UTILIZATION OF BARK WASTE
OREGON  STATE UNIVERSITY
PREPARED FOR
ENVIRONMENTAL PROTECTION AGENCY
JULY  1973
                                          PB-221 876
                               Distributed By:
                               National Technical Information Service
                               U." S.  DEPARTMENT OF  COMMERCE



-------
 BIBLIOGRAPHIC DATA
 SHEET
EPA-6 70/2-73-005
I. Title and SuBtitl?

        UTILIZATION OF  BARK  WASTE
7. Author(s)
   R.  A.  Currier,  M.  L,  Laver
V. Performing Organization Name and Address
  Department  of  Forest  Products
  School of  Forestry
  Oregon State  University
  Corvallis,  Oregon     	
            .                  .
            _   PB-221 876       _
           57 lU-fMMI l>:m-      "
             19 7 3-i s s ui ng date
                                                                6.
           8. Performing Organization Rept.
             No.
           10, Project/Task/Work Unit No,
           11. fnntrart /Grant No.

             R-EP  00276-04
12. Sponsoring Organization Name and Address
  U.S.  Environmental  Protection Agency
  National  Environmental Res <•.••.: i ch  Center
  Office of  Research  &  Development
  Cincinnati, Ohio  45268
           13. Type of Report & Period
              Covered
              Final
           14.
15. Supplementary Notes
16. Abstracts
      The prohleo of bark waste that le generated by the forest products industry in s-
 the  United States haft be coma Increasingly important-*  The major overall goal of the
 work covered in this report was to utilise physical end cheaical sciences in coor-
 dinated studies to promote eeononic uses of bark in order to relieve pollution
 created by present methods of disposal.  Physical utilization research included:
 (1)  investigating the preparation of bark pellets from bark in s series of exper-
 iments  designed to control several variables, such as particle size and configura-
 tion, moisture content and addition of chemical agents; (2) determining the compo-
 nents responsible for "self-bonding" of bark; and (3) investigating potential pro-
 ducts from or applications of "as-is" end modified bark wastes obtained from pro-
 duction sources.
      Chemical utilization research included: (.1) preparing, for chemical studies,
 natural bark, bark that had been ammoniated to contain 4 percent nitrogen,and bark
    . that had bean molded into pellets and then broken down into small particles;
 and  (2) Investigating tha chemical composition of each type of bark prepared.
17, Key Words mad Document Analysis. 17*. Descriptors

Pollution,  Waste  disposal,  *Wastes, *Bark>  Wood wastes,  Wood  products,
*Pelleting,  Pellets,  Chemical properties, Physical properties, Economics,
Moisture content,  Softwoods,  *DougIas  fir wood, Electric power, Agri-
cultural engineering,  Bonding, Carbohydrates
17b. Identifiers/Open-Ended Terms

*Bark waste>  *Ammoniscion,  n-Haxane-soluble  wax,  Benzene-soluble wax,
Solid waste  disposal,  Resource recovery
17c. COSAT! Field/Croup 13-B, 1 1 ~ L
18. Availability Statement

    Release to  public
I }?. Sorority Class (This
I   Hcf-.orc)
     UNjCLASSiFn-f?
                                                     21. No. o{ Pa«es
FORM NTlS-33 (REV. 3-72)
                                                        S«-.-. urity Chi.-:;, 'This    |22, Price

                                                        _  UNCLASSIFIED
                    7
                                                                          USCOMW-DC

-------
                       REVIEW NOTICE






     The Solid Waste Research Laboratory of the National




Environmental Research Center, Cincinnati, U.S. Environmental




Protection Agency, has reviewed this report and approved its




publication.  Approval does not signify that the contents




necessarily reflect the views and policies of this laboratory




or of the U.S. Environmental Protection Agency, nor does




mention of trade names or commercial products constitute




endorsement or recommendation for use.




     The text of this report is reproduced by the National




Environmental Research Center, Cincinnati, in the form  re-




ceived from the Grantee; new preliminary pages have been




supplied.
                             -ii-

-------
   .  Man and bis - envi conmen t must be p-^c.-rted  from the
adverse affects  Q.*  pesticides, radi'.tlon,  noise and oche
forms of . pollutidu ,  and the .unwise manc.~enent  of solid
waste.  Efforts  to  protect tne envi ror:m2n t  require a
focus th*t  recognizes  the inter-play between  the com-
ponente'ftof  pur 'physical envtron
-------
                                                      :                  Pag«
 SUMMAIY 07 PROOtWSTOWARD ORIGINAL* dbatfr .  .	   1

      Physical, UtiiizntioA Research  .  T'i ".  . •	   i
      Chndcal Utilization Resjlarcft - .  .v '+>	   3
      Publicmtiooe . ... .;.'..  ..... V.:	   5
      Plans for; Further Publication  .  . V'.v	   6
      .-  •< •   '"'.;  '    • :'  V>- '>'" ,  .   •
         .-   ' -  -  -     ' •"  •  /A
'PHYSICAL UtlLIZATION RESKAI^R	„'	   7

      Bark  Pelleting ..' .  . ';  .  .  ..  .  .  . . .  .    	   7
      Bark  Molding ....,../.	11
      Other Bark Products or  Osea-	14
                      	15
 CHEMICAL UTILIZAf ION RESEARCH *......	16
   I           • ,            ' j* -   »'
      Bark Carbohydrates f. .-.* .  .^,  .  ,	16
                       ,*••.••           •       '
           Historical Bevl*^  ......  ^.  .	16
  »';    -'•-  Bxperiaen^al,.. ^;\ •.  .  .  ..  . ,:V  .. ;	20
        '  Results and 'Di^cuaalim   ...;..,.....;..	78
      Douglas-Fir  Wax' .'.>., ^ ./.-  ;  .  .  .  ............. 118
                       '                       •
           Historical l^jdr.-.s-y;.  :.,'... ,'t  .: .......... us
           Ejc^ftimeotalY-a-^M^n^Soluble Waoc*  ........... 121
           Ext>earlaei^i(l.:;'^iBn|MHte^Soluble Wax  ............ 128
           Rasultv aa/j- DlaCtiiMibn  \  .  .  T-l ........  ...... 135
                   of-Bark,; ..',f  :  ...... v. :	145

           Historical Re^leVf;'.  .....  ^-'	145

           Results >an4 Dis'cusalon.\,'  .<•...'..	147
 BIBLIOGRAPHY   ....  ....  ,.%>1:.^.^.;. .,....'.'	150


 TABLES-. . ;....-.  .  >*• ..^. -rf,. «.. t=r.- ,«....*y •-.•••••••...•. . . '	157

           . .•'•;•/.  .  .  .-  , .  .  .«*.,*.....  ......f	173
                                  -v-

-------
                               -1-
                SU11MARY OF PROGRESS TOWARD ORIGINAL COALS




     A recent estimate of the amount of bark waste penerated  by  the  forest




products industry in the entire United States totals some 14  million tons,




oven-dry basis.  Other estimates show six million tons ar« available within




the west coast states of California, Oregon and Washington.  Approximately




half of the bark in these states now is used in some manner,  leaving three




million tons for which disposal problems remain.  The major overall  goal




of the work covered in this report was to utilize physical and chemical




sciences in coordinated studies to promote economic uses of bark in  order




to relieve pollution created by present methods of disposal.   We believe




substantial progress has been made toward finding economic uses for  bark.




Discussion of progress toward specific goals as originally outlined  follows.









Physical Utilization Research




     Goal 1.  To investigate the preparation of bark pellets from bark in




a seTies of experiments designed to control several variables, such as




particle size and configuration, moisture content, addition of chemical




agents, etc.




     Progress in fulfilling Goal 1 has been very good.  We have found that




all species of softwood barks investigated will form acceptable pellets.




This includes some eastern and northern species of bark,  in addition to




those from the west coast.  We have found that Douglas fir bark, in particular,




may be added to other types of forest industry residues  (such as sawdust) in




order to make a mixture which will result in formation of pellets.  One




significant finding was that pelletizing of bark greatly  increases its bulk




density and results in a product capable of being easily  conveyed, stored




and spread, compared to bark in natural form.

-------
                                    -2-
     There are at least three areas where basic information resulting from




our studies of bark palletizing may be applied directly to future commercial




Venturas utilizing waste bark.  These are as follows:




     1.  Palletizing bark residues to form a fuel for production of electrical




         poorer in a centralized plant.




     2.  Pelletizing of bark mixed with fertilizers or other chemicals for




         agricultural applications.




     3.  Pelletizing bark to increase its bulk density and handling




         characteristics for easier transport of this renewable raw material.




     Goal 2.  One. of the original goals was determination of the components




responsible for "self-bonding" of bark.  By a process of selective chemical




extraction of Douglas fir bark followed by pelletlzing (molding), it was




hoped to pinpoint the compounds essential to activation of bonding (cohesion).




With departure of the original Principal Investigator, this portion of the




research could not b® accomplished since we did not have an extractives




chemist on the grant staff.  He did learn that Douglas fir bark extracted




by one cheaical solvent sysfcsta did retain its ability to form pellets and




other saoldod  itesss.  The Forest Research Laboratory now has a separate




project, F858, "Cohesions in Consolidated Bark Products" with the objective




of determining the mschanism responsible for "self-bonding" of bark.  We




have cooperated by palletizing bark materials.




     Goal 3.  To Investigate potential products from or applications of "as-ls"




and modified  bark wastes obtained from production sources.




     We consider this phase of our physical utilization research efforts to




be the most rewarding since it has the potential of culminating in the




establishment of more than one commercial enterprise resulting directly from




efforts expended in pursuing Goal 3.  At this time, there are two products

-------
                                    -3-
on the market in the area of pelletized and extruded bark;  both obtained




from us samples and technical information generated during the course of




our studies.  In addition, there is at least one other excellent possibility




for the eventual establishment of a large-sized plant centered upon the




chemical-physical utilization of bark wastes.









Chemical Utilization Research




     Goal 1.  To prepare, for chemical studies, natural bark, bark which




has been ammoniated so as to contain about 4 percent nitrogen, and bark




which has been molded into pellets and then broken down into small particles.




     Goal 1 was completed in the 01 year of the research.  The samples were




then used for further chemical utilization research.




     Goal 2.  To investigate the chemical composition of each type of bark




prepared as outlined in Goal 1.




     Progress in fulfilling Goal 2 has been very good.  The carbohydrate




content of natural inner bark has been found to be about 50-60%.  The




carbohydrates have been fractionated into a soluble fraction and an insoluble




"holocellulose" fraction.  The ratio of sugars in the hydrolyzate from the




soluble fraction has been determined as follows:  glucose, 59.1; arabinose,




11.9; galactose, 3.9; mannose, 3.7; xylose, 1.0; rhamnose, 1.0.  Rhamnose




has not been previously reported in Douglas fir bark.  The results show




that the soluble fraction is composed mo.stly of a glucan-like polymer.




     The insoluble "holocellulose" carbohydrate fraction is very easy to




isolate and comprises some 35-40% of the inner bark.  This fraction could




become of commercial interest as a source of carbohydrates.  Some suggestions




for uses have been as an animal feed, and in the food industry.  Hydrolysis




showed the "holocellulose" to be composed of the following ratio of sugars:

-------
                                    -4-
glucose, 26.5; mannose, 4.1; xylose, 2.7; galactose, 1.0;  arabinose, 1.1.




The "holocellulose" has been further fractionated into a xylan, a glucomannan,




a mannan,, and a glucan.




     Douglas fir bark contains an n-hexane-soluble fraction and a benzene-




soluble fraction.  These are called "waxes."  The waxes have been studied




by column chromatography, thin-layer chromatography, gas-liquid chromatography




and gas-liquid chromatography combined with rapid-scan mass spectrometry.




We have found sterols  (6-sitosterol and campesterol) in the n-hexane wax as




well as evidence for terpenes.  These classes of compounds are in addition




to those reported in the early literature (see text of following report).




The benzene wax contains monocarboxylic fatty acids, dicarboxylic acids and




hydroxy fatty acids in addition to phenolic compounds.




     Whole bark was treated with gaseous ammonia to a nitrogen content of




A.08%.  Extraction of  the bark with organic solvents and hot water removed




1.10% of the nitrogen  but did not leach out 2.98% of the nitrogen.  This




suggests that the ammoniation of bark may provide a material for soil




amendment which possesses a quick release nitrogen supply followed by a slow-




release nitrogen supply as  the bark decomposes.




     Bark which had been pelletized and  then broken down into  small particles




was extracted with the organic solvents, ri-hexane, benzene, ethyl ether,




ethyl alcohol and hot  water.  These extracts were investigated by ultraviolet




spectroscopy.  No significant differences were noted between these extracts




and similar extracts prepared from non-pelletized bark.  Therefore, pelletizing




does not appear to grossly  alter the chemical make-up of the bark.  Thus,




most of the efforts of Goal 2 was directed to studies of the raw bark as




described above.

-------
                                    -5-
Publications




     1.  Currier, R. A.  1971.  What is bark?  Physical considerations.




Proceedings of .conference "Converting Bark into Opportunities," Oregon




State University, Corvallis, Oregon, March 8-9, 1971.




     2.  Currier, R. A. and W. F. Lehmann.  1971.  Bark as an Ingredient




in molded items, particleboards, adhesives and other products.  Proceedings




of conference "Converting Bark into Opportunities," Oregon State University,




Corvallis, Oregon, March 8-9, 1971.




     3.  Currier, R. A.  1972.  An assessment of current bark utilization




opportunities.  27th Proceedings Northwest Wood Products Clinic, Spokane,




Washington.




     4.  Laver, M. L.  1971.  What is bark?  Chemical considerations.




Proceedings of conference "Converting Bark into Opportunities," Oregon




State University, Corvallis, Oregon, March 8-9, 1971.




     5.  Laver, M. L.  1971.  Chemicals from bark.  Proceedings of conference




"Converting Bark into Opportunities," Oregon State University, Corvallis,




Oregon, March 8-9, 1971,




     6.  Fang, H. H-L.  1971.  Douglas-fir bark; n-hexane soluble fraction.




Master's thesis, Oregon State University, Corvallis, Oregon.




     7.  Laver, M. L., H. H-L. Fang and H. Aft.  1971.  The n-hexane-soluble




components of Pseudotsuga menziesii bark.  Phytochemistry 10(12) :3292.




     8.  Lai, Y. C. L.  1972.  Douglas-fir bark; carbohydrates solubilized




by the acidified sodium chlorite delignification reaction.  Master's thesis,




Oregon State University, Corvallis, Oregon.




     9.  Laver, M. L., J. V. Zerrudo and H. Aft.  1972.  Treatment of Douglas-




fir bark with gaseous ammonia.  Forest Products Journal 22:82.

-------
                                    -6-
     10.  Loveland, Patricia M.  and M.  L.  Laver.   1972.   Monocarboxylic




and dicarbosylic acids from Pseudotsuga menziesii bark.   Phytochemistry




11(1):430.




     11.  Loveland0 Patricia M.  and M.  L.  Laver.   1972.   w-hydroxy fatty




acids and fatty alcohols from Pseudotsuga  menziesii bark.  Phytochemistry




11 (10):3080.




     12.  Zerrudo, J. V.  1973.   Douglas-fir bark; water-soluble carbohydrates




and alkaline degradation of a xylan.  Doctoral thesis.  Oregon State University,




CorvalliSs Oregon.









Plans for Further Publications




     Results of research covered by the bark pelleting studies will be




presented in a paper to be published in the Forest Products Journal; title




of this report will be "Pelletizing bark residues."




     The research area covering bark molding appears to be appropriate for




preparation of a "Technical Note" to be published in the Forest Products




Journal.  Plans are not finalized as yet,  however.




     A paper has been accepted for presentation at the IV Canadian VJood




Chemistry Symposium, Quebec City, Quebec,  next July.  It will be published




in Carbohydrate Research and cover the alkaline degradation of a xylan.




     Manuscripts on "Water-soluble Carbohydrates of Bark," "Holocellulose




Characteristics and Properties," "The Tannins in Douglas fir Bark" and "The




Terpenes in the n_-Hexane Soluble Components of Douglas-fir Bark" are in the




early stages of preparation.

-------
                                    -7-
                      PHYSICAL UTILIZATION RESEARCH




I.  Bark Pelleting




     Using a laboratory model pelleting mill (California Pellet Mill Company,




Model CL, Type 3), approximately 200 pelleting trials have been made.  Raw




material involved 16 species of bark plus numerous mixtures of species.




In addition, runs were made on mixtures of bark plus woody residues.  See




Table 1 for a list of bark species and the mixtures utilized in pelleting




trials.




     Variables studied other than raw material mix included particle size,




moisture content, fraction of bark, addition of fertilizers and seeds,




pellet diameter and degree of pellet densification.  Most trials consisted




of 10-15 pound lots, but several weighed over 100 pounds.  Various types




of pellets produced are depicted in Figure 1; also shown is the hammermilled




Douglas fir bark serving as standard pelletlzing raw material.  Pelleting




concentrated on Douglas fir .and western hemlock barks, since both are




available in large quantities as a residue material.




     Suitable pellets were formed from all species of bark studied,




although problems were encountered with a few species.  Where mixtures of




barks or bark plus woody residues were pelleted, addition of 25% or less




Douglas fir bark to the mix often spelled the difference between success




and failure.  Douglas fir bark always pelletized readily under a variety




of conditions.




     Pelleting trials were made of selected barks from the northern,




eastern and southern portions of the United States in order to learn if




the techniques we developed could be applied to waste barks from other




regions of the nation having disposal and pollution problems.  Results of




our limited tests indicated bark from the species selected could be made

-------
                                    -8-
into well-formed pellets quite readily, although the southern oak gave




trouble since it was about half wood.




     A potentially significant application of data collected on bark




pelleting is in transportation of bark wastes.  Compared to raw bark in




chunk or comminuted form,, the weight of pelletized bark per unit of




volume occupied is 2.0 to 3.8 times as heavy.  Compression factors and




bulk densities for a number of bark species and pellet diameters are




listed in Table 2.  In addition, pelletized bark is easily conveyed, stored




and spread compared to bark in natural form.




     One of the most promising potential commercial applications for




pelletized bark is to utilize bark as a carrier for fertilizers or other




chemicals used in agriculture.  During the last year of this grant,




practically all the pelletizing trials were run on bark-fertilizer mixes.




Small-diameter (1/8" or 3/16") pellets were made using thin dies, since




this combination gave good rates of production and resulted in a product




which could be spread by present standard methods of fertilizer application.




Several drums of bark-fertilizer pellets were made for growing trials




during the 1972 season.




     Cooperation has continued with  Schroeder Sales Company, Division of




Pacific Kenyon Corporation, Long Beach, California, who have provided




fertilizer and conducted field trials.  Results of tests on a golf course




in  the Los Angeles area were encouraging, as were trials on home lawns in




the Corvallis area and on the Forest Research Laboratory lawn.  Pellets




for lawn use were formulated to yield either 7.5-0-0 or 9-3-6 fertilizer




values.  The source of nitrop.cn was  urea in prill form.  This is a fast-




acting form of nitrogen, but when mixed with bark and pelletized, the




rate of nitrogen release appears to  be slowed significantly.  Possible

-------
                                    -9-
economic benefits could result since slow release types of fertilizers




are priced considerably higher compared to urea.   Another potential source




of slow release nitrogen in bark is the ammoniation process discussed in




this report.




     Cooperation also has continued with Bohemia Lumber Company (now




Bohemia, Inc.), Eugene, Oregon.  In fact, when this grant concluded on




June 30, 1972, Bohemia, Inc. made a grant to us in order to continue




research on a bark-fertilizer pellet, utilizing Douglas fir bark previously




extracted chemically in their pilot plant.  Work planned will involve




production of pellets and conducting field trials on the product.  Chances




of constructing a commercial plant to manufacture pelletized bark-fertilizer




appear very promising.  A bark-plus-fertilizer pellet offers a means of




complete utilization of the residue from chemically extracted bark and




definitely adds to the feasibility of such an operation.




     Several hundred pounds of bark-fertilizer pellets have been given to




George D. Ward & Associates, Portland, Oregon, pollution control consultants,




They have a contract with the Naval Facilities Engineering Command, Seattle,




Washington, for reclaiming and stabilizing blow sand areas in eastern




Oregon.  The bark-fertilizer pellets may have the attributes of retaining




moisture and slowly releasing nutrients to trees, shrubs, or grasses




planted on the sand dunes.




     Samples of bark pellets have been distributed widely.  Included have




been forest products manufacturing concerns, universities, governmental




laboratories, chemical and plastics manufacturers and private individuals.




A partial list of recipients of bark pellet samples may be found in Table 3.




     Some limited commercial application of pelleted bark has resulted




aince our research work commenced.  It has been learned that the largest

-------
                                   -10-
forest products company in Canada,,  MacMillan Bloedel Ltd.  has test marketed a



bark-ccmtaining-fertilizer pallet product.   The original idea appears to



have come from our exploratory pelleting studies, and we have maintained



contact with their research personnel.



     Another company in Oregon has developed a fuel product from bark.



TriWest Products of Springfield, Oregon, makes the pelletized fuel sticks.



A principal of TriWest Products has consulted with us several times



regarding bark pelleting, and has informed us the product idea was initiated



by our research.



     Trials on use of 1/2-inch diameter pelletized Douglas fir and western



hemlock bark for fuel have been made by Forest Research Laboratory personnel.



To date, pellets have been tried both as a stove fuel and in fireplaces.



They are somewhat difficult to ignite, but once started, give off excellent



heat.  There has been recent discussion regarding the possibilities of



using pallatized bark as the fuel source for a centrally located electric



power generating plant.  Instead of disposing of bark wastes through the



present practices of incineration at  the sawmill sites, the residues would



be pelletized to improve their bulk density and shipping characteristics.



Transportation to the power generator could be by means similar to the



way coal is handled today.  Since bark is a renewable raw material, this



idea may gain more proponents as time passes.



     Limited trials have been made of palletizing bark plus seeds such as
                                         i


clover or grass seed.  Techniques have been developed whereby the mixture



can be pelletized with some degree of germination resulting.  No detailed



study has been made, but this is an area we hope to investigate soon.



     Another use of bark in pellet form is being investigated by the U.S.



Forest Service, Forest Products Laboratory, Madison, Wisconsin.  Some of

-------
                                    -11-
our 1/4-inch Douglas fir bark pellets have been used as the large aggregate




in cement-bark building blocks.  Tests are continuing on the process and




resulting blocks.









II.  Bark Molding




     Much of the research on molding of bark has consisted of demonstration-




type projects in cooperation with various segments of the forest products,




adhesives, and plastics industry.  This method of attacking the problem




was utilized in order to probe the potentials of various end markets for




comminuted Douglas fir and western hemlock barks and to increase awareness




that such material was available from bark residues.  A brief description




of the various investigations follows.




A.  Planter block from bark




     1.  Using compression molding, trays, each consisting of six planter




blocks, were prepared from western hemlock bark.  In the preparation of




these blocks, the following variables were studied:  moisture content of




the bark, binder use, and addition of other materials such as lime, clay,




chemical fertilizer  (Vigoro) and fungicides.  Greenhouse studies, followed




by field planting, indicated that the bark planter blocks could grow




tomatoes and pansies from seed.




     2.  Douglas fir and red alder barks were extruded into blocks and




limited growth trials were made utilizing tomatoes.  Blocks were formed




with and without added urea-formaldehyde binder; chemical fertilizer




(Vigoro) was added to all barks prior to forming.




B.  Bark added to plastics




     1.  One investigation involved a three-way cooperative study with




industry.  We prepared dry corminuted bark, a lumber company (Dant and

-------
                                    -12-
Russell, Inc.5 Portland, Oregon) provided the raw bark and financed




experimental time on molding machines at a plastics producer (Grant and




Roth Plastics, Inc., Hillsboro, Oregon).  Several types of extruded,




sheet formed and injection molded products have been produced, with bark




extension of the plastic in the order to 40-60 percent.  Types of plastics




extended by bark include polyethylene, cellulose butyrateB cellulose




acetate, polyurethane, polystyrene and poly-vinyl chloride.  Photographs




showing some experimental products may be found in Figure 2.




     2.  The plastics company involved in study (1) above (Grant and Roth




Plastics) sent some of the bark extender to Eastman Chemicals, Inc.,




Kingsport, Tennessees for their evaluation.  A laboratory report indicated




promise for a cellulose acetate + plasticizers + bark mix.  Cost per pound




of this mix was estimated to be lower than polyvinyl chloride„




     3.  Large samples  (60 Ib) of -20 mesh Douglas fir and western hemlock




bark were shipped to Borden,, Inc., Thermoplastic Division, Leominster,




Massachusetts, where they were evaluated as fillers for extruded rigid




PVC pipe.  The resulting laboratory report by Borden, Inc. indicated the




bark contained an excessive amount of volatiles, leading  to porous and




lumpy extrusions.  On the positive side, addition of bark lowered specific




gravity of the product, resulting in a cost-volume advantage.  A follow-up




study now is underway whereby Borden, Inc. has been provided another




sample of Douglas fir bark which has had some potential volatiles removed




by chemical extraction.  No results have been received to date.




     A.  The Hysol Divison of the Dexter Corp., Industry, California, has




formulated a knofchole or other void filling compound from -20 mesh




Douglas fir bark plus epoxy plastic materials.  The compound has been




used commercially to fill voids in rough-sawn textured plywood panels.

-------
                                    -13-
We have provided over 100 Ib of comminuted bark for these experimental




formulations, which have resulted in commercial production of products




numbered AE1502 and AE1504.




     5.  Preliminary runs have been made extruding Venetian blind slats




from polystyrene plus -20 mesh Douglas fir bark in both plain and




extracted form.  This experimentation was sponsored by Omega Industries




of Salem,..Oregon.  Five different color slats have been run, with




promising results.  The extracted bark appeared to be the more suitable




extender.




     6.  During the last few weeks of the grant period, an excellent




contact was made with B. F. Goodrich Chemical Company, Avon Lake, Ohio.




This company is one of the largest suppliers of plastics molding compounds.




Large samples of both unextracted and extracted Douglas fir bark, and




bast fibers have been shipped for trial runs on soil pipe, conduit and




etc.  No results have been received yet.




     7,  A few furniture drawer guides were made from 50% -20 mesh Douglas




fir bark and 50% general purpose polystyrene at Modcom, Inc., Canby,




Oregon.  The management there is willing  to conduct more evaluation tests




when some free plastic machine time is available.




C.  Other molded products from bark




     1.  Molded bark products not involving plastics have resulted in




cooperative efforts with two different commercial  companies  interested




in production of extruded "fuel logs" from bark wastes.  A sample lot of




Douglas fir bark was prepared for Glomera-American, Inc., Reedsport, Oregon,




who shipped it to Switzerland where the log forming machines are made.




Trail runs indicated a log containing mostly bark  can be formed by this




process.  Another company, Oregon Timber  Products  Company, Albany, Oregon,

-------
                                    -14-
has purchased log extrusion machines from Japan; we cooperated with them




in developing a log containing substantial quantities of Douglas fir and




western hemlock barks.  Unfortunately, this company has suffered financial




problems and is not operating at the present time.  Photographs depicting




these various logs may be found in Figure 2.









Ill.  Other Bark Products or Uses




     Samples of bark or bark fractions have been prepared for a number of




other companies and researchers.  Some examples follow.




     1.  Bark particles have been prepared for experimental work on a




trickling filter system for disposal of animal wastes.  This is being




done by Dr. Myron Cropsey, Agricultural Engineering Department, Oregon




State University.  We have provided all of the 3/4- to 3-inch Douglas fir




bark chunks used in his research.  Dr. Cropsey presented a paper covering




his results to date at a meeting of the American Society of Agricultural




Engineers in Portland, Oregon, October 8, 1971.  His tests will continue




during  the 1972-1973 academic year.




