APPENDIX C
STANDARD for
AIR-CONDITIONING &
REFRIGERATION
iNSTlTUTE
ANALYTICAL .
PROCEDURES
FOR ARI
STANDARD
700-99
Appendix C I to ARI 700-99 1
4301 NORTH FAIRFAX DRIVE ARLINGTON, VIRGINIA 22203
IMPORTANT
SAFETY RECOMMENDATIONS
It is strongly recommended that the product be designed, constructed. assembled and installed in accordance with
nationally recognized safety requirements appropriate for products covered by this standard.
ARI, as a manufacturers' trade association, uses its best efforts to develop standards employing state-of-the-art and
accepted industry practices. However, ARI does not cenify or guarantee safety of any products, components or
systems designed, tested, rated, installed or operated in accordance with these standards or that any tests conducted
under its standards will be non-hazardous or free from risk.
FOREWORD
The intent of this Appendix is to establish definitive test procedures for determining the quality of new, reclaimed
and/or repackaged refrigerants for use in new and existing refrigeration and air-conditioning equipment within +e
scope of ARI.
These test merhods are established as referee test methods. If alternative test methods are employed, the user must
be able to demonstraE that they produce results at least equivalent to the specified referee test methods (see Section
7 of each test method for information concerning the sensitivity. precision and accuracy of that test method).
Note:
This Appendix C to ARI Standard 700-99 supersedes Appendix C to ARI Standard 700-95.
TABLE OF CONTENTS
Part 1.
Part 2.
Part 3.
Part 4.
Part 5.
Part 6.
Part 7.
Part 8.
Part 9.
Part 10.
Part 11.
Part 12.
Part 13.
PAGE
Determination of Acidity in New and Reclaimed
Refrigerants by Titration ................................................................................... 1
Determination of Water in New and Reclaimed
Refrigerants by Karl Fischer Coulometric Titration ......................................... 4
Determination of High Boiling Residue in New and
Reclaimed Refrigerants by Volumetric and/or
Gravimetric Measurement and Determination of
Particulate Residue by Visual Indication .......................................................... 9
Determination of Chloride in New and Reclaimed
Refhgerants by Silver Chloride Precipitation ................................................. 13
Determination of Non-Condensable Gas in New and
Reclaimed Refiigerants by Gas Chromatography .......................................... 16
Determination of Purity of New and Reclaimed
Refngerant 1 1 by Gas Chromatography ......................................................... 25
Determination of Purity of New and Reclaimed
Refngerant 12 by Gas Chromatography ......................................................... 32
Determination of Purity of New and Reclaimed
Refhgerant 13 by Gas Chromatography ......................................................... 41
Determination of Purity of New and Reclaimed
Refrigerant 22 by Gas Chromatography ......................................................... 49
Determination of Purity of New and Reclaimed
Refngerant 23 by Gas Chromatography ......................................................... 59
Determination of Purity of New and Reclaimed
Refhgerant 32 by Gas Chromatography ......................................................... 67
Determination of Purity of New and Reclaimed
Refngerant 1 13 by Gas Chromatography ....................................................... 77
Determination of Purity of New and Reclaimed
Refrigerant 1 14 by Gas Chromatography ....................................................... 84
TABLE OF CONTENTS (CONTINUED)
PAGE
Part 14.
Part 15.
Part 16.
Part 17.
Part 18.
Part 19.
Part 20.
Part 21.
Part 22.
Part 23.
Part 24.
T
Determination of Purity of New and Reclaimed
Refngerant 123 by Capillary and Packed Column
Gas Chromatography ...................................................................................... 93
Determination of Punty of New and Reclaimed
Refhgerant 124 by Gas Chromatography ..................................................... 103
Determination of Purity of New and Reclaimed
Refhgerant 125 by Gas Chromatography ..................................................... 1 12
Determination of Purity of New and Reclaimed
Refrigerant 134a by Gas Chromatography ................................................... 120
Determination of Purity of New and Reclaimed
Refngerant 143a by Gas Chromatography ................................................... 13 1
Determination of Composition of New and Reclaimed
Refrigerant 401 Blends of R-22, R-152a and R-124
by Gas Chromatography ............................................................................... 14 1
Determination of Composition of New and Reclaimed
Refhgerant 402 Blends of R-125, R-22 and R-290
by Gas Chromatography ............................................................................... 146
Determination of Composition of New and Reclaimed
Refngerant 404A, A Blend of R- 125, R- 134a and
R-143a by Gas Chromatography .................................................................. 15 1
Determination of Composition of New and Reclaimed
Refrigerant 405 Blends by Gas Chromatography ......................................... 157
Determination of Composition of New and Reclaimed
Refhgerant 406 Blends by Gas Chromatography ......................................... 158
Determination of the Composition of New and Reclaimed
Refrigerant 407 Blends of R-32, R-125 and R-l34A by
Gas Chromatography ....................................................... .;. .......................... 163
11
..
TABLE OF CONTENTS (CONTINUED)
Part 25.
Part 26.
Part 27.
Part 28.
Part 29.
Part 30.
Part 31.
Part 32.
Part 33.
Part 34.
Part 35.
Part 36.
Part 37.
Determination of the Composition of New and Reclaimed
Refngerant 408 Blends of R-125, R-22 and R-143A by
Gas Chromatography .................................................................................... 170
Determination of the Composition of New and Reclaimed
Refngerant 409 Blends of R-22, R142b and R-124 by
Gas Chromatography .................................................................................... 175
Determination of the Composition of New and Reclaimed
Refrigerant R-410 Blends of R-32 and R-125 by Gas
Chromatography ........................................................................................... 1 8 1
Determination of Composition of New and Reclaimed
Refngerant 41 1 Blends by Gas Chromatography ......................................... 186
Determination of Composition of New and Reclaimed
Rehgerant 412 Blends by Gas Chromatography ......................................... 191
Determination of Azeotropic Composition of New and
Reclaimed Refrigerant 500 by Gas Chromatography ................................... 192
Determination of Azeotropic Composition of New and
Reclaimed Refngerant 502 by Gas Chromatography ................................... 197
Determination of Azeotropic Composition of New and
Reclaimed Refngerant 503 by Gas Chromatography ................................... 202
Determination of Azeotropic Composition of New and
Reclaimed Refrigerant 507 by Gas Chromatography ................................... 207
Determination of Azeotropic Composition of New and
Reclaimed Rehgerant 508 Blends by Gas Chromatography ....................... 212
Determination of Azeotropic Composition of New and
Reclaimed Refngerant 509 Blends by Gas Chromatography ....................... 217
Determination of Composition of New and Reclaimed
Refngerant 403 Blends by Gas Chromatography ........................................ 222
Determination of Composition of New and Reclaimed
Refrigerant 413 Blends by Gas Chromatography ........................................ 223
...
111
LIST OF FIGURES
RAGE
Part 5 .
Part 6 .
Part 7 .
Part 8 .
Part 9 .
Part 10 .
Part 11 .
Part 12 .
Figure 1 . Oven Column Design for a SRI 8610 Gas
Figure 2 . Gas Chromatogram of Non-Condensable Gas
Figure 3 . Vapor Pressure/Temperature Correlations
Figure 4 . Evacuated System Method of Introducing
Chromatograph ............................................................................ 20
(Air) in R- 12 ................................................................................ 21
for Refhgerants ............................................................................ 22
Vapor Sample into Gas Chromatograph ...................................... 23
Figure 1 . Apparatus used for Calibration Standard Preparation ................. 28
Figure 2 . Gas Chromatogram of R-1 1 ......................................................... 29
Figure 1 . Apparatus used for Calibration Standard Preparation
Figure 2 . Gas Chromatogram of R-12 ......................................................... 38
and for Cylinder Sampling ........................................................... 37
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ............................................................ 46
Figure 2 . Gas Chromatogram of R-13 ......................................................... 47
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ........................................................... 54
Figure 2 . Packed Column Gas Chromatogram ofR-22 .............................. 55
for Determination of R-3 1 Impurity ............................................ 56
Figure 3 . Capillary Column Gas Chromatogram of R-22
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ........................................................... 64
Figure 2 . Gas Chromatogram of R-23 ......................................................... 65
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ........................................................... 72
Figure 2 . Packed Column Gas Chromatogram of R-32 .............................. 73
Figure 3 . Capillary Column Gas Chromatogram of R-32 ........................... 74
Figure 1 . Apparatus used for Calibration Standard Preparation ................. 80
Figure 2 . Gas Chromatogram of R-1 13 ....................................................... 81
iv
Part 13 .
Part 14 .
Part 15 .
Part 16 .
Part 17 .
Part 18 .
Part 19 .
Part 20 .
Part 21 ..
Part 23 .
Part 24 .
Figure 1 . Apparatus used for Calibration Standard Preparation
Figure 2 . Gas Chromatogram of R-114 ..................................................... 90
and for Cylinder Sampling ........................................................... 89
Figure 1 . Apparatus used for Calibration Standard Preparation ................. 97
Figure 2 . Packed Column Gas Chromatogram of R-123 ............................ 98
Figure 3 . Capillary Column Gas Chromatogram of R-123 ......................... 99
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 107
Figure 2 . Gas Chromatogram of R-124 ..................................................... 108
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 116
Figure 2 . Gas Chromatogram of R-125 ..................................................... 117 ..
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling .......................................................... 125
Figure 2 . Packed Column Gas Chromatogram of R-134a ........................ 126
Figure 3 . Capillary Column Gas Chromatogram of R-134a ..................... 127
Figure 1 . Apparatus used for Calibration Standard Preparation
Figure 2 . 1 % SP- 1000 Packed Column Gas Chromatogram
Figure 3 . Combination Packed Column Gas Chromatogram
and for Cylinder Sampling ......................................................... 136
of R-143a ................................................................................... 137
of R-143a ................................................................................... 138
Figure 1 . Gas Chromatogram of R-401 Refrigerant Blend ....................... 145
Figure 1 . Gas Chromatogram of R-402 Refiigerant Blend ....................... 150
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 155
Figure 2 . Gas Chromatogram of R-404A .................................................. 156
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 161
Figure 2 . Gas Chromatogram of R-406A .................................................. 162
Figure 1 . Apparatus used for Calibration Standard Preparation
Figure 2 . Gas Chromatogram of R-407 ..................................................... 169
and for Cylinder Sampling ......................................................... 168
V
Part 25 .
Part 26 .
Part 27 .
Part 28 .
Part 30 .
Part 31 .
Part 32 .
Part 33 .
Part 34 .
Part 35 .
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 173
Figure 2 . Gas Chromatogram of R-408A ................................................. ~ 7 4
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 179
Figure 2 . Gas Chromatogram ofR-409A .................................................. 180
Figure 1 . Gas Chromatogram of R-410 ..................................................... 185
Figure 1 . Apparatus used for Calibration Standard Preparation
and for Cylinder Sampling ......................................................... 189
Figure 2 . Gas Chromatogram of R-411 ..................................................... 190
Figure 1 . Gas Chromatogram of R-500 ..................................................... 196
Figure 1 . Gas Chromatogram of R-502 ..................................................... 201
Figure 1 . Gas Chromatogram of R-503 ..................................................... 206
Figure 1 . Gas Chromatogram of R-507 ..................................................... 211
Figure 1 . Apparatus Used for Sampling Calibration Standards
and Samples ............................................................................... 215
Figure 2 . Gas Chromatogram of R-508 Blends ........................................ 216
Figure 1 . Apparatus Used for Sampling Calibration Standards
and Samples ............................................................................... 219
Figure 2 . Gas Chromatogram of R-509 Blends ........................................ 220
vi
. .
LIST OF TABLES
PAGE
Part 2 .
Part 3 .
Part 5 .
Part 6 .
Part 7 .
Part 8 .
Part 9 .
Pati 10 .
Part 11 .
Part 12 .
Part 13 .
Table 1 . Single Operator Method Precision .................................................. 8
Table 2 . Testing for Percent Water Recovery ................................................ 8
Table 1 . Densities of Liquid Rehgerants .................................................... 12
Table 1 . K, Values for Rehgerants at 75°F (23.9"C) ................................. 24
Table 1 . Retention Time Data for Identified Impurities
Not Normally Observed ................................................................. 30
Table 2 . Component Statistical Parameters ................................................. 30
Table 3 . Primary Calibration Standard Components ................................... 31
Table 1 . Component Statistical Parameters ................................................. 39
Table 2 . Retention Time Data for Identified Impurities
Table 3 . Primary Calibration Standard Components ................................... 40
Not Normally Observed ................................................................. 39
Table 1 . Component Statistical Parameters ................................................. 48
Table 2 . Primary Calibration Standard Components ................................... 48
Table 1 . Component Statistical Parameters ................................................. 57
Table 2 . Additional Impurities Observed in R-22 ....................................... 57
Table 3 . Primary Calibration Standard Components ................................... 58
Table 1 . Component Statistical Parameters ................................................. 66
Table 2 . Primary Calibration Standard Components ................................... 66
Table 1 . Component Statistical Parameters ................................................. 75
Table 2 . Additional Impurities Observed in R-32 ....................................... 75
Table 3 . Primary Calibration Standard Components ................................... 76
Table 1 . Component Statistical Parameters ................................................. 82
Table 2 . Primary Calibration Standard Components ................................... 83
Table 1 . Component Statistical Parameters ................................................. 91
Table 2 . Primary Calibration Standard Components ................................... 92
vii
.
LIST OF TABLES (CONTINUED)
PAGE
Part 14 . Table 1 . Component Statistical Parameters ............................................... 100
Table 2 . Additional Impurities Observed in R.123.
Quantitation by Effective Carbon Number Method .................... 101
Table 3 . Primary Calibration Standard Components ................................. 102
Part 15 . Table 1 . Retention Time Data for Identified Impurities
Not Normally Observed ............................................................... 109
Table 2 . Component Statistical Parameters ............................................... 110
Table 3 . Primary Calibration Standard Components ................................. 111
Table 4 . Primary Calibration Standard Liquid Impurities ......................... 111
Part 16 . Table 1 . Component Statistical Parameters ............................................... 118
Not Normally Observed ............................................................... 118
Table 2 . Retention Time Data for Identified Impurities
Table 3 . Primary Calibration Standard Components ... ; ............................. 119
Part 17 . Table 1 . Additional Impurities Observed in R.134a.
