REFINERY GAS ANALYSIS BY

GAS CHROMATOGRAPHY

UOP Method 539-97

SCOPE

This automated method is for determining the composition of refinery gas samples or expanded liquefied
petroleum gas (LPG) samples obtained from refining processes or natural sources. Non-condensable gases,
hydrogen sulfide, C1 through C4 hydrocarbons and C5 paraffins are reported individually, while C5 olefins
and C6+ hydrocarbons are reported as a composite. Oxygen is not separated from argon. Results for
hydrogen sulfide, if present, may not be quantitative on some analyzers. The method yields quantitative
results from 0.1 to 99.9 mol-% for a single component or composite; except for hydrogen sulfide that yields
quantitative results between 0.1 and 25 mol-%. Results may also be reported in mass-%

OUTLINE OF METHOD

The method requires the use of a dedicated gas chromatographic system that is configured for automated
refinery gas analysis, and is capable, via valve switching, of multi-column, multi-detector operation. The
sample is injected using two sampling valves, and the analysis is performed under isothermal conditions.
Hydrogen and helium are determined on a 13X molecular sieve column using nitrogen carrier gas and a
thermal conductivity detector (TCD). The remainder of the sample components are determined using
hydrogen carrier gas, a series of four columns connected by 6-port and 10-port rotary valves, and a second
TCD. The four columns separate specific portions of the total sample. The first two columns resolve the
gases in the C3-C5 boiling range, carbon dioxide, hydrogen sulfide and the C5 olefins and/or C6+
hydrocarbon composite. The third column resolves the components in the intermediate boiling range,
ethylene and ethane. The light gases, oxygen and/or argon composite, nitrogen, methane and carbon
monoxide, are resolved by the fourth column. Quantitative results are obtained from the measured areas of
the recorded peaks by the application of individual relative response factors, followed by normalization to
100%.

APPARATUS

References to catalog numbers and suppliers are included as a convenience to the method user. Other
suppliers may be used.

IT IS THE USER'S RESPONSIBILITY TO ESTABLISH APPROPRIATE PRECAUTIONARY PRACTICES AND TO

DETERMINE THE APPLICABILITY OF REGULATORY LIMITATIONS PRIOR TO USE. EFFECTIVE HEALTH AND
SAFETY PRACTICES ARE TO BE FOLLOWED WHEN UTILIZING THIS PROCEDURE. FAILURE TO UTILIZE THIS
PROCEDURE IN THE MANNER PRESCRIBED HEREIN CAN BE HAZARDOUS. MATERIAL SAFETY DATA SHEETS
(MSDS) OR EXPERIMENTAL MATERIAL SAFETY DATA SHEETS (EMSDS) FOR ALL OF THE MATERIALS USED IN
THIS PROCEDURE SHOULD BE REVIEWED FOR SELECTION OF THE APPROPRIATE PERSONAL PROTECTION
EQUIPMENT (PPE).

© COPYRIGHT 1963, 1973, 1987, 1997 UOP LLC

ALL RIGHTS RESERVED

UOP Methods are available through ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken PA 19428-2959,
United States. The Methods may be obtained through the ASTM website, www.astm.org, or by contacting Customer Service at
service@astm.org, 610.832.9555 FAX, or 610.832.9585 PHONE.
 2 of 17

Chromatographic columns:

Column 1A, 30 foot, 0.125-inch OD, 20% Sebaconitrile on 80/100 Chromosorb PAW, modified with
phosphoric acid, Hewlett Packard, Cat. No. 19006-80095

Column 1B, 2 foot, 0.125-inch OD, 20% Sebaconitrile on 80/100 Chromosorb PAW, modified with
phosphoric acid, Hewlett Packard, Cat. No. 19006-80090

Column 2, 6 foot, 0.125-inch OD, Porapak Q, 80/100 mesh, Hewlett Packard, Cat. No. 19006-80015

Columns 3 and 4, 10 foot, 0.125-inch OD, molecular sieve 13X, 45/60 mesh, Hewlett Packard, Cat.
No. 19006-80020, two required

Gas purifier, hydrogen, used to remove oxygen from carrier gas, UOP Mat/Sen, Cat. No. P200-1

Gas purifier, nitrogen, used to remove CO2, CO, H2O, O2, and hydrocarbons, UOP Mat/Sen, Cat. No.
P300-1

Integrator, electronic, or equivalent equipment for obtaining peak areas (may be included with the gas
chromatographic system)

LPG expansion apparatus, for quantitative expansion of LPG from a liquid to a gas phase, see LPG
Sampling and list immediately below:

Fitting, male connector, stainless steel, 0.25-inch tube fitting to 0.25-inch male NPT, Swagelok,
Dearborn Valve & Fitting, Cat. No. SS-400-1-4, four required. Sample cylinders having an outlet
fitting other than 0.25-inch female NPT will require a different fitting.

Fitting, port connector, stainless steel, 0.25-inch tube fitting, Swagelok, Dearborn Valve & Fitting,
Cat. No. SS-401-PC, two required

Fitting, union tee, stainless steel, 0.25-inch, Swagelok, Dearborn Valve & Fitting, Cat. No. 400-3

Tubing, stainless steel, 9 inches of Type 304, 0.25-inch OD x 0.21-inch ID, Alltech Associates, Cat.
No. 30301

Vacuum pump, capable of achieving a vacuum of 0.1-mm Hg, Fisher Scientific, Cat. No. 01-115-2

Valve, stainless steel, 0.25-inch Swagelok, Whitey, Dearborn Valve & Fitting, Cat. No. SS-1RS4

LPG expansion cylinder, sample cylinder for containing expanded LPG sample:

Cylinder, 4- x 6-inches, 316 stainless steel, 1380 kPa (200 psi) internal pressure, double connection,
0.25-inch pipe hex bored through, Arthur Harris, Cat. No. B-270

Fitting, hex nipple, stainless steel, 0.25-inch NPT, Cajon, Dearborn Valve & Fitting, Cat. No. SS-4-
HN, four required

Fitting, tee, stainless steel, 0.25-inch NPT, Cajon, Dearborn Valve & Fitting, Cat. No. SS-4-T

Gauge, stainless steel, vacuum-pressure, -100 through +200 kPa (-14.5 to +29.0 psi) range, Matheson
Gas Products, Cat. No. 63-2204

539-97
 3 of 17

Valve, stainless steel, 0.25-inch NPT inlet, 0.25-inch tube fitting outlet, Whitey, Dearborn Valve &
Fitting, Cat. No. SS-1RM4-S4, two required

Recorder (optional), used to supplement integrator plot

Refinery gas analyzer. The analyzer used in this method is based on a commercially available 6890
Hewlett Packard Gas Chromatograph with electronic pneumatic control, dual thermal conductivity
detectors, configured for automated refinery gas analysis, complete with four rotary valves, two
restrictor valves and five columns to perform the method as written. Fig. 1 shows a flow diagram of the
system. Various vendors that provide pre-configured refinery gas analyzers are listed in the
APPENDIX. Other vendors also supply similar systems. Confirm with the selected vendor that the
required separations are provided for the specific sample types to be analyzed.

Regulator, air, two-stage, high purity, Matheson Gas Products, Model 3122-590

Regulator, hydrogen, two-stage, high purity, Matheson Gas Products, Model 3122-350

Regulator, nitrogen, two-stage, high purity, Matheson Gas Products, Model 3122-580

Restrictor, fine metering valve, 0.0625-inch Swagelok, Nupro, Dearborn Valve & Fitting, Cat. No. SS-1-
SG, two required

Leak detector, gas, Alltech Associates, Cat. No. 21-250

Valve, 6-port (two required), 8-port and 10-port rotary valves, Valco Instrument, models C6UWE,
C8UWE and C10UWE, respectively

Sample loop, stainless steel, 100-µL, Valco Instrument, Cat. No. SL100CUW, two required

Tubing, stainless steel, 0.0625-inch OD, Alltech Associates, Cat. No. 300010

REAGENTS AND MATERIALS

All reagents shall conform to the specifications established by the committee on Analytical Reagents of
the American Chemical Society, when such specifications are available, unless otherwise specified.

References to catalog numbers and suppliers are included as a convenience to the method user. Other
suppliers may be used.

Air, compressed, to actuate column switching valves

Hydrogen, 99.95% minimum purity, total hydrocarbons less than 0.5 ppm as methane (zero gas)

Nitrogen, 99.99% minimum purity, total hydrocarbons less than 0.5 ppm as methane (zero gas)

Blend, qualitative, for determining cut times, containing approximately equal concentrations of hydrogen,
argon, nitrogen, carbon monoxide, carbon dioxide, methane, ethane, ethylene, propane, n-pentane and
1,3-butadiene, Matheson Gas Products. CAUTION: 1,3-Butadiene is a suspected human carcinogen.
Avoid exposure while sampling, handling or venting any blend or sample that may contain this
component. If analysis of 1,3-butadiene is not required, delete it from the Qualitative Blend and Blend
2 (increase nitrogen to 16 mol-%). Substitute n-pentane for 1,3-butadiene in the Qualitative Blend and
use the n-pentane peak in place of the 1,3-butadiene peak where cited under Cut Time Determination.

539-97

--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
 4 of 17

Blends, calibration, quantitative, primary standard, Matheson Gas Products, at the nominal levels shown
in Table 1. If the composition of the samples to be analyzed varies significantly from the specified
Calibration Blends, an alternative blend may be utilized that more closely resembles the composition
of the samples. Single point calibrations are acceptable for normalized composition. If hydrogen
sulfide is not present in the sample types being analyzed, it may be deleted from Blend 3 (increase
nitrogen to 50.0 mol-%).

Table 1
Calibration Blend Nominal Concentrations, mol-%

Component Blend 1 Blend 2 Blend 3

Hydrogen 50.0 --- 20.0
Nitrogen 15.0 15.0 45.0
Argon 5.0 --- 5.0
Methane 5.0 10.0 ---
Ethane 5.0 10.0 ---
Ethylene --- 5.0 ---
Propane 12.0 12.0 ---
Propylene --- 6.0 ---
Propadiene --- 1.0 ---
n-Butane 3.0 10.0 ---
Isobutane 3.0 10.0 ---
1-Butene --- 5.0 ---
Isobutylene --- 5.0 ---
trans-2-Butene --- 5.0 ---

--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
cis-2-Butene --- 5.0 ---
1,3-Butadiene --- 1.0 ---
n-Pentane 1.0 --- ---
Isopentane 1.0 --- ---
Carbon Dioxide --- --- 5.0
Carbon Monoxide --- --- 10.0
Helium --- --- 10.0
Hydrogen Sulfide --- --- 5.0

PREPARATION OF APPARATUS

If the pre-configured refinery gas analyzer is purchased, follow the instrument set-up procedure provided
by the manufacturer. For a refinery gas analyzer built in-house, refer to the following procedures for the
instrument set-up on a HP 6890 based system.

539-97
 5 of 17

Instrument Set-up

Install the four rotary valves, two restrictor valves and five columns on the GC as shown in Fig. 1.
CAUTION: Leakage of hydrogen into the confined volume of the column and valve compartments can
cause a violent explosion. It is, therefore, mandatory to test for leaks each time a connection is made and
periodically thereafter. All connecting lines are to be of minimum length and must be in the heated zone.
The restrictors are required to minimize any flow disruption when the flow path through the rotary valves is
changed and must be set to provide constant system pressure at Columns 2 and 3 when they are cut out of
the system.

Use the electronic pneumatic control in the constant flow mode. Establish the column flows in the
following manner. Rotate valve 4 to the off position (solid line position, flow through the column) and set
the column 4 flow rate to 25 mL/min (nitrogen carrier gas flow rate). Rotate valves 1, 2, 3 to the off
position (solid line position, flow through the column) and set the column flow rate on columns 1A, 1B, 2,
and 3 to 40 mL/min (hydrogen carrier gas flow rate).

Restrictor Adjustment

Needle valve restrictors are plumbed into Valves 2 and 3 to provide constant pressure when columns 2
and 3 are out of the flow path. Restrictors are adjusted by monitoring the inlet pressure. Record the inlet
pressure when all valves are off, then turn on Valve 2 and allow 5 to 10 minutes for the flow to equilibrate.
Adjust the restrictor to bring the inlet pressure back to the original value. Allow 5 minutes for flow to
equilibrate between restrictor adjustments. Turn Valve 2 off. Repeat the procedure for Valve 3.

Cut Time Determination

Prior to sample injection, all valves are in the off position as shown in Fig. 1.

Analysis of Hydrogen

The analysis of hydrogen is accomplished by injecting samples from the sample loop on Valve 4.
Hydrogen eluted from Column 4 (13X sieve column) is detected by TCD A and the rest of the components
injected are back flushed before the next analysis.

Cut Time A - the time the signal switches from TCD A to TCD B after hydrogen has been completely
eluted from Column 4 and the time to close Valve 4, back flushing the rest of components after the signal is
switched.

Enter the following commands into the Run Table:

Run Time 0.01 min Valve 4 On

Run Time 1.40 min Signal 1 Switch to TCD B Cut Time A
Run Time 1.43 min Signal 1 Zero
Run Time 1.43 min Valve 4 Off Back flush
Run Time 5 min Stop

Flush the qualitative blend containing hydrogen, nitrogen, argon, carbon monoxide, carbon dioxide,
ethane, ethylene, propane, n-pentane and 1,3-butadiene through the sample loops and start the run. Check
the elution time of hydrogen. The time to switch the signal to TCD B should be after hydrogen has

539-97
 6 of 17

completely eluted from Column 4. Readjust signal switch time and Valve 4 “Off” time in the Run Table if
it is needed. Record this time as Cut Time A.

Analysis of Fixed Gas and Light Hydrocarbons

The analysis of fixed gas and light hydrocarbons is established by injecting samples through the sample
loop on Valve 1 right after hydrogen is injected from Valve 4. Note: Valve 1 is switched shortly after Valve
4 to prevent hydrogen carrier gas from backing up into the sample loop on Valve 4 eliminating the error in
the hydrogen analysis. When sample is introduced into two analysis paths after injection, hydrogen is eluted
first and measured at TCD A. Then, the signal is switched to TCD B to measure fixed gases and
hydrocarbons. The determination of the following cut times is required for the separation of fixed gases and
hydrocarbons.

Cut Time B - the time that Valve 1 closes to back flush C6+ heavies so that all the 1,3-butadiene and the
components lighter than 1,3-butadiene enter into Column 1A, and C6+ heavies elute in the Column 1B back
flush.

Enter the following commands into the Run Table:

Run Time 0.01 min Valve 4 On Injection

Run Time 0.10 min Valve 1 On Injection
Run Time 0.10 min Valve 2 On
Run Time 0.10 min Valve 3 On
Run Time 1.40 min Signal 1 Switch to TCD B Cut Time A
Run Time 1.43 min Signal 1 Zero

--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
Run Time 1.43 min Valve 4 Off
Run Time 1.50 min Valve 1 Off Cut Time B
Run Time 30.0 min Stop

Flush the same qualitative blend used to determine Cut Time A through the sample loops and start the
run. Valves 2 and 3 are switched to the “On” position during the injection to isolate Columns 2 and 3 from
the flow path, and fixed gases (except for hydrogen) and hydrocarbons are separated only by the
Sebaconitrile columns. A chromatogram similar to that shown in Fig. 2 should be obtained. Identify the
peaks by comparing your chromatogram to that shown in Fig. 2. Check the chromatogram for the
appearance of a peak in the C6+ heavies region. If there is a peak in the C6+ region and the 1,3-butadiene
peak is smaller than expected or does not appear, delete the Run Time and enter a later time for switching
Valve 1 to the “Off” position. Repeat the run as above until there is no C6+ peak and the maximum area is
obtained for the 1,3-butadiene peak. Record the final time for switching Valve 1 “Off” as Cut Time B.

Cut Time C - the time that Valve 3 is switched to the “Off” position while Valve 2 stays on. The 13X
sieve column is in the flow path to collect the composite Ar/O2/N2/CH4/CO peak.

Based on the chromatogram obtained in Cut Time B, determine Cut Time C by subtracting 0.1 minutes
from the time that the composite Ar/O2/N2/CH4/CO peak started. Record this time as Cut Time C.

Cut Time D - the time that Valve 2 is switched off while Valve 3 is turned on. The Porapak column is in
the flow path to collect the composite C2=/C2 peak.

539-97
 7 of 17

Based on the chromatogram obtained in Cut Time B, also determine the time at the valley between the
carbon monoxide peak and the ethane peak. This time can be established quite accurately by taking one half
of the difference between the retention times of the two peaks and adding this value to the retention time of
the first peak. Record this time as Cut Time D.

Cut Time E - the time that Valve 2 is returned to the “On” position while both Columns 2 and 3 are
isolated from the flow path. Components CO2 to 1,3-butadiene are eluted from the Sebaconitrile columns
and detected by TCD B.

Based on the chromatogram obtained in Cut Time B, also determine the time at the valley between the
ethane/ethylene peak and carbon dioxide peak and record this as Cut Time E.

Cut Time F - the time that Valve 2 is turned back off to elute the C2=/C2 from Column 2 after 1,3-
butadiene has completely eluted from the Sebaconitrile column.

Based on the chromatogram obtained in Cut Time B, also determine the time for Valve 2 to switch to the
“Off” position by adding three minutes to the retention time of the 1,3-butadiene peak. Record the time as
Cut Time F.

Cut Time G - the time that Valve 3 is turned back off to elute O2, N2, CH4 and CO from Column 3 after
ethane has completely eluted from the Porapak column.

Based on the chromatogram obtained in Cut Time B, also determine the time for Valve 3 to switch to the
“Off” position by adding five minutes to Cut Time F. Ethylene and ethane are expected to be eluted in five
minutes, and then Valve 3 is turned off to elute O2, N2, methane and carbon monoxide from Column 3.
Record this time as Cut Time G.

Enter the Cut Times determined above in a new Run Table, such as:

Run Time 0.01 min Valve 4 On

Run Time 0.10 min Valve 1 On
Run Time 0.10 min Valve 2 On
Run Time 0.10 min Valve 3 On
Run Time 1.40 min Signal 1 Switch to TCD B Cut Time A
Run Time 1.43 min Signal 1 Zero
Run Time 1.43 min Valve 4 Off
Run Time 1.50 min Valve 1 Off Cut Time B
Run Time 2.60 min Valve 3 Off Cut Time C
Run Time 3.20 min Valve 3 On Cut Time D
Run Time 3.20 min Valve 2 Off Cut Time D
Run Time 3.70 min Valve 2 On Cut Time E
Run Time 15.0 min Valve 2 Off Cut Time F
Run Time 20.0 min Valve 3 Off Cut Time G
Run Time 30.0 min Stop

Re-inject the qualitative blend to check the cut times. If necessary, adjust the timing to ensure the proper
separations.
--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---

539-97
 8 of 17

PROCEDURE

Chromatographic Technique

The first time the columns are installed, or any time the columns are replaced, condition the columns
according to the manufacturers’ instructions.

Table 2

Operating Conditions for In-House Built Analyzer

Oven temperature (isothermal) 55oC
Injection port temperature 100oC
Detector temperature 160oC
Carrier gas (A) nitrogen
Flow rate 25 mL/min
Carrier gas (B) hydrogen
Flow rate 40 mL/min

Detector A* TCD
Reference gas nitrogen
Flow rate 40 mL/min
Makeup gas nitrogen
Flow rate 3 mL/min

Detector B* TCD
Reference gas hydrogen
Flow rate 55 mL/min
Makeup gas hydrogen
Flow rate 3 mL/min
___________________
*Consult the manufacturer’s instrument manual for suggested flow rates.

1. Install the gas purifiers in the supply line between the carrier gas source and the carrier gas inlet on
the gas chromatograph.
• The column life is significantly reduced if the gas purifiers are not used.

2. Establish the recommended operating conditions for the in-house built analyzer (see Table 2).
• Other conditions may be used if they produce the required sensitivity and chromatographic separations
equivalent to those shown in the typical chromatogram (Fig. 3).

3. Connect the sample or calibration blend cylinder to the sample inlet (Fig. 1) and purge the sample
loops with the gas to be analyzed.

4. Stop the flow, allow 5 to 10 seconds for the pressure to equilibrate, and start the analysis.

5. Identify the sample components by comparing the resultant chromatogram with the typical
chromatogram (Fig. 3).

539-97
--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
 9 of 17

LPG Sampling

Liquefied petroleum gas (LPG) must be carefully expanded to ensure that a representative sample is
analyzed. Various procedures are used to quantitatively expand LPG from a liquid phase into a
representative gas phase prior to analysis. The following is the recommended procedure that has been
proved to be satisfactory.

1. Assemble the LPG Expansion Cylinder consisting of a small stainless steel expansion cylinder, a
stainless steel gauge with a reading range from vacuum to 200 kPa (gauge) and two stainless steel
shut-off valves (see Fig. 4).
• Some expansion cylinders have two valves (C and D) as shown in Fig. 4, some have only one (Valve C).
The version shown in Fig. 4 is easier to clean, but either may be used.

2. Connect the apparatus to the vacuum system and evacuate the cylinder assembly to 0.013 kPa (0.1-
mm Hg).

3. Connect two small pieces of clean stainless steel tubing, a tee and Valve B to the evacuated cylinder
as shown in Fig. 4 (LPG Expansion Apparatus).

4. Determine if the LPG sample cylinder contains a dip tube. If not, place the LPG sample cylinder in a
vertical position in a hood or well-vented area. Briefly open the bottom valve (A) to check that no
water or sediment is present in the LPG. If the sample cylinder contains a dip tube, invert the cylinder
(both valves on the bottom) and briefly open the valve not connected to the dip tube to check that no
water or sediment is present. If water or sediment is determined to be present, discontinue the analysis
and obtain a clean sample.
• LPG samples are usually contained in a cylinder having valves on both ends or, in some cases, a cylinder
where one of the valves is connected to a dip tube.

5. Connect the bottom valve or the valve connected to the dip tube of the LPG sample cylinder with a
short stainless steel connector to the expansion apparatus. Fully open Valve B in Fig. 4.

6. Open (about 1/4 turn) Valve A, rapidly, on the sample cylinder until only liquid comes out of Valve
B. Important: The valve must be opened wide enough so that a portion of liquid sample enters the
stainless steel tubing before it vaporizes. Fractionation must not take place at the valve, or the
composition of the sample will change.

7. Close Valve B and then Valve A and open Valve C (Fig. 4).

8. Close Valve C on the apparatus and disconnect the apparatus from the sample vessel. A positive
pressure of 69 to 103 kPag (10 to 15 psig) should be displayed on the expansion cylinder gauge. If
not, repeat Steps 1 through 8 with a longer or shorter piece of stainless steel tubing in the expansion
apparatus. The cylinder now contains a gas phase sample that is representative of the LPG sample in
the original pressurized cylinder.

9. Inject the expanded sample following Steps 3 through 5 under Chromatographic Technique.

Calibration

Response factors are required to relate detector response for each sample component to mol-%. Response
factors for hydrogen, oxygen and/or argon composite, isopentane and n-pentane are calculated from

539-97
 10 of 17

Calibration Blend 1. The response factors for C1 to C4 hydrocarbons and nitrogen are calculated from
Calibration Blend 2, while the response factors for hydrogen sulfide (if present), carbon monoxide and
carbon dioxide are determined from Calibration Blend 3.

Analyze each calibration blend three times as described under Chromatographic Technique. The three
replicates should repeat with 3% (relative). If not, rerun until three replicates are obtained with the desired
repeatability. Based on the average of the three replicate analyses, determine the average relative response
factor for each component, to three significant figures, using nitrogen as reference and the following
formula:

AB
F= (1)
CD

where:
A = component of interest, mol-%
B = area of nitrogen peak
C = nitrogen, mol-%
D = average peak area for a component of interest
F = relative response factor

The response factor for argon is used for the unresolved oxygen and/or argon composite peak.
Extrapolate a relative response factor for a C6 hydrocarbon from the relative response factors of propane, n-
butane and n-pentane. Use that factor for the C5 olefin and/or C6+ hydrocarbon peak.

CALCULATIONS

Calculate the actual mol-% concentration of each component or composite (assuming all components
present in the sample are detected) to the nearest 0.1 mol-% using Eq. 2.

FG
Component or Composite, mol-% = 100 (2)
H

where:
F = previously defined, Eq. 1
G = peak area of the component
H = sum of the products FG for all recorded peaks
100 = factor to convert to mol-%

When mass-% concentrations are required, the conversion can be made using Eq. 3. Report results to the
nearest 0.1 mass-%.

YZ
Component or Composite, mass-% = 100 (3)
T

539-97
 11 of 17

where:
T = sum of the products YZ for all components
Y = component concentration, mol-%
Z = molecular weight of component, g/mole
100 = factor to convert to mass-%

PRECISION

Repeatability
Based on two tests performed by each of two analysts on each of two different days (8 tests), the within-
laboratory estimated standard deviations (esd) were calculated for components at specific concentrations in
a synthetic refinery gas blend and are listed in Table 3. Two tests performed in one laboratory by different
analysts on different days should not differ by more than the allowable differences in Table 3 at the
concentrations listed (95% probability).

The data listed in Table 3 are an estimate of short-term repeatability of the method. When the test is run
routinely in the field, a control standard and individual-range chart should be used to develop a better
estimate of the long-term repeatability.

Reproducibility

There is insufficient data to calculate reproducibility of the test at this time.

TIME FOR ANALYSIS

The elapsed time for the analysis of a gas sample is 0.5 hour, with a 0.1 hour labor requirement. The
elapsed time for the analysis of a LPG sample (including expansion of the sample) is 1.0 hour, with a 0.5
hour labor requirement.

Table 3
Repeatability

Component or Concentration, Within-Lab esd, Allowable

Composite mol-% mol-% Difference, mol-%
Hydrogen 50.2 0.15 0.6
Nitrogen 15.0 0.10 0.4
Methane 5.0 0.06 0.2
Ethane 4.9 0.04 0.2
Propane 12.0 0.06 0.2
n-Butane 2.9 0.02 0.1
Isobutane 3.0 0.03 0.1
n-Pentane 1.0 0.02 0.1
Isopentane 1.0 0.01 0.1
Oxygen/Argon 5.0 0.04 0.2
Hydrogen Sulfide 16.4 0.09 0.4

539-97
--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
 12 of 17

SUGGESTED SUPPLIERS

The suggested suppliers for the refinery gas analyzer are listed in the APPENDIX.

Alltech Associates, Inc., 2051 Waukegan Rd., Deerfield, IL 60015 (847-948-8600)

Arthur Harris and Co., 210 N. Aberdeen St., Chicago, IL 60607 (773-666-6832)
Dearborn Valve & Fitting Co., 1540 Old Rand Rd., P.O. Box 847, Wauconda, IL 60084-0847 (847-526-
6900)
Fisher Scientific, 711 Forbes Ave., Pittsburgh, PA 15219 (412-562-8300)
Hewlett Packard Co., 2850 Centerville Rd., Wilmington, DE 19808-1610 (302-633- 8000)
Matheson Gas Products, Inc., P.O. Box 96, Joliet, IL 60434 (815-727-4848)
UOP Mat/Sen, 4509 Golden Foothill Pkwy., El Dorado Hills, CA 95762 (916-939-8800)
Valco Instruments Co. Inc., P.O. Box 55603, Houston, TX 77255 (713-688-9345)

--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---

539-97
 13 of 17

APPENDIX
List of Suggested Suppliers for Refinery Gas Analyzers

Suggested Supplier,
Model # Application Instrument Specifications Address
Hewlett-Packard Separates hydrogen, C6+, C1-C5 Equipped with HP 6890 Series 426 Gallimore Dairy
Company, GC with EPC, HP 3365 Rd., Greensboro, NC
Model HP/AC, ChemStation or HP 3396B 27409

--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
Model G2329A Integrator, Packed Columns, Tel: (800) 227-9770
TCD/TCD
AC Analytical Separates hydrogen, C6+, C1-C5 range Equipped with HP 6890 Series 3448 Progress Dr.,
Controls Inc., per component and fixed gases GC with EPC, HP 3365 Bensalem, PA 19020
AC/HP RGA, ChemStation or HP 3396B Tel: (215) 638-7078
Model 1029 (Turnkey) Integrator, Packed Columns, Fax: (215) 638-7096
TCD/TCD
EG&G Chandler Refinery Gas Analysis for Low C5+ TCD/TCD P.O. Box 470710,
Engineering, Samples (enhanced C4 unsaturates Tulsa, OK 74147-0710
Carle Series 400: separation) Tel: (918) 627-1740
Model 04196-A, Fax: (918) 627-1748
Cat. No. 72030-11
Model 04124-A, Refinery Gas Liquids Analysis of C1-C5 TCD/TCD
Cat. No. 72020-01 saturates and unsaturates, with initial
backflush of C5= and C6
Wasson ECE Standard Refinery Gas: analysis of C1- HP 5890 II GC, one capillary 1305 Duff Dr.,
Instrumentation, C5 paraffins & olefins with initial and one packed column, Fort Collins, CO 80524
Model 397-00 backflush of C6+ hydrocarbons TCD/FID Tel: (303) 221-9179
Fax: (303) 221-9364
Model 383-00 Extended Refinery Gas Analysis: same as above
analysis of C1-benzene paraffins &
olefins followed by initial backflush of
toluene & C8+ heavies as composite
Model 196-00 Refinery Gas by TCD: C1-C5 paraffins TCD/TCD
and olefins, an initial composite C5
olefin/C6+ backflush, fixed gases and
hydrogen
Varian Analytical It provides the separations of oxygen, Varian 3800 GC, Varian Chromatography
Instruments, nitrogen, carbon dioxide, H2S and TCD/TCD, Systems, 2700 Mitchell
Varian 3800 GC hydrocarbons from C1 through C16 Star Chromatography Dr., Walnut Creek, CA
Workstation 94598
Tel: (510) 939-2400
Fax: (510) 945-2102
Renaissance Extended Accelerated Refinery Gas HP6890 GC, two capillary and P.O. Box 373,
Analytical, LLC, Analysis: Initial composite backflush of one packed column, Pearland, TX 77581
System 1 toluene & C8+ followed by C1-C7 dual TCD/FID Tel: (281) 412-0900
paraffins & olefins and benzene (fast Fax: (281) 412-0770
analysis)
System 2 Standard Refinery Gas Analysis: HP6890 GC,
analysis of C1-C5 paraffins & olefins packed columns only,
with initial backflush of C6+ heavies dual TCD/FID
System 3 Refinery Gas Analysis: analysis of C1- HP6890 GC,
C5 paraffins & olefins with initial packed columns only,
backflush of C6 heavies (slow analysis) TCD/TCD
Separation C3-C6 hydrocarbons are determined by HP GC, 100 Nightingale Ln.,
Systems, Inc., the FID. Hydrogen, nitrogen, oxygen, TCD/FID Gulf Breeze, FL 32561
Refinery Gas Analyzer carbon monoxide, carbon dioxide, and Tel: (904) 932-1433
C1 to C2 are determined by TCD Fax: (904) 934-8642

539-97
 14 of 17

Figure 1
Instrument Configuration and Flow Diagram

--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---

539-97
 15 of 17

Figure 2

539-97
Valve Timing, First Chromatogram (A,B, C and D refer to Cut Times, see text)
--`,,```,,`,``,,`,``,``,`,`,`,`-`-`,,`,,`,`,,`---
16 of 17

Figure 3
Typical Chromatogram (E,F,G and H refer to Cut Times, see text)

539-97
 17 of 17

Figure 4
Sample Expansion Setup

539-97