The proton free induction decay of a portland cement paste in an advanced stage of hydration can be roughly divided into three main components: (1) a component with a very short spin-spin relaxation time, T,, representing the protons of the solid OH groups and the water of crystallization, (2) a component with an intermediate T , representing the bonded water in the gel phase, and (3) a third component with a relatively long T2 representing the water in the micropores and layers. The dependences of the intensities, Tz's, and spin-lattice relaxation times (Ti's) of these three components on the cement hardening time have been determined. The proton spin-lattice relaxation time of the "solid" component increases with hardening time whereas T1 decreases for the other two components. The observed time dependence of the diffusion coefficient, D , of water in a tricalcium silicate paste supports the findings of the above correlation study.

The nuclear magnetic resonance spin-grouping technique has been applied to dentin from human donors of different ages. The apparent T2, T1, and T1 rho have been determined for natural dentin, for dentin which has been dried in vacuum, and for dried dentin which has been rehydrated in an atmosphere with 75% relative humidity. All apparent spin relaxation has been analyzed for exchange between the spin groups in which the dentin protons exist; the analyses incorporate the results of selective inversion recovery T1 measurements which better probe the effects of exchange. The exchange analyses of the high fields and rotating frame spin-lattice relaxation have also been correlated to determine uniquely the inherent relaxation parameters of the proton spin groups constituting the dentin magnetization. The natural dentin contains protons on water, protein, and hydroxy apatite; these spins contribute 50%, 45%, and 5% to the total dentin proton magnetization, respectively. The water exists in three distinct environments, the dynamics of each environment has been modeled. In the natural dentin 30% of the water undergoes uni-axial reorientation. 52% of the water has similar relaxation characteristics to bound water hydrating a large molecule, and the majority of the remaining water acts as bulk water undergoing isotropic reorientation. The results are independent of the age of the donor.

The spin-spin, T2, and spin-lattice, T1, relaxation times and the magnetization of protons were measured in human enamel. The proton free induction decay was analyzed into solid-like interstitial water, enamel apatite, and semiliquid-like water components. The solid-like interstitial water was evaluated to be approximately 5 wt% and the semiliquid-like water to be approximately 1 to 2%. Neither in wet nor in dry natural enamel does the solid water exchange upon deuteration nor can it be extracted in vacuum. The semiliquid natural water, which is in the closed pores of the structure in the two samples above also remains unexchanged upon an 8 hr deuteration and cannot be extracted in a vacuum. With the lineshape-relaxation correlation NMR the free induction decays from heads and tails of the enamel rods were resolved. The solid-like water with T2 of approximately 14 microseconds and the apatite with a T2 of approximately 61 microseconds have T2's too short to be observed in an NMR zeugmatogram. Therefore only the semiliquid water component having an apparent T*2 of about 240 microseconds would contribute to the NMR image of human enamel. Since the relative intensity of this proton magnetization component in tooth is quite small the NMR image of tooth would show primarily the dentin and liquids within the tooth and on its surface.