Determining the Fluoride, Chloride, Bromide and Sulfur Concentration in Liquid Butane and Propane Using Metrohm Combustion IC

This article shows how the concentrations of sulfur, chloride, fluoride, and bromide in liquid propane and butane are determined using the Metrohm combustion IC. Two commercial gas suppliers provided two sets of calibration gases: one was used for calibration purposes and another was treated as an unknown.

Four calibration standards were included in the calibration set that contained different concentration of analytes spanning from 25 to 300ppm. The unknown sample spanned from 2 to 400ppm in concentration. Three combustion IC vendors performed an initial inter-laboratory study (ILS) whose results have been included in this study.

Samples

First, the two sets of calibration gas standards were examined. The composition of individual standard sets supplied by the vendors is given in Table 1.

Table 1. Calibration standard set – ISGAS, Houston, TX

Table 2. Standard set #2 (Sample “unknowns”) – DCG Partnership, Pearland, TX

*Values for chloride and bromide were not provided by the manufacturer. These were calculated using certified fluoride values and mass.

However, it should be noted that the values for bromide and chloride were not given by the manufacturer. Chlorobenzene, fluorobenzene, acetonitrile, dimethyl sulfide and bromobenzene in n-butane were used to prepare the calibration standard set (ISGAS) by weight.

Acetonitrile, dimethyl sulfide, and 2-bromo-2-chloro-1,1,1-trifluoroethane in propane were used for preparing the standard set #2 (DCG). In the case of nitrogen, certified values were available but these were not examined in this analysis.

Instruments, Reagents and Solutions

The following instruments were used in this study:

• 930 Compact IC Flex Oven/SeS/Deg — 2.930.2460
• MSM rotor A — 6.2832.000
• IC conductivity detector —2.850.9010
• 920 absorber module — 2.920.0010
• Direct coupling LPG to oven — 6.730.4030
• Combustion module (oven + LPG/GSS) — 2.136.0730
• Adapter to MSM —6.2842.010
• Metrosep A Supp 5 - 150/4.0 — 6.1006.520
• MagIC Net 3.0 Compact — 6.6059.301
• Metrosep A PCC 1 HC/4. 0— 6.1006.310
• Metrosep A Supp 4/5 Guard/4.0 — 6.1006.500
• Metrosep A Trap 1 - 100/4.0 — 6.1014.000
• Metrosep I Trap 1 - 100/4.0 — 6.1014.200

The reagents used in the study were hydrogen peroxide, 30% H₂O₂, EMD SupraPur, from VWR, EM1.07298.0250; IC Eluent concentrate (100x) from ERA (REAIC1100) (sodium carbonate, 0.32M Na₂CO₃/sodium bicarbonate, 0.1M NaHCO₃); and ultrapure water, resistivity >18 MΩ•cm (25°C), type I grade (ASTM D1193).

The following solutions were used:

• Eluent: c(Na₂CO₃) = 3.2mmol/L; c(NaHCO₃) = 1.0mmol/L
• Absorber solution: 150ppm H₂O₂
• Suppressor rinsing: Stream, detector effluent
• Suppressor regeneration solution: c(H₂SO₄) = 500mmol/L

Calibration

The four gas standards, which were illustrated under the calibration standard set (ISGAS) were used to calibrate the combustion IC system. Similar instrument parameters were used for studying all the four gas standards in triplicate.

An average of three replicate analyses at an individual level were used for creating the calibration curve and the same was plotted with a quadratic regression fit for individual analytes. Analytes, such as sulfur, bromide, chloride and fluoride were used and for each analyte certified values were given in units of mg/kg (ppm).

The following parameters were used for ion chromatography:

• Column: Metrosep A Supp 5 - 150/4.0
• Recording time: 20min
• Polarity: +
• Flow: 0.7mL/min
• Sample loop: 250µL
• Temperature coefficient: 2.3 %/°C
• Column temperature: 30°C
• MiPT injection volume: 200µL
• MSM regeneration: Dosino, 0.1mL/min x 9min

The below parameters were used for combustion IC:

• Oxygen (4.6): 300mL/min
• Argon (5.0): 100mL/min
• Post-combustion time: 120s
• Post-combustion rinse: 0.2ml/min
• Oven temperature: 1050°C
• Water inlet: 0.1mL/min
• LPG/GSS Sample Volume: 50µL
• Final rinse after combustion finished: 1mL

Calculation

The peak area was used for calculations for all components through automatic integration with the MagIC Net 3.0 software. Blank subtraction was not carried out and only anion results were used to carry out the calculations. Also user-defined results were not applied to report the end sample concentrations.

Comment

The flame sensor is not used for combustion control in LPG analysis. As a result the combustion time is governed by the length of time needed by the LPG/GSS module to inject the sample as well as the post-combustion time given in the instrument technique. A 1µL sample loop is included in the LPG/GSS module that can be filled up and injected by as much as 50 times. This configuration can inject a changeable volume of LPG sample spanning from 1 to 50µL.

The time needed to finish the sample injection will differ for different volumes of LPG sample. Here 50 µL of the LPG sample was injected for samples and calibrants. The combustion time difference was considered to be insignificant and calculations were not done to account for this source of inconsistency. It took about 5 minutes to inject the sample (1 µL x 50 loop-fills).

As part of an initial ILS with the American Society for Testing and Materials International (ASTM) similar calibration and sample testing was carried out by two more instrument vendors. The ILS was performed within the WK24757 work item to show the preliminary repeatability for a test method development for LPG analysis with the combustion IC. ILS results are illustrated in the following section.

A full, round-robin ILS is likely to be completed in 2015 to establish reproducibility (R) as well as repeatability (r) for the proposed ASTM technique.

The 10mL-Dosino is used for liquid handling of the absorber solution. In other words, the starting volume is set within the absorber vessel before the combustion takes place. Figure 1 shows the setup scheme of ion chromatography, liquid handling and absorption combustion respectively.

Figure 1. Setup Scheme of ion chromatography, liquid handling, and absorption combustion.

During combustion instant absorption of the combusted products is achieved by dosing the absorber solution to the T-piece situated at the combustion tube end. A connection exists between the combustion tube and absorption vessel and the same is given a final rinse with 1mL once the combustion has ended. With the aid of the 5mL-Dosino, water inlet within the combustion tube for pyrohydrolytic environment is performed at a speed of 0.1mL/min.

The MagIC Net software can accurately capture all of the volumes mixed into the absorber solution both during and after combustion eliminating the need for an internal standard. With the help of the 5mL-Dosino, both standards and samples are transferred to the 920 Absorber Module loop using partial loop injection (MiPT). This provides complete flexibility in injection volumes, spanning from 4 to 200µL. For all MiPT injection volumes a permanent volume of 200µL was utilized.

Results and Discussion

Figure 2 shows the chromatogram acquired from 300ppm mixed-anion LPG standard, and Figures 3, 4, 5, and 6 show the fluoride calibration curve, chloride calibration curve, bromide calibration curve and sulfur calibration curve respectively. Figure 7 shows a chromatogram overlay acquired from the unknown sample (standard set #2), using Metrohm CIC.

Figure 2. Chromatogram obtained from 300ppm mixed-anion LPG standard.

Figure 3. Calibration curve - Fluoride

Figure 4. Calibration curve – Chloride

Figure 5. Calibration curve – Bromide

Figure 6. Calibration curve – Sulfur

Figure 7. Overlay of chromatograms from standard set #2 (sample unknowns) using Metrohm CIC.

Tables 3, 4, 5, 6 and 7 shows the results of the unknown samples (standard set #2) using Metrohm CIC, and Table 8 shows the results of the ILS (n = 3) acquired from various CIC vendors.

Table 3. Results Sample 1 – Standard Set #2 (Sample “unknowns”) using Metrohm CIC

Table 4. Results Sample 2 – Standard Set #2 (Sample “unknowns”) using Metrohm CIC

Table 5. Results Sample 3 – Standard Set #2 (Sample “unknowns”) using Metrohm CIC

Table 6. Results Sample 4 – Standard Set #2 (Sample “unknowns”) using Metrohm CIC

Table 7. Results Sample 5 – Standard Set #2 (Sample “unknowns”) using Metrohm CIC

Table 8. Results – Inter-Laboratory Study (n = 3) obtained from multiple CIC vendors

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

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