Focus:

Chemicaland ElectrochemicalAssessmentof Tannins

and AqueousPrimers ContaniningTannins
V FVetere and R Romagnoli

CIDEPINT - Research and D e v e l o p m e n t Center for Paint Technology.

Calle 52 e / 1 2 1 y 122. (1900) La Plata, Argentina

densed tannin is shown in Figure la, Crosslinking of the

Introduction flavanoids units takes place at the carbon atoms in the 4
The perfbrrnance of an organic coating d e p e n d s mainly on and 6 positions or, if a hydroxyl group is present on the 5
the condition of the metal surfhce as well as on the aggres position, crosslinking occurs at the carbon atoms in the 4
siveness of the environment, The application of a paint to a and 8 positions, giving rise in each case to a very complex
contaminated or rusted surfhce results in early failure of polirneric structure with a wide molecular weight distribu
the coating, In some cases, a thorough cleaning of the sur- tien, Both categories are unrelated appart from the fact that
face to eliminate rust is impossible and chemical treat they contain polyhydric phenols within their structure,
ments are an important alternative, There are more than
100 commercial products for application on poorly pre- Tannins react with the oxide layer giving a black, non-crys-
pared steel substrates, They work by difIerent mocha talline and highly reticulated film of ferric tannate, 10 pm in
nisms, impregnation, stabilization, conversion or isolation thickness, containing free tanninl the formation of mag-
of the oxide layer, ~Recently, derusting of the metal surfhces netite could not be demonstrated with certainty 4 s The fbr
by fungii has b e e n reported, 2 Also, the sealing of metal- ric tannate film has protective properties, 7The iron content
oxide structures with plastics cured in situ has b e e n pro of the film is about 2,5 %,8
posed, ~
The o-dihydroxyphenolic groups of the cathecol ring are
Pretreatment solutions, wash prkners and andcorrosive capable of chelating fbrric ions, giving, under acidic con
paints containing tannins are employed in order to convert ditions, a bis chelate was represented in Figure lb for a
surfhce oxide films, Tannins are polyhydric phenols condensed tannin, The tris chelate is unlikely to fbrrn
obtained from the extracts of certain plants, a distinction is under acidic conditions, ~,~
usually being made between hydrolysable tannins which
undergo hydrolysis in the presence of acids or enzymes Aqueous acid suspensions of tannins may be used directly
and condensed tannins which instead will tend to polymer- on rusted surface after wire brushing; the remaining mate-
ize, The fbrrner, in which ester and depside links predorni rial is generally removed by washing, 9 ~0 These fbrrnula
hate, belong to the group of gallic acid', the latter belong to tions have, as a general rule, bad wettability and surfac-
the group of flavanoids, A typical flavanoid unit of a con tants must be used together with low evaporation rate sol
vents to achieve good brushability, The phosphoric acid
Figure la: A typi~l flavanoid unit ~,~ and tannin concentrations may reach 55 % and 10 %
respectively ~~ Several researchers have suggested that
OH phosphoric acid would lead to the formation of a passive
layer of fbrric phosphate, but it was demonstrated that fbr
HO ric phosphate is not formed at acid concentrations below
OH 8,5M} 1

Wash primers are formulated using a resin to fix the non-

HsC adherent products resulting fkorn the conversion of the
oxide layer and to help in the subsequent painting of the
metal surface, The penetration rate of these fbrmulations
d e p e n d s on the solvent, the tannin concentration and the
Figure lb: A ferric tannate complex~,~
structure of the oxide layer', thus, a 10% butanol solution
proved to be the best solvent and 10% of tannin is recom
mended, The higher the tannin concentration the slower
HzO the penetration rate, The optimum phosphoric acid con
centration is 10%, Free resin formulations penetrate deep-
er inside the oxide layer, 9,~2,~The time fbr tannins to react
with iron oxides is 3 hours as rninknurn, however, 12 24
hours should be allowed before painting, Tannins yield
20~30 rnZ,1-~ and are compatible with red lead, zinc chro
(OH) H20 (OH) mate and zinc phosphate primers, Tannins may be
ernployed with: alkyd, chlorinated rubber, epox~ phenolic,
polyurethane and polyester resins, ~4

Surface Coatings International 1998 (8) 385

 Focus:

These pretreatments are useful in the following conditions: 10 ml of a 10 g,Pl ferric nitrate solution and 10 ml of the 1 M
a) for poor substrate preparation, b) when it is necessary potassium nitrate solution were placed in a 250 ml beaker,
to wait some days after cleaning before painting the sur- Afterwards, 100 ml of the tannin solution was added and
face, c) for reduced anticorrosive primer thickness, and d) the pH adjusted to 2, 4, 6 or S, After 24 hours the pH was
when it is necessary to paint in wet conditions, In the case controlled and adjusted to the original figure; 24 hours
of an oxide layer contaminated with chloride, sulphate, etc, later the pH was measured again and the suspensions fil-
the employment of rust converters combined with an anti tered, The amount of both tannin and iron precipitates were
corrosive primer is essential, &~4'~s determined by drying at 100~ and calcination at 1000~
respectively
DesLauriers 9clakns that the performance of these systems
d e p e n d s both on the amount of oxide present on the sur- The working temperature was 20 + l~
face and the barrier properties of the complete painting
system, Commercial formulations, tested by other The precipitated of tannin and ferric ions is a ferric tannate
researchers, fhiled in all laboratory tests, ~s whose IR spectrum coincides with the spectrum of tannin
alone', magnetite was not detected in this case, ~~
Rust converters b a s e d on a resin and phosphoric acid,
without tannins, which can act on cleaned and rusted sub The results plotted in Figure 2 show that the maximum
strates have b e e n developed, In some cases, they showed reaction capacity of tannins with soluble fbrric ions
a perfbrmance similar to that of zinc chromate and red increased with pH, thus the mass of tannin precipitated fbr
lead primersJ 7-2~Substances other than tannins able to sta- a given amount of iron increased as the pH increased,
bilize iron oxide films have been reported, 2~24 Chestnut tannin showed higher reactivity at low pH values;
which was e x c e e d e d by mimosa tannin as the pH
Also tannins may be employed in anficorrosive coats, increased, Finally, at alkaline pH values sulphitated tannin
Commercial rust converters formulated with an alkyd resin e x c e e d e d the reactivity of the other tannins, From the data
were compared with a conventional anticorrosive paint', all in Figure 2, at pH 4, it may be calculated that chestnut
formulations showed a sknilar protective action when precipitated Y00 rng of tannin per c r e o l e of fbrric ion,
applied on clean substrates, As time elapsed, oxides gave while mimosa, 'quebracho' and sulphitated 'quebracho'
rise to the most harmfhl lepidocrosite and the presence of p r e d p a t e d 560, 500 and 195 rag, respectively The higher
magnetite was not detected, Anticorrosive paints pigrnent the amount of tannin precipitated the better the quality of
ed with ferric tannate have been proposed in the literature, the film obtained, s
for poorly prepared substrates, with prornising results, ~ s
As stated in the literature, 3~ ferric tannates are insoluble
The p u r p o s e of this p a p e r is to study the reaction between compounds, which are precipitated from tannins suspen
tannins and cleaned and rusted steel, It attempts to specify
some features of the iron oxide-tannin interaction in order
to improve current rust converters formulations, Four tan Figure2:Massoftanninpredpitotedfrom1.79rnrnolesofsolubleferr[don,asa
nins were chosen for this research: chestnut (Ca~anea ~p,), functionofpH,forthedifferenttonnins
mimosa (Acac~a~'p,), 'quebracho' (Schlnop~us~'p,) and sul
phitated 'quebracho', Chestnut is a hydrolisable tannin
derived from elagic acid and to which the constitution of a 1.8 V
monogalloylelagic acid with a quercetin rest linked to an
hydroxyl group is attributed, Mimosa and 'quebracho' are
condensed tannins, 29 1.6 A
/A
The first stage of this investigation comprised the study of 1.4
the reaction between tannins, the ferric cation and ferric
hydroxide at the following pH values 2,0, 4,0, 6,0 and 8,0:
l.2 /i~o/!
i
the influence of the phosphate anion was also studied, The
nature of the products formed on cleaned and on rusted
steel surfaces by tannins was determined employing opti
cal microscopy and scanning electron microscopy During 1.0
the second stage the corrosion rate of the steel in tannin
suspensions was determined, Finally a primer containing "a_ 0.8
tannin was formulated and its anticorrosive performance
evaluated by corrosion potential measurements,
0.6
Experiments and results
0.4 ~~//V!uebracho
~ ~ :~iilh,i uebracho
'
Reactivityamong tannins and ferric cation as a function
of pH
A solution of each of the four selected tannins, chestnut,
0.2 -/ 9 Chestnut
A Mimosa
mimosa, 'quebracho' and sulphitated 'quebracho' was pre
pared as follows: 10 g of tannin was placed in a 400 ml
beaker and suspended in 400 ml of hot water (90~ then,
0.0 V
2
i
3
I
4 5
I
6
I I
7
I I
8 9 10
I

50 rnl of 1 M potassium nitrate was added to coagulate col pH

loidal materials and the system was left undisturbed for 24
hours, filtered and transfbred to a 500 rnl volumetric flask,

386 SurfaceCoatingsInternational1998 (8)

 Focus:

sions when ferric cation is added, They are non-adherent

compounds and all surfEces treated with aqueous acid
Reactivity among tannins and ferric hydroxide as a
suspensions of tannins must be washed to eliminate loose function of pH
reaction products, On the other hand, primers are formu- 10 rnl of a 10 g,1-1 ferric nitrate solution and 10 ml of the 1 M
lated with resins to fix the loose tannates resulting from the potassium nitrate solution were placed in a 250 ml beaker,
treatment of the steel surface, 9Ferrous tannates are soluble The pH value was adjusted to 2,0, 4,0, 6,0 and 8,0, respec
in water, colourless and easily oxidizable} ~ tivel~ The system was left undisturbed for 24 hours, A sim-
ilar procedure was fbllowed but using a solution of each
tannin in water, After 24 hours the contents of each beaker
Reactivity between tannins and the ferric ion, in the were stirred and the pH adjusted, The pH adjustement was
presence of phosphate ion, as a function of pH repeated 24 hours later and the systems allowed to rernian
The procedure employed in this case was similar to the undisturbed for one more da~ Finally the solids present in
former one with the exception that 2,0 g of phosphoric each beaker were filtered and both the tannin and fbrric
acid was added to the beaker containing the fbrric ion, The contents were determined according to the procedure
precipitate was analysed to determine the amount of tan- described in the first paragraph of this section, The results
nin, iron and phosphate present, The phosphate content are shown in Figure 4,
was determined by a gravirnetric technique, ~2 The results
are plotted in Figure 8, Comparing Figures 2 and 4, it may be seen that the reac
tivity of the tannins with ferric hydroxide is less than that
The phosphate ion and the tannins compete for the ferric observed with the soluble ferric ion; also it d e p e n d s on the
ion; therefbre, the presence of the fbrmer diminishes the tannin tested, Chestnut tannin is the most reactive while
amount of tannin precipitated for a given amount of iron, 'quebracho' and sulphitated 'quebracho' are the least
Chestnut tannin is less affbcted by such competition reactive tannins, Fiorn the data in Figure 4, at pH = 4, it may
because it precipitated 87,8% (at pH = 4) of the amount of be calculated that chestnut precipitated 81,8% of the pre-
tannin mentioned in the p r e c e e d i n g paragraph: while dpitable tannin per rnmole of ferric ion; while mimosa,
mimosa, 'quebracho' and sulphitated 'quebracho' precipi 'quebracho' and sulphitated 'quebracho' precipitated 78,8,
tated 49,5, 7,80 and 5,70% respectivel~ These calculations 44,4 and YT,1%, respectivelK The ability of tannins and
were made from the data in Figure 8, However, the pres related substances of much lower molecular weight to
ence of phosphoric acid is necessary b e c a u s e it acts on chelate rust had been studied previously and was related
clean steel surfaces by contributing to the dissolution of with chemical structure, It increased as molecular weight
iron oxides, generating the ions to fbrm the film of ferric and the number of hydroxyl groups increased, ~ However,
tannate, Likewise, it improves the wetting properties of the in this case it is difficult to make such a comparison
suspensions of tannins, because bso types of tannins (hydrolisable and con-

Figure 3: Mass of tannin precipitatedfrom 1.79 mmoles of suluble ferric[on in the Figure 4: Mass of tannin precipitated fram 1.79 mmoles af ferric hydraxide, as a
3resenceof 20 mmoles of phosphoricacid, as a fundion of pB, for the different lannins function of pH, for the different tannins

