERWRAP COMPOSITE REPAIRS OF OFFSHORE RISERS AT TOPSIDE AND SPLASH ZONE

A.Y.L. Leong1, K.H. Leong1,*, Y.C. Tan1, P.F.M. Liew2, C.D. Wood, W. Tian3 & K.A. Kozielski3

1 PETRONAS Research, Lot 3288/3289 Off Jalan Ayer Itam, 43000 Kajang, Selangor, MALAYSIA. 2 PETRONAS Carigali Sdn Bhd, Sarawak Operations, Jalan Sekolah, 98100 Lutong, Sarawak, MALAYSIA. 3 CSIRO, Bayview Avenue, Bag 10, Clayton, VIC 3169, AUSTRALIA.

* Corresponding author (kh_leong@petronas.com.my)

Keywords: overwrap repair, wet and dry conditions, offshore, riser, pipeline

SLIDE 2

re in w (p m a c e re h w a p p 2 in fo s c T in a li w re s e N w re v w d fo fa w e p p ru d c s eaction vesse nto a chopp woven E-gl prepreg) tap mechanical, and testing. In add characterisati effectiveness epair system have been pla wet/splash z applications. protection fie paper. 2.1 Resin che The com ntroduction formulations ystems that constraints Therefore, a nvestigated t as a hardene

• ife. Modera

was utilised esins such a

• lid epoxy

ethyl bisphen Nanoclay w whilst a silan esin adhesion A com versions of p wet/submerge development formulated fo fabrics and h whereas the epoxies with primer is preparation o usting whilst defects and d contact betw teel substrate

• el. The resin

ped strand lass fabric. es were used corrosion an dition to ion, this rese

• f the ne

m in the field anned to prog zone and fi Only the dry eld demonstr emistries mposite requ preclude and simp are generally and/or rel a new chem together with r to provide ately reactive with a mix s a high mol (DER 662) nol A (DGEB was added f ne cure prom n strength to mposite imp primer resin ed applicatio and testing.

• r factory pre

has an ambie primers w h a pot life applied im

• f the substra

t at the same displacing wa ween the com e, in this cas n was then p mat (CSM) The p d to produce nd thermal c materials earch also de ewly develo

• d. The field

gress in stag inally to fu y and splash rations are r uirements de the use ple epoxy/al y unsuitable latively po mical appro h an active a e high Tg an e aromatic ha ture of less lecular weigh and a stand BA) liquid re for viscosity moter was u

• wet steel su

regnation r n, one for dr

• n, were sele

. The comp e-impregnati ent pot life o were formula

• f 45 - 90

mmediately ate to protect e time filling ater, thereby mposite ove e the riser. pre-impregna ) backed pl pre-impregna e specimens characterisat testing a emonstrated

• ped overw

demonstratio ges; from dry ully submerg h zone corros reported in t escribed in

• f

uretha liphatic am due to their

• or

durabil

• ach has be

aromatic am nd adequate ardener (E 1 soluble epo ht di-functio dard diglyci esin (DER 33 y modificat used to enhan ubstrate [10]. resin and t ry and one ected for furt posite resin w ion into E-gl

• f about 3 da

ated as 2-p

• minutes. T

after surf t it against fl g up any surf ensuring be erwrap and ated lain ated for tion and the wrap

• ns

y, to ged sion this the ane mine r Tg lity. een mine pot 00)

• xy
• nal

idyl 31). tion nce two for ther was lass ays, part The face ash face tter the 2.2 E- bac ma he (se res no pre sea tem 2.3 Dy we con the ma sta wh we in sub

• p

ma An ISO in in AS the cor 2 Prepreggin The resin glass fabric cking using achine equip ating statio ee Fig. 1). A sin content o minal tape w e-impregnate aled in alum mperatures b

• Fig. 1. Prepr

3 Character Differenti ynamic Mec ere used t nversion and e neat resin aterial. Coupon t andards spec hich cover te ere carried ou air or in bmerged co erational env Hydrostat achined 80% nnex C, and O 24817, An this research this paper. Salt spra STM B117 ( e overwraps rrosion prote ng n was pre-im c with a th g a comm pped with

• ns and a

After an ini

• f 50wt% w

weight of 46 ed 300-mm w minised bag before they ar

regging of cho backed wo

risation and ial Scanning chanical Th to evaluate d to determin n matrix ma testing were cified in the ensile, shear ut on lamina artificial se

• nditions, to

vironments. tic test pi % wall thinnin d a through nnex D, were h, but only se ay tests w (unscribed) t s to rehabil ection coatin mpregnated hin chopped mercial pre- a (modified fabric tak itial process as achieved 60g per linea wide tapes (p gs and kept re delivered

• pped strand m
• ven tape.

testing g Calorimetr hermal Anal e the perc ne the resulti aterial and t e conducted e ISO 24817 and lap she ates that wer eawater, i.e.

• simulate

ipe specim ng defect, pe wall (hole e also prepar elected result ere carried

• evaluate th

litate damag

• ngs. The sp

into a wove d strand m

• impregnatio

d) resin bath ke-up system

• ptimisatio

resulting in al metre. Th prepreg) wer t at sub-zer to site.

mat (CSM)

ry (DSC) an lysis (DMTA centage cur ing Tg of bot the composi based on te

• 7. The test

ear evaluation re either cure . under full its intende mens with er ISO 2481 e) defect, p red and teste ts are reporte

• ut as p

he efficacy o ged and age pecimens wer en mat

• n

h, m n, a he re ro nd A) re th te est ts, n, ed ly ed a 7, er ed ed er

• f

ed re

SLIDE 3

fa d w c a w a la 3 3 b a c d F m re s

F

fabricated in diameter pipe wrapped with curing underw an accelerated was periodica any softening aminates. 3 Test Result 3.1 Glass tra In an u been reported and hence co curing [11]. duration and

• Fig. 2a. The

measured to eaction and hown as Fig

• Fig. 2. The Tg

duration

artificial se e steel section h the 300-mm water, the sp d salt spray ally measure g due to plas ts and Discu ansition temp unfilled therm d to be indep

• uld be used

The Tg for different st degree of cu tal and res this was plot . 2b.

(

• f neat resin (

n and (b) for d

awater and ns grit blaste m wide prepr ecimens wer

• chamber. Pr

ed to determi sticisation of ussion perature mosetting re pendent of cu to determin r different i tages of cure ure was calc sidual exoth tted as a fun

(a) (b) (a) for differen different degre

comprised 3 ed to SA2.5 a reg tape. Up re transferred repreg hardn ine if there w f the compos sin, the Tg ure temperat ne the degree isothermal c e are shown ulated from hermic heat nction of the

nt isothemal c ees of cure.

3-in and pon d to ness was site has ture e of cure n in the

• f

Tg,

cure

we pro the rea res

• b

tem set am we pre 35 Fig 50 3.2 3.2 fab Ta red mo str ch

• r

sur AS tha gre spl in- tha Yi 3.2 pre Ta fur wh He com pip Three di ere investiga

• gresses, Tg

e molecular action rate de sulting in th served for th Although mperature cu t of field d mbient tempe ere used to edicted mini °C in order

• g. 2, this co

°C. 2 Mechanica 2.1 Coupon t Selected bricated in a able 1. In- duction in

• duli remain

rength was un Applying allenge, as m bond to, rfaces [13]. STM D 4541 at of the out- eater than th lash zone co

• water prime

an the under i et al. [15]. 2.2 Pipe spoo Three essurised to able 1). Upo rther pressur hen signs of ence, the tes mposite rep pe to its unda fferent isoth ated; 30°C, 4 increases du network an ecreases rapi he slow dev he 30°C cure h initially ure system, it demonstratio

• erature. Re

develop a c imum cure d to achieve 9

• rresponds to

al properties tests properties o and out of

• water spec

strength at ned unchange naffected at g adhesive most epoxy a water im Adhesion 1) of the in-w

• of-water spe

he minimum

• ating [14].

er adhesion s rwater adhes

• l tests

pipe spoo

• the test p
• n holding

re was appli f the test pip st results con pair is capab amaged state

PA

hermal cure 40°C and 90 ue to the grad d crosslink idly after the velopment o in Fig 2a. formulated t was decide

• ns curing

sults from D ure kinetic duration of 4 90% cure; b

• a Tg of a

s

• f the comp

water are su imen showe higher tem

• ed. Out-of-w

55°C. underwater adhesives wil mmersed or strength (de water specim ecimens but t m 3.45MPa r It is notew strength is 3 sion strength

• ls

were pressure of at 326bar f ied until hal pe yielding w nsistently sh ble of fully e.

APER TITLE

temperature 0°C. As cur dual growth o

• density. Th

e resin vitrifie

• f the Tg, a

as a high ed that for th would be DSC analyse model, whic 4 days [12] by referring t approximatel posite system ummarised i ed a great mperature bu water lap shea r remains ll not wet-ou r wet ste etermined p mens was ha they were sti required for

• rthy that th

times greate h reported b successfull 326bar (se for a minut lted at 380b was observe

• wed that th

restoring th

3 E

es re

• f

he es as h- his at es ch at to ly m in er ut ar a ut, el er alf ill a he er by ly ee te, ar d. he he

SLIDE 4

3 T s u s la

• Long. M

Trans.

• Long. S

Trans. Poisson Lap Sh Transv Tg (°C) Short te surviva

a With PR

3.3 Salt spra The composi alt spray e underneath th ignificant c aminate over

(a) Unexpose

• Fig. 3. A repr

and (ii) afte which reveal steel pipe; th

• Fig. 4. Aver

after Table Modulus (GPa Modulus (GP Strength (MPa Strength (MP n’s Ratio hear (MPa) verse Adhesion ) erm pipe spoo al test

R25 out-of-water

ay tests ite overwrap exposure wi he wrap (see hange in h r the same pe

ed. (b resentative sa er, the compos ls no signs of he grit blasted raged Shore D exposure to c e 1. Mechanic a) Pa) a) Pa) n (MPa)

• l
• primer. b With P

p repair pass ith no corr e Fig. 3), nor hardness of eriod of time

(i) ) After 2000h in salt sp alt spray specim site wrap has b f corrosion atta surface still a D hardness of F continuous sal cal properties o Out Tested @ RT 16.7 16.4 233 214 0.19 14.7a 18.5a 76 Passed to 326bar c

PR30 in-water pr

sed 4500hrs rosion form r was there a the compos e (see Fig. 4)

(ii) hrs exposure pray men: (i) befor been removed ack on the bas appears pristin FRP overwrap lt spray.

• f the compos

t-of-Water T Tested @

16 15

22 21 0. 14

• N

N

• rimer. c Repair w

s of ming any site .

re, d, se ne. p

4 F 4.1 thr dam at to rep tha Th pla reg by are de Th ins con wi Ma wa pro a 2 the the 4.2 pro pro ma

site laminates @ 55°C T

6.7 5.3

29 14 19 4.1

• NA

NA

was still holding p

Field Demon 1 Backgroun The over ree risers tha mage at two the topside ( demonstrate pair conditio Topside l at is dry and he splash zo atform abov gularly splas y ocean wate eas only dur finition of sp his area henc sofar as corr nstant expos ith the warm alaysian wat as in the rang The repa

• duction plat

2-hour boat r e original ex e risers has s 2 Access Scaffolds

• vide a 36
• cess (see F

anually carri

used in the re In Tested @ RT 16.7 15.3 226 189 0.19

• 8.6b

100

• pressure at 380ba

nstrations nd rwrap repair at have suff

• areas of the

(R3) and the e its efficacy ns. loosely refer d beyond the

• ne, on the

ve the high shed, but not er; seawater ring storms/h plash zone re ce represents rosion attack sure to oxyg m sea breeze

• ters. Ambie

ge of 28-32°C air candidat tform situate ride from sh x-factory neo everely deter s were erect 60-degree a

• Fig. 6). All

ied down to

epairs. n-Water Tested @ 16. 16. 20 16 0.1

• NA

NA

ar when steel pip

r system wa fered corrosi e production splash zone for both dry rs to the part e direct reach hand, is the h tide sea t continuous penetrates t high tides. F efer to Semer the harshest k on the rise gen rich seaw which is ch ent temperat C. tes are on ed off the coa

• hore. It is no
• prene rubb

riorated (see ted around ccess for t materials an

• the sea le

@ 55°C .7 .8 5 8 8 A A

pe yielded.

