Contents lists available at ScienceDirect Vacuum journal homepage: www.elsevier.com/locate/vacuum Direct determination of oxygen and other elements in non-conducting crystal materials by pulsed glow discharge time-of-flight mass spectrometry with potassium titanyl phosphate as an example Anna Gubal a,* , Alexander Ganeev a,b , Victoria Bodnar a , Nikolay Solovyev a , Yegor Lyalkin a , Oleg Glumov a , Viktor Yakobson c a St. Petersburg State University, Institute of Chemistry, 199034, St. Petersburg, Universitetskaya nab. 7/9, Russia b Institute of Toxicology of Federal Medico-Biological Agency, 192019, St. Petersburg, ul. Bekhtereva, 1, Russia c Joint Stock Company "Research and Production Corporation S.I. Vavilova", Russia, 192171, St. Petersburg, ul. Babushkina 36-1, Russia ARTICLE INFO Keywords: Mass spectrometry Pulsed glow discharge Oxygen Potassium titanyl phosphate Direct analysis Dielectrics ABSTRACT Direct quantification of oxygen in dielectric materials using non-destructive or nearly non-destructive techniques still remains a nontrivial task. Simultaneous assessment of oxygen with other elements in a single analytical procedure is even more challenging. In the current study, a method of direct determination of oxygen and other matrix elements in solid samples, based on time-of-flight mass spectrometry with pulsed glow discharge in combined hollow cathode (CHC) is designed and tested. The possibility to effectively ionise oxygen owing to the electron impact mechanism under short repelling pulse delays has been shown. Stable sputtering and ionisation of dielectric samples were obtained via sample coating with thin conducting layer of silver. The parameters of oxygen quantification were optimised: duration and voltage of the discharge pulse, cell pressure, repelling pulse delay and material of the auxiliary cathode. The calibrations of oxygen, phosphorus and potassium are pre- sented. The intensity of 16 O + was shown to be highly dependent on discharge cell pressure. The limits of de- tection were 0.001, 0.001, and 0.002 mass% for oxygen, phosphorus and potassium respectively. The designed approach enables direct, fast and accurate quantitative and in depth analysis of oxygen-containing samples. 1. Introduction Oxygen is an omnipresent element and its quantitative determina- tion, both as necessary constituent and admixture, in various natural and technological materials claims high requirement to the analytical techniques employed [1–4]. Currently, the determination of oxygen in gaseous phase or that of dissolved oxygen usually does not pose a ser- ious challenge. Recent review by Wang et al. [5] comprehensively de- scribe the use of spectroscopic techniques in the determination of oxygen. The established methodology includes CNO-analysers [6], Winkler's titration [7,8], electrochemical [9] and optical methods [5]. Nevertheless, for the analysis of solids more problems arise, since the majority of methods applicable for oxygen quantification requires sample dissolution or analyte transfer into gaseous phase [10]. That causes high analytical uncertainty owing to analyte loss and con- tamination risks. Thus, simple direct techniques are needed in order to fulfil the high requirements to the material testing methods. The crystals of potassium titanyl phosphate KTiOPO 4 (KTP) and solid solutions containing it are well-known nonlinear optical materials [11,12]. The properties of KTP-containing materials are highly depen- dent on the growth method and conditions as well as on concentration and distribution of alloying ions and admixtures in the crystal [13–16]. However, owing to the complexity of the analysis and calibration, there are no reliable direct methods for the quantification of all constituting elements in KTP. Amongst these, oxygen is the most complicated ana- lyte, owing to its pronounced non-metal properties, volatility and om- nipresence in the environment. The following techniques may be applicable for the determination of oxygen in solids: micro-FTIR [17], nuclear techniques (neutron ac- tivation analysis and proton induced gamma ray) [18,19], laser-in- duced breakdown spectroscopy (LIBS) [20], glow discharge optical emission spectrometry (GDOES) [21,22], X-ray techniques (X-ray fluorescence – XRF, energy dispersive X-ray scanning electron micro- scopy – EDX SEM) [23] and mass spectrometry [21]. X-ray based ap- proaches usually have relatively high limits of detection (LoDs), which are mostly related to the high matrix effects for the light elements [24]. Such LoDs are usually inacceptable for the determination of oxygen at admixture level. Nuclear techniques are capable to quantify oxygen https://doi.org/10.1016/j.vacuum.2018.04.034 Received 5 March 2018; Received in revised form 6 April 2018; Accepted 20 April 2018 * Corresponding author. E-mail address: a.r.gubal@spbu.ru (A. Gubal). Vacuum 153 (2018) 248–253 Available online 22 April 2018 0042-207X/ © 2018 Elsevier Ltd. All rights reserved. T