Anal. Chem. 1997, 69, 1766-1770
Thermospectrophotometric Analysis of Alkylamines Utilizing Ion Association with Tetrabromophenolphthalein Ethyl Ester Tadao Sakai,*,† Shio Watanabe,† and Shunzo Yamamoto‡
Department of Applied Chemistry, Aichi Institute of Technology, Yachigusa, Yakusa-cho, Toyota 470-03, Japan, and Department of Chemistry, Faculty of Science, Okayama University, Tsushima, Okayama 700, Japan
Primary, secondary, and tertiary alkylamines react with tetrabromophenolphthalein ethyl ester dye (HTBPE) to form reddish charge transfer (CT) complexes in 1,2dichloroethane (1,2-DCE). Absorption maxima of the complexes with primary amines occur at ∼560 nm, with secondary amines at ∼570 nm and tertiary ones at ∼580 nm. CT complex formation constants between amines and HTBPE in 1,2-DCE at 25 °C increase in the order of primary, secondary, and tertiary amines but decrease with an increase in temperature. This phenomenon (thermochromism), which depends on the basicity of amines, was thermodynamically studied and used for the selective analysis of alkylamines. For example, tri-n-butylamine could be determined without interference from n-butylamine. Aliphatic alkylamines are widely used as industrial and pharmaceutical chemicals. Recently, alkylamines such as mono-, di-, and tri-n-propylamine were found to act as chemiluminescent reducing agents.1 Tatsuzawa et al. proposed the determination of aromatic and bulky tertiary amines with Bromophenol Blue.2 Also, Irving and Markham reported the extraction-spectrophotometric determination of long-chain tertiary alkylamines using Bromocresol Green.3 Since these methods were not sensitive, the utilization of potassium tetrabromophenolphthalein ethyl ester (KTBPE), which has a molar absorptivity of 105 L mol-1 cm-1, was investigated for the analyses of trace amounts of quaternary ammonium salts.4,5 However, the selectivity of KTBPE was inferior because it reacted with both quaternary ammonium salts and aromatic tertiary amines to form deeply colored ion associates in the organic solvent. To enhance selectivity while retaining the large molar absorptivity, we have taken advantage of the color changes (thermochromism) of charge transfer (CT) complexes with temperature changes.6,7 Other approaches, such as liquid chromatography of aliphatic trialkylamines8 and electrochemiluminescence detection of primary amines,9 have also been reported. However, there are few reports on spectrophotometric †
Aichi Institute of Technology. Okayama University. (1) Uchikura, K.; Kirisawa, M. Anal. Sci. 1991, 7, 803. (2) Tatsuzawa, M.; Nakayama, S.; Okawara, A. Bunseki Kagaku 1970, 19, 761. (3) Irving, H. M. N. H.; Markham, J. J. Anal. Chim. Acta 1976, 39, 7. (4) Sakai, T. Bunseki Kagaku 1975, 24, 135. (5) Sakai, T.; Hara, I.; Tsubouchi, M. Chem. Pharm. Bull. 1977, 25, 2451. (6) Sakai, T.; Ohno, N. Analyst 1982, 107, 640. (7) Sakai, T.; Ohno, N. Talanta 1986, 33, 415. (8) Noffsinger, J. B.; Danielson, N. D. J. Chromatogr. 1987, 387, 520. (9) Uchikura, K.; Kirisawa, M.; Sugi, A. Anal. Sci. 1993, 9, 121. ‡
1766 Analytical Chemistry, Vol. 69, No. 9, May 1, 1997
analysis of alkylamines. Although Naga et al. determined ionpair formation constants of alkylamines with Methyl Orange,10 this method is not applicable to the determination of alkylamines. In this paper, we discuss CT complex formation constants of primary, secondary, and tertiary alkylamines with the tetrabromophenolphthalein ethyl ester molecule (HTBPE) and describe the selective determination of alkylamines utilizing the thermochemical characteristics of CT complexes. EXPERIMENTAL SECTION Reagents. Primary amines (n-propylamine, n-butylamine, n-amylamine, n-heptylamine, n-octylamine, n-decylamine, isopropylamine, isobutylamine, and isoamylamine), secondary amines (din-propylamine, di-n-butylamine, di-n-amylamine, di-n-hexylamine, di-n-octylamine, diisopropylamine, diisobutylamine, and diisoamylamine), and tertiary amines (triethylamine, tri-n-propylamine, trin-butylamine, tri-n-amylamine, tri-n-hexylamine, tri-n-heptylamine, tri-n-octylamine, triisobutylamine, and triisoamylamine) supplied from Tokyo Kasei Kogyo Co. (Tokyo, Japan) were used without further purification. Stock solutions of amines in 1,2-dichloroethane (1,2-DCE) were prepared with concentrations of 1 × 10-21 × 10-3 mol L-1. Tetrabromophenolphthalein ethyl ester (HTBPE) solution was prepared as follows. A 4 × 10-3 mol L-1 solution of KTBPE was prepared by dissolving 0.280 g of potassium tetrabromophenolphthalein ethyl ester salt (KTBPE) (Nacalai Tesque, Osaka, Japan) in 100 mL of hot ethanol. After this, 50 mL of 4 × 10-3 mol L-1 KTBPE ethanol solution, 50 mL of distilled water, and 3 drops of 3 mol L-1 sulfuric acid were mixed with 100 mL of 1,2-DCE in a 200 mL separating funnel and shaken for a few minutes. The organic phase was filtered through a filter paper, giving 2 × 10-3 mol L-1 HTBPE-DCE solution. TBPE anion was completely extracted as HTBPE (yellow) below pH 3 into 1,2-DCE. Apparatus. A Hitachi U-3000 double-beam spectrophotometer with a temperature-controlled cell holder was used for absorbance measurements. Cell temperature was controlled by circulating thermocontrolled water delivered by Lab Thermo Cool Model LCH-10 (Advantec Toyo Kaisha, Ltd., Tokyo, Japan). The actual temperature of the solution for absorbance measurements was confirmed by a thermometer (Anritsu, Model HP-4F, Tokyo, Japan). Pyrex cells with Teflon stoppers were used for measurements. Measurement of CT Complex Formation Constants of Amine-HTBPE. To 10 mL of 6 × 10-5 mol L-1 HTBPE-DCE solution was added 0-1.5 mL each of amine. The mixture was (10) Naga, S.; Shioya, H.; Koike, H. Anal. Sci. 1995, 11, 113. S0003-2700(96)00690-7 CCC: $14.00
© 1997 American Chemical Society
Table 1. Charge Transfer Complex Formation Constants (log Kassoc) of Amine-HTBPE at 20, 35, and 45 °Ca log Kassoc ( SD
Figure 1. Absorption spectra of CT complexes with increasing n-amylamine concentration. HTBPE concentration: 2.4 × 10-5 mol L-1. n-Amylamine concentration: 1, none; 2, 8 × 10-5 mol L-1; 3, 1.6 × 10-4 mol L-1; 4, 3.2 × 10-4 mol L-1; and 5, 6 × 10-4 mol L-1. Reference: water. Temperature: 20 °C.
diluted with 1,2-DCE to 25 mL. After the temperature in the cells was controlled to 20, 35, and 45 °C, absorption spectra were recorded, and absorbance was measured at 410 nm. Selective Analysis of Secondary and/or Tertiary Amines in the Presence of Primary Amines. To 1 mL of a 2.5 × 10-4 mol L-1 secondary amine in a 25 mL calibrated flask was pipetted 0.5-4 mL of a 2.5 × 10-4 mol L-1 primary amine solution, and 7 mL of a 2 × 10-3 mol L-1 HTBPE solution was added. The mixture was diluted with 1,2-DCE, and absorbances were measured at 20 and 60 °C. For the determination of tertiary amine in the presence of primary amine, 4 mL of 2 × 10-3 mol L-1 HTBPE solution and 2 mL of 2.5 × 10-4 mol L-1 of tertiary amine solution were added. RESULTS AND DISCUSSION Determination of CT Complex Formation Constants of Amine-HTBPE Complexes. HTBPE reacted with n-amylamine (RNH2) in 1,2-DCE to form red-violet RNH2‚HTBPE CT complexes. When the n-amylamine concentration was varied in the range of (0-6) × 10-4 mol L-1, absorbance at 561 nm increased quantitatively (Figure 1). In contrast, absorbance due to HTBPE at 410 nm decreased. An isosbestic point occurred at ∼480 nm. Consequently, the CT complex formation in DCE is as follows:
RNH2 + HTBPE a RNH2‚HTBPE yellow colorless red-violet
(1)
The complex formation constant, Kassoc, is defined as
Kassoc ) [RNH2‚HTBPE]/[RNH2]R[HTBPE]R
(2)
[RNH2]R and [HTBPE]R refer to the remaining concentrations of RNH2 and HTBPE after formation of the red CT complex. [HTBPE]R can be calculated as follows:
[HTBPE]R ) A410nm/HTBPE,410nm
(3)
HTBPE obtained for 2.4 × 10-5 mol L-1 HTBPE-DCE solution was 27 500 L mol-1 cm-1, and A410nm is the absorbance of HTBPE at 410 nm. Conversely, [RNH2‚HTBPE] can be obtained as follows:
amine
20 °C
35 °C
45 °C
n-propylamine n-butylamine n-amylamine n-heptylamine n-octylamine di-n-propylamine di-n-butylamine di-n-amylamine di-n-octylamine tri-n-propylamine tri-n-butylamine tri-n-amylamine tri-n-heptylamine tri-n-octylamine isobutylamine diisobutylamine triisobutylamine
3.20 ( 0.03 3.26 ( 0.06 3.34 ( 0.03 3.45 ( 0.01 3.43 ( 0.01 4.44 ( 0.02 4.70 ( 0.03 4.61 ( 0.03 4.73 ( 0.04 4.95 ( 0.03 5.32 ( 0.09 5.12 ( 0.26 5.51 ( 0.12 5.16 ( 0.12 3.01 ( 0.01 3.93 ( 0.06 1.95 ( 0.07
2.76 ( 0.08 2.69 ( 0.04 2.79 ( 0.02 2.88 ( 0.07 2.84 ( 0.06 3.88 ( 0.03 4.12 ( 0.04 4.08 ( 0.02 4.18 ( 0.02 4.37 ( 0.02 4.68 ( 0.08 4.66 ( 0.10 4.81 ( 0.08 4.67 ( 0.06 2.54 ( 0.07 3.24 ( 0.04 1.56 ( 0.08
2.32 ( 0.02 2.38 ( 0.04 2.46 ( 0.02 2.68 ( 0.09 2.40 ( 0.05 3.64 ( 0.08 3.78 ( 0.14 3.72 ( 0.04 3.83 ( 0.03 4.05 ( 0.01 4.40 ( 0.05 4.30 ( 0.08 4.49 ( 0.02 4.38 ( 0.07 2.20 ( 0.02 2.91 ( 0.13 1.35 ( 0.06
a HTBPE, 2.4 × 10-5 mol L-1. SD, standard deviations for three determinations.
