Experiment 1: Results
Example of Student Work

A standard curve was performed to determine the calibration of the machine. The standards of potassium acid pthalate were made in the following concentrations of 0.5, 1.3. 2.0, 3.0, 4.0, 5.0, 10.0 ppm C. A stock solution of 100 ppm C was initially prepared by adding 0.2127 g of KHP to 1 Liter of ultrapure water. In order to make the appropriate dilutions respectively, 0.5, 1.3, 2.0, 3.0, 4.0, 5.0, 10.0 mL of the stock solution were added to a volumetric flask and filled to the 100 mL mark.

Standard and Sample Results

In the table below are the KHP standards and the sample TOC measurements. Included are the reading from the TOC Analyzer for the standards and samples as well as the true TOC concentration for the samples calculated from the linear regression of the standards. There were three replicate measurements for each sample (n=3).

Standards of KHP (ppm C)	TOC Machine (raw counts) Average	TOC minus blank (raw counts)	Samples (NPOC)	TOC Machine (raw counts) Average	True TOC (ppm C) Error = ±0.5
Blank	3414	0	Tap Water	22825	1.83
0.5	8928	5514	Glutamic Acid (2 ppm C)	31871	2.55
1.3	20762	17348	Salicylic Acid (5 ppm C)	61681	4.94
2.0	25800	22386	Soil Extract	39210	3.14
3.0	38122	34708	Wetland Unfiltered	43853	3.51
4.0	50212	46798	Wetland Filtered DOC	39661	3.18
5.0	68250	64836	Wetland Filtered POC	4192	0.33
10.0	127484	124070	(Unfiltered-Filtered)

Calibration Curve

The calibration curve can be seen below. Initially, the standard curve for this experiment relied on the preset calibration factor of 8.25E-05. The data was then converted to the appropriate values if the calibration factor had been set to one. According to the manual, it is required that the desired modes of analysis should be calibrated before the analysis from the following formula:

CAL factor = [ppm C standard] * [volume injected in mL]/[peak area]

With the calibration factor set to one, the sample or standard concentration equals the integrated peak area as the volume injected is 1 mL.

Statistical Assessment of the Slope Intercept

The error analysis on the slope and intercept was based on the calibration curve derived.

Slope = a = 1.03454 ± 0.0619 Intercept = b = -0.092 ± 0.292 where b was not statistically different from 0.

The errors associated with the samples based on regression were quite high relative to concentrations being measured. This demonstrates that great time and care is necessary when calibrating the machine, especially if lower concentrations are to be quantified. The standard deviations performed by the analyzer are much lower and not as accurate since the standard deviations are based on the mean of the three replicates, by the following formula, and not on the regression:

1 standard deviation = sqrt[sum(x_n-x)²/(n-1)]
where x = the mean
x_n = individual results
n = the total number of individual results

Approximately two-thirds of the individual values will be within the range of one standard deviation.

Overall, the results seem reasonable. The glutamic and salicylic concentrations were close to the desired concentrations. With this technique there appeared to be no effect on the analysis on whether the compounds were aliphatic or aromatic in nature. The Prado wetland sample was a mixture of inlet and outlet samples; thus, the value obtained seemed reasonable. If the sample just contained the effluent of the wetland, the TOC value would be expected to be much higher. Since the soil extract was a 1:2000 dilution, the TOC value was accurate. The tap water sample seemed reasonable since the water treatment plant does not attempt to remove it.

Interferences

Precaution had to be taken in cleaning the glassware in which the samples and standards were prepared and stored. Any contamination of the glassware, would affect the standard measurements, especially the blank and low level standards. The glassware was thoroughly cleaned with a dichromate-sulfuric detergent, rinsed three times with tap water, rinsed three times with distilled water, and rinsed three times with ultrapure water. In addition to contamination, degradation of the standards had to be taken into consideration. The standards had to be prepared and run in the same day otherwise the sample would degrade and the carbon concentration in the standards will decrease.

It was found that there was some carry over in the TOC Analyzer from previous samples, if the samples had a high concentration of carbon present. Therefore after the higher concentration samples (10 ppm and 5 ppm), a blank was rinsed through the analzyer twice.