96 



solutions containing 214 ng/tiL methyl lactate were analyzed on a 25 m x 0.2 mm i.d. 



HP-5 FSOT column (d(=0.33 p.m). Tlie transfer line and injection port temperature 

 were both set to be isothermal at 200°C and 220°C, respectively. The GC oven was 

 initially held at 40°C for 1.0 min after injection, then ramped at 20°C/min for 8 min 

 up to 200°C and held at that temperature for 1.0 min. The GC was run in splitless 

 mode, with the split valve opening after 0.5 min. The column carrier gas was helium 

 at a head pressure of 20 psig. Methanolic 120 ng/\iL L-lactic acid solutions spiked 

 with acid or base were analyzed on a 10.5 m x 0.178 mm i.d. DB-5 FSOT column 

 (d(=0.4 ^lm). The transfer line and injection port temperature were both set to be 

 isothermal at 250°C and 200°C, respectively. The GC oven was initially held at SO^C 

 for 0.25 min after injection, then ramped at 30°C/min for 4.0 min up to 150°C and 

 held for 0.25 min. The GC was run in splitless mode, with the split valve opening 

 after 0.2 min. The column carrier gas was helium at a head pressure of 5 psig. 



The ion source temperature was set at 170°C for EI analyses and 150°C for 

 CI analyses; the manifold temperature was set at 70°C. Data acquired using 

 PPINICI employed methane at 1650 mtorr. Data were acquired in single mode 

 operation (NCI or PCI) with methane reagent gas at an indicated pressure of 1610 

 mtorr. The filament emission current was set to 200 |j,A with an electron energy of 

 either 100 eV for CI or 70 eV for EI operation. Scanning at 0.5 s scan over Q3 was 

 used to obtain all data. The conversion dynode was set at +5 kV or -5 kV 

 depending upon the desired collection of negative or positive ions. The electron 

 multiplier was set at -800 V to -1000 V for experiments in this section. 