Quantitation by Effective Carbon Number Method .................... 128
Table 2 . Component Statistical Parameters ............................................... 129
Table 3 . Primary Calibration Standard Components ................................. 130
Table 4 . Primary Calibration Standard Liquid Impurities ......................... 130
Part 18 . Table 1 . Component Statistical Parameters ............................................... 139
Table 2 . Primary Calibration Standard Components ................................. 140
...
VI11
APPENDIX C TO ARI STANDARD 700
PART 1
DETERMINATION OF ACIDITY IN NEW AND
RECLAIMED REFRIGERANTS BY TITRATION
Section 1. Purpose
The purpose of this test method is to determine the amount
of acidity in new and reclaimed refigerants.
Sectlon 2. Scope
This test method is for use with low, medium and high
pressure fluorocarbon refrigerants.
Section 3. Definitions
Definitions for this part are identical to
Standards 700-95 and 740-95.
those of ARI
Section 4. Principle
A known quantity of a liquid refrigerant sample is added
to, or bubbled through, a mixm of toluene, isopropanol
and water with bromothymol blue indicator. The acidity
i m p d to the extraction solvent is titrated with
standardized potassium hydroxide to the indicator endpoint.
The acidity is reponed in ppm as HCl.
Section 5. Applicability
This method is applicable to the routine quantitative
determimion of acidity in low, medium and high pressure
refrigerants, including alternative replacements for CFCs.
Section 6. Limitations and interferences
None of the refrigerants tested interfere with the acidity
determination. The test must be performed quickly after
the indicator solution is brought to its p n endpoint to
avoid interferences from atmospheric carbon dioxide.
Section 7. Sensitivity, Precision and Accuracy
7.1 Sensitivity. The sensitivity of the acidity test using
50 g of sample in 100 g of extraction solvent is 0.1 ppm.
Care must be taken in sample handling and in avoiding
cross contamination when perfcmning this test.
7 3 Precision. (To be determined)
7 3 Accuracy. (To be determined.)
Section 8. Special Apparatus and Reagents
NOTE: EQUIVALENTS MAY BE SUBSTITUTED.
1.
2.
3.
4.
5.
-6.
7.
8.
9.
10.
11.
Stainlessteel capillary tubing: Cat# 3021 1,
Alltech, Inc., M e l d , IL.
Toploading balance, lo00 g x .01 g: Cat#
258-464, CUmn Matheson Scientific, Broadview
Heights, OH.
100 ml stainless steel cylinder: Cat##
304L-HDF4-100, Swagelok, Inc., Solon, OH.
Two 1/4" stainless steel valves with male pipe
fittings: Cat# SS-1RM4, Swagelok, Inc., Solon,
OH.
Two 1/4" female pipe x 1/4" flare fittings: Cat#
2P-199, W. W. Grainger, Chicago, E.
1/16 Swagelok x 1/4" Swagelok reducing union:
Cat# SS400-61, Swagelok, Inc., Solon, OH.
1/4" Swagelok x 1/4" flare AN adaptor CaH
SS400-A4ANF, Swagelok, Inc., Solon, OH
1/4" x 1/4" flare connector.
1/4" inlet NPT male x 1/4" outlet NPT female
Hoke pressure relief vaive: Cat# 67 1 1UY. Hoke,
Inc., Cresskill, NJ.
250 ml Erlenmeyer flask: Cat# 121-954, Curtin
Matheson Scientific, Broadview Heights, OH.
Bromothymol blue sodium salt Cat# 423-247,
Curiin Matfieson Scientific, Broadview Heights,
OH.
1
APPENDIX C TO ARI STANDARD 700
. .
12.
13.
14.
15.
16.
17.
18.
19.
Specuophotomehic grade isopropanol: Cat#
831-155, Curtin Matheson Scienrific, Broadview
Heights, OH.
Spectrophotometric grade toluene: &I#
MTXO737-3, Curtin Matheson Scientific,
Broadview Heights, OH.
0.1N Potassium hydroxide in methanol: Cat#
RP-395-500, Curtin Matheson Scientific,
Broadview Heights, OH.
0.1N Sulfuric acid Cat# RS-640-500, Curtin
Matheson Scientific, Broadview Heights, OH.
Absolute methanol (Anhydrous, reagent-grade):
Cat## MMXO485-1, Curtin Matheson Scientific,
Broadview Heights, OH.
Stir platehtir bar Cat# 267-914, Curtin Matheson
Scientific, Broadview Heights, OH.
Glass distilled water.
Buret: Cat# 040-642, Curtin Matheson Scientific,
Broadview Heights, OH.
Section 9. Procedure
9.1 Srainless Steel Capillary Tubing Connector. Take
1/16" x .007" Teflon tubing and "swage" on a 1/16" nut
and ferrule. Connect this to a 1/16" x 1/4" Swagelok
reducing .union and then connect it to the 1/4" Swagelok x
1/4" flare adaptor. The 1/4" flare adaptor can then be
connected to the 1/4" flare fitting on the cylinder assembly
just before each acidity determimion.
9.2 Cylinder Assembly. The cylinder assembly is used as
the sampling appmtus for each acidity determination.
Before assembly, all pipe fittings must be Teflon taped to
ensure a proper seal at each joint.
Attach the pressure relief valve to the 100 ml stainless steel
cylinder. Atrach one of the 1/4" NFT x 1/4" NPT stainless
steel valves to the pressure relief valve. Connect a 1/4"
NPT x 1/4" flare fitling to the 1/4" NPT valve. To the
other side of the 100 ml cylinder, atwh another 1/4" NPT
x 1/4" NIT valve. Sampling should occur from the side of
the 100 ml cylinder which does not employ the pressure
relief valve.
93 Reagent Preparation. -
9.4 Sample Analysis.
a.
b.
C.
d.
-
.OlN KOH Solution. Pipet 100 ml of 0.1N KOH
solution into a lo00 ml volumemc flask Dilute to
the mark with absolute methanol and mix
thoroughly.
.OlN Sulfuric Acid Solution. Pipet 100 ml of O.1N
sulfuric acid solution into a lo00 ml volumemc
flask. Dilute to the mark with distilled water and
mix thoroughly.
Emaction Solvent. Add 495 ml of toluene to
495 ml of isopropanol. Add 10 ml of water to the
toluene/soppanol solution and mix thoroughly.
Bromothymol Blue Indicator Solution. Dissolve
1 g of bromothymol blue sodium salt in 100 ml of
methanol Mix thoroughly and store in a dropper
bottle.
a.
b.
C.
d.
e.
f.
Thoroughly clean the 100 ml stainless steel
cylinder, the valve, the capillary tube, the copper
connector and the 250 ml Erlenmeyer flask before
initiating testing. Oven heat all of the components
to 23OOF (1 10°C) and pull a vacuum.
Weigh the cylinder assembly to the nearest 0.01 g
and designate this weight as X.
Attach the 1/4" copper fitting to the vapor valve of
the sample cylinder and to the cylinder assembly.
Open the vapor valve; loosen the connector and
quickly tighten the fitting. This will purge the 1/4"
COMeCtM Of &.
Invert the sample cylinder with the attached
cylinder assembly. Open the sample cylinder
valve and then the cylinder assembly valve. Allow
the refrigerant to be inuoduced into the cylinder
assembly until 50 to 100 g of refrigerant is
sampled.
Close the cylinder assembly valve and set the
sample cylinder upright. Close the sample cylinder
valve; loosen the 1/4" connector and remove the
cylinder assembly.
Reweigh the cylinder assembly with the refigerant
and designate this value as Y. The weight of the
refrigerant is given by Y - X = grams of
refrigerant sampled. (Value for X is in step b
above.)
2
APPENDIX C TO ARI STANDARD 700 ~
Note: For the low pressure refiigerants, weigh
the sample cylinder before sampling and
then add the refrigerant directly to the
extraction solvent Reweigh the sample
cylinder and subtracthe initial weight
f b n the fmal weight to obtain the total
weight of the refrigerant sampled.
g. Add 100 mi of the prepared extraction solvent to
a 250 ml Erlenmeyer flask. Add a clean magnetic
stining bar. Add 6 drops of the indicator solution
to the extraction solvent and initiate moderate
stirring.
h. If the extraction solventiindicator solution is
yellow, add .01N pomssium hydroxide through the
buret until a just noticeable difference of green is
seen in the extraction solvent Half drops from the
buret may be necessary to achieve the real
endpoint.
i. If the extraction solvent/indicator solution is green
or blue, add .01N sulfirric acid through the buret
dropwise until the solution is yellow and then
proceed as in step h immediately above.
j. Attach the cleaned capillary connector to the
cylinder assembly containing the refrigerant sample
and slowly introduce the entire sample into the
extraction solvent/indicatm solution by gradually
opening the cylinder valve until it is fully open.
The cylindex assembly should be clamped to a ring
stand throughout the procedure.
k. After all of the refrigerant sample has been added
to the extraction solvent, if the color of the
solution is green or blue the result is reported as
"nondetect" If the solution is yellow, record the
buret volume to the nearest .01 ml (designate this
value as V, ) and add .01N KOH dropwise until
the green endpoint is reached. Record the final
buret volume to the nearest .01 ml (designate this
value as V, ).
Final volume of .01N KOH added:
v, = v, - v,.
1. Calculation of total acidity expressed in ppm as
HC1 is given by:
Section 10. References
1. Allied Signal Inc. Method GP GEN-2A.
2. ASTM Standrad D974-87.
3. Integral Sciences Inc. Method ISI-A401.
3
APPENDIX C TO ARi STANDARD 70Y
PART 2
DETERMINATION OF WATER IN NEW AND RECLAIMED
REFRIGERANTS BY KARL FISCHER COULOMETRIC TITRATION
Section 1. Purpose Section 6. Limitations and interferences
The purpose of this test method is to determine moisture in . None of the refrigerants tested interfere with the titration.
new and reclaimed refrigerants by the Karl Fischer Oxidizing agents such as MnO;, Cr20,-2, €I&, Fe (IJI), Cu
coulometric titration method. 0 and reducing agents such as S', thiosulphates and
Sn@) will interfere. Also, certain compounds such as
basic oxides and salts of weak acids (NaHC03, for example)
Section 2. Scope can fonn water with the KF reagent. None of these
interferences are nonnaUy present in new or reclaimed
This test method is for use with low, medium and high refrigerants.
pressure'rerefrigerants.
Section 7. Sensithrity, Precision .and Accuracy
7.1 Sensitivity. The sensitivity of the analyzer in this
Definitions for this pm are identical to those of ARI method using a 10 g sample is 1 ppm. E x m e care must
. Section 3. Definitions
Standards 700-95 and 740-95. be used in sample handling in order to achieve this
sensitivity.
Section 4. Principle 7.2 Precision and Accuracy. The average analysis (X),
standard deviation (s) and 95% confidence limits (95% CL)
Karl Fischer (KF) tihimetry is based upon the redox establishedfor the single operator precision of this method
reaction of water, iodine and sulfur dioxide: are shown in Table 1. '
&O+I,+SO, -+ 2HI+s4
The solvent is typically a mixture of methanol and a weak
organic base (imidazole, pyridine, etc.) with the base
serving to neutralize the reaction products. In coulometric
KF titrimelry, iodine is generated at the anode in direct
proportion to the amount of water introduced; and the
endpoint is detected biamperamemcally as the first
appearance of excess fke b. The added reiiigerant
eventually evaporates; hence, the solvent can be used
repeatedly until either the SO, or the base solution is
consumed.
The dam in Table 1 were calculated from 17 replicate
analyses of one sample (approximately 10 g) performed by
one analyst over a period of two days.
The- samples in Table 2 were tested for total percent
recovery. They were prepared by 'analyzing CFC-12 and
CFC-22 to 4.8 and 7.1 ppm, respectively, and then
contaminating the refrigerants with known amounts of
water. Both samples were then mixed for a period of
24 hours before analyzing. Res& are shown in Table 2.
The total percent recovery for each sample was 99.3% for
R-12 and 99.7% reco~ery for R-22.
Section 5. Applicability
This method is applicable to the routine quantitative
determination of small amounts of water in low, medium,
and high pressure refrigerants, including alternative
replacements for CFCs.
4
APPENDIX C TO ARI STANDARD 7oC
Section 8. Special Apparatus and Reagents
NOTE: EQUrVALENTs MAY BE SUBSTITUTED.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
KF coulomemc titrator Aquastar, EM Science
Model Cu)oo, Curtin Matheson Scientific,
Broadview Heights, OH.
75 ml stainless steel cylinder: Cat#
304L-HDF4-75, Swagelok, Inc., Solon, OH.
Two 1/4" stainless steel valves with male pipe
fittings: Cat# SS-lRM4, Swagelok, Inc., Solon,
OH.
Two 1/4" female pipe x 1/4" flare fittings: Cat#
2P-199, W. W. Grainger, Chicago, IL.
Generator solution: Aquastar Coulomat A, Curtin
Matheson Scientific, Broadview Heights, OH.
Cathode solution: Aquastar Coulomat C. Curtin
Matheson Scientific, Broadview Heights, OH.
Stainless steel capillary tubing: Cat# 30211,
Alltech, Inc., Deerfield, IL.
1/4" NPT x 1/4" FPT pressure relief valve: Cat#
67 1 IUYD, Hoke, Inc., Cresskill, NJ.
1/16" Swagelok x 1/4" Swagelok reducing union:
Cat# SS400-6-1, Swagelok, Inc., Solon, OH.
1/4" Swagelok x 1/4" flare AN adaptor Cat#
SS4OO-A4ANF, Swagelok, kc., Solon, OH.
Toploading balance, lo00 g x 0.01 g: Cat#
258-464, Curtin Matheson Scientific, Broadview
Heights, OH.
.Ring stand with clamps: Cat# 274-969 & 058-909,
Curtin Matheson Scientific, Broadview Heights,
OH
Heat gun: Cat# 4A162, W. W. Grainger, Chicago,
IL.
Nalgene desiccator: Cat# 324-277, Currin
Matheson Scientific, Broadview Heights, OH.
Syringe, 10 pk Cat# 222-950, Curtin Matheson
Scientific, Broadview Heights, OH
1/4" x 1/4" copper flare connector.
17. 10 ml gastight syringe with deflected point needle:
CaH282-021, Curtin Matheson Scientific,
Broadview Heights, OH.
18. Septum: Included as an accessory with the Karl
Fischer titrator.
Section 9. Procedure
9.1 Cylinder Assembly. The cylinder assembly is used as
the sampling apparatus €or &e determination of water in
medium and high pressure refrigerants. Before assembling,
aJl pipe fittings must be stainless steel taped to ensure a
proper seal at each joint.
Attach the pressure relief valve to the 75 ml stainless steel
cylinder. Attach one of the 1/4" W x 1/4" NPT stainless
steel valves to the pressure relief valve. Connect a 1/4"
FPT x 1/4" flare fitting to the 1/4" NPT valve. To the
other side of the 75 ml cylinder, attach mother 1/4" NPT
x 1/4" NFT valve, and the 114" FPT x 1/4" flare fitting to
the 1/4" NPT valve. Sampling should always occur from
the side of the 75 ml cylinder which does not employ the
pressure relief valve.
9.2 Stainless Steel Capillary Tubing Connector. Take
1/16" x .007" stahless steel tubing and "swage" on a 1/16"
nut and ferrule. Connect this to .a 1/16" x 1/4" Swagelok
reducing union and then connect it to the 1/4" Swagelok x
1/4" flare adaptor. The 1/4" flare adaptor can then be
connected to the 114" flare fitring on the cylinder assembly
just before each moisture determination.
93 Sunrple Analysis of Medium and High Pressure
Rqrigeronts.
- a. Place the cylinder assembly, capillary tube and the
1/4" copper flare connector in an oven for one
hour at 2upF (1 1OOC).
b. Remove the cylinder assembly from the oven and
pull a vacuum on the entire assembly, while hot,
down to 30 microns (3.99 kpa).
c. Close the valve and cap the flare fitting. Mow
the cylinder to cool until it reaches room
temv in the desiccator.
d. Weigh the cylinder assembly to the nearest 0.01 g.
Designate this value as X.
e. With a heat gun, heat the flare fitting on the valve
of the cylinder assembly and the flare fitting on
5
aPPENDlX C TO ARI STANDARD 700
f.
gh.
' 1.
j.
k.
1.
m.
n.
0.
the sample cylinder for 30 seconds. This will
remove residual moisture.
Remove the 1/4" copper flare connector from the
oven, attach it immediately to the vapor side of the
sample cylinder and connect it to the cylinder
assembly. The cylinder assembly must be inverted
in relation to the sample cylinder.
Purge a small amount of the vapor phase
refigerant through the 1/4" copper flare connector
to remove air fYom the fitting. Tighten the fitting.
Allow the 1/4" copper flare fitting to cool until it
reaches mom temperature.
Invert the sample cylinder with the cylinder
assembly anached.
Open the valve of the cylinder assembly and
introduce 5 to 30 g of liquid ref'rigerant into the
Cylinder assembly. Close the valve to the cylinder
assembly. The greater the suspected water value,
the smaller the amount of sample necessary. For
detenninatons of water in new and reclaimed
refrigerants, a sample weight of 10 g is acceptable.
With the vapor side of the sample cylinda open
and the cylinder assembly valve closed, set the
sample cylinder upright, allowing liquid refrigerant
in the 114" copper connector to drain into the
sample cylinder. Heat the flare connection.
Close the vapor side valve of the sample cylinder
and disconnect the 1/4" copper connector from
both the sample cylinder and the cylinder
assembly.
Weigh the cylinder assembly (to the nearest
0.01 g) and the refrigerant sampled. Designate this
value as Y.
Determine the refrigerant sample weight by
Y - X = grams of refrigerant sampled. (Value for
X is in step dabove.)
Remove the capiuary tube h m the oven and
connect it to the cylinder assembly. Install the
capillary tube through a new septum to the Karl
Fischex tintor so that the end of the capiuary
tubing is level with the membrane of the cathode.
(See the Aquastar opemional manual for the
location of the cathode membrane.)
P.
9.
r.
S.
t.
U.
Attach the capillary tube fitting to the 1/4" flare of s
the cylinder assembly. The cylinder assembly
should be upright and secured to the ring stand.
Introduction of the sample is ready when the
Aquastar background moisture is stable (i.e., the
"S" on the liquid crystal display is blinking and the
background is between 0 to 50 pg HzO). The
detector pilot lamp should also be yellow.
Depress "Sample" on the EM Science Aquastar
C2000. Immediately open the valve of the
cylinder assembly and allow the vapor phase
refiigerant to pass into the Karl Fischer reagent.
The flow will be very slow due to the resmction
which occurs in the capillary tubing. Slight
chilling will occur in the cylinder assembly. Use
a heat gun to keep the cylinder tempemure
between 70OF and 90°F (21OC and 32OC)
throughout the test. DO NOT PROVIDE
CONTINUOUS HEAT TO THE STAINLESS
STEEL CYLINDER. HIGH PRESSURE COULD
RESULT IN SERIOUS INJURY.
When essentially all of the liquid refrigerant has
passed into the Karl Fischex reagent (when bubbles
begin to slow down considdly), use the heat gun
to raise the cylinder assembly temperature to
2509 (121OC) so as to drive off all moisture from
the walls of the cylinder assembly.
Once all of the sample has been introduced into
the titrator (as evidenced by vapor flow subsiding
in the tinator), close the valve, depress "Titration"
and allow the Karl Fischer titrator to report the
value of micrograms of water present.
When this value is reported, divide it ' by the
weight of the refrigerant in order to obtain the ppm
value for the sample.
Note: For high press= refrigerants (R-13, R-23
and R-503), cool the cylinder in an ice
water bath at 40°F (4.4'33 for 30 minute3
before beginning this test
6
APPENDIX C TO ARI STANDARD 7W
9.4
a.
b.
C.
d.
e.
f.
g.
Sample Analysis of Low Pressure Refrigerants.
Clean and dry the 10 ml syringe for use in the
moisture determination. When the syringe and
needle are not in use, they should be stored in a
desiccator.
Prepare the Karl Fischer analyzer as prescribed by
the device -on Manual.
Sample 5-8 ml of the low pressure refrigerant and
immediately attach a septum to the syringe needle
so as to cover the opening on the deflected point
needle. (See Note below.)
Weigh the syringe (with the attached septum) to
the nearest .01 grams. Designate this value as X.
Depress "Sample" on the KF titrator.
Remove the septum from the needle, and
immediately insert the needle through the sampling
port septum of the Karl Fischer vessel. Inject the
sample. The titration will begin automatically.
Reattach the septum cap to the syringe needle and
weigh to the nearest -01 grams. Designate this
value Y. From step 4 above, X - Y = grams of
refrigerant sampled.
h. The concentration of water, if not calculated by the
KF t i m r , is given as follows:
micrograms H20
grams refnserant sampled
ppm H20 =
NOTE: Low pressure refrigerant samples normally
arrive at the laboratory in many different
types of containers. Because of this, the
sampling apparatus for each type of
container is different and the details of
each are not provided in this method.
Whatever sampling apparams is chosen, it
must provide a hermetic seal between the
sample and atmospheric humidity to avoid
any water contamination during sampling.
Section 10. References
1. National Refrigerants Inc. Method NR300.0.
2. . DuPont Chemicals Dept. Method F3200.085.01LV.
3. EM Science Moisture Analyzer Instruction Manual.
4. Integral Sciences Inc. Method ISIW605.
APPENDIX C TO ARI STANDARD 700
I 1
Table 1. Single Operator Method Precision
Average 95 % Standard
Value Confidence Deviation
0
.ll 10.6 Water (R-12), ppm by wt.
Limit (SI
0.26
Water (R-22). p ~ m by wt. 0.77 .29 28.1
1 Table 2. Testing for Percent Water Recovery 1
original Percent Recovered Calculated Amount
Value Recovery Total Total Contaminated
+
Water (R-12), ppm by wt.
99.7 38.1 382 19.0 192 Water (R-22), ppm by wt.
99.3 13.6 13.7 6.9 6.8
8
APPENDIX C TO ARi STANDARD 700
PART 3
DETERMINATION OF HIGH BOILING RESIDUE
IN NEW AND RECLAIMED REFRIGERANTS BY
VOLUMETRIC AND/OR GRAVIMETRIC MEASUREMENT AND
DETERMINATION OF PARTICULATE RESIDUE BY VISUAL INDICATION
Section 1. Purpose
The purpose of this test method is to determine high boiling
residue and visible particulates in new and reclaimed
refiigerants.
Section 2. Scope
This test method is for use with low, medium and high
pressure refiigerants.
Section 3. Definitions
Definitions for this part are identid to those of ARJ
Standards 700-95 and 740-95.
Section 4. Principle
High boiling residue (HBR), also called non-volatile
residue, is determined by evaporating a hown amount of
refrigerant in a Goetz bulb at an ambient or elevated
temperature and then visually measuring or weighing the
remaining residue. If greater than specification volume is
observed, the bulb is placed in a 140OF (6OOC) oven for
30 minutes and, after cooling, the volume of residue is
again measured. For gravimetric determination, the residue
is redissolved in high purity R-11 or R-141b and
quantitatively transferred into a small tared aluminum pan.
The solvent is removed by evaporation and the pan
reweighed to obtain the weight of residue.
Prior to evaporation, the measured volume of liquid
refiigerant is visually examined for the presence of
insolubles such as packing fibers, rust dirt, etc. The
residue Erom high pressure samples is redissolved in a clean
solvent, swirled, and then visually examined for any
insoluble particulates.
Section 5. Applicability'
This method is applicable to the routine quantitative
determination of HBR and visible evidence of particulates
in all low, medium and high pressure new and reclaimed
refrigerants, including alternative replacements for CFCs.
The method was developed to measure HBR and
particulates in compliance with AlU Standard 700
specifications for HBR and particulates.
Section 6. Limitations and Interferences
In orda to achieve the statistical parameters stated for this
method, at least 100 ml of refrigerant sample is required.
There are no known interferences to this method.
Section 7. Sensitivity, Precision and Accuracy
7.1 Sensitivity. Based upon a 100 ml volume of sample,
the method will detect 0.01 ml of HBR which is the first
mark on the Goea bulb buret. This 0.01% value is the
ARI Standard 700 specification for most refiigerants. The
detection limit by weight is generally 4.01% due to the
sensitivity of the analytical balance and because 0.01 ml of
residue (usually oil) weighs 4.01 g. Also, except for very
high pressure refrigerants, the weight of 100 ml of liquid
refrigerant weighs >lo0 g.
7.2 Precision. The precision for the HBR determination
at 0.03 volume percent was found to be +0.005 at the 95%
confidence limit. This was based upon an analysis of R-1 1
by two analysts and using silicone oil as the residue.
73 Accuracy. The relative mean error at the 0.03%
volume level was found to be 3.3%.
NOTE: THESE STATISTICAL PARAMETERS ARE
NOT APPLICABLE TO VISUAL
OBSERVATIONS OF PARTICULATES.
9
APPENDIX C TO ARI STANDARD 700 .
Section 8. Special Apparatus and Reagents
NOTE. EQUIVALENTS MAY BE SUBSTITUTED.
1. Goetz graduated centrifuge tube: 100 ml, cat#
21123-003, VWR Scientific, Bridgeport, NJ.
2. Boileezers, carborundum crystals: Cat# C-189,
Fisher Scientific Company, King of Prussia, PA.
3. Disposable aluminum dish: Cat# 25433-090.
VWR Scientific, Bridgeport, NJ.
Section 9. Procedure
9.1 Calibration. For the HBR procedure, a caLibrasion
solution of 0.03% by weight of silicone oil in R-1 may be
prepared by weighing 0.220 g of silicone oil and dissolving
in 500 ml(738 g) of high purity R-11, mixing thoroughly,
labeling and storing in a screw-capped glass bottle in a
refrigerator. Alternatively, weigh 0.187 g of oil and
dissolve in 500 rnl of R-141b solvent.
9.2 Sample Analysis, HBR Volume Percent Measurement
and Particulates.
a. Measure 100 ml of refiigerant sample into the
Goetz bulb as follows:
1. For low pressure refrigerants (R-11, R-113,
R-123): add 100 ml of liquid refiigerant l3om
a glass graduate into the Gcetz bulb.
Alternatively, add liquid refiigerant hm the
sample container to the 100 ml mark of the
Goetz bulb.
2. While firmly holding the bulb, gently swirl
the sample solution and then position the bulb
in front of a source light, window, etc. and
visually examine for the presence of
paniculate mater. Record as "pass" if not
particulates are observed. Proceed to step
9.2b.
3. For medium and high pressure refrigerants:
tare the sample cylinder (to the nearest 0.1 8).
invert the cylinder and, by positioning the
valve opening just inside the neck of the
Goetz bulb, carefully open the valve and
allow the liquid phase to discharge inside the
bulb. Except for very high pressure
refrigerants (R-503, for example), liquid
Continue to add sample until 60-75 of liquid
reiiigerant will begin to accumulate.
has been collected. Tum off the sample
valve. Reweigh the sample cylinder and
record the difference as the weight of sample
added.
4. Repeat step 9.2a2 above. Use either isopropyl
alcohol, "thumb," or paper towel to wipe frost
off the outside of the bulb (create a window)
to facilitate the visual observation.
5. For very high refrigerants (R-503,
R-13, R-23), the sample cylinder is pre-cooled
to 4OoF (5'C) in ice water (especially on hot
days) before carefully flashing the liquid
phase into the Goetz bulb. Continue to add
liquid phase until the sample cylinder
weigh-back shows that between 100 and
130 g of refiigerant has been flashed into the
bulb. At this point, little or no liquid phase
refiigerant will have accumulated in the bulb.
Record this weight as the g of sample added.
Add 100 ml of either high purity R-1 1 or
R-141b to the bulb, stopper, swirl to dissolve
any residue on the inner walls of the bulb,
remove the stopper.
Note: DO NOT allow any stopcock grease
to be present on the glass stopper or
on the neck of the bulb.
6. Repeat step 9.2a2 above.
b. Add one small boiileezer, and then place the Goeb
bulb in a 115OF (45OC) constant temperahue bath
(140OF (60°C) for R-113). Position the bulb such
that it is immersed in the bath to about the 20-25
- ml mark. Do not remove the bulb from the bath
until all of the refiigerant has completely
evaporated (detennined by observing the
disappearance of refiigerant condensation around
the neck of the bulb).
c. Remove the Goek bulb from the bath, wipe the
outside dry and visually measure the ml residue (if
any) at the bottom of the buret (ignm the
boileezer). Measure to the nearest 0.005 ml.
d. If the observed residue is SO.01 ml, proceed to the
calculation section below. If the observed residue
is M.01 ml, proceed to step e.
e. Place the Goep bulb upright in a 14OOF (WC)
oven for 30 minutes, remove, cool, then measure
and record the volume of residue (to the nearest
0.005 ml) in the buret as above. Far weight
10
. .
percent measurement, save the Goetz bulbresidue
for 9.3.
HBR V01.96 = A x 100
B
Where:
A = volume of residue (ml) in buret.
B = ml of sample added to bulb (step 923 above).
Note: To determine the ml of high and very high
pressure refrigerant samples, the weight of
the sample is divided by the liquid density
of the refrigerant at the ambient sample
temperature (see Table 1).
Report all results to the nearest 0.01% vol. If
results are 4.01%, report as "4.01% vol."
93 Sample Analysis, Weight Percent Measurement
a. Prepare an aluminum pan by rinsing it in acetone
and placing it in a 1WF (60'C) oven for at least
30 minutes. Remove (use tweezers) and place in
a desiccator until cool (usually 15 to 20 minutes).
b. Using tweezers, remove the pan from the
desiccator and determine the rare weight (to the
nearest O.OOO1 g).
c. Add 20 ml of either high purity R-11 or R-l4lb to
' the Goetz bulb saved from 9 2 . Stopper the bulb
and shake to redissolve the residue and/or to
resuspend the PQmcuIates (if present) in the
solvent.
d. Carefully pour the solution from the G o e ~ bulb
into the pan. Use two approximately 8 ml portions
of the solvent to effect a quantimtive transfer. Do
not permit the boileezer to fall into the aluminum
pan, but if that should occur, carefully remove
using metal tweezers. ,
APPENDIX C TO ARI STANDARD 700
e. Carefully place the aluminum pan inside a hood
and allow the R-1 1 or R-141b solvent to evaporate
(alternatively, the pan may be placed on the hot
water bath).
f. Place the pan in the 14OOF (6OOC) oven for
30 minutes, remove and place in the desiccator
until cool (20 to 30 minutes).
g. Using tweezers, remove the pan, reweigh and
record the difference in weight (from step b) as the
weight of residue.
Calculation
HBR Wt.96 = A x 100
B
Where:
A = g of residue from 9.3g above.
B = g of sample taken from 9.21.
Note: To determine the g of a low pressure
sample Rfiigerant (R-11. R-113, R-123,
etc.), multiply the volume taken times the
density (see Table 1).
Report results to the nearest 0.01% wt. If results
are <0.01% WL, report as "4.01% wt."
Section 10. References
1. DuPont Chemicals Dept Method F3200.037.01 CW
(P).
2. -National Refrigerants Inc. Method NR 305.0.
3. National Refrigerants Inc. Method NR 398.0.
11
,APPENDIX C TO ARI STANDARD 700
Table 1.
Densities of Liquid Refrigerants
Refrigerant
11
12
13
13B 1
22
23
32
113
114
115
123
124
125
134a
14 lb
143a
15%
40 1A
40 1B
402A
402B
403B
404A
405A
407A
407B
407C
408A
409A
410A
500
502
503
507
Density (g/ml), at 77°F (25°C)
1.476
1.311
0.907
1.538
1.194
0.670
1.095
1.565
1.456
1.29 1
1.468
1.364
1250 at 68OF (20OC)
1210
1.244
0.946
0.91 1
1.188
1.188
1.151
1.156
1.150
1.048
1.173
1.142
1.166
1.134
1.062
. 1.223
1.03 1
1.168
1.217
0.795
1.170
12
APPENDIX C TO ARI STANDARD 700
PART 4
DETERMINATION OF CHLORIDE IN
NEW AND RECLAIMED REFRIGERANTS BY
SILVER CHLORIDE PRECIPITATION
Section 1. Purpose
The purpose of this test method is to qualitatively
determine the presence of chloride in new and reclaimed
refiigerants.
. Section 2. Scope
This test method is for use with low, medium and high
pressure refiigerants.
Section 3. Definitions
Definitions for this part are identical to those of ARI
Standards 700-95 and 740-95.
Section 4. Principle
The qualitative determination of chloride in refrigerants is
based on precipitation of the anion as silver chloride:
Ag' + C l + Ag ChS)
The refrigerant is added to a solution of silver nime in
methanol. Visual turbidity indicates the presence of
chloride and is reported as "fail." If no turbidity is
observed, chloride is within acceptable limits and reported
as "pass."
Section 5. Applicability
This method is applicable to the routine qualitative
determination of chloride in high, medium and low pressure
refiigerants, including alternative replacements for CFCs.
Section 6. Limitations and Interferences
None of the refrigerants tested interfere with the chloride
determination. Anions of weak acids can'be an interference
in the determination, but these interferences are not
normally present in new or rechimed refrigerants. Samples
containing insoluble lubricants and oils may show a visual
haze or slight turbidity, however, such levels of lubricant or
oil necessary to show such visual turbidity are not normally
present in new or reclaimed refrigerants.
Section 7. Sensithrity, Precisbn and Accuracy
7.1 Sensitivity. The sensitivity of the chloride turbidity
test using 5 ml of sample in 5 ml of methanol containing
three drops of sahxated AgNQ is approximately 3 ppm.
Care must be taken in sample handling to avoid cross
contamination when performing this test.
7.2 Precision. (To be determined.)
73 Accuracy. (To be determined.)
Section 8. Special Apparatus and Reagents
NOTE: EQurvALENTs MAY BE SUBSTlTUTED.
1.
2.
3.
4.
5.
6.
Stainless steel capillary tubing: Cat# 30211,
AUtech, Inc., Deerfield, IL.
Toploading balance, 10o0 g x 0.01 g: Car#
258-464, Curtin Matheson Scientific, Broadview
Heights, OH.
Methanol anhydrous reagent: Car# 563-189,
Cllmn Matheson Scientific, Broadview Heights,
OH.
Silver nitrate: Cat# 83 1-916, Curtin Matheson
Scientific, Broadview Heights, OH.
75 ml stainless steel cylinder Cat#
304L-HDF4-75, Swagelok, Inc., Solon, OH.
Two 1/4" stainless steel valves wirh male pipe
fiaings: Cat# SS-lRM4, Swagelok, Inc., Solon,
OH.
13
APPENDIX C TO ARI STANDARD 700
7. Two 1/4" female pipe x 1/4" flare fittings: Cat#
2P-199, W. W. Grainger, Chicago, IL.
8. 1/16" Swagelok x 1/4" Swagelok reducing union:
Cat# SS400-&1, Swagelok, Inc., Solon, OH.
9. 1/4" Swagelok x 1/4" flare AN adaptor Cat#
SS4OO-A4ANF, Swagelok, Inc., Solon, OH.
10. 1/4" x 1/4" copper flare connector.
11. 1/4" inlet NPT male x 1/4" outlet NPT female
pressure relief valve: Cat# 671 1L4Y. Hoke Inc.,
Cresskill, NJ.
Section 9. Procedure
9.1 Stainless Steel Capillary Tubing Connector. Take
1/16" x .007" stainless steel tubing and "swage" on a 1/16"
nut and fermle. Connect this to a 1/16" x 1/4" Swagelok
reducing union and then connect it to the 1/4" Swagelok x
1/4" flare adaptor. The 1/4" adaptor can then be
co~ected to the 1/4" flare fitting on the cylinder assembly
just before each chloride determination.
93 Cylinder AssembZy. The cylinder assembly is used as
the sampling apparatus for chloride determination of
medium and high pressure refrigerants. In order to
complete this assembly, all pipe fittings must be Teflon
taped to ensure a proper seal at each joint.
Attach the pressure relief valve to the 75 ml stainless steel
cylinder: Attach one of the 1/4" NIT x 1/4" stainless
steel valves to the pressure relief valve. Connect a 1/4"
NPT x 1/4" flare fitting to the 1/4" N€T valve. To the
other side of the 75 ml cylinder, attach another 1/4" NPT
x 1/4" NPT valve. Sampling should always occur from the
side of the 75 ml cylinder which does not employ the
pressure relief valve.
93 Sample Analysis.
a. Thoroughly clean the 75 ml stainless steel cylinder,
the valve, the capillary tube; the copper connectof
and the 100 ml beaker before initiating testing.
Heat all of the components to 23OOF (1 10°C) and
pull a vacuum.
b. Weigh the cylinder assembly to the nearest 0.1 g
and designate this weight as X.
c. Attach the 1/4" copper fitting to the gas valve of
the sample cylinder and to the cylinder assembly.
Loosen the connector and quickly tighten the
fitting.
d. Invert the sample cylinder with the attached
cylinder assembly., Open the sample cylinder
valve and then the cylinder assembly valve. Allow
the refrigerant to be introduced into the cylinder
assembly until 30 to 40 g of refrigerant have been
sampled.
Note: For very high pressure refrigerants (R-13,
R-23, R-503), it is necessary to precool the
sample cylinder and the cylinder assembly
to 40OF (4OC) in order to provide sufficient
liquid phase sample for this test.
e. Close the cylinder assembly valve and set the
sample cylinder upright. Close the sample cylinder
valve, loosen the 1/4" connector and remove the
cylinder assembly.
f. Reweigh the cylinder assembly with the refrigerant
and designate this value as Y. The weight of the
refrigerant is given by Y - X= grams of
r e f i g a t sampled. (Value for X is in step b
above.)
g. Calculate the volume of refrigerant sampled by:
volume = gram sampled
&nsity *
* The values of the densities for each refrigerant
can be found in Table 1 of Pan 3,
"Determination of High Boiling Residue in New
and Reclaimed Refiigerants By Volumetric
and/or Gravimetric Measurement and
Determination of Particulate Residue by Visual
Indication."
h. Add the same volume of methanol as the volume
ofrefiigerantfoundinstepgtoa100mlbeaker.
For each 5 ml of methanol used, add three drops
of saturated silver nitrate solution to the methanol.
Also, add one dmp of concentrated Nhic acid to
the solution before adding the re€iigerant sample.
Note: This Cr test is valid only if the sample
solution being tested is on the acid side.
This prevents: Ag' + OH- + Ag(OH),,) +
Ag,O,, if the sample pH > 7.
i. Attach the cleaned capillary connector to the
cylinder assembly containing the refiigerant sample
14
APPENDIX C TO ARI STANDARD 700
and slowly introduce the entire sample into the Section 10. References
methanolic silver nitrate.
1. General Services Administration Federal Specification
j. If turbidity is present, the test is reported as "fail." BB-F-1421B.
If no turbidity exists, the test is reported as "pass."
Note: For low pressure refrigerants, pour
approximately 25 ml of the refrigerant into
a 100 ml beaker and proceed as in step h.
After adding the methanol and saturated
silver nitrate solution, stir the mixture for
30 seconds. If any turbidity is present in
the methanol layer, the result is reported as
"fail."
2. Integral Sciences Inc. Method Number IS1 C610.
15
APPENDM C TO ARI STANDARD 700
PART 5 -
DETERMINATION OF NON-CONDENSABLE GAS
IN NEW AND RECLAIMED
REFRIGERANTS BY GAS CHROMATOGRAPHY
Section 1. Purpose
The purpose of this test method is to determine
non-condensable gas levels in new and reclaimed
refrigerants using gas chromatography.
Section 2. Scope
This test method is for use with medium, high and very
high pressure rehigerants.
Section 6. Limitations and Interferences
None of the refrigerants interfere with the determination as
all chromatographically elute after the air peak Methane
elutes about 0.10 minutes after the air peak and, if present
in amounts >0.10% by volume begins to slightly interfere.
However, the amounts of methane (formed during
compressor bum-out) in reclaimed re&gerants normally
ranges from 0 to 50 ppm by weight and does not interfere
at these levels.
Section 7. Sensitivity, Precision and Accuracy
Section 3. Definitions
Definitions for this part are identical to
Standards 700-95 and 740-95.
those of ART
Section 4. Principle
Noncondensable gas (NCG) is measured in the vapor
space above the refkigerant liquid phase by isothermal gas
Chromatography using a thermal conductivity detector
(TCD) and an external standard calibration. By definition,
NCG includes gases such as oxygen and nitrogen (air),
carbon dioxide, argon and carbon monoxide. However, in
the typical refkigerant sample, air is the only NCG present
in significant amounts and the other gases are not routinely
analyzed. Very high pressure refiigerants (R-13, R-23,
R-503) often contain no liquid phase and these are analyzed
directly. NCG equilibrium between refkigerant liquid and
vapor phases 'is temperature dependent and appropriate
sample tern- comctions are applied to report results
at the 75OF (23.9OC) specification temperature.
Section 5. Applicability
This method is applicable to the routine quantitative
analysis of NCG in medium, high and very high pressure
refiigerants, including the alternative replacements for
CFCs. ?he method was develw to measure
non-condensable gas as required by ARI Standard 700.
7.1 Sensitivity. The method will detect about 0.02% by
volume NCG in any of the ARI Standard 700 listed
refrigerants.
7.2 Precision. The precision was determined at 5.2% by
volume concentration and was found to be f 0.07% by vol.
at the 95% confidence limit (CL.). This was based upon
12 repetitive analyses of an R-12 sample by two
technicians over a two-day period.
73 Accuracy. A 5.1% by volume certified calibration
standard (air in helium), purchased from Scoa Specialties,
was-analyzed 9 times following the Wtial calibration during
a one-day period by one technician. The relative mean
e m was 1.63%.
Section 8. Special Apparatus and Reagents
NOTE:
1.
2.
3.
EQWALEN'IS MAY BE SUBS'ITTUTED.
Gas chromatograph Model SRI 8610, equipped
with a manual sample injection valve, 1-ml sample
loop and TCD, Buck Scientific Inc., East Norwalk,
CT.
GC column: 6' x 1/8" Parapack Q, 80-100 mesh,
Supelco, Bellefonte, PA.
Electronic integrator: Model SP4290,
Spectra-Physics, San Jose, CA.
16
APPENDIX C TO ARI STANDARD 700
4. Calibration standard: 1.5% by volume, air in
Helium. 30 lb. cylinder, Scott Specialty Gases,
Plumsteadville, PA.
5. Digital thermometer Cat# 61220-670, VWR
Scientific, Bridgeport, NJ.
6. Temperature probe: YSI 409A, Cat# 61254-002,
VWR Scientific, Bridgeport, NJ.
Section 9. Procedure
9.1 Calibration.
a.
b.
C.
d.
e.
f.
g-
Refer to the operaling manual for the SRI 8610 to
gain familiarity with the gas chromatograph (GC).
Anach a 20-inch (51 cm) section of 1/4" inside
diameter flex line (or Tygon tubing) to the GC
sample inlet line (need 1/8" Swagelok to 1/4" flare
adaptor) and terminate the other end with a 1/4"
female flare connector.
Attach a short piece of Tygon tubing to the GC
sample exit line and terminate the tubing by
placing it inside a small beaker of water.
Connect the sample inlet line to the valve of the
1.5% NCG calibration standard cylinder.
Set the GC and electronic integrator as follows:
TC Bridge Current low sensitivity
Carrier Gas Flow 3Occ H4min.
Attenuator x1
Detector Temp. lOOT
Inj. Port Temp. lOOT
Head Pressure mpsi
Column Temp. lOOT
Sampling Valve Load Position
Integrator Ext Std. Method, 96 vol.,
Atten. x8, (Refer to the
SP4290 Operating
Manual).
Slowly open the standard cylinder valve, and
slowly purge the sample vapor through the sample
loop as indicated by bubbles in the exit line beaker
of water. Purge for about 10 seconds so as to
expel air h m the system. One IO-second purge
should be equivalent to about 10 ml of vapor.
Close the cylinder valve and, when the bubbling
stops, immediately rotate the sampling valve to the
h.
1.
j.
k.
"Inject" position and immediately press the "Run"
switch on the integrator.
After the air peak has eluted (about 0.4 minutes),
r e m the sampling valve to the "Load" position
and terminate the integration.
Repeat steps f through h until three consecutive
analyses yield essentially reproducible peak areas
for the air peak.
Calculate the air peak Absolute Response Factor
(ARF) for each of the three analyses as follows:
ARF = 4
96 by VoE. Air in Cali Std.
Where:
q=areaofairpeak
typical ARF = 12,000.
Average the three ARF values and assign the
average value as the ARF for the method. The
three ARF values should agree within about 1.6%
Relative Error.
Note: The calibration standard should be
analyzed at least on a daily basis and the
ARF updated as necessary.
9.2 Sample Analysis.
a.
b.
C.
d.
e.
Record the temperature &l0F (*X) of the sample
source liquid phase when the vapor phase is taken
for analysis. If this information is &own
(customer samples, for example), record as 75OF
(23.9OC) (see Notes 2 and 3 in Section 10).
Connect the sample inlet line to the sample
cylinder valve which directly accesses the sample
vaporphase.
Set the GC and integrator as described in the
calibration section (9.1).
Slowly open the sample cylinder valve and slowly
purge vapor (about 10 seconds) to expel air from
the sample loop and lines (refer to Note 5 in
Section 10 for samples with c500 ml total
headspace vapor).
Close the valve and, when the bubbling stops in
the exit line beaker of water, immediately rotate
17
APPENDIX C TO ARI STANDARD 700
the sampling valve to the "Inject" position and
immediately press the "Run" switch on the
integrator.
f. Continue the chromatographic separation until the
large refrigerant peak returns to the original
baseline (Figure 2). Stop the integration.
g. Repeat steps d through f until the air peak area is
reproducible (i.e., until all system air has been
expelled). This may quire two or three
additional consecutive determinations.
93 Calculation.
a. The integratm will print-outhe result for air
(NCG) in % by volume which represents the
temperature at which the sample was taken for
analysis:
% by VOL NCG = - ARF
4
b. Correct the result to % NCG at 75OF (23.9OC) as
below. Use the Vapor Ressure-Temperature
graphs in References 5 and 6 (also shown in
Figure 3). For R-403B, use the curve for R-125.
Cz x P2 x Ki c1 = T
'2
See Table 1.
Where:
C, = NCG, 9% by vol., at 75OF (239OC).
C, = NCG, 76 by vol., at the sampling temperame.
P, = Vapor pressure (psis) of the refrigerant at the
sampling tempemure, T,, in OF(OC). This value
is determined from Figure 3 (also from the cham
in References 5 and 6).
Rankine (i.e., T, = OF + 459.67 ("C + 273.15)).
75OF (23.9OC) (see Table 1).
T, = Sampling tern- in degrees absolute
IC, = Temperame/pssure ratio for refrigerant i at
c. Report results to the nearest 0.01 % by vol. If
results are 4.02 % by vol., report as "4.02 % by
VOL"
Section 10. Notes
1. When analyzing cylinders containing both liquid and
vapor phases, it is especially importanthat, when
purging air from the chromatographic system (9.2d), a
too-rapid purge is not used. A too-rapid purge may
cause some liquid refiigeranto "bump," and such
droplets may evaporate resulting in a
too-rich-in-refrigerant vapor purge. This could result in
NCG values which are too low.
-
2. To reestablish equilibrium in a liquid/vapor phase
sample cylinder brought into the laboratory and which
has changed temperature to a significant degree from
the original temperature (standing several hours, for
example), the cylinder must be rolled (to mix) for
several minutes before sampling the vapor phase for
GC analysis. The outer wall temperature of the
cylinder below the liquid level should be nearly
equivalent to that of the refiigerant contents and can be
measured using a suitable thermocouple probe.
3. If the vapor phase of a storage tank, road tanker. ton
cylinder, e&., is sampled into a small evacuated
cylinder, regardless of what temperature the small
sample cylinder vapor may be when analyzed by GC,
the contents will represent the vapor temperature at the
original sample location point.
4. In all liquid/vapor phase refrigerants, the NCG
concentration in the vapor phase increases with
decreasing ternpeame of the liquid phase. This is
because the vapor concentration of the refrigerant
decreases more so than that of air as the temperature
drops.
5. For samples containing very small total headspace
vapor (<500 ml), the sampling lines, loop, 'etc., are
evacuated to 1 mm of Hg (0.1333 kPa) to the sample
cylinder valve. The vacuum line is then closed and the
system brought to the desired pressure (usually 1 am.)
by slowly opening the sample cylinder and metering
valves and then injecting into the GC as described. In
this way, less total volume of headspace vapor is
consumed compared to the purging method. See
Figure 4.
Section 11. References
1. Spectra-Physics, SP4290 Computing Integrator
Operators Manual, PN AOO99-158, 1984.
2. SRI Instruments 8610 Gas Chmatograph Users
Manual, 1990.
18
APPENDIX C TO ARI STANDARD 700
3. DuPont Chemicals Dept Method LL-F-1266.
4. Downing, R. C., Fluorocarbons Refrigerants Handbook,
Prentice Hall, 1988.
5. DuPont Chemicals Dept. "Freon" products Laboratory
Information Bulletin G-1, 1984.
6. DuPont Chemicals Dept. "Freon" Products Laboratory
Information Bulletin AG-1, 1991.
7. National Refrigerants Inc. Method NR 280.0.
19
APPENDIX C TO ARI STANDARD 700
Part
"A" injection End Connection (1 " up from bottom).
"B" Flexible 1 /I 6" S/S line to detector.
"C" Column Exit end connection,
Figure 1. Oven Column Design for a SRi 8610 Gas Chromatograph
20
APPENDIX C TO ARI STANDARD 700
1
Figure 2. Gas Chromatogram of Non-Condensable Gas (Air) In R-12
21
APPENDIX C TO ARI STANDARD 700
,
22
n u ' u r n u w m w
Figure 3. Vapor Pressurmemprature Correlations for Refrigerants
. . ._
. . APPENDIX C TO ARI STANDARD 700
Vacuum/Pressure
GC Sampling Gauge
Valve
U Vacuum
A = Sample Cylinder Valve
B = Metering Valve
C = Vacuum Pump Valve
Figure 4. Evacuated System Method of Introducing Vapor Sample into Gas Chromatograph
23
APPENDIX C TO ARI STANDARD 700 .
Table 1. & Values for Refrigerants
at 75°F (23.9”C)
Refrigerant
R- 12
R-13
R-114
R-124
R- 125
R-22
R- 134a
R-115
R-14%
R-403
R-500
R-502
R-13B1
R- 15h
K, (“Npsia)
5.941
1.08
. 17.94
9.90
2.785
3.687
5.456
4.227
11.256
2.78
5.00
3.34
2.397
6.855
' APPENDIX C TO ARI STANDARD 700
PART 6
DETERMINATION OF PURITY OF
NEW AND RECLAIMED REFRIGERANT 11
BY GAS CHROMATOGRAPHY
Section 1. Purpose Section 7. Sensitivity, Precision and Accuracy
The purpose of this test method is to determine the purity Statistical parameters for each impurity are listed in
of new and reclaimed fluoromchloromethane (R- 1 1) by gas Table 2. The data was obtained by analyzing an R-11
chromatopphy. calibration mixture 7 times during one day by one operator.
Section 2. Scope Section 8. Special Apparatus and Reagents
This test method is €or use with R- 1 1.
Section 3. Definitions
Definitions for this part are identical to those of ARI
Standatds 700-95 and 740-95.
Section 4. Principle
The organic purity of new and reclaimed R-11 is
determined by programmed temperature gas
chromatography using a packed column and flame
ionization detector (FID). Component peak areas are
integrated electronically and quantified by the area
normalization-response factor method.
Section 5. Applicability
This method is applicable to the determination of the
impurities typically present in commercially manufactured
R- 1 1 and in R- 1 1 recovered and reclaimed from operating
refiigeration systems.
Section 6. Limitations and Interferences
This method is calibrated for only those impurities
commonly present in R-11. Other impurities which have
been detected on occasion are listed (with retention times)
in Table 1. This method wil not detect any impurities
which may elute within the comparatively large R- 11 peak
matrix.
NOTE: EQUIVALENTS MAY BE SUBSTITUTED.
1. Gas chromatograph: Model 5890, equipped with
FID, Hewlea Paclcard. Valley Forge, PA.
2. Electronic integratoc Model 3396, Hewlett
Paclcard, Valley Forge, PA.
3. Packed column: 24 ' x 1/8" (7.32 m x 3.17 mm)
OD stainless steel, 1% SP-lo00 on Carbopack B,
6OBO mesh, Supelco, Bellefonte, PA.
4. Serum bottle: 125 ml, (Note: BottIe holds 160 ml
when liquid full.) Cat# 223748, Wheaton Glass,
Vineland, NJ.
5. Septa, No. 2-3244, 20 mm: seals, No. 3-3250,
Supelco, Bellefonte, PA.
6. Glass collecting tube: 125 ml, Cat# LG-8601, Lab
Glass Inc., Vineland, NJ. (Enlarge side outlet
opening to accommodate a crimp-on 2-cm septum.
Apply fiberglass tape outside for protection.)
7. Deflected point needles: Cat# 7174, #22, Popper
and Sons, Inc., New Hyde Park, NY.
8. R-11 and impurities for calibration standard
preparation: purchase the CCI,, CHCl,, C&Cl,
and TCE from Aldrich, Milwaukee, WI; all other
fluorochemicals from PCR Inc., Gainesville, FL.
Note: The purity of each calibration component
must be predetermined by gas
chromatography flame ionization detector
(FID) and/or thermal conductivity detector
(TO) and, if necessary, by gas
25
APPENDIX C TO ARI STANDARD 700
. .
9.1
chromatography/mass spectroscopy
(GC-MS).
Section 9. Procedure
Chromatographic Operating Conditions.
Detector
Carrier Gas
Init Column Temp.
Initial Hold
program
Final Column Temp.
Post Hold
Sample
Detector Temp.
Inj. Port Temp.
Max. Safe Column Temp.
FID
3Occ He/Min.
125OC
4 Min.
lOK/Min.
18OOC
14 Min.
1 pl (liquid syringe)
250°C
200oc
225OC (for
conditioning purposes)
f. Calculate the ppm added (to the nearest 1 ppm) for
each component by dividing the pg added by the
total weight of sample in the serum bottle (step e).
g. Calculate the ppm present for each component by
combining the ppm present in the stock R-11 (if
any) and the pprn component added (refer to
Note 1 in Section 10). The pprn component
present values are those used for determining the
method response factors.
h. Place the serum bottle standard in an ice bath and,
after it is ice cold, remove it and immediately
replace it with a new septum.
i. Write the ppm present vaiues for each component
on the label, date of prepamion, gross weight and
total grams of calibration sample. Store in a
refiigerator. Discard and prepare a new standard
when the sample weight falls below 60% of the
original weight.
Externally cool the syringe and sample to 5OOF (10OC)
before 93 Determinatwn of Component Response Factors
(see Note 3 in Section 10).
9.2 Calibration Standurd, Preparation and Analysis.
a.
b.
C.
d.
e.
Obtain a stock of the highest purity R-11 available
as evidenEd by the chromatograms using the
above procedure (see Note 1 in Section 10).
DeteImine the tare weight (to the nearest 0.01 g)
of a 125 ml serum bottle with septum and cap
loosely attached, then fill with stock R-11 to
within about 5B" of the top. Crimpon the
Septum.
Reweigh and subtract the tare weight in step b to
obtain the grams of R-11 added.
