Chromatography is a process that separates the components of a substance into two phases, stationary and mobile phase. Chromatography was first introduced by Mikhail Tswett, who used separation to explore the chemistry that takes place in chlorophyll. However, Gas Chromatography (GC) did not come about till years later. In 1952, Martin and Synge received the Nobel Prize for the invention of partition chemistry. They used partition chromatography to explain how the makeup of wool. They came to the conclusion that they could separate the individual amino acids and peptides which led to the structure of wool. This method is still very important today because it is practical in trying to unravel gene sequencing complications.1 This discovery lead to the creation of a new method, Gas Chromatography which is used specifically for the division of volatile substances by filtering a gaseous mobile phase through a stationary phase. Gas Chromatography has many applications to daily life; it can trace certain hydrocarbons/pollution in air, drugs in blood/urine samples, and contaminants in alcohol. Also GC can be used in the detection of explosives by separating the different measurements taken from an electron-capture detector.  This serves as a very useful tool in such things as airports, which require high levels of security.
Within a commercial Gas Chromatograph a sample of a substance is injected into a separating column, stationary phase, and is constantly being forced through the column by a constant flow of gas. Temperature, carrier gas flow, column length, amount of material injected effect of the gas and intermolecular forces of the gas and mobile phases are the reasons that the separation takes place.2 If the substance is polar it will interact with a polar stationary phase much better translating into a greater retention time than if it was non-polar; this follows the principle of “like dissolves like”.  After this the components will exit the column and enter the detector unit. In the commercial GC there were three different detectors: thermal conductivity, flame ionization and electron capture but in the experiment that was conducted the Beilstein detector was used. 3 The Beilstein detector emits different colored light when separated halocarbons go through it. A Beilstein is made up of a column and at the end of the column there is a copper wire coil. A constant stream of gas is pushed through the column, just like the commercial GC, but rather than a detector unit a Beilstein detector has a flame. The flame reacts with the electrons that are excited by the increase in temperature, and as the electrons fall back to the ground state the copper wire emits a blue-green light; this light indicates that a halogen-containing component. This process can be very useful because it can detect refrigerator and air conditioner leaks. 3ta
The goal of this experiment was to experimentally show that Freon 22 will separate and have a smaller retention time than Freon 123 because it has the smaller intermolecular forces and molecular weight that leads to a lower boiling point and dipole moment. These forces are shown on Table 2.
This experiment followed the procedure in PSU Chemtrek Small-Scale Experiments for General Chemistry for Experiment #19. 3 The small scale gas chromatograph that was built in this experiment consisted of a glass column packed with Tide (stationary phase), a copper coil and latex tubing. The latex tubing was used to connect the carrier gas to the glass column and the glass column to the copper coil. The carrier gas was sent through the column and came out where the copper coil was; a match was taken to the gas which caused a small flame. The coil acted as the Beilstein Detector in this homemade GC. The gas flow rate was then calculated by an injection of air. After the injection the flame would dip then return to normal, the amount of time that elapsed was taken which is the gas flow rate.
Both Freon 22 and 123 were injected after a dilution of 6% into the tubing. To dilute the sample, the syringe was filed with 3 cc of gas then 7 cc of air was added (total of 10). The syringe was then pushed back to 2 cc then finally pulled back to 10 cc. At this time the Freon 22 and Freon 123 were added at separate times to see how long it took for the flame to initial show a color change, show a maximum intensity of that color change and disappearance of that color change. These times were then used to calculate the retention times and bandwidth. Also a mixture of Freon 22 and 123 was injected and the same calculations were taken.
A commercial GC was also used to test the sample of unknown concentrations Freon 22 and 123. The two gases were put through the commercial GC and calibration graphs were given to detect the gases.
Table 1. Air flow times/Gas Flow Rate
Average = (5.28 + 5.95 + 5.92)/ 3 = 5.72 sec
Gas Flow Rate = ū (cm/sec) = 31cm / 5.42 sec
ū = 5.42 cm/sec
The time gives the amount of time that the flame dipped. The glass column that was filled with Tide, was measured to be 31cm long (L). The Gas Flow Equation Below uses the average air flow time calculated above.
Table 2. Physical Data for Freon 22 and Freon 123
(ū = 7.05 cm/sec)
(ū = 7.05 cm/sec)
*-    Peter Racioppo, Caitlyn Parry, Matt Palmar Lab Manual
**-  Alex Bohen, Lab Manual
Appearance, Max Intensity, Disappearance were the averages of the color change over three trials. These values are used to find the retention time (TR) and the bandwidth (WB). The dipole moments, molecular weights and boiling points were from the Merck Index. 
Retention Time (tR) = Maximum Intensity (s)
Freon 123: tR = 19 s
Band Width (Wb) = Disappearance(s) - Appearance (s)
Freon 123: Wb = 30.2 - 7.8 = 22.4 sec.
Table 3: Freon 22 and Freon 123 Mixture
.1 mL of Freon 22 and .25 mL of Freon 123 were mixed to form a .35 mL solution. The same procedure was followed as in the Procedure Section.
Table 4: Resolution using the Homemade GC
Resolution = ( tR (Freon 123) - tR (Freon 22)) / ((Wb (Freon 22) + Wb (Freon 123) / 2)
3Resolution for homemade GC = (16.6 - 4.4) / ((.8 + 22.2) / 2) = 1.06
Resolution using the Homemade GC was measured using the data previously collected with the equation that was given in the Chemtrek.3
Table 6: Calibration Graph Data from Commercial GC
Freon 123 (area)
The data from Table 6 was used in the Calibration Graphs of both Freon 22 and 123.
Table 7: Commercial GC Concentration, Retention times, and Peak for Freon 123 and 22
Concentration before dilution (ppm)
Retention time (s)
Sample calculation using the trend line for Freon 22
y = 26.46x - 83
3268 = 26.46x - 83
x = 126.64 ppm
3cc à 10 cc = 30% Fr
2cc à 10cc = 20% Fr
60% of the 30 % à 10 cc
= 6% dilution
126.64 ppm / .06 dilution = 2110.67 ppm before dilution
Using the data given from the commercial GC, plug in the concentration into the equation given from the trend line. This gives the concentration but a dilution was done. The conversion on the right is for the dilution factor. The original concentration was before it was diluted.
The precision of this experiment was very important; building the GC is as important as regulating the gas flow rate and amount of material injected. Table 1 shows the gas flow rate calculated from the air flow times; if a different gas flow rate was used for each trial than the data from the experiment would be flawed. Another important part of the gas flow is the column length; the column length helps determine the gas flow. The gas flow rate was 5.43 cm/s, if this rate was too high the gas would not have time to separate in the glass tube. If too much material was injected the results would tail off too much therefore the right amount of material has to be injected to get correct values.
The Physical Characteristics of Freon 22 and 123 are shown in Table 2. As shown in the first three columns of the table, Freon 123 spent more time traveling through the glass column. The Freon 22 appeared quicker, but only turned the flame green for 2.07 s compared to the 22.4 s for Freon 123. The time it takes for maximum color intensity, retention times and time between appearance and disappearance, band width, were very different. The retention times and the band width are related because of the dispersion of gases, which cause the peaks to get wider
Freon 123 had a much longer retention time by almost five times. This could have been due to many different reasons. While Fr. 22 had a very low boiling point of -40.7 ̊C, Fr. 123 had a larger boiling point, 27.82 ̊C. Freon 123 also had a greater dipole moment and molecular weight. The greater molecular weight, results in stronger intermolecular forces between the stationary phase (Tide) and the sample (Fr. 22/123) causing a longer time to separate the components of the sample. The difference in intermolecular forces is clearly shown in the difference between boiling point and dipole moments. London Dispersion Forces (LDF's) were also a factor in the retention times. LDF's depend solely on the size of the molecule therefore Freon 123 will have greater LDF's than Freon 22. LDF's are created because electrons are constantly moving which produces dipole moments therefore an attraction to a nearby molecule is started. If the molecule is bigger, it will have more electrons which will increase the LDF's which will increase the dipole moment thus attracting it to other molecules more.
