Quantitative Determination of Caffeine in Soft Drink by HPLC
The quantity of caffeine found in a 16-oz bottle of soda was determined by using High Performance Liquid Chromatography (HPLC). A series of standard dilutions containing varying amounts of soda were synthesized and analyzed through the chromatogram. Data collected from the series of dilutions were used to produce a calibration curve, from which a linear regression equation was determined. Two soda samples of unknown caffeine concentration and unknown caffeine amount were also analyzed through the chromatogram, from which two peak areas were produced. Using the linear regression equation and the peak areas for both unknown solutions, the caffeine concentration and the amount of caffeine per 16-oz soda bottle were determined for each of the unknown. The concentrations for Unknown 1 and 2 were determined to be 0.06286 mgmL and 0.05791 mgmL , respectively. The amount of caffeine present in a 16-oz. bottle for each unknown was calculated to be 74.33 mg and 68.48 mg. The r2 value, generated from the calibration curve, was 0.9969, suggesting that the data collected was both accurate and reliable.
High Performance Liquid Chromatography is an instrument that is designed to both separate and analyze the components of a sample. Due to the nature of the instrument, HPLC requires that the analyte is in a liquid state in order to be analyzed. The instrument is connected to a computer, which illustrates the data collected from the machine. HPLC combines column chromatography and quantifiable spectroscopy in order to identify compounds in an analyte (1). Column chromatography can be utilized for separation and purification of a target compound based on its polarity (2). The polarity of the target compound in an analyte plays a role in determining what form of HPLC, forward-phase or reversed-phase, is most appropriate to use. If the compound of interest in an analyte is non-polar, forward-phase HPLC is the most appropriate option. In forward-phase HPLC, the mobile phase is non-polar, and the stationary phase is polar. Once the analyte is injected into the machine, the non-polar compound of interest will have a shorter retention time in the column compared to polar compounds in the analyte. The polar compounds prefer to stay in the column longer because they are more comfortable being surrounded by other compounds of the same polarity. If the compound of interest in an analyte is polar, such as caffeine in this experiment, reversed-phase HPLC is the most appropriate option. In reversed-phase HPLC, the mobile phase is polar, and the stationary phase is non-polar. Once the analyte, soda, is injected into the machine, the polar compound of interest, caffeine, will have a shorter retention time in the column compared to non-polar compounds in the soda. The non-polar components of the soda prefer to stay in the column longer because they are comfortable being surrounded by other compounds of similar polarity (3).
To begin, approximately 30 mg of caffeine was weighed and deposited into a dry and clean 150-mL beaker. A small amount of methanol was used to collect any residual caffeine possibly left on the weighing paper into the 150mL beaker to reduce transfer loss. Approximately 100 mL of methanol was added to the 150mL beaker to dissolve the caffeine. Once caffeine was thoroughly dissolved, the solution was carefully transferred into a 250.0-mL volumetric flask and diluted to the mark with water. After diluting the caffeine solution with water, the concentration for the caffeine solution was calculated and recorded in mgmL. The concentration determined was recorded as the concentration of the standard stock solution and labeled as C12. Approximately 300 mL of a 40%:60% solution of methanol:water was prepared and was used as the solvent for the diluted standard solutions, labeled as C3, C6, and C9 (Figure 1), along with the unknown sample of soda.
Approximately 15 mL of soda was obtained and was poured into a clean and dry 50-mL beaker. Next, a second clean and dry 50-mL beaker was obtained. The soda was decarbonated by pouring the soda back and forth between the two beakers until the bubbling ceased, meaning that the carbon dioxide that was initially present in the sample was liberated into the surrounding atmosphere. Exactly 10.00 mL of soda was pipetted into a 25.0-mL volumetric flask and diluted to the mark with the previously prepared 40%:60% methanol:water mixture, creating a diluted soda solution.
After all the dilutions were synthesized (Table 1), each solution, all dilutions and both unknowns, were extracted into a 20.0 mL syringe and injected into the chromatograph one at a time. After the first dilution was analyzed, each subsequent dilution was injected into the chromatograph only when the monitor indicated that it was ready to analyze the next sample. Each dilution standard was analyzed twice to eradicate random error and to enhance precision and reliability of the data collected. Each unknown sample was analyzed once in the chromatograph. After each solution was analyzed, the chromatogram generated a retention time and peak area for each sample.
After all of the data from the chromatogram was recorded, the average peak area was averaged based on the two analyzations of each diluted solution. A calibration curve was generated by graphing peak area against concentration in mg/mL for the series of standard dilutions. A linear regression equation as well as an r2 were obtained from the calibration curve (Graph 1). Using the linear regression formula and the average peak area calculated for the unknowns, the unknown concentration of caffeine was calculated in mgmL. Once the concentration of the unknown was calculated, the amount of caffeine, in mg, in a 16-oz. bottle of soda was determined.
The linear regression line obtained from the calibration curve was determined to be y=6×106x+31122 (Figure 3) and the r2 value generated from the calibration curve was 0.9969 (Figure 3). The chromatograph generated retention time and peak area for each of the solutions analyzed (Figure 2). The peak areas of the series of dilutions allowed the formulation of a calibration curve. The peak areas for the series of dilutions were crucial for determining the concentration of the unknown samples. The peak areas generated from the HPLC for both unknowns were applied to the linear regression equation in order to calculate the respective caffeine concentrations for each unknown. The caffeine concentration of Unknown 1 was calculated to be 0.06286 mgmL and the caffeine concentration of Unknown 2 was calculated to be 0.05791mgmL. Based on the concentrations of the unknown samples, the amount of caffeine in a 16-oz. bottle of coke was determined to be 74.33 mg and 68.48 mg for Unknown 1 and 2, respectively.
Figure 1: Standard Dilutions
Dilution Label Volume of Stock Volume of Methanol:WaterConcentration (mg/mL)
C12 250.0 mL 0 mL 0.124
C9 75.0 mL 25.0 mL 0.093
C6 50.0 mL 50.0 mL 0.062
C3 25.0 mL 75.0 mL 0.031
Figure 2: Retention Time, Peak Area, and Average Peak Area of Caffeine in Standard Solutions and Unknowns
Retention Time (min) Peak Area Average Peak Area
C3 2.27 204871 208101
2.26 211331 C6 2.26 418354 392565.5
2.26 366777 C9 2.26 613969 595098
2.26 576227 C12 2.25 770254 1493046
2.25 722792 Unknown Sample 1 2.28 408287 408287
Unknown Sample 2 2.28 378628 378628
Figure 3: Average Peak Area v. Concentration (mg/mL)
Concentration of Standard Solutions:
C12: 31mg250mL=0.124mgmLC9: 0.124mgmL ×75 mL100 mL=0.093mgmLC6: 0.124mgmL × 50 mL100 mL=0.062mgmLC3: 0.124mgmL × 25 mL100 mL=0.031mgmLConcentration of Unknown 1:
408287=6×106x+31122= 0.06286 mgmLConcentration for Unknown 2:
378628=6×106x+31122= 0.05791 mgmLUnknown 1: Amount of Caffeine in a 16-oz. Soda
0.06286 mgmL ×25 mL10 mL×473 mL= 74.33 mgUnknown 2: Amount of Caffeine in a 16-oz. Soda
0.05791 mgmL ×25 mL10 mL×473 mL= 68.48 mgDiscussion and Conclusion
The amount of caffeine in a 16-oz. bottle of soda was determined by using a series of dilution standards and analyzing them through High Performance Liquid Chromatography. The HPLC analyzed each of the dilution standards and generated peak areas for each sample. Each dilution was analyzed through the chromatograph two times in order to maximize precision and accuracy. Both unknown solution was analyzed once through the chromatogram and their peak areas were recorded. A calibration curve was created by plotting the peak areas and the concentrations of the series of dilution standards against each other. A linear regression equation was generated from the calibration curve, which was then used to identify the concentration of the unknown’s caffeine amounts by using their respective peak areas. The caffeine concentration for Unknown 1 was calculated to be 0.06286 mgmL . Concerning Unknown 1, the amount of caffeine in one 16-oz bottle of soda was determined to be 74.33 mg. The caffeine concentration for Unknown 2 was calculated to be 0.05791 mgmL. Concerning Unknown 2, the amount of caffeine in one 16-oz. bottle of soda was determined to be 68.48 mg. The r2 value, also generated from the calibration curve, was 0.9969, which suggests that the data collected is accurate and reliable. Assuming that there were no mathematical errors made, the concentrations of the unknowns and amounts of caffeine in a 16-oz. soda for both unknowns is equally as accurate. Possible sources of error could be due to the preparation of stock solutions by another group, transfer loss when pouring liquid from one beaker to another or from a beaker to a volumetric flask, and not liberating all of the carbon dioxide from the soda samples.
High Performance Liquid Chromatography: HPLC Basics, Theory, Definition. https://chemdictionary.org/hplc/ (accessed Nov 11, 2018).
Libretexts. B. Column Chromatography. https://chem.libretexts.org/Textbook_Maps/Analytical_Chemistry/Supplemental_Modules_(Analytical_Chemistry)/Instrumental_Analysis/Chromatography/V._Chromatography/B._Column_Chromatography (accessed Nov 11, 2018).
Types of HPLC. http://hplc.chem.shu.edu/NEW/HPLC_Book/Introduction/int_typs.html (accessed Nov 13, 2018).