1: \begin{abstract}

2: Strong and weak approximation errors of a spatial finite element method are analyzed for  stochastic partial 

3: differential equations(SPDEs) with one-sided Lipschitz coefficients, including the stochastic Allen--Cahn equation, driven by additive noise.

4: In order to give the strong convergence rate of the  finite element method,

5: we present an appropriate  decomposition

6: and some a priori estimates of the discrete stochastic convolution. To the best of our knowledge, there has been no essentially sharp weak convergence rate of spatial approximation for parabolic SPDEs

7: with non-globally Lipschitz coefficients. 

8: To investigate the weak error,

9: we first regularize the original equation by the splitting technique and derive the 

10: regularity estimates of the corresponding regularized Kolmogorov equation.

11: Meanwhile, we present the regularity estimate in Malliavin sense and the refined estimate of the  finite element method. 

12: Combining the 

13: regularity estimates of regularized Kolmogorov equation  with Malliavin integration by parts formula, the weak convergence rate  is shown to be twice the strong convergence rate. 

14: %In particular, when $T$ or noise effect is small enough, the strong convergence rate can reach the same order as the weak convergence rate.

15: \end{abstract}

16: