Glass microfabricated nebulizer chip for mass spectrometry Ville Saarela,* a Markus Haapala, b Risto Kostiainen, b Tapio Kotiaho bc and Sami Franssila a Received 4th January 2007, Accepted 13th March 2007 First published as an Advance Article on the web 4th April 2007 DOI: 10.1039/b700101k A microfluidic nebulizer chip for mass spectrometry is presented. It is an all-glass device which consists of fusion bonded Pyrex wafers with embedded flow channels and a nozzle at the chip edge. A platinum heater is located on the wafer backside. Fabrication of the chip is detailed, especially glass deep etching, wafer bonding, and metal patterning. Various process combinations of bonding and metallization have been considered (anodic bonding vs. fusion bonding; heater inside/outside channel; metallization before/after bonding; platinum lift-off vs. etching). The chip vaporizes the liquid sample (0.1–10 mL min 21 ) and mixes it with a nebulizer gas (ca. 100 sccm N 2 ). Operating temperatures can go up to 500 uC ensuring efficient vaporization. Thermal insulation of the glass ensures low temperatures at the far end of the chip, enabling easy interconnections. Introduction Coupling of a mass spectrometer (MS) to microfluidic systems necessitates approaches where flow rates of the fluidics and absolute sample amounts are compatible with the ionization methods. We have used microfabrication techniques to implement four different ionization methods: atmospheric pressure chemical ionization (APCI), 1 atmospheric pressure photoionization (APPI), 2 electrospray ionization (ESI), 3 and desorption/ionization on silicon (DIOS). 4 Different glasses are widely used in analytical devices. In general, glasses are chemically stable, electrically insulating, transparent over a wide spectrum, and can be used in a wide temperature range. Microscale structures can be made in glass by wet etching in hydrofluoric acid–based solutions. Simple photoresist masks endure etches up to a few tens of micro- meters deep, but deeper trenches require a hard mask, such as gold or silicon thin films. 5 Although etch selectivity between mask material and glass is usually good, problems can arise due to stress-related pin-holes and notch defects at the pattern edges. 6 In contrast, for deep reactive ion etching (DRIE) of glass, the mask selectivity usually becomes the limiting factor. 7 Plasma etching of glass requires high power which also con- sumes the masking layers. Glass-to-glass bonding is used to make enclosed channels in glass. Anodic bonding of two glass wafers requires a suitable ion barrier layer. 8 Fusion bonding of glass is done by heating the wafers near to their softening point. We have previously presented nebulizer chips made of silicon and glass wafers. 9 This paper describes an all-glass nebulizer chip that can be operated at considerably higher temperature, which is beneficial for high boiling point samples. The nebulizer chip is operated with low sample flow rates which makes it suitable for microfluidic to MS interfacing. Combined with versatile operation modes (APCI and APPI) it can also be used with less polar and even non-polar samples, whereas ESI is limited to polar compounds. Experimental Nebulizer chip fabrication The nebulizer chips are fabricated by bonding two wafers together. The bottom wafer has etched channel and nozzle structures, and the top wafer accommodates a platinum metal heater element. Fabrication process of the all-glass nebulizer chip is illustrated in Fig. 1. First, a 400 nm silicon hard mask is deposited on a 500 mm thick borosilicate glass (Pyrex) wafer by low pressure chemical vapour deposition (LPCVD) (I). The silicon masking layer is patterned using double sided litho- graphy (AZ5214E photoresist, Clariant) and an isotropic silicon wet etchant (44 : 1 : 18 HNO 3 –NH 4 F–H 2 O, etch rate ca. 100 nm min 21 ). After resist removal, the glass is etched at about 8 mm min 21 in 10 : 1 HF–HCl solution 10 (II). When the etch has proceeded through the wafer the remaining silicon is removed in 25% tetramethylammonium hydroxide (TMAH) solution at 80 uC. An RCA-1 (80 uC 5 : 1 : 1 NH 4 OH–H 2 O 2 – H 2 O) cleaning step is done prior to fusion bonding to another glass wafer in a furnace at 650 uC (III). A 17 nm chromium adhesion layer and a 200 nm platinum film are sputtered onto a Micro and Nanosciences Laboratory, Helsinki University of Technology, P.O. Box 3500, FI-02015 TKK, Finland. E-mail: ville.saarela@tkk.fi; Fax: +358 9 451 6080; Tel: +358 9 451 4627 b Division of Pharmaceutical Chemistry, Faculty of Pharmacy, P.O. Box 56, FI-00014 University of Helsinki, Finland c Laboratory of Analytical Chemistry, Department of Chemistry, P.O. Box 55, FI-00014 University of Helsinki, Finland Fig. 1 Fabrication process of the glass nebulizer chip. TECHNICAL NOTE www.rsc.org/loc | Lab on a Chip 644 | Lab Chip, 2007, 7, 644–646 This journal is ß The Royal Society of Chemistry 2007