4.21: Calculating the AB Proton NMR Using Tensor Algebra
- Page ID
- 150937
The purpose of this tutorial is to deviate from the usual matrix mechanics approach to the ABC proton nmr system in order to illustrate a related method of analysis which uses tensor algebra. For a discussion of the traditional approach for the ABC system visit http://www.users.csbsju.edu/~frioux/nmr/Speclab4.htm. This site also provides general information on the quantum mechanics of nmr spectroscopy.
\[ \begin{matrix} \text{Nuclear spin and identity operators:} & I_x = \frac{1}{2} \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix} & I_y = \frac{1}{2} \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix} & I_z = \frac{1}{2} \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix} & I = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix} \\ \text{Chemical shifts:} & \nu_A = 250 & \nu_B = 300 & \text{Coupling constant:} & \text{Jab} = 10 \end{matrix} \nonumber \]
Hamiltonian representing the interaction of nuclear spins with the external magnetic field in tensor format:
\[ \begin{matrix} \hat{H}_{mag} = \nu_A \hat{I}_z^A - \nu_B \hat{I}_z^B = - \nu_A \hat{I}_z^A \otimes \hat{I} + \hat{I} \otimes \left( - \nu_B \hat{I}_z^B \right) & \text{where, for example,} & \nu_A = g_n \beta_n B_z(1- \sigma_A) \end{matrix} \nonumber \]
Implementing the operator using Mathcad's command for the tensor product, kronecker, is as follows.
\[ H_{mag} = - \nu_A \text{kronecker}(I_z,~I) - \nu_B \text{kronecker}(I,~I_z) \nonumber \]
Hamiltonian representing the interaction of nuclear spins with each other in tensor format:
\[ \hat{H}_{spin} = Jab \left( \hat{I}_x^A \otimes \hat{I}_x^B + \hat{I}_y^B \otimes \hat{I}_y^B + \hat{I}_z^A \otimes \hat{I}_z^B \right) \nonumber \]
Implementation of the operator in the Mathcad programming environment:
\[ H_{spin} = Jab \left( \text{kronecker}(I_x,~I_x) + \text{kronecker}(I_y,~I_y) + \text{kronecker}(I_z,~I_z) \right) \nonumber \]
The total Hamiltonian spin operator is now calculated and displayed.
\[ H = H_{mag} + H_{spin} \nonumber \]
\[ \begin{matrix} \begin{array} \alpha \alpha \alpha & & \alpha \beta & & \beta \alpha & & \beta \beta \end{array} \\ \begin{pmatrix} -272.5 & 0 & 0 & 0 \\ 0 & 22.5 & 5 & 0 \\ 0 & 5 & -27.5 & 0 \\ 0 & 0 & 0 & 277.5 \end{pmatrix} \begin{array} \alpha \alpha \alpha \\ \alpha \beta \\ \beta \alpha \\ \beta \beta \end{array} \end{matrix} \nonumber \]
Calculate and display the energy eigenvalues and associated eigenvectors of the Hamiltonian.
\[ \begin{matrix} i = 1 .. 4 & E = \text{sort(eigenvals(H))} & C^{<i>} = \text{eigenvec}(H,~E_i) \end{matrix} \nonumber \]
\[ \begin{matrix} \text{augment}(E,~C^T )^T = \begin{pmatrix} -272.5 & -27.995 & 22.995 & 277.5 \\ 1 & 0 & 0 & 0 \\ 0 & -0.099 & 0.995 & 0 \\ 0 & 0.005 & 0.099 & 0 \\ 0 & 0 & 0 & 1 \end{pmatrix} \begin{array} \alpha \alpha \alpha \\ \alpha \beta \\ \beta \alpha \\ \beta \beta \end{array} \end{matrix} \nonumber \]
The nmr selection rule is that only one nuclear spin can flip during a transition. Therefore, the transition probability matrix for the AB spin system is:
\[ \begin{matrix} T = \begin{pmatrix} ' & \alpha \alpha & \alpha \beta & \beta \alpha & \beta \beta \\ \alpha \alpha & 0 & 1 & 1 & 0 \\ \alpha \beta & 1 & 0 & 0 & 1 \\ \beta \alpha & 1 & 0 & 0 & 1 \\ \beta \beta & 0 & 1 & 1 & 0 \end{pmatrix} & T = \begin{pmatrix} 0 & 1 & 1 & 0 \\ 1 & 0 & 0 & 1 \\ 1 & 0 & 0 & 1 \\ 0 & 1 & 1 & 0 \end{pmatrix} \end{matrix} \nonumber \]
Calculate the intensities and frequencies of the allowed transitions.
\[ \begin{matrix} i = 1 .. 4 & j = 1 .. 4 & I_{i,~j} = \left[ C^{<i>} \left( TC^{<j>} \right) \right]^2 & V_{i,~j} = \text{if} \left( I_{i,~j} > .001,~ \left| E_i - E_j \right|,~ 0 \right) \end{matrix} \nonumber \]
\[ \begin{matrix} \text{Intensity matrix:} & I = \begin{pmatrix} 0 & 0.8 & 1.2 & 0 \\ 0.8 & 0 & 0 & 0.8 & \\ 1.2 & 0 & 0 & 1.2 \\ 0 & 0.8 & 1.2 & 0 \end{pmatrix} & \text{Frequency matrix:} & V = \begin{pmatrix} 0 & 244.5 & 295.5 & 0 \\ 244.5 & 0 & 0 & 244.5 \\ 295.5 & 0 & 0 & 254.5 \\ 0 & 305.5 & 254.5 & 0 \end{pmatrix} \end{matrix} \nonumber \]
Display the calculated AB nmr spectrum: