**3. Results and discussion**

#### **3.1. Characterization of the sorbent**

The spectrum in the infrared region of the material PUF-Me-BTAP (Figure 2) shows that the absorptions in the range of 3600-3300 cm-1, centered at 3448 cm-1 and 3358 cm-1, can be attributed to the stretches of the -OH and-NH groups, respectively. The bands between 2970 and 2920 cm-1 are characteristic of the aliphatic part (-CH2 and CH3) of the sorbent structure. The peaks between 1500 and 1448 cm-1 are characteristic vibrations of the -CS- group, related to the segment molecular organic reagent for the Me-BTAP. The absorption at 1708 cm-1 was assigned to the axial deformation of the urethane carbonyl group conjugated by hydrogen bonding. There were no bands featuring free –NCO of urethane groups (1730-1720 cm-1) (Radhakrishnan Nair, 2008)]. Solubility tests showed that the sorbent is insoluble in the following solvents: chloroform, methanol, ethanol, tetrahydrofuran, benzene, toluene, acetone, diethyl ether, isopropanol, dioxane and acetic acid. In the presence of pyridine and dimethyl sulfoxide, the material swelled. Briefly, the particles formed a gel with a volume greater than 3.2 times the initial volume.

**Figure 2.** Infrared spectrum of PUF-Me-BTAP.

The graphs of TGA for Me-PUF BTAP in N2 (Figure 3) and O2 (Figure 4) show that there are no significant differences related to the atmosphere used during the degradation process for the amount of steps, temperature ranges and loss of mass.

268 Polyurethane

(absorbance).

**2.5. Procedure for preconcentration** 

**3. Results and discussion** 

**3.1. Characterization of the sorbent** 

greater than 3.2 times the initial volume.

**Figure 2.** Infrared spectrum of PUF-Me-BTAP.

Solutions containing Cd and Pb were adjusted to pH 7.5 with borate buffer. These solutions were passed through the sorbent minicolumn that contained the PUF-Me-BTAP. At this stage, the elements are sorbed onto the solid phase. After preconcentration (120 s), the position of the six-port valve was changed, and an eluent flow was passed through the minicolumn. The eluent transported Cd (II) or Pb (II) to the nebulizer of the flame atomic absorption spectrometer. The analytical signal was then measured as the peak height

The spectrum in the infrared region of the material PUF-Me-BTAP (Figure 2) shows that the absorptions in the range of 3600-3300 cm-1, centered at 3448 cm-1 and 3358 cm-1, can be attributed to the stretches of the -OH and-NH groups, respectively. The bands between 2970 and 2920 cm-1 are characteristic of the aliphatic part (-CH2 and CH3) of the sorbent structure. The peaks between 1500 and 1448 cm-1 are characteristic vibrations of the -CS- group, related to the segment molecular organic reagent for the Me-BTAP. The absorption at 1708 cm-1 was assigned to the axial deformation of the urethane carbonyl group conjugated by hydrogen bonding. There were no bands featuring free –NCO of urethane groups (1730-1720 cm-1) (Radhakrishnan Nair, 2008)]. Solubility tests showed that the sorbent is insoluble in the following solvents: chloroform, methanol, ethanol, tetrahydrofuran, benzene, toluene, acetone, diethyl ether, isopropanol, dioxane and acetic acid. In the presence of pyridine and dimethyl sulfoxide, the material swelled. Briefly, the particles formed a gel with a volume

**Figure 3.** Thermogravimetric curve of the material PUF-Me BTAP under an N2 environment.

**Figure 4.** Thermogravimetric curve of the material PUF-Me BTAP under an O2 environment.

The degradation of the material occurs in two general stages. In Stage I, the degradation is mainly due to the decomposition of rigid segments and involves the dissociation of urethane and the original chain extender, which then form primary amines, alkenes and carbon dioxide. Stage I is influenced by the amount of rigid segments. In the subsequent stage II, depolymerization and degradation of the polyol occur. Therefore, this stage is affected by the content of flexible segments. According to Figure 5, there is a maximum of degradation, indicated by the first derivative curve of DTA at 364 C in air (O2). The first stage of degradation occurred concomitantly with a phase transition, possibly because part of the polymer changed from a crystalline to an amorphous phase. An important observation is that the sorbent Me-BTAP-PUF has a high thermal stability at the final temperature for the first stage of degradation in air at 280 C. The observed mass reduction, which occurred between 0 and 120 C, was attributed to loss of water.

Synthesis of a New Sorbent Based on Grafted PUF

Cd

Pb

for the Application in the Solid Phase Extraction of Cadmium and Lead 271

**Figure 6.** Influence of pH on the determination of Cd and Pb using solid phase extraction.

**3.3. Type and concentration of the eluent** 

experiments.

0.000

0.020

0.040

Absorbance

0.060

0.080

0.100

According to Figure 6, the best pH range for the extraction of cadmium is between 6.8 and 8.2. The extraction of lead is maximal when performed at pH values between 7.0 and 7.8. Thus, the extraction of both metals was performed at pH 7.5 in all subsequent experiments.

4.5 5.5 6.5 7.5 8.5 9.5 10.5 pH

Polyurethane may be dissolved by concentrated sulfuric acid or oxidized by concentrated nitric acid and potassium permanganate solutions. This material was resistant to solvents such as water, hydrochloric acid (up to 6 mol L-1), ethanol and glacial acetic acid (Navratil et al., 1985). Thus, HCl was chosen to prevent a reduction in the lifetime of the PUF that was grafted with Me-BTAP. PUF-Me-BTAP is resistant to ethanol. However, the use of this solvent in the elution of cadmium and lead presented pressure problems in the on-line system in this work. When ethanol was used as the eluent, there was a swelling of the sorbent, which caused backpressure on the minicolumn. This increase in pressure caused leaks throughout the system on-line. Thus, the use of this solvent was discontinued. Hydrochloric acid solutions were then used as the eluent in all further

Figure 7 illustrates the phenomenon of desorption of cations from the solid phase when the concentration of HCl is varied. It was observed that the hydrochloric acid solutions that provided the highest analytical signals were those at concentrations ranging between 0.01 and 1.00 mol L-1 (Cd) and 0.10 and 1.00 mol L-1 (Pb). The use of low concentrations of acid is beneficial because it can increase the lifetime of the minicolumn. Moreover, in this work, we chose to use an eluent of identical concentration for both metals, with the aim of simplifying the operation of the on-line system. Therefore, a solution concentration of 0.10 mol L-1 for the

desorption of both chemical species was used in all subsequent experiments.

**Figure 5.** DTA curve of the material PUF-Me BTAP in an O2 environment.

#### **3.2. Effect of pH**

Many complexing agents are Lewis bases (capable of donating electron pairs) and Brönsted bases (capable of receiving protons) and, as such, will be affected by changes in pH. The reaction for the formation of the chelate is influenced by pH because the chelating agent is not presented entirely in the form of a free ion. Thus, the effect of hydrogen concentration was studied to observe the pH range over which the cations cadmium and lead are absorbed by PUF-Me-BTAP. Figure 6 shows the influence of pH on the extraction of cadmium and lead by PUF-Me-BTAP.

**Figure 6.** Influence of pH on the determination of Cd and Pb using solid phase extraction.

According to Figure 6, the best pH range for the extraction of cadmium is between 6.8 and 8.2. The extraction of lead is maximal when performed at pH values between 7.0 and 7.8. Thus, the extraction of both metals was performed at pH 7.5 in all subsequent experiments.

#### **3.3. Type and concentration of the eluent**

270 Polyurethane

The degradation of the material occurs in two general stages. In Stage I, the degradation is mainly due to the decomposition of rigid segments and involves the dissociation of urethane and the original chain extender, which then form primary amines, alkenes and carbon dioxide. Stage I is influenced by the amount of rigid segments. In the subsequent stage II, depolymerization and degradation of the polyol occur. Therefore, this stage is affected by the content of flexible segments. According to Figure 5, there is a maximum of degradation, indicated by the first derivative curve of DTA at 364 C in air (O2). The first stage of degradation occurred concomitantly with a phase transition, possibly because part of the polymer changed from a crystalline to an amorphous phase. An important observation is that the sorbent Me-BTAP-PUF has a high thermal stability at the final temperature for the first stage of degradation in air at 280 C. The observed mass reduction,

which occurred between 0 and 120 C, was attributed to loss of water.

**Figure 5.** DTA curve of the material PUF-Me BTAP in an O2 environment.

Many complexing agents are Lewis bases (capable of donating electron pairs) and Brönsted bases (capable of receiving protons) and, as such, will be affected by changes in pH. The reaction for the formation of the chelate is influenced by pH because the chelating agent is not presented entirely in the form of a free ion. Thus, the effect of hydrogen concentration was studied to observe the pH range over which the cations cadmium and lead are absorbed by PUF-Me-BTAP. Figure 6 shows the influence of pH on the extraction of cadmium and

0 200 400 600 800

C)

Temperature (o

**3.2. Effect of pH** 

DTA (uV)


lead by PUF-Me-BTAP.

Polyurethane may be dissolved by concentrated sulfuric acid or oxidized by concentrated nitric acid and potassium permanganate solutions. This material was resistant to solvents such as water, hydrochloric acid (up to 6 mol L-1), ethanol and glacial acetic acid (Navratil et al., 1985). Thus, HCl was chosen to prevent a reduction in the lifetime of the PUF that was grafted with Me-BTAP. PUF-Me-BTAP is resistant to ethanol. However, the use of this solvent in the elution of cadmium and lead presented pressure problems in the on-line system in this work. When ethanol was used as the eluent, there was a swelling of the sorbent, which caused backpressure on the minicolumn. This increase in pressure caused leaks throughout the system on-line. Thus, the use of this solvent was discontinued. Hydrochloric acid solutions were then used as the eluent in all further experiments.

Figure 7 illustrates the phenomenon of desorption of cations from the solid phase when the concentration of HCl is varied. It was observed that the hydrochloric acid solutions that provided the highest analytical signals were those at concentrations ranging between 0.01 and 1.00 mol L-1 (Cd) and 0.10 and 1.00 mol L-1 (Pb). The use of low concentrations of acid is beneficial because it can increase the lifetime of the minicolumn. Moreover, in this work, we chose to use an eluent of identical concentration for both metals, with the aim of simplifying the operation of the on-line system. Therefore, a solution concentration of 0.10 mol L-1 for the desorption of both chemical species was used in all subsequent experiments.

Synthesis of a New Sorbent Based on Grafted PUF

for the Application in the Solid Phase Extraction of Cadmium and Lead 273

If the flow rate of the metal solution is too high, there is a possibility that the metal ions can pass through the minicolumn at a speed so quickly that a portion of the analyte passes through without being sorbed. Conversely, an excessively low flow rate of the metal solution can also cause problems with the analytical signal. A solution that passes through the minicolumn with a very low flow rate can result in leaching of the complexed species and significantly increase the analysis time. Considering the curve that corresponds to the lead solution, we observed a similar behavior to that of cadmium. However, the decrease in the amount of extracted metal was smoother. The range of flow rate that produces the maximum extraction of lead was between 5.5 and 7.4 mL min-1. Based on these results, flow rates of 4.5 and 7.0 mL min-1 were used in further experiments for solutions of cadmium and

The inconsistency between the rate of aspiration of the nebulizer of the spectrometer and the flow rate of eluent of the on-line system could result in peak broadening of the analytical signal. This broadening will result in a decrease in the analytical signal (Lemos et al., 2007). Thus, the flow of the eluent for desorption of cadmium and lead ions was adjusted to 8.0 ml min-1 to match the flow rates of elution and aspiration of the nebulizer of the spectrometer.

A linear relationship between preconcentration time and analytical signal is dependent on the flow of metal solution and the mass of sorbent. The graph in Figure 9 shows the variation of the analytical signal when the sample is inserted into the on-line preconcentration system at various time intervals. It is observed that the analytical signal is linear for preconcentration periods up to 210 and 120 seconds for cadmium and lead,

**Figure 9.** Influence of preconcentration time on the determination of Cd and Pb using solid phase

20 60 100 140 180 220 260 300 Preconcentration time (s)

lead, respectively.

respectively.

extraction.

Absorbance

0.000

0.050

0.100

0.150

0.200

Cd

Pb

0.250

**3.5. Preconcentration time** 

**Figure 7.** Influence of eluent concentration on the determination of Cd and Pb using solid phase extraction.

#### **3.4. Flow rate of solutions**

In on-line preconcentration systems, it is crucial to study the flow of the sample to meet the appropriate speed at which the ions pass through the minicolumn. The results of the influence of flow rate in on-line preconcentration systems of Cd (II), shown in Figure 8, show that the extraction is maximal when the flow rate ranges between 3.3 and 4.6 mL min-1. Values outside this range cause a decrease in the analytical signal.

**Figure 8.** Influence of flow rate of the Cd and Pb solutions for the determination of elements using solid phase extraction.

If the flow rate of the metal solution is too high, there is a possibility that the metal ions can pass through the minicolumn at a speed so quickly that a portion of the analyte passes through without being sorbed. Conversely, an excessively low flow rate of the metal solution can also cause problems with the analytical signal. A solution that passes through the minicolumn with a very low flow rate can result in leaching of the complexed species and significantly increase the analysis time. Considering the curve that corresponds to the lead solution, we observed a similar behavior to that of cadmium. However, the decrease in the amount of extracted metal was smoother. The range of flow rate that produces the maximum extraction of lead was between 5.5 and 7.4 mL min-1. Based on these results, flow rates of 4.5 and 7.0 mL min-1 were used in further experiments for solutions of cadmium and lead, respectively.

The inconsistency between the rate of aspiration of the nebulizer of the spectrometer and the flow rate of eluent of the on-line system could result in peak broadening of the analytical signal. This broadening will result in a decrease in the analytical signal (Lemos et al., 2007). Thus, the flow of the eluent for desorption of cadmium and lead ions was adjusted to 8.0 ml min-1 to match the flow rates of elution and aspiration of the nebulizer of the spectrometer.

#### **3.5. Preconcentration time**

272 Polyurethane

extraction.

phase extraction.

0.000

0.020

0.040

0.060

Absorbance

0.080

0.100

0.120

**3.4. Flow rate of solutions** 

0.000

0.020

0.040

0.060

Absorbance

0.080

0.100

0.120

**Figure 7.** Influence of eluent concentration on the determination of Cd and Pb using solid phase

min-1. Values outside this range cause a decrease in the analytical signal.

In on-line preconcentration systems, it is crucial to study the flow of the sample to meet the appropriate speed at which the ions pass through the minicolumn. The results of the influence of flow rate in on-line preconcentration systems of Cd (II), shown in Figure 8, show that the extraction is maximal when the flow rate ranges between 3.3 and 4.6 mL

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00

Cd

Pb

Cd Pb

Concentration of eluent (mol L-1)

**Figure 8.** Influence of flow rate of the Cd and Pb solutions for the determination of elements using solid

0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 Flow rate (mL min-1)

A linear relationship between preconcentration time and analytical signal is dependent on the flow of metal solution and the mass of sorbent. The graph in Figure 9 shows the variation of the analytical signal when the sample is inserted into the on-line preconcentration system at various time intervals. It is observed that the analytical signal is linear for preconcentration periods up to 210 and 120 seconds for cadmium and lead, respectively.

**Figure 9.** Influence of preconcentration time on the determination of Cd and Pb using solid phase extraction.