as applied o

• n attack an

platform, vi e (R1 and R2 y and wet liv

• f a platform

h of seawate e area of th level that sly submerge these elevate For a detaile rad et al. [13 t environmen ers due to th water couple haracteristic o ture measure an offshor ast of Borne

• teworthy th

ber coating o e Fig. 5). the risers t the wrappin nd tools wer

• evel. Due t
• n

nd iz. 2), ve m er. he is ed ed ed 3]. nt he ed

• f

ed re

• ,

hat

• n

to ng re to

SLIDE 5

w

• s

d th 4

• T

g 4 w w p a s la d fo fo d w m

• Fig. 5. The sp

here by R1 to their e

weight limita

• n the scaffo

plash zone, daily to prev he tide rises 4.3 Surface p All the

• ther marine

The surface w grit blasting. 4.4 Repair in The cle with the epox with the pre prepared by actual repair,

• ection. It

anding (clos destructive t follow-up p following 12 Visual i discernible e wrap over th marine growt

• Fig. 6.

topside (left) plash zone can 1, have sustain ex-factory rub

ations, only

• lding at any

the planking vent them fr

• vernight.

preparation exposed o growth wer was then clea nstallation eaned surfac xy primer fol

• epreg. A

applying the

• nto a piece

was tied to se to the sp testing cou periodic insp months. inspection of external dam he past 9 mo th was obse

Repair install ), and splash z ndidate risers, ned long-term bber protective

3 personnel

• ne time. N

g is required

• m being w
• ld coating,

re removed aned to a SA e was imme llowed by sp “dummy” t e prepreg, as e of steel slic and left o plash zone r uld be perf pection pla f the repairs h mage or degr

• nths. Non

erved during

lation in progr zone (right), o , as exemplifie m degradation e coatings.

l were allow Note that in to be remov washed away barnacles a by hand too A2.5 finish us ediately coa pirally wrapp test plate w s it was on ced from a p n the platfo repairs) so t formed dur anned for has revealed radation of netheless, so the 7th mo

ress at the

• f the platform

ed

wed the ved y as and

• ls.

sing ated ping was the pipe

• rm

that ring the d no the

• me
• nth

m. Ha (Sh Do (M Tg

* D

ins har tes rem in rem str 9 m pla an wa the are 5 S sys

• ff

we are reg co be dev ver qu

Table 2 “dum ardness hore D)

• lly adhesion

MPa) (˚C)

Dolly adhesive fa

• spection. D

rdness measu st plate whil moved from Table 2. It mained leve rength and T months, thu

• ace. This re

alyses (see F as proven ermomechan e scheduled f Summary an A newly stem has b fshore riser r ell as dry off e shown in F Substanti gard to the p ating system en carried velopmental rification th ualification pr The two

(i) R1 (a) Risers a

• Fig. 7. Pi
• 2. Properties d

mmy” test plat 5 days 2 81.2 3.6 47.4

ailure.

Dolly adhesi urements we le Tg was d the plate. Th t can be see eled at aroun Tg have imp us indicating esult correlat

• Fig. 2a) and t

to be a nical requirem for the comin nd Concludi developed r been succes repair/mainte fshore condi

• Fig. 7 below.

al amount o erformance o m for both dr

• ut prior t

l testing i hrough a co rogram.

• sets of

(ii) R at splash zone ictures showin PA determined fro e placed offsh 2 mths 7 m 77.3 75 10.4 >1 65.8 75

• n strength

ere made on determined fr he results ar en that Shor nd 80 but b proved over g post-curin tes well with the specified adequate to

• ments. Furth

ng months. ing Remark resin compo ssfully dem enance, both

• tions. The r
• f testing, es
• f the materi

ry and wet c

• the field

is currently

• mprehensiv

repairs are

R2 e (b) R ng the repaire APER TITLE

• m the

hore. mths 9 mths .5 80.0 2* >12* .2 76.9

and Shore the “dummy from materia re summarise re D hardne both adhesio the period o ng had take h earlier DS d cure duratio

