1 INTRODUCTION
Many important decisions are based on the results of chemical quantitative analysis, so the quality of the results is very important. This has led the use of "official methods" to fulfill legislative and trading requirements. The "Guide to the Expression of Uncertainty in Measurement" was publicized by ISO [1]. The guide established general rules for evaluation and expressing uncertainty in measurement across a broad spectrum of measurements. Most of previous uncertainty researches were atomic absorption spectrometry (AAS)[2] and liquid chromatography (LC) [3]. Recently, a procedure to calculate the uncertainty associated to a binary response, of the type yes/no, originated from an instrumental screening system was described by Pulido and his coworkers, and its application in differential pulse anodic stripping voltammetry (DPASV) screening system to analyse tap water, specifically, for determining Pb²⁺ and Cd²⁺ above the UE legislation limits [4]. In this study, uncertainty in determination of selenocystine(SeCys) by differential pulse voltammetry (DPV) using selenium-gold film modified glassy carbon electrode ((Se-Au)/GC electrode) is evaluated.

3. RESULTS AND DISCUSSION
3.1 Analysis of uncertainty sources                     
As it can be seen from the Fig. 1, the uncertainty sources mainly include concentration, volume and mass of the solution and repeatability of sample treatment and electrode.

　
Fig. 1 Initial cause and effect diagram

3.2 Quantification of the uncertainty sources
3.2.1 SeCys concentration C_x                    
For standard addition method, u (C_x) is given by
（1）,
with the residual standard deviation s given by （2）,
and （3）.
where
b :slope
n : number of measurement for the calibration
C_x : determined SeCys concentration of the solution
: mean value of the different calibration standards (n number of measurements)
j : index for the number of measurements to obtain the calibration curve
    For selenium-enriched yeast, the uncertainty of SeCys concentration is ; for selenium-enriched tea, the uncertainty of SeCys concentration is .

3.2.2 Volume V (V₁, V₂, V₃)
Calibration: The volume is calibrated according to the manufacturer's specification within the range of ±0.06mL and ±0.04mL for 50mL graduated cylinder and for 10mL transfer pipet, respectively. The standard uncertainty is obtained assuming a triangular distribution. The uncertainties of volume calibration are given by _,  .
    Temperature: The temperature of the experiment has to be 20±2 ^oC. This temperature range leads to an uncertainty in the determined volume, due to a considerable larger volume expansion of the liquid compared with the vessel. The coefficient of volume expansion for water is a=2.1×10^{-4 o}C^-1. The standard uncertainty of volumes of 50mL and 10mL, assuming a rectangular temperature distribution, is
_,   .
    The two contributions are combined to give the standard uncertainty u (V) of the volume V: （4）.
    For selenium-enriched yeast or tea, the uncertainty of 10 mL transfer pipet (V₁) is given by ; two 50 mL graduated cylinders were used for stored solution (V₂) and determination solution (V₃), respectively, and the uncertainty is given by .
3.2.3 Mass m
Because of the combined repeatability term identified above, there is no need to take into account the weighing repeatability.
    The weighing procedure is a weight by difference. There are two branches, one branch is for the determination of the tare (m_tare) and the other branch is for the gross weight (m_gross).
    Sensitivity: If the weighing is done on the same scale and over a small range of weight then the sensitivity contribution can be neglected.
    Linearity: The calibration certificate of the balance quotes ±0.2mg for the linearity. The balance manufactures own uncertainty evaluation recommends the use of a rectangular distribution to convert the linearity contribution to a standard uncertainty.
    Readability: The readability of analytical balance is 0.1mg.
    These give, for the standard uncertainty u (m) of the mass m, a value of
.
3.2.4 Repeatability R
The uncertainty for repeatability (U(R)) can be expressed as a relative standard deviation (RSD). u(R) is given by （5）,
where u(R_e) is the uncertainty for repeatability of electrode expressed as RSD (4.19% in table 1) when using three modified electrodes prepared at different times for determination the same sample, and u(R_s) is the uncertainty in sample treatment expressed as RSD when using the same electrode for determination samples prepared at different times or by different persons. u(R_s) presented in Table 2 in determinations selenium-enriched yeast and tea are 9.15%, 5.63% respectively. So the uncertainty for electrode repeatability is smaller compared with the repeatability uncertainty in sample treatment. The results of u(R) in determinations selenium-enriched yeast and tea are shown in Table 3.

3.2.5 Combined standard uncertainty u(r)
The concentration of SeCys in samples is given by （6）.
    In order to calculate the combined standard uncertainty of a multiplicative expression (as above) the standard uncertainties of each component are as follows:
    （7）.
3.2.6 Expanded uncertainty U
To increase the confidence level , the uncertainty is expressed by the expanded uncertainty (U), which is given by U =u (r)×k = u (r)×2. k is a coverage factor and when k is 2, confidence level is 95%.