October 2022

Biofuels, Alternative/Renewable Fuels

ASTM D5453 vs. D7039 and the importance of oxygen correction for B100 samples

The use of biodiesel is rapidly becoming more popular due to growing trends both inside and outside the petroleum industry.

McHenry, L., X-Ray Optical Systems

The use of biodiesel is rapidly becoming more popular due to growing trends both inside and outside the petroleum industry. The U.S. Energy Information Administration’s (EIA’s) data shows that biodiesel production in the U.S. has increased by more than 100 MM gal/mos since 2011, with biodiesel making up 4% of total diesel consumption in 2016. The EIA forecasts biofuels production to increase between 18%–55% over the next 30 yr. Responding to the evolving industry and social landscape, many traditional refineries have begun to incorporate biodiesel into their finished products.

While biofuels typically contain low amounts of sulfur, they are still required to adhere to fuel quality compliance specifications either for use in vehicles or as a blending feed for traditional refinery fuels. As such, biorefineries must measure the sulfur in their product to ensure that it is below regulatory limits, typically less than 15 parts per million (ppm). Biofuel analysis can be challenging due to the variety of feedstocks and to changing sample compositions.

According to ASTM D6751-20a [Standard Specification for Biodiesel Fuel Blend Stock (B100) for Middle Distillate Fuels], there are several options for testing sulfur in biodiesel. ASTM D5453 is listed as the referee method, but D7039 may also be used. This article examines these methods in more detail, using data from the ASTM B100 Proficiency Testing Program (PTP), and discusses renewable diesel vs. biodiesel, as well as the importance and methodologies for oxygen correction.

ASTM B100 PTP overview

The ASTM B100 PTP allows biodiesel laboratories to improve their testing performances by comparing their biodiesel test results with other laboratories. The statistical analysis also provides a valuable tool to assess test method performance on a particular matrix type and allows comparison between two or more test methods that measure the same property. For the B100 PTP, ASTM sends 1-gal samples out three times per year for analysis of approximately 24 biodiesel properties. This article will focus on sulfur analyses from 2018–2021, using ASTM D5453 and ASTM D7039. First, understanding the test methods is critical to interpreting the data presented.

ASTM D7039 (Monochromatic wavelength dispersive x-ray fluorescence)

Monochromatic wavelength dispersive x-ray fluorescence (MWDXRF) is a subset of WDXRF that utilizes similar principles. Rather than using filters or traditional crystals that are flat or singly curved, MWDXRF incorporates doubly curved crystal (DCC) optics to provide a focused, monochromatic excitation x-ray beam to excite the sample. A second DCC optic is used to collect the sulfur signal and focus it onto the detector. This modified methodology delivers a signal-to-background ratio that is 10 times more precise than traditional WDXRF, which improves method precision and limit of detection (FIG. 1).

FIG. 1. MWDXRF delivers a signal-to-background ratio that is 10 times more precise than traditional WDXRF, which improves method precision and limit of detection.
FIG. 1. MWDXRF delivers a signal-to-background ratio that is 10 times more precise than traditional WDXRF, which improves method precision and limit of detection.

ASTM D5453: MWDXRF. In ultraviolet fluorescence (UVF) technology, a hydrocarbon sample is either directly injected into a high-temperature (1,000°C) combustion furnace or placed in a sample boat that is cooled and then injected into the combustion furnace. The sample is combusted in the tube, and sulfur is oxidized to sulfur dioxide (SO2) in the oxygen-rich atmosphere. A membrane dryer removes water produced during the sample combustion, and the sample combustion gases are exposed to ultraviolet (UV) light. SO2 is excited (SO2*), and the resulting fluorescence that is emitted from the SO2* as it returns to the stable state is detected by a photomultiplier tube. The resulting signal is a measure of the sulfur contained in the sample.

ASTM test method scope and precision

Within the ASTM test method, the scope defines the test method parameters, including matrices of interest and ranges of applicability. This scope is defined by an interlaboratory study (ILS), which also determines the precision (repeatability and reproducibility) of the test method. (Note: This is a separate study from the PTP). Both ASTM D7039 and D5453 include diesel, biodiesel and biodiesel blends—see TABLE 1 for the applicable range of these test methods, along with precision equations for each test method.

The ILS study is a discrete study used to define the repeatability and reproducibility of the test method. The advantage of these studies is that they cover multiple sample matrices spanning the entire concentration of the test method. The disadvantage is that these studies are from a discrete point in time, and they typically do not provide in-depth data on a particular sample type. For this information, it is better to look at ongoing ASTM PTP studies, which are organized around a particular sample type, rather than around sample properties (test methods). By filtering multiple PTP test cycles for a sample property, it is possible to get in-depth looks at particular test methods.

ASTM defines precision in terms of repeatability and reproducibility:

  • Repeatability is the difference between successive results obtained by the same operator in the same laboratory with the same apparatus and the same test method under constant operating conditions on identical test material.
    • A lower repeatability value correlates to a better level of precision and to a higher likelihood of obtaining the same or similar test results over multiple measurements of different aliquots of the same sample.
      • Reproducibility is the difference between two single and independent results obtained by different operators who are applying the same test method in different laboratories, using different apparatus on identical test material.
    • A lower reproducibility value correlates to a better level of precision, which can minimize risks (such as incurring regulatory fines and contract disputes) that can result from inaccurate reporting.

Next, we will look at sulfur data from an ASTM biodiesel (B100) PTP.

ASTM B100 PTP results

There were 12 biodiesel program cycles (or data points) from 2018–2021. On average, there are three times as many D5453 participants vs. D7039 participants— although, if participants are submitting data using both sulfur methods, this value may be skewed. Only one result is submitted per laboratory for each test method; therefore, the program statistics cannot include sulfur repeatability, thus limiting this discussion to sulfur reproducibility. The sulfur data and statistics can be summarized as follows:

  • The average sulfur concentration ranged from 0.27 ppm–6.7 ppm (FIG. 2, TABLE 2).
    FIG. 2. ASTM B100 PTP sulfur data, January 2018–June 2021.
    FIG. 2. ASTM B100 PTP sulfur data, January 2018–June 2021.
  • Half of the sulfur data points (six for D5453 and seven for D7039) are below the test method scopes (TABLE 2).
  • Approximately 58% of the data (0.27 ppm–1.12 ppm sulfur) has a lower sulfur concentration than its associated reproducibility (FIG. 2, TABLE 2).
  • Of the remaining 42% of data (FIG. 2, TABLE 2), D7039 has consistently equal or better reproducibility and is biased lower than D5453.
  • The author’s D7039 reported results with sulfur concentrations within the D7039 method scope (as shown in the FIG. 2 red Xs right of the dotted line, and in TABLE 2 results below the dotted line) were closer to the average D5453 sulfur concentration than the rest of the D7039 data.

What does all this mean? In short, this data snapshot suggests the following:

  1. Neither D5453 nor D7039 are suitable for B100 samples ≤ 1 ppm sulfur.
  2. D7039 has equivalent or better precision than D5453 for B100 samples within the D7039 method scope.
  3. There is some evidence to suggest that PTP D7039 method users may not be correcting for oxygen matrix effects.

It is not surprising that the PTP reproducibility is poor for B100 samples ≤ 1 ppm sulfur, as this concentration range is at or below the lower limit of both methods. This is because the lower limit of an ASTM method is based on the precision of the ILS data (it is not just based on the lowest concentration in the ILS). While D7039 has equivalent or better precision than D5453 for B100 samples within the D7039 method scope, it would have been interesting to see the reproducibility statistics for PTP data in the > 1 ppm–3.5 ppm sulfur range. More data within this range would have solidified whether D7039 was equivalent or better than D5453, as the lower limit for D7039 is 3.2 ppm.

Lastly, there is limited evidence suggesting that PTP participants using D7039 may not be correcting for oxygen matrix effects.

Renewable diesel vs. biodiesel and oxygen effects on x-ray fluorescence (XRF)

Biofuels are any liquid fuels made from renewable biomass, including ethanol, biodiesel and renewable diesel. While the terms “renewable diesel” and “biodiesel” are sometimes used interchangeably, they are actually different. According to the Alternative Fuels Data Center, renewable diesel is a biomass-derived hydrocarbon that meets ASTM D975 specifications for diesel fuel, and is produced through various processes such as hydrotreating, gasification, pyrolysis, and other biochemical and thermochemical technologies. Biodiesel is a mono-alkyl ester that meets ASTM D6751 specifications for biodiesel and is produced via transesterification.

Another difference between renewable diesel and biodiesel is that biodiesel contains oxygen, typically around 10 wt%–12 wt%, whereas finished renewable diesel does not contain oxygen and is considered a drop-in product. Feedstocks for biodiesel and renewable diesel may contain varying amounts of oxygen, depending on the type of feedstock and where in the process the intermediate stream has been sampled.

From an ease-of-use standpoint, drop-in products are easy to measure using XRF, as no additional precautions are needed, and the sample can be measured on a typical hydrocarbon calibration. For diesel-like matrices, samples above 2.5% oxygen must be addressed through matrix-matched calibration standards or correction factors. The high oxygen content in these samples leads to significant absorption of sulfur Kα fluorescence, and, if uncorrected, to low sulfur results (see Section 5.2 in D7039).

Matrix matching uses calibration standards with the same or similar elemental composition as the samples being measured. For biodiesels, it is possible to make or obtain calibration standards in a biodiesel matrix. However, it should be noted that true biodiesel blanks are difficult to find, as they are usually sulfur contaminated. Consider using methyl oleate or octanol for a biodiesel blank, instead of the biodiesel blank that comes in the calibration set. Chances are it is not blank and may cause issues when measuring low-concentration samples.

For oxygenated feedstocks or samples with varying oxygen content, it may be advantageous to use correction factors instead. ASTM D7039 Table 2 (TABLE 3) has correction factors for varying amounts of oxygen in biodiesel measured on a mineral oil calibration. The correction factor is applied by multiplying the uncorrected measured result by the correction factor to obtain the oxygen-corrected result. Note: The correction factors are limited to D7039-compliant MWDXRF systems, such as the author’s company’s proprietary analyzersa. Also, these correction factors can be used on these analyzers when in both 7039 and 2622 modes because the correction factors in TABLE 3 are applied to the sulfur ppm values calculated from the total counts per second (cps) in 7039 mode or from net cps in 2622 mode (background counts are subtracted), at which time the basic analyzer geometry is identical.

Working through a couple of examples, consider two biodiesel samples containing 10 wt% oxygen measured on a mineral oil calibration:

  • (uncorrected measured value) × (correction factor) = sulfur corrected value
  • 1 ppm sulfur (uncorrected) × 1.1740 = 1.2 ppm sulfur (corrected)
  • 10 ppm sulfur (uncorrected) × 1.1720 = 11.7 ppm sulfur (corrected).

Because the correction factors in TABLE 3 are multiplicative, as the sulfur concentration increases, the difference between the oxygen corrected and uncorrected values is greater, which creates a widening of the gap between the measured value and the true value of the sample. A visual representation of this would look like the line graph in FIG. 2, wherein the average concentration difference between D5453 and D7039 widens as sulfur concentration increases.

To be clear, we do not know for certain if the bias between D7039 and D5453 is due to oxygen correction issues because D7039 PTP participants do not report their calibration matrix and method of oxygen correction. However, it can be surmised that this is at least part of the issue based on the red Xs in FIG. 2. These red Xs represent PTP samples measured at the author’s company by using D7039, a mineral oil calibration and correcting the measurement result for 10 wt% oxygen. In this instance, it is known that the submitted results were corrected for oxygen, and it can be observed that, as the sulfur concentration increases, these results stay more consistent with the average D5453 sulfur concentration than the rest of the D7039 data. However, this theory is based on limited data from a single user, so it will be interesting to see if this trend continues as more data is collected. If there can be a takeaway from this, it is that it becomes increasingly important to correct for oxygen as the sulfur concentration increases.

Takeaways

Despite D5453 being the referee method for B100, data from the ASTM B100 PTP shows that D7039 has equivalent or better precision than D5453 for samples above 3 ppm. Data from this program also suggest that D7039 participants are not correcting for oxygen content, which not only becomes more important as sulfur concentration increases, but also may be responsible for the low sulfur bias relative to D5453 seen on the higher-concentration samples in this ongoing study.

Additionally, this article discussed the difference between renewable diesel and biodiesel, and how renewable diesel is a drop-in product that complies with the diesel specification and does not require the oxygen correction or matrix matching of biodiesel samples. HP

NOTE

a XOS’s Sindie 7039, Sindie 2622 and Sindie+Cl analyzers

The Author

Related Articles

From the Archive

Comments

Comments

{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}