New Standards in Corrosion Control

As a means to improve their margin spread, many refiners often look at discounted opportunity crudes. The growing variety of discounted opportunity crudes on the market contain one or more risks for the buyer, such as high sulfur content and naphthenic acid. Sulfur compounds and napthenic acids are among the many species that contribute to the corrosive nature of crude oils and refined fractions. Therefore, opportunity crudes with high sulfur content and naphthenic acid come with a processing risk of increased corrosion.

When processing these crudes, the refiner needs to balance the cost benefit against the cost of corrosion control and the risk. A reliable acid number determination is a crucial part of corrosion control. Guest Authors Bryce McGarvey, Bert Thakkar and Colette McGarvey of Imperial Oil and Lori Carey and Larry Tucker of Metrohm USA developed the new ASTM Method D8045 for the determination of acid number in crude oil and refinery distillation fractions. Here, they describe the method and explain how it came about.

Occurrence of Naphthenic Acids

Naphthenic acids are found in many types of crude oil and can be present at varying concentrations. Naphthenic acids are found in crudes of various global origins such as Venezuela, California, India, China, Brazil, Mexico, West Africa, Western Canada, North Sea and other regions. With increasing volume and availability of naphthenic-acid-containing crudes, the risk of experiencing high-temperature corrosion of refinery equipment also increases in refinery operations. Piping, furnaces, atmospheric and vacuum distillation columns, overhead systems and side strippers are particularly at risk.

Naphthenic Acid and Sulfur Corrosion

W.A. Derungs1 established the relationship between refinery corrosion and naphthenic acids. He observed that differentiating between sulfide and naphthenic acid corrosion is very difficult. Both sulfide and naphthenic acid produced high corrosion rates at high temperatures. The mechanism of corrosion from combined sulfur and naphthenic acid content has been explained by the following chemical reactions.2, 3, 4

Fe + 2RCOOH ⇌ Fe(RCOO)2 + H2 (1)

Fe + H2S ⇌ FeS + H2 (2)

Fe(RCOO)2 + H2S ⇌ FeS + 2RCOOH (3)

During the initial reaction, iron naphthenates are formed as a result of the reaction between naphthenic acids and steel. Since iron naphthenates are soluble in oil, they are carried in the fluid flow and at the same time hydrogen sulfide or other sulfide-containing species form an iron sulfide coating on the surface (reaction 2), after reacting with steel. Hydrogen sulfide even reacts with the iron naphthenates to produce iron sulfide and release the naphthenic acids (reaction 3).

Although these three reactions indicate the established mechanism of naphthenic acid corrosion in sulfur-containing crudes, naphthenic acid corrosion is actually more complex and influenced by a number of factors such as stream velocity, temperature, as well as sulfur and acid content.5 At process temperatures of more than 200 °C, corrosion risk from naphthenic acids is much greater. At operating temperatures of more than 420 °C, naphthenic acids are assumed to break down into shorter-chain organic acids. These acids can end up in distillation fractions, and their corrosivity is a major concern.6 With the increase in the operating process temperature, the corrosion rate may also increase because of these short-chain organic acids.

Physical Parameters Affecting Corrosion

Flow-induced wall shear stress can separately affect the corrosion rate by both naphthenic acid and sulfur species. Areas of high turbulence and refinery units with process stream flow velocities greater than 2.7 m/s are more sensitive to naphthenic acid corrosion. A thin iron sulfide film, which was formed through the reaction of hydrogen sulfide present in crude with the steel refinery units, protects the steel from naphthenic acid attack. However, the sulfide film can be dissolved by turbulence and high-velocity flow, which means the metal is directly exposed to attack by naphthenic acid.

Desalter Upsets Caused by Naphthenic Acids

Naphthenic acids can cause upsets in the crude oil desalter through the formation of emulsions: as the pH of the water within a desalter increases, naphthenic acids form highly stable sodium naphthenate emulsions. These emulsions should be broken down to reduce fouling and restore the efficiency of the desalter.

The current test method was developed for the analysis of lubricants and presents the analyst with a number of challenges when applying it to crude oil and fractions.

Corrosion Control: Monitoring the Acid Number

To control corrosion during the processing of crude oil, the acid number and the sulfur content of the crude or refinery fraction are measured. The acid number (AN) is defined as the total acidity, i.e., the amount of potassium hydroxide in milligrams needed to neutralize one gram of sample. Traded or crude fractions such as vacuum gas oil (VGO) can have acid numbers up to 4 mg KOH/g. An acid number of less than 1 mg KOH/g is present in most crudes or refinery fractions. Experience from corrosion studies and refineries demonstrates increased corrosion risk when the naphthenic acid content is greater than 1.0 mg KOH/g in fractions, and greater than 0.5 mg KOH/g in crude. If the acid number of a fraction or crude goes beyond these values, it is believed to be a high-acid-number stream. The acid number of crude oil and fractions is currently determined by titrating the total acidity in accordance with ASTM D664.

This test method was first developed for analyzing new and used lubricants and presents the analyst with several analytical difficulties when applying it to fractions and crude oil. For example, inadequate dissolution of fractions and crudes in the titration solvent can present a challenge: bitumen, paraffinic (waxy) and asphaltic materials are not all readily dissolved in the titration solvent that has been stipulated by the D664 Method. When the sample is incompletely dissolved, the titrant will not be able to react with all of the acid present in the sample. Moreover, the undissolved sample also forms a coating on the glass membrane of the electrode in the titration cell, lowering its ability to correctly sense voltage fluctuations during the titration and thus leading to poor accuracy and imprecision.

ASTM D974, a colorimetric titration method, has been used to calculate some clear refinery fractions. However, this test method cannot be used to determine crudes and front-end refinery fractions because they exhibit intense dark color when dissolved in the titration solvent.

A New Method for Determining the Acid Number

A better method is required by the industry to determine acidity in crudes and refined fractions. To resolve the sample and method challenges, Metrohm USA has been working with industry partners to develop an improved, reliable titration method that measures acidity in crudes and refinery fractions. Based on thermometric endpoint detection, the new titration method uses the Metrohm thermometric titrator 859 Titrotherm.

Thermometric Titration: Measuring Principle

The acidity measurement of fractions and crudes in accordance with ASTM D664 employs a potentiometric titration wherein a pH electrode detects the reaction between the naphthenic acid and the titrant. During the titration, the sensor becomes coated with heavier fractions which cannot be easily dissolved in the titration solvent specified by the standard, and thus results in measurement imprecision. A thermometric sensor is used by the new Titrotherm method that overcomes this limitation in two ways: first, there is no glass membrane to coat, and second, the analyst can vary the composition of the solvent to aid the dissolution of bitumen and other heavier crudes

Metrohm has developed a new thermometric sensor for non-aqueous titration of acidity in crude. The sensor uses a thermistor to determine the temperature in the titration vessel. Since the neutralization reaction of naphthenic acid is exothermic, the temperature increases during the course of the reaction. In order to achieve a sharp discontinuity in the temperature curve at the endpoint, a thermometric indicator that reacts endothermally with surplus hydroxide after the endpoint is added to the sample solution. As depicted in Figure 1, the thermometric titration provides an inverted V-shaped curve. The tiamo software handles the evaluation of the titration endpoint.

As titrant is added, an exothermic reaction is measured, i.e. the temperature in the titration vessel increases. After reaching the neutralization or endpoint, excess titrant endothermically reacts with the thermometric indicator present in the solvent, resulting in an abrupt temperature decrease.

Figure 1. As titrant is added, an exothermic reaction is measured, i.e. the temperature in the titration vessel increases. After reaching the neutralization or endpoint, excess titrant endothermically reacts with the thermometric indicator present in the solvent, resulting in an abrupt temperature decrease.

The thermistor’s measuring response time is less than 0.003 seconds which is much shorter than that of a pH glass membrane which means the titration can be performed much faster than the titration performed according to the ASTM D664 method which employs pH indication – without affecting accuracy or precision. Thanks to the thermometric sensor, even nonpolar solvents such as xylene can be used which enhances the solubility of many different oils, including crude. No insulated reaction chamber is required by thermometric titration because a relative temperature change is tracked to indicate completion of the reaction.

