Example of Student Work
The absorption spectra from 200-700 nm for the samples of glutamic acid, salicylic acid, tap water, soil extract and wetland water (original pH and pH 2) are shown in Figures 5-7.UV-Vis Absorption Spectra of Humic Substances
FIGURE 5: Prado Wetlands Spectra
FIGURE 7: Salicylic and Glutamic Acid Spectra
Table 2 lists the concentration and absorbance data used for the standard curves shown below.
KAP (ppm C) pH 4 A254 pH 2 A254 0.0 0.000 0.000 0.5 0.026 0.028 1.0 0.030 0.041 2.0 0.042 0.050 3.0 0.051 0.053 4.0 0.063 0.080 5.0 0.091 0.091 10.0 0.163 0.157
Parameters from the linear regression analysis used to generate the equations relating ppm C and A254 are tabulated in Table 3. All confidence limits were calculated using t= 2.45 at the 95% confidence interval.
pH 4 pH 2 Equation (conc in ppm C) A254 = 0.0156(conc) + 0.007 A254 = 0.0134(conc) + 0.0224 slope 0.01654 ± 0.002 0.01341 ± 0.009 intercept 0.00704 ± 0.007 0.02239 ± 0.044 Sxx 73.97 62.38 Sxy 1.16 0.84 Syy 0.02 0.02 ESD 0.006 0.030
Table 4 summarizes the A254 results for the samples at both initial pH and at pH 2. Using the KHP standard curve at the appropriate pH, the concentration corresponding to the A254 value was calculated and is compared to the value measured in the TOC analysis. These results are also shown in Table 4. The y-intercept of both equations in Table 3 was determined to be indistinguishable from zero according to the t-test at the 95% confidence interval. Therefore, the intercept was neglected when the concentrations of each sample were calculated.
samples pH A254 c (ppm C) pH A254 c (ppm C) TOC (ppm C) glutamic acid 4.2 0.030 0.9 ± 3.1 2 0.030 1.4 ± 6.5 2.0 salicylic acid 3.9 0.090 1.6 ± 1.3 2 0.095 3.2 ± 6.1 5.0 tap water 7.0 0.175 1.9 ± 3.0 2 0.198 2.2 ± 6.3 1.8 Histosol soil extract 7.3 0.014 11.2 ± 2.1 2 0.018 14.7 ± 8.9 3.1 natural water (Prado Wetlands) 8.2 0.02 5.8 ± 2.6 2 0.043 7.1 ± 6.0 3.1
Table 5 lists the E2/E3 Ratios, E4/E6 ratios and the A280 of all samples at both pH's. The percent aromaticity for the variable pH samples are also tabulated based on A280 (1). The data in Table 5 are used to make qualitative statements about the source, degree of aromaticity and molecular weight of the DOC and/or NOM in each sample and are discussed in the next section.
variable pH pH 2 samples E2/E3 E4/E6 A280 % Aro. E2/E3 E4/E6 A280 glutamic acid 1.73 1.99 0.089 68.6 % 2.13 2.09 0.118 salicylic acid 2.03 2.04 0.196 79.4 % 2.59 2.02 0.199 tap water 1.19 1.97 0.102 38.0 % 2.28 2.00 0.480 Histosol soil extract 2.68 2.72 0.226 18.0 % 2.38 2.69 0.256 natural water (Prado Wetlands) 2.52 2.07 0.147 22.0 % 2.75 2.08 0.165 4 ppm KAP 2.80 1.99 0.129 34.0 % 3.22 2.04 0.184
Humic substance make up a large portion of NOM, and can be divided into two operational groups based on sedation procedure: fulvic acids and humic acids (Malcolm, 1985). Fulvic acids are the fraction that is soluble in both acidic and basic solution, where as humic acids are nonsoluable in acid solution, but are soluble at a higher pH. Fulvic acids have, in general, a lower molecular weight and a higher percentage of carboxyl and hydroxyl groups than the humic acid fraction. The humic acid fraction is usually more aromatic, and hydrophobic than that fulvic fraction. Both fractions tend to be complex mixtures of carbohydrates, aliphatic chains, polyphenols, carboxy benzene and other simple organic acids.Sample Concentrations
The UV-Vis absorption spectra of humic substances is usually featureless with absorption increasing at lower wavelengths due to the overlapping absorption spectra of the many functional groups in NOM (4). This trend is observed for the soil and wetland spectra shown in Figure 4 and 5. Inspection of these spectra also show that near maximum absorbance is measured at around 254 nm making it a good wavelength for applying Beers law.
UV-Vis spectra can be used to detect for the presence of absorbing functional groups or chromophores. The UV-Vis spectra of aromatics are defined by a sharp peak called the B-band at 256 - 312 nm due to the excitation of pi electrons in the benzene ring. The B-band can be seen in Figure 6 and 7 for the 4 ppm C KHP, glutamic acid and salicylic acid where there is a sharp peak just less than 300 nm. As shown, KAP and salicylic acid both contain substituted benzene rings. The presence of the COO- group on KHP may destablize the benzene ring, and raise the energy level of the pi electrons. Thus, the absorptivity of the KHP B-band is less than that of the salicylic acid. Also, the presence of a second carboxyl group in KHP makes the peak less pronounced as there may be overlap between the absorption of energy by electrons in each double bond. Glutamic acid has two carboxyl groups, and also unbonded electrons associated with the amino group. These groups absorb weakly in the 200-400 nm range (8).
The addition of acid raises the absorption for all samples. Both Edzwald et al. (5) and Chen et. al. (1) found that absorption of humic substance was affected by pH. Humic substances, salicylic and glutamic acid all contain carboxylic, and other acid functional groups. The structures of these compounds can vary with pH. The unbonded electrons associated with the deprotonated functional groups tend to stabilize the pi electrons lowering the energy necessary for transition. When protonated, the stabilizing effect is absent, and the transition takes more energy the acidified samples have a higher absorbance. Since both the soil extract and wetland water contain humic substances that have acid/base properties, the trend of higher absorbance at pH 2 is observed in Figure 4 and 5. However, other references have reported different effects of measuring absorbance at pH 2. Edzwald et. al. (5) reports that lowering the pH decreases the absorptivity. Chen et. al. (1) found that changes in pH did affect A254 of humic and fulvic substances isolated from soil samples. They hypothesized that the change was due to compaction or expansion of the average molecule size, which implies a scattering effect.
Sample concentrations are shown in Table 4. The data from the non-acidified samples will be used in this discussion as the most consistent results for TOC analysis by UV-Vis absorbance have been obtained using samples at pH 5-7 (4, 5). There are significant error associated with the concentrations determined from the standard curve of KHP (i.e. 7.1 ± 6.0 ppm for wetland water). The spectra for KHP in Figure 6 shows that the absorbance is steadily increasing at 254 nm. Therefore, using the absorbance of KHP at 254 nm for a standard curve results in error since the absorbance will change over the narrow wavelength range measurement by the spectrophotometer. Increasing the sophistication of the instrument so that it measures an even narrower wavelength range should decrease this error.E4/E6 ratios, E2/E3 ratios and % Aromaticity
The TOC calculated for the glutamic and salicylic acid samples is much lower than the target 2 ppm C and 5 ppm C, respectively. Both Edzwald et. al. and Dobbs et. al. determined that UV-Vis absorbance could be used to determine the TOC of an unknown sample provided that an appropriate standard curve was created. Both papers used a catalytic combustion TOC analyzer to verify the actual TOC of their samples. However, uncertainties arise when the standard curve is prepared with a substance other than the actual analyte. It seems that using KHP as a standard for salicylic acid would be appropriate since the structure are similar, however, the salicylic acid spectra in Figure 4.2 shows that the absorption dips around 254 nm. Thus, concentrations determined from absorbance data at this wavelength will underestimate the true concentration.
Typical values for TOC in river water are 4-7.5 ppm (5), and 4.5-7.7 ppm (4). Both sources had a large amount of NOM. The first source was from a river high in humic substances and had a strong brown color. The second source was from a river polluted by agricultural waste. Thus, they can be used as a valid comparison to the wetland water sample which has a concentration determined to be 5.8 ppm C. Both the wetland sample and the soil extract sample with 11.2 ppmC are have a greater DOC than tap water with 1.9 ppm C.
In general, the concentration of NOM containing a large percentage of aromatics will be overestimated by absorbance measurements at 254 nm. Similarly, the concentration of NOM containing mainly caboxylic acid functional groups will be underestimated. This trend is caused by the strong absorbance of aromatics and weak absorbance of carboxylic acids at this wavelength.
E4/E6 ratios and E2/E3 ratios are determined by sample absorbances at 254, 365, 465 and 665 nm. E4/E6 ratios have been used by soil scientists to indicate aromaticity: a high E4/E6 ratio corresponded to a low aromaticity and vice versa. However, Chen et. al. found no relation between E4/E6 ratios and the degree of aromaticity (1). Instead, they determined E4/E6 ratios varied inversely with molecular weight for soil samples. They also found that E4/E6 ratios were not concentration dependent and were correlated with %C, % O, acidity and amount of COOH groups. E4/E6 ratios for humic acids have been determined to be in the 5.44-5.7 range and for fulvic acids in the 8.88-9.9 range (1). Kukkonen also found E4/E6 values for humic and fulvic acids from a wide variety of sources to be in the 3.8-5.8 and 7.6-11.5 range respectively (7). Kukkonen et. al. found that fulvic acids from strongly colored water has E2/E3 ratios around 4.
The E4/E6 and E2/E3 ratios for our samples (1.96-2.71) are much lower than those either from either of the references. Chin et. al point out that there can be error in making reliable absorbance measurements at these longer wavelengths (2). According to our results, the soil sample would have the lowest molecular weight . However, soil extract should have a high amount of lignin from decaying plant material resulting in a higher molecular weight than the other water samples. There seems to be some disagreement as to how E4/E6 ratios are correlated with molecular weight as other sources determined E4/E6 ratios to be positively correlated with molecular weight (7).
The A280 for all samples is shown in Table 5. Chin et. al used A280 to as an indicator of % aromaticity of soil samples based on moles of carbon (2). Using this data to calculate the absorbtivity (e), the % aromaticity was determined according to, % aromaticity = 0.05e + 6.74. This relationship was generated by correlating NMR data for the % aromaticity with molar absorptivities at 280 nm (2). The % aromaticity results for our samples seem somewhat consistent. Salicylic acid has the highest % aromaticity, and also has the highest number of carbon molecules in a benzene ring. The soil, wetland and tap water sources have the lowest % aromaticity indicating NOM with a large fulvic acid character. The high % aromaticity of the glutamic acid sample is unexplainable except that the correlation used was developed for soil samples containing fulvic and humic acids. It may not be applicable to amino acids such as glutamic acid. It seems that any experimental correlations, such as this one for aromaticity, are purely empirical and only quantitatively valid for the samples used to make correlations.