1.6 1.2
9 'Quebracho'
9 'Quebracho' V Sulfltated 'Quebracho'
.A
1.4
V Sulfitated 'Quebracho' 9 Chestnut 9 " ~
1.0
..m
1.2 A Mimosa
0.8

i i
1.0

0.8 0.6

0.6
//- 0.4

///
0.4

0.2
0.2

0.0 I I I
vI I I I I I I I
0.0
2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 10
pH pH

Surface Coatings International 1998 (8) 387

 Focus:

densed) are employed and it is considered that the gravi- Figure 5: Phatographaf lhe ferric lannate film formed on a cleanAISI 1010 sleel
metric test is very useflll to assess tannin reactivity, surJaceby mirnor~alannin suspemional its natural pH (40 •
According to curves in Figure 4 all tannins b e h a v e in a sim-
ilar way in the sense that the amount of tannin reacting with
iron increased with the pH up to the range 3.0-S.0; in this
zone the curves present a m a x - - . The initial increase is
attributed to the tannin de-protonization which favours the
formation of ferric ion complex. At higher pH values the
precipitation of the ferric hydroxide c o m p e t e s with the for-
mation of ferric tanate such that the amount of tannin pre-
cipitated is reduced. The pH range 3 . 0 4 . 0 is the o p ~ n u m
for the formation of the ferric tannate film. If the tannins
react with precipitated iron hydroxide to a lesser extent,
with respect to the soluble ferric ion, unreacted tannin m a y
combine with soluble ferric compounds formed lately b y
corrosion,

Nolure of lhe surface of the [Link] formed by

lannim on pi&led AISI 1010 sleel [Link] Figure 6: SEM micrographaf lhe ferric lannate film formed on a ck~anAISI 1010
This study was carried out employing pickled AISI 1010 sleel surlace by mimosalannin suspensionat its natural pH (350x)
steel panels (C: 0.12 %, Si: 0.01%, Mn: 0.35 %, S: 0.02 %,
P: 0.02 %), d e g r e a s e d with calciurn hydroxide and w a s h e d
with distilled water. Panel dimensions were 10.0 x 1.8 x 0.1
cm,

The panels were lzeated with 8 % tannin suspensions at

their natural pH (Table 1) or by changing it to 2.0 and 4.0
with nitric and phosphoric acid, respectively I%ne reaction
was 48 hours. Once the reaction occured, the panels were
viewed by means of an opffcal microscope (40x) and a
scanning e l e c t o n m i c r o s c o p e (SEM, 3S0x in most cases).
Thereafter, loose reaction products were r e m o v e d b y
brushing and the observations repeated.

'Quebracho' tannin showed the b e s t spreading ability on

steel surfaces while chestnut and sulphitated 'quebracho'
tannins the worst, The spreading ability is greatly improved
when phosphoric acid is added, Corrosion of the b a s e Figure 7: SEM micrographaf lhe remainingof lhe ferric annate film formed on a
metal was o b s e r v e d in those areas where the liquid film clean AISI 1010 steel surlace by mirnosalannin suspension,at its natural pH, after
did not cover the metal surface, brushing (3500 •

A microscopic analysis (40><) showed that 'quebracho' tan-

nin suspensions (nattral pH) formed the b e s t film on the
steel surface, mimosa tannin b e h a v i e d similarly; this m a y
b e due to their spreading properties and highest reactivity
for pH values up to neulzality (Figure 2). Underfilm corro-
sion was o b s e r v e d in the case of ches~ut and sulphitated
'quebracho' tannins. For the sake of simplicity, pho-
tographs and micrographs of panels lxeated only with
mimosa tarmin (Figure 8) are included in this paper.

According to SEM analysis, tannate films exhibited a lamel-

lar struch~re, which varied for the differem tannins (Figure
6, 380><). 'Quebracho' tannin p r o d u c e d small plates and a
highly reticulated surface; sulphitated 'quebracho' origi-
nated a film with b i g g e r plates and fewer cracks. Some of
the ferric tannate films remained strongly adherent to the
steel substrate even when the panel was b r u s h e d with a
hard brush. The tarmate film in contact with the metal sur- However, in all cases, underfilm corrosion took place,
face a p p e a r e d to b e m o r e compact and with fewer cracks Chestnut tannin treated panels showed the lowest d e g r e e
(Figure 7, 3500x). The reactivity of tannins with cleaned of corrosion, For this the reason the studies employing tan-
steel surfaces ensures the adherence of the tarmate film to nin suspensions in niiric acid were discontinued. When
the b a s e metal. ~ phosphoric acid was u s e d instead of nilxic acid, it was
o b s e r v e d that the tannate film s e e m e d to have cracks
Suspensions of tannins in nitric acid (pH = 2) also give a (Figure 9, 350><), The p r e s e n c e of phosphoric acid
ferric tanate film with a plate-like structure (Figure 8, 40><). markedly r e d u c e d the b a s e metal corrosion, Both niiric

388 Surf~Ke Coatings International 1998 (8)

 Focus:

Figure 8: Photographof the ferric tannale film tormed on a clean AISI 1010 steel Figure 10: SEM micrographo| th~ untrec~edferric o• structure(350x)
sur[ace by mimosatannin sus~nsion at pH = 2 of nnrk acid (40 x)

Figure 9: SEM rnicrographof tl~ ferric tannc~efilrn lorrmd on a clean AISI 1010 Figure 11: S~ rnicrographof th~ ferric tannate filrn for~d on a rustedAISI 1010
steel surface by mirn~a tannin smpensional pH = 2 of phosphorkacid (350x) steel surlace by mirnosatannin suspensional its nalural pH (350x)

and phosphoric acids favoured b r m a t i o n of the the tarmate with phosphoric acid, Time reaclion was 48 hours, Once
film b e c a u s e the acids allowed the dissolution of the b a s e the reaction h a d taken place, the panels were o b s e r v e d in
metal causing a h o m o g e n e o u s precipitation throughout the s a m e way as pickled ones,
the metal surface and, consequently a tannate layer of
better quality After brushing, the p r e s e n c e of a strongly Tabk~ I: Cor~ion tale of pickledAISI 1010 panelsin aquBoussus~nsJonsof
adherent tannate film was o b s e r v e d on the metal surface, tannim

The Chestnut tannin p r o d u c e d the b e s t films from the sus- Tannin pH Corrosion rate [Link]-]
pensions acidified with phosphoric acid, On reducing the 'Quebracho' 3.83 0.227
natzral pH of tannin suspensions, it m a y b e seen that chest- Sulphitated 'Quebracho' 4.14 0.402
nut tannin reacts with ferric ion to a greater extent (Figures ~imesa 4.40 0.227
3~4), giving rise to a better quality film, Chestnut 3.09 0.505
Blank 7.00 0.140
Slrudure of lhe surface compoundsformed by tannins on Note: pH values measured in a suspensioncontaining 0.5 g of tannin in 50 rnl of distilled
water
rusted AISI 1010 steel panels
Rusted panels were p r e p a r e d from pickled AISI 1010 steel When rusted steel panels were treated with the tannin sus-
panels e x p o s e d for 3 weeks in a c h a m b e r at 40 +I~ wet- pensions at their natzral pH, the oxide structure (Figure 10,
ring them periodically with distilled water. The oxide film 380x) was r e p l a c e d by a plate-like struchse (Figure 11,
obtained b y this technique was 180 [Jm thick FeOOH 380x), The oxide present on the surface s u p p o r t e d the tan-
b e i n g the most important product, which is normally b u n d hate film leading to film h a n g fewer cracks, A significant
in atmospheric exposures2 The rusted panels were amotmt of non-converted rust was found under the tarmate
b r u s h e d with a wire b r u s h to remove weak and non- film formed by the sulphitated 'quebracho' or chestnut tan-
adhered corrosion products. The thickness of the remain- tins; however, these react slcongly with the outer oxides,
ing layer was 30 pro, Mimosa and 'quebracho' tannins gave films of higher
qualitz, this could b e atlcibuted to their better spreading
The panels were 1teated with 8 % tannin suspensions at properties, also, the mimosa variety is highly reactive with
their n a ~ r a l pH (Table 1) or modifymgit to either 2,0 or 4.0 ferric hydroxide (Figure 4),

Surlace Coatings International 1998 (8) 389

 Focus:

Figure 12: SEM micrographof the ferric iannale film termed on a ru~siedAISI 1010 Figure 13: Cor~ion polemial varJalionas a funclien of lin~ for steel coaled wnh an
'~eel surface by mimosalannin s,Kpensionat pH = 2 of phosphoricacid (350x) aclueouspri~r containinglannin. Elecirolyle:sodium ~rchbrc~e 0,SM
''

.. _.~ 4;.;"~ _.;O::b;~'~_ ~ ' ~ ' ; : . k S : , , ~ - , . J [ "~"

, V.,.~,~ 9 A', (~.,%..~;",~?~,:~,..-, f ~ 2 " " , . 9 ,. -550

.~"~:'~ ".4~."~ 9 " ,'

' ~4,~~'.*:. - - 4 ~~,~,'.'.',~,"
I'"~ - - ~ _ . : , l ' ~ - ~,~'.~
j' , -~.
.d[.'~L~i
,
"~- I-~ ~
, *: "
~ , ' ~ ,
. ~i--a~-.~ ~,.' . _~-,~. ~ ~ ~-~- , L ~ . ~, " - 9 ~ " ~ - 9 "
~'.~.,~.,,,.,~'~.~ : ~ , ~ 9 , ,, , - , ~ - < . .
~..[...~.~',%,, . ~ ~.'-,~ k.:,~[,'~ .-~
9~,,'~,.
,: -_-;."
~.. i~_~,~
,~.-4
~. ~- ~;..~- ~-;. , ~.L.'~
.1-,t~
~'~1.~,:, ;r~..~..,. , -~ - ' . "',--~r
- , . ~- , . , -650.
i;r ~ 9 ,.~ ~r~ ~ . . ~ , - ~ - ~ ,...~-;~{.,.~*,~ , - ~ . . _ ~ ,,
,~..p._'~L, e,, ~.-,~,,, R ~ i - . " " L , ~'.~.~'~ ,, ~.;i~.,. ~l~'~. ~ '
~.., ,... '= .~;,~ ~,, :~:-~-~,~-~ ~r-~.~,L,~cr...~,, ;~' %
i."~_ ' ~- ,v--'-%,~ ~ " , . x . ~ , .
9"'~".~.~,..~ ~.[I ,'., ~,~ , . , ~ , ._r.,.....;"/ t~~. ~' J. '. ,, ~ . ~ ~'~
~
,~c
s.,
~ -.'~K'~
, . ',.-
" -7gO
: ; . . ; ~ ..~.~!.~.- ~'.+~,.T ~,'.~ ?',,:+-.' .,,2 ~ ~ . " .

-750
V ~mer ~thouttonnin
When phosphoric acid suspensions were employed,
chestnut gave very g o o d films while 'quebracho' and 9 Primer~th tannin
mimosa tannins p r o d u c e d films of fair quality. This behav- -800
iour could b e explained partially by considering the 1'
curves in Figure 4. Again, the p r e s e n c e of phosphoric acid t days
improved the film quality and r e d u c e d the b a s e metal cor-
rosion (Figure 12,350x).
It may b e seen that both primers protected steel during the
Adherence to the steel ~bslxate and r e p l a c e m e m of the first hours of immersion, the potential of coated steel was
iron oxide struc~re by a plate one, indicate that tannins: a) displaced to higher values with respect to bare steel. After
actually convert the steel surface, b) stabilize surface oxide one day of immersion, p r i m e r without tannin lost its protec-
layers and c) m a y b e employed as pre-treatments on rust- tive action and eleclxode potential derived quickly to bare
e d steel surfaces. steel corrosion potential. However, the primer containing
tannin continued to protect steel from corrosion after a
Eleclrochemical lesls week of immersion and eleclxode potential remained + 100
The corrosion rate of steel was determined by placing 10.0 mV higher than that of the p r i m e r without tannin. It is con-
x 1 .S x 0.1 cm steel panels in a 2 g tannin suspensions dur- eluded that this exlxa protection was due to the p r e s e n c e of
ing 5 days, s ~ r i n g them periodically After 5 days, the pan- tannin, the simple addition of phosphoric acid was not
els were removed and the total amount of dissolved iron enough to stop corrosion as immersion time went on. The
d e t e r m i n e d colorime~ically initial protection achieved with the p r i m e r without tannin
m a y b e alxibuted to phosphoric acid which by itself is not
The results 9wen in Table 1 show that tannins suspensions able to produce an effective phosphatizing of the steel
a r e more corrosive than disWled water (blank), as was stat- surface; other species such as zinc cation must b e incor-
e d in the literature, 5 but the corrosion rate also varied for porated in the phosphate film to improve its anticorrosive
the different tannins, chestnut suspensions b e i n g the most performance. White ferrous (vivianite) and s o m e ferric
corrosive, due probably to the m o r e acidic na~_re of this phosphate m a y a p p e a r on a steel surface ~:eated with
tannin. phosphoric acid; this is noticed b e c a u s e a white and/or a
light brown film develops when the primer is applied. In
Finally, an aqueous p r i m e r containing chestnut tannin was the p r e s e n c e of tannins these phosphates are readily con-
b r m u l a t e d and its anticorrosive p e r b r m a n c e evaluated b y verted into the m o r e stable ferric tannate.
open circuit potencial measurements. Chestnut tannin was
selected b e c a u s e of its high reactivity with ferric hydrox-
ide, its superior behaviour in the p r e s e n c e of phosphoric
Conclusions
acid and the highest corrosion rate of steel in tannin sus- 1 T h e a n a l y t i c a l p r o c e d u r e o u t l i n e d p r e v i o u s l y is r e c o m -
pensions. The composition of the p r i m e r was as follows, m e n d e d for the evaluation of the interaction of tannms with
acrylic resin 20% b y volume, tannin 6% by weight, phos- iron species. Prior to formulation, each tannin must b e eval-
phoric acid 8% by weight and corrosion inhibitor 0.8% b y uated to assess its usefulness in formulating anticorrosive
weight. The inhibitor to inhibit flash rusting was selected primers.
among: butyraldehyde, benzaldehyde, cyclohexanone, as
is s u g g e s t e d in the literature9 The p r i m e r was applied b y 2. The determinatbn of pH and corroslvity of tannin sus-
brushing on a slightly rusted steel panel up to a thickness pensions m a y help in understanding its interaction with
of 10 pm and allowed to dry for 7 days. The m e a s u r e m e n t s steel. The more acidic and the m o r e corrosive the suspen-
were carried out following a procedure d e s c r i b e d previ- sion, the greater the interaction with steel.
ously? ~The variation of the open circuit potential, as a trac-
tion of time, was m e a s u r e d and is presented in Figure 18 3. The optimum pH b r the tannate film lies in the range
together with the curves of the p r i m e r without tanning and 3.0~.0. The values are close to the nahtral pH of the tan-
the b a r e steel elect:ode. nms suspensions.

390 SurFace Coatings International 1998 (8)

 Focus:

4, The p r e s e n c e of phosphoric acid in the formulations 19, Watanabe T, Kanda S and Oono H, JP 82,280,888
passivates the steel surf%~ces and favours the fbrmation of a [87,260,866J (C1. C09D3/58), 1987.
tannate film on clean steel surfaces, Also it assists dissolu-
20. S i e g e E S i e g e FM Patel and S i e g e K, Hung, TeIjes HU
tion of the oxides and further reaction with tannins,
43,I00 (G1 G08L33/08) 288EP1987,
5, Tannins react with clean and rusted steel surfaces creat- 21, NIPPON OILS & FATS GO,, Jap, Pat, Abs 92 (2I) Gp G,
ing a layer with a plate like structure replacing the globu 54, Patent Number 04/] ] 0357.
lar one of iron oxides,
32. PENZOIL PRODUCTS CO. United States Pas163
501550Z: Off, Gaz, (2), 1036, 1991,
6, Aqueous primers containing tannins inhibit steel corro
sion as was revealed by potential corrosion measure- 23, Balmforth B, Proc, 'Corrosion 9I ', Cincinnati, Paper 561,
rnents,
24. Baknforth B, ChernJcaIs for @e Autorno~Jve k~dusiry,
Drake J. G. (Ed.), p. 38 46, Royal Society of Chemistr~
Acknowledgemenls London, 1991,
The authors thank the GIG (Comisi6n de Investigaciones 25, Bruzzoni We, Aznar A and Ifiiguez Rodrfguez A, Rez
Cientfficas de la Provincia de Buenos Aires) and to GON- / b e t Cortes, yPros G (2), 3 10, 1975.
ICET (Consejo Nacional de Investigaciones Cientificas y 26, Morcillo M, Gracia M, Gancedo JR and Feliti S,
T%cnicas) for their sponsorship of this investigation and Memonas deI II Congreso Iberoamencano de
Lic, Mirta Stupak and Tco, Qufmico Pedro Luis Pesi for their CorrosJSn y ProteccJSn, 221 228, Venezuela, 1986,
help witht he photography
27, Seavell AJ, ], Oi7 Color ChemAssoc, 28, (8) 298~306,
1992,
References 28, Draper PA, S O~ Color, Chern, Assoc,, GS, (10),
1, Thomas NL,J, Protect, Coat, &Lm,, (12), 63-71, 1989, 243 250, 1985,
2, Winkel R Galvano~echnJk, 83, (9), 3095 97, 1992, 29, Haslam E, Chemistry of YegetaMe tannins, Academic
Press Inc, Ltd,, London, 1966,
3, White J, CorrosJonRewews, 7, (2 & 3), 235 257, 1987,
30, Ross TK and F~ancis IRA, Corr, Sci, 18,351 361, 1978,
4, Joseph G and Vallejos R, RevIber, Cortes, yPro~,, X I X ,
(6), 379-381, 1988, 31. Noller CR, QuknJca de los Cornpues~os Org~nros, p.
758-759, L6pez Libreros Edkores S, R, L,, Buenos
5 Kuzniecow JI and Iwanow J, [Link] Pried Eoroz,, 33, Aires, 1968,
(s 9), 1990,
82, Bermejo Martinez F and Prieto Bouza A, Aplicaciones
6 Morcille M, Feliti S, Simancas], Bastidas JM, GalvanJ, analfficas del AEDT y similares, p, 28, Imprenta del
Felhi (Jr) S and Almeida EM, Corrosion (NACE), 48, Seminario Conciliar, Santiago de Cornpostela, 1960,
(12),1031 1039,1992,
88, Gust] andWawer I, CorrosJon (NACE), 51, (1), 87 44,
7 Oonz~lez MI and Abreu A, s de ClencJas 1995,
Quimicas, 1 5 , 3 1 5 ~ 1 7 , 1984,
34, Brandau AH, Introduction to coatings technology, p, 40,
8 A.J. Seavell, S O17 Col, Chem Assoc,, 61, (12), 439-462, Federation Series fbr Coatings Technolog5 USA, 1990,
1978.
35, Uhlig HH, The CorrosJon Handbook p, 910 912,John
9 DesLauriers PJ, Mat, P e i , ~-G, (11), 35-40, 1987, Wiley & Sons, Inc,, N,Y, 1948,
10 Alvarez ZE, Gallozo I and Vald~s D, Rex Iber, Corros y 36, del Amo B, Romagnoli R and Vetere VE Corrosion
Prot,, XVIII, (1), 3 5 4 8 , 1987. Rewews, XIV, (1 2), 121 134, 1996,
11 Nigam AM, Tripathi RP and Dhoot K, Corr, Sol,, 30, (8
9), 799~903, 1990,
12 Oonz~lez MI and Oorvo E Rev, Iber Corros yProt,,
XVIII, 39 41, 1987,
13 Kurkurs O, Produlc~ a~rnoufernoJ korrozh zheIeza i
obraska porzhauchme, Ed, Zinatne, Riga, 83-86, 1980, Water-borne Resins
14 Gianelli R, Memorias deI II Congreso Iberoamericano by G A Howarth& G R Hayward
de CorrosJ6n y Proteccl6n, 253 257, Venezuela, 1986. Monograph Number3 in the series publishedby lhe Oil & @our Chemists'
15 Morcillo M, Felhi S and SimancasJ, Farbe + Lack, 98, Association,on behalf of the BritishCoatingsFederationis now available,
(10), 726-728, 1989, price s includingpostageand packing.
16 Emeric DA and Miller GE, Proc, ADVMAT/gl (NACE, This publicationis aimed at providingadditionalreadingmaterial for
Hous~on), 212, San Diego, 1991, studentswishing to take up studieson lhe BCF'sOpenTed1
17 Guruviah S and Sundaram M, Proc, A d z SurfT~-eatment distancelearningprogramme,whilst also being suitable
Metals, 216-20, Bombay 1987, for lhose personsseeking basic backgroundinformation on aspects
of the scienceand tedlnology of surfacecoatings
18 Lin Ch, Lin R Hsiao M, Meldrum DA and Martin FL,
Ind, Eng Chem, Res 31, (1), 424-430, 1992, To orderyour copy make cheque payable to SURFEXLimitedand send to:
OCCA(Book Sales), Priory House,967 Harrow Road,Wembley, HAO2SF, UK.

Surface Coatings International 1998 (8) 391