     2.  Bohemia,, Inc., Eugene, Oregon, is interested  in developing a




process for extraction of wax from Douglas fir bark, and has built a




pilot plant.  After wax extraction, over  90% of the bark still remains;




we have cooperated with Bohemia, Inc.  in  determining potential uses for




this spent bark.  One effort has been  tb  turn the  residue into an extender




for phenolic plywood glues.  Extracted bark was shipped to us and we




ground  and screened over 400 Ibs to prepare -100 mesh material.  This was




successfully evaluated as an extender  for plywood  glues, undergoing




standard test procedures prescribed by the American Plywood Association.




Several of the chemical companies manufacturing plywood glues are interested

-------
                                    -15-
in the bark extender.  Cooperation is continuing with Bohemia,  Inc.  in




separating the physical components (cork and fiber) of  extracted Douglas




fir bark prior to grinding the remaining portion as an extender.  Possibilities




of commercialization of the process appear encouraging.




     3.  Samples of the -100 mesh extracted bark extender mentioned  in




(2) above, as well as -100 mesh unextracted whole bark have been sent to




Colorado State University, where a graduate student is evaluating their




possible use as an extender for binders currently used in manufacture of




particleboard.  Preliminary results look promising; additional  bark  has




been requested.  No published report is available as yet.









IV.  Miscellaneous




     During the period January 1, to June 30, 1972, the physical utilization




research done under this grant was presented orally at four meetings as




follows.




     1.  Western Builders Short Course, Portland, Oregon.  February  11, 1972.




     2.  Twenty-seventh Annual Northwest Wood Products Clinic meeting




jointly with the Inland Empire Section of the Forest Products Research




Society.  Spokane, Washington.  April 17-19, 1972.




     3.  Pacific Northwest Section of the Forest Products Research Society,




Vancouver, B. C., Canada.  May 9, 1972.




     4.  Twenty-sixth Annual Meeting, Forest Products Research  Society,




Dallas, Texas.  June 19-22, 1972.




     In addition, a report for local radio broadcast also was made;  this




covered both physical and chemical research sponsored by this grant.

-------
                                   -16-
                      CHEMICAL UTILIZATION RESEARCH









                              INTRODUCTION




     There are a number of general review articles on bark utilization,




but thrse pertinent ones a,re by Harkin and Rowe (1), the Bark Committee




of the Forest Products Research Society (2) and J. Alfred Hall (3).  Hall




cosasents that one cannot w;rite a meaningful "Chemistry of Douglas-fir




Bark" in the present state of our knowledge.  This statement demonstrates




the great gaps in the knowledge of the chemistry of bark.  The research




herein reported is an effort designed to provide a better understanding




of the chemistry of Douglas-fir bark with the goal of a more complete




and better utilization of 'this abundant natural raw material.




     The work has been organized into three areas:  the carbohydrates;




the waxes; aad the ammoniation of whole bark.  For clarity of reporting,




each of these subjects is presented separately.  Each section Includes




an historical review,, an experimental, and a results and discussion.









                           BARK CARBOHYDRATES




I.  Historical Review




     Authors have referred to Douglas-fir as Pseudotsuga taxifolia  (Poir.)




Britt. and other names.  However, the presently preferred botanical name




is Pseudoteuga menziesii (Mirb.) Franco,  The names all refer to the same




genus and species.  Mention is made of this to avoid confusion about the




exact species investigated.




     The present work is concerned with Douglas-fir inner bark.  This is a




specific anatomical part of the bark and a brief description of bark




anatotay is included for purposes of definition.

-------
                                   -17-
     For a detailed anatomical description of Douglas-fir bark, reference




is made to Grilles (4), Grilles and Smith (5), Chanp. (6) , and Ross and




Krahmer (7).  Briefly, however, bark can be considered to consist of




inner bark and outer bark (Figure 4).  The inner bark (phloem cells) is




the portion from the vascular cambium to the innermost cork layer.  The




outer bark (rhytidome) is'.everything to the outside of the innermost




cork cambium (Figure 4).




     The inner bark comes from the vascular cambium, that layer of living




cells between the wood and bark which divide to form wood to the inside




and bark to the outside.  The inner bark is composed mainly of sieve




cells, axial and ray parenchyma, and sclereids (Figure 4).  Much of the




inner bark is living in the living, tree because many of the parenchyma




and sieve cells remain alive as long as they are components of the inner




bark.




     Douglas-fir sclereids are short, sharply pointed, spindle-shaped




fibers of a red brown color (Figure 4).  They are often referred to as




bast fibers.  They are lignifled cells and develop from axial parenchyma




cells some distance from the vascular cambium.  In becoming sclereids,




axial parenchyma cells approximately 0.1 to 0.5 mm in length elongate to




1 to 2 mm by apical intrusive growth, and form thick walls.  The sclereids




are commonly straight and somewhat cigar-shaped.  Kiefer and Kurth (8)




and Ross and Krahmer (7) describe and illustrate the general appearance




and position of the sclereids in Douglas-fir bark.




     The outer bark of Douglas-fir consists of layers of cork in which




growth increments are usually visible, as shown by Ross and Krahmer (7).




Interspersed among the corky layers are areas of phloem tissue that contain




the sclereids and other ce^l types found in the inner bark (Figure 4).

-------
                                   -18-
The cork layers form frorv the cork cambia which are livinp cells that v/ere




once living parenchyma cells of the inner hark.  New cork cambia form in




the inner bark and cut away part of the inner bark, which now becomes part




.of the outer bark.  The cork cambium produces cork cells to the outside




and a few storage cells to the inside.  Cork cells have thin cellulose




walls which are coated with suberin.  Suberin is essentially an ester




condensation polymer of hydroxylated, saturated and unsaturated stralp.ht-




chnln fatty acids (9).  The cork layers may also contain tannin,




clihydroquercetin and starch.  All cells outside the Innermost cork cambium




arc dead because no food supply can pass through this layer of cork cells.




This then results in an outer bark composed of cork cells and dead phloem




cells, which were once inner bark.




     The studies on the anatomy of bark have shown it to be a complex




physical material.  The chemical composition of bark is equally complex.




It is knowns howevers that the major constituents of bark are the




carbohydrates j, just as they are the major constituents in v;ood  (9, 10).




However, little attention has been devoted to  the carbohydrates present




In the hark of trees.   In contrast, the carbohydrates in the wood of




trees have been extensively studied.  Polysacch.ir ides such as pectinlc




acids, galacturonogalactans, arabinopalactans, 4-0-methylp,lucuronoxylans,




arabino-4-O-methylglucuronoxylans, glucomannans , galactop.lucomannans ,




and cellulose have been isolated  from the wood of numerous species of




both gymnosperras and arborescent angiosperns and their chemical structures




elucidated  (11, 12, 13, 14, 15 p. 447)




     One reason for this has undoubtedly been  the greater economic




importance of wood as compared with bark.  Another is probably  to be




found in the occurrence in bark of non-carbohydrate constituents such as

-------
                                   -19-
suberin, tannins, phlobaphenes, and various phenolic compounds, all of



which make the isolation of polysaccharides from bark difficult.  Wood



contains none or few of these components.



     Segall and Purves (9) discussed these difficulties in a review of



the early literature on the chemistry of bark.  The so-called "extractives"



were known to interfere unless they were removed by exhaustive, successive



extractions with alcohol, water or other neutral, chemically inert liquids,



such as diethyl ether or petroleum ether.  This early literature showed



that after all of the solvent-soluble materials had been removed by



exhaustive extraction, some 70 to 90% remained as an undissolved residue.
                                                                         *


     After removing the extractive materials it was possible to acid



hydrolyze the remaining residue and investigate some of the monosaccharides



released.  It appeared that the polysaccharides in bark were based



predominantly upon glucose, and to a lesser extent on galactose, mannose,



xylose, arabinose and rhamnose.  The existence of true cellulose could



not be established at that; time.



     Kurth (10) also reviewed the early literature on the chemical



composition of barks.  He reported that barks contained hemicelluloses



and "cellulose"; the sugars associated with the hemicelluloses were



glucose, galactose, mannose, arabinose, and xylose.  The "cellulose" was



simply that fraction which remained after treating extracted bark with



four successive portions of a mixture of one part nitric acid and four



parts alcohol.  No additional evidence that the material was cellulose



was given.



     Kiefer and Kurth (8) isolated a holocellulose fraction from the bast



fibers, or sclereids, of Douglas-fir bark.  Whole bark was collected,



ground and screened.  All fractions larger than 40-mesh were reground and

-------
                                   -20-
rescreened.  The  fibers were obtained  in an almost pure  state by  stirring




tlio crude  fiber fraction  in five  times  its volume of distilled water  at




room  t etnpernturs?.   By  vlrtui1 of their  high apeclClr gravity, t:l>t-  fibers




readily  sank to the  bottojm of  the container, whereas cork  and other




impurities  remained  on tHe surface and  were skimmed off.




      The overall  percentage composition of the bast fibers was determined




and is presented  in  Table 4.   For sake  of comparison,  analyses of Douglas-




fir wood,  taken from the  literature (16), are  included in  the table.   The




data  indicate  that  the bajst fibers may  be a lignocellulose material with




a  composition  similar  to  that  of  wood.  The higher  lignin  content of  the




bast  fibers and the  lower, methoxyl content of  this  lignin, however, are




notably  dissimilar  from those  found in  wood.




      Paper  chromatograns  of an acid hyilrolyzate  of  the holocellulose




'showed the  presence  of glucose, pa lactose, mannose  and xylose.   The limited




techniques  of  paper  chrortatography at  that  time  did not resolve  mannose




and arabinose.  Their  fermentation procedures  for  quantitative analyses




indicated  an apparent  absence  of  arabinose  in  Douglas-fir  bark.   This




paper by Keifer and  Kurth (8)  and the  one by Holmes and Kurth  (17) appear




to be the  major works  published on the carbohydrates  in Douglas-fir bark.




These materials  seem to have been largely  ignored  to  date.









I1.   'Experimental




A.  Collection of Bark Samples




      On  May 22,  1969,  the outer bark was chipped from a standing Douglas-




fir of diameter  18.8 inches at breast  height at  Black Rock,  Oregon.   The




inner bark plus  cambiup were then carefully stripped  from  the tree and




 Immediately brought  to ttye  laboratory  where the  cambium layer was removed.

-------
                                   -21-
Tlie cambium-free inner bark (4832.0 g, moisture content 44.9%, hot air oven




at 110°C) was immersed in 18 liters of 95% ethanol.  Water (1207.0 ml) was




later added to adjust tlie solution to ethanol-w.it IT (4:1 v/v) wJth




calculations for the moisture content of the inner hark.




     The tree was later cut and by count of the annual rings was 130 years




old.                                                                  .




B.  Sample Preparation




     The inner bark, after soaking in the ethanol-water (4:1 v/v) for three




days, was recovered by filtration and washed well with fresh ethanol-water




(4:1 v/v).




     The combined filtrate and wash liquors were concentrated on a rotary




evaporator (Buchi, Rotavapor, Switzerland) to less than 2 liters,  Water




was added to exactly 2 liters in a volumetric flask and three 10-ml aliquots




of the adjusted solution were removed and the nonvolatile solids determined




by the Celite-vacuum drying method (18).  The values obtained were 2.057 g




per 10-ml aliquot which corresponded to a total of 411.4 g of material in




the ethanol-water (4:1 v/lv) extract or 15.4% of the original inner bark on




a dry weight basis.




     The extract was tesrjed for monosaccharides by paper chromatography




using the solvent system ethyl acetate-pyridine-vater  (8:2:1 v/v/v) .  A




trace of glucose was detected by spraying the chromatograms with o-




aminodiphenyl reagent (0.4 g o-aminodiphenyl dissolved in a solution




prepared from 100 ml of jjlacial acetic acid and 20 ml of distilled water)




and heating at 100±2° in an oven for 5 min. (19).




     The residue of innei} bark (air-dried) was ground in a Wiley Hill (A.




H. Thomas Company, Philadelphia, PA) and fractionated according to particle




size hy screening with'.screens of an increasing number of meshes per  inch

-------
                                   -22-
(The W. S. Tyler Company, Cleveland, Ohio).  All fractions were examined




under a microscope.  It was observed that those materials retained on the




screens of 20 and 35 meshes per inch were bundles of fibers.  These




fractions were reground arid rescreenecl.  All materials (1612.8 g) v.'hich




passed through 35, 60, and. 80 mesh per inch screens and were retained on




n 100 mesh per inch screen were used in prcp.irinp. tin- sample for future




investigations.  In addition 238.3 p. out of the 390.u p. which passed through




a 100 mesh per inch screert but were retained on a 150 mesh per inch screen




were included.  None of the larper material (23.6 g) retained on the 35




mesh per  inch screen was used.




C.  Benzene-Ethanol Extfraction




     A part of the recombincd bark fractions  (1500.0 ?., dry weight) was




divided into three portions (618.0 g,  618.0 g, and 264.0 g dry weight).




Each portion was extracted with a solution of benzene-ethanol  (2:1 v/v;




1400 ml of benzene and 700 ml of ethanol for  the large portions, and




598 ml of benzene and 299 ml of ethanol  for the small portion) in a




Soxhlet extractor.  Each Extraction was  continued for 37.5 hr  (a minimum




of 50  solvent exchanges).




     The  non-volatile solids in the benzene-ethanol extract were determined




by the Celit-vacuum drying method  (18);  weight, 77.9 g or 4.4% of the




original  inner bark on a dry weight basis.




     The  extract was  tested for monosaccharides by paper chromatography




using  the solvent  system ethyl acetate-pyridine-vater  (8:2:1 v/v/v).  No




sugars were detected  by  the o_-aminodinhenyl spray reapent described earlier.




0.  Hot-Water Extraction




     A' part (1458.3 p, dr^ weight) of  the air-dried residue remaining, from




the benzene-ethanol  (2:1 v/v) extraction was  divided  into four portions

-------
                                   -23-
(three of 363.8 g, and one of 366.9 p, dry weight).  F,ach portion was stirred




into 3 liters of distilled water at 55° in a 4-liter beaker.  The extraction




was continued at 50°-6Q° for 24 hr with intermittent stirring.




     The mixture was separated by filtration using a Buchner funnel.  The




filtrate was condensed on a rotary evaporator, freeze-dried and the solids




weighed; weight, 202.5 g or 11.1% of the original inner bark on a dry




weight basis.




K.  Anmonium-Oxalate Extraction




     A part  (1458.3 g, dry weight) of the residue (dried in a hot-dry room




at 32° and relative humidity 31% for 4 days) was divided into 4 fractions




each containing 312.0 g.  Each fraction was extracted with 3 liters of




0.5% aqueous ammonium oxalate in a 4-liter beaker at 70-80° for 26 hr.




     The insolubles were recovered on a Buchner funnel, washed thoroughly




with water and dried in a hot-dry room at 32° for 3 days; weight 1198.0 g




or 66.3% of  the original inner bark on a dry weight basis.




     The filtrate was concentrated on a rotary evaporator, freeze-dried,




.•mil the sollfls (3.5 g nlr-dry weight) were stored for future reference.




F.  Acidified Sod i urn Chlorite Uelign If ication; laolat ion c.' ,a Holocell'ulose




    Fraction




     An amount (1194.0 g dry weight basis) of the above residues from the




ammonium oxalate extraction was divided into six batches.  Five batches




contained 196.8 g each and one contained 210.0 g.  Each batch was stirred




into 3.0 liters of distilled water at 75-80° and the temperature was




maintained throughout the reaction.  A steady stream, of nitrogen was




bubbled through the mixture to prevent the accumulation of gases and the




mixture was  stirred mechanically.  Glacial acetic acid (20.0 ml) was added,




followed by  sodium chlorite (60.0 g).  Fresh glacial acetic acid and sodium

-------
                                    -24-
chlorite were added two more  times  at  one  hour  intervals.  During, the
                                                          1

second addition foaming occurred on top  of the  beaker so 2-5 ml of isoamyl


alcohol was added  into the  foaming  mixture to prevent additional foaming.


At the end of 4 hours, the  yellow solids were recoverd by filtfd.feion using


a Biichner funnel with Whatman No. 1 filter paper, and wasrw*Vi*h'distilled
                                                           •. ' »-..-.
                                                         ...^aa, V/ •<•.••    • • •
water.  The yellow solids were dialyzed  for one week, wfl8nca,;with distilled


water, dried with  ethanol and finally  dried in  the air for two weeks;


weight 797.6 g, dry weight  basis or 44.3%  of the original- firmer bark.  The  -;


filtrate was dialyzed for one week, concentrated on a rpjtflty evaporator    ,;


and the solids recovered by freeze-dryinp.  They were s'ttiFOtft'of1 future


reference (sec section II-H).                       •-" —••*—>-••  - t" "


     The yellow color of  the residue from  the acidified1'Wdiufiv chlorite


treatment indicated  incomplete delipriification.  Therefore, the delignification


was repeated on 783.9 g  (dry weight basis) of the yellow eoHd.8  (20).  The

                                                          •'."•'' v
residue was recovered by  filtration, dialyzed for one? vefefe'$
f_ the Hot-Water-Soluble Solids (Isolation described in section Il-D) ' . ^'^--:. 's;'V;-'-;V . 1. Elemental Analyses- for Nitrogen, Sulfur, Phosphorus; jidA^tne Halogens '•\ '^*•/•.;• •- + :••'*• i . V'' '•• . -l A small piece of freshly cut sodium was wiped tlioroufenKt ta. remove * T ^^if^f'tt *. » • "• •' V'.' ' all traces of kerosene and placed in a small glass teat tuffeX The tuhe was
-------
                                   -23-
gently heated-in a flame until the sodium melted and the vapors rose 1-2 cni




up the walls of the tube.  A snail amount of the hot-water-soluble sol ids




was added to the molten sodium and the tube was heated strongly over an open




flame.  Heating was continued for 1 to 2 minutes.  After the entire end of




the tube was red hot, the tube was plunged into an evaporating dish which




contained about 10.0 ml of distilled water.  The hot end of the tube




shattered and the resulting mixture was heated to boiling, the tnsolubles




removed by filtration, and the filtrate recovered for elemental analyses.




     An aliquot (2.0-3.0 ml) of the filtrate was added to powdered ferrous
                     •



sulfate (0.1 g) in a test tube.  The solution was heated gently with




shaking until it boiled.  Sufficient dilute sulfuric acid was added to




dissolve the iron hydroxides and to p,ive an acid solution.  A precipitate




of Prussian blue formed, which indicated the presency of nitrogen.  For




purposes of comparison, alanine was fused with sodium, and tested for




nitrogen.  A Prussian blue color formed which showed the presence of nitrogen.




     A second aliquot (2.0 ml) of the filtrate was acidified with dilute




acetic acid, and a few drops of lead acc'tate were added.  No precipitate




formed, indicating that sulfur was not present.  As a control for the




analysis of sulfur, cystine was fused with sodium and analyzed exactly as




described above.  A yellow precipitate formed which confirmed the presence




of sulfur.




     A third aliquot  (1.0 ml) of the filtrate was acidified with concentrated




nitric acid (3.0 ml) and boiled for 1 r.inute.  The solution was cooled and




an equal volume of ammonium molybdate reapent was added.  The solution was




warmed to 40-50° and allowed to stand.  No precipitate formed, indicating




that phosphorus was not present.  Olucose-1-phosphate was fused with sodium




and tested for phosphorus.  A yellov precipitate (avimonium phosphomolybdate)

-------
                                   -26-
formed which Indicated the presence of phosphorus.'ft

     A fourth aliquot (2.0 ml) of the filtrate was acidified with dilute

sulfurlc acid, boiled gently to remove any hydrogen cyanide which might

be present, and the solution was treated with a few drops of aqueous

silver nitrate.  No precipitate formed indicating that none of the halogens

were present.                                             ;

     A small sample of the hot-water-soluble solids was quantitatively

analyzed for nitrogen (Kjeldahl, 3.63%; Pascher and Pascher, 53 Bonn,

Buschstrasse 54, West Germany).                           ;

2.  Test for Tannins and Starch

     A small amount of the hot-water-soluble solids was dissolved in

distilled water.  An aliquot  (1.0 ml) of this solution was treated with

a few drops of a ferric chloride-potassium ferricyanlde solution (21, p. 227)

(1% solutions are mixed prior to use).  A blue color developed which

Indicated the presence of phenolics.                      ......

     A second aliquot (5.0 ml) of the solution was created with a few drops

of iodine indicator.  A deep blue color developed which Indicated the

presence of. starch.

3.  Strong Acid Hydrolysis

     A portion of the hot-water soluble solids (0.07 g) was treated with

72% sulfuric acid (0.9 g) and allowed to stand for 45 minutes at room

temperature.  Water  (20.1 ml) was added slowly with stirring to provide

a final concentration of 3.0% acid.  The .solution was refluxed for 5 hours,
                                                          Jr
cooled to room temperature, and neutralized to pH 5.0 with saturated   >

aqueous barium hydroxide solution.  The resulting barium  sulfate

precipitate was removed by centrifuge, washed well with water and the

washings were added to the decantate.  The combined decantate was

concentrated under vacuum on  a rotary evaporator to about 50.0 ml.

-------
                                   -27-
4.  Mild Acid Hydrolysis




     A portion of the hot-water-soluble solids (0.32 g) was dissolved in




3.0% sulfuric acid (96.0 ml) and the solution refluxed for 5 hours.  After




cooling to room temperature, the solution was neutralized to pH 5.0 with




saturated aqueous barium hydroxide solution.  The precipitate of barium




sulfate was removed by centrifuge and washed with water.  The decantate




plus the washings were concentrated to about 100.0 ml under vacuum on a




rotary evaporator (22) .




5.  Qualitative Araino Acid Analysis by Paper Chromatopraphy




     The hydrolyzates from the mild acid hydrolysis were subjected to




ascending two-dimensional paper chromatography (23, 24, p. 93) on Vlhatman




No. 1 paper, using water-saturated phenol as a developer in one direction




in an atmosphere of ammonia, and n-butanol-forraic acid-water (20:6:5 v/v/v)




as developer in the second direction.  A solution of ninhydrin (1.0 g)




dissolved in n-butanol (500.0 ml) was used as the spray reagent.




6.  Carbohydrate Analysis by Paper Chromatography




     A portion of the hot-water-soluble solids was dissolved in distilled




water, spotted on Whatman No. 1 filter paper and paper chromatographed as




described below to determine if free sugars were present.  No free sugars




were detected.




     The hydrolyzates from the strong acid hydrolysis and the mild acid




hydrolysis were applied at intervals of about 1 inch along one edge of a




Whatman No. 1 filter paper, 18 x 22.5 inches, in such amounts as to produce




a colored spot easily detectable with the naked eye in a natural light or




under ultraviolet light.  A standard solution cbntaininp about 1.0% each




of the known monosaccharides, glucose, mannose, galactose, arabinose, xylose,




and rhamnose v?as also spotted on the filter papier.  The sugars were

-------
                                   -28-
separated by descending development with ethyl acetate-pyridine-vater (8:2:1




v/v/v) (25) as developer.  The solvent was allowed to migrate almost to the




bottom of the papers at which time they wets removed from the tank and




air-dried for at least 6 hours.  Some of the papers were returned to the




tank and developed as before (repeated up to 3 or 4 tines) in order to




obtain a better separation.




     The paper chromatograms were sprayed with <>-aminodiphenyl reagent




(0.4 p £-aminodiphenyl dissolved in 100.0 ml of glacial acetic acid and




20.0 ml of distilled water) and heated at 100i29 in an oven for 5 minutes




(26, 19).  The spots were outlined in pencil untfer ultraviolet light.  The




rates of movement of the hydrolyzate sugars werfe compared with those of




authentic samples when run simultaneously on the same chromatograms.




     The o-aminodiphenyl was purified by recryskallizing the technical




grade material twice from aqueous ethanol.  Activated charcoal was used




as a decolorant.  The purified crystals were dried at room temperature




under vacuum.




7.  Purification of the Hot-Water-Soluble Solids




     A portion of the hot-water-soluble solids (3,0-4.0 g) was stirred into




a small amount of distilled water until it was thoroughly wetted.  Distilled




water at room temperature was then added slowly with stirring until the




mixture had a concentration of 1.0%.  Stirring was continued for at least




2 hours at room temperature.  The mixture was ail lowed to stand and was




then cenfcrifuged.  The supernatant liquid was recovered.  The undissolved




solids were again stirred into water and the whole process repeated 3 more




times.  The liquor from the fourth washing was colorless and clear.




     The undissolved solids which remained were freeze-dried; yield 29.8%




of the original hot-water-soluble solids.  Thesis were labeled "Fraction A."

-------
                                    /
                                    /

                                   -29-
A portion of "Fraction A" was tested for starch (positive) and tannins




(negative).  A second portion of "Fraction A" was hydrolyzed by strong acid,




and the hydrolyzate was examined on paper chronatograms.  The chromatograms




showed glucose only.  A large sample of "Fractiqn A" was later prepared for




additional research purposes.




     All of the liquors from the above treatments were combined and




concentrated to about 1.0 liter under vacuum at less than 30° temperature




in a rotary evaporator.  Ethanol (95%) was added to provide a final solution




of 70.0% ethanol (1.0 liter each tine).  After the fourth washing, the




precipitate was dissolved in water and traces of ethanol were removed under




vacuum in the rotary evaporator.  The polysaccharide was then recovered by




freeze-drying and labeled "Fraction B"; yield 25.9% of the original hot-




water-soluble solids.  A portion was tested for starch (positive) and




tannins (faintly positive).  Another portion was hydrolyzed under mild




acid conditions, and examined by paper chromatography.  The chromatoprams




showed the presence of amino acids and sugars.




     The mother liquor and washings from "Fraction B" were combined,




concentrated, and freeze-dried.  This sample was labeled "Fraction C".




The yield of "Fraction C" (by difference) was 44.3% of the original hot-water-




soluble solids.  A portion of this sample was tested for starch  (negative)




and tannins (positive).




8.  Enzyme Hydrolysis to 'Remove Starch




     The enzymes used were two commercial preparations purchased from




Marschall Division, Miles Laboratories, Elkhart, Indiana.  The first




enzyme, HT-1000 (27) is a mixture of anylolytic and proteolytic enzymes,




capable of faster and more economical liquifactions of starch than many




other a-amylases.  It has been derived from Bacillus sub tiles, and is in

-------
                                   -30-
the form of a white, dry powder.  The second enzyme (commercial name,


Diazyme L 30) is an amyloglucosidase (28).   It is sold in liquid form.


     A part (15.0 g) of "Fraction B" was dissolved in distilled water  (P^.l


ml) and the pH was adjusted to 5.5-7.0 with sodium carbonate solution


(1.0 N).  HT-1000 (7.5 mg, 0.05% based on sample weight) dissolved in  a


small amount of distilled water was added to the sample solution.  The

                              i
mixture was heated to 75° in a wofer bath with continuous agitation and


held at that temperature for 15 minutes.  Heating was then continued


until the temperature reached 85-87° and held at this level for 30-40


minutes.  At the end of this time, the sample and water bath were cooled


to 60°, and the pH of the sample was adjusted to 3.8-4.2 with hydrochloric


acid (0.1 N).  Diazyme  (0.09 raJ , equivalent to 80 units/lb starch) was


added directly to the cooled carbohydrate solution.  The mixture was


incubated at 60° with stirring for 72 to 96 hours.  At the end of the


incubation period, the mixture was transferred to a dialysis bap. and


dialyzed for two days in distilled water, one week in running tap water1


and another day in distilled water.  The non-dialyzahle portion was


concentrated in a rotary evaporator at less than 40° and the concentrate


was freeze-dried; yield 9.9% of the original hot-water-soluble solids.


     The materials passing out of the dialysis bag during the first two


days were collected, concentrated, and freeze-dried.  A portion of the


freeze-dried sample was dissolved in watert spotted on Whatman No. 1 filter


paper and paper chromatographed.  The chromatograms showed glucose only.


Another portion of the  sample was hydrolyzed under mild acid conditions


and the hydrolyzate was tested by paper  chromatography.  The chromatograms


showed glucose only.

-------
                                   -31-
9.  Enzyme Hydrolysis to Remove Protein


     The non-dialyzable, freeze-dried material recovered from the enzyme


hydrolysis to remove starch was dissolved in 1.5 liters of distilled


water.  Tris (hydroxymethylamino)-methane was added to the solution and


the pH was adjusted to 8.5 with 0.1 N hydrochlotic acid.  Chymotrypsin


(160 mg) and trypsin (150 ing), dissolved in a small amount of water, were


added to the buffered solution and the volume adjusted to 2.0 liters.  The


solution was placed in a hot-water bath and heated at a constant temperature


of 30-40° with constant stirring for 12 days.  At the end of the incubation


period, the solution was transferred to a dialysis bap. and dialyzed for


12 days against running tap water.  The precipitate which had formed


during the reaction was separated from the mother liquor by centrifupation


and washed three times with distilled water.  The mother liquor and the


washings were combined, concentrated in a rotary evaporator and freeze-dried


(29).  The freeze-dried material had a tan, fluffy appearance.  This sample


was labeled "Fraction D"; yield 1.5% of the original hot-water-soluble


solids.  "Fraction D" was hydrolyzed with 3.0% sulfuric acid and investigated


by paper chromatography.  The chromatograms showed the presence of galactose,


glucose, arabinose and traces of rhamnose, xylose, and mannose.


10.  Carbohydrate Analysis by Gas-Liauid-Chromatography


     The gas-chromatograph used was a Hewlett-Packard 5751B Research


Chromatograph (Hewlett-Packard Company, Palo Alto, California) equipped


with dual flame ionization detectors.  The conditions were:  column, 6.5%


ECNSS-M on Gas Chrom 0 100/120 mesh, 6 ft x 1/6 in. O.D. stainless steel;


injection port 200°, detector 235°; column temperature 175° isothermal;

                                       2
helium flov 30 ml/rain; range setting 10 ; attenuation setting 16.


     The various samples were hydrolyzed and th£ derivatives prepared for

-------
                                   -32-
injection into the gas chromatogsraph as follows.   The polysaccharide sample




(0.32 g) was dissolved in 3.02 sulfuric acid (96.0 ml) and refluxed for 5




hours (mild acid hydrolysis).  The solution was cooled and authentic




ayo-inositol (0.1000 g) was added.  The hydsrolyzate was then neutralized




to pH 5.0 with a saturated aqueous barium hydroxide solution.  The resulting




barium sulfate precipitate was removed by centrifuge.  An aliquot (25.0 ml)




of the clear supernatant solution was transferred to a round-bottomed flask




(IGOoO ml).  Sodium borohydride (0.08 g) was added to the flask and allowed




to react for 2 hours at room temperature (308 31).




     The excess sodium borohydride was decomposed by adding acetic acid




until gas evolution ceased.  The solution was concentrated to a sirup in




a rotary evaporator, and methanol (10.0 ml) was added and re-evaporated.




The addition and removal of methanol was repeated five times (32).  The




resulting sirup was dried in an oven at 105° for 15 minutes to ensure




complete removal of water.




     Acetic anhydride  (7.5 ml) and concentrated sulfuric acid (0.5 ml) were




added to the sirup and the solution was heated for 1 hour at 50-60° in a




water bath.  After cooling for 5 minutes0 the acetylation mixture was




poured slowly with stirring into about 70.0 ail of ice-water.  The mixture




was transferred to a separatory funnel and the alditol acetates were




extracted with three successive amounts of freshly distilled methylene




chloride (25.0 sal5 1590 ml and 10.0 ml).  The methylene chloride extract




was concentrated to dryness on a rotary evaporator at 75°.  Distilled




water (1.0 ml) was added to the residue and re-evaporated.  The alditol




acetates were dissolved in 2.0 ml of freshly distilled methylene chloride




and about 4.0 pi of the solution were injected into  the gas chromatograph




(21, 33).  The peaks In the resulting spectra wery identified by  comparison

-------
                                   -33-
of retention times with authentic known alditol acetates and by peak




enhancement techniques.  The areas under the peaks in the resulting




spectra were measured by means of a planimeter.








H.  Characterization of the Carbohydrates Solubilized by the Acidified



    Sodium Chlorite Delignification Reaction




     (Isolation described in section II-F)




1.  Ash Determination




     The inorganic ash was determined by a modification of the procedure




of Paech and Tracey (3A).




     Six samples (0.170 g to 0.180 g dry weight) were weighed into silica




dishes.  The samples were saturated with concentrated sulfuric acid (0.2 ml).




The mixtures were stirred thoroughly with glass rods and set aside for 1-1*5




hours.  They were placed in a muffle furnace and heated gently until




charring occurred.  The temperature was increased to 300° to drive off the




sulfuric acid (about 30 min).  The temperature was increased to 500° and




heating continued until ashing was complete (about 5 hours).  The ashed




samples were placed in a dessicator to cool and were weighed after 20 min.




The ash content was 13.39±1.57%.




2.  Dialysis and Precipitation with Ethanol




     In an attempt to separate and purify the carbohydrates in the acidified




sodium chlorite soluble solids, a sample was dialyzed and re-precipitated




(35).




     An aliquot (22.2 g, dry weight) was dissolved in distilled water (500




ml) and dialyzed against distilled water.  The water was changed each day




for four days, concentrated to about 200 ml under reduced pressure on a




rotary evaporator and lyophilized to yield a brov/n solid; weight 5.6 g or

-------
                                   -34-
22.4% of the starting material.  An aliquot of the fraction was  subjected




to paper chromatography, under conditions to be described in  detail  later.




     The solution in the dialysis bag was concentrated under  reduced




pressure on a rotary evaporator to a volume of 400 ml.  The concentrate




was added, with stirring, into 1200 ml of 95% ethanol to provide a  70-71%




etHanoi concentration resulting in the formation of a flocculent white




precipitate.  The mixture was allowed to stand at room temperature  for




several hours and then centrifuged.  The decantate was concentrated  and




lyophilized to yield a fluffy white powder; weight 7.4 g or 33.3% of the




starting material.




     The white residue was lyophilized to yield a fluffy white powder;




weight 7.7 g or 34.7% of the starting material; ash content 11.88±2.83%.




     Dialysis and precipitation did not lower the ash content of the




original material beyond the degree of error of the ash determinations,




and so all further work was done on the original material.




3.  Elemental Analysis for Nitrogen, Sulfur, Phosphorus and the Halogens




     The procedures used are slight modifications of the Lassaigne's sodium




fusion method (36, p. 1039).  A small glass test-tube was supported in a




clamp and a small cube of sodium metal was added.  The tube was gently




heated in a flams until  the sodium melted and the vapors rose 1-2 cm up




the walls of the tube.  A small amount of solid was added directly




onto the molten sodium.  The tube was strongly heated over an open flame




until the entire end was red hot.  The heating was continued for one or




two minutes and the tube was then plunped into an evaporating dish




containing about 10 ml of distilled water so that the hot tube shattered.




The mixture was heated to boiling, the insolubles removed by filtration,




and the filtrate recovered for elemental analysis.

-------
                                   -35-
     An aliquot (2-3 ml) of the filtrate was added to a test-tube containing



about 0.1 g of powdered ferrous sulfate.  The mixture was gently heated



with shaking until it boiled.  Without cooling, sufficient dilute sulfuric
          #


acid was added to dissolve the iron hydroxide and give the solution an



acid reaction.  A precipitate of Prussian blue formed, indicating the



presence of nitrogen.



     A small aliquot (15.0 mg dry weight) of the acidified sodium chlorite



soluble material was quantitatively analyzed (Pascher and Pascher, 53 Bonn,



Buschstrasse 54, West Germany); nitrogen content, 0.84%.



     A second aliquot (2.0 ml) of the filtrate from the sodium fusion



reaction was acidified with dilute acetic acid and a few drops of lead



acetate was added.  No reaction resulted, indicating that no sulfur was



present.



     A third aliquot (1.0 ml) of the filtrate from the sodium fusion reaction



was acidified with 3.0 ml of concentrated nitric acid and boiled for one              :



minute.  The solution was cooled and an equal volume of ammonium molybdate



reagent was added.  The solution was warmed to 40-50° and allowed to stand            j



but no yellow precipitate formed, indicating that no phosphorus was present.



     A fourth aliquot (2.0 ml) of the filtrate from the sodium fusion                 j



reaction was acidified with dilute sulfuric acid and boiled gently until              >



it had been reduced to about 1 ml to remove any hydrogen cyanide which



might be present.  A few drops of aqueous silver nitrate was added.  No

                                                                                      i

precipitate formed, indicating that no halogens were present.                         >

                                                                                      !

4.  Strong Acid Hydrolysis                                                            i



     An aliquot (177.6 mg dry weight) of the solids solubilized by the



acidified sodium chlorite reaction was dissolved in 77% sulfuric acid



(3.0 g) and allowed to stand for 30 minutes at room temperature.  Water

-------
                                   -36-
(56.29 g) was slowly added with stirring to provide a 3.9% sulfurlc acid




solution.  The solution was refluxed for 5 hr., cooled to room temperature,




and neutralized to pH 5.0 by tltration with a saturated aqueous barium




hydroxide solution.  The resulting precipitate of barium sulfate was




removed by centrifuge and washed well with water.  The decantate plus




washings were concentrated on a rotary evaporatory to 50 ml.




5.  Mild Acid Hydrolysis




     A part  (17706 mg dry weight) of the solids solubilized by the acidified




sodium chlorite reaction was dissolved in 3% sulfurlc acid (50 ml) and the




solution was refluxed 5 hr.  After cooling to room temperature the solution




was titrated to pH 5.0, with a saturated solution of aqueous barium




hydroxide.  The resulting precipitate of barium sulfate was removed by




centrifuge and washed well with water.  The decantate plus washings were




concentrated to 25 ml on a rotary evaporator.




6.  Qualitative Amino Acid Analysis by Paper Chromatography




     The hydrolyzates from the strong acid treatment and the mild acid




treatment were subjected to two dimensional paper chromatography  (Whatman




No. 1 paper) using water-saturated phenol  (beakers of 0.3% ammonium hydroxide




were placed  in the bottom of the developing tank) as developer in one




direction and rv-butanol-formic acid~water  (20:6:5 v/v/v) as developer in




the second direction.  After air-drying, the papers were sprayed with




ninhydrin spray reagent  (0.02% ninhydrin in n-butanol) and heated at




100±5° in an oven for 5 minutes  (37, pp. 93-96).




7.  Qualitative Carbohydrate Analysis by Paper Chromatography




     The hydrolyzates from the strong acid hydrolysis and the mild acid




hydrolysis were subjected to paper chromatography using ethyl acetate-




pyridine-water (8:2:1 v/v/v) as developer.  The solvent was allowed to

-------
                                   -37-
mlgrate almost to the bottom of the papers at which time they were removed


from the tank and air-dried.  The papers were returned to the tank and


developed as before (repeated 3 tines).  The paper chromatograms were


sprayed with o-aminodiphenyl reagent (0.4 g o-aminodiphenyl dissolved in


a solution prepared from 100 ml glacial acetic acid and 20 ml distilled


water) and heated at 10012° in an oven for 5 min (19).


8.  Quantitative Carbohydrate Analysis by Gas-Liquid Chromatography


     The gas-chromatograph used was a Hewlett-Packard 5751B Research


Chromatograph (Hewlett-Packard Company, Palo Alto, California) equipped


with dual flame lonlzation detectors.  The conditions were: column, 6.5%


ECNSS-M on Gas Chrom Q 100/120 mesh, 6 ft. x 1/8 in. O.D. stainless steel;


injection port 200°; detector 230°; column temperature 180° isothermal;

                                       2
helium flow 30 ml/min; range setting 10 , attenuation setting 16.


     The solids solublllzed by the acidified sodium chlorite reaction were


hydrolyzed and the derivatives prepared for injection into the gas


chromatograph as follows.  A portion of the solids (0.2721 g, dry weight)


was dissolved in 32 sulfuric acid (80 ml) and refluxed for 5 hours.  After


cooling, authentic myo-lnosltol (0.1004 g) was added.  The solution was


neutralized to pH 5.0 with a saturated aqueous solution of barium hydroxide.


The resultant precipitate of barium sulfate was removed by centrifuge and


the decantate plus washings were concentrated to about 25 ml and transferred


to a 100-ml round-bottomed flask.  Sodium borohydride (0.08 g) was added


to the flask and allowed to react for 2 hr. at room temperature (30).


     The excess sodium borohydride was decomposed by adding acetic acid


until gas evolution ceased.  The solution was concentrated to a sirup on


a rotary evaporator and methanol (10 ml) was added and re-evaporated.  The


addition and removal of methanol was repeated five time (32).  The resulting

-------
                                   -38-
sirup was dried in an oven at 105° for 15 min to ensure complete removal.




of water.




     Acetic anhydride (7.5 ml) and concentrated sulfuric acid (0.5 ml)




were added to the sirup and the solution was heated for 1 hour at. 50-60°




in a water bath.  After cooling for 5 min. the acetylation mixture was




poured slowly with stirring into about 70 ml of ice water.  The mixture




was transferred to a separatory funnel and the alditol acetates were




extracted with three successive amounts of methylene chloride (25 ml, 15




ml, and 10 ml).  The methylene chloride extract was concentrated to dryness




on a rotary evaporator at 75°.  Water (1 ml) was added to the residue and




re-evaporated.  The alditol acetates were dissolved in 2 ml of methylene




chloride and about 1.0 pi of  the solution was injected into the gas




chrotnatograph for quantitative analysis (31) .




     The areas under the peaks in the resulting spectrum were measured




with a planimeter.




9.  Qualitative Uronic Acid Analysis by Infrared Spectroscopy




     A sample (0.5 mg) of dried solids solublllzed by the acidified- sodium




chlorite with potassium bromide (200 mg) and pressed into a pellet.  A




spectrum was taken of the pellet over the range of 4000 to 300 cm




(Beckman IR-20A).  The residue from the 0.5% ammonium oxalate extraction




was also analyzed, by Infrared spectroscopy.




     An infrared spectrum was also obtained  in a Nujol mull prepared by




grinding it to a thick paste  in Nujol oil (paraffin oil).




10.  Qualitative Uronic Acid  Analysis by, Color Reactions




     Concentrated sulfuric acid (6.0 ml) was added slowly to an aqueous




solution (1.0 ml) of the 3% sulfuric acid hydrolyzate of the solids solubilized




by the acidified sodium chlorite reaction and cooled under tap water.  The

-------
                                   -39-
reaction mixture was heated for 20 min in a boiling v/ater bath and cooled.




An aliquot (0.2 ml) of a 0.1% ethanolic solution of carhazole was added




and the test sample allowed to stand for 2 hours at room temperature.  The




development of a purple color indicated a uronic acid.  The solution was




scanned in the ultraviolet range and showed maximum absorption at 535 nm




indicating the presence of a uronic acid (38, p. 497).




     An aliquot (177.6 g dry weight) of the solubilized solids was dissolved




in 3% sulfuric acid (50 ml) and the solution was refluxed for 5 hr.  The




hydrolyzate was neutralized by passage through a column of Amerlite IRA-400




anion exchange resin in the -OH form.  An amount (1.0 ml) of the neutralized




solution was tested for uronic acids by the carbazole-sulfuric acid test




described above.  The results were positive.




     A small aliquot of the 0.5% ammonium oxalate residue was also tested




with carbazole-sulfuric acid.  The results showed a strong positive test.




Color tests were also run on authentic galacturonic acid (positive test),




oxalic acid and benzoic acid (negative tests).




11.  Qualitative Uronic Acid Analysis by Paper Chromatography




     The hydrolyzate prepared from the mild acid hydrolysis of the solids




solubilized by the acidified sodium chlorite reaction was subjected to




paper chromatography.  Whatman No. 1 filter paper was employed in conjunction




with the following solvent systems:




     1.  water saturated n-butanol-absolute ethanol-water (10:9:1 v/v/v)




     2.  water saturated n-butanol-acetone-water (4:5:1 v/v/v)




     3.  ethyl acetate-pyridine-water (8:2:1 v/v/v).




     Authentic galacturonic acid v/as chromatographed simultaneously with




the unknown hydrolyzate.




     For the first two solvent systems, the papers were first impregnated

-------
                                   -40-
vlth a phosphate buffer solution prepared by titrating a 0.01 M solution


of disodlum hydrogen phosphate to pH 5.0 with a 0.1 M solution of



phosphoric acid (39).



     The chromatograms developed with solvent three were alternately



irrigated and air-dried for a total of 6 times.  They were thus in the

                                                                      'ln!^: 1 :.

solvent atmosphere for a total of about 30 hr.              .:           ,
                                                            ;            . i .

     After development, the paper chromatograms prepared from the first



two solvent systems were sprayed with aniline hydrogen phthalate reagent



(1.66 g of phthalic acid dissolved in 100 ml of water-saturated n-butanol



containing 0.93 g of freshly distilled aniline) (40).  The chromatograms


prepared from the third solvent system were sprayed with the o_-aminodiphenyl


reagent.  After.spraying, all chromatograms were heated in. an oven at


100±2° for 5 min.



12.  Preparation of Methyl Ethers



     An aliquot of the solids solubilized by the acidified sodium chlorite


reaction (1.461 g dry weight) was dissolved in 18% aqueous sodium hydroxide


at 0° with stirring followed by the addition of 6.0 g of sodium hydroxide



pellets.  Sodium hydroxide  (100 ml, 30%) and dimethyl sulfate  (50 ml) were



added simultaneously over a period of 3 hours while maintaining the



temperature at  0°.  Acetone  (150 ml) was added to prevent foaming and the


mixture was stirred for an additional 45 hr at room temperature.  The


solution was cooled to 0°, neutralized with 10% sulfuric acid, dialyzed



against water  (3 days), concentrated under reduced pressure  (200 ml),  .


lyophilized, and the entire methylation sequence repeated (41).         '


     The product was further methylated by the method of Falconer and


Adams (42) by solution in 150 ml of tetrahydrofuran followed by treatment


over a period of 60 hours with seven 10 g portions of crushed  sodium

-------
                                   -41-
hydroxide, each followed by a 12 ml portion of dimethyl sulfate.   The




solution was stirred vigorously and additional solvent was added  as needed




to maintain fluidity.  The product was recovered as described above for




the first methylation, and the concentrated dialyzate (600 ml) was




extracted with three 300 ml portions of chloroform.  The chloroform layer




was concentrated under reduced pressure, yielding a light brown solid




which was dissolved in acetone and filtered.  The concentrated filtrate




(50 ml) was poured into petroleum ether (b.p. 30-60°, 200 ml) and the white




precipitate which formed was recovered by centrifuge (after the mixture




was refrigerated for 6 days) and lyophllized, affording a white,  fluffy




powder; yield 80 mg.




13.  Acid Hydrolysis of the Methyl Ethers




     An aliquot (40 mg) of the methylated material prepared above was




hydrolyzed for 3 hr at 97° with 88% formic acid (10 ml).  The formic acid




was removed by evaporation under reduced pressure followed by the addition




and removal of water, then hydrolyzed with 0.5 N sulfuric acid (5 ml for




2.5 hr at 97°).  Upon cooling the solution was neutralized with water-




saturated barium hydroxide and the solids were removed by cehtrifugation.




The decantate was concentrated to about 2 ml of sirup.  The sirup was




chromatographed on paper using the developer ethyl acetate-pyrldine-water




(8:2:1 v/v/v).




I.  Characterization of_ the Holocellulose Fraction




1.  Elemental Analysis for Nitrogen, Sulfur, Phosphorus, and the Halogens




     To a small test tube (75 x 12 mm), supported in a clamp, was added




0.06 g of sodium metal.  The tube was heated in a Bunson Burner flame until




the sodium melted and the vapors rose 1-2 cm up the walls of the tube.  The




tube was strongly heated over an open flame until the entire end was red hot

-------
                                   -42-
The tube was then plunged into a small beaker containing 10 ml of distilled




water.  The hot tube was shattered.  The mixture was heated to ..boiling.  The




insolubles were removed by filtration, and-the filtrate recovered for




elemental analysis.                                         .  •




     A second 10 ml of filtrate from the sodium fusion reaction was similarly




prepared.                                                    ^




     An aliquot (2.5 ml) of the above filtrate was added:to a test tube




containing about 0.1 g of powdered ferrous sulfate.  The mixture-was ;gently




heated with shaking until it boiled.  Without cooling, justcenough'dilute




sulfuric acid .was.added to dissolve the iron hydroxide and give the solution




an acid reaction.  No.precipitate of Prusslon Blue formed, indicating the




absence of nitrogen.                                          ,,




     A known nitrogen containing compound, alanine (0.02 g)., was fused with




sodium aa described above.  A.precipitate of Prussian Blue formed upon




reacting the.acidified.sodium .fusion solution with ferrous sulfate:indicating




the presence of nitrogen.




     A second aliquot  (2.0 ml) of the filtrate'from the sodium fusion reaction




was acidified.with dilute acetic acid.  A few drops of lead acetate were




added.  No yellow precipitate formed which indicated no sulfur.




     A known sulfur-containing compound, cystine, was fused ;wi.th sodium as




above.  A yellow precipitate formed in the filtrate by acidifying with




acetic.acid and .adding a few drops of lead acetate.  This indicated the




presence of sulfur.




     A third aliquot  (2.0 ml) of the filtrate from the.sodium fusion reaction




was acidified with 3.0 ml of concentrated nitric acid and boiled 'for one




minute.  The solution was cooled and to it was added an equal volume 'of




sjipmonium molyb.date reagent.  The solution was warmed to 40-50" .and allowed

-------
                                   -43-
to stand.  No yellow precipitate formed which Indicated the absence of




phosphorous.




     A known phosphorus containing compound, p.lucose-1-Phosphate was fused




with sodium as described above.  A yellow precipitate formed upon reacting




the acidified sodium fusion solution with ammonium molyhdate Indicating the




presence of phosphorus.




     A fourth aliquot  (2.0 ml) of the filtrate from the sodium fusion




reaction was acidified with dilute nitric acid and added to an excess of




silver nitrate solution.  No precipitate formed Indicating the absence of




halogens.




     Known sodium chloride (0.02 g) was fused with sodium as described above.




A precipitate formed when the acidified filtrate was added to a silver




nitrate  solution, indicating the presence of a halogen.  The mother liquor




was decanted and the precipitate was treated with dilute aqueous ammonia




solution.  A white precipitate formed which was readily soluble in the ammonia




solution indicating the presence of chlorine.




2.  Determination of Lignin




n.  Acid-Insoluble Lignin




     A portion  (0.4566 g, dry weight) of bark holocc-llulosc was placed  in




a 500-ml,  three-necked round-bottoned flask submerged  in a cold-water bath




(18-20°).  An awount (10.0 ml) of cold  (13-15") 72% siilfuric acid was added




slowly with stirring.  The sample was allowed to stand, with frequent




stirring,  for 2 hou:t>  at 18-20°.  The holocellulose became completely




dispersed  in the acid.  The transparent sirup was diluted  to 3.0% sulfuric




acid concentration by  slowly adding 376.0 ml of distilled water.  The




sample was refluxed for 4 hours.  The Insoluble material was recovered  by




filtration using a Cooch crucible  (fine porosity) which had previously

-------
                                   -44-
been dried and weighed.  The residue In the crucible was washed free of


acid with 150.0 ml of hot water and dried In an oven at 105±2° until the


weight became constant  (18 hours) (43).  The average of three determinations


was 0.0139 g or 3.05% of the holocellulose.
               t      -    •

b.  Acid-Soluble Lignin
               •-'                                             :

     The filtrates from  the above Klason lignin determinations were cdmbined


to yield a solution of 1544.0 ml.  The solution was analyzed, for the acid-


soluble lignin content by the characteristic lignin absorptions; at 280 nm


and 210 nm (44, 45, p. 259).  The instrument used was a Beckmari AGTA TM 111


UV-Visible spectrophotometer.  The instrument was standardized by placing


3.0 ml of 3.0% sulfuric  acid solution in both the reference and sample cells


(cell widths, 1.0 cm) and scanning over the ultraviolet spectral range from


370 nm to 200 nm.  The sample, cell was cleaned, dried and 3.0 ml of the acid


filtrate from the Klason lignin determination was added.  The sample was


scanned over the, ultraviolet spectral range from 320 nm to 200-rim.  Absorption


peaks were recorded at 280 ran (absorbance 2.60) and at 210 nm (absorbance 3.77)


3.  Hydrolysis with  7.7..0% Sulfuric Acid
                (

     Holocellulpse (913.2 mg, dry weight) was dissolved in 12/.4 g, of 77.0%


sulfuric acid in a 1-liter three-necked, round-bottomed flask, at ice-water


temperature for 1 hour. ' The transparent sirup of dissolved holocellulose


was diluted to 32 sulfuric acid by the dropwise addition of 247.0 ml of


distilled water from a dropping funnel.  Small, insoluble particles were


observed In the dilute  solution.  The dilute solution was refluxed for six


hours.  The small, insoluble particles remained.  The solution was filtered.


The filtrate wae neutralized to pH 5 with saturated aqueous.barium hydroxide


solution.  The resulting precipitate of barium sulfate was removed*by•


centrifuge and washed, well with water.  The decantate plus washings were  •
                 *

concentrated on a rotary evaporator to a sirup.

-------
                                   -45-
4.  Qualitative Carbohydrate Analysis by Paper Chromatography




     The hydrolyzate from the sulfuric acid hydrolysis was separated into




its component oonosaccharides by paper Chromatography.  Three drops of the




hydrolyzate sirup was dissolved in 1.0 ml of distilled water.  Five




micropipettes of the above dilute hydrolyzate were spotted on Whatman No. 1




chromatographic paper.  A solution of a known monosaccharide mixture




containing 1.0% each of glucose, galactose, mannose, arabinose and xylose




were placed along with the unknown hydrolyzate on the front of the




chromatographic paper 3 cm from one end.  The chroraatographic sheet was




developed by the descending method in a chromatographic tank of dimensions




61 cm x 67 cm x 82 cm in height, pre-saturated for one day with the developer




ethyl acetate-pyridine-water (8:2:1 v/v/v).  The solvent was allowed to




migrate almost to the bottom of the papers (6-8 hr).  The papers were




removed from the tank and air dried.  They were placed back in the tank




and developed again as above (repeated 3 times).  The papers were sprayed




with o-aminodiphenyl reagent (0.4 g o-aminodipher.yl dissolved in a solution




prepared from 100.0 ml of glacial acetic acid and 20.0 ml of distilled




water) and heated at 100±2° in an oven for 5 min to develop the color.  The




monosaccharides after separation and reaction with the color reagent were




investigated under ultraviolet light.  Hexoses showed a white color and




pentoses showed a red brick color under the ultraviolet light.




