Vincent+J.+Gatto1++Alfonso+Ortiz++Charles+Zha+Jason+Chen++Jeffrey+Wang

Abstract:Oxidation stability of lubricating oil is critical for the smooth operation of industrial equipment such as turbines. Sludge formation is a major issue in turbine operation. Currently, the turbine oil industry employs antioxidants, especially, the combination of aminic and phenolic antioxidants, as the key additives to reduce sludge formation. Aromatic amines, primarily PANA, are considered as the most effective antioxidants in terms of oxidation induction time (OIT) measured by the RPVOT test. However, PANA is also a sludge former. Phenolic antioxidants are normally used in combination with the aminic antioxidants. It is widely believed that there exists a synergistic effect between the phenolic and aminic antioxidants. Regarding the interaction mechanism between the two types of antioxidants, it is generally accepted that phenolics act as a hydrogen donor to help regenerate the consumed aminic antioxidants. This mechanism can only explain some of the phenomena observed in field practice, but there are many other questions remaining unanswered including the antagonistic effect between PANA and phenolics. In this study, we used HPLC, FTIR and LC-MS to monitor the degradation products of the oil systems that were formulated with various aminic and phenolic antioxidants. For the APANA alone system, we identified the formation of dimers and trimers of APANA. For the PANA alone system, we identified the formation of dimers and tetramers. The dimers, trimers, and tetramers of PANA have very limited solubility. These findings therefore help explain why PANA is a sludge former. The oligomerizations (dimerization, trimerization, etc.) regenerate the N-H functionality, thus explaining why the PANA as well as other amines are very effective antioxidants. For the PANA/phenolic system, the oligomers of PANA were greatly reduced. Instead, we identified species that were formed by the recombination of the PANA radicals, or the PANA oligomeric radicals with the phenolic radicals. This, to our knowledge, has not been reported so far. The recombination of aminyl radicals and the phenolic radicals decreases the capability of regenerating the N-H functionality, which helps explain the antagonistic effect observed in the PANA/phenolic system. It also suggests that the phenolic antioxidants are very important in suppressing the sludge formation in the turbine oil system because the recombination products of PANA and phenolics have much better solubility than the PANA oligomers.endprint

Key words：deposit control; turbine oil; phenolic antioxidant; aminic antioxidant; PANA; synergistic effect; antagonistic effect

中图分类号：TE624.82 文献标识码：A

0 Introduction

Sludge formation has emerged as a major issue for the turbine oil industry in the past decade［1-3］. There are a number of factors that contribute to this issue［4］. First of all, the advance of turbine technology results in new turbine designs with smaller lubricating oil reservoirs, hotter operational temperatures and wider operational temperature ranges, and increased bearing loads［5］. These designs increase the power output of the turbines, but also increase the stress to turbine oil. Secondly, increased use of highly refined basestocks such as Group Ⅱ oil has significantly changed the degradation chemistry. Compared to Group I oil, Group Ⅱ and Ⅲ basestocks have better oxidation stability due to the removal of the reactive functionalities and compositions. On the other hand, this removal also reduces the solvency of the basestocks, worsening the solubility of the degradation products, especially the polar products, in the basestocks. Thirdly, the increasing demand on greater oxidation stability of turbine oil has driven the use of PANA, which offers a much longer OIT on the RPVOT test compared to other antioxidants. However, PANA is a sludge former, and is a major contributing factor to varnish formation.