• meet

th her inspection ks

• site overwra

monstrated fo h under wet a repaired rise specially wit ial as a repai

• nditions, ha
• trials. Th

y undergoin ve third-part e continuall

(iii) R3 Riser at topsid d risers. 5 E s

D y” als ed ss

• n
• f

en SC

• n

he ns ap for as rs th ir/ as he ng ty ly

de

SLIDE 6

monitored for their health to ensure their durability under

• ffshore

conditions. Post-installation inspections carried out after 2, 7 and 9 months have revealed no discernible sign of any damage or degradation of the composite overwrap. This was confirmed through visual inspection and destructive tests that were carried out on the “dummy” test plate set adjacent to the riser repairs. Results showed that transverse adhesion and Tg have improved over the first seven months while hardness remained constant. Acknowledgements Thanks are due to personnel of PETRONAS Carigali, PETRONAS Research, CSIRO and PETRONAS GTS for technical support. In particular, the assistance with chemical characterisation and salt spray testing from S. Gao of CSIRO and R.M. Johar of PETRONAS Research, respectively, deserves special thanks. Technical discussions with Dr J.H. Hodgkin of CSIRO and Drs P.J. Falzon,

• L. Djukic

and X.L. Liu

• f

ACS Australia Pty Ltd have been immensely useful and are gratefully acknowledged. References [1] O.O. Ochoa and M.M. Salama. “Offshore composites: Transition barriers to an enabling technology”. Composites Science and Technology, Vol. 65, pp. 2588-2596, 2005. [2] C. Alexander and O.O. Ochoa. ”Extending

• nshore pipeline repair to offshore steel risers

with carbon-fiber reinforced composites”. Composites Structures, Vol. 92, pp. 499-507, 2010. [3] L.C. Hollaway. “A review of the present and future utilisation of FRP composites in the civil infrastructure with reference to their important in-service properties”. Construction and Building Materials, Vol. 24, pp. 2419-2445, 2010. [4] A.R. Mableson, K.R. Dunn, N. Dodds and A.G. Gibson. “Refurbishment of steel tubular pipes using composite materials”. Plastics, rubber and composites, Vol. 29, pp. 558-565, 2000. [5] R. Scott and D. Elder. “Surface preparation for pipeline repairs”. TR11003, CRC-ACS, 2011. [6] R. Sen and G. Mullins. “Application of FRP composites for underwater piles repair”. Composites Part B, Vol. 38, pp. 751-758, 2007. [7] G. Marsh. “Marine composites - drawbacks and successes”. Reinforced Plastics, Vol. 54,

• pp. 18-22, 2010.

[8] J. Verhoff,

• K. Ramani,
• N. Blank

and

• S. Rosenberg. “Moisture durability of four

moisture cure urethane adhesives”. Journal of Adhesion Science and Technology, Vol. 16,

• No. 4, pp. 373-393, 2002.

[9] B. Dao, J. Hodgkin, J. Kristina, J. Mardel and

• W. Tian. “Accelerated aging versus realistic

aging in aerospace composite materials. II. Chemistry of thermal aging in a structural composite”. Journal of Applied Polymer Science, Vol. 102, pp. 3221-3232, 2006. [10] S.B. Kim, H.Y. Yi, H.D. Phan, J.W. Nam and J.J. Kim. “Development of aqua epoxy for repair and strengthening of RC structural members in underwater”. Construction and Building Materials, Vol. 23, pp. 3079-3086, 2009. [11] J.J. Imaz,

• A. Valea,
• M. Cortazar

and

• I. Mondragon.

“Correlations between rheological and thermal behaviour

• f

TGDDM/m-PDA epoxy systems”. European Polymer Journal, Vol. 30, No. 5. pp. 561-565, 1994. [12] X.L. Liu, A.Y.L. Leong, K.H. Leong and P.J. Falzon. Unpublished work. July 2010. [13] V.A.W. Semerad, F.A. Corsiglia, D. Weaver and G. Cox. “Testing of epoxy adhesives for a splash-zone coating retrofit system for marine pipeline riser application”. Proceedings of Corrosion 2003. San Diego, Paper No. 03042, 2003. [14] PTS 30.48.00.31-P. “Protective coatings and linings”, PETRONAS Technical Standards, 2008. [15] N.H. Yi, J.W. Nam, S.B. Kim, I. S. Kim, and J.J. Kim. “Evaluation of material and structural performances of developed Aqua-Advanced- FRP for retrofitting of underwater concrete structural members”, Construction and Building Materials, Vol 24, pp. 566-576, 2010.