High-speed cryo-focusing injection for gas chromatography: Reduction of injection band broadening with concentration enrichment Ryan B. Wilson, Brian D. Fitz, Brandyn C. Mannion, Tina Lai, Roy K. Olund, Jamin C. Hoggard, Robert E. Synovec n Department of Chemistry, Box 351700, University of Washington, Seattle, WA 98195-1700, USA article info Article history: Received 26 January 2012 Received in revised form 9 March 2012 Accepted 20 March 2012 Available online 25 April 2012 Keywords: Thermal injection Cyrogenic trapping Gas chromatography High speed abstract In order to maximize peak capacity and detection sensitivity of fast gas chromatography (GC) separations, it is necessary to minimize band broadening, and in particular due to injection since this is often a major contributor. A high-speed cryo-focusing injection (HSCFI) system was constructed to first cryogenically focus analyte compounds in a 6 cm long section of metal MXT column, and second, reinject the focused analytes by rapidly resistively heating the metal column via an in-house built electronic circuit. Since the cryogenically cooled section of column is small ( 750 nl) and the direct resistive heating is fast ( 6000 1C/s), HSCFI is demonstrated to produce an analyte peak with a 6.3 ms width at half height, w 1/2 . This was achieved using a 1 m long column with a 180 mm inner diameter (i.d.) operated at an absolute head pressure of 55 psi and an oven temperature of 60 1C, with a 10 V pulse applied to the metal column for 50 ms. HSCFI was also used to demonstrate the head space sampling and fast GC analysis of an aqueous solution containing six test analytes (acetone, methanol, ethanol, toluene, chlorobenzene, pentanol). Using Henry’s law constants for each of the analytes, injected mass limits of detection (LODs) were typically in the low pg levels (e.g., 1.2 pg for acetone) for the high speed separation. Finally, to demonstrate the use of HSCFI with a complex sample, a gasoline was separated using a 20 m  100 mm i.d. column and the stock GC oven for temperature programming, which provided a separation time of 200 s and an average peak width at the base of 440 ms resulting in a total peak capacity of 460 peaks (at unit resolution). & 2012 Elsevier B.V. All rights reserved. 1. Introduction Gas chromatography (GC) is often used in repetitive, routine, time sensitive applications to analyze complex mixtures of volatile and semi-volatile analytes. For such applications, the reduction of analysis time is desired, from a traditional time scale of 10 min–60 min down to emerging applications in the minutes to seconds time frame, and is commonly achieved by using short (1 m–10 m), narrow (100 mm–180 mm inner diameter) columns at high linear flow velocities and either fast temperature program ramp rates or an isothermal oven. However, unless off-column sources of band broadening (due to injection, detection, electro- nics, etc.) are minimized, the peak widths obtained are not minimized, and the resulting chromatograms may lack the peak capacity and separation power of GC performed on a longer, more traditional time scale. It has been supported from GC theoretical considerations [1], and demonstrated experimentally [2] with a state-of-the art injector, that an unretained peak, eluting from a long (40 m  180 mm) column under optimal flow rate conditions and with no off-column band broadening, should and does have a width of only  250 ms. Various commercial instruments for common practice, equipped with a standard auto-injector, often produce peaks typically  2s wide, even on 180 mm wide columns [2]. This gap between GC theory and state-of-the-art experimentation versus common prac- tice (and the desire for faster analysis) has prompted researchers to devote a significant amount of attention to reducing off-column sources of broadening, particularly the injection pulse width. The resulting reports cover a wide variety of techniques for producing narrow injection bandwidths (fluid logic gates [3,4], split injection with high split ratios [5], microloop systems [6] and micro gas valve inlets [7], etc.). Recent reports from our group demonstrated that single high-speed diaphragm valves are extremely capable injection systems, producing peaks  20 ms wide [8], and dual diaphragm synchronized-injection valve systems are capable of 0.5 ms wide peaks [9]. Unfortunately, for a valve-based injection system, a GC- like separation can inadvertently occur in the transfer capillary between the GC inlet and the valve when oven temperatures are low, as is typically the case at the beginning of a temperature programmed separation. Unless special attention is given to the Contents lists available at SciVerse ScienceDirect journal homepage: www.elsevier.com/locate/talanta Talanta 0039-9140/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.talanta.2012.03.054 n Corresponding author. Fax: þ1 206 685 8665. E-mail address: synovec@chem.washington.edu (R.E. Synovec). Talanta 97 (2012) 9–15