In order to solve this issue, a number of measures have been placed to try to reduce the varnish formation, and meanwhile keep good oxidation stability, among which, the optimization of antioxidant system is at the center of the efforts［6］. In the turbine oil community, it is commonly accepted that aminics, primarily PANA (phenyl-alpha-naphthylamine) are excellent antioxidants in terms of RPVOT oxidation induction time［7］. However, rapid deposit formation has been observed in all aminic systems at a very early stage of testing. From Figure 1, one can see the amine content remaining level for a long period, which helps retain a long OIT, but the deposit formation starts almost instantly after the test starts［8］. On the other hand, phenolic antioxidants are not as effective as the aminics in keeping the oxidation stability but are helpful in suppressing sludge and varnish formation. To achieve the optimal performance, the combination of aminic and phenolic antioxidant has been widely used in turbine oil formulations due to the well-believed synergistic effect between aminic and phenolic antioxidants. Figure 2 shows the typical behavior of the aminic and phenolic antioxidant system. From Figure 2, we can see the phenolic antioxidant is steadily depleted after the test starts, but the amine level stays relatively constant. The amines are significantly more active, and they should be depleted first. The reason why their level remains steady is because the phenol transfers its hydrogen to the newly generated aminyl radical, regenerating the amines. When the phenolic level reaches a critical value, deposits start to form. This demonstrates the importance of the phenolic level in keeping the system from forming deposits.

Vincent J. Gatto,et al.Study on the Degradation and Synergistic/antagonistic Antioxidizing Mechanism of Phenolic/aminic Antioxidants and their Combinations

Another interesting phenomenon is that some antioxidant combinations show synergism behavior, and others show antagonism behavior, as shown in Figure 3. On the left chart, the combination of PANA and DTBP gives a break time that is shorter than the expected additive value of the original two antioxidants, indicating an antagonistic effect. On the right chart, the combination of NDPA and PANA gives a break time that is longer than the expected additive value, indicating a synergistic effect. With all these interesting phenomena, one can ask what is the reaction mechanism that causes such a performance difference.

Figure 3 Synergistic and antagonistic effect of antioxidant

combinations in RPVOT test

To elucidate the interaction mechanism between aminic and phenolic antioxidants, we conducted a series of kinetic studies on the oil samples formulated with different antioxidants and antioxidant combinations. The degradation process was monitored by HPLC. Key degradation products of the antioxidants were separated by using chromatographic methods, and the isolated products were characterized by FTIR, HPLC, NMR and LCMS. These data allowed us to identify the degradation species, and to determine the concentrations of the parent antioxidants, and the relative intensity of the degradation products at different degradation stages, therefore provided clues to the degradation mechanism of the antioxidants in the process of oil oxidation. In this paper, we report the results, and discuss the implications of data to the formulation of turbine oil with improved performance.

1 Experimental Methods

1.1 Antioxidants

The antioxidants chosen for this study are shown in Figure 4. These antioxidants are:

(1) The traditional hindered phenolic antioxidants represented by DTBP and pHPE, are common hindered phenolic antioxidants used in many industrial oils. The high molecular weight hindered phenolic represented by pHPE is used in applications where high temperatures are achieved such as high performance engine oils and specialty industrial fluids. To simplify analytical work, a single isomer of pHPE was used for the study.

(2) Alkylated diphenylamine represented by oDPA (octylated diphenylamine) is a multi-purpose aminic antioxidant used extensively in all types of engine oils and industrial fluids.

(3) Phenyl-α-naphthylamines represented by PANA and alkylated PANA (APANA).

Figure 4 The chemical structure of the antioxidants used in this study

1.2 Formulation

All tests were done on a simple turbine oil formulation comprising a Group Ⅱ basestock and an antioxidant, or combination of antioxidants. There are no rust or corrosion inhibitors included in the formula for the purpose of simplicity. The load of the antioxidant(s) for each formula is listed in Table 1.

1.3 Oxidation Procedure

Oils (300 mL) were aged at 120 ℃ with 10 L/hr oxygen flow without an oxidation catalyst present. Samples were taken at designated times, and subjected to analysis.

1.4 Analysis Procedure

HPLC was used to follow the loss of parent antioxidants and the formation of degradation products. LC-MS was used to identify mass of the parent antioxidant and the degradation products. FTIR was used to identify the tautomers of the degradation products of pHPE.

2 Results and Discussion

2.1 Degradation of DTBP System

Under the experimental conditions, DTBP was quickly depleted. No significant degradation product was detected. Results of the replicate runs were very similar. It was possible that the loss of DTBP was mainly caused by vaporization. The same experiment was repeated by replacing oxygen with nitrogen, and the result was very similar, confirming the quick depletion was mainly the result of evaporation (Figure 5).

2.2 Degradation of pHPE System

For the pHPE system, the depletion of pHPE was immediately accompanied by the appearance of a degradation product. LCMS and FTIR analyses confirmed that the degradation product was the corresponding quinone, as expected (Figure 6). This quinone can exist in the form of two equilibrating tautomers, but FTIR indicated the quinone tautomer was preferred. Interestingly, in the path of degradation, after the area counts of the quinone and the parent antioxidant crossed, the parent antioxidant continued to deplete, but the quinone area started going down slowly, suggesting that there might be other side reaction paths involving other degradation products.

2.3 Degradation of oDPA

Similar to DTBP, oDPA was quickly lost in the oil aging test (Figure 7), but the mechanism may be different. oDPA has a very high boiling point (509 ℃ at 760 mmHg) so its quick loss was unlikely caused by evaporation. Instead, it might have followed the following mechanism(Figure 8). Under the experimental condition, it should follow the cool temperature path to eventually form the benzoquinone, and the nitroso product. The benzoquinone would evaporate at the experiment temperature, and the nitroso might decompose under the experimental condition. Indeed, the nitroso compound is not stable to photolysis of sunlight, or similar light source. Meanwhile, the viscosity of the oil was also observed to increase as the antioxidant was depleted, indicating significant degradation had occurred.

2.4 Degradation of APANA System

From Figure 9, one can see the depletion of APANA started immediately after the oil aging started. Not long after the oil aging started, the dimer emerged, and the concentration increased steadily. Soon after, the trimers started to form, and its concentration increased steadily as the degradation moved forward.

Figure 9 The depletion of APANA

APANA and PANA are very effective aminic antioxidants. They can provide extended Oxidation Induction Time (OIT) on the Rotating Pressure Vessel Oxidation Test (RPVOT). This outstanding performance is originated from the special properties of the aromatic amines. First of all, aromatic amines such as PANA, APANA, and DPA are very reactive toward peroxy radicals due to the weak N-H bond dissociation energy (BDE). Compared to the hindered phenolic antioxidants, aromatic amines are ten times more reactive toward peroxy radicals. The scavenging of peroxy radicals during oxidation is the key step of inhibiting action of antioxidants. The generated amino radical, however is very reactive; it can quickly abstract the hydrogen from the hydro peroxide and give back to N-H bond. So reaction (1) (Figure 10) is an equilibrium reaction. Therefore, the high reactivity of aromatic amine alone cant ensure the good antioxidant performance. The further reaction of the amino radical drives reaction (1) to the right, shifting the equilibrium in the first step reaction. As shown in Figure 10, the amino radicals can couple to each other to form dimers, and regenerate the N-H bond［9-10］. The dimers can capture radicals just like the monomer, although the rate constant may be smaller［9］. The dimer radical can further couple with a monomer radical to form trimer, and regenerate an N-H bond, which can further capture a radical. This mechanism slows down the reverse reaction of (1), and regenerates the N-H bond. Both factors are critical for the superior antioxidant activity of aromatic amines.

However, there is a negative side to this mechanism. The dimers and trimers have lower solubility in hydrocarbon oil, especially GroupⅡ and Ⅲ oil, and are likely to form sludge more easily, which explains why when PANA is used as antioxidant, deposit formation starts very early in oil aging. The alkylation of the phenyl-ring of PANA forms APANA, of which the dimers and trimers have better solubility compared to PANA, thus reducing the deposit forming tendency.

2.5 Degradation of oDPA/APANA System

Figure 11 shows the degradation behavior of the oDPA and APANA system. Synergism between alkylated diphenylamine and APANA/PANA is well known and used in the stabilization of hydrocarbon oils. The results shown are a clear demonstration of this mechanism. In the previous section, we discussed the depletion of oDPA (Figure 7). Without APANA, oDPA disappeared quickly. Here, the concentration of oDPA remained steady, indicating the regeneration of oDPA by abstracting hydrogen from the APANA N-H bond. The mechanism can be depicted as in Figure 12. While oDPA was regenerated, the APANA radical continued its journey to dimerization, and trimerization as shown in Figure 10. As a result, APANA worked more like a hydrogen radical donor, but the formation of the dimers and trimers proceeded as in the APANA alone system. So the synergism between alkylated DPA and PANA, or APANA, may increase the oxidation stability of the system, but it cant reduce the sludge formation, as demonstrated by many examples.

2.6 Degradation of APANA plus DTBP plus pHPE System

Figure 13 shows the degradation behavior of the three components, APANA plus DTBP plus pHPE system. Looking at the degradation curve of DTBP, one finds it is very similar to Figure 1. Even the half-life is similar, suggesting that evaporation remains the major cause for loss of DTBP. Looking at the degradation curves of pHPE, one finds the first part of the curve is similar to Figure 2. After the two curves cross, the degradation product curve continues to climb which is different from the pHPE alone system shown in Figure 2 where the degradation product curve starts to go down after the two curves cross. These differences indicate that maybe different reaction mechanisms were involved in the two systems. The biggest difference is the degradation curve of APANA. In this three components system, the dimer and trimer curves of APANA disappeared. Instead, two new species were formed, indicating different reaction paths occurred. The dimerization and trimerization paths were detoured by the phenols. Its possible that the phenoxy radicals interfered with the formation of the APANA dimer and trimer radicals to form new species. We also noticed that the decrease of APANA concentration between the starting point and the end point is significantly more than the APANA alone system.

2.7 Degradation of APANA plus oDPA plus DTBP plus pHPE System

This is a very complex system (Figure 14). Looking at the degradation curve of DTBP, one notices that the loss of DTBP seems to be faster than the previous two systems. This may be related to the presence of the active oDPA radical, which abstracts hydrogen from the phenols and APANA. So in addition to the evaporation loss, DTBP is also depleted by oDPA, and possibly by APANA as well. As expected, the oDPA concentration remains relatively constant, indicating the regeneration of oDPA by other antioxidants. Interestingly, there were no additional species formed, which suggests at the present stage, and under current experimental conditions, no oDPA radical was combined with other radicals. The degradation of APANA may follow a few paths: (1) donate hydrogen to oDPA and become a radical, (2) scavenge the carbon peroxy radical, and (3) the generated APANA radical combines with the phenoxy radicals. These reactions are depicted as in Figure 15.

Figure 14 The degradation of APANA plus oDPA plus 2,6-DTBP plus pHE system

2.8 LC-MS Analysis of the Degradation Products

To identify the degradation products, Samples 6, 7, 10 and 12 were subjected to LC-MS analysis. The results are listed in Tables 2, 3 and 4. For the PANA alone system (sample 10), the formation of the dimer is expected. This also suggests that the PANA dimer aminyl radical has strong tendency to recombine with each other. As mentioned before, the products of the recombination, the dimers and the tetramers, regain the N-H functionality, thus the peroxy radical scavenging capability, and this is an important contributing factor to its high antioxidant capacity. Meanwhile, this result is critical to the understanding of the sludge formation behavior of PANA. The solubility of the tetramers is likely to be worse than the dimers; the direct formation of tetramers from dimers helps explain why PANA is a sludge former.

The oxygen containing species were formed by the recombination of the dimer, or tetramer radicals, with peroxy radicals, as shown by the following general formula.

Different from the PANA system, the APANA system formed APANA dimers, trimers, and tetramers. Obviously, alkylation alters the chemistry properties, making APANA have less of a tendency to form tetramer than PANA (Table 3).

Figure 15 Proposed reactions involved in the APANA plus oDPA plus pHPE plus DTBP system

Table 3 Species formed in the degradation of

When phenolics are included in the formula, the degradation products are significantly different (Table 4). We still see the parent antioxidants, but the recombined products of APANA, or PANA radical with phenoxy redicals are also present as degradation species. The recombination of aminyl and phenoxy radicals forms new species. The attachment of phenolic groups to PANA and APANA blocks PANA and APANA from further recombination and prevents regeneration of the aminic NH functionality. This also shuts down the radical scavenging capability of PANA or APANA. Thus this recombination mechanism reduces the radical scavenging capability of PANA, or APANA. So in terms of activity, the phenolic/PANA system is an antagonistic system, as shown in the introduction section. However, this result cant be generalized to all aminic/phenolic antioxidant systems. Synergism between antioxidants involves very complex reactions［11-15］. The conditions necessary for antioxidant synergism to occur between aminics and phenolics are well documented.

On the other hand, the new species formed from the PANA aminyl radical and phenoxy radical have much better solubility due to the attachment of the highly hydrophobic phenoxy moiety. This plus the decreased oligomerization of PANA help sludge control.

Table 4 Species formed in the degradation of

sample 11 (PANA plus pHPE plus DTBP) and sample 6

In the turbine oil community, it has long been accepted that the mechanism of aminic/phenolic function involves consumption of the phenolic to regenerate the aminic. To our knowledge it has not been reported that the aminyl radical and the phenoxy radical recombine to form totally new species. This discovery leads us to rethink the concepts that used to formulate turbine antioxidant systems. It also provides us with a new window to view the behavior of various existing aminic/phenolic antioxidant systems.

3 Conclusions

The experimental results clearly demonstrate the formation of dimers, trimers and tetramers of PANA, or APANA in the degradation of the PANA and APANA-containing oil systems. The formation of the oligomers regenerates the N-H functionality, boosting the antioxidant activity, which, in addition to the high activity of N-H, makes PANA and APANA effective antioxidants. However, the low solubility of the oligomers, especially the PANA oligomers also accounts for the high sludging potential.

The degradation behavior of oDPA/APANA system suggests oDPA is a more active peroxy radical scavenger than APANA, and the oDPA can be regenerated by abstracting hydrogen from APANA. The oligomerization of APANA proceeds as in the APANA alone system. This result suggests that combination of oDPA and APANA may improve its antioxidant activity, but cannot reduce the sludge formation of such a system.

The recombination of APANA, or PANA radical with phenoxy radical blocks the recombination of the APANA and PANA radicals, thus killing the oligomerization. So the antioxidant activity of PANA, or APANA may be reduced (antagonism). But the solubility of the adducts is improved versus the PANA oligomers, thus helping suppress the formation of sludge.

References：

［1］ Day L. Solving Varnish Problems in Gas Turbine Lubricants［J］. Tribology and Lubrication Technology, 2008, 64(1):32-38.

［2］ Farooq L. Effective Varnish Removal from Turbine Lubrication Systems［J/OL］. Machinery Lubrication,August 2012［2013-03-04］.http://www.machinerylubrication.com/Read/29033/turbine-varnish-removal.

［3］ Fitch JC.Sludge and Varnish in Turbine Systems［J/OL］.Machinery Lubrication,May 2006［2013-03-04］.http://www.machinerylubrication.com/Read/874/sludge-varnish-turbine.

［4］ Livingston G, Wooten D, Thompson B. Finding the Root Causes of Oil Degradation［J/OL］. Practicing Oil Analysis, January 2007［2013-03-04］.http://www.machinerylubrication.com/Read/989/fluid-degradation-causes.

［5］ Gatto VJ,Moehle WE,Cobb TW,et al.Oxidation Fundamentals and Its Application to Turbine Oil Testing［J/OL］.Journal of ASTM International,2006,3(4)：20 ［2013-04-12］.http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI13498.htm.

［6］ Livingston G, Ameye J, Thompson B. Rethinking Condition Monitoring Strategies for Today's Turbine Oils［J/OL］.Machinery Lubrication, May 2010［2013-04-12］.http://www.machinerylubrication.com/Read/24987/rethinking-condition-monitoring-strategies-turbine-oils.

［7］ Gatto VJ, Liu B, Wang J. Development of a New Sustainable Antioxidant Technology for Low Phosphorus Passenger Car Engine Oils ［C］//Symposium of Dalian Lubricant Technique and Economy Forum, 2010：212-220.

［8］ Livingstone G, Thompson B, Okazaki ME. Physical, Performance and Chemical Changes in Turbine Oils from Oxidation［J/OL］. ASTM Symposium on the Oxidation and Testing of Turbine Oils, 2007, 4 (1)：1-18［2013-04-17］http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI100465.htm.

［9］ Bridger RF. Kinetics of Inhibition of Hydrocarbon Autoxidation by 1,1'-Bis(N-phenyl- 2-naphthylamine)［J］. Journal of Organic Chemistry, 1971, 36(9):1214-1216.

［10］ Zeman A, Romer R, von Roenne V. Fate of Amine Antioxidants during Thermal Oxidative Ageing of Neopentylpolyol Ester Oils: Part 1［J］. Journal of Synthetic Lubricants, 1986, 3(4): 309-325.

［11］ Slobodin YM, Gonor AA. Synergism of Antioxidants in Synthetic Ester Oils［J］. Chemistry and Technology of Fuels and Oils, 1966, 2(3):198-201.

［12］ Mahoney LR, DaRooge MA. Inhibition of Free-Radical Reactions, IV. The Synergistic Effect of 2, 6-Di-t-butylphenols on Hydrocarbon Oxidation Retarded by 4-Methoxyphenol［J］.Journal of American Chemical Society, 1967, 89(22): 5619-5629.

［13］ Brimer MR. Composition Useful as Gum Inhibitor for Motor Fuels: US, 2496930［P］. 1950.

［14］ Chao TS, Hutchison DA, Kjonaas M. Some Synergistic Antioxidants for Synthetic Lubricants［J］. Industrial & Engineering Chemistry Product Research and Development, 1984, 23(1):21-27.

［15］ Davis TG, Thompson JW. Synergistic Antioxidants for Synthetic Lubricants［J］. Industrial & Engineering Chemistry Product Research and Development, 1966, 5(1):76-80.

［3］ Fitch JC.Sludge and Varnish in Turbine Systems［J/OL］.Machinery Lubrication,May 2006［2013-03-04］.http://www.machinerylubrication.com/Read/874/sludge-varnish-turbine.

［4］ Livingston G, Wooten D, Thompson B. Finding the Root Causes of Oil Degradation［J/OL］. Practicing Oil Analysis, January 2007［2013-03-04］.http://www.machinerylubrication.com/Read/989/fluid-degradation-causes.

［5］ Gatto VJ,Moehle WE,Cobb TW,et al.Oxidation Fundamentals and Its Application to Turbine Oil Testing［J/OL］.Journal of ASTM International,2006,3(4)：20 ［2013-04-12］.http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI13498.htm.

［6］ Livingston G, Ameye J, Thompson B. Rethinking Condition Monitoring Strategies for Today's Turbine Oils［J/OL］.Machinery Lubrication, May 2010［2013-04-12］.http://www.machinerylubrication.com/Read/24987/rethinking-condition-monitoring-strategies-turbine-oils.

［7］ Gatto VJ, Liu B, Wang J. Development of a New Sustainable Antioxidant Technology for Low Phosphorus Passenger Car Engine Oils ［C］//Symposium of Dalian Lubricant Technique and Economy Forum, 2010：212-220.

［8］ Livingstone G, Thompson B, Okazaki ME. Physical, Performance and Chemical Changes in Turbine Oils from Oxidation［J/OL］. ASTM Symposium on the Oxidation and Testing of Turbine Oils, 2007, 4 (1)：1-18［2013-04-17］http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI100465.htm.

［9］ Bridger RF. Kinetics of Inhibition of Hydrocarbon Autoxidation by 1,1'-Bis(N-phenyl- 2-naphthylamine)［J］. Journal of Organic Chemistry, 1971, 36(9):1214-1216.

［10］ Zeman A, Romer R, von Roenne V. Fate of Amine Antioxidants during Thermal Oxidative Ageing of Neopentylpolyol Ester Oils: Part 1［J］. Journal of Synthetic Lubricants, 1986, 3(4): 309-325.

［11］ Slobodin YM, Gonor AA. Synergism of Antioxidants in Synthetic Ester Oils［J］. Chemistry and Technology of Fuels and Oils, 1966, 2(3):198-201.

