Mobile phase generators

Detectability may be a significant problem with homologous series of unsaturated compounds, particularly //-alkanes. For these compounds, refractive index detection or evaporative light-scattering, both of which are described elsewhere in the book, may be of use. Indirect photometry is a useful detection scheme for compounds that do not absorb in the UV. Acetone, methylethyl ketone, methyl propyl ketone, methyl isopropyl ketone, methyl isobutyl ketone, and acetophenone are added to an acetonitrile/water mobile phase, generating a negative vacancy peak when the nonchro-mophoric analyte emerges and a positive peak if the ketone is adsorbed and displaced.70 Dodecyl, tetradecyl, cetyl, and stearyl alcohols also have been derivatized with 2-(4-carboxyphenyl)-5,6-dimethylbenzimidazole and the derivatives separated on Zorbax ODS in a mobile phase of methanol and 2-propanol.71... 

It is cmcial that the separation mechanisms in HPSEC are determined by the size of the molecules and not by their charge, nor by ion exclusion effects or adsorption into the column. These remarks are particularly relevant to chitosan which is a polyelectrolyte. The addition of sodium acetate or ammonium acetate in the buffCT used as the mobile phase generates enough ionic strength to overcome these unwanted features. 

The optimization process generates the actual separation at each point (i.e., mobile phase composition) and eliminates those that produce the least acceptable results. In many cases the initial decision parameter is resolution. As a consequence, the three mobile phases generating the best resolution between all peaks ate used as the apices of a new triangle. From those three the next series of experiments is constructed. For instance, in the above example, if the results at points 4, 2 and acetonitrile were most acceptable, then the next set of experimoits include points 5, 6, 7, and 8 (see Figure 2.5b). Successive iterations of this process will lead to the best separation and the sq)aration is ( timized. This entire process can be visualized by through a 3D plot as shown in Figure 2.6. The higher the 2 value, the better the separation [59]. 

Thiophanate-methyl and its metabolites (2 aminobenziniidazole, carbendazim) were isolated from water and separated on a C g column (A = 270ran). A 50/50 methanol/water with 0.6% ammonia mobile phase generated elution in 10 min [187]. A linear working range of 0.25-10 pg/L with a detection limit of 0.25pg/L (S/N = 3) was reported. 

Seven benzophenone sunscreen agents (benzophenones 1, 2, 3, 4, 5, 8, 10 often hydroxylated and/or methoxylated benzophenones) were extracted from suncreams, lipsticks, and other personal care products and separated on a C g column (A = 286nm). A 30/30/40 methanol/acetonitrile/water mobile phase generated good peak shapes and resolution in 40 min [255]. All detection limits were similar and were reported as about 1 mg/L. 

Eleven folylmonoglutamates were separated on a phenyl column and quantitated using electrochemical detection (-1-900 mV vs. Ag/AgCl), fluorescence detection (A = 295 nm, ex 365 nm, em), or photodiode msy detection (A = 200-400 nm). A 15/85 methanol/water (50 mM phosphate buffer pH 3.5) mobile phase generated... 

Vitamin K was also isolated from emulsified nutritional supplements [343]. Quantitation was accomplished using post-column reduction on a platinum oxide catalyst. A 50/50 ethanol/methanol mobile phase generated a 10-min retention time on a 40°C C j column (A = 320nm, ex 430nm, em). The reported detection limit was 0.1 pg injected (S/N = 3) with a linear range of 0-2 pg injected. It is important to remember that many types of HPLC-grade ethanol are available and, although this is probably not important in this separation, care should be taken to ensure that this information is provided. 

Trichothecenes (nivalenol, deoxy- and 15-acetyldeoxynivalenol) were extracted fi om bananas and analyzed on a C g column (A = 225 nm). A 35/65 methanol/ water mobile phase generated a 10 min baseline-resolved sq>aration [400], Standards were run from 2 to 20 pg/g and detection limits of 5ng/g were reported. Similarly, in the same study, a- and ) -zearalenol and zearalenone were well resolved using a 65/35 methanol/water mobile phase. Standard levels were 2-20 pg/g and detection limits of 9 ng/g were reported. 

