Copper and Steel in Turbine Oil
Why RPVOT Uses Copper, How Metals Affect Oxidation, and How to Set Meaningful Oil-Analysis Alarms
By Khash
MLE, CLS, MLA III, MLT II, VIM, VPR
Copper and steel deserve attention for two connected reasons: they can provide evidence about the machine, and some of the metal species released into oil can participate in its oxidation chemistry.
A copper result does not automatically identify a damaged bearing. An iron result does not automatically identify a damaged steel cage. The presence of copper in an oxidation test also does not prove that a brass cage will shorten lubricant life more than a steel cage in every application.
The useful approach connects the laboratory result to the installed metallurgy, the form of the metal, the oil condition and the machine’s operating history.
The central answers are these: RPVOT uses a prescribed copper catalyst to impose a reproducible accelerated oxidation challenge; iron can also participate in oxidation chemistry; and meaningful wear-metal alarms must be specific to the element, equipment and measurement method. store.astm.org
This article concerns industrial turbine lubrication and relevant circulating-oil systems. Its examples should not be transferred automatically to aviation turbine oils, separate hydraulic control fluids or unrelated auxiliary sumps.
1. First clarify what the laboratory measures
Steel is an alloy. Routine elemental oil analysis normally reports iron, Fe, together with other measured elements; it does not directly report a quantity called “steel.”
Copper is reported as Cu. That result may represent copper from mechanical wear, corrosion, leaching or a mixture of processes. ExxonMobil’s technical discussion of elevated copper distinguishes particulate wear from dissolved copper associated with oil/material interactions. A copper-containing oil cooler is one possible source, as are oil-wetted copper-alloy components. www.mobil.com
Therefore, Cu and Fe are element identities, not complete diagnoses.
A result of 5 ppm Cu does not establish whether that copper came from a brass cage, a Babbitt lining, a bronze bush or a cooler. It also does not establish its chemical form. Routine ICP elemental analysis alone cannot resolve those questions.
Throughout this article, ppm means parts per million by mass. Confirm the laboratory’s reporting basis before comparing results or limits.
Three pieces of information should accompany the metal number:
- The components and alloys that can contact the sampled oil.
- Whether the evidence indicates dissolved material, fine particles or larger debris.
- The change from the machine’s established condition, including recent maintenance and oil additions.
2. Why does RPVOT use copper rather than steel?
RPVOT means Rotating Pressure Vessel Oxidation Test. ASTM D2272 specifies a copper catalyst coil, water, oxygen pressure and a test temperature of 150°C. Its scope concerns turbine oils of the same composition, including base stock and additives. ASTM explicitly says it should not be used to compare the service lives of new oils with different compositions, and it does not replace ASTM D943. store.astm.org
Copper is chemically capable of promoting reactions involved in oxidation. The prescribed exposure also controls an important experimental variable.
Replacing that coil with steel changes the chemical challenge. Such an experiment could be useful in research, but the result would belong to a modified method. It would not be an unmodified D2272 result.
That is the defensible explanation for using copper in this test. A stronger claim about the original committee’s historical reasoning would require its development records.
The test conditions matter
| Feature | Representative D2272 conditions |
|---|---|
| Oil sample | Approximately 50 g |
| Water inside the glass sample beaker | 5 mL |
| Initial oxygen charge | Approximately 620 kPa gauge, or 90 psig, at room temperature |
| Test temperature | 150°C |
| Rotation | 100 rpm; the arrangement depends on the apparatus |
| Reported endpoint | A pressure decrease of 175 kPa, approximately 25.4 psi, from the maximum recorded pressure |
The quantities and charging conditions are documented in a primary RPVOT study; Koehler documents its apparatus and rotation speed. The pressure-drop endpoint is also described by a lubricant manufacturer reporting its own tests. elibrary.narr.digital
The 175 kPa decrease is measured from the maximum pressure, not from the initial room-temperature charge. This distinction matters because pressure rises during heating.
The test exposes oil to a controlled combination of heat, oxygen, water and copper. As oxygen is consumed, the pressure response provides the basis for the reported test time.
Apparatus details should follow the applicable standard. For example, the published Method B study also used water outside the sample beaker. A brief description saying “5 mL water” is not a complete laboratory procedure. A stainless-steel pressure vessel, where used, has a structural function distinct from the deliberately introduced copper catalyst. elibrary.narr.digital
Iron is also relevant to oxidation
ASTM D943 uses both copper and iron, together with oxygen and water, to evaluate oxidation stability. ASTM also cautions that the relationship between its laboratory result and field behavior varies with the oil and operating conditions. store.astm.org
These methods support two conclusions: copper is a relevant catalytic challenge, and iron cannot be treated as chemically inactive.
They do not establish a universal ranking in which every form of copper is more harmful than every form of iron under every operating condition.
3. How a metal catalyst accelerates oil oxidation
The easiest way to understand this is to follow the reaction chain.
Represent an oil hydrocarbon as RH. The dot in R• indicates a radical: a species with an unpaired electron.
Initiation and propagation
The balance between radical generation, propagation and removal influences the oxidation rate. Metal-assisted production of new radicals can shift that balance and increase the demand on the antioxidant system.
A hydroperoxide is an intermediate oxidation product. It is not itself a radical, and it is not an antioxidant. Its subsequent reactions help determine whether oxidation accelerates.
Depending on the reaction pathways, oxidation can lead to oxygen-containing compounds, acids and larger molecules. Some products remain dissolved; others can contribute to deposits. The extent and consequences depend on the formulation and operating environment.
Copper and iron can help generate more radicals
A simplified formal representation of copper-mediated hydroperoxide reactions is:
The practical message is that metal redox cycling can convert hydroperoxides into reactive species that sustain or accelerate oxidation.
These equations are simplified electron-transfer bookkeeping. They do not mean that turbine oil contains free aqueous ions behaving exactly as these individual equations suggest. Actual reactions involve metal complexes, surfaces, ligands and competing pathways.
Original iron research directly detected alkoxyl radicals during Fe²⁺-induced hydroperoxide decomposition. Modern copper research observed a Cu(I)/Cu(II) sequence involving hydroperoxide-derived radicals and copper–oxygen complexes. These experiments support the chemistry, but neither measures a universal oxidation rate for a turbine reservoir. advanced.onlinelibrary.wiley.com
A catalyst participates in reaction cycles and provides faster pathways for particular reactions. It need not be consumed molecule-for-molecule with the oil. Nevertheless, real metal species can become passivated, bound, precipitated or otherwise less active. “Catalyst” does not mean that the material remains indefinitely unchanged and equally effective.
The antioxidant package interrupts the process
Chain-breaking phenolic and aminic antioxidants intercept propagating radicals. A generic representation is:
Here, AH represents an antioxidant. The resulting antioxidant-derived radical is stabilized sufficiently to avoid continuing the oil-oxidation chain in the same way.
Other protective functions can include decomposing hydroperoxides through nonradical pathways and reducing harmful metal participation through metal deactivation or surface protection. The behavior of the finished lubricant depends on how its components work together.
That interaction explains why a metal concentration alone cannot predict antioxidant depletion or oil life.
Equal metal concentrations do not imply equal catalytic effects
The relevant variables include:
| Variable | Why it changes the interpretation |
|---|---|
| Chemical form | A metal complex, oxide-covered fragment and fresh metallic surface have different reactivity. |
| Accessible surface | Fine debris can expose much more surface per unit mass than large fragments. |
| Temperature | Reaction kinetics and local thermal exposure change with operating conditions. |
| Oxygen availability | Air contact and oil circulation affect the supply of an oxidation reactant. |
| Water and interfaces | Water changes the local chemical environment and can participate in corrosion-related processes. |
| Oil formulation | Base-oil chemistry, antioxidants and metal deactivators influence the response. |
| Existing oxidation products | Hydroperoxides and other products change the chemistry encountered by the metal. |
For a simple geometric illustration, equal masses of spherical particles have a surface-area-to-mass ratio proportional to 1/diameter. Reducing diameter from 10 μm to 1 μm gives ten times the surface area, assuming the same material.
That is a surface-area calculation, not proof of a tenfold oxidation rate. Surface condition and reaction chemistry still matter.
Original model studies have shown that metal-binding agents can suppress copper-driven oxidation while enhancing iron-driven oxidation, depending on the system. Those were aqueous lipid experiments, not turbine-oil formulation trials, but they demonstrate why a universal metal ranking is unreliable. pubmed.ncbi.nlm.nih.gov
In practice, investigate both the source of an increasing metal trend and whether the oil is showing accelerated deterioration. One does not establish the other automatically.
4. What an RPVOT result can—and cannot—tell us
RPVOT is an accelerated oxidation-performance measurement. It is not a direct measurement of antioxidant concentration.
In a published in-situ experiment, rapid oxidation developed while an aminic-antioxidant signal remained. That result concerned a particular formulation, but it directly challenges the assumption that the RPVOT endpoint means every antioxidant molecule has disappeared. elibrary.narr.digital
Consider a hypothetical oil with:
- Fresh-oil RPVOT: 800 minutes.
- Used-oil RPVOT: 240 minutes.
The used-to-fresh test-time ratio is:
This means 30% of the original RPVOT time under that comparison. It does not establish 30% antioxidant concentration, 30% calendar life or permission to operate for 30% of the original service interval.
Linear-sweep voltammetry, commonly associated with RULER testing, addresses a different question by measuring specified antioxidant responses relative to a suitable reference. ASTM D6971 itself explains that it does not capture every active intermediate or establish overall oxidation stability on its own. store.astm.org
Use a suitable fresh-oil reference and the same formulation when interpreting trends. A top-up, formulation change or mixed charge can make a simple before-and-after comparison misleading.
Combine RPVOT with the evidence needed for the decision:
| Measurement | Main contribution |
|---|---|
| RPVOT | Oxidation resistance under the prescribed accelerated challenge |
| Antioxidant analysis | Changes in measurable antioxidant components |
| FTIR oxidation and acid number | Evidence concerning chemical deterioration |
| MPC | A trend in extracted insoluble color bodies associated with deposit concerns |
| Viscosity and water | Changes affecting lubricant condition and operation |
| Elemental and debris analysis | Evidence concerning metals and particulate contamination |
| Machine observations | Bearing temperatures, vibration, oil delivery and operating events |
ASTM D7843 describes MPC as a trending tool within a broader condition-monitoring program. Its result is not a direct inventory of all varnish already deposited inside the turbine. FTIR oxidation similarly benefits from appropriate reference and trend interpretation. store.astm.org
5. Rolling bearings: identify the rings, rolling elements and cage separately
Many large industrial turbines use hydrodynamic journal and thrust bearings. Rolling bearings may instead be present in other turbine designs, geared equipment or auxiliaries. Confirm which components share the sampled oil circuit. Kingsbury identifies steam and gas turbines among hydrodynamic-bearing applications. www.kingsbury.com
Where rolling bearings are installed, the material map commonly looks like this:
| Component | Possible material | Elements relevant to investigation |
|---|---|---|
| Rings, balls and rollers | Common chromium-bearing steel, such as the 100Cr6/52100 family | Mainly Fe, with Cr |
| Pressed steel cage | Typically a different, lower-carbon steel from the raceways | Mainly Fe |
| Machined steel cage | Grade depends on the design | Fe and applicable alloying elements |
| Brass cage | Copper–zinc alloy | Cu and Zn, plus grade-specific additions |
| Tin-bronze component | Copper–tin alloy | Cu and Sn, plus grade-specific additions |
SKF describes the familiar bearing-steel family as approximately 1% carbon and 1.5% chromium. NSK separately identifies low-carbon steels for pressed cages and brass or carbon steel for machined cages. cdn.skfmediahub.skf.com
“Brass” and “bronze” should not be used interchangeably. Bronze also includes alloy families other than tin bronze, such as aluminium bronze. Obtain the actual specification. www.ggbearings.com
Practical example: a brass-caged auxiliary bearing
Suppose an oil-lubricated auxiliary bearing has steel rings and rolling elements with a brass cage.
An increasing Cu/Zn trend makes the cage one candidate source. Increasing Fe makes several steel components candidates. Neither pattern identifies the failed part by itself.
If the same reservoir supplies a gearbox, oil pump and copper-containing cooler, the investigation must include those components. A bearing lubricated by separate grease cannot explain metal in this oil circuit.
Do not expect the oil’s Cr/Fe or Cu/Zn ratio to reproduce the alloy certificate. Different sources, particle retention, dissolution and analytical response prevent that simple interpretation.
