(at http://www.jcse.org/)

Volume 2 Extended Abstract 6

 Submitted 31st August 1999

The Corrosion Resistance of the Different States of Polyaniline Compared with Strontium Chromate in Powder Coating Epoxy on Mild Steel

*M.M..Attar, J.D. Scantlebury and J. Marsh *Polymer Engineering Department, Amirkabir University of Technology,
P.O.Box 15875-4413, Tehran, Iran

Corrosion and Protection Centre, UMIST, P.O.Box 88, Manchester M60 1QD
England

Keywords: Poly-aniline, Powder epoxy coatings, Strontium chromate, Corrosion prevention, Cathodic disbonding, Accelerated testing

Introduction

 Previous work by the same authors has looked at dispersion of protonated polyaniline, PANI, (42% and 50% protonated[1],[2]) in a solvent borne epoxy system. This produced severe flocculation in the final film and further study on these systems was abandoned. When attempting to use powder techniques it was found that the problem of flocculation disappeared and this paper reports data on the powder coating system. The purpose of this research is to examine the corrosion resistance of the different states of PANI in comparison with strontium chromate in a powder epoxy coating, under three conditions of exposure; namely full immersion in 3%w/w NaCl solution at –1000 mV(SCE), hot salt spray testing (ASTM B117-95) and a wet-dry cyclic test employing artificial Manchester rain water[3].

Preparation of Powder Epoxy Paint

 Different states of PANI(0%, 42% and 50% protonated) were synthesised based on the oxidation reaction using ammonium persulphate[4]. The powder epoxy was produced according to the following process:

All materials including epoxy resin, hardener, flowing agent and pigment (different states of polyaniline, strontium chromate) were premixed in 2%w/w concentration, 5 min 1800 rpm, in order to prepare for the next stage of extrusion which is a homogenous blend of ingredients.

Mixed materials were extruded so that the pigment blends and disperses more efficiently. The extruded polymer was chipped using a Cumberland Granulater and granulated and ground to a fine powder by a mini-kek pin disk mill.

The final epoxy powder paint was sieved obtain small particles less than 75 microns in size.

The powder coatings were applied to a thickness of 22± 2 microns using an electrostatic spray gun. The curing schedule used for all powder coated panels was 15 minutes at 180° C.

Types and formulation of the powder epoxy paint are as follows:

Sample number 1 2 3 4 5
Epikote 3003 855 g 835.9g 835.9g 835.9g 835.9g
Epicote 3003-FC-A-10 100g 100g 100g 100g 100g
Epicure 107FF 45g 44.1g 44.1g 44.1g 44.1g
0% Protonated PANI

20g

42% Protonated PANI     20g    
50% Protonated PANI       20g  
Strontium chromate         20g
Epikote 3003 = Epoxy resin
Epikote 3003 -FC-A-10 = Epoxy resin + 10% flow control agent
Epikote 107FF = Accelerated dicyandiamide curing agent

Cathodic disbondment test

The samples were prepared with 2 weight % concentrations of the pigments. A single holiday of 250 microns was made in the coating of each panel using a laser ( Quanty Ray-Laser, Model GCR-150-20 NBI YAG ). Each sample was coated with a 3:1 mixture of beeswax and colophony resin rounds the edges leaving a central area of 36 cm2 unmasked. The sample, a saturated calomel reference electrode with a luggin agar gel bridge and a platinum counter electrode were dipped in 1 litre 3%w/w NaCl solution. The working electrode, counter electrode and reference electrode were connected to the terminals of a potentiostat, Thomson model Ministat 253 and maintained at -1000 mV (SCE) The disbonded area was calculated by weighing tracing paper of an area equal to the separated layer of coating.

Accelerated testing

Two accelerated atmospheric corrosion tests were performed. Firstly a 1000h 5% NaCl continuous salt spray test according to ASTM: B117-95. Both scribed and unscribed samples were tested. Various degradation features were selected and assigned a number between 1 and 6 based on visual observation where the higher the number indicates the more corrosion. Specific features include; corrosion product at the edge, corrosion products at the scribe, blistering size and extent, black spots not at scribe and edge. Five samples were tested for each situation and the numbers were added. Thus the number for best is 6 and worst is 30. Secondly, a 2000h wet dry cyclic test was performed using Artificial Manchester Rain Water. Because of the benign nature of the environment only scribed samples were tested. Cabinet conditions were 2h at 250C with spray followed by 2h at 350C without spray. Visual assessment was carried out as previously described except no blistering took place and was therefore not considered.

Results and Discussion

 Figures 1, 2 and 3 show the rates of disbonding over 15 days for the five systems. Clearly the greatest disbonding occurs with the 0% protonated PANI, the least disbonding with the strontium chromate. Further, the strontium chromate system is the only system to exhibit a delay time before disbonding. The rest of the samples are not too dissimilar. The lack of alkali resistance of the 0% protonated PANI [5], is the probable explanation of the poor performance of this system. The resistance of the chromate system may be explained by a variety of mechanisms including enhanced adhesion[6], resistance to cathodic reduction[7], improved cross linking of the adjacent coating[8].

Table 1 is the corrosion performance of the scratched samples after a 1000h continuous hot salt spray test

Sample Corrosion at edge Corrosion at scribe Size/extent of blistering Isolated black spots
Blank 22 26 20 14
Strontium chromate 25 25 15 10
0% PANI 20 17 17 10
42% PANI 19 17 30 20
50% PANI 22 22 30 28
Table 2 is for the same exposure conditions as Table 1 except that the panels were unscribed and an assessment of general corrosion is made.
Sample Corrosion at edge Size/extent of blistering General corrosion
Blank 13 9 17
Strontium chromate 25 12 12
0% PANI 10 8 13
42% PANI 30 30 22
50% PANI 22 30 30
  Table 3 is the visual assessment of the various systems after 2000h wet-dry cyclic testing.
Sample Corrosion at edge Corrosion at scribe Isolated black spots
Blank 20 23 22
Strontium chromate 15 27 13
0%PANI 13 18 12
42%PANI 18 18 29
50%PANI 20 22 25
 

Examination of the salt spray data produces the following interesting observations.

To achieve a similar degree of degradation, the test period for the cyclic test had to be twice as long as the hot salt spray test. Even so, the degree of discrimination is somewhat less in the cyclic test. The two protonated PANI samples behave significantly worse than the rest of the samples including the blank in both the cyclic and continuous tests. Behaviour of the exposed steel adjacent to the coating is an important feature of an anticorrosion system, i.e. at an edge or a scribe. With the systems tested, the 0% protonated PANI provides a significant improvement compared with the blank and the two protonated forms and in certain situations is comparable or even better than the conventional strontium chromate pigment. However, the cathodic disbonding experiments provide completely contrary data. The 0% protonated PANI is now the system with the highest rate of cathodic disbonding; i.e. the lowest performing system. It is necessary explain why under free corrosion this system is good and at a relatively negative potential this system is the worst. It is likely that the poor performance is associated with the oxygen reduction reaction which is accelerated at negative potentials. One suggestion is that the 0% protonated PANI is in contact with the electroactive steel surface and it catalyses the peroxy intermediates that are a feature of the oxygen reduction reaction. A second suggestion is that this particular PANI is unstable at the alkali pH’s generated by the cathodic reduction reaction. A third suggestion is that this PANI accelerates the oxygen reduction reaction in general[9]. At this stage it is not possible to be more definitive about which mechanism is most likely.

Conclusions

 Unlike solvent-borne epoxy binders, 0%, 42% and 50% protonated PANI may be incorporated into powder epoxy systems, provided the PANI is added during the extrusion stage of manufacture of the powder coating.

The coating containing 0% protonated PANI showed the good protective properties against a corrosive environment including a conventional hot salt spray test and a more unconventional wet dry cyclic test Performance was assessed visually and seemed as good as a more traditional anti-corrosion pigment containing strontium chromate.

In a cathodic disbonding test, the 0% protonated PANI came out worst. Suggestions are made to explain why this might be so.

References:

 1. M.M. Attar, J.D. Scantlebury, The Journal of Corrosion Science and Engineering.

2. M.M. Attar, J.D. Scantlebury, J.Marsh, Proceeding of the symposium on Advances in Corrosion Protection by Organic Coating I I I , Proceeding, 97-41, 1 (1998).

3. S.B. Lyon, J.B. Johnson, J.D. Scantlebury, Realism in cyclic cabinet corrosion tests - the use of artificial acid rain solution and objective materials assessment", in "Cyclic cabinet corrosion testing", ASTM STP 1238, ed. G.S. Haynes (Proceedings, ASTM Fall Meeting, Fort Worth, TX, USA, 15-16 November 1993), American Society for Testing and Materials, Philadelphia, PA, USA, 3-17 (1995).

4. A.G.MacDiarmid, J.C.Chiang, A.F.Richter, N.L.D.Somasiri, Conducting polymers, Alcacer, L., Ed., Reidel Publications Co: Holland, 105 (1987).

5. Y. Cao, A. Andreatta, A.J. Heeger, P. Smith, Polymer, 30, 2305, (1989)

6. A.T. Evans, J.D.Scantlebury, L.M.Callow, These Cambridge Conferences, (1994).

7. E.L. Koehler, Localized Corrosion, eds Staehle, R. et al. NACE (1974).

8. D.J. Mills, PhD Dissertation, University of Cambridge (1973).

9. J. Marsh, Private Communication (1999).

Send Mail to the

Journal of Corrosion Science & Engineering Home Page
Corrosion Information Server
Centre for Electrochemical Science and Engineering, University of Virginia (JCSE Mirror Site)