Taxus and Taxol - A Compilation of Research Findings

Special Circular 150-99

Thin-Layer Drying of Cultivated Taxus Clippings

Robert C. Hansen,
Harold M. Keener,
Department of Food, Agricultural, and Biological Engineering,
Ohio Agricultural Research and Development Center,
The Ohio State University,
Wooster, Ohio.

Hala N. ElSohly,
Research Institute of Pharmaceutical Sciences,
School of Pharmacy,
The University of Mississippi,
University, Miss.

Originally published in the Transactions of the ASAE. Nov.-Dec. 1993, Vol. 36 No. 6, 1873-1877. American Society of Agricultural Engineers. St. Joseph, Mich. Reprinted with permission.

Abstract

The ornamental yew, Taxus x media 'Hicksii,' has been identified as a renewable source of Taxol. However, clippings from the plant must be properly and efficiently harvested, dried, and stored. Thin-layer drying studies of Taxus clippings, needles, and stems were conducted. Parameters for the thin-layer drying equation were successfully determined for drying temperatures of 30°, 40°, 50°, and 60°C. The results showed that drying rates increased 28-, 15-, and 3-fold as drying temperatures increased from 30°, 40°, and 50°C to 60°C, respectively. Also, stems dried at a faster rate than needles, and needles dried at a faster rate than whole clippings. Taxol yields (g/100 g db) from stems were nearly constant for the four temperatures tested. However, yields from needles increased linearly as drying temperatures increased from 30° to 60°C. The highest Taxol yields were obtained from clippings. Nearly constant yields were obtained (~ 0.014 g/100 g db) for drying temperatures of 40°, 50°, and 60°C. The lowest yields for all three plant components occurred when the drying temperature was set at 30°C. The results suggested 60°C was the best temperature set point for drying Taxus x media 'Hicksii,' but temperatures higher than 60°C should be evaluated.

Introduction

Taxol has been obtained from the bark of the Pacific yew tree, Taxus brevifolia, which grows in forests primarily located in Oregon and Washington. Because of low concentrations of Taxol in the bark (less than 0.01%) and a limited supply of trees, research was directed to finding additional sources for the drug. The ornamental yew, Taxus x media 'Hicksii,' was found to have concentrations of Taxol on the order of 0.02% in clippings (Croom, 1991). Croom (1991) stated Taxol yields from Taxus clippings are very sensitive to drying temperatures and procedures. Ornamental yews are typically pruned on an annual basis as a part of standard nursery practices. For commercial Taxol production, the clippings must be properly and efficiently harvest- ed, dried, and stored. However, no published data exist on the effect of drying temperature on rate of drying and on extractable Taxol yield.

Objectives

The objectives of this research were to:

• Measure the rate of drying for Taxus clippings, needles, and stems as a function of temperature.
• Quantify parameters for the thin-layer drying equation.
• Measure the effect of drying temperature on Taxol yield.

Literature Review

Other than for purposes of propagation, Taxus clippings have not been routinely harvested or stored. Efforts to measure drying rates of any form of Taxus biomass have not been reported in research literature. However, drying of cereal grains and forage crops has been studied extensively (Hukill, 1947; Henderson and Perry, 1966; Brooker et al., 1974; Hall, 1980; Parker, 1991).

Conceptualization of drying involves the single particle alone and the particle en masse. Many drying models have been developed for both cases. When drying a single particle (or thin-layer drying), the empirically fitted curves as well as theoretical curves based on heat and mass transfer for rate of moisture loss take the general form:

(1)

where j takes values between 1 and 4.

(See nomenclature on page 52.)

Studies have shown the parameter values to be dependent upon material, type, temperature, moisture level and its distribution within the kernel, and airflow. For the case of j = 1, k is known as the drying constant and is evaluated from thin-layer drying curves by plotting on semilog paper (Hall, 1980).

For corn, Henderson and Henderson (1968), Thompson et al. (1968), and Barre et al. (1971) have expressed k as a function of the saturation vapor pressure and air velocity:

k = bp_s^mv_aⁿ      (2)

Using low-temperature (10°C) thin-layer drying data for corn, Sabbah et al. (1979) determined the constants b and m as 0.0722 and 0.722, respectively. The value for m derived by Henderson's and Thompson's data for shelled corn was 0.83.

According to Henderson and Pabis (1961) and Henderson and Henderson (1968), the effect of air velocity on the drying constant is trivial and, hence, the value of n becomes zero. However, when applying the drying constant to deep-bed drying, results of Barre et al. (1971) and Sabbah et al. (1979) give the value of n as 0.8 for corn.

Keener (1991) summarized values derived for k for grains based on published research. Most k values are reported as a function of temperature only. Other forms of thin-layer dryer equations have been proposed and are discussed by Brooker et.al. (1974). For this study on thin-layer drying, velocity was fixed and no evaluation of n was attempted.

Materials and Methods

Clippings were collected from the cultivar Taxus x media 'Hicksii' on a research farm near Wooster, Ohio. The plants had been growing on the site for 13 years. Cuttings were obtained December 17, 1991, and were representative of one full season of growth. Within one hour after cutting, the material was packaged in one-gallon sealable plastic storage bags after which the samples were stored and refrigerated at 3.5°C.

The laboratory dryer consisted of a blower that directed air past three electrical resistance heating coils with one connected to a variable voltage controller. The resistance heating unit was adjusted and calibrated to operate at four temperatures - 30°, 40°, 50°, and 60°C.

Figure 1. Schematic illustration of the thin-layer dryer showing the relative location of the drying trays

The drying unit contained nine stacked trays (see Figure 1). Thermocouples located above each tray were used to monitor air temperatures at each tray. An orifice plate was used to monitor airflow rate. Temperature of the air at the orifice plate, barometric pressure, and ambient dew point temperatures were also recorded. Temperature readings from the thermocouples were recorded every 15 minutes with a Digi III Kaye Datalogger and a MFE tape recorder.

Previous research results showed Taxol yields were diminished if Taxus needles were separated from stems before drying (Croom, 1991). Therefore, trays 1 to 5 were dedicated to drying clippings. Trays 6 and 7 were used to dry needles only, and trays 8 and 9 were used to dry stems only. This arrangement permitted a rough comparison of drying rates for clippings, needles, and stems while at the same time permitting trays 1 and 2 to provide data for evaluating thin-layer drying rates. In addition, Taxol yields obtained by separating needles and stems prior to drying could be compared to earlier findings reported by Croom (1991), and trays 3, 4, and 5 could provide a first look at the variability of Taxol yield within a drying zone.

One batch of clippings, needles, and stems was dried at each of the four temperatures noted previously. Each tray was 26.7 cm x 26.7 cm x 3.5 cm deep. The weight of each sample of standard clippings before drying was approximately 180 g. After drying, the final weight was about 75 g. Subsamples of needles were stripped from clippings and separated from stems. The needles that were dried in trays 6 and 7 were similar in weight to the whole clippings dried in trays 1 to 5. The stems, dried in trays 8 and 9, weighed approximately 100 g before drying and about 50 g after drying. Average cross-sectional airflow velocity was 0.039 m/s. The trays were 6.4 cm apart (Figure 1).

The Taxus samples were weighed at various times throughout each of the four tests. For example, for the test where the temperature was set at 60°C, samples were weighed at 0, 2, 4, 16, 19, 22, 28, and 46.5 h. At 40°C, weights were recorded at 0, 2, 4, 8, 12, 23.25, 27.25, 31.25, 38, 65, 77, 82.5, 86, and 101 h. The dryer door was opened and remained open during the time period required to remove each tray, weigh it, record it, and return it to the appropriate location in the dryer. The fan and the heater continued to operate during this time. The time period required to weigh all nine trays was about two minutes. When a constant weight was obtained for three consecutive readings, it was concluded that all moisture was volatilized, and the test was terminated. Relative humidity levels ranged from 21.8% for the 30°C test down to 0.3% for the 60°C test.

For calculations of moisture ratios in the results presented in this article, the equilibrium moisture content (M_e in Equation 1) was set equal to zero. Setting M_e equal to zero was appropriate since the 180 g samples dried at 30°, 40°, and 50°C reached equivalent final weights and were equal to the final weight for plant samples dried at 60°C. If the equilibrium moisture contents were significantly different than zero, final weights would have been different for one or more of the four drying conditions. Initial moisture content of fresh Taxus clippings ranged from 55 to 60% wet basis.

Taxol yields for the tests were determined by the Research Institute of Pharmaceutical Sciences, School of Pharmacy, The University of Mississippi. After a drying run was completed, the contents of each tray were sealed in plastic bags in preparation for shipment. All samples were stored at room temperature for about six weeks prior to shipment to Mississippi.

Results and Discussion

In general, the lowest trays dried at a faster rate than the higher ones, indicating that location was a factor and that the drying air picked up sufficient moisture to reduce rates of drying in the upper trays. However, temperature variations within trays 1 to 5 were less than 1°C after 8-12 hours for all tests.

In order to compare thin-layer drying rates for the four drying temperatures studied, moisture ratios as a function of time were averaged for trays 1 and 2 and plotted in Figure 2. The results showed that drying time decreased dramatically as drying temperatures increased. For example, total drying time for 40°C was nearly three days while 50°C required about one day; i.e., total drying time was reduced to one-third with only a 10°C increase in temperature. These conclusions are, of course, based on data for thin-layer drying and do not account for the effects of deep beds. Figure 3 illustrates the effect of temperature on total drying time.

Figure 2. Drying curves for Taxus x media 'Hicksii' clippings as a function of drying temperature	Figure 3. Total drying time for Taxus x media 'Hicksii' clippings as a function of drying temperature

Hall (1980) described a procedure for determining the drying constant (k) by drawing a semilogarithmic plot of moisture ratio (MR) vs. time (t). Since Equation 1 is an exponential relationship, the data should plot as a straight line for j = 1, where the slope is equal to k. Figure 4 is a semilogarithmic plot of the same results that are plotted in Figure 2, with the exception that the curves were truncated for MR < 0.1. The data come very close to fitting straight-line relationships.

Results of a regression analysis are shown in Table 1 where parameters are tabulated for the thin-layer drying equation (Equation 1) for each of the four drying temperatures tested. A regression analysis of k vs. p_s for drying temperatures tested gave:

k = 1.62 x 10^-10 p_s^2.233 (r² = 0.967)  (3)

A comparison of estimated drying rates for clippings, needles, and stems is shown in Figure 5 for a drying temperature of 40°C. To minimize the effect of bed depth, the results were obtained by averaging the drying rates for trays 4 and 5 for clippings, trays 6 and 7 for needles, and trays 8 and 9 for stems. Even though the stems were in the top two trays, the drying rate was much higher than it was for needles or clippings.

Figure 4. Semilogarithmic plot of drying curves for Taxus x media 'Hicksii' clippings as a function of drying temperature.	Figure 5. Drying curves for Taxus x media 'Hicksii' clippings as a function of plant parts. Temperature set point for dryer was 40#deg;C

Taxol yields (g/100 g db) are compared for clippings, needles, and stems as a function of drying temperature in Table 2 and Figure 6. The results indicated that yields from stems were unaffected by drying temperature; the yield from needles increased linearly as the temperature increased from 30° to 60°C; and the yield for clippings was nearly constant at 0.014% for 40°, 50°, and 60°C while being much lower at 0.008% for 30°C.

When comparing 30°, 40°, 50°, and 60°C drying temperatures, Figures 3 and 6 clearly showed that Taxus x media 'Hicksii' clippings should be dried at 60°C. Although the effects of depth of the drying bed have not been evaluated here, the results indicated that drying clippings at 60°C could be accomplished under ideal conditions in less than 16 hours with Taxol yields equal to or better than drying at the lower temperatures tested. A drying temperature of 30°C may bring about enzymatic activity which destroys Taxol. This effect, along with the extra time required, would suggest that drying should be done at 60°C or higher.

Figure 6. Taxol yields from Taxus x media 'Hicksii' clippings
needles, and stems as a function of drying temperature.

Although preliminary results indicated that needles should not be separated from stems prior to drying because Taxol would be lost (Croom, 1991), the results in Table 2 and Figure 6 showed increasing yields from separated needles as temperatures increased from 30° to 60°C. These results suggested that drying temperatures greater than 60°C should be evaluated for both clippings and needles.

Conclusions

Thin-layer drying studies of clippings, needles, and stems were conducted in a laboratory dryer. The thin-layer drying equation was found to be applicable for the prediction of drying rates for the Taxus materials tested at 30°, 40°, 50°, and 60°C drying temperatures. Parameters for the drying equations were successfully determined. The results showed that the drying constant at 60°C was 3.2 times greater than k at 50°C and 15.5 times greater than k at 40°C. Also, stems were found to dry at a faster rate than needles, and needles dried at a faster rate than whole clippings.

Comparison of Taxol yields for four drying temperatures from 30° to 60°C suggests that 60°C is the most ideal temperature for drying clippings and separated needles and stems. Drying at 30°C resulted in low yields. Future testing should be directed at drying temperatures greater than 60°C. Perhaps Taxus biomass should be dried a few hours at a high temperature to stop enzymatic activity followed by a step down to a lower temperature to finish the process.

Acknowledgments

The authors gratefully acknowledge Jill Bero, student, Department of Food, Agricultural, and Biological Engineering, Ohio Agricultural Research and Development Center, Wooster, Ohio; Kenneth D. Cochran, Curator, Secrest Arboretum, Ohio Agricultural Research and Development Center, Wooster, Ohio, and the Ohio State University Agricultural Technical Institute; and Dr. Edward M. Croom Jr., Associate Professor, Research Institute of Pharmaceutical Sciences, The University of Mississippi, for assisting with the research.

References

Barre, H. J., G. R. Baughman, and M. Y. Hamdy. 1971. Application of the logarithmic model to cross-flow deep-bed grain drying. Transactions of the ASAE. 14(6): 1061-1064.

Brooker, D. B., F. W. Bakker-Arkema, and C. W. Hall. 1974. Drying Cereal Grains. Westport, Conn.: AVI Publishing Co., Inc.

Croom, E. M. Jr. 1991. Private communication.

Hall, C. W. 1980. Drying and Storage of Agricultural Crops. Westport, Conn.: AVI Publishing Co., Inc.

Henderson, J. M. and S. M. Henderson. 1968. A computational procedure for deep-bed drying analysis. J. of Agric. Engr. Res. 6(3): 169-174.

Henderson, S. M. and S. Pabis. 1961. Grain drying theory II: Temperature effects on drying coefficients. J. of Agric. Engr. Res. 6(3): 169-174.

Henderson, S. M. and R. L. Perry. 1966. Agricultural Process Engineering. 2nd Ed. 294-327. University of California, Davis.

Hukill, W. V. 1947. Basic principles in drying corn and grain sorghum. Agricultural Engineering. 28(8):335-338.

Keener, H. M. 1991. Solar grain drying. In: Solar Energy in Agriculture. B. F. Parker, Ed. Energy in World Agriculture. Vol. 4. B. A. Stout, Ed. 295-314. Amsterdam: Elsevier Science Publishers B.V.

Parker, B. F. 1991. Design of solar-heated forage dryers. In: Solar Energy in Agriculture. B. F. Parker, Ed. Energy in World Agriculture. Vol. 4. B.A. Stout, Ed. 315-334. Amsterdam: Elsevier Science Publishers B.V.

Sabbah, M. A., H. M. Keener, and G. E. Meyer. 1979. Simulation of solar drying of shelled corn using the logarithmic model. Transactions of the ASAE. 22(3):637-643.

Thompson, T. L., R. M. Peart, and G. H. Foster. 1968. Mathematical simulation of corn drying: a new model. Transactions of the ASAE. 11(4):586-588.

Witherup, K. M., S. A. Look, M. W. Stasko, T. J. Ghiorzi, and G. M. Muschik. 1990. Taxus spp. needles contain amounts of Taxol comparable to the bark of Taxus brevifolia: analysis and isolation. J. Nat. Prod. 53(5): 1249-1255.

Nomenclature