Table 3, shows the data collected from a combination of Freon 22 and Freon 123. The mixture consisted of .1 mL of Freon 22 and .25 mL of Freon 123. There was .25 mL of Freon 123 used because that was the amount needed to see a change of color, when used alone. The procedure was done exactly the same as when the two gases were done individually, so that the measurements stayed constant. This proved to be very helpful in confirming that there was a difference between Freon 22 and Freon 123. While Freon 22 turned the flame green at 3.0 s it only stayed for .8 seconds until it disappeared. Freon 123 followed the disappearance of Freon 22 because it only took 5.5 seconds to appear but kept the flame green-blue for 22.2 s. This experiment clearly showed that Freon 123 has stronger intermolecular forces because it takes so much longer to disappear.
The data that was collected in Table 3 was used in the equation given to find the Resolution, found in table 4.3 The resolution shows the time between the disappearance of one gas and the appearance of the second. If there was a resolution of 1, it would indicate that as soon as the first gas disappeared the second gas would appear. It would make sense that our resolution value was so close to 1 (1.06) because as soon as Freon 22 disappeared, Freon 123 appeared.
The data in Table 6 shows the concentration of the two gases that have moved across a certain length of the column. This process takes much longer than the homemade since the column is so much longer in the commercial GC. The data is used in the calibration graphs, s 1 and 2. From these graphs both trend lines and correlation values can be found. Both graphs have high R-squared values, correlation values very close to 1. R-squared values range from + 0-1, the closer the number is to + 1 the more accurate the data is.  1 has an R-squared value of .999 and 2 has a value of 1.0 thus showing that the data was very accurate.
The concentrations of the two unknown halocarbons, Table 7, were determined by plugging in the peak area into the equation given from the trend line found from the correlation graphs. The different halocarbons were identified by their retention times. The peaks on the graphs from the commercial GC were higher for the second gas, and since the retention time for Freon 123 was greater in the homemade, Freon 123 must bet the second halocarbon. However, this concentration gives the concentration after the dilution took place. As stated in the Procedure, a dilution of 6% was taken by also taking in air into the syringe as well as either Freon 22 or 123. This dilution factor is shown in the calculation; after the syringe is pulled back to 10 cc, the concentration is 30% of what it started out to be, then when it is pushed to 2 cc it becomes 20% of the original and finally 20% of the original concentration when pulled back to 10 cc.
The goal of this experiment was accomplished because Freon 123 had a longer retention time because it had a larger molecular weight and larger intermolecular forces; such as LDF's that resulted in a higher boiling point and dipole moment. Also the experiment showed that the commercial GC was more accurate than the homemade GC even though it took much longer for the reactions to take place.
 University of Virginia, Chromatography Website. http://galileo.phys.virginia.edu/outreach/8thGradeSOL/Chromatography.htm (accessed February 26, 2008).
 Scott, Raymond; Scott, Scott Introduction to Analytical Gas Chromatography, 2nd ed; Crc Press: New York, 1997.
PSU Chemtrek, “Small-Scale Experiments for General Chemistry”, Stephen Thompson, Hayden McNeil: Publishing Inc., Plymouth, MI, Aug. 2009-July 2010: Exp. 18, 19 (pg. 18-2: 19-1 - 19-21)
Collin,L. Oliver. Fast Gas Chromatography of Explosive Compounds Using a Pulsed-Discharge Electron Capture Detector.http://oak.cats.ohiou.edu/~jacksong/Research/Fast%20GC%20of%20explosive%20compounds%20using%20a%20PDECD.pdf. July 2006, Vol. 51.
 UCLA Chemistry Gas Chromatography.http://www.chem.ucla.edu/~bacher/General/30BL/gc/theory.html. Aug 06
 Peter Racioppo Lab Manual pgs 7 - 10
 Caitlyn Parry Lab Manual
 Matt Palmer Lab Manual
 Alex Bohen Lab Manual
 The Merck Index: An Encyclopedia of Chemicals, Drugs and Biologicals, 12th ed. Budavari, S; Merck & Co: Whitehouse Station, NJ, 1996.
 Printouts from the Commercial Chromatograph
 Brown Theodore L.; LeMay, H. Eugene; Bursten, Bruce E., Chemistry the Central Science, Prentice Hall, Upper Saddle River, NJ, 10th edition, 2006
 Decision 411 Forecasting: What's good value for R-squared? http://www.duke.edu/~rnau/rsquared.htm