Si:WO3 sensors for highly selective detection of acetone for easy diagnosis of diabetes by breath analysis

Marco Righettoni, Antonio Tricoli, Sotiris E. Pratsinis

Research output: Contribution to journalArticlepeer-review

572 Citations (Scopus)


Acetone in the human breath is an important marker for noninvasive diagnosis of diabetes. Here, novel chemo-resistive detectors have been developed that allow rapid measurement of ultralow acetone concentrations (down to 20 ppb) with high signal-to-noise ratio in ideal (dry air) and realistic (up to 90% RH) conditions. The detector films consist of (highly sensitive) pure and Si-doped WO3 nanoparticles (10-13 nm in diameter) made in the gas phase and directly deposited onto interdigitated electrodes. Their sensing properties (selectivity, limit of detection, response, and recovery times) have been investigated as a function of operating temperature (325-500 °C), relative humidity (RH), and interfering analyte (ethanol or water vapor) concentration. It was found that Si-doping increases and stabilizes the acetone-selective e-WO3 phase while increasing its thermal stability and, thus, results in superior sensing performance with an optimum at about 10 mol % Si content. Furthermore, increasing the operation temperature decreased the detector response to water vapor, and above 400 °C, it was (≤0.7) always below the threshold (10.6) for fake diabetes detection in ideal conditions. At this temperature and at 90% RH, healthy humans (≤900 ppb acetone) and diabetes patients (≥1800 ppb) can be clearly distinguished by a remarkable gap (40%) in sensor response. As a result, these solid state detectors may offer a portable and cost-effective alternative to more bulky systems for noninvasive diabetes detection by human breath analysis.

Original languageEnglish
Pages (from-to)3581-3587
Number of pages7
JournalAnalytical Chemistry
Issue number9
Publication statusPublished - 1 May 2010
Externally publishedYes


Dive into the research topics of 'Si:WO3 sensors for highly selective detection of acetone for easy diagnosis of diabetes by breath analysis'. Together they form a unique fingerprint.

Cite this