5.  Isolation of Crystalline Sugar Derivatives




a.  Isolation of'Sugars by Preparative Paper Chromatography




     A part (6.0 g, dry weight) of bark holocellulose was added to 77.0%




sulfuric acid (90.0 g) in a 5-liter round-bottomed flask which was submerged




in an ice-water bath.  The mixture was stirred for 1 hr at the end of which




time it appeared as a transparent sirup.  Water (1688.0 ml) was added slowly

-------
                                   -46-
to the sirup to dilute the sulfuric acid to 3% concentration..  The solution


was refluxed for a total of six hours.  It was allowed to cool  to room

                                                             'I
temperature and neutralized to pH 5.0 by the careful addition of aqueous


saturated barium hydroxide solution.  The resulting precipitate of barium


sulfate was removed by centrifugalion and washed well with water. The


decantate plus washings were concentrated to about 20.0 ml on a rotary
                      ,j           ~  -                   •.•,•'

evaporator.                                                  ...


     Aliquots of approximately 0.1 ml each of the concentratedjhydrolyzate


were streaked with a syringe across large pieces of Whatman Njj.. 3 MM


chromatographic paper, 9 cm from one end of the paper.  After ,the streaks,


were dry, the papers were developed by the descending method using the


solvent system ethyl acetate-pyridine-water (8:2:1 v/v/v).  The solvent


was allowed to migrate almost to the bottom of the papers  (about 6-8 hr).


The sheets were removed and dried in the air overnight.  The sheets were


developed and air-dried again (repeated 5 more times).  A  strip 1.0 cm in


width was cut from the center of each sheet along the direction of solvent


migration.  The strips were sprayed with o_-aminodiphenyl reagent and -dried


in an oven at 105±3° for 10 rain.  The degree of separation was_examined


under ultraviolet light.  It required six solvent migrations .to the bottom


of the sheets to adequately separate the monpsaccharides  (glucose, galactose,


mannose, arabinpse, and xylpse) in the holocellulpse hydrolyzate,


     After assurance that th* sugars were well separated,  1.0 cm strips  were


cut from the papers in the direction of solvent flow at 6  cm intervals across


the width of the papers.  These strips were sprayed with o-aminodiphenyl
                                                                          «
indicator as before.  After location of the sugars, the strips, were re-fitted


back into the original paper chromatograms and the unsprayed cross^bands pf


each sugar cut out.  The sugars were collected from 24 such paper chronatpgrams.

-------
                                      -47-
        The  paper  bands  thus  collected were placed  in  1  liter beakers  (a




   different beaker  for  each  sugar),  and  the  sugars eluted with 800.0  ml of




   distilled water.   The eluate  of  each isolated sugar was concentrated on a




   rotary  evaporator to  a sirup.  Methanol was added to  the sirup and  then




   re-evaporated (repeated 3  tines),  and  final drying  was achieved  in  a




   dessicator.




        Each isolated sugar sirup was re-chromatographed on paper simultaneously




   with a  solution of known sugars  (glucose,  galactose,  mannose, arabinose and




   xylose; each 1% concentration).   In this way each Isolated sugar sirup was




   of  assured purity before preparation of a  crystalline derivative was




   attempted.




   b.   Diethyl  Dithloacetal Acetate Derivatives of  Galactose and Arabinose




        Diethyl dithioacetal  acetate derivatives of known monosaccharides were




e  synthesized  for direct comparison with derivatives  of the sugars from the




   holocellulose hydrolyzate.  The  general procedure for each known sugar was




   that reported by  Wolfrom and  Karablnos (46).  Galactose is used  as  a typical




   reaction  illustration.  Galactose (100.0 mg) was dissolved in concentrated




   hydrochloric acid (1.0 ml,  12 N) in a  100-ml round-bottomed flask submerged




   in  an ice-water bath.  Ethyl  mercaptan (1.0 ml)  was added under  a well




   ventilated hood and the mixture  was stirred for  one hour.  The mixture was




   neutralized  at  ice-bath temperature by the addition of ammonium  hydroxide




   (13-14  drops, 15  N).   During  neutralization, white  smoke resulted and a




   green precipitate was formed.  The mixture was concentrated to dryness under




   aspirator vacuum  on a rotary  evaporatory at 40°.  Dryness was accomplished




   by  the  addition and re-evaporation of  absolute ethanol (repeated 5  times).




   The greenish-yellow color  disappeared  after the  second or third  evaporation




   of  ethanol,  and a white solid resulted.

-------
                                   -48-
     The white solid was acetylated with a mixture (6.0 ml) of acetic anhydride



and anhydrous pyridine (2:1 v/v).   The reaction was allowed to continue



overnight at room temperature.  The solution was poured into distilled water



(10.0 ml) and extracted twice with chloroform (15.0 ml each time).   The



chloroform extract was washed three times with saturated aqueous sodium



bicarbonate (10.0 ml each time).  The acid-free chloroform extract  was



concentrated on a rotary evaporator to a sirup.  Crystallization was attempted
              i


by dissolution, of the- sirup in methanol followed by the dropwise addition of



water until a permanent cloudiness resulted.  Crystallization.of some of the



derivatives was also attempted from 95% ethanol.



     Diethyl dithioacetal acetate derivatives of known mannpse, galactose,



arabinose and xylose were similarly prepared.  However, only galactose diethyl



dithioacetal acetate (m.p. 75/.0-77.00; literature value 76.5-77.0°), and



arabinose diethyl dithioacetal acetate (m.p. 78.0-79.0°; literature valfle



79.0-80.0°) crystallized.  Therefore, similar derivatives of only the



galactose and arabinose sirups isolated from the holocellulqse were prepared:



galactose diethyl dithioacetal acetate, m.p. 76.5-77.5°, unchanged on


                                     23 5
admixture with, authentic material [a]  '  + 10.0°  (£ 4.0, chloroform)



arabinose diethyl dithioacetal, m.p. 77.0-78.5°, unchanged on admixture


                           23.5
with authentic material [ot]_  *  - 26.40 (c_ 1.54, chloroform)  literature



value -30.0°.



c.  Acetate Derivitave of Glucose



     An amount (5.4 ml) of acetic anhydride was heated to about 100° and



0.44 g of anhydrous sodium acetate was added.  In  small portions, 1.0 g of



D-glucose was added with.-stirring over the course  of 30 min..



     When the addition was. completed, the solution was cooled, to room



temperature, poured into 16. ml of ice-water, and stirred for  30; min.  The

-------
                                   -49-
acetylated product was extracted with three 20-ml portions of chloroform.



The extracts were washed with 20 ml of distilled water and concentrated



on a rotary evaporator to a sirup.  Glucose pentacetate was crystallized



by dissolution of the sirup in methenol followed by the dropwise addition



of water until a permanent cloudiness resulted.  After two recrystallizations



from 95% ethanol, white crystalls of glucose pentaacetate were obtained;



(yield 234.0 mg, m.p. 133.5-135.0; literatuve value 135°) (47).



     Glucose sirup (250 mg) isolated from the bark holocellulose hydrolyzate



by preparative paper chromatography was similarily acetylated.  White



crystals of glucose pentaacetate were obtained; yield 112 mg, m.p. 133-134°,


                                                   23 5
unchanged on admixture with authentic material, [a]  "  + 3.9° (c 3.4,
                                                   L)


chloroform), literature value 3.8°.



d.  Di-()-benzylidene Dimethyl Acetal Derivative of D-Xylose



     The reagent for preparing di-0-benzylidene dimethyl acetal derivatives



of D-xylose was prepared by dissolving 2 ml of redistilled benzaldehyde in



a mixture of 2.5 N methanolic hydrogen chloride (1.0 ml) and spectroquality



methanol (6.0 ml).  An aliquot (2.5 mg) of D-xylose was treated with 5.0 ml



of the reagent at room temperature.  After standing for 1 hr, silky needles



appeared in the solution.  After standing for 6 hr the solution solidified.



The white crystals were recovered by filtration, washed with 200 ml of ice-



water and dried in a dessicator for 3 days.  When recrystallized from



spectroquality methanol, silky needles of di-0-benzylidene-D-xylose dimethyl



acetal were obtained; m.p.210-211.5°, literature value 211-212° (48, p. 88).



     Xylose sirup isolated by preparative paper chromatography from bark



holocellulose hydrolyzate was similarily treated with the prepared reagent



and crystalline needles were obtained.  After recrystallization from



spectroquality methanol, silky needles of dt-0-benzylidene-D-xylose dimethyl

-------
                                   -50-
acetal were obtained; yield 90.7 nip,, m.p. 211-212.5°, unchanged on admixture


                            23 5
with authentic material, [al  '  - 8.0°  (c 1.0, chloroform) literature
                            1)            —


value -7.0° (48, p. 88).



e.  Phenylhydrazone Derivative of Mannose



     D-raannose (1.0 g) was added to a solution of phenylhydrazine (1.0 ml)



in 95% ethanol (15.0 ml) in a 50-ml round-bottomed flask.  The mixture was



warmed for 30 min  in a hot-water bath (60°) and then placed in the



refrigerator overnight.  The resulting crystals of mannose phenylhydrazone



were removed by filtration and washed successively with a few drops each of



water, ethanol,, and diethyl either; yield 94.5 tng, m.p. 191-192°.  The



crystals were again washed with a few drops each of water, ethanol and



diethyl ether; yield 90.5 mg of glistening snow-white crystals, m.p. 194-195°,



[a]~3'5 + 2.5° (c  0.1 pyridine), literature value 199.0-200.0° [u]p6(c 0.1



pyridine) (49, p.  147).



     An aliquot (9.0 g air-dried weight, moisture content 8.68%) of



holocellulose was  hydrolyzed with 77.0%  sulfuric acid as previously



described.  One-third of the resulting hydrolyzate was concentrated to



a sirup.  The sirup was dissolved in 95% ethanol (45.0 ml) followed by the



addition of phenylhydrazene  (3.0 ml) (49, p. 107).  The mixture was warmed



for 30 min in a hot-water bath  (55°), cooled, and kept in the refrigerator



overnight.  The resulting white crystals were recovered by filtration and



washed successively with water, ethanol  and diethyl ether; m.p. 160-180°.



These crystals were dissolved in water (30.0 ml) and kept in the refrigerator



overnight.  The insoluble white crystals were recovered by filtration and



dried; yield 226.7 mg, m.p.  193.5-194.5, unchanged on admixture with


                                      235
authentic mannose  phenylhydrazone;  [a]   *  + 25° (£0.1, pyridine) literature



value +26° (49, p. 147).

-------
                                   -51-
     Another aliquot (5.0 g air-dry weight, moisture content 8.68%) of



holocellulose was refluxed with 90% formic acid (50.0 ml) for 1 hr (41).



The residue was removed by centrifugation and filtration.  The filtrate



was concentrated to a sirup which was hydrolyzed with 0.5 N sulfuric acid



(50.0 ml) under reflux for 2.5 hr.  The acid hydrolyzate was neutralized



to pH 5.0 with a saturated aqueous solution of barium hydroxide and the



resulting precipitate was removed by centrifugation.  The remaining



inorganic salts were removed by refluxing the concentrated sirup with



absolute methanol followed by filtration (repeated four times) and removal



of solvent.



     The resulting sirup was dissolved in 95% ethanol (15.0 ml) followed by



the addition of phenylhydrazine (1.0 ml) (49, p. 147).  The solution was



warmed  (55°) for 30 min and then kept in a refrigerator overnight.  The



resulting white crystals were recovered by filtration washed successively
with water, ethanol, diethyl ether and dried; yield 132.4 mg; m.p. 194-196°,



                                                                  12:
                                                                  'U
                                                                  23.5
unchanged on admixture with authentic mannose phenylhydrazone; [a]n *   + 25C
 (c 0.1, pyridine), literature value +26°  (49, p. 147).



 6.  Quantitative  Determination of Reducing Sugars by Copper Reduction



 a.  Standardization of Sodium Thiosulfate Solution



     Potassium  iodate  (1.41387 g) was dissolved  in distilled water and the



 solution was diluted to  1  liter In a volumetric  flask.  An aliquot (100.0 ml)



 of the solution was removed with a pipette.  A 1.8 M potassium iodide



 solution (10.0 ml) and a 12 M hydrochloric acid  solution  (3.0 ml) were



 added to the aliquot.  This solution was  titrated with the sodium thiosulfate



 solution to be  standardized (about 0.005 M) to a starch end-point.  The



 exact molarity  (0.0058 M)  of the sodium thiosulfate solution was calculated



 from the average  of triplicate samples (50, p. 460; 51, p. 379).

-------
                                   -52-
b.  Preparation of Somogyi Copper Reagent




     Rochelle salt (40.0 g) (potassium sodium tartrate), disodium hydrogen




phosphate heptahydrate  (53.0 g), and 1.0 N sodium hydroxide (100,0 ml) were




dissolved in 500.0 ml of water and 80.0 ml of an aqueous solution containing




8.0 g of cupric sulfate pentahydrate was stirred in.  Finally, 180.0 g of




anhydrous sodium sulfate was added and dissolved.  The solution was diluted




to 1 liter and allowed  to  stand for 3 days (52, p. 383-386).  Potassium




iodate  (37.2 mg) was added to a 100-ml aliquot of the reagent to provide a




concentration of 3.72 mg per liter




c.  Determination of Total Reducing Sugars in the Holocellulose Hydrolyzate




     Holocellulose (300 mg air-dry weight; moisture content 8.68%) was




dissolved in 77% sulfuric  acid  (4.50 g) in a 150-ml three-necked round-bottomed




flask in an ice-bath.   The mixture was diluted to 3.9% sulfuric acid




concentration by the dropwise addition of water.  The solution was heated




to reflux and aliquots  (2.0 ml) were withdrawn at 30 min intervals.




     Each aliquot was neutralized to pH 2.0 with 3% aqueous sodium hydroxide




solution  (1.28 ml).  Each  neutralized solution was diluted to 25.0 ml  in a




volumetric flask.  Three aliquots (5.0 ml each) of these solutions were




added with a pipette to test tubes  (20 cm x 2.4 cm).  Similarly, 5.0 ml of




three standard glucose  solutions containing 1.0 mg, 0.75 mg and 0.5 mg per




5 ml of solution were added to  identical test tubes.  A blank of 5.0 ml of




water was added  to another test tube.  Somogyi reagent  (5.0 ml) was pipetted




into each of these test tubes under a stream of nitrogen gas.  The solutions




were heated in a boiling water  bath for 10 min and cooled in a water bath




for 10  min.  Each solution was  oxidized by adding 2.5% aqueous potassium




iodide  (0.5 ml)  and  2 N sulfuric acid  (1.5 ml) and each was allowed to stand




for 5 min with occasional  shaking.  The solutions were  titrated to a starch

-------
                                   -53-
end point with standarized sodium-thiosulfate (0.0058 M)  within 30 min.   The




amount of copper reagent reduced by the reducing sugars was obtained by




subtraction of the titer of the sample from the tlter of  the blank.  The




amount of reducing sugar in the test tubes of the unknown hydrolyzate was




calculated by direct comparison with the known amount of glucose in the




standard solution (22).




     The initial sample of holocellulose was refluxed a total of 10 hr and




then heating was discontinued while the aliquots which had been withdrawn




were analyzed.  The titration results showed that monosaccharides were still




being released after 10 hr of reflux.  Therefore, a second sample of




holocellulose (300 mg) was hydrolyzed by sulfuric acid as before but the




reflux time was extended to 20 hr.  Aliquots (2.0 ml) were removed every




two hours and analyzed for reducing sugars with the Somogyi reagent.  The




reducing power of the solution ceased to increase after 12 hr of reflux,




indicating that hydrolysis was complete at this time.




7.  Quantitative Carbohydrate Analysis by Gas-Liauid Chromatography




a.  Preparation of Authentic Aliditol Acetates




     Galactitol (0.50 g) was added to 10.0 ml of an acetylating reagent




composed of pyridine-acetic anhydride (1:1 v/v).  The mixture was refluxed




for 1 hr and poured into ice-water (40.0 ml).  A white precipitate formed




which was recovered on filter paper and washed well with water.  The white




solid (galactitol hexaacetate) was recrystallized 3 times from 95% ethanol;




m.p. 171°, literature value 171°  (30, 53).




     Glucitol (sorbitol), mannitol, arabinitol, and myo-inositol were




similarily acetylated:  glucitol hexaacetate, m.p. 99.5°, literature value




99-99.5° (30, 53); mannitol hexaacetate, m.p. 125.5°, literature value 126°




(30, 53); arabinitol pentaacetate, m.p. 76°, literature value 76°  (30, 53);

-------
                                   -54-
inositol hexaacetate, m.p. 215-216°, literature value 214-215° (54, p.  85).




     Xylitol was acetylated as above but the derivative would not crystallize




from 95% ethanol.  The oil of xylitol pentaacetate was poured into ice  water




and extracted 3 times with chloroform.  The chloroform extract was condensed




to a sirup and dissolved in 95% ethanol, 85% methanol, acetone, diethyl ether,




or a mixture of acetone, diethyl ether and ligroin.  The sirup was dried by




the addition and removal of methanol under vacuum (repeated 3 times).  The




xylitol pentaacetate then crystallized from acetone and was recrystallized




two times, m.p. 57°, literature value 61-62° (30, 53).




     Rhamnose (0.1 g) was dissolved in 100 ml of water in a 100-ml round-




bottomed flask and was reduced with sodium borohydride (0.8 g) for 2 hr at




room temperature.  Acetic acid was added to the solution until no gas evolved.




The solution was concentrated to a sirup and dried by the addition and removal




of methanol on a rotary evaporator (repeated 5 times).  The sirup was dried




in an oven at 105° for 15 min.  The dry sirup was acetylated with an




acetylating reagent of acetic anhydride (75 ml) and concentrated sulfuric




acid (5 ml) for 1 hr at 50-60°.  The acetylated mixture was cooled for 5 min




and poured into  ice water and extracted with chloroform (300 ml, 300 ml,




200 ml).  The chloroform  extract was condensed to a sirup and dissolved in




95% ethanol.  No crystals formed.  The sirup was dried several times by the




addition and removal of methanol on a rotary evaporator.  It still did not




crystallize from any solvent  tried.




     Each of the above crystalline alditol acetates was dissolved  in turn




in methylene chloride  (2.0%  solution) and  injected (2 pi) separately into




the gas chromatograph  (Hewlett-Packard 5751B Research Chromatograph equipped




with dual flame  ionization detectors, Hewlett-Packard Company, Palo Alto,




California).  The conditions were:  column, 6.5% ECNSS-M on Gas Chrom Q

-------
                                   -55-
100/120 mesh, 6 ft x 1/8 in O.D. stainless steel; injection port 180°;


detector 240°; column temp 180° isothermal; helium flow 30 al/min; range

          2
setting 10 , attenuation setting 16. .


     In this way the retention time of each alditol acetate was determined.


A mixture of the alditol acetates were dissolved in methylene chloride (2.0%


concentration of each) and an aliquot (2 yl) of the solution was injected


into the gas chromatograph.  Resolution of the six compounds was excellent,


and under the chromatographic conditions outlined above the retention time


ot each compound in the mixture was: xylitol pentaacetate 16 min; arabinitol


pentaacetate 12 min; mannitol hexaacetate 28 min; galactitol hexaacetate


33 min; glucitol hexaacetate 38 min; myo-inositol hexaacetate 48 min.  These


retention times were used to identify alditol acetates prepared from the


holocellulose hydrolyzate.


b.  Determine of Optimum Gas-Chromatographic Conditions


     Glucitol hexaacetate  (4.63 mg), galactitol hexaacetate (0.25 mg), mannitol


hexaacetate  (0.75 mg), arabinitol pentaacetate (0.25 mg), xylitol pentaacetate


(0.50 mg), and myo-inositol hexaacetate (6.25 mg) were dissolved  together  in


methylene chloride  (1.0 ml).  Aliquots  (2 ul) of the solution were injected


into the gas-chromatograph under various ranges of injection port temperatures


(180-230°), detector  temperatures (210-270°), and column temperatures (180-


185°).  The areas under the peaks of the resulting spectra were measured


with a planimeter.  The percent recovery was calculated by comparing the


peak areas of the other sugars to myo-inositol hexaacetate as 100%.  The


best recoveries were  realized under  the following conditions:  injection port


temperature, 180°; detector temperature, 240° and column temperature, 180°.


These were the conditions generally used in this work.

-------
                                   -56-
c.  Determination of an "Instrument K Factor"




     The response of the gas chromatograph to the alditol acetates was




determined by preparing 5 mixtures of the known alditol acetates in various




ratio concentrations as shovn in Table 5.




     Each of the mixtures in Table 5 was dissolved in methylene chloride (1.0




ml) and an aliquot  (2 yl) was injected into the gas chromatograph.  The




areas under the peaks of the resulting spectra were measured by a planimeter.




The percent recovery was calculated on the basis of myo-inositol hexaacetate




as 100%.  The "Instrument K Factor" for each alditol acetate was obtained




as the slope of the line obtained by plotting the ratio of the area of each




alditol acetate to  that of the standard (myo-inositol hexaacetate) against




the ratio of the weight of the alditol acetate to that of the standard




(myo-inositol hexaacetate).  The resulting "Instrument K Factors" were as




follows:  xylitol pentaacetate 0.95; arabinitol pentaacetate 0.96; mannitol




hexaacetate 0.96; galactitol hexaacetate 0.98; glucitol hexaacetate 0.92.




d.  Analysis of the Holocellulose Hydrolyzate




     An amount  (300 mg, air-dry weight, moisture content 8.68%) of holocellulose




was dissolved in 4.5 g of 77.0% sulfuric acid in an ice-bath for one hour




with occasional stirring.  After one hour, the translucent sirup was diluted




to 3.9% sulfuric acid by dropwise addition of 84.42 nil of distilled water




from a dropping funnel.  The dilute solution was refluxed for 6 hr.  The




solution was diluted to 100 ml from which 25 ml of hydrolyzate was removed.




To this' 25.0 ml of  hydrolyzate, 25 mg of myo-inositol was added.  The




hydrolyzate was neutralized to pH 5.0 with saturated aqueous barium hydroxide




solution, centrifuged, concentrated to 25 ml and transferred to a 100-nl




round-bottomed  flask.  Sodium borohydride (0.08 g) was added to the flask




and allowed to  react for 2 hr at room temperature  (30).

-------
                                   -57-
     The excess sodium borohydride was decomposed by adding acetic acid until




gas evolution ceased.  The solution was concentrated to a sirup on a rotary




evaporator, and methanol (10.0 tng) was added and re-evaporated.  The addition




and removal of methanol was repeated eight times (32).  The resulting solid




was dried in an oven at 105* for 15 min to ensure complete removal of water.




     An aliquot (8.0 ml) of a mixture containing acetic anhydride (7.5 ml)




and concentrated sulfuric acid (0.5 ml) were added to the dry solid and the




mixture was heated for 1 hour at 50-60° in a water bath.  After cooling for




5 min the acetylation mixture was poured with stirring into about 70 ml of




ice-water in a 150-ml beaker.  Two immiscible layers  (water layer on top




and acetate layer on the bottom) formed.  These were transferred to a 150-ml




separatory funnel and the alditol acetates were extracted with three




successive amounts of methylene chloride (25 ml, 15 ml, and 10 ml).  The




methylene chloride extract was concentrated to dryness on a rotary evaporator




at 75°.  Water (1.0 ml) was added to the sirup to remove remaining acetic




anhydride.  The resulting alditol acetates were dried in a dessicator.  The




alditol acetates (20.0 mg) were dissolved in methylene chloride and diluted




to 1.0 ml in a volumetric flask (1.5%).  An aliquot (2.0 ul) of the solution




was injected into the gas chromatograph for quantitative analysis (31).




     The areas under the peaks were measured with a planimeter and compared




to the area under the peak of myo-inositol hexaacetate which represented




the area from 25 mg of inositol.  The weights of the other sugars were thus




calculated as: glucose, 156.8 mg; galactose, 5.9 mg; mannose, 26.2 mg;




arabinose, 4.6 mg; xylose, 16.5 mg; total 198.6 mg.




     A second sample of holocellulose  (300 mg) was hydrolyzed as above except




that it was refluxed for 12 hr instead of 6 hr in 3.9% sulfuric acid.  The




alditol acetates were prepared and injected Into the gas chromatograph as above.

-------
                                   -58-
     The areas under the peaks were measured with a planimete'r and compared




to the area under the peak of mjro-inositol which represented the area from




25 mg of sugar.  The weights of the other sugars were thus calculated as:




glucose, 160.4 mg; galactose, 5.8 mg; mannose, 27.9 mg; arabinose, 7.8 mg;




xylose, 21.8 mg; total 223.6 mg.  This increase in sugar return after 12 hr




of reflux over the 6 hr of reflux is in agreement with the Somogyi values




for total reducing sugars which had indicated that a minimum reflux time




of 12 hr was required to completely hydrolyze the holocellulose.




J.  Fractionation of the Holocellulose into its Component Polysaccharides




1.  Impregnation with 2.0% Barium Hydroxide, then Extraction with 10.0%




    Aqueous Potassium Hydroxide Solution




     Air-dried holocellulose  (54.3 g, moisture content 8.68%) from Douglas-fir




inner bark was slurried for 20 min at 25° in 779.0 g of aqueous barium




hydroxide containing 63.4 g of barium hydroxide octahydrate.  The consistency




of the slurry was 6.0% based  on the holocellulose content and the concentration




of barium hydro::ide in the liquor was 4.4%.  At the end of the 20 min period




952.0 g of 18.38% aqueous potassium hydroxide solution was added which lowered




the slurry consistency to 2.81% and the barium hydroxide concentration to  2%.




The potassium hydroxide concentration of  the extraction liquor was 10.0%.




Treatment was continued for a second 20 minute period at 25°.  Separation




was attempted by  filtration through Whatman No. 1 filter paper and a filter




crucible.  It was difficult to filter the colloidal solution.  The difficulty




was overcome by filtration through milk filters-disks (rapid).  The solids




were washed with  distilled water, dialyzed for three days, freeze-dried  and




weighed; weight 42.3 g or 84.6% of the starting holocellulose.  They were




termed "Residue A"  (55).




     The filtrate plus the washings were  acidified with acetic acid to a pH

-------
                                   -59-
of 5.0 and condensed to 500.0 rol on a rotary evaporator.  The polysaccharide




solids were precipitated from the filtrate by the addition of 500.0 ml of




methanol.  The precipitate was recovered by centrifuge and redispersed in




700 ml of 70% methanol.  The re-precipitation was repeated 4 times.  Methanol




was removed from the precipitate by the addition of water followed by




evaporation using a ratuvapor.  The solids were recovered by freeze-drying;




weight 3.5 g or 7.0% of the original holocellulose.  They were labeled




"Hemicellulose A" (55).




2.  Extraction with 1.0% Aqueous Sodium Hydroxide Solution




     Residue A (41.4 g dry weight) was extracted with a 1% aqueous sodium




hydroxide solution by adding it to 990.0 g of a 9.97% aqueous sodium hydroxide




solution to give a 4.0% slurry consistency based on the content of Residue A.




The mixture was stirred for 20 min at 25°.  The solids were recovered by




filtration using milk filter disks.  The residue was washed with 250 ml of




1% sodium hydroxide solution followed by 1 liter of distilled water.  The




solids were transferred with water to a dialysis bag and the mixture was




dialyzed for 3 days against running water.  The solids which remained in the




dialysis bag were recovered by freeze-drying and weighed; weight 39.1 g or




78.2% of the original holocellulose.  This residue was labeled "Residue B"




(55).




     The filtrate and combined washings were condensed to 1.0 liters and




acidified with acetic acid to pH 5.0.  The solids were precipitated by the




addition of 3 liters of methanol, and then recovered by centrifuge.  The




precipitate was redispersed in 700 ml of 70% methanol.  This redispersion




purification procedure was repeated two more times.  The methanol was removed




from the final precipitate on a rotary evaporator.  The solids were slurried




in water and the mixture was freeze-dried.  The weight of solids recovered

-------
                                   -60-
was 0.84 g or 1.7% of the original holocellulose.  These solids were labeled


"Hemicellulose B".


3.  Extraction with 15.0% Aqueous Sodium Hydroxide Solution


     To "Residue B" (38.6 g dry weight) was added 923 g of a 15.05% aqueous


sodium hydroxide solution to give a 4.0% slurry consistency based on the


content of Residue B and a solution concentration of 15.0% in sodium hydroxide.


After stirring for 20 min at 25°, the insolubles were recovered by filtration


using milk filter disks.  The residue was washed with 250 ml of 15.0%


aqueous sodium hydroxide solution and 1.0 liters of distilled water.  The


residue was dialyzed for 5 days against running water and freeze-dried;


weight 31 o 3 or 62.6% of the original holocellulose.  The residue was labeled


"Residue C"  (55).


     The filtrate and washings were combined, acidified with acetic acid,


condensed to  1.0  liters and the solids were precipitated by the addition of


3  liters of methanol.  The precipitate was redissolved in alkali and


redispersed  three times in 700 ml of 70% methanol.  The methanol was removed


on a rotary  evapofator and the precipitate freeze-dried; weight 1.46 g or


2.9% of the  starting holocellulose.  These solids were labeled  "Hemicellulose

                                                                 23 5
C".  The rotation was  taken in 1 N sodium hydroxide solution,  [ot]   '  - 40°


(c_ 0.25, N sodium hydroxide).


K.  Characterization of_ Hemicellulose A; Isolation of a_ Xylan


1.  Qualitative Carbohydrate Analysis by Paper Chromatography


     Hemicellulose A  (262.2 mg dry weight) was dissolved  in 4.5 g  of 77%


sulfuric acid at  ice-bath temperature  in a 150.0 ml round-bottomed three-


necked  flask.  The mixture was stirred until  complete, dissolution  of


Hemicellulose A was  achieved  (1.0 hr).  The translucent sirup was  diluted


to 3.9% sulfuric  acid  by  the addition  of water  (84.42 ml).  The dilute

-------
                                   -61-
solution was refluxed, neutralized, and paper chromatographed as described



above for holocellulose hydrolysis.  Paper chromatograms of  the hemicellulose



A hydrolyzate showed large amounts of xylose, minor amounts  of glucose,



galactose, and arabinose and trace amounts of mannose.  Hemicellulose A  was



thus considered to be a "xylan".



     The isolation procedure outlined in section J-l was repeated and a



large amount of the crude xylan was isolated.



2.  Ash Determination of the Crude Xylan



     The ash content of the crude xylan was determined according to a



modified TAPPI standard method (56).



     Duplicate samples (0.2 g dry weight) were weighed into previously ignited
                                                                            , a


and weighed porcelain crucibles.  The crucibles were placed in a muffle



furnace and heated gently (200°) until charring occurred.  The temperature



was increased to 600° and heating continued until ashing was complete.  The



ashed samples were placed in a dessicator to cool and were weighed after 1.0



hr.  The samples were then reignited and reweighed until constant weight was



obtained.  The ash content was 27.9%.



3.  Precipitation of Excess Barium and Dialysis of the Crude Xylan



     For purposes of calculation, it was assumed that the ash content of the



crude xylan was due entirely to the presence of barium as barium oxide.   From



this value, the percent barium (25.0%) present in the sample was calculated.



The large barium residue was due to the use of barium hydroxide in the



isolation of the crude xylan.



     A part (5.06 g) of the crude xylan was dissolved in distilled water



(100.0 ml).  A slight excess of 2 M sulfurlc acid (5.0 ml) was added to



precipitate all of the barium in the sample.  The precipitate was allowed



Lo settle, recovered by filtration and washed several times with hot distilled

-------
                                   -62-
water (51, p. 134).  The washings and the filtrate were recotnbined, placed



in a dialysis bag and dialysed against tap water for 5 days.  The material



left in the dialysis bag was concentrated on a rotary evaporator and



freeze-dried (yield,69.8%) .  A portion of the freeze-dried material was



ignited to determine its ash content (4.4%).  This purified, freeze-dried



material was labeled "xylan".



4.  Optical Rotation of the Xylan



     A part  (0.18235 g) of the xylan was dissolved in water (10.0 ml).  Ten



different readings were taken on the polarimeter (Rudolf and Sons) at 25°.



The specific rotation was calculated according to the relationship (57, p.



415):


          r  ,25   IQOct   IQOa

          la]D '" TC"c~~





              25
     where  [a]n  = specific rotation



               a = optical rotation in degrees



               b = cell length in dm = 1



               c = concentration in g/100 ml



     The  specific rotation was -30.5,  U 1.74 g/100 ml, water)  (average of 10



readings) .



5.  Qualitative Uronlc Acid Analysis by Color Reaction



     Concentrated sulfurlc acid  (6.0 ml) was added slowly to an aqueous



solution  (1.0 ml) of the 3.0% sulfuric acid hydrolyzate of  xylan and  cooled



under tap water.  The reaction mixture was heated for 20 minutes in a boiling



water bath and cooled.  An aliquot  (0.2 ml) of a 0.1% ethanolic solution  of



carbazole was added and the test sample allowed to stand for 2 hr  at  room



temperature.  The development of a purple color indicated that  the xylan



contained a  uronlc acid.  The solution was scanned in the visible  region

-------
                                   -63-
and showed maximum absorption at 535 ran indicating the presence of a uronic




acid (38, p. 497).  Color tests were also run on authentic galacturonic acid




(positive test), oxalic acid (negative test), and benzoic acid (negative test).




6.  Paper Chromatography and Gas-Liquid Chromatography of the Xylan Hydrolyzate




     A part (0.30 g) of the dialyzed xylan was hydrolyzed in 3.0% sulfuric




acid (mild acid hydrolysis) for 5 hours, cooled to room temperature,




neutralized to pH 5*0 with saturated aqueous barium hydroxide and centrifuged




to remove the barium sulfate precipitate.  An aliquot of the supernatant




liquor was concentrated under vacuum in a rotary evaporator.  The concentrated




liquor was spotted on Whatman No. 1 filter paper.  A solution of known sugars




was also spotted on the same paper.  The paper chromatogram was developed




and sprayed as described previously.  The sugars detected were xylose (strong),




arabinose (weak), galactose (trace), glucose (trace), and rhamnose  (trace).




     Another aliquot (25.0 ml) of the hydrolyzate was reduced with sodium




borohydride and alditol acetates were prepared as described previously.




The alditol acetates were injected into a Hewlett Packard Gas Chromatograph




using the same column and conditions as described previously.  The sugars




detected were the same as those reported above for paper chromatography.




The areas under the peaks were determined by planimetry.




7.  Reducing End-Group Analysis (Somogyi Method)




     A Somogyi determination of the reducing end group was carried out on a




portion of the dialyzed xylan.  The alkaline copper reagent (1945) was




prepared aa follows:  Rochelle salt (40.0 g), disodiuro hydrogen phosphate




dodecahydrate (71.0 g) (or 53.0 g of the heptahydrate), and 1 N sodium




hydroxide (100.0 ml).  A solution containing cupric sulfate pentahydrate




(8.0 g) was added with stirring followed by a solution of potassium iodate




(0.372 g in 100 ml water).  This amount of potassium iodate is good for

-------
                                   -64-
determining up to 1.25 rag glucose.  Finally, anhydrous sodium sulfate (180.0




g) was dissolved; the solution was diluted to 1.0 liter and allowed to stand




3 days so that the Impurities would settle out.  The clear supernatant solution




was further clarified by filtration through a fritted glass filter.  The pH




of the solution was about 9.5 as reported in the directions for preparation.




The solution is supposed to be stable for at least one year (52, p. 383).




     A sample 8.40 mg of the unhydrolyzed xylan in solution (5.0 ml) was




placed in a 25 x 200 iron test tube.  Alkaline copper reagent (5.0 ml) was




added by pipette and mixed thoroughly.  The tube was closed with a glass




bulb and placed in a rack.  Blanks were prepared with water (5.0 ml).




Standard xylose solutions (1.25 mg, 0.75 mg and 0.25 mg) were also added to




aliquots (5.0 ml) of the alkaline copper reagent.  The rack of tubes was




immersed in a vigorously boiling water bath to a depth of about 5.0 cm above




the solution inside the tubes.  The solutions were heated for 30 minutes,




removed from the bath and allowed to cool.  Care was taken that the tubes




were not agitated during the heating or cooling periods.  Triplicate




determinations were carried out in all cases.




     Potassium iodide (2.5%, 2.0 ml) was added to each tube without mixing.




From a fast flowing buret, 2 N sulfuric acid (1.5 ml) was run into each  tube




with shaking so that the liberated iodine would oxidize all reduced copper.




After 5 minutes, the tubes were reshaken.  The excess of  liberated iodine




not reduced by cuprous ions was then titrated with sodium thiosulfate




(0.005 N).  When the solution was light yellow, two drops of starch indicator




(1.0%) and two drops of phenol red indicator were added, and the titration




continued until the starch-iodine blue disappeared.  The difference in the




amount of titer consumed by the blank and the xylan was  attributed to the




reducing end group of the polysaccharide.

-------
                                   -65-
8.  Viscosity Measurements




     A part (0.19156) of the dialyzed xylan was weighed into a stoppered




volumetric flask (25.0 ml) which had been previously swept free of dry air




by a stream of nitrogen.  Diethylenediamine copper II ion (cupriethylenediamine)




(General Chemical Division, Allied Chemical, Columbia Road and Park Ave.,




Morristown, NJ or Ecusta Paper Division, Olin, P.O. Box 200, Pisgah Forest,




NC) at 25° was added and the solution was shaken until the xylan was completely




dissolved.  The flask was filled to the mark with the solvent and mixed




thoroughly.  The filled flask was carefully weighed to determine the density




of the solution in g/25 ml.  An aliquot (10.0 ml) of the solution was




transferred to a Cannon-Ubbelhode dilution viscometer previously placed in




a water bath at 25°±0.01° and flushed with nitrogen.  After 5 minutes the




solution was drawn into the bulb of the viscometer by applying pressure with




nitrogen.  The time for the miniscus to pass between the two calibration




marks was measured to 0.1 seconds.  Duplicate measurements were made until




they agreed to within ±0.3% (58, p. 537, 59, 60).




     The solution v»as diluted directly in the viscosimeter with a suitable




amount of solvent and mixed by stirring with a stream of nitrogen.




Measurements were taken with each dilution.  A total of six concentrations




were measured.  The viscosity of the pure solvent was also measured.  The




intrinsic viscosity was found to be 0.42 dl/g in diethylenediamine copper II




ion.




     The intrinsic viscosity was similarly determined in M aqueous sodium




chloride (0.48 dl/g) and in distilled water (0.84 dl/g).




9.  Gel Permeation Chromatography - Determination of Molecular Weight




    Distribution




     A chromaflex column (Kontes Glass Co., Vineland, NJ) 400 x 1000 mm with

-------
a total bed volume (V ) of 1250.0 ml was used for the determination of the



molecular weight distribution of the xylan.



     Dry sephadex H-75 (40-120 \i) v/as stirred into a 4- liter beaker with



distilled water and swollen for 30 hours.  During this time the suspension



was decanted four times to remove the fine particles.  The suspension was



well stirred and was transferred through a funnel into the vertically



mounted column which had previously been filled with water.  After a layer



of a few centimeters had formed, the capillary outlet at the bottom of the



tube was opened to release a slow stream of water.  When all of the gel had



settled, the eluant was allowed to flow through the bed overnight to complete



the stabilization of the column (40) .  The void volume (V ) of 369 ml was
                                                         o
determined by use of Blue Dextran 2000 (V^ 2,000,000).



     A part  (500 mg) of the dialyzed xylan was dissolved in water in a 50.0



ml volumetric flask.  After the xylan had completely dissolved the solution



was diluted  to the  50 ml mark and mixed thoroughly.  The xylan solution



(20.0 ml) was added dropwise to the top of the column with great care and



when it had  disappeared below the surface a small quantity of water was



applied to wash the surface.  Then more water was added to a height of 2-3



era to start  the elution.  The elution was carried out under a low hydrostatic



pressure at  a flow  rate of 0.5-0.7 ml/min.  The concentration of the eluted



fractions  (10,0 ml  each) vas estimated by the phenol-sulfuric acid method



to be described later.



10.  Acidity of the Xylan



     Three samples  (49.90 mg, 81,76 mg, 133.43 mg) of the xylan were each



dissolved  in separate aliquots  (10.0 ml each) of distilled water.  An



aliquot (10.0 ml) of aqueous sodium hydroxide solution (0.0526 N) was added



to each of the three samples.   Each solution was back titrated to a methyl

-------
                                   -67-
red end point with hydrochloric acid solution (0.0406 N).   Three blank




solutions were similarly titrated.  The differences between the titers of




the blanks and the titers of the xylan samples in the order of the weights




given above were: 0.15 ml, 0.25 ml, 0.35 ml.




11.  Analysis of Periodate Consumed




     A portion (0.26959 g) of the dialyzed xylan was placed in a volumetric




flask (50.0 ml), dissolved in water (10.0 ml), after which, 0.5 M sodium




periodate solution (25.0 ml) was added and finally water was added to bring




the volume up to the 50.0 ml mark.  The solution was mixed thoroughly.  A




blank having the same periodate concentration was also prepared.  Both




reaction mixtures were incubated in a constant temperature bath at 5° in




the dark (61).  The time at which the periodate was added was taken as time




zero.  Allquots (1.0 ml) of each reaction mixture were withdrawn at timed




intervals and added to 250.0 ml iodine flasks containing 1 N sulfuric acid




(5.0 ml) and 20.0% potassium iodide solution  (5.0 ml).  The liberated iodine




was titrated with 0.1 N sodium thiosulfate solution until only a pale yellow




color remained.  Starch indicator (1.0 ml, 1% solution) was added and the




titration continued until the blue color disappeared.




12.  Analysis of Formic Acid Released




     When the consumption of periodate had stopped, aliquots  (5.0 ml) of




each reaction mixture were added to Florence flasks (150.0 ml) containing




ethylene glycol (1.0 ml).  The mixture was allowed to stand at room temperature




for 5 minutes with occasional swirling until the unreacted periodate was




consumed.  The solution was titrated with 0.05 N sodium hydroxide solution




using methyl red as the indicator.




13.  Complete hydrolysis of the Polyalcohol




     The oxidized xylan from the periodate reaction was placed in a dialysis

-------
                                   -68-
bag and dialyzed against running tap water for two days.  At the end of




this period, it was taken out and concentrated under vacuum in a rotary




evaporator to about 100.0 ml.  The concentrate was neutral to litmus




paper.




     An excess of sodium borohydride (1.5 g) was added to the concentrate




and the reduction was allowed to proceed at room temperature for 24 hours




with occasional shaking.  At the end of the reduction period the reaction




mixture was returned to the dialysis bag and dialyzed apainst running tap




water for 3 additional days (100).




     The polyalcohol was hydrolyzed under mild acid hydrolysis conditions as




described previously.  The hydrolyzate was spotted on Whatman No. 1 filter




paper and paper chrotnatographed.  The sugars in another aliquot (25.0 ml) of




the hydrolyzate were converted to their alditol acetates and analyzed by




gas-liquid chromatography as described previously.  The spectrum showed




xylitol pentaacetate and arebinitol pentaacetate and three small unidentified




peaks which came out early.  The areas under the peaks were measured with a




planimeter.




14.  Degradation of the Xylan in Sodium Hydroxide Solution




     A part  (4.34 g) of the dialyzed xyIan was placed in a volumetric flask




(25.0 ml), which had previously been evacuated and flushed with nitrogen,




and dissolved  in sodium hydroxide solution  (1.0 N).  After the xylan was




completely dissolved the solution was diluted to the mark with sodium




hydroxide solution and shaken vigorously  to attain complete mixing.  The time




when the sodium hydroxide was added was recorded.




     Class ampules  (5.0 ml) were evacuated and filled with nitrogen In a




glove box.  Aliquots (1.0 ml each) of the xylan solution were placed in the




ampules.  The  ampules were taken out of the glove box, sealed, and placed

-------
                                   -69-
in boiling water (62, 63).  The ampules were withdrawn from the constant




temperature bath at definite time intervals (30, 60, 90, 120, 180 min and




so on).  Each withdrawn ampule was allowed to cool for 1 to 2 min and placed




In ice water to stop the reaction.  Normal hydrochloric acid (1.0 ml) was




added to the contents of the ampule to neutralize the sodium hydroxide




reaction mixture.




     The degradation was also carried out in sodium hydroxide solutions of




0.001 N, 0.01 N, 0.1 N, 4 N and 6 N concentration.




15.  Phenol-Sulfuric Acid Method of Analysis of the Xylan




     The analysis for the undegraded xylan was carried out by the phenol-




sulfur ic acid method of Smith and coworkers (64, 65).  A Beckman Model DB




spectrophotometer was used for this investigation.




     Reagent grade sulfuric acid (95.5%, specific gravity 1.84) conforming




to ACS specifications was used.  Phenol (80% by weight) was prepared by




adding glass distilled water (20.0 g) to redistilled reagent grade phenol




(80.0 g).  This mixture formed a water-white liquid which was readily




pipetted and has been known to stay the same color after one year of storage




(30).  A pale yellow color developed after several weeks but this did not




interfere in the determination since a blank was included.




     A fast delivery pipette (5.0 ml) was used to deliver the concentrated




sulfuric acid.  The pipette was prepared by cutting a portion off the tip




of a standard 5.0 ml pipette.  A high maximum temperature was desired because




it increased the sensitivity of the reagent.




     Water (1.7 ml) and 80% phenol (0.05 ml) were added to an aliquot (0.3 ml)




of the neutral xylan solution in a test tube.  Concentrated sulfuric acid




(5.0 ml) was added rapidly.  The stream of acid was directed against the




liquid surface in order to obtain good mixing.  Intense heat was generated

-------
                                   -70-
at this time.  The tubes ware allowed to stand for 10 minutes, then shaken.




end placed in a water bath at 25 to 30° for another 10 to 15 rain.  The test




tubes were shaken again and samples were transferred to spectrophotometer




cells (1.0 cm light path) and readings taken at 480 nm.  The color developed




was stable for several hours.  The weight of hydrolyzed pentoses and uronic




acids was expressed as xylose equivalents by using percent transmission in




conjunction with a standard curve prepared from known xylose.




L.  Characterization of Hemicellulose Ji; Isolation of a_ Glucomannan




1.  Qualitative Carbohydrate Analysis by Paper Chromatography




     Hemicellulose B (250.8 mg, dry weight) was dissolved in 4.5 g of 77%




sulfuric aicd for 1.0 hr at ice-bath temperature.  The resulting translucent




solution was diluted to 3.9% sulfuric acid by the addition of 84.42 ml of




distilled water.  The dilute solution was refluxed, neutralized, and paper




chromatographed as described above for holocellulose hydrolysis.  Paper




chromatograms of the Hemicellulose B hydrolyzate showed large amounts of




glucose and mannosep minor amounts of galactose and xylose, and trace amounts




of arabinose.




M.  Characterization of Kami eel lulose C_; Isolation of a Mannan




1.  Qualitative Carbohydrate Analysis by Paper Chromatography




     Hemicellulose C (273.3 mg, dry weight) was hydrolyzed, neutralized, and




paper chromatographed as described above for Hemicelluloses A and B.  The




Hemicellulose C hydrolyzate contained large amounts of mannose, small amounts




of glucose and galactose, and  trace amounts of arabinose and xylose.




2.  Quantitative Carbohydrate  Analysis by Gas-Liquid Chromatography




     A portion of Hemicellulose C  (150 mg, air-dry weight, moisture content




8.9%) was acid hydrolyzed, neutralized, reduced and acetylated as described




previously for the quantitative carbohydrate analysis of the holocellulose

-------
                                   -71-
fraction by gas-liquid chromatography.  An aliquot (20 mg) of the resulting




alditol acetate sirup was dissolved in methylene chloride and diluted to the




mark in a 1.0 ml volumetric flask.  An aliquot (2 pi) of the solution was




injected into the gas chromatograph under the general conditions previously




established.  The areas under the peaks of the resulting spectrum were




measured by a planimeter.




3.  Methylation of the Mannan




     A portion of Hemicellulose C (1.4 g, air-dry weight, moisture content




8.9%) was dissolved in 182 aqueous sodium hydroxide solution (50.0 ml) at




ice-bath temperature with stirring.  The solution was methylated by the




simultaneous addition of 30% aqueous sodium hydroxide solution (100.0 ml)




and dimethyl sulfate (50.0 ml) over a period of 3 hr while maintaining the




solution at ice-bath temperature.  Acetone (100.0 ml) was added to prevent




foaming.  The mixture was stirred for an additional 45 hr at room temperature.




The solution was cooled to 0° in an ice-bath, neutralized with 10% sulfuric




acid, and dialyzed for 3 days against running tap water.  The solution was




concentrated on a rotary evaporator, freeze-dried and the entire methylation




sequence was repeated (41).




     The product was further methylated by dissolution in tetrahydrofuran




(150.0 ml) followed by treatment over a period of 60 hr with seven 10.0 g




portions of crushed sodium hydroxide, each followed by a portion (12.0 ml)




of dimethyl sulfate (42).  The product was dialyzed against running tap water




for 3 days.  The concentrated dialyzate was extracted with three 300.0-ml




portions of chloroform.   The chloroform extract was concentrated on a rotary




evaporator, dissolved in acetone and filtered.  The concentrated filtrate




(50.0 ml) was poured into petroleum ether (200.0 ml), and the white




precipitate which resulted was recovered and freeze-dried; yield of a white

-------
                                   -72-
powder 40.7 mg.  An infrared spectrum (potassium bromide pellet) of a portion




of Che methylated Hemicellulose C shoved no absorption in the regions of




3200 cm   or 1750 cm   indicating that there were no free hydroxyl groups in




the material.  Therefore, methylation was considered complete.




4.  Hydrolysis of the Methylated Mannan




     The methylated mannan  (40.7 mg) was hydrolyzed with 10.0 ml of 90% formic




acid at 97° for 3 hr.  The  formic acid was removed by evaporation under




reduced pressure followed by the addition and removal of water.  The sirup




so obtained was hydrolyzed  with 0.5 N sulfuric acid (5.0 ml, for 2.5 hr at




97°).  Upon cooling, the solution was neutralized with barium hydroxide, and




the solids were removed by  centrifugation.  The remaining inorganic salts




were removed by refluxing the concentrated sirup with absolute methanol,




followed by filtration (repeated four times) and removal of solvent (41).




     The sirup so obtained  was dissolved in a few drops of water and subjected




to paper chromatography using 2-butanone saturated with water as the developer




(67).  The chromatogram was allowed to run until the solvent front had moved




32 cm from the origin where the sample was applied (2.5 hr).  The developed




chromatogram was allowed to air-dry and then sprayed with aniline phthalate




indicator  (1.66 g of phthalic acid dissolved in 100.0 ml of water-saturated




n-butanol containing 0.93 g of freshly distilled aniline) (68, 69).  The




papers were allowed to air-dry again for 15 minp heated in an oven at 105°




for 15 min and viewed under ultraviolet light.  Three spots were evident




with the following R  values: 0.89  (trace), 0.51 (very strong), 0.24 (trace).




(R.. value is the ratio of the distance the spot moved divided by the distance




the solvent front moved, both measured from the point of application of the




material).




     A second aliquot of the hydrolyzate of the methylated mannan was subjected

-------
                                   -73-
to paper chromatography using n-butanbl-ethanol-water (5:1:4 v/v/v) as



developer.  The solvent front was allowed to migrate 40 cm past the origin



where the spot was applied (19 hr).  The papers were sprayed with aniline



phthalate indicator as described above.  Seven spots were evident which had



the following R  values (R  values determined by dividing the distance the
               o          o


sugars have moved from the starting line by the distance moved by 2, 3, 4,



6-tetra-O-methyl-D-glucopyranose): 0.96 (trace), 0.82 (very strong), 0.64



(trace), 0.52 (trace), 0.38 (trace), 0.28 (trace), 0.21 (trace).



5,  Molecular Weight Determination of the Mannan by Viscosity Measurements



     An amount (125 mg, dry weight) of the mannan (Hemicellulose C) was



weighed into a 25.0 ml volumetric flask which had been previously weighed



and swept free of air by a stream of nitrogen gas.  Diethylenediamine copper



II reagent at 25° was added and the mixture shaken.  However, not all of



the mannan dissolved.  The mixture was transferred to a 100 ml volumetric



flask and diluted with additional diethylenedlamlne copper II reagent.



However, the mannan still did not completely dissolve.



N.  Characterization of Residue C_\ Isolation of a_ Glucan



1.  Hydrolysis with 77.0% Sulfuric Acid



     A portion (200 mg, air-dry weight, moisture content 11.1%) of the residue



from the 15.0% aqueous sodium hydroxide extraction (Residue C) was hydrolyzed



with 77.0% sulfuric acid according to the procedure previously described.



     Paper chromatography of the hydrolyzate showed the presence of a large



amount of glucose, medium amounts of mannose and galactose, and trace



amounts of xylose and arabinose.



2.  Four Hour Formic Acid Extraction



     A portion (1.0 g, air-dry weight) of Residue C remaining after the 15.0%



aqueous sodium hydroxide extraction was refluxed (97°) with 100.0 ml of 95%

-------
                                   -74-
fonnic acid for k hr.  The insoluble residue was recovered by centrifuge,




washed with water and freeze-dried; weight 0.18 g.




     A portion (0.15 g) of the above residue was hydrolyzed with 77.0%




sulfuric acid and the hydrolyzate paper chroroatographed as previously




described.  The chromatograras showed a large amount of glucose, a small




amount of mannose, and trace amounts of arabinose and xylose.  Thus the




formic acid did not completely remove the non-glucose containing polysaccharides,




3.  Four Day Formic Acid Extraction




     A portion (10.0 g) of Residue C was extracted with 200.0 ml of 95%




formic acid for 4 days in a Soxhlet extractor using a porcelain extraction




thimble.  An amount  (25.0 ml) of additional formic acid was added each day.




At the end of 4 days the residue was recovered by washing with distilled




water, centrifuged, and freeze-dried; weight 3.6 g.




     An amount (150 mg, air-dry weight, moisture content 11.1%) of the above




residue was hydrolyzed with 77.0% sulfuric acid and the hydrolyzate was




paper chromatographed as previously described.  The paper chromatogram




showed glucose only.




4.  Quantitative Carbohydrate Analysis by Gas-Liquid Chromatography




     An amount (150 mg, air-dry weight, moisture content 11.1%) of the




residue from the formic acid extraction was hydrolyzed with 77.0% sulfuric




acid as previously described.  The hydrolyzate was reduced with sodium




borohydride „ and the alditols were acetylated and analyzed by gas-liquid




Chromatography as previously described.  The areas under the peaks of the




resulting spectra were measured with a planimeter.




5.  Acetolysis of the Glucan




     An amount (2.0  g air-dry weight, moisture content 11.1%) of the residue




from the formic acid extraction was kneaded into a mixture of acetic anhydride

-------
                                   -75-
(8.0 ml) and concentrated sulfuric acid (0.2 ml), at ice-bath temperature



(3°), until completely wetted.  The mixture was placed in an oven at 50°



for 14 days (70).  The resulting dark brown sirup was mixed with acetic



acid (10.0 ml).  The suspension was stirred with 500.0 ml of cold water



and an additional 250.0 ml of water was added.  The mixture was stirred



for 30 min and filtered through a sintered-glass funnel.  The residue was



washed free cf acid with water, air-dried, and extracted with boiling



ethanol (95%, 200.0 ml).



     The aqueous filtrate from the above filtration was extracted with



chloroform (250.0, 250,0, 100.0 ml).  The chloroform extract was washed



with 10.0 ml of a saturated aqueous solution of sodium carbonate and



concentrated on a rotary evaporator to yield a solid.  However, the material



resisted crystallization.



     Since no crystals resulted from the above acetolysis, a second acetolysis



was attempted simultaneously with the acetolysis of filter paper.  The



acetolysis of filter paper (2.0 g) resulted in crystals of cellobiose



octaacetate (1.56 g), m.p. 228-228.5°, literature value 223-224°.



     However, still no crystals resulted from the acetolysis of the residue



from the formic acid extraction.  Therefore, the reaction was repeated



but the reaction time was shortened from two weeks to one week.  Crystals



of cellobiose octaacetate (550 mg) were obtained.  The material was



recrystallized three times from 95% ethanol; yield 275 mg, m.p. 224-225°,


   23
[a]   + 40° (c. 3.0, chloroform); unchanged on admixture with authentic


                                        22
material, literature value 223-224°, [a]   +40° (£3.0, chloroform).



6.  Molecular Weight Determination of the Glucan by Viscosity Measurements



     An amount (100 mg, dry weight) of the residue from the formic acid



extraction was placed into a 25.0 ml stoppered volumetric flask which had

-------
                                   -76-
bsen previously weighed and swept free of air by a stream of nitrogen gas.




Dlethylensdiamine copper II reagent (cupric ethylenediatnine) (General




Cheaical Division, Allied Chemical, Columbia Road and Park Ave., Morristown,




NJ or Ecustic Paper Division,, Olin, P.O. Box 200, Pisgah Forest, NC) at




25° was addad and the solution was shaken until the glucan was completely




dissolved.  The flask was filled to the mark with the solvent and mixed




thoroughly.  The filled flask was carefully weighed to determine the density




of the solution in g/25 ml.  An aliquot (10.0 ml) of the solution was




transferred to a Cannon-Ubbalhode dilution viscometer previously placed in




a water bath at 25±0.01° and flushed with nitrogen.  After 5 min the solution




was drawn into the belt of the viscometer by applying pressure with nitrogen.




The pressure was released and the time required for the miniscus to pass




between the two calibration marks was measured to 0.1 seconds.  Triplicate




measurements were made until duplicates agreed to within ±0.3%  (58, p. 537,




59, 60).




     The solution was diluted directly in the viscometer with a suitable




amount of solvent and mixed by stirring with a stream of nitrogen.  Measurements




were taken with each dilution.  A total of six concentrations were measured.




The viscosity of the pure solvent was also measured.  The intrinsic viscosity




was found to be 0.432 dl/g in diethylenediamlne copper  II ion.




7.  Mathylation of the Glucan




     A portion  (1,4 g, air-dry weight, moisture content 11.1%)  of the residue




from the formic acid extraction was methylated by the procedure previously




outlined for the methylation of Hetnicellulose C; yield  605 mg.  An infrared




^spectrum (potassium bromide pellet) of a portion of the methylated material




showed no absorption in the regions of 3200 cm   or 1750 cm   indicating




that there were no free-hydroxyl groups.  Therefore, methylation was

-------
                                   -77-
considered complete.




8.  Hydrolysis of the Methylated Glucpn




     A.part of the methylated glucan (150 rag, dry weight) was dissolved in




2.25 p of 77.0% sulfuric acid at the ice-bath temperature for 1 hr and then




diluted with 42.2 ml of water to 3.9% sulfuric acid concentration.  The




solution was refluxed for 6 hr.  There remained considerable insoluble




materials in the solution which were removed by centrifugation.  The




decantate was neutralized with saturated aqueous barium hydroxide solution




to pH 5.0 and the precipitate was removed by centrifugation.  The resultant




hydrolyzate was condensed to a sirup on a rotary evaporator.




     The sirup so obtained was dissolved in a few drops of water and subjected




to paper chromatography using 2-butanone saturated with water as the developer




(67).  The sirup from the hydrolysis of the methylated tnannan was chromatographed




on the same sheet of paper.  The chromatogram was allowed to run until the




solvent front had moved 32 cm from the origin where the sample was applied




(2.5 hr).  The developed chromatogram was allowed to air-dry and then




sprayed with aniline phthalate indicator (1.66 g of phthalic acid dissolved




in 100.0 ml of water-saturated n-butanol containing 0.93 g of freshly




distilled aniline)  (68, 69).  The papers were allowed to air-dry again for




15 min, heated in an oven at 105° for 15 min and viewed under ultraviolet




light.  Four spots were evident from the hydrolyzed glucan with the following




Rf values: 0.82 (trace), 0.56 (very strong), 0.21 (trace), 0.06 (trace).




     A second aliquot of the hydrolyzate of the methylated glucan and of the




methylated tnannan were subjected to paper chromatography using ri-butanol-




ethanol-water (5:1:4 v/v/v) as developer.  The solvent front was allowed to




migrate 40 cm past the origin where the spot was applied (19 hr).  The papers




were sprayed with aniline phthalate indicator as described above.  Five spots

-------
                                   -78
were evident from the hydrolyzate of the methylated glucan with the following




R  values (Rp values determined by dividing the distance the sugars have




moved from the starting line by the distance moved by 2, 3, A, 6-tetra-O-.




methyl-D-glucopyranose): 1.0 (tr.ace), 0.83 (very strong), 0.67 (trace),




0.41 (trace), and 0.28 (trace).









Ill.  Results and Discussion




     A sample of inner bark free from outer bark and cambium was desired




because it provided a relatively homogeneous starting material.  Outer bark




contains considerable cork material which would interfere with experimental




studies of the carbohydrates.  The cambium layer contains proteins as well




as carbohydrates and the two are best separated at the outset.




     In the spring of the year from April through June, the outer bark of




Douglas-fir is easily separated from the inner bark by simply chipping it




away.  The inner bark for the present work was taken from a standing tree




to reduce contamination from other sources.  The cambium layer was then




carefully separated from the inner bark and a relatively homogeneous




sample resulted.




     The inner bark after the ethanol-water extraction was carefully ground,




screened and reground to ensure that most of the fibers were separated and




that the surface areas were exposed.  Some of the fine material was




included in the fraction to provide a large starting weight to ensure that




sufficient polysaccharides would be isolated for later studies.  All of




the  fines were not included because preliminary experiments showed that




they plugged the extraction apparatus and filter systems and made laboratory




procedures time-consuming and difficult.




     It is desirable, before proceeding with polysaccharide separation, to

-------
                                   -79-
rcmove ao much as possible of the low-molecular weight  materials (simple




sugars, organic acids, liplds, waxes, and so on) present in the bark.   Some




of these materials arc readily oxidized and would interfere with the




delipnlficatIon reaction.




     Figure 5  outlines the scheme followed in the treatment of the inner




bark and the isolation of the different fractions.  The yields for each




extraction are shown.  None of the extraction procedures shown in this




chart accomplished a clear-cut separation of one type of material from




another, but In general each step in the sequence performed a definite




function.




     The ethanol-water (4:1 v/v) extraction served two purposes.  It




prevented possible enzyme action which might change the natural materials




from their native state, and  it also solubilized the simple sugars which




could be identified  to provide a more detailed  investigation of the total




carbohydrate fraction in the  bark.  The ethanol-water  (A:1 v/v) soluble




solids were shown to contain  the free sugar, glucose,  by paper chromatography.




     The benzene-ethanol  (2:1 v/v) azeotrope removed lipids and viaxes which




would  Interfere with later reactions.  The extract contained no free sugars




when  Investigated by paper chromatography.  Having de-fatted the bark  it




was possible to carry out a  hot water extraction and remove a water-soluble




fraction which amounted  to 11.1% of  the  original bark  sample.




     Compounds  isolated  from natural  sources often contain one or more  of




the elements, nitrogen,  sulfur, phosphorus or  the halogens.  Therefore,  it




it> best  to  qualitatively determine  if  they arc present.




     The sodium  fusion  test  on  the water-soluble  fraction  failed  to detect




the presence of  sulfur,  phosphorus,  and  the halogens.   Nitrogen was shown




to be  present as  evidenced by the  formation of a precipitate of Prussian

-------
                                   -80-
blue with ferrous sulfatc.




     A portion of the sample was sent to an analytical laboratory (Pascher




and Pasoher, Mikroanalytisches Lahorntor iuri 53 Bonn, Buschstrasse 54, West




Germany) for analysis of nitrogen by the Kjeldahl method, and was found to




contain 3.632 nitrogen.  A part of this nitrogen is in the form of proteins




as discussed later.




     A ferric chloride-potassium ferrlcyanide test produced a dark-blue




color.  This color  (Turnbull's blue) was due to the formation of a 1:1




complex of  ferric  ion and pnenolic hydroxyl which give the ferrous reaction.




The iodine  test,  likewise resulted in a blue color characteristic of a




starch-iodine complex.




     These  strongly positive  color tests showed the presence of considerable




amounts of  tan.^ns  and  starch.  Kiefer and Kurth  (8)  indicated the possibility




of  phenolic materials  in Douglas-fir inner bark and starch has been  reported




 in  the hot-water-soluble  fractions from several Inner barks  (63).  Therefore,




starch was  expected to  be  present  in Douglas-fir  bark.




     A  fundamental  aspect  of  polysaccharides  is the component monosaccharides




which are  linked together  to  form  the  polymer chain.  .Jften  polysaccharides




contain  linkages which  are resistant to acid  cleavage.   Thus,  to  ensure




complete  hydrolysis, the  hot-water-soluble  solids were  treated with  strong




acid according  to the  ptocedure  of Laver, Root,  Shafizadeh and Lowe  (22).




A portion  of  the hot-water-soluble solids would not dissolve in  72.0%




 sulfurlc  acid.   After  hydrolysis,  the  insoluble  portion had  a  dark-red color




 and presumably  were phenolic  acids.  This  precipitate was not  investigated.




 The hydrolyzatc was analyzed  for monosaccharidcvs  as discussed  below.




      If  the polysaccharides under Investigation  arc water soluble and  if




 hydrolysis can  be accomplished under mild  acid  conditions,  treatment with

-------
                                   -81-
strong acids as discussed above is not desirable because of deprodative




side reactions.  Since the hot-water-extracted solids were water soluble,




a mild ncid hydrolysis (3.0% sulfuric acid) was compared to the 72.0%




suit uric acid hydrolysis.  The hydrolyzate was analyzed for amino acids




and tnonosaccharides as discussed belov.




     As was the case in the strong acid hydrolysis, the insoluble portion




of  the extract had a reddish appearanc .  This precipitate was found to




persist later even after purification of the water-soluble solids.




     Quantitative analysis showed that the hot-water-soluble solids




contained  3.63% nitrogen.  Bark has been known to contain both protein and




non-protein nitrogen, as well as free atnino acids.  At least 5 water-soluble




proteins have been detected.  The cambium  layer of Douglas-fir bark has




also been  shown to contain protein  (17), and possibly some cambium layer




may have remained with the inner bark during the sairple preparation.




     The hydrolyzate  from the mild  acid hydrolysis  (3.02 sulfuric acid) of




the hot.-water-soluble solids was examined  for amino acids by two-dimensional




paper  chromatography.




     The chromatograns showed several purple spots but the spots were not




well  resolved.  Lai  (62)  found  that  3.0Z sulfuric acid hydrolysis did not




appear to  he  strong enough to 'lydrolyze the proteins  into  their component




amino  ac.'ds.   The  strong acid  treatment  (72.02 sulfuric acid) better




hydrolyzed the proteins  into  their  component amino  acids.  The evidence  in




 the present  woric  shows that proteins  were  extracted by  the hot-water  treatment,




      Paper chromatography of  the original  extract failed  to detect free




mononaccharidos.   Therefore,  the carbohydrates present  in  the hot-water-




 soluble  fraction  are  considered to  be in the  form of  polysaccharides.




      The mixture  of munosaccharide  sugars  resulting iron  acid hydrolysis

-------
                                   -82-
of the hot-water-soluble solids was well resolved by paper chromatography.




The chromatograms showed a very large amount of glucose, a moderate amount




of arabinose, a slight amount of galactose, trace amounts of xylose and




rhamnose and almost no mannose.




     The hydrolyzates from both the strong acid hydrolysis and mild acid




hydrolysis contained what appeared to be identical monosaccharides by paper




chromatography.  Therefore,  strong acid hydrolysis was not necessary, and




to avoid degradation, the 3.0% sulfurii- acid method was used in all




subsequent hydrolyses.




     The five major sugars,  glucose, mannose, galactose, arabinose and




xylose are those ordinarily  found  in wood and wood pulp (20).  Rhamnose in




Douglas-fir  bark was  first  reported oy  Lai  (62)  in the acidified acid




hydrolyzate  of  the  sodium chlorite-soluble  fraction.  Its  identification




in the acid  hydrolyzate  of  the hot-water-soluble solids supports its




finding  in  the  polysaccharides cf  Douglas-fir  inner hark.




      Figure  6   outlines  the scheme followed for  the purification of  the




polysaccharides extracted by hot water.  When  the freeze-dried sample was




redissolved  in  distilled water at  room temperature, it was found  that  29.8%




of  the sample would not  dissolve.   The insoluble portion  (Fraction A)  gave




a strong positive  reaction  to the  iodine test  shoving the  presence of




starch and  a negative test  to the  ferric chloride-potassium fcrrfcyanidc




solution showing the absence of  phenolics.   Strong acid  hydrolysis and




paper  chromatography indicated  the presence of  glucose  only.  No  traces




of  other monosacchjrides could  be  detected. Microscopic  examination of




this sample  showed  some  fiber debris and irregularly  shaped materials




which were  presumably starch granules.  Since  only glucose was detected




and  the  most obvious sources of  this munosaccharide were  starch and

-------
                                   -83-
cellulose, this fraction was not investigated further.




     Ethanol was added to the water-soluble portion to provide a solution




of 70.0% In ethanol.  A precipitate formed which amounted to 25.9% of the




original water-soluble solids.  These precipitated solids were labeled




Fraction B (Figure 6).  Those solids which remained soluble in the 70.0%




ethanol solution (44.3% of the original water-soluble solids) were labeled




Fraction C (Figure 6).  Fraction C was found to react strongly with ferric




chloride-potassium ferricyanide indicating the presence of phenolics but




gave no reaction with iodine indicating the absence of starch.  This




fraction was not studied further.  Polysaccharides which may be in this




fraction would be expected to possess a very low molecular weight.




     The fraction insoluble in the 70.0% ethanol-water mixture (Fraction B)




gave a positive reaction with iodine indicating the presence of starch.  It




also gave a slightly positive reaction with ferric chloride-potassium




ferricyanide, thus indicating that .separation of the polysaccharides




from the phenolics was not complete.




     Hydrolysis of Fraction B and two-dimensional paper chromatography of




the hydrolyzate showed the presence of amino acids.  Six major spots,




seven light spots, five very light spots and two large smears showed on




the paper chromatogram.  The identification of these amino acids was not




investigated further because the main objective of this study was the




carbohydrates.  However, the paper chromatogram definitely showed the




presence of protein in the inner bark of Douglas-fir.




     Paper chromatography of the monosaccharides showed the same relative




amounts of sugars as found in the whole hot-water-soluble fraction.




     Fraction B (Figure 6) was treated with ot-amylase enzymes to remove




starch.  At the end of the hydrolysis the reaction mixture was dialyzed

-------
                                   -84-
(Figure 6).  The dialyzate showed the presence of glucose only, both before




and after hydrolysis with acid.  No other tnonosaccharides were detected.




     The enzymea used in this Investigation were u-amylases and therefore




specific for the hydrolysis of the (l->4)-0-a-D-glucopyranosyl-type bond.  The




fact that glucose was released proves the presence of an a-D-(l-»-4)-glucan,




undoubtedly the amylose portion of starch.  This represents the first time




that starch has been shown by chemical analysis to be present in the inner




bark of Douglas-fir.  The starch undoubtedly acts as a food reserve for




the living cells in the inner bark.




     The non-dialyzables froin the enzyme hydrolysis showed a blue color




with iodine indicator.  Thus the d-amylases did not hydrolyze all of the




starch.  This was undoubtedly due to the presence of some amylopectin in




the starch molecule.  Amylopectin possesses some cx-D-(l-*6)-glucan branch




chains which are not cleaved by a-amylases.  The residue was, therefore,




expected to contain sone residual starch molecules.




     The original hot-water-soluble solids were found to have a nitrogen




content of 3.63%.  The protein content of natural products is generally




calculated as the product of the nitrogen percentage and the factor 6.25,




which  is the usual factor used for conversion of nitrogen content to




protein content.  In accordance with this calculation the hot-water-soluble




solids of  the inner bark of Douglas-fir contained 21.8% protein.  This




appeared to be a very high protein content, but when it was corrected to




the original bark weight (Figure 5) the value was 2.42% protein in  the  inner




bark of Douglas-fir.  Values for bark proteins which have been reported




are 5.0% for the inner bark of birch and an average of 21.6% for the inner




bark of black locust (58, p. 625).  Therefore, in comparison, Douglas-fir




bark appears quite low in protein content.

-------
                                   -85-
     The quantitative nitrogen content (Kjeldahl) of Fraction D (Figure 6)


was 1.57%.  This proved to be a reversal in nitropen content.  The original


water-soluble solids contairied 3.63% nitrogen as previously mentioned.


Kxtraction with water at room temperature and precipitation with 70% ethanol


provided Fraction B (Figure 6) which contained 0.28% nitrogen.  Apparently


in the re-precipitation of the water-soluble solids, a large amount of the


protein remained in solution with the phenolic compounds.  In an effort to


obtain purer polysaccharide samples, and in particular to remove the starch,


Fraction B was treated with starch hydrolyzing enzymes (HT-1000 and Diazyme


T.30) and with protein hydrolyzing enzymes (chymotrypsin and trypsin;


Figure 6).  The nitrogen content of the final product (Fraction D) increased


co 1.57%.  It would appear from this increase in nitrogen content that not


all of the enzymes were removed by dialysis.  Therefore, the net result of


the treatments was to introduce protein impurities into the sample.


     Fraction D was hydrolyzed with 3.0% sulfuric acid and paper chromato-


graphed.  The paper chromatogram showed a large reduction in the amount


of glucose present, relative to the amount of the other monosaccharides.


f.lucose was still the major monosaccharide present, although the spots for


arabinose and galactose were now more prominent relative to the glucose


spot.  A part of the glucose residues may have come from residual starch


that was not destroyed by the enzyme.  It was thought, however, that


another treatment with the starch enzyme would not completely remove the

                                         l
starch and that it might even prove detrimental to the other polysaccharides


that were present.  Starch enzyme treatment was carried out at a pH of 3.8-


4.2 and at this slightly acidic condition, there was a distinct possibility


that acid hydrolysis might destroy the polysaccharides that were the object


of the purification process.

-------
                                   -86-
     Carbohydrates are neither heat resistant nor volatile and so derivatives




must be prepared which will volatilize without degradation in order to




perform gas-liquid chromatographic analysis.  There are numerous derivatives




which have been tried but the ones most commonly used today are the "alditol




acetates."  In the preparation of the alditol acetates the monosaccharides




are first treated with sodium borohydride to reduce the aldehyde function




to the alcohol function.  This has the result of preventing ring isomerization




of the pyranose and furanose forms and so prevents the formation of alpha




and beta forms of the sugars.  The end result is only one form, the alditol,




for each of the monosaccharide sugars, rather than three, four or even five




isomers.  The acetates are synthesized from the alditols to make a volatile,




heat resistant derivative.




     The gas chromatographic resolutions of the alditol acetates prepared




from Fractions B and D  (Figure 6) were excellent.  The areas under the peaks




of the different sugars were compared to each other to give relative weight




values.  The areas under the peaks of the other sugars were compared to the




area of the derivative  for galactose which was taken as 1.0.  For Fraction




B  (Figure 6) the areas under the curves were:  galactitol hexaacetate, 1.0;




arabinitol pentaacetate, 1.4; and glucitol hexaacetate, 32.A.  Only traces




of rhamnitol pentaacetate  (0.1), xylitol pentaacetate and mannitol hexaacetate




were found.  For Fraction D  (Figure 6) the areas were galactitol hexaacetate,




1.0; arabinitol pentaacetate, 1.3; glucitol hexaacetate, 2.9; and traces of




rhamnitol pentaacetate, xylitol pentaacetate and mannitol hexaacetate.




These relative areas are in agreement with the relative area and intensity




of color developed by the spots on the paper chromatograms of both samples.




     The starch enzyme treatment resulted in a decrease of glucitol




lu-xaacetate from 32.4 units to 2.9 units when compared with galactitol

-------
                                   -87-
hexaacetate.  The ratio  (1.4:1.0) of  arabinose  to galactose  In Fraction  B




is  the  same as  in Fraction D  (1.3:1.0), allowing for errors  in measurements.




The  results show that  the starch enzyme was  effective  in  reducing  the




amount  of  starch in  the  sample.  The  almost  constant ratio of arabinose  to




galactose  in Fraction  B  and Fraction  D confirms what the  paper chromatograms




have already shown,  that the  starch enzymes  acted'to hydrolyze the  starch




only.   All other polysaccharides were left intact by the  starch  enzyme




treatment.  Therefore, it was concluded that the major carbohydrate component




of  the  hot-water-extract was  starch.




     The arabinose and galactose monosaccharides released by acid  hydrolysis




undoubtedly exist in the inner bark as part  of  L-arabino- U-galactans which




are  known  to be present  in the wood of conifers.  These polysaccharides  are




the  only major wood  glycans that can  be isolated in good  yield by  extraction




of wood with water before delignification.   Their ease of extraction and




their useful qualities as gums have brought  them into  commercial production




marketed as the commercial gum, Stractan.




     Since the L-arabino-D-galactans  are common to the wood in conifers,




it  is considered possible that the arabinose  and galactose containing




polysaccharides isolated in the present work  are similar.  This represents




the  first  time that possible  L-arabino-D-galactan polysaccharides have been




reported In Douglas-fir bark.  Because of the large volumes of Douglas-fir




bark which are waste products, it is  possible that the bark could become




of interest as a raw-material  source  for these  polymers since they  have




aJready been marketed as commercial gums.




     Not all of the pectic material of plants is extractable with water.




The insoluble part,  protopectin (calcium pectate)  was removed from  the




water-insoluble residue of Douglas-fir inner bark  by extraction with a

-------
                                   -88-
0.5% ammonium oxalate solution (Figure 5).  Presumably a cation exchange




occurred in which insoluble calcium oxalate was formed along with soluble




ammonium pectate.  The latter was extracted with the filtrate.




     Through all of the extractions the bark retained its characteristic




brown color.  However, the strong oxidizing conditions of the acidified




sodium chlorite reaction bleached the reaction mixture to a pale yellow




color.  Fumes of yellow gases, undoubtedly chlorine and chlorine dioxide,




were visible in the reaction vessel above the mixture.  These gases




emphasize the necessity of performing the delignification reaction in a




fume hood and of bubbling nitrogen through the reaction to sweep away




these toxic materials.  Fresh sodium chlorite was added to the reaction




at one hour intervals rather than all of it at the beginning to prevent




an excess of gas formation and to give the reaction time to proceed.  The




residue from the delignification, termed "holocellulose" was a pale yellow




color.




     The filtrate was yellowish in color and was dialyzed to remove




low-molecular weight impurities and the inorganic salts and ions which




resulted from the reaction of the sodium chlorite.  The solution which




remained in the dialysis bag after several days was almost colorless and




the solids which were recovered by lyophilization were white.




     The delignification reaction has been developed to solubilize the




non-carbohydrate components of plants.  Thus the insoluble residue




(holocellulose) remaining is relatively pure carbohydrate material.  However,




the reaction does dissolve some carbohydrates and it is these materials




that are part of the subject of this study.




     The solubilized solids of the present investigation represented a




complex mixture of carbohydrates, lignin-like compounds, tannins, proteins,

-------
                                   -89-
and so forth.  Therefore, as described below, a variety o> approaches were




used to characterize the carbohydrates Ln the mixture.




     The ash content was 13.39'].57%, as Hul.fati>.  Although a r.onsidvrabl.e




amount of inorganic sodium chlorite had been added to the delignification




reaction, this was a high ash content considering that the fraction had




been dialyzed.  Although it would be of interest to know what elements




were present in the ash, no effort was made to determine them in the present




work.




     In an effort to lower the ash content and remove more of the low-




molecular weight materials, the acidified sodium chlorite soluble materials




were again dialyzed.  More than 22.4% passed through the dialysis membrane.




Paper chromatography of the dialyzate showed a relatively large  spot for




glucose, indicating that carbohydrate substances, probably monosaccharides




and short chain oligosaccharides, had passed through the dialysis membrane.




     The solution retained in the dialysis bag was made to 70% ethanol, a




method often employed for the purification of polysaccharides (15, p.  449),




and resulted in the precipitation of a white, flocculent material.




     However, the sulfated ash content of the precipitate was 11.88+2.83%




indicating that dialysis and precipitation did not lower  the ash content




of  the original material below the experimental error of  the ash




determination.  The inorganic materials were either chemically attached




to  the polysaccharides, possibly as salts, or were physically wrapped  in




the polysaccharides so  that diffusion through the dialysis bag was




restricted.




     Because of the fact that carbohydrates which might be important to




the present  investigation passed through the dialysis bag and might be




lost, and that dialysis and precipitation did nothing to  lower the ash

-------
                                   -90-
content, It was concluded that simple purification was not feasible.


Therefore, all additional investigations were done on the original


material.


     The original material showed a positive test for nitrogen.  However,


quantitative analysis showed only 0.84% nitrogen, an amount not considered


to represent an interfering contaminant and no special efforts were made


to separate the nitrogenous material.


     The tests for sulfur, phosphorus and the halogens were negative.


     A fundamental aspect of polysaccharides is the component monosaccharides


which are linked together to form the polymer chain.  Often polysaccharides


contain linkages which are resistant to acid cleavage.  Thus, to ensure


complete hydrolysis the material was treated with strong acid according to


the procedure of Laver, Root Shafizadeh and Lowe (22).  The hydrolyzate


was analyzed for amino acids and monosaccharides as described below.


     If the polysaccharides under investigation are water soluble and if


hydrolysis can be accomplished under mild acid conditions, treatment with


strong acids as described above is not desirable because of degradative


side reactions.  Since the material was water soluble, a mild acid hydrolysis


(3% sulfuric acid) was compared with the 77% sulfuric acid hydrolysis.


The hydrolyzate was analyzed for amino acids and monosaccharides as


described below.


     Elemental analysis showed the presence of nitrogen.  It could possibly

                                        I
be a result of some residual ammonium oxalate from the ammonium oxalate


extraction or it could be part of protein material.  The cambium layer of


Douglas-fir bark has been shown to contain protein and possibly some cambium


layer remained with the inner bark in sample preparation.


     The hydrolyzates from the strong acid hydrolysis  (77% sulfuric acid)

-------
                                   -91-
and the tnlld acid hydrolysis (3% sulfuric acid) were examined for amino




acids by two-dimensional paper chromatography.  It was found that the




second solvent, n^-butanol-formic acid-water (20:6:5 v/v/v)  muBt be




prepared just prior to use otherwise gradual esterification occurred and




altered its chromatographic usefulness.




     Chromatograms of -the hydrolyzate from the mild acid treatment did




not show amino acids.  However, chromatograms of the hydrolyzate from the




strong acid treatment showed ten purple spots.  Quantitative analysis showed




only 0.84% nitrogen content indicating that the amount of protein material




was not great.  Therefore, no effort was made to identify the amino acids.




     The mixture of monosaccharide sugars resulting from acid hydrolysis




was well resolved by paper chromatography.  The chromatograms showed that




the hydrolyzates contained a large proportion of glucose, a moderate amount




of arabinose, slight amounts of mannose and galactose and trace amounts of




xylose and rhamnose.




     The hydrolyzates from both the strong acid hydrolysis and mild acid




hydrolysis contained what appeared to be identical monosaccharides by




paper chromatography.  Therefore, strong acid was not necessary and to




avoid degradation, the 3% sulfuric acid method was used in all subsequent




hydrolysis.




     The five major sugars, glucose, mannose, galactose, arabinose and




xylose are those ordinarily found in wood and wood pulp.  The trace amount




of rhamnose represents the first time it has been found in Douglas-fir bark




although it has been reported in other barks.




     Since the monosaccharides resulting from the hydrolysis of bark are




the same as those found in wood, it is expected that the polysaccharides of




bark are similar to those in wood.  If this is so then the major carbohydrate

-------
                                   -92-
component is cellulose, but it also contains some hemicelluloses  such  as




xylans, glucomannans, glucoRalactomannans and so forth.   The isolation of




the cellulose fraction and an evaluation of its properties and its  degree




of polymerization will be of future interest.




     The gas chromatographic resolution of the alditol acetates prepared




from the acid hydrolyzate was excellent.  The myo-inositol peak at  the end




of the spectrum resulted from the addition of an accurately measured




amount of myo-inositol as an internal standard.  The areas under the peaks




of the other sugars in the spectrum are compared to the area under  the




peak of myo-inositol for quantitative results.




     The sample taken  (moisture-free, ash-free basis) yielded the following




percentages of monosaccharides:  glucose, 41.5%; arabinose 8.3%; galactose,




2.7%; mannose, 2.6%; xylose, 0.7%; rhamnose, 0.7%.  This total of 56.5% is




less than 100% because the material contains organic substances other than




carbohydrates.




     The acidified sodium chlorite delignification reaction is an oxidation




reaction.  Therefore,  it was anticipated that the primary hydroxyl groups




in position six of the hexose containing polysaccharides would most likely




be oxidized to uronic  acids as demonstrated below.  Therefore, several




qualitative tests were applied to determine if uronic acids existed.  Since




uronic acids are known to be present in natural materials, all of the tests




were run in direct comparison with the residue from the ammonium oxalate




extraction, the material from which the fraction was prepared.




     The infrared spectrum of the 0.5% ammonium oxalate insoluble material




showed a strong absorption at 1590 cm   , corresponding to absorption by




carboxylate ion.  Thus the starting material from which the fraction was




prepared contained uronic acids.  The  fraction isolated showed an absorption

-------
                                   -93-
band at exactly the same place.  However, the treatment with acidified


sodium chlorite, although an oxidation reaction, did not appear to increase


the uronic acid content, at least not as detected by infrared spectroscopy.


     The carbazole-sulfurlc acid color reaction is said to be reasonably


specific for hexuronic acids.  This was verified by testing monosaccharides


and oxalic acid and benzole acid.  These compounds shoved no reaction,


indicating that the simple sugars and carboxylic acid functions other than


those in uronic acids did not interfere with the color reaction.


     The 3% sulfuric acid hydrolyzate showed a positive purple color,


indicating the presence of hexuronic acids.  The reactions showed a maximum


absorption in the ultraviolet region at 535±1 nm as described for the test.


     The color was not particularly strong and it was considered possible


that since the sulfuric acid had been neutralized with barium hydroxide,


some of the uronic acids had formed barium salts and had precipitated


with the barium sulfate.  Therefore, a part of the acid hydrolyzate was


neutralized with ion exchange resin but the results were the same.


     The pale yellow color of the "holocellulose" from the first delignifi-


cation (Figure 5) reaction indicated incomplete delignification.  The


holocellulose was therefore treated a second time with acidified sodium


chlorite solution.  A white holocellulose was obtained in a yield of 30.6%


based on the original starting bark sample.


     This was a considerable reduction in yield from the 44.3% isolated


after the first acidified sodium chlorite treatment.  However, no effort


was made to bleach the material under milder conditions and it is believed


that the second oxidation could be performed under conditions which would
                                                           \

provide a white holocellulose with little loss of material.


     Organic compounds isolated from natural sources often contain one or

-------
                                   -94-
more of the elements nitrogen, phosphorus, sulfur or the halogens.  Therefore,




it is best to qualitatively determine if they are present because purification




techniques are usually required when they are found.  However, the tests for




these elements in the holocellulose fraction were negative.  Therefore,




plant materials, such as proteins, which contain these elements were not




present in the hdlocellulose fraction.




     The separation of the carbohydrate materials from other plant




substances, particularly lignin, is difficult and results in a choice




between two possibilities.  Most of the polymeric phenols can be removed




by strong oxidation reactions but in such a reaction the degradation of




the desired carb'ohydrates is extensive.  However, when mild oxidative




conditions are used, as in the present work, considerable polyphenolic




substances are not separated from the carbohydrate  (holocellulose) fraction.




     It is thus necessary to analyze for these polymeric phenols  in order




to have a complete knowledge of the holocellulbse fraction.  The usual




analysis is the "Klason lignin" determination which involves the hydrolysis




of the carbohydrates with strong  (72%) sulfuric acid and weighing the




insoluble solids which remain as  "Klason lignin."  The holocellulose




isolated from the inner bark of Douglas-fir was found to contain  3.1%




Klason lignin.  This is not a particularly high amount of this type of




material and so the holocellulose fraction seemed to be quite free of




highly polymeric lignin-like molecules.




     The digestion of a holocellulose material with 72% sulfuric acid




(Klason lignin determination) results in the solubilization of some lignin-




like substances which are thus not measured as "Klason lignin."  These




materials are referred to by the  term, "acid-soluble lignin" and  they must




be determined to obtain a complete analysis of a holocellulose fraction  (45).

-------
                                   -95-
     These materials are best analyzed by ultraviolet absorption.   Lignin-



like compounds strongly absorb energy at 280 nm and this band has  been



used for the quantitative determination of acid-soluble lignins.  However,



the formation of 5-(hydroxylmethyl)-furfural from hexoses and furfural



from pentoses formed during the refluxing step in the Klason lignin



determination was found to interfere with this determination.  Browning



and Bublitz (44) showed that interference by these compounds could be



minimized by determining the absorbance of the filtrate at two wavelengths.



     Using absorptivity values for lignin and for carbohydrate degradation



products obtained, respectively, from spruce Brauns lignin and from a



synthetic mixture of glucose, xylose, mannose, and glucuronolactone which



had been subjected to the hydrolysis conditions of the lignin determination,



Browning and Bublitz (44) were able to write the following equations:




     A280 = °'68 S + 18 CL
            0.15CD+70CL



where A_on and A  c were the absorbance values of the lignin filtrate, 0.68
       /oU
and 0.15 the absorptivities of carbohydrate degradation products, 18 and 70



the absorptivities of lignin at 280 and 215 nm, respectively, and C  and C
                                                                   D      Li


the concentrations in grams/liter of carbohydrate degradation products and



of soluble lignin in the filtrate.  Goldschmid (45) showed that by solving



the simultaneous equations, the following expression for the soluble lignin



concentration in the filtrate can be obtained:



               4 5T A    - A
               t.jj A215    280

           L          300



     In the present work the acid-soluble lignin content of the holocellulose



Isolated from Douglas-fir inner bark was determined by scanning the filtrate

-------
                                   -96-
from the Kalson lignin determination over the wavelength range from 320 to

200 run.  Absorption maxima were observed at 280 nra and 210 run.  The 280 nm

peak was the usual one observed for lignin compounds and the 210 nra peak was

slightly shifted from the 215 run peak reported by Browning and Bublitz (44) .

However, a shifting of wavelength due to different materials and different

analytical techniques are not uncommon (45).

     The absorbances at 280 nm and 210 nm were 2.62 and 3.77 respectively.

These values were used to calculate the acid-soluble lignin as follows:
          r    *'53 A215 " A280   ..   4.53 x 3.77 - 2.62
          C
           L         300                     300



     Using this figure in conjunction with the filtrate volume (1544.0 ml)

and the dry weight of the sample (1.8264 g) the value of 4.07% acid-soluble

lignin in the holocellulose was calculated.  Thus the total lignin content

was 3.1% (acid-insoluble) plus 4.1%  (acid-soluble) to equal 7.2%.

     The amount of acid-soluble lignin thus exceeded the amount of acid-

insoluble lignin as is generally the case with holocellulose material (39).

Therefore, correcting the lignin content for soluble lignin was essential.

Although the absorptivity values for lignin and carbohydrate degradation

products may be somewhat uncertain, as stressed by Browning and Bublitz

(44), this error is small compared with that resulting from not correcting

for soluble lignin at all.

     An important step in determining the structures of polysaccharide

materials is to acid hydrolyze them  to their monosaccharide components.

For water-insoluble polysaccharides  two steps are usually required.  The

first is to dissolve the material in a strong acid usually sulfuric, and

    second is to dilute the acid with water followed by reflux to hydrolyze

-------
                                   -97-
the glycosidic bonds.  It is important to perform the hydrolysis with




complete dissolution of polysaccharides and with minimum decomposition of




sugars.




     The holocellulose fraction from Douglas-fir inner bark was found to




completely dissolve In 77.0% sulfuric acid, the concentration used to




dissolve cellulose (22) .  It was necessary to add water to the solution




slowly with strong stirring to prevent local heating which might have




caused degradation of the sugars.  After dilution to 3.9% sulfuric acid




concentration, the solution was refluxed to bring about hydrolysis.




     The acid solution was neutralized to pH 5.0 with aqueous barium




hydroxide, resulting in a heavy precipitate of barium sulfate.  This method




of neutralization was preferred because the pH could be controlled easily.




A final pH of about 5.0 was desired because monosaccharide solutions should




not be allowed to become alkaline.  The action of alkali on monosaccharides




follows three general courses  (22):  isomerization, fragmentation, and




internal oxidation and reduction.  Such reactions interfere with the




qualitative and quantitative results of monosaccharide analyses and consider-




able care was taken to avoid them.




     The identification of the monosaccharides released from polysaccharides




on acid hydrolysis is fundamental to an understanding of these polymers.




Paper chromatography has become the standard way of tentatively identifying




the monosaccharides.  Several  solvent systems are well established and




operation of the individual sugars can usually be achieved with relative




ease.




     Paper chromatograms of the hydrolyzate from the holocellulose fraction




of Douglas-fir inner bark showed the presence of glucose (very strong spot),




mannose (medium spot), xylose  (medium spot), galactose (trace spot), and

-------
                                   -98-
arabinose (trace spot) .   These sugars moved the same distance as authentic




sugar samples on the same chromatogram.




     A trace spot was also evident near the origin.  This spot was attributed




to the presence of glucuronic acid.




     Qualitative paper chromatography provided an indication of the




monosaccharides present in the holocellulose hydrolyzate.  However, it is




possible for two or more sugars to migrate the same distance and so paper




chromatography did not unambiguously identify the monosaccharides.  The




chromatography did not show whether the sugars were of the D or the L




configurations and hence did not completely designate the sugars.  Therefore,




it was considered necessary to isolate crystalline derivatives of the sugars




to unambiguously prove their presence and their configurations.




     To prepare crystalline derivatives it was necessary to first separate




the mixture of monosaccharides into their individual, isolated sirups.




Since large amounts of each sirup were not required, the isolation was




accomplished by preparative paper chromatography as described above.  The




only limitation to the procedure was the time required to develop a




sufficent number of paper chromatograms to obtain enough milligrams of




sirup.  Each isolated sirup was reacted with reagents which provided




derivatives which crystallized readily.  In every case derivatives of




authentic monosaccharides were also crystallized for comparison purposes.




     Although the diethyl dithioacetal acetate derivatives of many




monosaccharides have been reported  in  the literature (46) not all of  them




crystallize easily.   In the present work only the derivatives from authentic




D-galactose and L-arabinose crystallized easily although diethyl-dithioacetal




acetate sirups of D-glucose, D-xylose  and D-mannose were also prepared.




     Because only a few milligrams  of  each monosaccharide sirup was available

-------
                                   -99-
from the holocellulose hydrolyzate, only the sirups from the easily




crystallized galactose and arabinose fractions were reacted to prepare




diethyl dithioacetal acetate materials.  In this way crystalline penta-0-




acetyl-D-galactose diethyl dithioacetal and crystalline tetra-0-acetyl-L-




arabinose diethyl dithioacetal were obtained.  These materials proved the




presence of D-galactose and L-arabinose residues in the polysaccharides




from Douglas-fir inner bark.




     The acetylation of glucose is a common reaction and in the present work




white crystals of penta-0-acetyl-B-D-glucopyranose were readily prepared




from authentic D-glucose.  The glucose sirup isolated from the holocellulose




hydrolyzate also yielded penta-()-acetyl-8-D-glucopyranose upon acetylation.




This derivative proved the presence of D-glucose residues in the polysaccharide




from Douglas-fir inner bark.




     The acetates and the diethyl dithioacetal acetate derivatives of xylose




do not crystallize easily.  However, D-xylose can be easily identified as




its di-0-benzylidene dimethyl acetal and in the present work di-O-benzylidene-




D-xylose dimethyl acetal was readily obtained from authentic D-xylose.  The




xylose sirup isolated from the holocellulose hydrolyzate also yielded




di-0-benzylidene-D-xylose dimethyl acetal upon reaction.  This derivative




proved the presence of D-xylose residues in the polysaccharides from




Douglas-fir inner bark.




     Mannose is a difficult sugar to crystallize and there are also few




derivatives of mannose that crystallize easily.  However, D-mannose




phenylhydrazone crystallizes readily and can be used to identify and to




isolate the sugar.  The reaction must be carefully done, or a second molecule




of phenylhydrazine will add to the sugar and the phenylosazone will result.




In fact, this is what occurs with the other monosaccharides.  Phenylosazones

-------
cannot be used to identify sugars because both Cl and C2 are involved and

sugars which differ only about Cl and C2S such as mannose, glucose, and

fructose, cannot be differentiated.  Thus to identify mannose, the reaction

must be stopped after the addition of only one mole of phenylhydrazine to

give mannose phenylhydrazone.  Isbell and Frush  (49) have reported the

reaction conditions to perform this step.

     In the present work, D-mannose phenylhydrazone was readily prepared

from authentic D-mannose.  Because D-mannose phenylhydrazone crystallizes

so readily from the reaction system outlined by  Isbell and Frush  (49), it

can be preferentially crystallized from mixtures of monosaccharides.  Thus,

in the present work, a complete hydrolyzate sirup (77.0% sulfuric acid

hydrolysis) of the holocellulose from Douglas-fir bark was reacted with

phenylhydrazine.  After standing overnight in the refrigerator, white

crystals of D-mannose phenylhydrazone were obtained.  This derivative

proved the presence of D-mannose residues in the polysaccharides  from

Douglas-fir bark.

     There are numerous modifications of the standard Fehling's copper

reduction test for the presence of reducing sugars.  However, one of the

more common methods for the quantitative analysis of reducing sugars is

the Soraogyi (52) procedure.  This method is based on the ability  of certain

sugars (reducing sugars) to act as reducing agents.  The sugar reacts with

Cu   in the aqueous alkaline medium to produce cuprous oxide.  The cuprous
                                      j_i_
oxide ie oxidized by iodine back to Cu   and the excess iodine is titrated
                                         i
with thiosulfate.  The reactions are:

          RCHO -»- 2 Cu"*"* -»- 5 OH~ -»• RCO~ + Cu-0 +  3 H~0

          I0~ + 5 I~ + 6 H* •*  3 I2 + 3 H 0

          Cu20 + 2 H"*" + I2 -» 2 Cu""" +21" + H20

-------
                                  -101-
          I2 + 2 S203  + 21  + S406




     This method has been widely and successfully applied on both milligram




;md microgram quantities, both titrimetrically and colorimetrically.  Its




accuracy over a wide range of sugar concentrations, the ease and rapidity




of operation, and its proven reliability place it above other micro-




oxidation methods (52) .




     As evidenced from the above equations, thiosulfate anion is used as the




standard tlter to measure the excess iodine concentration.  In the present




work the sodium thiosulfate solution was standardized by titration to a




starch end-point of an accurately diluted solution of potassium iodate




which had been prepared from recrystallized and dried potassium iodate




solids.




     Several different alkaline copper reagents have been recommended for




use in conjunction with the iodometrlc determination of reduced copper.




The most common ones are Somogyi's 1945 (52) reagent and his 1952 reagent.




Somogyi's 1945 phosphate-buffered reagent possesses an advantage over the




1952 carbonate-buffered reagent, in that amyloses are held in solution in




the 1945 reagent but are precipitated from the 1952 reagent.  Therefore,




the 1945 reagent has become the standard procedure and was the method of




choice in the present work.  The amount of potassium iodate added to the




reagent determines the amount of iodine released and hence the amount of




sodium thiosulfate titer needed.  Therefore, some preliminary titrations




are often required to determine the best' amount of potassium Iodate to add.




The blank should require between 9.0-10.0 ml of 0.005 M sodium thiosulfate




solution.  When performed properly the method is very good for determining




the overall reducing power of a solution and the precision is ±0.01 mg for




D-glucose or about ±2%, averaged throughout the range 0.3 to 3.0 mg of glucose.

-------
                                  -102-
     When determining the quantitative amounts of monosaccharides released




from a holocellulose fraction on acid hydrolysis, it is important that the




time of refluxing be carefully determined.  If the refluxing time is too




short, complete hydrolysis is not realized and the quantitative analysis




will be erroneous because those monosaccharide residues linked by the more




acid-resistant glycosidic bonds will not be determined.  If the refluxing




time is too long, acid degradation of the hydrolyzed monosaccharides will




be extensive.  Acid reversion to new, complicated polymers may also occur




on extended heating.  The importance of the correct reflux time to achieve




maximum monosaccharide yield has been demonstrated by Laver, Root, Shafizadeh




and Lowe (22) in their studies of pulp analyses.




     The simpliest way to measure the proper time to reflux the acid solution




is to measure the reducing power of the hydrolysis solution as the reaction




progresses.  In the present work this was accomplished by removing aliquots




of the refluxing solution at specific times and determining the reducing  «




power of the aliquots by the Somogyi method.  Glucose was used as a




convenient standard reducing sugar in which terms the concentration of




monosaccharides was measured.  A plot of the reducing power in glucose




equivalents versus the time of reflux showed clearly that 12 hours is the




optimum time to reflux the hydrolysis solution to achieve maximum yield




of monosaccharides.




     The reflux time of 12 hours was considerably longer than the 4.5 hours




of reflux time which Laver9 Root, Shafizadeh and Lowe  (22) showed to be




necessary for wood pulps.  Therefore, the holocellulose fraction from




Douglas-fir inner bark contained glycosidic bonds which were more acid




resistant than those in wood pulps.  These acid resistant bonds were




attributed to the mannose containing polysaccharides in the holocellulose.

-------
                                  -103-
     Carbohydrates are neither heat resistant nor volatile and so derivatives




must be prepared which will volatilize without degradation in order to




perform gas-liquid chromatographic analyses.




     The derivatives most commonly used today are the "alditol acetates."




In the preparation of the alditol acetates the monosaccharides are first




treated with sodium borohydride to reduce the aldehyde function to the




alcohol function.  This has the result of preventing ring isomerization to




the pyranose and furanose forms and so prevents the formation of alpha and




beta forms of the sugars.  The end result is only one form, the alditol,




for each of the monosaccharide sugars rather than three, four or even five




forms.  The acetate esters are synthesized from the alditols to make volatile,




heat resistant derivatives.




     Identification of compounds by gas-liquid chromatography is realized




by comparing the time required for the unknown material to pass through the




column with the time required for a sample of the authentic material to pass




through the column.  These times are called "retention times."  Qualitative




paper chromatography of the hydrolyzate of the Douglas-fir holocellulose




had shown the presence of glucose, mannose, galactose, xylose, and arabinose.




Therefore, authentic crystalline alditol acetates of each of these sugars




were synthesized to positively determine retention times and instrument




response for quantitative analyses.  myo-Inositol hexaacetate was also




synthesized to be used as an internal standard.  Each of these crystalline




materials was passed through the gas chromatdgraph and the retention time




for each was determined. • The crystalline materials were then mixed and




again passed through the gas chromatograph to ascertain the resolution.




The separation of each alditol acetate was excellent and the retention




times of each was measured at the center of the peak.

-------
                                  -104-
     There are nsany parameters which can be varied in gas-liquid chromato-


graphic techniques.  The resolution of the peaks will be changed by changes


in the carrier gas flow and changes in the column temperature.   Peak heights


and peak areas will also change with changes in injection port  and detector


temperatures.  Therefore, in the present study several conditions were


systematically changed until the optimum conditions were determined.  The


conditions used throughout the rest of the work were:  column,  6.5% ECNSS-M


on Gas Chrom Q 100/120 mesh, 6 ft x 1/8 in O.D. stainless steel; injection


port 180°; detector 240°; column temperature 180° isothermal; helium flow

                           2
30 ml/min; range setting 10 ; attenuation setting 16.


     Gas chromatographic detectors respond differently to different compounds,


These response factors must be known to obtain quantitative results.  The


recorder is also a possible source of error and the precision obtainable


with standards should be determined.  A good way to reduce these sources of


error is to add an accurately weighed amount of an internal standard to the


mixture to be analyzed and compare the peak areas of the compounds to be


measured against the peak area of the internal standard.  The internal


standard used in the present work was myjv-inositol hexaacetate.


     The response  of the gas chromatography to each of the alditol acetates


to be measured was calibrated by analyzing varying, but accurately weighed,


amounts of each crystalline, authentic, alditol acetate with authentic,


crystalline myo-inositol hexaacetate.  The calibration factor for each


alditol acetate was obtained by plotting' the ratio of the area of each


alditol acetate to the area of the standard (myo-inositol hexaacetate) as


ordinate against the ratio of the weights of each as the abscissa.  Straight


lines passing through the origin resulted for arabinitol pentaacetate,


xylitol pentaacetate, mannitol hexaacetate, galactitol hexaacetate, and

-------
                                  -105-
glucitol hexaacetate.  The slope of each of these lines is the "Instrument
K Factor" and represents the overall response of the gas chromatographic
system to each of the alditol acetates calibrated against the internal
standard, myo-inositol hexaacetate.  The "Instrument K. Factors" were:
arabinitol pentaacetate 0.96; xylitol pentaacetate 0.95; mannitol hexaacetate
0.96; galactitol hexaacetate 0.98; glucitol hexaacetate 0.92,  These
"Instrument K Factors" were used in subsequent calculations.
     A sample of the holocellulose from Douglas-fir inner bark was carefully
hydrolyzed to minimize acid degradation.  The hydrolyzate so obtained was
reduced by the addition of sodium borohydride and the excess sodium boro-
hydride was decomposed by acetic acid.  The solution was concentrated to
a sirup.  The sirup was dissolved in methanol and the methanol was
reevaporated.  This procedure was repeated a total of eight times.
Albersheim, Nevins, English and Karr (32) showed that the boric acid level
could be reduced to a convenient level by a number of additions and
reevaporations of methanol.  The boric acid was presumably converted to
volatile methyl borate which was evaporated from the sirup.  The removal
of boric acid is important otherwise the acetylation reaction does not
proceed.  Acetylation and the gas-liquid chromatographic analysis was
accomplished as outlined above.
     The areas under the peaks were measured with a planimeter and were
found to be (duplicate analyses):  arabinitol pentaacetate, 13.2; xylitol
pentaacetate, 39.7; mannitol hexaacetate', 62.0; galactitol hexaacetate,
15.A; glucitol hexaacetate, 386.0; myo-inositol hexaacetate, 232.2.
     The following equation was used to determine the individual anhydro-
monosaccharide residue in the holocellulose fraction (31):
                                            C x I x F x 100
          % anhydromonosaccharide residue
                                             R x S x H x K

-------
                                  -106-
where


     C = chromatographic area of the1 component alditol acetate peak,


     R = chromatographic area of the myo-inositol hexaacetate peak,


     I = weight of the myo-inositol originally added, in grams,


     S = dry weight of the original holocellulose sample, in grams,


     F = factor to convert weight monosaccharide to anhydromonosaccharide


         residue  (0.88 pentose)(0.90 hexose),


     H = hydrolysis survival factor (90),


     K = "Instrument K Factor".


Using the area under the peak of glucitol hexaacetate as an example the


calculations become:

          ,.   ,   ,   ,          . ,       386.0 x 100 mg x 0.90 x 100      ,, ,
          L anhydroglucose residue = 232.2 x 273.96 mg x 0.974 x 0.92 = 6l«




     In a similar way the percentages of anhydromonosaccharide residues in


the Douglas-fir inner bark holocellulose were calculated (average of two


determinations):  anhydroarabinose residues9 2.6; anhydroxylose residues,


6.3; anhydromannose residues, 9.5; anhydrogalactose residues, 2.3;


anhydroglucose residues, 61.1.  The total yield of anhydromonosaccharides


was 81.8%.  This  value is comparable to the values of 7A.9%, 78.0%  and


81.9% reported by Borchard and  Piper (31) for the total carbohydrate


content of chlorite holopulps and appears to be favorable for materials of


this type.


     The value of 81.8% carbohydrates plus  the value of 7.2% lignin amounts


for 89.0% of the  solids in the  holocellulose fraction.  This total  value  is


similar to the values usually reported for  the solids content of materials


of  this type.


     The study of the structure of bark hemicelluloses requires homogeneous

-------
                                  -107-
polytners isolated in high yields with minimal modification.  This is usually




accomplished, as in the present work, by delignifying the bark and extracting




the resulting holocellulose with aqueous alkali.  Solubility differences




can be exploited for the separation of individual hemicelluloses by judicious




use of cations and concentrations.  Thus, relatively dilute alkalis suffice




to dissolve xylans and galactoglucomannans, but higher concentrations are




required for the extraction of glucommannans.




     The procedure used in the present work to separate the polysaccharides




which comprise the holocellulose from Douglas-fir inner bark is outlined in




Figure 7.  The key feature was selective blocking of the dissolution of




mannose-containing polysaccharides in the first extraction step.  This was




accomplished by impregnating the holocellulose with aqeuous barium hydroxide




(55).  The impregnated holocellulose was then contacted with 10.0% aqueous




potassium hydroxide which was known to be a good solvent for xylan-rich




polysaccharides.  The extract was neutralized and methanol was added to




70.0% concentration.  The resulting precipitate was labelled "Hemicellulose




A."  Hemicellulose A was recovered in 7.6% yield based on the starting




holocellulose (Figure 7).




     Residue A (Figure 7) remaining from the 10.0% potassium hydroxide




extraction was expected to contain most of the mannose-containing hemicelluloses




because of the impregnation with barium hydroxide (55).  After removal of




barium ions by dialysis, the residue was extracted with 1.0% aqueous sodium




hydroxide which has been shown to be a gbod solvent for galactoglucomannan




hemicelluloses (55).




     The extract was neutralized and methanol was added to 70.0% concentration.




The resulting precipitate was labelled "Hemicellulose B."  Hemicellulose B




was recovered in 1.7% yield based on the starting holocellulose (Figure 7).