Three neuraminic acids (7V-glycoyl-, IV-acetyl-, and 7V-propionylneuraminic acids) were derivatized with 1,2-diamino-4,5-(methylenedioxy)benzene and separated on a 40°C Cjg column (A = 373nm, ex 448nm, em). A 100/15/10 water (lOmM acetate buffer at pH 5)/methanol/acetonitrile mobile phase generated a 9-min baseline resolved chromatogram [447]. Solutions of 0.1-1 nmol analyte were used in the procedure. 

Four components of gentamicin (termed C, C2, Cia< C2,) were extracted from bovine milk, derivatized with o-phthalaldehyde, and separated on a Cjg coluirm (A = 340 nm, ex 430 nm, em). An 82/18/0.1 water (11 mM pentanesulfonic acid with 5.6mM sodium sulfate)/methanol/acetic acid mobile phase generated baseline resolution and elution in 20min [S05]. Standards ran fi om 30 to 240ng/mL and detection limits were reported as 0.4ng/mL. 

Four polyether antibiotics (lasalocid, monensin, saliomycin, narasin) were extracted from feed, derivatized with 2,4-dinitrophenylhydrazine (DNPH), and quantitated on a Cjg colutrm (A = 305 nm for lasalocid and 393 nm for all other compounds). A 90/10 methanol/water (1.5% acetic acid) mobile phase generated baseline resolution, good peak shape, and elution in < 15 min. It should be noted that... 

Auraptene, an anticancer phytochemical candidate, was extracted from a wide range of citrus fruits and analyzed on a 40°C Cjg column (A = 325 nm). A 75/25 methanol/water mobile phase generated elution in 6 min. A detection limit of 0.1 ig/g was claimed [522]. 

Diazepam, three metabolites (A -desmethyldiazepam, temazepam, oxazepam) and clonazepam [IS] were isolated from plasma or urine and separated on a base deactivated C g column (2 = 232nm). A 50/10/40 methanol/acetonitrile/water (50 mM KH2PO4 pH 3.5) mobile phase generated baseline resolution in 12 min [552]. Calibration standards ran from 10 to 2000ng/mL with detection limits of lOng/mL reported. Similarly, diazepam and three metabolites (nordiazepam, oxazepam, temazepam) were baseline resolved in 12min on a Cjg column (A = 257 nm) using a 60/35/5 methanol/water/acetonitrile mobile phase [553]. 

Severe chronic pain is sometimes treated with co-administered ketamine/ bupivacaine. These compounds were isolated from plasma and analyzed on a cyanopropyl column (A = 215nm). A 718/220/80/2 water (lOmM NaH2p04 buffer at pH 2.34)/methanol/acetonitrile/H3P04 mobile phase generated baseline separation in <8 min. The peaks were quite tailed. Introduction of TEA and/or a TFA/triethylamine buffer syston in place of the phosphate system may remedy this. Calibration curves from 10 to 400ng/mL (ketamine) and 125 to 4000ng/mL (bupivacaine) were stated, with the lowest concentration serving as the quantitation limit. Detection limits (S/N = 4) of 1 ng injected and 0.8 ng injected, respectively, were reported [563]. 

Undecylenic acid was extracted from ointments and powders, derivatized with 4 -nitrophenacyl bromide, and analyzed on a Cg column (2 = 265 nm). A 50/30/20 methanol/acetonitrile/water mobile phase generated a 7 min elution [573]. A table of the effect on retention times of changing phenacyl derivatives (e.g., 4 -methyl, 4 -phenyl, 4 -bromo) and the column to a C,g is presented. The linear range was reported as 12.5-300 pg injected. 

Four antibacterial compounds (dimetridazole, ronidazole, metronidazole, hydto-xymetridazole) were isolated from poultry meat and separated on a C g column (positive ion electrospray MS, capillary T — 220°C quadrapole T = 70°C no auxiliary gas of et voltage —0.2 V spray voltage 5kV, N2 sheath gas SOpsi collision voltage lOV). An 81/13/6 water (formic acid)/methanol mobile phase generated complete elution in <8 min. A linear range of 0.025-0.04 mg/L and detection limits of 2-5 pg/kg were reported [577]. 

Heavy petroleum was characterized with respect to its aromatic (e.g., benzoie, I-methylnaphthylene), and nonaromatic (e.g., tetradecane) content These results were comparable to the lengthier ASTM D2549-91 meOiod. Two 300 x 4.1mm 35°C aminopropyl columns (Rl detector) and a 95/5 hexane/IPA mobile phase generated complete elution in <15 min [627]. 

The stereoisomeric pairs of ricinoleate, methyl isoricinoleate, and methyl esters of three additional bis-homoallylic hydroxy fatty acids were separated as their a-naphthylethyl isocyanate derivatives [696]. A silica column and a 95/5/1 hexane/ ethyl acetate/THF mobile phase generated the separation. Elution was generally complete in <30 min and resolution was excellent in all cases. 

In a nice study by Ye et al. [704] the comparative performance of narrow-bore and analytical silica columns on the sqiaration of a-, fi-, y-, and -tocopherols was determined. Temperature was held at 29°C and peaks were detected at A = 298 run. A 99.2/0.8 hexane/IPA mobile phase generated baseline resolution and elution in 14 min. Linear ranges of 0.33-1 S.Sng injected was established and quantitation limits (analyte dependent) were 14-51 pg injected and 72-307 pg injected for narrow-bore and analytical, respectively. 

Four Aconitum alkaloids (aconitine, mesaconitine, hypaconitine, jesaconitine) were extracted from blood and separated on a 40°C C,g colunrn (2 = 260nm for jesaconitine and 235 nm for all others). A 14/86 THF/water (0.2% TEA) mobile phase generated baseline resolution and elution in 15 min. Calibration curves from 100 to 10000ng/mL and detection limits of 50ng/mL were reported [833]. 

Triglycerols were determined in human milk in a similar fashion. A C g (R1 detector) column and 64/36 acetone/acetonitrile mobile phase generated the separation [872]. Components with equivalent carbon numbers ranging from 36 to 54 were eluted in 45 min. Greater resolution was obtained by placing two Cjg columns in series and increasing the analysis time to 90 min. Thirty-nine peaks were clearly delineated and identified. 

Trace levels of vanadium(V) were extracted from water and coal fly ash samples and analyzed as the 2-(8-quinolylazo)-5-(dimethylamino)phenoI complex [922]. A Cjg column (2 = 550nm) and a 50/50 acetonitrile/water (lOmmol/kg tetrabutyl-ammonium [TEA] bromide) mobile phase generated the separation. A number of potentially co-eluting metal ions were studied as well Ni, Co, and Fe. Detection limits of 3 pg were reported. Vanadium(IV) and (V), Pb(II), Cu(II)and Cr(VII) were resolved on a Cg colunrn (2 = 245nm) with a 12/78 acetonitrile/water (50 mM TEA with 2mM EDTA to pH 6) mobile phase [923]. Elution was complete in <20 min and peak shapes were good. Detection limits of 1 ng and linear concentration ranges of 1-30 pg/mL were reported. 

The diasteriomers of metolachtor were separated on a 10°C graphitized carbon column (A = 210 run). The authors noted that the low temperature was necessary for improved resolution. A 1/1 acetonitrile/water mobile phase generated baseline resolution with complete elution in 17 min [952]. Standard solutions of 1-10 mg/mL were used. 

In a very simple and elegant method by Peikins et al. [989], hexazinone was analyzed in ground water through direct injection on a Cg column (A = 247 nm). A 60/25/15 water/acetonitrile/methanol mobile phase generated elution in 8 min. The linear range was 0.3-33 pg/L and the quantitation limit was 10 pg/L. 

