Basic Amine Analytes

A series of 10 tobacco alkaloids (e.g., (/, 5)-anatabine, (S)-(—)-cotinine, -nicotyrine, -nicotine, and -anabasine, and syn-nicotine- -oxide) were separated on 

Analysis of urinary nicotine and cotinine has been widely studied. A study by Reynolds and Albazi [294] reported extraction of the analytes from urine followed by baseline resolution on a 40°C cyanopropyl column (A = 260 nm). Complete elution was achieved in 15 min when using a 3/97 IPA/water (50 mM sodium phosphate buffer at pH 2 with 0.1 M sodium dodecyl sulfate [SDS]) mobile phase. Plots of k vs. percent IPA show a decrease in retention from 0 to 5% IPA then an increase in retention starting at 8% IPA. The authors attribute this increase in K to the disruption of the SDS micelles. The mechanism appears to change from a micelle-based to an ion pair-mediated separation over this IPA range. Standard concentrations for both nicotine and cotinine covered the range from 0.2 to 3 ppm. 

The uiinaiy and hepatocyte metabolites of tobacco-specific nilrosamines (e.g., 4-[methylnilrosainino]-l-[3-pyridyl]-l-butanone and butanol, 4-hydroxy- and 4-oxo-4-[3-pyridyl]butyric acid, 4-[methylnitrosamino]-l-(JV-oxy-3-pyridyl)-l-butanone and butanol) were well resolved on a C g column (A = 254nm) using a complex 75-min 0/100 - 30/70 methanol/water (20 mM sodium phosphate buffer at pH 7) gradient [2%]. Excellent peak shapes were obtained and elution was complete in 70 min. Electrospray MS detection was also used with a nearly identical setup the sodium phosphate buffer was replaced with a lOmM ammonium acetate buffer at pH 6.7. 

Norharman () -caiboline), barman (l-methyl-) -caiboline), and l-ethyl-) -carbo-line (as internal standard) were separated in 10min on a C,g column (thermospray MS) using a 23/77 methanol/water (0.1 M ammonium formate with 0.1 M formic acid buffer at pH 3.4) mobile phase [297]. Alternatively, fluorescence detection (A = 300 nm, ex 433 nm, an) and 32/68 methanol/water (0.1 M potassium phosphate buffer at pH 3.0) were used. Detection limits of 5-lOng/L were reported (analyte dependent). 

Kotzabasis et al. [298] used a C g column (A = 254run) and a 23-min 55/45 —y 84/16 methanol/water gradient to separate the benzoyl derivatives of putrescine, cadaverine, spermidine, spermine, and agmatine extracted fiom leaves. Levels as low as 0.2 nmol were easily detected. Peak shapes were dramatically dependent upon the sample solvent chosen. Samples dissolved in 100% methanol generated extremely tailed peaks, whereas samples dissolved in 80/20 methanol/ water generated peak doublets. Odd peak shapes are not uncommon when a sample solvent-mobile phase mismatch occurs. A sample solvent similar to the mobile phase composition almost invariably gives the best peak shape and the most reproducible results. Whenever possible, a sample solvent identical in composition to the mobile phase is generally recommended. 

---

The NH + species can then react with a basic amine analyte compound, RNH2, to create an ionized species with an m/s ratio 1 Da in excess of the parent compound s molecular weight as follows ... 

Photoluminescence experiments with ni-V wafers of Ino.5o(Gao.9o A1o.io)o.5oP were conducted 114]. The Lewis basic gaseous analytes ammonia, methylamine, dimethyl amine, and trimethylamine all yielded reversible PL enhancements. The Lewis acid sulfur dioxide, in contrast, caused reversible quenching of the semiconductor s PL intensity. These PL intensity changes were consistent with analyte-induced modifications of the dead-layer thickness. 

Competing amines such as triethylamine and di-rc-butylamine have been added to the mobile phase in reversed-phase separations of basic compounds. Acetic acid can serve a similar purpose for acidic compounds. These modifiers, by competing with the analyte for residual active sites, cause retention time and peak tailing to be reduced. Other examples are the addition of silver ions to separate geometric isomers and the inclusion of metal ions with chelating agents to separate racemic mixtures. 

The FAB plasma provides conditions that allow to ionize molecules by either loss or addition of an electron to form positive molecular ions, M" , [52,89] or negative molecular ions, M, respectively. Alternatively, protonation or deprotonation may result in [Mh-H]" or [M-H] quasimolecular ions. Their occurrence is determined by the respective basicity or acidity of analyte and matrix. Cationization, preferably with alkali metal ions, is also frequently observed. Often, [Mh-H]" ions are accompanied by [MH-Na]" and [Mh-K]" ions as already noted with FD-MS (Chap. 8.5.7). Furthermore, it is not unusual to observe and [Mh-H]" ions in the same FAB spectmm. [52] In case of simple aromatic amines, for example, the peak intensity ratio M 7[Mh-H] increases as the ionization energy of the substrate decreases, whereas 4-substituted benzophenones show preferential formation of [Mh-H]" ions, regardless of the nature of the substituents. [90] It can be assumed that protonation is initiated when the benzophenone carbonyl groups form hydrogen bonds with the matrix. 

The use of HPLC to analyze biogenic amines and their acid metabolites is well documented. HPLC assays for classical biogenic amines such as norepinephrine (NE), epinephrine (E), dopamine (DA), and 5-hydroxytryptamine (5-HT, serotonin) and their acid metabolites are based on several physicochemical properties that include a catechol moiety (aryl 1,2-dihydroxy), basicity, easily oxidized nature, and/or native fluorescence characteristics (Anderson, 1985). Based on these characteristics, various types of detector systems can be employed to assay low concentrations of these analytes in various matrices such as plasma, urine, cerebrospinal fluid (CSE), tissue, and dialysate. 

What is the reason for the overwhelming acceptance of stationary phases based on high-purity silicas in the pharmaceutical industry The answer is simple superior peak shapes for analytes with basic functional groups, which has been a problem with older phases. The older, low-purity silicas contain metal ions buried in the matrix of the silica. These contaminants acidify the surface silanols, and the consequence is a strong and non-uniform interaction with basic analytes. This in turn results in tailing peaks, which is an impediment for accurate peak integration and peak resolution. Of course, adding appropriate additives, such as amine modifiers, to the mobile phase can solve these difficulties. But this is an unnecessary and undesired complication in methods development. Therefore, silicas that are free from this complication are much preferred. 

With some stationary phases at low pH values (<4) benzyl amine as benzyl ammonium ion can be excluded by a Donnan potential from the pores, when positive charges are present at the surface. These could have stemmed from the manufacturing process or could have been introduced on purpose to shield amines from interacting with silanols. With an increasing pH, the Donnan exclusion decreases and at pH >5 benzyl amine is retarded increasingly. An example of this effect with a modern RP with low silanophilic properties is demonstrated in Figure 2.22, where the elution peaks of benzyl amine are presented as a function of pH. With these stationary phases, basic analytes cannot be separated at low pH values. 

In comparing the various test procedures, there is always a good agreement found for hydrophobic retention and selectivity as well as for shape selectivity. However, the characterization of silanophilic interaction is still a matter of discussion. In part, the differences are due to the selection of the basic analyte. Therefore, the outcome of every test is different. It has been shown, that the peak asymmetry—used for detection of silanophilic interactions—does not correlate to the pA" value of the basic test solute [64]. A closer look at these data leads to the assumption, that the differences are related to the structure of the basic solute, irrespective of whether a primary, secondary, or a tertiary amine is used. The presence of NH bonds seems to be more important in stationary-phase differentiation than the basicity expressed by the pA value. For comparable test procedures for silanophilic interactions further studies seem to be required. 

Method. A standard amino-acid analyzer (Technicon or an equivalent) may be used. The reagents for development and the buffers are prepared as for analysis of amino acids. The analytical column (24 cm X 0.57 cm) consists of Zeocarb 226-4.5% DVB (average particle diameter, 24 jum). The two buffers are prepared by dissolving 8.74 g of potassium citrate, 60.36 g of potassium chloride, 10 ml of Brij and 100 ml of n-propanol (for the first buffer, 140 ml of n-propanol for the second buffer) in enough water to make a total volume of 11. The pH of each buffer is 7.4. For analysis the sample is adjusted to pH 7.4 and an aliquot portion is applied to the column. The column temperature is maintained at 43 °C for 103 min and is automatically switched to 75 °C for the remainder of the run. The flow-rate of the buffer is 42 ml/h. The first buffer is automatically replaced by the second after 120 min. The second buffer is necessary for the separation of tryptamine and cadaverine. The use of the increased temperature results in a shorter elution time. The retention times of some basic amino acids and amines are listed in Table 4.3. Absorption is monitored at 570 nm with a 1,5-cm flow cell. 

---