is apparently not water of hydration. Gua HCl 1 (originally the dihydrate) was still absorbing moisture after 3 days and the 6 percent gain in mass was approaching the 8 percent equivalent of 1 H2O.

All calorimetric samples of the guanine hydrochlorides were the dried material. Transfers to the sample holder were made in a dry atmosphere to prevent absorption of moisture by the samples.

2.2. Elemental Analysis of the Guanine Hydrochlorides

Portions of the guanine hydrochloride samples which had been heated under vacuum for 70 h or longer (see table 1) were transferred in a glovebox to glass vials with tight-fitting plastic caps for shipment to the analytical laboratory. Four days later, the microanalyses were performed; the results are given in table 3. The compositions of Gua⚫ HCl 1 and 2c correspond to that of the anhydrous monohydrochloride, but that of Gua HCl 2d is closer to the hemihydrochloride. This might be expected since the latter sample was recrystallized from a more dilute HCl solution than Gua HCl 2 c.

the dried materials were used in the calorimetric experiments. It will be assumed that samples 1 and 2c are the anhydrous monohydrochloride and that sample 2d is the anhydrous hemihydrochloride although these may be erroneous assumptions.

2.3. Density

The densities of several samples of guanine and the guanine hydrochlorides were measured by a displacement method in 25-cm3, Gay-Lussac-type pycnometers using Eastman ACS spectroscopic grade CCL (density = 1.5898 g.cm3 under the laboratory conditions; this is the mean of 3 measurements for which the average deviation was 0.0002 g.cm3). Details of the method are described in the first paper of this series [1]. The results of these measurements are given in table 4.

After the density measurement for Gua 1, the sample was collected on filter paper, dried in air, and weighed; there was a loss in mass of~ 1 mg or <0.1 percent. When the filtrate was evaporated to dryness there was no visible residue. Thus, the solubility of guanine in CCL, was insignificant in the density determinations.

The results of density measurements given in table 4 for Gua 1, 3, 4,and Gua HCl 1 were obtained on samples as received from commercial sources; the other measurements were made on the dried samples (see sec. 2.1). For calculating the buoyancy factor, 1.000546, used in this work to cor

This method for determining densities proved unsatisfactory for the guanine hydrochlorides. The densities measured did not agree with the elemental analyses. It was suspected that impurities in the CCL reacted with the dried samples to change their compositions during the density

measurements.

It was reported in Beilsteins Handbuch [8] that the hydrogen chloride salt of guanine may crystallize as the monohydrate (molar mass = 205.60), as the dihydrate (molar mass = 223.62), or as guanine dihydrochloride (molar mass = 224.05). Broomhead [9] prepared two crystalline guanine hydrochloride samples. The first was found to have molar mass = 223, density = 1.562 g cm3, and microanalysis indicated that it was the monohydrochloride dihydrate rather than the dihydrochloride. The second sample had density = 1.662 g cm3 and molar mass = 205.2, and was apparently the monohydrate. (Details of the preparations and measurements were not given.)

Our density measurements confirmed that Gua· HCl 1 (as received) was the dihydrate (although the anhydrous material prepared from it was used in the calorimetric experiments). However, the measurements on the dried samples did not agree with the elemental analyses. Therefore, these

tallographic lattice. It is not possible to draw conclusions about the states of hydration.

No impurities were identified or detected from analyses by paper and thin layer chromatography (TLC) in the samples of Gua 1, 2, 3, 4, and 4b. Details of the procedures and detection limits were described previously [1]. The spotting solutions contained~ 0.01 mol Gua/L of aqueous HCl or NaOH (1 mol·L-1). (Guanine was nearly insoluble in the hot NH4OH used for the other bases). The R, values (distance traveled by the major component/distance traveled by the solution) were determined for the 5 guanine samples using TLC plates coated with 250 nm of MN300 Cellulose(C), or (CF) which also contains a fluorescent indicator, and Whatman No. 1 and No. 40 chromatography papers, and 4 different carrier solutions described in table 5. The R, values obtained for guanine samples are compared with those given by the National Academy of Sciences (NAS) [11] for similar solutions and procedures. The NAS values are lower than in

volatile matter found was less than in the other three samples from commercial sources. It was assumed that the discoloration in Gua 3 was from impurity although none was detected or identified in the analytical work described in section 2. Gua 4 was also slightly discolored and was used in preparing other samples of guanine and guanine hydrochlorides by reprecipitation and recrystallization. A few measurements of the enthalpy of solution were made with samples other than Gua 1 for comparison.

In table 6 are the data from the measurements of the enthalpy of solution in HCl solutions of various concentrations; the arrangement of the experiments is in order of increasing HCl concentration. The mass of guanine was ~0.2 g (1.2 to 1.5 mmol) and the volume of the solution was ~ 300 cm3 in all experiments. The experiments are divided into 3 groups because the measured values for the AC, of the reaction were significantly different at different HCl concentrations.

1

Table 7 shows the Expt. Nos. (from table 6) for which values of Q, and T, were used in calculating Ac, for the reaction at five concentrations of HCl. The Ac, for group 1, 5.39 J.g1 K1, was derived from only one pair of experiments at the concentration, 0.1013 mol L-'. It was applied in the calculation of Corrsr for this entire group because a precipitate formed in the final solutions of all these experiments, but not in those of Groups 2 and 3. At first, this undissolved material was believed to be unreacted guanine. However, observation of a non-calorimetric experiment revealed the immediate formation of a voluminous white precipitate (presumably the hydrochlorides since there was an excess of Cl- in even the most dilute solution), only part of which dissolved in the dilute solutions. The concentration where the dissolution of this precipitate was complete is shown graphically in figure 1 which is a plot of the values given in table 6 for AH(298.15 K) as a function of HCl concentration. The values for the enthalpy of solution increase sharply with increasing HCl concentration until a precipitate is no longer visible in the final solution,~ 0.2 mol· L ̈1.

-1 •

The values measured for Ac, of the reaction (table 7) at the HCl concentrations, 0.307 and 3.46 mol L-', agreed, and the mean value, 0.65 J.g1 K-1, was used in calculating CorrST for the experiments in Group 2. At HCl concentrations, 5.45 and 6.09 mol· L-1, the values for Ac, also agreed, and the mean value, 1.06 J.g1.K', was used to obtain Corrsr for experiments in Group 3. It was questionable whether the Ac, value from Group 2 or 3 should be used for the 6 experiments using HCl solutions between the concentrations, 3.46 and 5.45 mol L-1. The use of either value for Ac, resulted in lower values for AH(298.15 K) than would correspond to the curve for Group 2 (see fig. 1). Therefore, it was assumed that the second protonation occurred at the HCl concentration of 4 mol·L1 or less, and the 6 experiments in question were included in Group 3.