Ginkgolic acid is a mixture of several 2-hydroxy-6-alkyIbenzoic acids, two of which (alkyl = 1-octene or 3-decene) were baseline resolved on a 45 C Cjg column (A = 308nm) with a 92/7/1 acetonitrile/water/acetic acid mobile phase [1053]. Elution was complete in 6 min and peaks of interest were well resolved from all extracted material. In a separate study, Pietta et al. [1054] isolated six metabolites (hippuric acid, 4-hydroxy- and 3-methoxy-4-hydroxyhippuric acid, 4-hydroxy-, 3,4-dihydroxy-, and 3-methoxy-4-hydroxybenzoic acid) of Ginkgo biloba leaf extracts from urine and blood samples and analyzed on a Cjg column (photodiode array detector). An 88/5/5/2 water/acetonitrile/methanol/acetic acid mobile phase generated good resolution and complete elution in 15 min. The standard used was at the 1 mg/mL level. 

Anatoxin-a, homoanatoxin-u and four degradation products (dihydroanatoxin-a, dihydrohomoanatoxin-a, epoxyhomoanatoxin-a, epoxyanatoxin-a) were isolated fiom blue-green algae, derivatized with NBD-F (4-fluoro 7-nitro-2,l,3-benzoxadia-zole), and separated on a 35°C C g column k = 470 nm, ex 530 nm, em). The use of a 55/45 water/acetonitrile mobile phase generated baseline resolution, excellent... 

Onion skin and wine extracts were analyzed for bioflavonoid content (quercetin, kaempferol, fisetin, rutin, myricetin, morin). A 50°C Cg column (A = 370 nm) and a 71/28/1 water/acetonitrile/TFA mobile phase generated a 15-min elution. Peaks were resolved and shapes were excellent. The wine extract had a large early-eluting interfering set of peaks (three analytes eluted on the shoulders of these peaks). A linear range of 0.1-30 pg/mL was reported [1139],... 

The metallointercalator-DNA conjugates ([Ru ][9.10-phenanthrone quinone diimine]2[4-butyric acid-4 -methyl bipyridyl]-DNA, where DNA is a 20-mer oligonucleotide) attached to the 5 or 3 terminus, were synthesized and separated on a C g column (A = 260 nm or 390 run). A 50/50 water (50 mM armnoniinn acetate buffer at pH 6)/acetonitrile mobile phase generated baseline resolution and elution in 35 min [1198]. 

Ferulic acid and liguistilide were extacted from Angelica sinensis roots and analyzed on a Cjg column (A = 303 nm for ferulic acid, 205 nm for liguistilide). A 55/45 methanol/water (1.5% acetic acid) mobile phase generated good peak shapes and elution in 17 min [1261]. Note that the Imguistilide work is conducted very near the UV cutoff for methanol. This could lead to reproducibility problems. Linear curves from 10 to 500pg/mL were reported. 

The analgesic morphine and three metabolites (morphine-3- and 6-glucuronide, hydromorphone) were isolated from plasma and separated on a 35°C C g column (multi-channel electrochemical detector set at 0.2 y 0.35 V, and 0.4 V and X = 275 nm, ex 345 nm, em for morphine-3-glucosonide only). A 25/75 acetoni-trile/water (50 mM phosphate pH 2.1 with 2.5 mM sodium dodecyl sulfate) mobile phase generated baseline resolution in 24 min. Linear ranges of 5-600 ng/mL and detection limits of 0.2 ng/mL were reported [1334]. 

Four macrolide antibiotics (azithromycin, erythromycin, roxithromycin, clarithro-inycin) were extracted om serum, derivatized with FMOC-Cl, and separated on a 50°C base deactivated C,g column (A = 225 nm, ex 31Snm, em). A 40/60 water (SOmM KH2PO4 with SOOpL triethylamine/L to pH 7.5 with 10% KOH)/ acetonitrile mobile phase generated baseline resolution and complete elution in 22 min [1346]. Calibration curves of 0.1-20mg/mL and quantitation limits of 1-12 mg/L were reported. 

Epirubicin and doxorubicin and their 13-dihydro metabolites were extracted from plasma and separated on a C,g column (electrochemical detector -1-400 mV guard and —300 mV signal). A 71 /29 water (50 mM Na2HP04 and 0.05% trietlqrlamlne to pH 4.6 with citric acid)/acetonitriIe mobile phase generated baseline resolution in 13 min [1384]. A linear concentration range from 1 to 500ng/mL was reported. 

---