TRISTIMULUS COLOR MEASUREMENTS IN FADING STUDIES 373 Throughout this discussion the major concern will be with "color," not the dye or colorant content in the dosage form. This point must be emphasized because it is the crux of the entire technique. "Color" and colorant content of a dosage form are not necessarily the same. Consider a new product which is to be colored green. There is a choice of two green colorants and a blue and yellow combination that will result in green. That is the total colorant gamut for this color. It is also desired to color this product with a colorant or colorants such that it will appear unchanged to the consumer for the longest possible period of time. It should be noted that it is important that the product need only appear to remain unchanged--not that its dye or pigment con- tent remain the highest or that its chemical degradation rate be the slowest. The objective is a measure of the total effect that the consumer sees when he looks at the product. This includes the background in the product caused by any and all additives. In other words, the colorant is of interest not as an isolated chemical entity but as an integral part of a specific, well-defined formulation. It is essential to achieve a situation where a consumer cannot tell any difference in color between the last of one purchased product and a fresh container of the same item. For this study a green insoluble colorant (non-FDA approved), an aluminum lake, and a combination of a blue and a yellow FD&C dye were used to achieve three blue-green colored compacts.* A white con- trol was also made. The insoluble colorants were first mixed with lac- tose and passed through a Mikro-Samplmill fitted with a number 0.010 HB screen. They were then granulated using an acacia solution as the binding agent, dried for sixteen hours at 54 øC, and subsequently screened through a fine screen on a comminution mill. The two soluble dyes were added to the lactose as a solution. This was then air-dried, finally screened, and subsequently similarly treated as the other two granula- tions. The granulations were compressed on a Carver Press using 1 •/• in. (32 ram) bevel-edge, flat faced punches, each compact being com- pressed at the same pressure. To effect the color change the compacts were mounted on a white board and placed in windows having a southern exposure. They were exposed to sunlight from the hours of 11 a.m. to 4 p.m. each day. The amount of sunlight that the tablets received each day was measured by the use of a Sunlight Integrator manufactured by the Self-Organizing Systems, Inc.• The instrument is designed to measure the solar energy * The percentage of colorant in each formulation is deliberately omitted because products are colored up to a certain depth, within certain legal limits, and the interest here is to follow the change in the color. } 6612 Denton Drive, Dallas, Tex.

374 JOURNAL OF THE SOCIETV OF COSMETIC CHEMISTS at the sensor over a period of time. A cadmium sulfide light-dependent resistor is used as the sensor and a solion tetrode as the integrator. The instrument is calibrated against an Eppley Pyrheliometer to give one count for each 2 cal/cm 2 or one count per two Langley units. The Langley unit is used by the U. S. Meteorological Service so that the number of units to which the tablets have been exposed may be trans- posed into the equivalent number of days or weeks of sunshine in any one of several major areas in the United States. Figure 5 is a photograph of the Sunlight integrator showing the sens- ing unit. The spectral response of the light-detecting unit is shown in Fig. 6. Its maximum sensitivity lies in the visible range. The compacts were exposed to the sunlight for varying periods of time. As soon as a difference in color was detected in each formulation, ::.•.. .•,• % .... ... .... . :/ ---...: , ? **.•-• - .:' ..: . . :•:.,.. ...... .:,,:•v•:... ....... ... :::: Figure 5. Sunlight integrator and sensing unit a compact was removed from the exposure area and placed in a case to await its subsequent evaluation by reflectance spectrophotometry. When a series of compacts had reached a point of nonacceptability as determined by laboratory personnel making a visual evaluation, the particular test series was discontinued. When all of the colored tablets, along with the uncolored control, had been exposed to the sunlight for the desired periods of time, the samples of an entire series were measured to determine their changes in reflectance. The tablets were measured us- ing a G. E. recording spectrophotometer equipped with a D&H integra-