The xylene–2-propanol titration solvent is efficient enough that only 20 to 40 mL are required compared to 120 mL when using ASTM Method D664.

Sample Preparation

When measuring relatively small sample sizes of 3 to 5 g, the heterogeneous nature of crude oil can indeed affect determinations. In order to enhance the precision of the test protocol, a shear mixer should be used to homogenize the crude prior to analysis. The work group carried out studies which demonstrated that this enhances the precision of the new test method.

The 859 Titrotherm with magnetic stirrer, two Dosinos, and the tiamo software: This simple setup allows you to measure the acid number of crudes and refinery fractions.

The 859 Titrotherm with magnetic stirrer, two Dosinos, and the tiamo software: This simple setup allows you to measure the acid number of crudes and refinery fractions.

A New Solvent System

A solvent study was preformed to address the problem of dissolving crudes and fractions. It was observed that a mixture of xylene and 2-propanol (also known as isopropyl alcohol or IPA for short) in the ratio of 75:25 by volume worked best to dissolve the range of crudes and refinery fractions. The titration solvent, xylene–2-propanol, is so efficient that just 30 to 40 mL of solvent was required when compared to 120 mL for ASTM Method D664. This saves considerable operating costs in terms of reduced total solvent volume and waste disposal.

At room temperature, crudes and refinery fractions are in liquid state and these are weighed directly into a beaker. The sample is dissolved by adding 30 mL of titration solvent containing the thermometric indicator. It is then titrated with 0.1 molar potassium hydroxide in 2-propanol. Sample preparation is required for samples that are not liquid at room temperature, such as high-paraffin-content fractions and asphalt.

High Paraffin Content: Particularly Challenging Samples

The analyst may come across crude oils with high paraffinic content which are called waxy crudes in the industry. These samples can be very complex as the paraffins are usually solid at room temperature. High paraffinic content samples should be fluidized and homogenized by heating to a temperature of 80 °C, so that a representative aliquot is acquired and analyzed. Next, the warm sample is weighed directly into the titration beaker and 10 mL of solvent is added (xylene or toluene). This first dissolution step was not required by most of the crudes analyzed, including bitumen samples. To ensure the dissolution of the paraffin content, about 30 mL of the xylene–2-propanol titration solvent is added and the sample aliquot is again heated to 65 °C. The aliquot is then analyzed instantly while it is still warm.

Recommended Sample Weights

Using a sample mass of 10 to 20 g, crudes and distillation fractions with a predicted acid number less than 1 should be analyzed. If the acid number is greater than 1, 5 g of sample should be used for the measurement. The amount of sample used can be adjusted to accommodate for solubility limitations. In the case of unknowns, it is recommended to begin with 5 g and adjust the sample size as required for following measurements. The volume of titrant consumed should be at least 0.15 mL, but if a titrant volume is smaller than 0.15 mL it means extra sample has to be used. Similarly, if a titrant volume is greater than 5.0 mL it means less sample is required. Table 1 presents the recommended sample weights according to the predicted value of the acid number.

Table 1. Recommended sample weights.

Expected acid number [mg KOH/g] Recommended sample mass [g] ± 10%
0.05–0.99 10–20
1.00–4.99 5
5.00–15.00 1

Blank Determination

It is important to determine a blank periodically. This should consume less than 0.1 mL titrant, especially when samples with acid numbers less than 1.0 are being measured. To obtain the blank value of less than 0.1 mL, it is recommended to use only ACS Reagent grade solvents. The blank can be determined by measuring a stable sample with a known acid number three or more times, using a different sample mass each time. The largest size of sample should not use a titrant volume greater than the burette volume.

The following example demonstrates a blank calculation for a crude oil sample with an acid number of about 0.9 mg KOH/g. In a plot of the titrant volume consumed until the endpoint versus the sample mass, the blank value is equal to the value of the titrant volume ‘y’ when the sample mass ‘x’ is set to zero, as shown in Figure 2. In this particular case, it equals 0.039 mL. The software can be configured to accommodate the data and measure the slope automatically.

The blank value is determined from three or more determinations of the same sample, each of which is done using a different sample mass. The titrant volume required in these determinations is plotted against the respective sample mass. After applying a linear fit, the blank value, equal to the titrant volume when the sample mass is 0, is extrapolated.

Figure 2. The blank value is determined from three or more determinations of the same sample, each of which is done using a different sample mass. The titrant volume required in these determinations is plotted against the respective sample mass. After applying a linear fit, the blank value, equal to the titrant volume when the sample mass is 0, is extrapolated.

The new thermometric method produces equivalent results to ASTM method D664.

Correlation of Thermometric and Potentiometric Methods

Using the new thermometric titration method, a wide range of crude types and refinery fractions has been analyzed. A study comparing this titration method to the ASTM D664 potentiometric method demonstrates good correlation (Table 2). To compare the method results, 89 samples were examined using thermometric and potentiometric methods in a three-laboratory study. The results demonstrated that the new thermometric method yields equivalent results to ASTM Method D664, as illustrated in Figure 3.

Table 2. Thermometric titration method (ASTM D8045) correlated to the ASTM D664 Method.

Sample Mean thermometric TAN [mg KOH/g] Relative standard deviation Potentiometric TAN [mg KOH/g] Difference
Desalted crude 0.76 2.1% 0.73 4.0%
Raw crude 0.73 1.1% 0.67 8.6%
Light vacuum gas oil 1.23 0% 1.20 2.5%
Heavy vacuum gas oil 1.25 0.8% 1.23 1.6%
Heavy atmospheric gas oil 1.15 1.2% 1.10 4.4%
650 Endpoint gas oil 0.73 1.1% 0.69 5.6%

Correlation of the results of the thermometric TAN determination and the potentiometric TAN determination according to ASTM D664.

Figure 3. Correlation of the results of the thermometric TAN determination and the potentiometric TAN determination according to ASTM D664.

Repeatability

The acid number repeatability of the thermometric method was studied for low-TAN samples in a laboratory. A refinery fraction, a crude oil and a mineral oil were examined. Table 3 shows the results, demonstrating excellent method precision for low TAN values.

Precision

For the development of an ASTM thermometric TAN test method, both the single-laboratory precision and the precision between multiple laboratories have been analyzed. This was performed in a 10-laboratory study of 12 crudes and refinery fractions. The single-laboratory precision, or repeatability, as well as the precision between multiple laboratories, or reproducibility, showed to be much better than that of the ASTM D664 method when measuring refinery fractions and crudes. This is largely due to the enhanced solubility of the sample in the xylene–2-propanol solvent which renders the sample fully accessible for reaction with the titrant.

Table 3. Repeatability of the acid number determination in different samples by thermometric titration*.

Sample Run n° Sample weight
[g]
End point
[mL]
TAN
[mg KOH/g]
Crude oil 1 5.1924 0.1367 0.15
Crude oil 2 5.1623 0.1333 0.14
Crude oil 3 5.0474 0.1400 0.16
Crude oil 4 5.0192 0.1400 0.16
Crude oil 5 5.0100 0.1333 0.15
Crude oil 6 5.0643 0.1333 0.15
Crude oil 7 5.0858 0.1400 0.15
Crude oil 8 5.0956 0.1333 0.15
Crude oil 9 5.0278 0.1467 0.16
Crude oil 10 5.1419 0.1367 0.15
Mean 0.1520
SD** 0.0063
Refinery stream 1 20.152 0.1733 0.05
Refinery stream 2 19.9484 0.1633 0.05
Refinery stream 3 20.0509 0.1600 0.04
Refinery stream 4 20.2944 0.1533 0.04
Refinery stream 5 19.3136 0.1467 0.04
Refinery stream 6 20.1019 0.1567 0.04
Refinery stream 7 20.1044 0.1467 0.04
Refinery stream 8 20.2357 0.1533 0.04
Refinery stream 9 20.1517 0.1533 0.04
Refinery stream 10 20.3568 0.1567 0.04
Mean 0.0420
SD 0.0042
Mineral oil 1 10.2058 0.1267 0.07
Mineral oil 2 10.1955 0.1300 0.07
Mineral oil 3 10.3425 0.1267 0.07
Mineral oil 4 10.1028 0.1167 0.06
Mineral oil 5 10.307 0.1167 0.06
Mineral oil 6 10.0383 0.1200 0.07
Mineral oil 7 10.0328 0.1200 0.07
Mineral oil 8 10.0974 0.1200 0.07
Mineral oil 9 10.0852 0.1167 0.06
Mineral oil 10 10.1142 0.1333 0.07
Mean 0.0670
SD 0.0048

*All samples were dissolved in 30 mL of solvent before being titrated at a rate of 2 mL/min.
**Absolute standard deviation.