［12］ Mahoney LR, DaRooge MA. Inhibition of Free-Radical Reactions, IV. The Synergistic Effect of 2, 6-Di-t-butylphenols on Hydrocarbon Oxidation Retarded by 4-Methoxyphenol［J］.Journal of American Chemical Society, 1967, 89(22): 5619-5629.

［13］ Brimer MR. Composition Useful as Gum Inhibitor for Motor Fuels: US, 2496930［P］. 1950.

［14］ Chao TS, Hutchison DA, Kjonaas M. Some Synergistic Antioxidants for Synthetic Lubricants［J］. Industrial & Engineering Chemistry Product Research and Development, 1984, 23(1):21-27.

［15］ Davis TG, Thompson JW. Synergistic Antioxidants for Synthetic Lubricants［J］. Industrial & Engineering Chemistry Product Research and Development, 1966, 5(1):76-80.

［3］ Fitch JC.Sludge and Varnish in Turbine Systems［J/OL］.Machinery Lubrication,May 2006［2013-03-04］.http://www.machinerylubrication.com/Read/874/sludge-varnish-turbine.

［4］ Livingston G, Wooten D, Thompson B. Finding the Root Causes of Oil Degradation［J/OL］. Practicing Oil Analysis, January 2007［2013-03-04］.http://www.machinerylubrication.com/Read/989/fluid-degradation-causes.

［5］ Gatto VJ,Moehle WE,Cobb TW,et al.Oxidation Fundamentals and Its Application to Turbine Oil Testing［J/OL］.Journal of ASTM International,2006,3(4)：20 ［2013-04-12］.http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI13498.htm.

［6］ Livingston G, Ameye J, Thompson B. Rethinking Condition Monitoring Strategies for Today's Turbine Oils［J/OL］.Machinery Lubrication, May 2010［2013-04-12］.http://www.machinerylubrication.com/Read/24987/rethinking-condition-monitoring-strategies-turbine-oils.

［7］ Gatto VJ, Liu B, Wang J. Development of a New Sustainable Antioxidant Technology for Low Phosphorus Passenger Car Engine Oils ［C］//Symposium of Dalian Lubricant Technique and Economy Forum, 2010：212-220.

［8］ Livingstone G, Thompson B, Okazaki ME. Physical, Performance and Chemical Changes in Turbine Oils from Oxidation［J/OL］. ASTM Symposium on the Oxidation and Testing of Turbine Oils, 2007, 4 (1)：1-18［2013-04-17］http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI100465.htm.

［9］ Bridger RF. Kinetics of Inhibition of Hydrocarbon Autoxidation by 1,1'-Bis(N-phenyl- 2-naphthylamine)［J］. Journal of Organic Chemistry, 1971, 36(9):1214-1216.

［10］ Zeman A, Romer R, von Roenne V. Fate of Amine Antioxidants during Thermal Oxidative Ageing of Neopentylpolyol Ester Oils: Part 1［J］. Journal of Synthetic Lubricants, 1986, 3(4): 309-325.

［11］ Slobodin YM, Gonor AA. Synergism of Antioxidants in Synthetic Ester Oils［J］. Chemistry and Technology of Fuels and Oils, 1966, 2(3):198-201.

［12］ Mahoney LR, DaRooge MA. Inhibition of Free-Radical Reactions, IV. The Synergistic Effect of 2, 6-Di-t-butylphenols on Hydrocarbon Oxidation Retarded by 4-Methoxyphenol［J］.Journal of American Chemical Society, 1967, 89(22): 5619-5629.

［13］ Brimer MR. Composition Useful as Gum Inhibitor for Motor Fuels: US, 2496930［P］. 1950.

［14］ Chao TS, Hutchison DA, Kjonaas M. Some Synergistic Antioxidants for Synthetic Lubricants［J］. Industrial & Engineering Chemistry Product Research and Development, 1984, 23(1):21-27.

［15］ Davis TG, Thompson JW. Synergistic Antioxidants for Synthetic Lubricants［J］. Industrial & Engineering Chemistry Product Research and Development, 1966, 5(1):76-80.