[RNH2‚HTBPE] ) [HTBPE]i - [HTBPE]R
(4)
where [HTBPE]i represents the initial concentration of HTBPE before reaction. [RNH2]R can be calculated from eq 5:
[RNH2]R ) [RNH2]i - [RNH2‚HTBPE]
(5)
By substituting eqs 3, 4, and 5 into eq 2, the complex formation constant can be obtained. Table 1 shows logarithms of the CT complex formation constants (log Kassoc) of amine-HTBPE associates at 20, 35, and 45 °C. The formation constants increased in the order of primary, secondary, and tertiary amines at 20 °C but were not dependent on the number of carbons. In iso-type amines, the formation constants as shown in the isobutylamine series were smaller than those of the normal butylamine series; the log Kassoc of triisobutylamine was particularly small. It is assumed that the hydrogen bond between the nitrogen atom in the amine and the hydroxyl group in the HTBPE molecule, which contributes to the formation of the red CT complexes, was prevented by steric hindrance. In addition, the formation constants of all amines decreased with elevating temperature, but their variation was inversely proportional in each group. Absorption Spectral Characteristics of CT Complexes with HTBPE. Table 2 shows absorption maxima for the amineHTBPE complexes investigated in this study. The absorption maxima for primary amines occurred in the range of 560-563 nm, while those for secondary amines were near 572 nm. For tertiary amines, absorption maxima occurred near 580 nm. Thus, a red shift was observed in the order of primary, secondary, and tertiary amine families. It is inferred that the nitrogen atom in amine attracts the hydrogen atom of the phenol group in the TBPE molecule and that the bonding distance from the TBPE molecule affects color development and the spectral band shift, as shown in Chart 1a. We speculate that the hydrogen atom is attracted to tertiary amines because of their higher basicity than other amines in the Analytical Chemistry, Vol. 69, No. 9, May 1, 1997
1767
Table 2. Absorption Maxima of CT Complexes of Primary, Secondary, and Tertiary Amines with HTBPE amine
λmax (nm)
amine
λmax (nm)
n-propylamine n-butylamine n-amylamine n-heptylamine n-octylamine di-n-propylamine di-n-butylamine di-n-amylamine di-n-octylamine
561 563 561 562 560 572 574 571 572
tri-n-propylamine tri-n-butylamine tri-n-amylamine tri-n-heptylamine tri-n-octylamine isobutylamine diisobutylamine triisobutylamine
579 579 579 580 580 561 569 604
Chart 1. Electronic Structures of Tri-n-butylamine-HTBPE Complex and Proton Sponge-TBPE Ion Associate
Figure 2. Changes in absorption spectra of di-n-butylamine and the mixture of di-n-butylamine + n-butylamine complexes with HTBPE at 20 and 60 °C. 1,1′: 1 × 10-5 mol L-1 di-n-butylamine. 2,2′: 1 × 10-5 mol L-1 di-n-butylamine + 5 × 10-5 mol L-1 n-butylamine. 3,3′: 1 × 10-5 mol L-1 di-n-butylamine + 1 × 10-5 mol L-1 n-butylamine. TBPE‚H: 5.6 × 10-4 mol L-1. Reference: reagent blank.
mochromism) caused by temperature changes. Alkylamines also reacted with HTBPE in 1,2-DCE and showed reversible thermochromism in a manner similar to aromatic amine complexes. When the temperature in the cell was elevated from 20 to 45 °C, absorbance at 410 nm, the λmax of HTBPE species, increased. That is, the CT band decreased. Since an isosbestic point was at ∼480 nm, it is possible that the equiliburium shift with elevating temperature is as follows: heat
RNH2‚HTBPE 98 RNH2 + HTBPE yellow colorless red-violet
gas phase11 and aprotic solvents such as 1,2-DCE, chloroform, and carbon tetrachroride and that it is disconnected from the oxygen atom of the TBPE molecule; the CT band changes depending on the position of the proton. On the other hand, in a previous paper,12 we showed that, when 1,8-bis(N,N-dimethylamine)naphthalene (Proton Sponge), which has extremely high basicity, was added to HTBPE (yellow) in a nonpolar solvent (benzene), λmax shifted to 620 nm, and the color changed to be blue. It seems that the proton is attracted to the Proton Sponge, and the dissociated ion-pair is formed, as shown in Chart 1b. Thermochromism of Amine-HTBPE CT Complexes. In our previous paper,6 we found that dibucaine and chlorpheniramine (aromatic tertiary amines) reacted with a blue TBPE anion to form red-violet CT complexes in 1,2-DCE and chloroform and that the complexes showed reversible color changes (ther(11) Smith, M. B. Organic Synthesis; McGraw-Hill, Inc.: New York, 1994; p 104. (12) Nishimura, N.; Misaki, K.; Miyake, J.; Sakai, T. Chem. Lett. 1988, 1239.
1768
Analytical Chemistry, Vol. 69, No. 9, May 1, 1997
Disappearance of CT Complexes at Higher Temperatures. Figure 2 shows changes in absorption spectra of di-n-butylamine and the mixtures of di-n-butylamine and n-butylamine at 20 and 60 °C. Curves 1 and 1′ are for 1 × 10-5 mol L-1 di-n-butylamine, 2 and 2′ are for the mixture of 1 × 10-5 mol L-1 di-n-butylamine plus 5 × 10-6 mol L-1 n-butylamine and 3 and 3′ are for the mixture of the same di-n-butylamine plus 1 × 10-5 mol L-1 n-butylamine. At 20 °C, the absorbance increased with addition of n-butylamine, but the absorbance at 60 °C was in agreement with that of only di-n-butylamine because the absorbance of the n-butylamine complex disappeared at 60 °C. This effect suggests that it is possible to detect only the di-n-butylamine complex, despite the presence of n-butylamine, if the measuring temperature is kept at 60 °C. This technique is applicable to the determination of n-tertiary amines and/or n-secondary amines without interference from n-primary amines, though amine concentrations mixed are limited. The relationship between absorbance and concentration of tertiary and secondary amines was linear at both 20 and 60 °C. The temperature error was (0.5 °C. Determination of Tertiary and/or Secondary Amine in Artificial Samples. A solution of (0.5-4) × 10-5 mol L-1 n-heptylamine was added to 2 × 10-5 mol L-1 tri-n-heptylamine, and the absorbance of the mixture was measured at 20 °C. Even 5 × 10-6 mol L-1 n-heptylamine gave a positive interference to the determination of tri-n-heptylamine at 20 °C, and the interference became stronger with an increasing concentration of n-
Table 3. Recovery Tests for Tertiary and/or Secondary Amines in the Presence of Primary Amine at 20 and 60 °C tri-n-heptylaminea added concn of n-heptylamine (mol L-1) 0 5 × 10-6 1 × 10-5 2 × 10-5 3 × 10-5 4 × 10-5
20 °C abs rec (%) 1.109 1.170 1.255 1.443 1.584 1.752
100 106 113 130 143 158
60 °C abs rec (%) 0.745 0.754 0.746 0.744 0.745 0.738
100 101 100 100 100 99
di-n-amylamineb added concn of n-amylamine (mol L-1)
abs
0 5 × 10-6 1 × 10-5 2 × 10-5
0.459 0.505 0.589 0.750
20 °C rec (%) 100 110 128 163
abs
60 °C rec (%)
0.228 0.226 0.209 0.221
100 99 92 97
Table 4. Thermodynamic Quantities for Amine-HTBPEa Complexes 25 °C
n-propylamine n-butylamine n-amylamine n-heptylamine n-octylamine di-n-propylamine di-n-butylamine di-n-amylamine di-n-octylamine tri-n-propylamine tri-n-butylamine tri-n-amylamine tri-n-heptylamine tri-n-octylamine isobutylamine diisobutylamine triisobutylamine
log Kassoc
∆G°
∆H°
∆S°
3.05 3.07 3.16 3.27 3.23 4.26 4.50 4.43 4.54 4.76 5.11 4.96 5.28 5.00 2.85 3.70 1.82
-17.4 -17.5 -18.0 -18.6 -18.4 -24.3 -25.7 -25.3 -25.9 -27.2 -29.2 -28.3 -30.1 -28.5 -16.3 -21.1 -13.2
-61.2 -62.6 -62.3 -55.4 -72.5 -57.5 -65.1 -63.5 -63.8 -64.5 -66.2 -57.7 -73.5 -56.0 -57.3 -73.2 -42.6
-147 -152 -149 -124 -181 -112 -132 -128 -127 -125 -124 -98.5 -146 -92.3 -138 -175 -108
a HTBPE, 2.4 × 10-5 mol L-1. ∆G°, ∆H°/kJ mol L-1; ∆S/J K-1 mol L-1. The errors were (3% for three determinations.
tri-n-butylaminec added concn of triisobutylamine (mol L-1)
abs
0 5 × 10-6 1 × 10-5 2 × 10-5 3 × 10-5 4 × 10-5
0.511 0.530 0.544 0.572 0.597 0.616
20 °C rec (%) 100 104 106 111 117 121
abs 0.380 0.378 0.377 0.388 0.394 0.386
60 °C rec (%) 100 99 99 102 104 102
a λ -5 mol L-1; HTBPE, 3.2 × 10-4 mol L-1. max ) 581 nm, 2 × 10 λmax ) 570 nm, 1 × 10-5 mol L-1; HTBPE; 5.6 × 10-4 mol L-1. c λmax ) 580 nm, 1 × 10-6 mol L-1; HTBPE, 3.2 × 10-4 mol L-1. b
heptylamine. However, when the measuring temperature was kept at 60 °C, no interference was observed up to the level of 4 × 10-5 mol L-1 (Table 3). The upper limit concentration of tri-nheptylamine that could be determined by the proposed technique was 2.2 × 10-5 mol L-1, and the lower limit was 5 × 10-6 mol L-1. The relative standard deviation was 3% for five determinations of 1 × 10-5 mol L-1 tri-n-heptylamine. As another example, at 20 °C, 5 × 10-6 mol L-1 n-butylamine gave strong interference to 1 × 10-5 mol L-1 di-n-butylamine at 570 nm. However, this interference was eliminated at 60 °C, and the tolerance limit of coexisting n-butylamine was 1 × 10-5 mol L-1. The upper concentration limit of di-n-butylamine for which a proper absorbance (less than about 1.20 absorbance) was observed at both 20 and 60 °C was 4 × 10-5 mol L-1, and the lower limit was 1.0 × 10-5 mol L-1. In the mixture of di-n-amylamine and n-amylamine, good recovery was also obtained within about 3% error for di-namylamine and the upper concentration limit was 2.5 × 10-5 mol L-1, and the lower was 1 × 10-5 mol L-1. In addition, Table 3 also shows the thermospectrometric effect for the selective analysis of steric isomers. Analysis between tri-n-butylamine and triisobutylamine was then attempted. As shown in Table 1, log Kassoc of tri-n-butylamine was large, while that of triisobutylamine was very small because of steric hinderance. The difference was 3.05 at 45 °C. Accordingly, 4 × 10-5 mol L-1 triisobutylamine in the mixture could be easy eliminated at 60 °C in the determination of 1 × 10-5 mol L-1 tri-n-butylamine, and discrimination among these amines was also possible. The upper limit concentration
Figure 3. Plots of ∆S° vs ∆H° for normal or iso primary, secondary, and tertiary amine complexes. 4, normal primary amines; 2, iso primary amines; O, normal secondary amines; b, iso secondary amines; and 0, normal tertiary amines.
for the tri-n-butylamine analysis was 2 × 10-5 mol L-1 and the lower limit was 5 × 10-6 mol L-1. The RSD was 2.5% for five determinations of 1.0 × 10-5 mol L-1 tri-n-butylamine. In summary, the positive interference at 20 °C from a primary amine, during the analysis of tertiary or secondary amines, can be eliminated at 60 °C, with recovery at almost 100%. It seems that the applicability of this technique depends on the formation constant differences among amines (Table 1). In the case of octylamines, the difference between n-octylamine and di-n-octylamine complexes was 1.30, and that between n-octylamine and tri-n-octylamine was 1.73. When the difference (∆ log Kassoc) is large and a certain level of energy is added to the complexes, the absorbance of one complex remains and the other one disappears. That is, selective analysis between a primary amine and a secondary amine and/or a tertiary amine is possible by controlling the measuring temperature. Thermodynamic Characteristics. A good linearity between log Kassoc of the amine-HTBPE complex and a 1/T was observed. ∆H° from the slope of the line and ∆S° from the intercept were Analytical Chemistry, Vol. 69, No. 9, May 1, 1997
1769
a s so c
Figure 4. Plots of proton affinity (PA) for CT complexes versus log Kassoc. 4, normal primary amines; 2, iso primary amines; O, normal secondary amines; b, iso secondary amines; and 0, normal tertiary amines.
calculated using the van’t Hoff equation as follows:
log Kassoc ) -
∆S° ∆H° + 2.303RT 2.303R
The results are shown in Table 4. As a reasonable result, the values of ∆G° became small in the order of primary, secondary, and tertiary amines. The ∆G° value for triisobutylamine complex was the smallest. Although it was not obvious which contributed to the formation of CT complexes, ∆H° or ∆S°, the contribution of ∆S° to the CT complex formation is presumably larger than that of ∆H°. The plot of ∆S° vs ∆H° is shown in Figure 3. It is assumed that (1) the formation constants are governed by one simple reaction mechanism because the plots are linear and (2) the CT complex formation ability of each amine with HTBPE depends on the electronic effect of each amine. Although the line for normal and iso primary amines is linear and the slope is almost the same as that for secondary or tertiary amines, the intercepts differ. It seems that the steric effect for formation of CT complexes has little influence. On the other hand, the slope for iso secondary amine complexes is different from that of normal (13) Aue, H. D.; Webb, M. H.; Bowers, T. M. J. Am. Chem. Soc. 1976, 98, 311.
1770
Analytical Chemistry, Vol. 69, No. 9, May 1, 1997
amine complexes. It seems that the steric effect on the CT complex formation is larger for iso secondary amines. In addition, the relationships between log Kassoc and pKa of amines10 and/or log Kassoc and proton affinity (PA)13 were investigated. A strong correlation was not observed between log Kassoc and pKa. This is because (1) the spread of pKa values is very small and (2) pKa values depend on the hydration effect in the aqueous phase, and its utilization was not appropriate for the CT complex formation reaction in the aprotic organic solvents with lower dielectric constants. Proton affinity of normal and iso amines versus log Kassoc was also plotted (Figure 4). The correlation coefficient was 0.920, and the plot was almost linear. The magnitude of log Kassoc was proportional to the PA and increased in the order of primary, secondary, and tertiary amine groups. Consequently, it seems that the CT complex formation by the hydrogen bridge depends on proton affinity. CONCLUSION Alkylamines react with HTBPE dye (yellow) to form CT complexes (red-violet) in 1,2-DCE. The R3N‚HTBPE (R2NH‚ HTBPE, RNH2‚HTBPE) complexes dissociate to HTBPE and R3N (R2NH, RNH2) with an increase in the measuring temperature. The degree of disconnection depends on the strength of the hydrogen bond for amines. When the measuring temperature was elevated from 20 to 60 °C, the absorbance of primary amine complexes became almost zero, while absorbances due to secondary and tertiary amines remained. The thermochromism phenomenon was applied to the selective determination of alkylamines. As a result, a tertiary and/or secondary amine in the presence of a primary amine could be analyzed without interference from the primary amine. The factors that govern the formation of CT complexes were also investigated thermodynamically. It was obvious that the steric effect of each amine group and proton affinity influenced the amine-HTBPE CT complex formation. ACKNOWLEDGMENT The authors thank Professor M. Tabata of Saga University for his valuable discussion. Received for review July 18, 1996. Accepted December 6, 1996.X AC960690I X
Abstract published in Advance ACS Abstracts, March 15, 1997.
