Experimental and theoretical investigations of the reactions NH(X 3Sigma-) + D(2S)-->ND(X 3Sigma-) + H(2S) and NH(X 3Sigma-) + D(2S)-->N(4S) + HD(X 1Sigmag+)

The rate coefficient of the reaction NH(X (3)Sigma(-))+D((2)S)-->(k(1) )products (1) is determined in a quasistatic laser-flash photolysis, laser-induced fluorescence system at low pressures. The NH(X) radicals are produced by quenching of NH(a (1)Delta) (obtained in the photolysis of HN(3)) with Xe and the D atoms are generated in a D(2)/He microwave discharge. The NH(X) concentration profile is measured in the presence of a large excess of D atoms. The room-temperature rate coefficient is determined to be k(1)=(3.9+/-1.5) x 10(13) cm(3) mol(-1) s(-1). The rate coefficient k(1) is the sum of the two rate coefficients, k(1a) and k(1b), which correspond to the reactions NH(X (3)Sigma(-))+D((2)S)-->(k(1a) )ND(X (3)Sigma(-))+H((2)S) (1a) and NH(X (3)Sigma(-))+D((2)S)-->(k(1b) )N((4)S)+HD(X (1)Sigma(g) (+)) (1b), respectively. The first reaction proceeds via the (2)A(") ground state of NH(2) whereas the second one proceeds in the (4)A(") state. A global potential energy surface is constructed for the (2)A(") state using the internally contracted multireference configuration interaction method and the augmented correlation consistent polarized valence quadrupte zeta atomic basis. This potential energy surface is used in classical trajectory calculations to determine k(1a). Similar trajectory calculations are performed for reaction (1b) employing a previously calculated potential for the (4)A(") state. The calculated room-temperature rate coefficient is k(1)=4.1 x 10(13) cm(3) mol(-1) s(-1) with k(1a)=4.0 x 10(13) cm(3) mol(-1) s(-1) and k(1b)=9.1 x 10(11) cm(3) mol(-1) s(-1). The theoretically determined k(1) shows a very weak positive temperature dependence in the range 250< or =TK< or =1000. Despite the deep potential well, the exchange reaction on the (2)A(") ground-state potential energy surface is not statistical.