Individually and in turn add the volumes of each
calibration component indicafed in Table 3 through
the septum and below the R-11 liquid surface in
the bottle. Use an appropriate sized ml gas tight
syringe with deflected point,needles for gases and
a liquid pl syringe for liquids (see Note 2 in
Section 10). Shake the bottle to mix after addition
of each component
Total the pg added column and combine this
weight with that of step c to obtain the total weight
(to the nearest 0.01 g) of calibration sample in the
bottle.
a. Set up thelectronic integrator for an area
normalization-response factor calibration.
b. Analyze the calibration standard bulb in aiplicate
using the chromatographic conditions described in
9.1.
c. Using R-11 as the reference peak, perform the
necessary functions to have the integmor
determine each component Relative Response
Factor (RRFi ) which is then stored. Response
. Factors are calculated as follows:
ARF. = Kt96 i in Cal. Std.
1 4
m 1 1 =
1oo.m-s
All
where:
ARFi = Absolute Response Factor of component i.
A, = peak area of component i (average of 3
S = Wt% sum of all impurities present.
determinations).
Then, using R-1 1 as the referem peak:
APPENDIX C TO ARI STANDARD 700
RRF, values are computed to the nearest O.OOO1
unit.
9.4 Sampling. Submitted sampleshould be in either
metal cylinders or in glass or plastic bottles such that the
containers are at least 80% liquid full.
9.5 Sample Analysis. Analyze the sample using the
chromatographic conditions described in 9.1. The sample
and syringe are pre-cooled (refrigerator, ice bath) to 50°F
(10OC) before sampling. This is to simplify loading into
the pl syringe. Use component spiking and/or GC-MS (if
available) to identify questionable peaks.
9.6 Calculations.
a. The weight percentage of each component is
calculated as follows:
Where:
wi = weight percent of component i.
Rw;, = Relative Response Factor for
component i.
peak area of component i.
%A, x RRFJ = sum of all component peak areas
times their respective Relative
Response Factors.
b. Report sample component concentrations to the
nearest O.O001% (or to the nearest 1 ppm). If
results are less than the individual detection limits
(see Table 2), then report < the detection limit
(DL) value given.
Section 10. Notes
1. The purest R-11 will contain some of the impurities
listed in Table 1. The ppm amounts of impurities
already in the stock R-1 1 are determined via the
method of Standards Addition. Individual impurity
peak areas in the stock are increased in the calibration
standard by the ppm amount of the correspondmg
impurity added. The ppm already present is combined
with the ppm added to give the total ppm component
present in the calibration standard.
2. To preserve the stock of calibration gases. it is
suggested to load a small evacuated 125 ml gas
collecting tube to 1 afm. from the liquid phase as
illusmted in Figure 1. The appropriate volume is then
withdrawn and injected into the serum bottle containing
the R-11. For impurities which are liquids at ambient
temperature, inject the indicated pl volumes of each
respective component into the serum bottle.
3. Depending upon the electronic integrator used, it is
often more desirable to convert the ppm values to wt%
foresponse factor calculations and foreporting
Purposes.
Section 11. Reference
National Refrigerants Inc. Method NR200.0.
APPENDIX C TO ARI STANDARD 700
Calibration Chemical Cylinder
Septum
2 -cm
Vacuum Gauge
Gas Sampling Bulb
Vacuum
Figure 1. Apparatus used for Calibration Standard Preparation
28
APPENDlX C TO ARI STANDARD 700
RT Wn.)
2.20
2.65
3.03
3.22
3.97
4.63
5.98
6.35
7.00
7.41
.8.83
9.79
10.32
1 1.67
12.56
14.96
15.64
19.79
22.09
Comp.
23
13
152a
22
115
12
133a
21
30
114
11
123a
123
20
113
140a
10
1120
112
i 2.333
t
1
t
Figure 2. Gas Chromatogram of R-11
29
APPENDIX C TO ARI STANDARD 700.
1
Table 1. Retention Time Data for Identified Impurities
Not Normally Obsenred
Impurity Retention Time (Mia.)
3W)
8.00 GH*
237
1114 4.10
(1) Coelutes with R-23. To separate, aaach 1 ft. (30.5 cm) section of
Porapak-T column to detector end of column and rechromatograph (23
elutes first).
Table 2. Component Statistical Parameters
Component
~ ~~~
23
13
152a
22
115
12
133a
21
30
114
123a
123
20
113
10
1120
Detection
Limit,
PPm
2
3
1
2
2
2
1
2
2
2
3
2
2
2
2
2
Range
PPm
15
20
30
50
30
60
25
30
25
40
25
50
25
60
25
25
Investigated,
Precision at 95%
Confrdence Limit,
PPm
0.37
0.53
0.47
0.98
0.80
1.10
0.33
0.67
0.33
1.91
2.70
1.33
0.73
2.31
1.70
1.77
Relative Mean
Error, 5%
-2.8
-3.1
1.7
-0.8
0.7
1.1
-2.5
1.2
-2.5
-2.7
4.8
3.3
0.7
22
-3.3
1.8
30
APPENDIX C TO ARI STANDARD 700
Component
23(4)
13(4)
152a(4)
22(4)
115(4)
W4)
133a(4)
21(4)
30
114(4)
123a
123
20
113
10
1120
Table 3. Primary Calibration Standard Components
Mol. Wt.
70
105
66
86
136
121
118
103
85
170
153
153
120
188
154
132
VOl.
Added, ml
1.2
1 .o
2.5
3.2
1.2
2.8
1.1
1.6
5 .OpL
1.3
5 . w
1o.opL
5 .OpL
10.w
10.w
5 -w
PI2
Added (1)
3436
4274
6748
11321
6650
13 845
5332
6740
6680
9061
7490
14750
7445
15650
15950
7278
PPm
Added (2)
15
19
30
50
29
61
24
30
29
40
33
64
33
68
70
32
Total ppm
Present (3)
(1) If necessary, correct the pg added for the purity of the calibration component previously established.
(2) Values shown are for illustration; exact values are determined at 9.2f.
(3) Column to be filled in (9.2g) after determining ppm present in stock R-11 (see Note 1 in
(4) These impurities are gases at ambient room temperature: the others are liquids with low boiling
Section 10).
points.
31
APPENDIX C TO ARI STANDARD 700
PART 7 -
DETERMINATION OF PURITY OF NEW AND RECLAIMED
REFRIGERANT 12 BY GAS CHROMATOGRAPHY
Section 1. Purpose
The purpose of this test method is to determine the purity
of new and reclaimed dichlorodifluoromethane (R-12) by
gas chromatography.
Section 2. Scope
This .test method is for use with R-12.
Section 3. Definitions
Delinitions for this part are identical to
Standards 700-95 and 740-95.
those of ARI
Section 4. Principle
The organic purity of new and reclaimed R-12 is
determined by programmed temperature gas
chromatography using a packed column and flame
ionization detector (FJD). Component peak areas are
integrated electronically and quantified by the area
normalization-response factor method.
Section 5. Appiicability
This method is applicable to the determination of the
impurities typically present in commercially manufactured
R-12 q d in R-12 recovered and reclaimed from operating
refiigerarion systems.
Section' 6. Limitatlons and Interferences
The method is c a l i i for only those impurities
commonly present in R-12. Other impurities which have
been detected on occasion are listed (with retention times)
in Table 2. The method will not detect any impurity which
may elute within the comparatively large R-12 peak matrix.
Section 7. Sensitivity, Precision and Accuracy
Statistical parameters for each impurity are listed in
Table 1. The data was obtained by analyzing an R-12
calibration mixture 7 times during one day by one operator.
' Section 8. Special Apparatus and Reagents
NOTE. EQWALEN'E MAY BE SUBSTITUTED.
1.
2.
3.
4.
5.
6.
7.
Gas chromatograph: Model 5890, equipped with
FID, Hewletf packard, Valley Forge, PA.
Electronic integrator Model 33%. Hewlea
Packad, VaIley Forge, PA.
Packed colwnn: 2 4 ' x 1/8" (7.32 m x 3.17 mm)
OD stainless steel, 1% SP-lo00 on Carbopack B,
60BO mesh, Supelco, Bellefonte, PA.
Glass collecting tubes: 1ZmL and 5oomL,
LG-8601, Lab Glass Inc., Vineland, NJ. (Enlarge
side outlet opening to accommodate a crimp-on
2-cm septum. Apply fiberglass tape outside for
protection.)
Deflected point needles: cat# 7174, #Q2, Popper
and Sons, Inc., New Hyde Park, NY.
Swivel union: 1/4" SAE, US44, United
Refrigeration Inc., Philadelphia, PA.
R-12 and impurities for calibration standard
preparation: purchase the hydrocarbons and
methyl chloride h m Scott Specialty Gases,
Plumsteaflville, PA 2-butan01, MEK, and pentane:
purchase from Aldrich, Milwaukee, WI. all other
fluorochemicals: purchase from PCR he.,
Gainesville, FL.
Note: The purity of each calibration component
must be predetermined by gas
chromatography (flame ionization detector
and/or thermal conductivity detector) and,
if necessary, by gas chmmatographylmass
spectroscopy (W-MS).
32
APPENDIX C TO ARI STANDARD 700
8. Steel cylinder: 22 lb., with a #9 valve (Part#
1014-C, Superior Valve (20.). 3B" pipe neck, E. F.
Britten Co., Cranford, NJ.
Section 9. Procedure
9.1 Chromatographic Operating Conditions.
Detector
Carrier Gas
Init Column Temp.
Initial Hold
Final Column Temp.
Post Hold
Sample
Detector Temp.
Inj. Port Temp.
Max. Safe Column Temp.
Bogram
FID
3Occ He/Min.
4OoC
6 Min.
1owMin.
16oT
18 Min.
0.50 ml (gas syringe)
25OOC
15OOC
225OC (for
conditioning purposes)
9.2 Primary Calibration Standard, Preparation and
Analysis.
a. Crimp-on the septum, then determine the internal
volume of the 500 ml gas bulb by weighing the
bulb empty, then filled to maximum capacity with
water. Record the grams of water (as ml volume
capacity) on the outside of the bulb (to the nearest
1.0 ml).
b. Assemble the apparatus as illustrated in Figure 1.
c. Attach a cylinder of high purity R-12 to the gas
sampling bulb (refer to Note 1 in Section 10).
d. With valve "A" closed, open all other valves and
evacuate to 1 mm of Hg (0.1333 kPa).
e. Cl&e valve "D" and monitor the gauge for several
minutes to ensure that the system is not leaking.
f. Slowly open valve "A" and flash liquid phase R-12
to bring the system to 1 atm. Close valve "A."
g. Repeat steps d through f.
h. Close valves "B" and "C" and remove the bulb
from the vacuum/sampling apparatus.
i. Calculate the grams of R-12 added to the bulb as
follows:
grams added = 120.9 x Int. VoZ. of bulb (mL)
24450
Where: 120.9 = mol. wt. of R-12.
24450 = volume (ml) occupied by 1 mole of
R-12 at 77'F (25OC) and at 1 arm.
j. Individually and in turn add the volumes of each
gaseous calibration component indicated in Table 3
to the calibration bulb. Use an appropriate sized p1
or ml gas tight syringe with deflected point needles
(refer to Note 2 in Section 10).
k. Into a 30 ml (37 ml) serum bottle, capped and
crimped with a septum, add the exact volumes of
the following liquid impurities in the order given.
Density
Comu. Vol. Added, ml at 2OoC g
;?-butanol 6.0 0.808 4.848
MEK 5.0 0.805 4.025
113 4.0 1.565 6.260
n-pentane 2.0 0.626 1.252
123 6.0 1.470 8.820
11 9.0 1.487 13.383
Add by syringe injection through the septum using
a #22 needle (or smaller) as a vent. After
addition, shake vigorously to mix. Label, date and
store in a refiigerator.
1. Refer to Figure 1. Evacuate a 125 ml bulb (whose
intemal volume has been measured) and fill to
. 1 am. with R-12 stock
m. Accurately withdraw and inject exactly 10.0 pl of
solution from the 30 ml serum bottle into the
125 ml bulb. Allow to equilibrate for 30 minutes.
n. Using a 5 ml gas tight syringe, withdraw vapor
fmm the 125 ml bulb and inject exactly 5.0 ml
into the 500 ml calibration bulb. The pg of each
component thus added is calculated as follows and
is added to column four of Table 3:
gi x 5oooo
pgi added = 32 x A
33
APPENDIX C TO ARI STANDARD 700
Where:
gi= g from the table in step k, above.
A = internal ml of 125 ml bulb.
32 = total approx. volume (ml) of solution in
50,000 = dilution ratio.
step k.
0.
P.
9.
r.
Total the pg added column and combine this
weight with that of step i to obtain the total weight
of sample (to the nearest 0.0oO1 g) in bulb.
Calculate the ppm added (to the nearest 1 ppm) for
each component by dividing the pg added by the
total weight of sample in the gas bulb (step 0).
Calculate the ppm present for each component by
combining the ppm present ih the stock R-12 (if
any) to the ppm component added (refer to Note 1
in Section 10). The ppm component present
values are those used for determining the method
response factors.
Allow the gas calibration bulb to stand for 20 to
30 minutes to equilibrate. The standard will be
stable for 3 to 4 days.
93 Deternuhation of Component Response Factors (see
Note 3 in Section 10).
Set up thelectronic integrator for an area
nannalization-response factor calibration.
Analyze the calibration standard bulb in triplicate
using the chromatographic conditions described in
9.1.
Using R-12 as the reference peak, perfonn the
n e c e functions to have the
determineach component Relative
Factor (RFG, ) which is then stored.
Factors are calculated as follows:
Where:
A R F i = Absolute Response Factor of component i.
4 = peak area of component i (average of 3
S = Wt% sum of all impurities present.
determinaiions).
Then, using R-12 as the reference peak:
FNF, values are computed to the nearest O.OOO1
unit.
9.4 Secondary Calibration Standard Preparation.
Note: A secondary calibration standard is
a.
b.
C.
d.
e.
prepared in much larger quantity due to
the comparatively short lifetime of the
primary bulb standard The primary bulb
standard is necessaq initially because of
inherent phase distribution of added
components if simply prepahg and
calibrating a standard such as described
here. The secondary standard is analyzed
as a sample against the primary standard
and then used subsequently as the daily
calibration standard.
Evacuate a clean, dry 2.2 lb. steel cylinder and
dete- the tare weight to the nearest 0.1 g.
Attach a Swagelok nut and septum to the valve
and then externally cool the cylinder in ice water.
Open the cylinder valve.
While keeping the cylinder in ice, individually and
in turn add 400 times the volume of each gaseous
component giveq in Table 3 to the cylinder by
syringe injection through the septum. Similarly
add 0.15 ml of the liquid refiigerant mixture from
92, step k. Close the cylinder valve and remove
the Swagelok nut and septum.
Evacuate a second clean, dry 2.2 lb. 'steel cylinder
and determine the tare weight of the cylinder to the
nearest 0.1 g.
Cool the cylinder externally in ice water for
30 minutes, then aaach a short (up to 24") section
of flex line fkom the stock cylinder supply of R-12.
34
APPENDIX C TO ARI STANDARD 700
Purge a small amount of R-12 through the flex line
before immediatelv attaching to the 2.2 Ib.
cylinder.
Note: The stock R-12 must be at 75'F (23.9'C)
or higher before the transfer. If necessary,
warm, but do not overheat, with a heat
gun.
f. While keeping the cylinder immersed in ice water,
open the 2.2 Ib. cylinder valve, then open the stock
cylinder valve and fill with 1200 g of liquid R-12.
It may be necessary to reconnect the flex line and
add more R-12 until the desired 1200 g has been
added. If significantly greater than 1200 g is
added, vent the cylinder to give approximately
1200 g (see Note 4 in Section 10).
g. Remove the 2.2 lb. cylinder from the ice bath and
allow it to warm to ambient temperature.
h. Place the 2.2 lb. secondary standard cylinder (the
cylinder mentioned in steps a, b and c) in the ice
bath and cool for 30 minutes.
i. Using a short 1/4" swivel union, invert the 2.2 Ib.
cylinder containing the 1200 g of R-12 and
connect ito the secondary standard cylinder.
Carefully open the valve and purge some of the
R-12 vapor to sweep the coupler before
immediately connecting it to the secondary
standard cylinder. Warm the cylinder containing
the R-12 with a heat gun. Do not overheat.
j. While keeping the calibration cylinder in ice water,
open the valves on both cylinders and allow all of
the R-12 to transfer into the calibration standard
cylinder. Close the cylinder valves and disconnect
from the swivel union.
k. Remove the calibration cylinder from the ice bath
and allow the cylinder to reach ambient laboratory
temperature before the final weighing. Wipe dry,
then reweigh to the nearest 0.1 g.
1. Subtract the tare weight (step a) from the total
weight (step k) to obtain the total grams of
standard R-12 in the cylinder. Record this weight
together with the cylinder tare weight and date of
preparation on the cylinder label.
m. Roll the cylinder for one hour to thoroughly mix.
0. Average the results calculated electronically (see
9.7, Calculations) and tabulate to the nearest
1 ppm. List each component on the cylinder label
with the ppm amount for each. This cylinder is
used henceforth as the calibration standard until the
loss of standard weight indicates that the internal
volume of liquid phase is less than 60% of the
total internal volume of the cylinder (liquid density
of R-12 equals 1.311 g/cc at 77'F (25OC)). The
internal volume of a 2.2 Ib. steel cylinder is
lo00 ml (water volume).
95 Sampling. Submitted sample cylinders must contain
sufficient liquid phase (80% liquid full is recommended)
for analysis.
9.6 Sample Analysis. Analyze the sample using the
chromatographic conditions described in 9.1. Load the
sample as illustrated in Figure 1 by flashing the liquid
phase into an evacuated gas bulb and bringing to 1 am.
pressure.
Note: Alternatively, the sample liquid phase may be
flashed into a Tedlar bag (1 L recommended)
and the 0.5 ml sample for GC analysis
withdrawn from the bag.
9.7 Calculations.
a. The weight percentage of each component is
calculated as follows:
RRFixAix 100 w = '' i WAi x RRF,)
Where:
Wi = WL% of component i.
RRFi = Relative Response Factor for component i.
A, = peak area of component i.
X... = sum of all component peak areas times their
respective Relative Response Facton.
b. Report the sample component concentdons to the
nearest O.O001% (or to the nearest 1 ppm). If
results are less than the individual detection limits
(see Table l), then repon < the detection limit
(DL) value given.
n. Chromatograph in triplicate the cylinder contents
as described in 9.1 loading into an evacuated gas
bulb as shown in Figure 1.
35
APPENDIX C TO ARI STANDARD 700.
Section 10. Notes
1. The purest R-12 will contain some of the impurities
listed in Table 1. The ppm amounts of impurities
already in the stock R-12 are determined via the
method of Standards Addition. Individual impurity
peak areas in the stock are increased in the calibration
standard by the ppm amount of the corresponding
impurity added. The ppm already present is combined
with the ppm added to give the total ppm component
present in the calibration standard.
3. Depending upon the electronic integrator used, it is
often more desirable to convert the ppm values to wt %
foresponse factor calculations and for reporting
Purposes-
4. During the R-12 addition to the 22 Ib. cylinder
(secondary standard preparation), it is unnecessary to
bring the cylinder to ambientemperature between
weighings as only an appxhate 1200 g weight is required.
2. To preserve the stock of calibrarion component, it is Section 11. Reference
suggested to load a small evacuated 125 ml gas
collecting tube to 1 atm. from the liquid phase as National Refrigerants Inc. Method NR202.l.
illustrated in Figure 1. The appropriate volume is then
withdrawn and injected into the 500 ml calibration bulb.
36
’ APPENDIX C TO ARi STANDARD 700
Calibration Chemical Cyfinder A Vacuum Gauge
Septum
2-cm
Gas Sampling Bulb
I
liquid phase \ \ \
Vacuum
Figure 1. Apparatus used for Calibration Standard Preparation and for Cylinder Sampling
37
APPENDIX C TO ARI STANDARD 700
2.40
3.19
3.92
4.62
5.36
6.27
7.23
7.49
7.97
8.60
10.91
11.19
12.92
13.61
15.17
15.77
. 16.31
16.69
17.58
18.19
18.94
19.54
20.85
2 1.63
24.18
24.65
25.34
26.34
Comp.
CH,
23
C,H,
c*J&
13
143a
15%
40
134a
22
115
12
142b
124
133a
21
114
c3H,
~ ~ 4 h l
n-C,H,,
2-C.h
11
123a
123
MEK
113
2-butanol
C5H12
I
38
Figure 2. Gas Chromatogram of R-12
APPENDIX C TO ARI STANDARD 700
Component
Methane
23
C,H,
c*H,
13
143a
152a
140
134a
22
115
142b
124
133a
21
isobutane
114
n-butane
2-butene-T
11
123
2-butan01
MEK
113
n-pentane
GH,
Table 1. Component Statistical Parameters
Detection
Limit,
PPm
0.5
2
0.5
0.5
3
1
1
1
1
2
0.5
2
1
1
1
2
0.5
2
0.5
0.5
4
2
2
2
2
0.5
Range
Investigated,
PPm
5
25
5
5
30
25
30
20
45
65
5
115
20
25
35
50
20
50
20
5
40
35
20
25
30
5
Precision at 95%
Confidence Limit,
PPm
0.07
0.54
0.13
0.10
0.47
0.30
0.63
0.37
0.27
1.75
0.10
1.67
0.23
0.37
023
0.83
023
0.83
0.18
0.06
0.87
1.05
0.33
0.47
0.87
025
Relative Mean
Error,
%
4.00
-2.30
-5.60
-4.10
-3.80
3.30
1.67
2.30
-3.30
2.73
3.37
1 .a0
-1.33
1.83
1.83
1.80
-2.77
2.03
-3.33
-3.80
1.05
-4.73
1.60
-2.33
-4.00
-3.73
-
Table 2. Retention TSme Data for Identified Impurities
Not Normally Observed
Impurity Retention Time Win.)
~~
3x11
11-60 C,H*
6.00 1114
3.45
30 16.93
2,2-Dimethylppane 19.80
hpentane 24.30
(1) Coelutes with R-23. To separate, attach 1 ft. (30.5 cm) section of
Porapak-T column to detector end of column and rechromatograph (23
elutes first).
39
APPENDIX C TO ARI STANDARD 700
Component
Methane
23
CZH,
CZH,
13
143a
152a
40
134a
22
115
142b
124
. 133a
21
isobutane
114
n-butane
2-butene-T
1U4)
123(4)
Mm4)
113(4)
n-pentane(4)
' C,H,
2-butan01(4)
Table 3. Primary Calibration Standard Components
16
70
28
30
104
84
66
50
102
86
42
154
100
136
118
103
58
170
58
56
137
153
72
188
74
72
Vol.
Added, pl
-~ ~
20
22
12
11
20
20
30
28
30
50
8
50
15
12
20
32
25
20
25
6
M
Added (1)
13.1
63.0
13.7
13.5
85.4
68.8
81.0
57.8
125.1
176.9
13.7
315.9
61.7
67.0
97.0
134.7
59.3
139.8
59.3
13.7
(5)
(5)
(5)
(5)
(5)
(5)
PPm
Added (2)
5
23
5
5
31
25
30
21
46
64
5
115
22
24
35
49
22
51
22
5
57
38
17
27
21
5
Total ppm
Present (3)
(1) If necessary, correct the pg added for the purity of the calibration component previously
(2) Values shown are for flustration; exact values are detennined at step p.
(3) Column to be filled in (step q) after determining ppm present in stock R-12 (see Note 1 in
(4) These components are liquids at ambient laboratory temperature and are added to the 500 ml
(5) From step n.
established.
Section 10).
bulb as described in 92, steps k through n.
40
APPENDIX C TO ARI STANDARD 7M)
PART 8
DETERMINATION OF PURITY OF NEW AND RECLAIMED
REFRIGERANT 13 BY GAS CHROMATOGRAPHY
Section 1. Purpose will not detect any impurity which may elute within the
comparatively large R-13 peak matrix. At this point in
The purpose of this test method is to determine the purity . time, virtually no experience with reclaimed R-13 has been
of new and reclaimed chlorolrifluoromethane (R-13) by gas documented; however, the impurities profiie should be
chromatography. that of new refrigerants.
Section 2. Scope ' Section 7. Sensitivity, Precision and Accuracy
This test method is for use with R-13.
Section 3. Definitions
Statistical parameters for each impurity are Listed in
Table 1. The data was obtained by analyzing an R-13
calibration mixture 7 times during one day by one operator.
Definitions for this part are identical to those of ARI Section 8. Special Apparatus and Reagents
NOTE: EQUIVALENTS MAY BE S U B S m D .
Standards 700-95 and 740-95.
Sectlon 4. Principle 1. Gas chromatograph: Model 5890, equipped with
TCD, Hewlett Packard, Valley Forge, PA.
The organic purity of new and reclaimed R-13 is
determined by programmed temperature gas 2. Electronic integrator: Model 33%, Hewlett
chromatography using a packed column and thermal packard Valley Forge, PA.
conductivity dewtor (TCD). Component peak areas are
integrated elecmnically and quantified by the area 3. Packed column: 2 4 ' x 1/8" (7.32 m X 3.17 mm)
normalization-response factor method. OD stainless steel, 1% SP-lo00 on Carbopack B,
60BO mesh, Supelco, Bellefonte, PA.
Section 5. Applicability
This method is applicable to the detemination of the
impurities typidy present in commercially manufactured
R-13 and in R-13 recovered and reclaimed from operating
refrigeration systems.
4. Glass collecting tubes: 125 ml and 500 ml,
LG-8601, Lab Glass Inc., Vineland, NJ. (Enlarge
side outlet opening to accommodate a crimpon
2-cm septum. Apply fiberglass tape outside for
protection.)
5. Deflected point needles: Cat# 7174, #22, Popper
and Sons, hc., New Hyde Park, NY.
Section 6. Limitations and interferences
The method is calibrated for only those impurities
commonly present in R-13. Because R-14 and R-116 have
very poor response to the flame ionization detector (FID),
the TCD is used here. This limits the detection for the
other impurities: however, in R-13 there is no compelling
need to detect the other impurities at very low
concentrations. R-13B1 and R-3 1 coelute. If the presence
of R-31 is suspected, reanalyze the sample using the test
method for R-23 purity (see Part 10). The R-13 method
6. Stainless steel cylinder: 1000 ml,
304L-WDF4-1O00, 1/4" pipe, Whitey Co.,
Highland Heights, OH.
7. Swivel union: 1/4" SAE, US44, United
Refrigeration Inc., Philadelphia, PA.
8. R-13 and impurities for calibration standard
preparation: The R-13, 13B1 and all of the
41
APPENDIX C TO ARI STANDARD 700-.
9.1
93
fluorochemical impurities may be purchased from
PCR Inc., Gainesville, FL.
Note: The purity of each calibration component
must be predetermined by gas
chromatography flame ionization detector
(FID) and/or thermal conductivity detector
(TCD) and, if necessary, by gas
chromatography/mass spectroscopy
(GC-MS).
Section 9. Procedure
Chromatographic Operating Conditions.
Detector
Canier Gas
Init. column Temp.
Initial Hold
program
Final Column Temp.
Post Hold
Sample
Detector Temp.
Inj. Port Temp.
Max. Safe Column Temp.
TCD, Low Sensitivity
(.%e Note 5 in
section 10)
3Occ He/Min.
4OoC
6 Min.
lOK/Min.
16OT
4 Min.
2.0 ml (gas syringe)
200T
l5OT
225OC (for
conditioning pmposes)
Primary Calibration Standard, Preparation and
Analysis.
a.
b.
C.
d.
e.
Crimp-on the septum, them determine the internal
volume of the 500 ml gas bulb by weighing the
bulb empty, then fill to maximum capacity with
water. Record the grams of water as ml volume
capacity on the outside of the bulb (to the neamt
1.0 ml).
Assemble the apparahls as illustrared in Figure 1.
Attach a cylinder of high purity R-13 to the gas
sampling bulb (refer to Note 1 in Section 10). To
ensure that sufficient liquid phase is present, cool
theR-13 cylinder to 4OOF (4.5OC) before sampling.
The critical tempemure of R-13 is 839OF
(28.8OC).
With valve "A" closed, open all other valves and
evacuate to 1 mm of Hg (0.1333 kPa).
Close valve "D" and monitor the gauge for several
minutes to ensure that the system is not leaking.
f.
gh.
1.
Slowly open valve "A" and flash liquid phase R-13
to bring the system to 1 atm. Close valve "A."
Repeat steps d through f. Allow the R-13 vapor in
the bulb to reach ambientemperature before
making the final pressure adjustment to 1 atm.
Close valves "B" and "C" and remove the bulb
from the vacuum/samphg apparatus.
Calculate the grams of R-13 added to the bulb as
follows:
gramsadded=
104.46 x Int. vol. of bulb (mt)
24450
Where:
104.46 = mol. wt of R-13.
24450 = volume (ml) occupied by 1 mole of R-13
j.
k.
1.
m.
n.
at 77OF (250C) and at 1 -am.
Individually and in turn add the volumes of each
calibration component indicated in Table 1 to the
calibration bulb. Use an appropriate sized pl or ml
gas tight syringe with deflected point needles (refer
to Note 2 in Section 10).
Total the pg added column and combine this
weight with that of step i to obtain the total weight
of sample (to the nearest O.OOO1 g) in bulb.
Calculate the ppm added (to the nearest 1 ppm) for
each component by dividing the pg added by the
total weight of sample in the gas bulb (step k).
Calculate the ppm present for each comrnnent by
combining the ppm p e n t in the stock R-13 (if
any) with the ppm component added (refer to
Note 1 in Section 10). The ppm component
present values are those used for determining the
method response factors.
Allow the gas calibration bulb to stand for about
20 to 30 minutes to equilibrate. The standard will
be stable for 3 to 4 days.
93 Determination of Component Response Factors (see
Note 3 in Section 10).
a Set up the electronic integrator for an area
normalimion-response factor calibration.
42
APPENDIX C TO ARI STANDARD 700
b. Analyze the calibration standard bulb in triplicate
using the chromatographic conditions described in
9.1.
c. Using R-13 as the reference pealL perform the
necessary functions to have the integrator
determine each component Relative Response
Factor (RRF, ) which is then stored. Relative
Response Factors are calculated as follows:
ARF. = Wt% i in Cal. Std.
1 4
Where:
ARF, = Absolue Response Factor of component i.
* A, = peak area of component i (average of 3
determinations).
S = Wt% sum of all impurities present.
Then, using R-13 as the refmnce peak:
RRF, values are computed to the nearest O.OOO1
unit.
9.4 Secondary Calibration Standard Preparation.
Note: A secondary calibration standard is prepared
in much larger quantity due to the
compiaatively sha lifetime of the primary
bulb standard. ?he primary bulb standard is
necessrny initially because of inherent phase
distribution of added components if simply
preparing and calibrating a standard such as
described here. The secondary standard is
analyzed as a sample against the primary
standard and then used subsequently as the
daily calibration standard.
a. Evacuate a clean, dxy 1 liter stainless steel cylinder
and detumine the tare weight to the nearest 1.0 g.
b.
C.
d.
e.
f.
h.
1.
Attach a Swagelok nut and septum to the valve
and then externally cool the cylinder in ice water.
Open the cylinder valve.
While keeping the cylinder in ice water.
individually and in turn add 350 times the volume
of each component given in the calibration table
(Table 2) to the cylinder by syringe injection
through the septum. Close the cylinder valve and
remove the Swagelok nut and septum.
Evacuate a second clean, dry 1 liter stainless steel
cylinder and determine the tare weight of the
cylinder to the nearest 1.0 g.
Cool the cylinder to about -3OOC extemally in
dry-ice (approximately 10 minutes), then attach a
short (up to W) section of flex line €ram the stock
cylinder supply of R-13. Purge a small amount of
R-13 through the flex line before immediately
attaching to the 1 L cylinder.
Note: The stack R-13 must be at 80°F (26.5OC)
or higher before the transfer. If necessary,
warm, but do not overheat, with a heat
gun.
While keeping immersed in dry-ice, open the 1 L
cylinder valve-then open the stock cylinder valve
and fill with 820 g of R-13. It may be necessary
to reconnect the flex line and add more R-13 until
the desired 820 g has been added. If significantly
greater than 820 g is added, vent the cylinder to
give approximately 820 g (see Note 4 in
Section 10).
Remove the 1 L cylinder from the dry-ice cooling
bath and allow to warm to ambient ternpemture.
Place the 1 L secondary standard cylinder (the fmt
cylinder) in the dxy-ice bath and cool for
10 minutes.
Using a short 1/4" swivel union, invert the 1 L
cylinder containing the 820 g of R-13 and connect
to the secondary standard cylinder. Carefully open
the valve and purge some of the R-13 vapor to
sweep the coupler before immediately connecting
to the .secondary standard cylinder. Warm the
cylinder containing the R-13 with a heat gun. Do - not overheat+
While keepjng the calibration cylinder in dry-ice,
open the valves on both cylinders and allow all of
the R-13 to transfer into the calibxation smdard
43
APPENDIX C TO ARI STANDARD 700
cylinder. Close the cylinder valves and disconnect
from the swivel union.
k. Remove the calibration cylinder from the dry-ice
bath and ailow the cylinder to reach ambient
laboratory temperature before the final weighing.
Wipe dry, then reweigh to the nearest 1.0 g.
1. Subtract the tare weight (step a) from the total
weight (step k) to obtain the total grams of
standard in the cylinder. Record this weight
together with the cylinder tare weight and date of
preparation on the cylinder label.
m. Roll the cylinder for one hour to thoroughly mix.
n. Cool the cylinder to 4OOF (4SoC), then sample
from the liquid phase by loading into an evacuated
gas bulb as shown in Figure 1. Chromatograph in
mplicate as described in 9.1.
0. Average the results calculated electronically (see
9.7, Calculations) and tabulate to the nearest
1 ppm. List each component on the cylinder label
with the ppm amount for each. This cylinder is
used henceforth as the calibration standard until the
loss of standard weight is 40% of the original
weight. At that time a new c a l i i o n srandard is
P P d -
95 Sampling. Submitted sample cylinders must contain
sufficient sample (minimum 700 g/liter) as to provide a
sufficient liquid/vapor ratio at 40°F (4.4'0. Note the
critical temperame of R-13 is 83.9OF (285OC).
9.6 Sample Analysis. Analyze the sample using the
chromatographic conditions described in 9.1. Externally
chill the sample cylinder to W F (4.4OC) and then load as
illustrated in Figure 1 by flashing the liquid phase into an
evacuated gas bulb and bringing to 1 atm. pressure.
Nok: Alternatively, the sample liquid phase may be
flashed into a Tedlar bag (1 L recommended)
and the 2.0 ml sample for GC analysis
withdrawn from the bag.
9.7 Calculations.
a. The weight percentage of each component is
calculated as follows:
RRFixAix 100 wi =
'(Ai X
1.
2.
3.
4.
5.
~ ~
Where:
Wi = weight percent of component i.
RRF, = Relative Response Factor for component i.
4 = peak area of component i.
X... = sum of all component peak areas times their
respective Relative Response Factors.
Note: Ignore the air peak throughout these
calculations.
b. Report sample component concentrations to the
nearest 0.0o01% (or to the nearest 1 ppm). If
results are less than the individual detection limits
(see Table l), then report c the detection limit
(DL) value given.
Section 10. Notes
The purest R-13 will contain some of the impurities
listed in Table 1. The ppm amounts of impurities
already in the stock R- 13 are determined via the
method of Standards Addition. Individual impurity
peak areas in the stock are increased in the calibration
standard by the ppm amount of the comsponding
impurity added. The ppm already present is combined
with the ppm added to give the totaI ppm component
present in the calibration standmi. The ppm amounts
added for calibmtion should be greater than the usual
amounts present in typical samples. This is because the
peak response of the T O becomes increasingly
non-linear as impurity concentration levels increase.
To preserve the stock of calibration component, it is
suggested to load a mail evacuated 125 ml gas
collecting tube to 1 am. fmm the liquid phase as
illusaated in Figure 1. The appropriate volume is then
withdrawn and injected into the 500 ml calibration bulb.
Depending upon the electronic integrator used, it is
often more desirable to convert the ppm values to wtQ
foresponse factor calculations and forepoIting
purpOs=
During the R-13 addition to the second lo00 ml
cylinder (secondary standard prepamion), it is
unnecessary to bring the cylinder to ambient
temperature.during intermediate weighin@ as only an
aumximate 820 g weight is required. Note that the
liquid density of R-13 is 0.907 g/cc at 77aF (25OC).
A 32-fold increase in peak sensitivity is possible by
operating the chromatographic TCD at the high
sensitivity position. However, at higher sensitivity, the
useful lifetime of the detector (hot wires) is diminished,
44
APPENDIX C TO ARI STANDARD 700
the baseline noise and background is often intolerable Section 11. References
and peak area reproducibility is generally degraded.
For these reasons, the lower sensitivity position was 1. National Refrigerants Inc. Method NR204.0.
chosen for routine application.
2. DuPont Chemicals Dept. Method F3275.165.01CC (P).
45
APPENDIX C TO ARI STANDARD 700
Calibration Chemical Cylinder
Septum
2 -cm i Vacuum Gauge
\ \ \ \ \ n
4 / "C" v D'
I 1
Gas Sampiing Bulb
Vacuum
Figure 1. Apparatus used for Calibration Standard Preparation and for Cyllnder Sampling
46
. . APPENDIX C TO ARI STANDARD 700
2 . 2 4
1
3 . 3 7
T
4.96
. cf9.53
I
I
I '17.91
19.94
Note: Methane, which is sometimes
present, elutes at 2.86 minutes.
Figure 2. Gas Chromatogram Of R-13
47
APPENDIX C TO ARI STANDARD 700
Component
14
23
116
125
22
31
115
12
114
11
Tall
Detection
Limit,
PPm
12
10
14
12
10
10
12
10
12
12
! 1. Component Statistical Parameters
Range
Confidence Limit,
~ InvestiEted,
Precision at 95%
PPm
300 72
200 2.7
250 3.1
150 6.6
I 175 5.3
100 2.6
200 4.7
200 3.6
100 63
100 3.1
~
Relative
Mean Error,
9%
1.7
-1.0 .
1.3
-2.0
23
2.0
-2.0
2.5
4 .O
-2.0
Table 2. Prlmary Calibration Standard Components
Component I Mol. Wt.
14
23
116
125
22
31
115
12
-1 14
11(4)
88
70
138
120
86
68
154
121
170
137
Vol.
Added, mi *
0.20
0.16
0.10
0.08
0.12
0.10
0.08
0.10
0.04
0.04
Pg
Added (1)
720
458
564
393
425
280
506
495
279
225
PPm
Added (2)
303
193
237
166
179
118
213
209
118
95
Total ppm
Present (3)
(1) If necessary, comt the pg added for the purity of the calibration component previously established.
(2) Values shown are for illudon; exact values are determined at step 1.
(3) Column to be filled in (step m) after determining ppm present in stock R-13 (see Note 1 in Section 10).
(4) Added by warming an uncapped vial of the liquid component to enrich the headspace vapor, capping,
cooling, then removing headqmx vapor via a gas syringe.
48
. APPENDIX C TO ARI STANDARD 700
PART 9
DETERMINATION OF PURITY OF NEW AND RECLAIMED
REFRIGERANT 22 BY GAS CHROMATOGRAPHY
Section 1. Purpose
The purpose of this test method is to determine the purity
of new and reclaimed chlorodifluoromethane, (R-22), by
gas chromatography.
Section 2. Scope
This test method is for use with R-22.
Section 3. Definitions
Definitions for this part are identical to those of ARI
Standards 700-95 and 740-95.
Section 4. Principle
The organic puxity of new and reclaimed R-22 is
determined by programmed temperature gas
chromatography using a packed column and flame
ionization detector 0). Because R-22 obscures R-3 1 on
the packed column, R-31 is determined separately by a
capillary column procedure. Component peak areas are
integrated electronically and quantified by the area
nomalization-response factor method.
Section 5. Applicabiilty
This method is applicable to the determination of the
impurities typically present in commercially manufactured
and in reclaimed R-22.
Section 6. Limttations and Interferences
This method is calibrated for only rhose impurities
commonly present in new and reclaimed R-22. Other
impurities which have been detected on occasion are listed
(with retention times) in Table 2. The method will not
detect any impurity which may elute within the
comparatively large R-22 peak matrix.. The R-31 peak
(capillary column method) will be obscured if n-butane is
present in exceptionally large concentrations.
Section 7. Sensitivity, Precision and Accuracy
Statistical parameters for each impurity are listed in
Table 1. The data was obtained by analyzing an R-22
calibration standard mixture 7 times during one day by one
operator.
Section 8. Special Apparatus and Reagents
NOTE: EQUIVALENTS MAY BE SUBSTITUTED.
1.
2.
3.
4.
5.
6.
7.
8.
9.
Gas chromatograph Model 5890, equipped with
FID and with capillary column split injection and
packed column capability, Hewlett Packard, Valley
Forge, PA.
Electronic integrator: Model 33%. Hewlett
Packard, Valley Forge, PA.
Packed column: 24 x 1/8" (7.32 m x 3.17 mm)
OD stainless steel, 1% SP-lo00 on Carbopack B,
60BO mesh, Supelco, Bellefonte, PA.
Capillary column: 120 m, two (2) 60 m DB-1301,
0.25mm, lp, J&W Scientific Co., Folsom, CA.
Glass collecting tubes: 125 ml and 500 ml,
LG-8601, Lab Glass Inc., Vieland, NJ. (Enlarge
side outlet opening to accommodate a uimp-on
2cm septum. Apply fiberglass rape outside for
protection.)
Steel cylinder 22 lb., with a single #9 valve
(#1014-C, Superior Valve), 3/8" pipe neck, E. F.
Britten Co., Cranford, NJ.
Deflected point needles: Cat #7174, #22, Popper
and Sons, Inc., New Hyde Park, NY.
Swivel 'union: US44, United Refrigeration Inc.,
Philadelphia, PA.
R-22 and impurities for calibration standard
preparation: aIl fluorochemimls may be purchased
from PCR Inc., Gainesville, FL; the hydrocarbons
may be purchased from Scott Specialty Gases,
49
APPENDIX C TO ARI STANDARD 700
. .
Plumsteadville, PA: methyl chloride may be
purchased from Aldrich, Milwaukee, WI.
Note: The purity of each calibration component
must