In choosing between possible inhibitors, the simplest tests should be done first to screen out unsuitable candidates. The philosophy of initial screening tests should be that poorly performing candidates are not carried forward. An inhibitor that does poorly in early screening tests might actually do well in the actual system, but the user seldom has the resources to test all possible inhibitors. The inhibitor user must employ test procedures that rigorously exclude inferior inhibitors even though some good inhibitors are excluded. (reference)
Inhibitor selection begins with the choice of physical properties. Must the inhibitor be a solid or liquid? Are melting and freezing points of importance? Is degradation with time and temperature critical? Must it be compatible with other system additives? Are specific solubility characteristics required? This list can be extensive, but is important because it defines the domain of possible inhibitors. It must be the first step of the inhibitor evaluation for any new system. These physical measurements are those routinely done as part of minimal quality acceptance testing.
The challenge in inhibitor evaluation is design of experiments that simulate the conditions of the real world system. The variables that must be considered include temperature, pressure, and velocity as well as metal properties and corrosive environment chemistry. System corrosion failures are usually localized and attributed to micro conditions at the failure site. Adequate testing must include the most severe conditions that can occur in the system and not be limited to macro or average conditions. Examples of micro environments are hot spots in heat exchangers and highly turbulent flow at weld beads.
Test specimens should be the same metal as that to be protected; even very small differences in metal chemistry can make major differences in inhibitor performance. Inhibitor performance can vary greatly on different metals and thus inhibitor rankings based on one metal are not universal. Much less obvious are differences between the "same" metal. These nonchemical differences include grain size and orientation, residual stresses, and surface condition. Surface preparation should, to the extent possible, provide a surface comparable to that in the system that is being modeled. Except in special tests, minimal cleaning includes a solvent wash to degrease the sample. More vigorous cleaning such as bead blasting or acid activation can markedly affect inhibitor response even while improving reproducibility of tests. Many experimenters activate test specimens in acid when doing electrochemical measurements. The purpose is to remove any protective or passive oxide layer so that metal solution equilibrium is reached rapidly. Methods for preparing specimens can be found in ASTM G 1, Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens.
Corrosion inhibitors are generally described by terms such as oil soluble, water soluble, oil soluble-water dispersible, etc. Such terms are generalizations, not rigorous descriptions. An oil soluble inhibitor, for example, in reality partitions between the liquid hydrocarbon phase and the water phase as do all other inhibitors; all that can be assumed is that it likely partitions more to the oil phase.
Partitioning of a single compound between two phases is clearly defined. Many commercial corrosion inhibitors, however, are not single compounds but complex mixtures of many compounds, each with its own unique partitioning coefficient. Thus a commercial corrosion inhibitor has no unique partitioning coefficient but rather one for each of the multiple components. Organic inhibitors are generally more soluble in aromatic hydrocarbons than aliphatic ones and more soluble in long chain aliphatics than short chain ones. The result is that partition coefficients must be measured for each "oil" of interest.
Corrosion rates are most commonly reported as penetration rates. The usual way of reporting protection efficiency is in terms of percent protection. Although this reporting method is useful for comparing inhibitor performance, it obscures the actual number of interest the inhibited corrosion rate.
Film persistency tests are more complex than constant concentration experiments. The test metal is exposed to an inhibited test solution for a fixed period of time, then the corrosion rate is determined in a similar solution containing no inhibitor. Test variables include inhibitor concentration in the initial filming solution and the number of rinse solution repetitions. A typical experiment might film for one hour with 1000 ppm inhibitor, rinse one time for an hour, and finally measure the corrosion rate in a third solution. Film, rinse, and corrode solution are the same composition except for inhibitor in the filming step.
Metal loss can be determined gravimetrically, volumetrically, or radiometrically; all are a direct measure of corrosion. Of these, gravimetric or weight loss methods are most used for inhibitor testing. Volumetric methods are associated with inspection or monitoring techniques such as ultrasonic inspection and electric resistance (ER) probe monitoring, although both are sometimes used in long-term inhibitor evaluations. Radiometric methods are used as monitoring methods such as in thin layer activation but could be used for inhibitor evaluation. The corrosion wheel test used to evaluate oilfield inhibitors is an example of weight loss testing.
Coupons from weight loss experiment should be examined visually for localized corrosion seen as pits or edge attack. Analysis can be as simple as "none, some, or lots" or as detailed as counting and depth measurement. ASTM G 46 (Practice for Examination and Evaluation of Pitting Corrosion) provides a complete procedure for evaluating pitting attack.
Electrochemical testing has two major benefits, one major limitation, and one lesser limitation. The benefits are short measurement time and mechanistic information. The severe limitation is the requirement for a conductive corrosive environment. Less troublesome from a testing perspective is the requirement for a corrosion model. Rapidity of measurement makes these techniques useful in characterizing inhibitor performance. Corrosion rates can be determined electrochemically in minutes while weight loss methods can take days. With the near instantaneousness of electrochemical methods, changes of inhibitor performance with time are readily measurable. Questions about inhibitor persistence and incubation time are thus experimentally accessible and experiments concerned with velocity effects become less cumbersome.