With improved measurement, refiners can better adjust their plant operation to control and mitigate corrosion risk.

Conclusion

The new thermometric titration method provides high accuracy for the analysis of the total acid content in both crudes and refinery fractions. Thanks to an improved solvent system, the titration method overcomes the sample solubility issues of the present standard method ASTM D664 and thus leads to better precision. The new method uses 75% less solvent, reduces the analysis time and also lowers the cost of testing significantly.

Table 4 gives an overview of the most major improvements of the new method compared to the ASTM D664 method.

From crude feedstock through to refinery fractions, the determination of the acid number by thermometric titration is made both simple and precise. With enhanced measurement, refiners can optimally adjust their plant operation to control and reduce corrosion risk from naphthenic acid.

Table 4. Method parameters of thermometric titration (ASTM D8045) and ASTM D664.

Parameter ASTM D664 Thermometric
Solvent system Toluene/IPA/H2O (120 mL) Xylene/IPA (30 mL)
Reagent cost per titration $ 4.09 $ 1.07
Titration time ~220 s ~60 s
Sample size for expected TAN of 0.05–1.0 mg KOH/g 20 g ± 2.0 g ~10 g
Sensor maintenance procedure 1. Solvent rinse
2. Two-minute rehydration
3. IPA dip

Fill with electrolyte fill solution.
The probe must not dry out during storage.
1. Solvent rinse

No rehydration
No fill solution
Store dry

References

[1] Derungs, W. A. Naphthenic Acid Corrosion – An Old Enemy of the Petroleum Industry. Corr. 1956, 12, pp. 617–622.

[2] Turnbull, A.; Slavcheva, E.; Shone, B. Factors Controlling Naphtenic Acid Corrosion. Corr. 1998, 54 (11), pp. 922–930.

[3] Slavcheva, E.; Shone, B.; Turnbull, A. Review of naphtenic acid corrosion in oil refining. Br. Corr. J. 1999, 34 (2), pp. 125–131.

[4] Babaian-Kibala, E.; Craig H. L.; Rusk, G. L.; Quinter R. C.; Summers M. A. Naphthenic Acid Corrosion in Refinery Settings. Mater. Perform. 1993, pp. 50–55.

[5] Bota, G. M.; Qu, D., Nesic, S.; Wolf, H. A. Naphthenic Acid Corrosion of Mild Steel in the Presence of Sulfide Scales Formed in Crude Oil Fractions at High Temperature. NACE 2010 Paper # 1035377

[6] Gutzeit, J. Naphthenic Acid Corrosion in Oil Refineries. Mater. Perform. 1977, 16 (10), pp. 24–35.

[7] Murray, D. TAN Thermometric Method Evaluation. Canadian Crude Quality Technical Association Technical Report. April 2014.

This information has been sourced, reviewed and adapted from materials provided by Metrohm AG.

For more information on this source, please visit Metrohm AG.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Metrohm AG. (2019, April 02). New Standards in Corrosion Control. AZoM. Retrieved on June 26, 2019 from https://www.azom.com/article.aspx?ArticleID=14361.

  • MLA

    Metrohm AG. "New Standards in Corrosion Control". AZoM. 26 June 2019. <https://www.azom.com/article.aspx?ArticleID=14361>.

  • Chicago

    Metrohm AG. "New Standards in Corrosion Control". AZoM. https://www.azom.com/article.aspx?ArticleID=14361. (accessed June 26, 2019).

  • Harvard

    Metrohm AG. 2019. New Standards in Corrosion Control. AZoM, viewed 26 June 2019, https://www.azom.com/article.aspx?ArticleID=14361.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit