Using Karl Fischer Gas Analyzers to Identify Moisture in Liquefied Petroleum

Alkylation units, which are found in refiniries, are used to change alkenes and isobutene into alkylates. Alkenes have a low molecular weight whilst alkylates are high octane, branched-chain hydrocarbons. In this process hydrofluoric or sulfuric acids are used as catalysts.

Hydrofluoric acid alkylation units can contain moisture content, which must be tracked accurately. Also, to prevent acid dilution the feedstock of alkene, HF, and isobutene should be tracked for the presence of water, otherwise this may result in unwanted formation of polymer or oily samples.

Currently no ASTM method is available for direct measurement of moisture content in the gas feedstock of alkylation units. In modern techniques online analyzers are used for measuring moisture in these gases but these instruments are known to provide ambiguous results.

Instrumentation

The following instruments were used for the analysis:

875 KF gas analyzer

Figure 1. 875 KF gas analyzer

Metrohm KF Gas Analyzer

Metrohm’s 875 KF gas analyzer employs a range of actuator valves that easily close and open to facilitate the entry of a sample into the system. A MFC is used to regulate the quantity of a sample entering the titration cell. During the initiation of each titration, a sample does not directly reach the MFC. Instead it flows via the sample valve, a fine control valve, an evaporator at 80 °C and then finally enters the MFC. Care must be taken to ensure that the sample remains in a gaseous phase at a steady temperature before reaching the MFC. This is because any liquid entering the MFC can lead to inaccurate results. Soon afterwards the sample is titrated and the amount of gas that flowed via the MFC is converted to mass in order to determine its moisture content. A nitrogen purge is integrated in the system that occurs automatically before and after sampling.

Gas Calibration

In order to determine the mass flow of nitrogen the MFC integrated in the Metrohm KF Gas Analyzer was fully calibrated (Figure 2). However if the mass flow for different gases has to be calculated a calibration factor must be determined first. Next, the balance was mounted with the sample container and a liquefied gas and a PTFE capillary was used to join the sample cylinder to the system. In order to measure the calibration factor, about 10 L of sample was made to pass via the system, and the instrument software was used to save the calibration factor.

Gas calibration

Figure 2. Gas calibration

Analysis and Calculations

Liquefied petroleum gases often include less than 100 µg/g of water. A non-dried isobutene sample is estimated to have over 10X that amount. Since the sample contains a high level of water content prior to drying, it is important to control the volume of sample reaching the system. Based on this a slight adjustment was made to the default analysis to actuate the sample valve for 1 ms and then close it instantly. As soon as the rate of flow reduces to less than 500 mL/min the valve opens and closes until the smallest sample size is recorded. Table 1 shows the calibration factor and water content of isobutene.

Table 1. Calibration factor and water content of isobutene

Calibration Factor Water content
Fcal = Vsample/Msample water(μ/g) = EP x Fcal / Vsample
Fcal: Calibration factor (L/g) water: water content in μg/g
Vsample: Measured (with MFC) volume of sample (L) EP: Determined amount of water at the end of titration (μg)
Msample: Weighed sample mass (g) Fcal: Calibration factor (L/g)
Vsample: Measured (with MFC) volume of sample (L)

This process made it possible to accurately control the amount of sample entering the system. If this modification has not been made, then the analysis time would have been longer because of the size of sample introduced. Once the samples had been dried, they were examined for consistency. The system functionality was ensured by examining the certified nitrogen gas standards.

Water Content in Isobutane Before and After Drying

The isobutane’s calibration factor was found to be 1.42mL/mg. The standard size of the sample registered before and after drying was 2300 and 300mg, respectively (Table 2).

Table 2. Sample size and water content of isobutene before and after drying

Sample Sample Size (mg) water content (μg/g)
Isobutane before drying 2370 4148.5
2320 4144.3
2190 4121.2
Mean: 4138 ±14.7
Isobutane after drying 294 20.7
324 19.9
324 18.4
Mean: 19.7 ± 1.17
Nitrogen 25.5 (mg/L) 4150 25.5
4180 25.7
4180 25.0
Mean: 25.4 ± 0.36 mg/L

For each and every replicate the analysis time was about 300 s. Before and after drying the Isobutane had 4138µg/g water and 19.7µg/g respectively. System functionality was tested with a 25.5 ± 5%mg/L certified nitrogen standard and the recovery result for this standard was recorded at  99.6%. Figure 3 shows Karl Fischer titration curve of isobutane after and before drying.

Karl Fischer titration curve of isobutane after drying and Karl Fischer titration curve of isobutane before drying.

Figure 3. Karl Fischer titration curve of isobutane after drying and Karl Fischer titration curve of isobutane before drying.

Conclusion

In this article both low and high water content samples were examined thoroughly. With the help of Metrohm’s 875 KF gas analyzer the moisture content in feedstock gases before and after drying was reliably and accurately measured, using nitrogen purging and automated sample delivery.

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

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

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