Most importantly, a bearing that begins releasing relatively large fatigue fragments may not produce a proportionate rise in routine spectrometric iron. Debris examination and machine condition remain essential.
6. Journal and thrust bearings: use a layer-by-layer metal map
A Babbitt bearing is usually a lined assembly. Its running surface, backing and mating shaft or collar can all have different compositions.
Kingsbury’s PJ journal bearing gives a concrete example: its standard shoes use 4100-class steel with a high-tin ASTM B23 Grade 2 lining. A copper-chromium backing is an available alternative. www.kingsbury.com
Waukesha documents Babbitt over steel or CuCr and also identifies other bearing-material options. Miba describes steel and CuCr carriers, including 16MnCr5 for standard tilt segments. www.waukbearing.com
| Location or construction | Relevant metals |
|---|---|
| Common tin-base Babbitt lining | Sn, with Sb and Cu |
| Lead-base Babbitt lining | Pb, with additions such as Sn and Sb |
| Steel backing | Fe and grade-specific alloying elements |
| Copper-chromium backing | Mainly Cu, with Cr |
| Steel journal or thrust collar | Fe and applicable alloying elements |
| Alternative bronze or aluminium-tin surface | Elements specific to that installed alloy |
Copper can already be present in the lining
AIM’s No. 2 Babbitt has nominal major constituents of:
- 89% tin.
- 7.5% antimony.
- 3.5% copper. www.aimalloys.com
Consequently, copper accompanying Sn and Sb does not prove that wear has penetrated through the lining into a copper backing. Copper already exists in this common lining composition.
Those percentages describe the alloy, not mandatory proportions in an oil sample.
Lead-base Babbitt is a different material. Belmont’s Grade 7 example contains nominally 10% Sn and 15% Sb, with lead as the balance and a small arsenic addition. www.belmontmetals.com
Therefore, a laboratory panel that monitors Pb but omits useful Sn and Sb measurements can be poorly matched to an installed tin-base bearing. Absence of Pb is not evidence that such a bearing is healthy. Use the material certificate, including any permitted residual elements.
Connect the metal to the oil-film event
Hydrodynamic pressure develops as relative motion draws lubricant into a converging clearance. The operating film separates the surfaces. Starting, stopping and insufficient speed can require particular attention; hydrostatic lift, where installed, provides another means of supporting separation. www.waukbearing.com
Waukesha identifies inadequate oil supply, excessive load, insufficient clearance and misalignment among causes of wiping. The lining can rub, smear or melt during such distress. www.waukbearing.com
A rise in Sn and Sb after a difficult start, accompanied by a pad-temperature excursion, therefore deserves a different response from an isolated Cu change after cooler replacement.
Iron together with Babbitt debris also needs careful interpretation. Kingsbury describes how particles embedded in Babbitt can score the steel counterface. Fe can originate from the journal or collar without proving exposure of the steel backing. www.kingsbury.com
7. Understand what the oil-analysis methods can miss
A metal alarm is useful only if the method can detect the material relevant to the failure.
Routine ICP is not a complete debris inventory
ASTM D5185 explains that its oil-soluble calibration does not quantitatively determine insoluble particles, and that particles larger than a few micrometres can produce low results. Routine ICP can therefore underrepresent larger wear debris. store.astm.org
A low result must not be interpreted as proof that the machine has released little metal of every size.
Complementary methods answer different questions:
| Method | Useful contribution | Important limitation |
|---|---|---|
| ICP elemental analysis | Element concentrations represented by the prepared sample | Does not by itself identify source, chemical species or all large particles |
| Ferrous-density/PQ measurement | Additional evidence of ferromagnetic debris | Does not clear a machine of Cu, Sn, Pb or Sb debris |
| Particle count | Number and size distribution of particles | Does not identify elemental composition |
| Filter or membrane examination | Captured particle appearance and size | Represents material retained by that sampling or filtration process |
| Microscopy and suitable particle-composition analysis | Morphology and elemental clues from selected debris | Requires representative particles and specialist interpretation |
Mobil’s turbine-analysis program includes both elemental analysis and PQ testing because they provide different information about wear. www.mobil.com
Low ICP Fe with a rising PQ response is a reason to investigate the particle population. A low PQ response does not exclude nonferrous bearing-lining debris.
Ask the laboratory which elements are actually measured, its limits of quantification and the preparation method. Do not assume Sb is included simply because the report says “wear metals.”
“Not reported,” “below the quantification limit” and “zero” are different statements.
8. Set alarms by metal and by failure mechanism
The available facts do not justify one universal Fe, Cu, Sn or Pb failure limit for every turbine.
ASTM D4378 explicitly treats its guidance as indicative and recognizes the importance of equipment, duty, oil circuit and oil additions. Its scope also distinguishes turbine lubrication from auxiliary-equipment monitoring. store.astm.org
A metal-specific framework should look like this:
| Metal | Sources to check against the installed equipment | Appropriate investigation emphasis |
|---|---|---|
| Fe | Steel bearings, gears, shafts, collars, pumps and corrosion sources | Baseline and rate of increase, ferrous debris, morphology and machine condition |
| Cu | Brass/bronze components, some Babbitt linings, CuCr backing and coolers | Separate wear evidence from leaching/corrosion; use companion elements and maintenance history |
| Sn | Tin-base Babbitt, tin bronze or other documented tin-containing materials | Confirm an emerging signal; correlate with start events, bearing temperatures and recovered debris |
| Sb | Antimony-containing bearing lining | Confirm laboratory capability; investigate with Sn or Pb according to the lining |
| Pb | Lead-base or lead-containing components where installed | Apply the actual metallurgy; an absent Pb signal cannot clear a tin-base lining |
| Cr | Chromium steels, CuCr backing or other documented sources | Interpret with Fe, Cu and particle evidence; do not assume one source |
| Zn | Brass and potentially lubricant additives or cross-contamination | Compare with fresh-oil chemistry before assigning it to wear |
| Al | Aluminium-containing bearings, housings or other oil-wetted components | Check construction and possible contamination sources |
| Ni and other alloying elements | Applicable alloy steels, coatings or special components | Include when the material inventory makes them diagnostic |
The source associations above follow the bearing constructions discussed earlier and the broader component-source guidance in turbine-oil monitoring practice. They are candidate lists; each site must remove sources that cannot contact its oil. latinamerica.chevronlubricants.com
An investigation threshold, an operational action limit and a machinery trip setting serve different purposes.
An investigation threshold identifies a change needing confirmation and diagnosis. An action limit belongs to a documented response plan. A shutdown decision requires the applicable machinery criteria and the developing evidence.
Do not turn a laboratory’s generic red flag into an automatic trip instruction.
9. A practical method for developing those alarms
Establish the reference conditions
Record the actual oil formulation and test a representative fresh sample. Identify the reservoir, sampling point, filtration arrangement and all oil-wetted alloys.
Build a baseline from sufficiently representative healthy operation. Include oil and machine operating hours, duty, starts, maintenance, top-ups, purification and filter changes.
Do not mix samples from a bearing return, a stagnant sump drain and a downstream-of-filter point into one interchangeable trend.
ASTM D7669 supports interpreting lubricant trends alongside equipment condition and applicable OEM guidance. ASTM D7720 addresses statistical alarm development and distinguishes methods appropriate to different data distributions. store.astm.org
Use statistics as an investigation tool
The following is an invented arithmetic example. It is not an ASTM-mandated limit, OEM limit or recommended ppm setting for another machine.
Assume a sufficiently established, approximately normal, stable healthy dataset, with laboratory precision and quantification capability suitable for these low concentrations.
| Metal | Healthy mean | Standard deviation | Illustrative mean + 3 SD investigation screen |
|---|---|---|---|
| Fe | 2.0 ppm | 0.5 ppm | 3.5 ppm |
| Cu | 0.5 ppm | 0.1 ppm | 0.8 ppm |
The calculations are:
These screens ask whether the result is unusual for this particular healthy baseline. They do not establish the concentration at which a bearing becomes damaged or operation becomes unsafe.
If the data are skewed, sparse, changing or dominated by results below the quantification limit, this simple calculation is unsuitable. Select a defensible method with the laboratory and reliability team.
For Sn that has consistently been below the laboratory’s quantification limit, do not enter every result as zero and calculate a zero standard deviation. Confirmed emergence above the method’s quantification capability can become an investigation criterion, with significance determined by the installed materials and corroborating evidence.
Keep the healthy baseline controlled. Continually including an upward fault trend in the baseline can raise the threshold and normalize deterioration.
Include the rate and pattern of change
Suppose Cu increases from 0.5 to 2.0 ppm over 200 operating hours:
That is the concentration-change rate in the sampled oil. It is not the component’s mass wear rate.
A substantial change from an established low baseline can deserve investigation even when the absolute number appears small. Confirm the result, consider analytical uncertainty, and examine whether several samples show sustained acceleration.
Conversely, a single unusual sample should be checked for sampling problems and relevant maintenance events.
Account for the reservoir and oil management
Consider a hypothetical 10,000-litre system with an assumed oil density of 0.86 kg/L. The oil mass is 8,600 kg.
At 1 ppm, the measured concentration corresponds to 8.6 grams of metal in the represented oil phase, assuming uniform mixing and representative measurement.
That calculation does not include all material already trapped in filters, settled in the reservoir or missed by the analytical method. A low concentration in a large reservoir is not a direct measurement of negligible component damage.
Partial oil replacement also changes concentration. Under a simple well-mixed assumption:
If 20% of the charge is replaced with oil containing negligible concentration of the metal, the immediate concentration becomes 80% of the previous value, even if the mechanical problem remains.
Filter changes, purification and changes in particle retention can also alter the trend. Record these events before interpreting a falling number as recovery.
Define the response before the alarm occurs
For each diagnostic metal, the site procedure should specify:
- The applicable OEM or established equipment limit.
- The normal baseline and chosen investigation criterion.
- A meaningful rate or pattern of change.
- The confirmation sample and supplementary tests required.
- The operating evidence that increases urgency.
- Who evaluates the result and authorizes the operational response.
Published limits from another turbine, oil volume or analytical method should be assessed for applicability before adoption.
10. Four practical interpretations
The following are illustrative scenarios, not documented plant case histories.
A. Copper increases after an oil-cooler replacement
Cu rises, while Sn and Sb remain near baseline and the bearing observations are stable.
Review the new cooler’s metallurgy, oil compatibility and the possibility of leaching or corrosion. Examine debris and repeat the sample. The timing strengthens a cooler-related hypothesis; it does not prove it or exclude a bearing source. www.mobil.com
If the copper is predominantly dissolved, ordinary particle filtration may not remove it effectively. Corrective action must address the demonstrated source and the actual oil condition.
B. Tin and antimony increase after an abnormal start
The bearing specification confirms a Sn-Sb-Cu lining. A difficult start was accompanied by a pad-temperature excursion.
Investigate lining distress, oil delivery and lift-system operation where fitted. Examine filters or recovered debris. Give the machine event weight even if the laboratory’s generic total-metal alarm has not activated.
C. Spectrometric iron stays low while ferrous debris increases
PQ or filter examination indicates increasing ferrous material, accompanied by abnormal vibration in an oil-lubricated rolling-bearing or geared component.
Investigate the size and morphology of the debris. The low ICP result cannot cancel independent evidence of larger particles or machine distress.
D. RPVOT decreases while antioxidant and deposit indicators also change
A declining RPVOT trend, decreasing measured antioxidant response and worsening deposit-related evidence justify evaluating the oil’s condition and operating stress.
If Cu or Fe is also increasing, assess it as a possible contributor or parallel symptom. The correlation does not by itself prove which event caused the others.
Check temperature, water, air exposure, contamination, oil mixing and the metal source. Replacing oil without resolving an ongoing source of degradation can leave the underlying problem active.
11. The decision that matters
Copper in RPVOT is an intentional, controlled part of a laboratory challenge. Copper in a turbine-oil report is a finding whose source and significance must be established.
Iron deserves the same disciplined interpretation. Both metals can be associated with wear, and suitable copper or iron species can participate in oxidation chemistry. Their practical effects depend on the system.
A good oil-analysis program therefore connects each result to a material, a mechanism, a measurement capability and an action.
The most useful report goes beyond colouring Cu or Fe red. It helps establish where the metal could have originated, how the pattern is changing, what the oil and machine are doing, and which response the evidence supports.
Khash
MLE, CLS, MLA III, MLT II, VIM, VPR
Discover more from Turbine Oil Reliability
Subscribe to get the latest posts sent to your email.