By A.W. Momber, Muehlhan AG, Hamburg, Germany; and S. Koller, Germanischer Lloyd AG, Hamburg, Germany

How Surface Preparation MethodsAffect Delamination in Ballast Tanks

between substrate and environment• Solvent retention

Some of these processes, namely cont-amination and adhesion, depend solelyon the method used to prepare the sub-strate before coating application, if allother parameters of the coating systemand environment are kept constant.There are a number of studies that eval-uate the effects of surface preparationmethods—basically dry abrasive blastcleaning (AB) and waterjetting (UHP)—on coating performance. These studiesalso include investigations of the long-term behavior of the coating systems.2,3

These authors, however, were con-cerned with adhesion assessment proce-dures (falling ball impact, pull-off testing,pen knife disbondment) only, and they

J P C L A u g u s t 2 0 0 8 43www.paintsquare.com

elamination is an infa-mous phenomenon fre-quently observed on

coated steel substrates. Delamination atan artificial scribe is usually used as acorrosion-resistance evaluation para-meter in conjunction with a salt spraytest per ISO 7253. Major reasons forthe delamination of organic coatingsinclude the following.1

• Penetration of a liquid between sub-strate and coating, caused by mechani-cal damage• Osmotic processes• Contamination of substrate surfaces• Insufficient adhesion between sub-strate and coating• Too much swelling of the binder• Large temperature differences

D

Continued

did not consider delamination on an arti-ficial scribe after salt spray exposure asa parameter characterizing the corro-sion protection performance of coatings.

The laboratory research described inthe present article focuses on delamina-tion at the scribe for coatings primarilysuitable for ballast tanks.

Objective of the InvestigationThe combination of AB and UHP is aninnovative approach in the surface pro-tection industry. Although thismethod—denoted UHPAB throughoutthis study—is already introduced intoship repair practice, there is limitedinformation available to evaluate quali-ty aspects associated with UHPAB.

All Photos © Muehlhan

Previous studies have shown thatUHPAB removes soluble salts andground abrasive debris from steel sub-strates, and that the method can guar-antee an excellent adhesion betweenorganic coatings and steel sub-strates.4,5 It is the objective of thisstudy to investigate the delaminationof various coating systems applied to substrates preparedwith three surface preparation methods—AB, UHP, andUHPAB. The investigation focuses on ship repair and steelstructure repair applications. The selected coating systemsare mainly suitable for ballast tank applications. However,one additional coating, originally designed for steel-water con-structions, was also included to check the ability of the inves-tigated methods to prepare surfaces of steel-water construc-tion, such as weirs and flood barrages.

The environment for evaluating coating performance wasselected according to the classifications listed in ISO 12944-2. In detail, the following categories were considered.• Atmospheric-corrosivity: C5-M (marine)—This categorycovers coastal and offshore environments with high salinityas well as areas with almost permanentcondensation• Categories for water and soil: Im2—This category covers sea or brackishwater; structures considered includesluice gates, locks, and offshore structures

These two categories determined thetype and intensity of the laboratory teststo be performed according to ISO 12944-6. These tests are listed in Table 1. Theresults of the neutral salt spray tests areof particular concern because they wereevaluated in terms of delamination.

Surface Preparation and Coating ProceduresThe following three methods were used for surface prepara-tion in our study.• AB (dry abrasive blast cleaning)

• UHP (waterjetting)• UHPAB (ultra-high-pressure abrasive blasting; µ®jet)

The performance parameters of these methods are listed inTable 2. The abrasive material applied for dry abrasive clean-ing and UHPAB was a commercial copper slag (NAstra®),according to ISO 11126-3, with the following properties:hardness: > 7 (Mohs); density: 3600 kg/m3; particle sizerange: 0.4-1.2 mm; particle shape: irregular. For the UHP andUHPAB applications, tap water was used with a specific elec-

trical conductivity of 650 µS/cm.Test panel size was 3 m in length and 0.5 m in width.

The area of each panel was separated into three parts:one for AB, one for UHP, and one for UHPAB. All testpanels were originally coated with an organic coatingsystem with an average DFT between 400 and 800µm. After coating removal, the surfaces were allowedto dry. Thereafter, the new coatings were applied tothe entire panel surface. Thus, any individual coatingapplication was performed identically to the threetypes of prepared surfaces.

Five commercial organic coating systems were used forcoating; their major properties are listed in Table 3. The coat-ings were applied with an airless spray device in a sprayingbooth with a controlled climate. The hardening period was 50days. After that period, specimens for the laboratory tests

were manufac-tured. Their sizewas 15 x 10 cm,per the require-ments of ISO12944-6. Thespecimens werecut with plateshears. Edge pro-tection was manu-ally applied toeach sample witha commercial pro-

tective coating system. Debonding was evaluated on the sam-ples after the salt spray test per ISO 7253. The coating wasthen artificially injured with a knife before the salt spray test-ing. The length of the artificial cut was 130 mm, and the cut

J P C L A u g u s t 2 0 0 844 www.paintsquare.com

Research News

Continued

Property Coating System1a 1b 2 3 4 5

Binder EP1) EP1) EP2) EP3) EP4) EP5)

Primer Zn - - - - -Density in kg/l 2.8 1.5 - - 1.3 -Solid content in vol.-% 67 100 96 83 60 80DFT in µm 70 490 346 253 273 374VOC in g/l - - free - 385 170

Table 3: Properties of Applied Coating Systems

1) two-pack; 2) two-pack epoxy-mastic; 3) two-pack modified;4) two-pack, polyamide-hardened; 5) high-build polyamide epoxy

Parameter MethodAB UHP UHPAB

Operating pressure in MPa 0.85 (air) 200 (water) 150 (water)0.8 (air)

Nozzle diameter in mm 10 6 x 0.3 19Water consumption in l/min 0 10 10Abrasive consumption in kg/min 12 0 12

Table 2: Surface Preparation Method Parameters

Corrosivity Durability TestsCategory Range ISO 2812-2 ISO 6270 ISO 7253

(water immersion) (water condensation) (neutral salt spray)C5-M medium - 480 h 720 hIm2 medium 2000 h - 720 h

Table 1: Test Procedures for Paint Systems Applied to Steel (ISO 12944-6)

PPG PMC provides coatings systems that target the growing market and needs of chemical

processors. Protecting critical steel assets, including storage tanks, exterior coatings and

interior tank linings, piping, pipe racks and exposed steel structures, is a primary focus of

ours. Our coatings are uniquely positioned to guard facilities against corrosion and enhance

their appearance while meeting and exceeding customer specifi cations.

Your facility deserves nothing but the best.

For more information, contact a PMC Specialist or visit www.ppgpmc.com.

WE KNOW...

CHEMICALPROCESSING

www.ppgpmc.com

PPG Protective and Marine Coatings

(PMC) protect customer assets in

some of the world’s most demanding

conditions and environments. Our

research and development teams,

in partnership with industry experts,

work tirelessly to develop new and

innovative systems that prevent

corrosion and provide protection

from the elements.

OUR SOLUTIONS

WE KNOW. . . PROTECTIVE COATINGS WE KNOW. . . PROTECTIVE COATINGS WE KNOW. . . PROTECTIVE COATINGS

Click our Reader e-Card at paintsquare.com/ric

P

Bayer did. It’s the Proven Power of Polyurethane CoatingsBuilt on Bayer Technology

Copyright © 2008, Bayer MaterialScience LLC

BBaayyeerr MMaatteerriiaallSScciieennccee.. WWhheerree VViissiioonnWWoorrkkss..

POLYISOCYANATES POLYASPARTICSDISPERSIONSPREPOLYMERS UV SYSTEMS

Personnel at the Atlanta Braves Spring Training Facility atWide World of Sports Disney in Orlando, Florida realizedthat a membrane between the 125,000 square-foot stadium’sstructural concrete slab and the topping slab had failed |inspots. As a result, water had seeped through cracks in theconcrete and started to damage the superstructure. Theyneeded an expert team to develop a winning game plan.

