Ashless Phosphorus-Containing Lubricating Oil Additives 85
was much more effective in a low viscosity white (paraffi nic) oil than in an aromatic base stock.
Aromatics are good AW agents and compete with the TCP for the surface. Under these conditions,
either the iron phosphate reaction products are less stable or perhaps a thinner and less complete
layer is produced and is worn away, leading to an increase in corrosive wear. Surprisingly, the AW
performance of the mixed aliphatic/aromatic base stock was better than either of the components
and was not improved by addition of TCP.
The behavior of TCP in different atmospheres focused on the effect of moisture in a wet-air
atmosphere and also under dry argon, that is, in the absence of oxygen and moisture. No signifi cant
differences were found in the results indicated previously for the different hydrocarbon base stocks.
However, in a further series of tests comparing the behavior under both wet and dry air and wet and
dry argon in an ISO 32 grade white oil, TCP was shown to have a slight AW effect. The exception
was in wet air, when it increased wear but also generally showed higher scuffi ng loads than when
used in dry argon. In a naphthenic oil of similar viscosity, the use of wet air (or wet argon) again
resulted in increased wear and exhibited higher scuffi ng loads. This behavior was also observed
with other phosphates and phosphites. The authors suggested that in dry air the TCP fi lm forms
very rapidly and metallic contact quickly falls. In dry argon the same thing happens, only at a
slower rate. In wet air the fi lm is not as strong, and metallic contract remains high, whereas in the
case of wet argon, it does not form at all. “Thus the formation of a protective fi lm is enhanced by
oxygen but hindered by the presence of moisture.” The observation [78] that air was necessary for
the action of TCP did not consider that moisture was present in the air and could have been respon-
sible for the improvement in wear performance.
The previous theory indicating it was necessary for the TCP to hydrolyze to form acid phos-
phates before it became active was also challenged. Wear tests on standard and very low acid TCP
in dry argon showed no signifi cant difference in activity. It was concluded that TCP was reacting
directly with the surface without fi rst hydrolyzing to acid phosphate and without being preferen-
tially adsorbed at the metal surface.
In 1972, Forbes et al. [85] summarized the current thinking on the action of TCP, which indi-
cated that TCP was an effective AW additive at high concentrations independent of the base oil, but
at low concentrations was adversely affected by the presence of aromatics. The acidic degradation
products have similar properties but show better performance at low concentrations. It was felt that
TCP adsorbed onto the metal surface decomposed to give acid phosphates that reacted with the
surface to give metal organophosphates.
The results of further investigations into the effects of oxygen and temperature on the frictional
performance of TCP on M-50 steel were published in 1983 [86]. The critical temperature at which
friction is reduced as surface temperature rises was measured under different conditions and was
found to be 265°C in dry air (<100 ppm water) when full-fl ow lubrication is used; 225°C under con-
ditions of limited lubrication and 215°C under nitrogen, also with limited lubrication. Analysis of
the surface indicated that TCP had reacted chemically at these temperatures, causing a substantial
increase in the amount of phosphate deposited (phosphide was not observed). Oxygen was said to
be necessary for this reaction, but the suggestion that prior hydrolysis of the phosphate was required
could not be substantiated.
The debate regarding the formation of iron phosphate or phosphide as reaction products in the
wear mechanism rumbled on into the late 1970s and early 1980s. In 1978, Yamamoto and Hirano
[87] carried out scuffi ng tests on several aryl and alkyl phosphates. The aryl phosphates showed bet-
ter scuffi ng resistance, and it was suggested that the alkyl phosphates reacted with the steel surface,
forming a fi lm of iron phosphate under mild lubricating conditions, but that the aryl phosphates
reacted only slightly until conditions became more severe with the formation of iron phosphide.
The implication was that the phosphide (formed as a result of a reaction between the phosphate and
the metal surface) acted as a good EP additive but that the iron phosphate had only AW activity.
Surface roughness measurements showed a polishing action for the aryl phosphates (particularly for
TCP) but not, under these conditions, for the alkyl phosphates.