Monitoring of trace heavy metal ions in ores samples before the mining, smelting and processing is of great importance due to it high toxicity and gradual accumulation in the environment as well as in animal or human organs. The well-known atomic spectrometry analytical instruments, such as atomic fluorescence spectrometry (AFS), atomic absorption spectrometry (AAS) and inductively coupled plasma-atomic emission spectrometry (ICP-AES), have been extensively employed for the determination of metal elements in various complex samples. However, these analytical instruments require bulky and costly devices, high power and large gases consumption. These shortcomings restrict their use within laboratory, preventing their use for field measurement and continuous monitoring. To meet the trend of miniaturization in analytical instrumentation and the requirements of on-line detection in field, electrolyte cathode discharge (ELCAD) has been developed by Cserfalvi in 1993 as an important tool in atomic spectrum analysis for element determination of liquid samples. In the original apparatus of ELCAD, the sample solution is acted as cathode, which overflows with typical flow rate of 8～10 mL·min-1 from a pipette into about 35 mL reservoir completely filled with electrolyte solution, and a counter-electrode (mostly W or Ti rod) above it (2～4 mm) is the anode. The pipette is immersed into electrolyte solution and then curved upwards about 1～3 mm from the reservoir containing a grounded graphite electrode to make it electrically conductive. Since then, in order to improve the emission efficiency and discharge stability, many improvements for excitation source of ELCAD have been developed. In the present work, a novel liquid cathode glow discharge (LCGD) was successfully constructed based on the principle of ELCAD, in which the glow discharge plasma was generated between the needle-like Pt anode (diameter 0.5 mm) and electrolyte (served as the liquid cathode) overflowing from a quartz capillary (1.0 mm inner diameter). The vertical gap between capillary and pointed Pt wire is 2 mm. The quartz capillary was inserted into a graphite tube and protruded from the graphite tube about 2.5 mm. The sample solution was introduced through the quartz capillary with the aid of a peristaltic pump at flow rate 4.5 mL·min-1, and then flowed over the top of capillary into the grooves on the graphite tube. This device can offer several advantages over conventional ELCAD. For example, sealed Pt wire into a quartz tube can form a Pt tip discharge and make the energy focus on a very tiny spot, which has lower energy consumption (<60 W) and higher excitation efficiency. In addition, several knots in peristaltic-pump tubing can reduce signal fluctuations of discharge induced by the peristaltic pump and improve the stability of discharge plasma. Furthermore, inserted the quartz capillary into graphite tube is excluded the reservoir of ELCAD, which can reduce the consumption of solution samples. To evaluate the analytical performance of LCGD, the simultaneous determination of Pb and Zn in digested refined copper ores samples with HNO3-HCl was carried out. The stability of LCGD and the effects of discharge condition, supporting electrolyte, solution pH and solution flow rate on emission intensity were systematically investigated. The limits of detections (LODs) of Pb and Zn were compared with those measured by closed-type ELCAD. In addition, the measured results of samples using LCGD were verified by ICP. Moreover, a t-test between the analytical results obtained by LCGD-AES and ICP-AES was also used for estimating the uncertainty in analytical measurements. The results showed that the emission intensities increase markedly with the increase of the discharge voltage from 620 to 680 V. This is because a higher discharge voltage creates more high-energy electrons which collide with gaseous water and metal vapor in the excitation source, thus improving the excitation efficiency of metal. Considering the discharge stability, excitation efficiency and lifetime of the electrodes, 650 V was selected as the optimal discharge voltage. The emission intensities are increased when the flow rate increases from 2.5 to 4.5 mL·min-1 before slightly reducing over 4.5 mL·min-1. The increase of the emission intensity may be assigned to increasing the amounts of analytes which entered into the discharge region. The reduction of emission intensity over 4.5 mL·min-1 may be ascribed to too much water loading and evaporating, which can consume the energy in the discharge region and reduce the efficiency of exciting the atoms. Therefore, 4.5 mL·min-1 was selected as the optimal flow rate. It is observed that pH=1 HNO3 has higher emission intensities. Hence, pH=1.0 HNO3 was selected as the optimum solution pH. Under the best analyzing conditions, the limits of detections (LODs) of Pb and Zn obtained from this method are 0.38 and 0.59 mg·L-1, respectively. The relative standard deviation (RSD) is 0.9% for Pb and 1.2% for Zn. The power consumption is below 60 W. LOD of this method has higher than that of other ELCAD. This may be associated with the selected spectrometer. The emission intensity remains almost unchanged under the same discharge condition, suggesting that the glow plasma is very stable. The recoveries of Pb and Zn are in the range of 87.6%~107.4%. The results of refined copper ores samples using LCGD are well consistent with the comparing values of ICP and there is no significant difference between the two methods. Compared with ICP, LCGD has several advantages, such as low power consumption, high excitation efficiency and easy miniaturization. It may be developed as a miniaturized analytical instrument for on-site, real-time and on-line determination of metal elements with further improvement.