Our vision for a waterproof topcoat built on polyasparticresin technology from Bayer MaterialScience was a homerun. The polyaspartic color topcoat was applied to anew external membrane installed above the concretetopping. Once applied, it provided a high film build todevelop the waterproof coat needed to protect the stadium’s structural concrete from further water damage.It’s exceptionally durable, protecting the external membranefrom wear or damage due to heavy cart and foot trafficfrom thousands of fans.

Initial performance tests conducted during the firstfour-and-a-half years of service have indicated that the polyaspartic color coat will outperform traditionalpolyurethane coats. It will retain its sheen muchlonger and will be easier to clean.

Plus despite a late start, the coating was appliedquickly and dried fast . . . just in time for the year’s400 professional and little league baseball games.

That’s the power of polyurethane coatings built on Bayer technology.

At Bayer MaterialScience, vision works. All it takes isthe know-how to transform that vision into productsyou can use.

Visit www.bayermaterialsciencenafta.com/5.Or call 412-777-3983.

Click our Reader e-Card at paintsquare.com/ric

J P C L A u g u s t 2 0 0 8 47www.paintsquare.com

Research News

of the widths of the delaminated areasat both sides of the artificial scribes.

Results of Delamination TestsThe results of the tests are summarizedin Table 4 and in Figs. 1 to 5. There werenotable differences in the degree of

pentrated through the coating system tothe substrate. Delamination was evalu-ated by removing all loose coating sec-tions very carefully with a small chiseland then assessed by measuring theabsolute width of the disbonded coatingsections along the cut. The delaminationof each sample was given as the average Continued

Clickour

Reader

e-Cardatpaintsquare.com

/ric

Table 4: Results of Delamination Tests

Coating Surface Delamination1 Values in mmSystem Preparation

Average Standard Deviation Variance1 AB 0 - -1 UHP 0 -1 UHPAB 0 - -2 AB 11.2 2.09 4.352 UHP 7.6 0.45 0.202 UHPAB 7.6 0.70 0.493 AB 13.9 0.76 0.623 UHP 9.1 0.54 0.293 UHPAB 6.8 0.21 0.044 AB 12.8 0.85 0.724 UHP 12.5 1.47 2.174 UHPAB 10.7 0.85 0.725 AB 12.2 0.29 0.085 UHP 13.0 1.22 1.505 UHPAB 8.7 0.88 0.78

1 Six measurements

Fig. 1: Effects of coating systems and preparation methods on delamination(Coating System 1 did not show any delamination at all.)

Coating System

Del

amin

atio

nin

mm

J P C L A u g u s t 2 0 0 848 www.paintsquare.com

Clic

kou

rR

eade

re-

Card

atpa

ints

quar

e.co

m/r

icCl

ick

our

Rea

der

e-Ca

rdat

pain

tsqu

are.

com

/ric

J P C L A u g u s t 2 0 0 8 49www.paintsquare.com

Research News

delamination. First, Coating System 1showed excellent performance com-pared to the other coating systems. Thisresult agrees with results of delamina-tion and underrusting tests performedby other authors.6-8 Pietsch andKaiser9,10 found that, at dry blastcleaned substrates (Sa 2½), replacing azinc phosphate primer with a zinc-dust-based primer reduced delamination.However, if the substrates were pre-pared with hand-held tools (St 2, sur-faces partly rusty) or needle guns (sur-faces partly rusty), the primer type didnot notably affect undercreeping. Thus,corrosion and delamination at defects Continued

depended on the condition of the sub-strate before coating and on the pigmen-tation of the coating systems. Coatingsystems with inert pigments tended toexhibit cathodic delamination on dryblast cleaned substrates, and thisprocess occurred at comparatively highrates.9

With respect to the surface prepara-tion method, a notable trend can be seenin Fig. 1. UHPAB always showed thelowest delamination values for any ofthe coating systems being tested.However, the beneficial effect ofUHPAB depended on the coating sys-tem. It was highest for Coating System 3

and lowest for Coating System 4. ABgenerated high delamination values inmost cases. The only exception wasCoating System 5, where UHP had high-er values than AB. These results con-firmed results obtained by Pietsch andKaiser,8 who found that delaminationwas more intense on dry blast cleanedsurfaces than on surfaces prepared withhand-held tools.

The differences in delamination forthe surfaces prepared by various meth-ods depended upon the coating systems.If Coating System 1 was applied, the sur-face preparation methods investigated

Fig. 3: Results of delamination testson Coating System 3

(A=AB; W=UHP; µ=UHPAB)

Fig. 4: Results of delamination testson Coating System 4

(A=AB; W=UHP; µ=UHPAB)

Fig. 2: Results of delamination testson Coating System 2

(A=AB; W=UHP; µ=UHPAB)

J P C L A u g u s t 2 0 0 850 www.paintsquare.com

1685 Baltimore PikeGettysburg, PA 17325sales@automation-usa.comwww.automation-usa.com

HEAVY-DUTY

QNIX®1500COATING THICKNESS GAUGE

1-800-678-4370

MADE IN GERMANY

EASY-TO-USEROBUSTACCURATERELIABLE

3 YEARWARRANTYFREE CERTIFICATE OF CALIBRATIONSEPARATELY INTEGRATED PROBESMEASURING RANGE (200 mil for both FE & NFe)

IDEALFORCORROSIONINSPECTION

Clic

kou

rR

eade

re-

Card

atpa

ints

quar

e.co

m/r

icCl

ick

our

Rea

der

e-Ca

rdat

pain

tsqu

are.

com

/ric

Research News

in this study did not seem to affect thedelamination process. No delaminationoccurred in that case at all. (Therefore,Coating System 1 is not displayed in Fig.1.) The high resistance to delaminationof Coating System 1 could be explainedpartly by the protective action of dis-solved zinc particles that penetrated theartificial scribe. The hydrogen ions gen-erated during zinc dissolution neutral-ized hydroxyl ions from oxygen reduc-tion. Thus, highly alkaline pH-values,responsible for delamination, were pre-vented.8 If, however, Coating System 3was applied, delamination was a verystrong function of the surface treatment

Fig. 5: Results of delamination testson Coating System 5

(A=AB; W=UHP; µ=UHPAB)

J P C L A u g u s t 2 0 0 8 51www.paintsquare.com

24/7/365 ServiceAggreko quickly and cost-effectively fulfills emergency or planned dehumidification needs – 24/7/365. We supply customized rental solutions to control dew points, eliminate dampness and adjust relative humidity levels. We’re your best source for efficient desiccant dehumidifiers and refrigerated dryers.

Prevent moisture and the damage it causes, call Aggreko.

866.684.5577aggreko.com/northamerica

Power/Temperature/Air

Moisture never rests.Neither do we.

Clickour

Reader

e-Cardatpaintsquare.com

/ric

Research News

method. This is illustrated in Fig. 3.Further examples are shown in Figs. 2,4, and 5.

Delamination on dry blast cleanedsubstrates seemed to be less sensitive tochanges in the coating type. The heightsof the columns for AB in Fig. 1 show thelowest fluctuations over the coatingtype range (26%), whereas the valuesfor UHPAB (57%) and especially UHP(71%) exhibited much higher devia-tions. For the latter two methods, acareful adjustment of coating type andsurface preparation method can help tominimize delamination.

It is not clear yet what mechanismsmay be responsible for the effectsobserved in this study. The difference indelamination between UHP andUHPAB can partly be explained by dif-ferences in profile roughness and coat-ing adhesion, which, according to thelist on page 43, affect delamination.

Concerning surface profile and rough-ness, the areas cleaned with UHP exhib-ited a lower profile. The steel plates hadalready been prepared before the appli-cation of the original coating. Becausethis coating was then removed by UHPwater jetting to prepare the surface forapplying the new paint, the original pro-file was deteriorated. The rather insuffi-cient performance of the AB-samplescould be due to the presence of abrasivedebris that were detected and describedin a previous study.5 Thus, fine cleaningafter AB may be an additional parame-ter that influences the delaminationprocess. However, other effects, namelythe ability of UHP and UHPAB toremove dissolved salts, may also con-tribute to the superior performance ofthe coatings applied to the substratesprepared with UHP and UHPAB.4 Theeffects of the coating types on delamina-tion can not easily be explained. All

coatings, except System 1, were basedon epoxy with inert pigments. This par-ticular problem will be addressed in asubsequent study.

SummaryIn our study, delamination of organiccoatings applied to metal substratesdepended upon the surface preparationmethod and coating properties. In mostcases, substrates prepared with ABshowed the highest delamination val-ues, whereas substrates prepared withUHPAB exhibited the lowest delamina-tion values. Zinc dust containingprimers prevented delamination duringthe testing period.

References1. STG-Richtlinie Nr. 2221:

Korrosionsschutz für Schiffe undSeebauwerke, Teil III: Instand-

Continued

J P C L A u g u s t 2 0 0 852 www.paintsquare.com

Clic

kou

rR

eade

re-

Card

atpa

ints

quar

e.co

m/r

ic

Research News

haltung von Korrosionsschutz-systemen, SchiffbautechnischeGesellschaft e.V., Hamburg, 1992.

2. Allen, B, 1997, “Evaluating UHPwaterjetting for ballast tank coatingsystems,” PCE, Vol. 2, No. 10, 38-92.

3. Morris, M., 2000, “Update: evaluat-ing UHP waterjetting as preparationfor ballast tank coating,” PCE, Vol. 5,

No. 9, 54-59.4. Momber, A.W., 2006, “Aspects of

salt concentration of prepared steelsubstrates,” JPCL, Vol. 23, No. 2, M2-M8.

5. Momber, A.W., Koller, J., Dittmers,H.-J., 2004, “Effects of surface prepa-ration methods on adhesion oforganic coatings to steel substrates,”

JPCL, Vol. 21, No. 2, 44-50.6. Baumann, M., 2003, Korrosions- und

Beschichtungsschäden an Stahl-wasserbauten. 2. Tagung Korrosions-schutz in der maritimen Technik,Hamburg, 124-142.

7. Binder, G., 2001, Korrosionsschutzan Kainanlagen, Schifffahrtswegenund Seebauwerken. 1. TagungKorrosionsschutz in der maritimenTechnik, Hamburg, 84-102.

8. Binder, G., 2003, “Examination ofaccelerated laboratory tests for cor-rosion protection,” PCE, Vol. 8, No.11, 8-16.

9. Pietsch, S., Kaiser, W.-D., 1997,Defect-induced undermining of anti-corrosive paints – influence of zincdust and rust. Proc. Eurocorr ’97,Trondheim, Norway, Vol. II, 367-371.

10. Pietsch, S., Kaiser, W.-D., 2001,Unterrostung am Defekt –Korrosion von beschichtetem Stahlan mechanischen Verletzungen.Farbe & Lack, Vol. 107, 141-145

Sven Koller hasworked for German-ische Lloyd AG,Department MCM,Hamburg, since 1996.His areas of expertise

include materials technique, non-fer-rous metals, corrosion, and corrosionprotection. He was given the 2007JPCL Editors’ Award for a paper he co-wrote with A.W. Momber.

Dr. Andreas Momberis head of research &development atMuehlhan SurfaceProtection InternationalGmbH and a lecturer atAachen University, Germany, in theDevelopment of Mining, Metallurgy andEarth Sciences. He was given the 2007JPCL Editors’ Award for a paper he co-wrote with Sven Koller.

Click our Reader e-Card at paintsquare.com/ric