-------
                                  -108-
     The concentration of the sodium hydroxide extraction medium was


increased to 15.0% sodium hydroxide because it has been shown (55)  that

polysaccharides rich in mannose residues become soluble in concentrated


alkali solutions.  The extract was neutralized and methanol was added to

70.0% concentration.  The resulting precipitate was labelled "Hemicellulose

C."  Hemicellulose C was recovered in 2,9% yield based on the original

holocellulose  (Figure 7).

     Residue C (Figure 7) was hydrolyzed by acid.  Paper chromatography

showed considerable amounts of mannose and galactose and traces of  xylose

and arabinose as well as glucose in the hydrolyzate.  It was thought possible

that increasing the sodium hydroxide concentration from 15.0% to 18.0%


would extract more of the mannose- and galactose-containing polysaccharides

and leave a relatively homogeneous glucan as a residue.  However, the

residue from the 18.0% sodium hydroxide still appeared to contain the


same ratios of sugars.  Therefore, further investigations were concentrated

on the extract and residue from the 15.0% sodium hydroxide extraction


(Figure 7) rather than those from the 18.0% sodium hydroxide extraction.

     The minimum hydrolysis time for Hemicellulose A in acid to achieve

the maximum yield of monosaccharide was determined to be 4 hours.  A plot

of glucose equivalents (%) versus reflux time in hours indicated complete

hydrolysis after 4 hours and in most cases additional aliquots of the

material were  refluxed for 4.5 to 6 hours.

     The maximum amount of glucose equivalents released was found to be

59.3%.  However, the fraction contained 29.7% ash determined as the oxides.
                                                i I
The major inorganic ion was considered to be Ba   because of the previous

impregnation with barium hydroxide.  By calculation the percentage of

barium ions in the fraction would be 26.6%.  Therefore, the total solids

-------
                                  -109-
accounted for in the fraction was 85.9%.  Acid degradation of xylose


containing polysaccharides is known to be extensive.  In fact, the


conversion of pentoses to furfural under conditions of acid reflux is the


basis of the TAPPI standard method for the determination of pentosans in


wood and pulp.  The relatively low total recovery of materials in the


Hemicellulose A fraction was, therefore, attributed to acid degradation.


     A paper chromatogram of the acid hydrolyzate showed a very strong


spot for xylose, weak spots for glucose, galactose, and arabinose and a


trace spot for mannose.  There was a slow moving spot which barely migrated


from the origin.  This spot was attributed to glucuronic acid.


     The trace amount of mannose indicated that the impregnation of the


holocellulose with barium hydroxide worked well in blocking the dissolution


of raannan-containing hemicelluloses.  These results are in agreement with


those previously reported (55) for the separation of materials of this type.


     The Hemicellulose A fraction was shown to be a "xylan" composed of


the following ratio of sugar residues:  xylose, 4.5; arabinose, 1.0;


glucuronic acid, 1.0.  The xylan possessed a specific rotation of -30.5.


The intrinsic viscosity of the polysaccharide in molar deithylenediamine


copper II reagent at 25° was 0.42 dl/g which corresponded to a degree of


polymerization of 89.  The molecular weight of the xylan as analyzed by

                               4
end-group analysis was 1.8 x 10  which, in combination with gas-liquid


chromatographic analysis, showed a degree of polymerization of 90.


     The xylan was oxidized with periodate anion.  It consumed 133.5 moles


of periodate anion and released 22.5 moles of formic acid per mole of xylan.


The oxidized xylan was hydrolyzed and analyzed by gas-liquid chromatography


and showed the following ratio of sugar residues:  xylose, 7.0; arabinose


1.0.

-------
                                  -110-
     The above data were consistent with a polysaccharide xylan structure



consisting of a backbone of 88 anhydro-D-xylopyranose units hooked to



8-jO-(l-*4) plus a reducing and a non-reducing end group on the backbone.



There were 20 anhydroglucuronic acid side chains and 20 anhydroarabinose



side chains on these 90 units.  At least ten of the anhydroarabinose units



were in the form of monoarabinose side chains and up to ten were in the



form of arabinoblose or longer side chains.



     The xylan was reacted at 100° with aqueous sodium hydroxide solutions



ranging in concentration from 0.001 N to 3.676 N.  The rate of degradation



did not change at alkali concentrations above 0.1 N in sodium hydroxide.



     Alkaline degradation of the xylan in 0.1 N aqueous sodium hydroxide



at 100° showed a rate constant for end-group peeling of k.. = k~ = 5.33


    -1                                                  -1
hour   and a termination rate constant of k, =» 0.66 hour



     The peeling reaction stopped when 19.81% of the xylan had been



degraded.  This small amount of degradation reflected the relatively high



termination rate constant compared to the peeling rate constant.



     Hemicellulose B was hydrolyzed with sulfuric acid and the hydrolyzate



qualitatively analyzed by paper chromatography.  The chromatogram showed



strong spots for glucose and tnannose, weak spots for galactose and xylose,



and a trace spot for arabinose.



     The fraction still contained some xylose residues indicating that  the



extraction of the xylose-containing hemicelluloses was not complete by  the



extraction with 10.0% potassium hydroxide'.  Beelik, Concac Hamilton and



Partlow  (55) also reported xylose residues in similar fractions isolated



from western hemlock holocellulosa and grand fir holocellulose.  Hemicellulose



B was considered to be a galactoglucomannan,



     Hemicellulose C was hydeolyzed, neutralized, and analyzed by paper

-------
                                  -111-
chromatography by the methods described for the holocelluloae fraction.
The chromatogram showed that the Hemicellulose C fraction was composed of
a large amount of mannose residues, small amounts of glucose and galactose
residues, and trace amounts of arabinose and xylose residues.
     The results indicate that all of the xylan hemicelluloses had been
extracted by the 10.0% potassium hydroxide and the 1.0% sodium hydroxide
treatments.  This is in agreement with the results reported by Beelik,
Conca, Hamilton and Partlow (55) for hemicelluloses from softwoods.
     The gas-liquid chromatographic spectrum of the alditol acetates
prepared from the acid hydrolyzate of Hemicellulose C showed that the major
peak present was from mannitol hexaacetate.  The areas under the peaks of
rhamnitol pentaacetate, arabinitol pentaacetate, and xylitol pentaacetate
were too small to be measured.  The areas under the remaining peaks as
measured by a disc integrator were (average of triplicate reading):
mannitol hexaacetate, 3407; galactitol hexaacetate, 315; glucitol hexaacetate,
240; myo-inositol hexaacetate, 11617.  The percentages of anhydromonosaccharide
residues were calculated similarily to those for the holocellulose fraction
from the formula:
          «,..,          u-ij     jj     C x I x F x 100
          % anhydromonosaccharide residue =• —	~	—
                                             K x b x n x K.
     The results were: anhydromannose residue, 85.7%; anhydrogalactose
residue, 7.0%; anhydroglucose residue, 4.6%; total recovery, 97.3%.
     The ratio of anhydrosugars in Hemicellulose C was therefore: anhydro-
glucose, 1.0; anhydrogalactose, 1.5; anhydromannose, 18.6.  Therefore, the
great majority of Hemicellulose C is composed of mannose-containing
polysaccharides and is therefore termed a "mannan."
                                                23
     The specific rotation of the mannan was [a]   - 40° (c_ 0.25, N sodium
hydroxide) .  This is close to the rotations of mannans from other plant

-------
                                  -112-
sources and supports the above conclusion that this fraction from Douglas-fir



inner bark is primarily a mannan.



     Methylation of a polysaccharide provides information about the positions



of the glycosidic bonds which unite the anhydromonosaccharide residues in



the polymer.  The usual procedure is to form methyl ethers on all of the



free hydroxyl groups.  These ethers are resistant to the acid hydrolysis



conditions ordinarily applied and so hydrolysis of the completely methylated



polysaccharide frees only the hydroxyl groups involved in the glycosidic



linkages.  Location of the positions of the freed hydroxyl groups determines



the point of attachment of the glycosidic bond in the original polymer.



     It is important that all of the hydroxyl groups in the original



polysaccharide be methylated.  The methylation reaction is usually analyzed



by infrared spectroscopy because the hydroxyl groups absorb readily and



their disappearance as methylation progresses can be followed.



     In the present work with the mannan from Douglas-fir inner bark,



raethylation proved difficult.  However, after extended reaction time the



infrared spectrum showed no absorption for hydroxyl groups in the usual



regions of 3800 ^ 3300 cm  .  Therefore, methylation was considered complete.



     Formic acid has been shown  to be a good medium for the hydrolysis of



mannans and was used in the present work-  The initial formic acid treatment



must be followed by an hydrolysis with dilute sulfuric acid to hydrolyze



any formate esters which might have been formed.



     Paper chromatography of the hydrolyfeate sirup in two different solvent



systems showed R, and R  values  consistent with the presence of 2,3,6-tri-
                r      o


0-methyl-D-mannopyranose (very strong spot), 293p4,6-tetra-0-methyl-p-



mannopyranose (trace spot) and 2.3-di-O-methyl-D-mannopyranose (trace spot).



There were several unidentified  spots which are possibly partially methylated

-------
                                  -113-
sugars resulting from an incomplete methylation of the mannan as is




evidenced by the presence of 2,3-di-0_-inethyl-D-mannopyrano8e.  They were




present in very trace amounts and could only be detected when the paper




chromatogram was heavily loaded with hydrolyzate sirup.  On the other hand,




the spot for 2,3,6-tri-O-methyl-D-mannopyranose was very strong.




     The crystallization of the above derivative of 2,3,6-tri-O-methyl-p-




raannopyranose shows that the hydroxyl group on C4 of the mannopyranose




compound was freed on hydrolysis of the methylated mannan.  The isolation




of the crystalline derivative supports the paper chromatographic evidence




for the presence of 2,3,6-tri-0_-methyl-D-mannopyranose in the hydrolyzate




of the methylated mannan.  Moreover the paper chromatograms show it to be




not only the major spot but essentially the only spot.




     Therefore, it is concluded the Hemicellulose C is primarily a mannose




containing polysaccharide and that the anhydroraannose residues are attached




by 3-D_-(l+4)-glycosidic bonds.  The B configuration is taken from the negative




optical rotation (-40°).




     A paper chromatogram of the acid hydrolyzate of Residue C showed a




strong spot for glucose, medium spots for mannose and galactose, and trace




spots for arabinose and xylose.  Residue C therefore appeared to be a




mixture of the hemicelluloses now known to be present in the holocellulose




from Douglas-fir inner bark.  While the extractions shown in Figure 7 tended




to dissolve relatively pure hemicelluloses, the final residue was still a




mixture.




     Formic acid has been shown to completely remove hemicelluloses from an




anhydroglucose rich material and leave a residue containing only anhydroglucose




residues (70).  The degradation of the residue in this procedure can be




quite extensive and relatively short treatment of 4-hour duration was first

-------
                                  -114-
attempted.  However, a paper chromatogram of the hydrolyzate of the residue


remaining after the 4-hour extraction showed a weak spot for mannose, and a


trace spot for xylose and arabinose as well as the strong spot for glucose.


Therefore, removal of the hemicelluloses was not complete.


     Longer extraction time in the refluxing formic acid was attempted.


The yield of residue was 36.0% and so degradation was quite extensive.


However, a paper chromatogram of the acid hydrolyzate of the residue showed


the presence of glucose only with no contamination from other sugars.


Therefore, the other hemicellulose linkages had been hydrolyzed by the
                                     i

formic acid in preference to those joining  the anhydroglucose residues.


Therefore, a glucan was left as the residue.


     Since the material under investigation was a glucan, the Somogyi copper


reduction method with glucose as the standard sugar would give a reliable


quantitative analysis of the amount of glucan in the sample.  The amount


of reflux time required for total hydrolysis was also determined by removing


aliquots as the reaction progressed and determining the reducing power.


     The hydrolysis was complete after 4 hours of reflux and the overall


yield in glucose equivalents was 103.3%.  Values of greater than 100%


recovery occur sometimes in this type of procedure due to the accumulative


sources of error in moisture content determinations, dilution measurements


and the error of the  Somogyi method.  The large yield does show that  the


residue remaining  after extracting Residue  C  (Figure 7) for 4 days with  formic

acid was a glucan.                       ,


     The. gas-liquid chromatographic spectrum of tha alditol acetate prepared


from the acid hydrolyzate of the glucan  showed trace peaks for rhamnitol

pentaacetate, arabinitol pentaacetate, and  xylitol pentaacetate.  The areas


of these peaks were too small to be measured.  Primarily the spectrum shows

-------
                                  -115-
peaks for glucitol hexaacetate and the added myo-inositol  hexaacetate.  The



areas under these peaks, as measured by a planitneter were  (average  of  three



readings):  glucitol hexaacetate, 41; myo-inositol hexaacetate,  165.   The



percentage of anhydroglucose residues was calculated similarly to those for



the holocellulose fraction from the formula:



          v   u j   -i         a     C x I x F x 100
          6 anhydroglucose residue =  R x s x H x K



The results showed the percent of anhydroglucose residues  to be  94.4%.



     A definitive method of establishing linkages in a polysaccharide  is



by fragmentation analysis.  If known dimers can be identified, then the



linkage Joining the two anhydromonosaccharide units is established.  In



the present investigation of the glucan, cellobiose octaacetate  was isolated



in crystalline form from the glucan.  Thus the presence of B-D.-(l-»-4)



glycosidic bonds in the glucan was established.  It is interesting  to  note,



however that the time of acetolysis had to be shortened from two weeks,



which is the usual reaction time for cellulose (70) to one week.  No



cellobiose octaacetate was obtained if the acetolysis of the glucan from



Douglas-fir bark was allowed to react for two weeks.  This would indicate



that the chain length was short, because if the acetolysis proceeds too



far on a glucan then glucose pentaacetate is produced rather than the  dimer.



Thus the acetolysis showed the presence of (3-I>-(l"*^) glucosidic  bonds,  but



indications were of a short chain.



     The calculations of the viscosity data were made as follows:



     n     - Kpt



     nr    = n/no





     nsp   = (T1-V/no = nr " l


     H  .  = reduced viscosity





                 

-------
                                  -116-







where



     n     = viscosity in centipoise (cp)



     n     a viscosity in pur® solvent            •  .
      o


     H     o relative viscosity



     [n]   a Intrinsic viscosity in deciliters/gran (dl/g)



     K     ° viscotneter COBStent



     p     ° density at 25<,0°±0»Q1



     t     o time in seconds for solution to flow through the viscometer



     c     a concentration in g/dl



     The plot of the reduced viscosities against the concentration showed a



straight line ma determined by linear regression analysis.  The value for



the intrinsic viscosity wa© obtained by extrapolating the plots to zero



concentration.  This was 
-------
                                  -117-
prior to hydrolysis.  The Infrared spectra of the methylated glucan showed



the disappearance of the hydroxyl absorption in the 3800 - 3300 cm   range



indicating that all of the hydroxyl groups had been methylated.



     The methylated glucan was hydrolyzed with sulfuric acid in the same



way that the glucan had been hydrolyzed.  Paper chromatography in two



different solvent systems showed R_ and R_ values consistent with the
                                  t      u


presence of 2,3,4,6-tetra-O-methyl-D-glucopyranose (trace spot), and



2,3,6-tri-O-methyl-D-glucopyranose (very strong spot).  The unidentified



spots are possibly partially methylated sugars resulting from slightly



incomplete methylation of the glucan.  These materials were present in only



trace amounts and could be detected only when the chromatograms were heavily



loaded with hydrolyzate sirup.  On the other hand the spot for 2,3,6-tri-O-



methyl-D-glucopyranose was very pronounced.



     The isolation of crystalline cellobiose octoaacetate from acetolysis



of the glucan proved the presence of B-^-Cl^A) linkages and hence established



the presence of 2,3,6-tri-O-methyl-D-glucopyranose in the hydrolyzate from



the methylated glucan.  Thus the identification of 2,3,4,6-tetra-O-methyl-



D-glucopyranose by comparing relative spot movements on paper chromatograms



was placed on firm ground.  This material came from the non-reducing end



group of the glucan.



     The above data on the glucan is consistent with the presence of a



linear polysaccharide of 65 repeating anhydroglucopyranose units attached



by 8-1)-(1+4) glucosidic bonds plus a reducing and non-reducing end group.

-------
                                  -118-









                          DOUGLAS-FIR BARK WAX




I.  Historical Review




     Kurth and Kiefer (71) found that the chemical components of the whole




hark could be partially separated by extraction with various solvents.  Table




6 indicates the nature and yields of the Douglas-fir extractives which they




removed by successive extraction with the solvents shown.




     Kurth, in a separate publication (72)p reported on the chemical




composition of the "waxes" In Douglas-fir bark.  The n-hexane-soluble wax was




saponified by refluxing a mixture of 20 g of the wax, 20 g of potassium




hydroxide and 300 ml of 70% ethanol for three hours.  Following this, 100 ml




of water was added, the alcohol was removed by evaporation, and the residue




extracted with n-hexane in a separatory funnel to remove unsaponifiable




compounds.  The unsaponifi&ble matter was dried and recrystallized from




acetone.  A white crystalline solid was obtained which was identified as




lif.noceryl alcohol, also named tetracosanol-1, a straight-chained, saturated




alcohol, C24H5QOH.  The yield was 19.5% of the "wax."  The filtrate from




the lignoceryl alcohol crystallization gave a positive Liebermann-Burchard




test for sterols (73, p. 80; 74, p. 100) and a precipitate with digitonin.




After removal of the acetone, the residue was crystallized as white needles




from dilute alcohol.  The material was identified as "phytosterols."  The




yield was 0.3% of the "wax."  "Phytosterols" are now known to be mixtures




of closely related sterols of which the most common one in bark is 8-sitosterol.




     The alkaline solution from the separation of the unsaponifiable matter




was acidified with sulfuric acid and extracted with n-hexane in a separatory




funnel.  The t^-hexane solution was washed with water, dried over anhydrous




sodium sulfate, and evaporated to dryness.  The residue was recrystallized




from acetone and then from ii-hexane.  The white crystals were identified as

-------
                                  -119-
lignoceric acid, also named tetracosanoic acid, a straight-chained, saturated


acid, C2,H,gO .  The yield was 60.5% of the "wax."  The small amount of


yellow solid residue left in the filtrate from this crystallization was


oxidized with cold potassium permanganate to dihydroxystearic acid; m.p.


129-130°.  This indicated the presence of oleic acid in the original wax.


     A brown resin remained suspended in the aqueous liquor after the

                            f
n-hexane extraction of the lignoceric acid.  The resin was extracted with


diethyl ether.  Evaporation of the ether left a brown residue which


crystallized from benzene as fine crystals.  Recrystallization from dilute


ethanol gave yellow prisms.  These crystals were identified as ferulic acid


(4-hydroxy-3-methoxycinnamic acid).  The yield was 22% of the original wax.


     The fraction of the bark which was n_-hexane insoluble but benzene soluble


(benzene-soluble "wax") was reported by Kurth and co-workers (75, 76) to


possess a more complicated composition than the n-hexane-soluble "wax."


It had a melting point of 60° to 63°, and Kurth tentatively isolated, after


saponification, a fatty acid mixture in about 25% yield, a dark-colored


phlobaphene in 24% yield, a dark colored diethyl ether-soluble acid fraction


in 26% yield, unsaponifiable matter in 5% yield and glycerol.


     The cork fraction of Douglas-fir bark was studied by Hergert and Kurth


(76).  The extractives were separated according to their solubilities in


n-hexane, benzene, diethyl ether, ethanol and hot water.  The rv-hexane and


benzene fractions were "waxes."  The yield of n_-hexane-soluble "wax" was


5.78% and of benzene-soluble "wax" was 1.75% based on an average of nine


trees sampled.  The n-hexane-soluble "wax" was found to contain lignoceric


acid, 49.3%; lignoceryl alcohol, 27.5%; ferulic acid, 9.8%; "phytosterol",


0.6%; and n-hexane-insoluble, benzene-soluble acidic material, 8.1%; a


benzene-insoluble phenolic material, 3.6%.  The n-hexane-insoluhle but

-------
                                  -120-
benzene-soluble "wax" proved to have a complicated composition.  The authors




did not locate the position of the hydroxyl group in the hydroxypalmitic




acid, C..,H_20 , but reported that it was not terminal.




     It is interesting to note that dihydroquercetin was the extractive




present in greatest amount from the cork.  Although not extracted with




n-hexane or benzene, and not considered a part of the "wax-portion," Douglas-




fir bark represents a major source of this compound.  It is easily obtained




from the "wax"-free (benzene extracted) bark or cork by extraction with




diethyl ether.  Upon evaporation of the diethyl ether it can be recrystallized




from hot water as long white needles, m.p. 241-248 (71).  The quantity of




dihydroquercetin in cork varies from 5% in samples of cork from second-growth




trees to 23% in cork from mature trees (76) .  This indicates the desirability




of separating the cork from the other bark components when the extraction of




dihydroquercetin is intended.




     In 1967 Kurth reported more recent work on the n-hexane-insoluble,




benzene-soluble "waxes" of Douglas-fir bark and cork  (75).  In addition to




the compounds previously mentioned, he found an hydroxybehenic acid c?oH,,0~,




and an hydroxyarachidic acid (hydroxyeicosanoic acid), C Jfl,-.0-.




     The hydroxybehenic acid was crystalline but gave a range of melting points




depending upon how it was.recrystallized.  Kurth comments that this is




characteristic of those hydroxy-fatty acids that form lactones and lactides.




He did not report the position of the hydroxyl group.  Likewise, he did not




report the exact position of the hydroxyl' group in the hydroxyarachidic acid.




He also mentions the presence of a dicarboxylic acid of tentative formula




C20H38°4 ^ut ^ic* not reP°rt further characterization.



     The chemical composition of the bast fibers of Douglas-fir bark were




also studied by Kiefer and Kurth (77).  Table 6 shows the yields of extractives

-------
                                   -121-
resulting  from  successive extractions with  the  solvents  shown.   Thef$ indicated




•that the bast fibers did not  contain as much extractive  material,  particularly




dihydroquercetin  and "wax," as did  the cork fraction  of  Douglas-fir^bark.




Further separation  of  the extractives  into  their  individual  components was




not reported.




II.  Experimental:  a,-Hexane-Soluble Wax




A.  Collection  of_ Bark Samples




     Bark  samples used in this study were collected  from indigenous  sources




of Douglas-fir.   They  consisted  of  bark from trees of various apes and




diameters.   The specimens were stripped from the  trees,  sealed in  polyethylene




bap,s,  and  brought to the Forest  Research  Laboratory,  Oregon  State  University.




B.  Sample Preparation and ja.-Hexane Extraction




     Bark  chips were ground to pass a  screen of ten meshes to the  inch  (The




W. S.  Tyler Company, Cleveland,  Ohio).  Since it  was  not necessary to dry




the bark before extraction  (57) , the ground bark  was  packed  into a Soxhlet




thimble, placed in  a Soxhlet  extractor and  extracted  for 48  hours  with £-hexane.




The solvent was evaporated under aspirator  vacuum on  a rotary evaporator




(Buchi, Rotavapor,  Switzerland). The  residual  "wax-like" solid was  transferred




to a sample jar with a minimum of n-hexane  solvent.   The excess solvent  was




evaporated by passage  of a stream of nitrogen.  The ultraviolet absorption




spectrum of the extracted solids was measured in  a chloroform-n-hexane  (3:1




v/v) solution.




C.  Separation  of the  n-Hexane-Soluble Components




1.  Separation  by Column Chromatography




     Column chromatrography was  used as a first step  in  separating the




n-hexane-soluble materials.   A typical separation is  described belov?.  Silica




Oel G  was  packed  uniformly into  a glass column  two feet  in length  and one

-------
                                  -122-
inch in diameter.  A portion (2.00 g) of the r^-hexane-soluble fraction was




dissolved in chloroform (15.0 ml) and added to the top of the Silica Gel G




column.  Chloroform-n-hexane (3:1 v/v) was used as the developing solvent.




A series of bands were located by use of ultraviolet light.




     The fastest moving two bands (blue and light blue bands) were investigated




by H. Aft (personal communication).  The present investigation was concerned




with the next two fastest moving bands, the light bluish-green band and the




bright-blue band (designated in combination as the "yellow band").  To




collect these bands the fastest moving zones were first washed from the




column and stored.  Becuase of the difficulty of separating the next two




hands, they were washed from the column and collected as a mixture.  The




mixture was added to the top of a second Silica Gel G column and developed




with the same solvent system.  Then  the bright-blue band (fastest moving)




approached the sand in the bottom of the column, fractions of effluent  (5.0




ml) were collected.  These fractions were tested by thin-layer chromatography.




The fractions were tested in two different solvent systems:  diethyl ether-




n-hexane (1:4 v/v) and chloroform-carbon tetrachloride  (6:1 v/v).  The




chromatograms showed several spots for each fraction, indicating that the




column chromatographic separation was incomplete.




     However, as a source of starting meterial for further separations, the




"yellow band" was collected from several Silica Gel G columns.




     The Liebermann-Burchard test (59, p. 70; 60, p. 100) was employed  to




determine if the material contained  unsaturated sterols.  A sample of the




"yellow band" from the column chromatographic separation was dissolved  in




chloroform.  An aliquot (1.0 ml) of  the solution was added to a test tube




and dried.  Acetic anhydride (2.0 ml) was added followed by the dropwise




addition of concentrated sulfuric acid.  A dark green color change indicated

-------
                                  -123-
a positive test.


2.  Separation by Thin-Layer Chromatography


     The "yellow band" from the column chromatographic separation was


subjected to thin-layer chromatography in several solvent systems in an


effort, to more completely resolve the nixture.


     Solutions of the samples to be chromatographed were applied to the


thin-layer plates with capillary tubes in such a way that the surfaces of


Che plates were disturbed as little as possible.  The chromatography tanks


used for irrigating the thin-layer plates were saturated with vapors from


the developing solvent at least two hours before the plates were placed in


the tanks.


     Aliquots of the "yellow band" in chloroform-n-hexane (3:1 v/v) solution


were added to the thin-layer plates and developed with the following solvent


systems  (all hexane solvents were n_-hexane) :


     1.  Chloroform-carbon tetrachloride  (6:1 v/v)


     2.  Chloroform-carbon tetrachloride  (1:1 v/v)


     3.  Chloroform-carbon tetrachloride  (1:4 v/v)


     4.  Chloroform-hexane (4:1 v/v)

             '.' .. 'I
     5.  Ben^ene-hexane (2:1 v/v)


     6.  Diethyl ether-hexane (3:1 v/v)


     7.  Diethyl ether-hexane (1:4 v/v)


     8.  ;Chloroform-benzene  (6:1 v/v)


     9.  Chloroform-benzene  (1:1 v/v)


    10.  Chloroform-benzene  (1:4 v/v)


    11.  Benzene-methanol-acetic acid (45:8:4 v/v/v)


    12.  Hexane-diethyl ether-acetic acid  (70:30:1 v/v/v)


    13.  Hexane-diethyl ether-acetic acid  (85:15:1 v/v/v)

-------
                                  -124-
    14.   Chloroform-ethyl acetate-formic acid (5:4:1 v/v/v)




     Some thin-layer plates were developed two dimmensionally in the following




solvent  systems:




     ii-hexane-dlethyl ether-acetic acid (70:30:1 v/v/v) in one direction




followed by chloroform in the second direction; diethyl ether-n_-hexane (1:4




v/v) in one direction followed by chloroform-carbon tetrachloride (6:1 v/v)




in the second direction.




     After development, the thin-layer plates were dried at room temperature.




Those plates which were impregnated with silver nitrate were dried in the




dark.  The following methods were used to examine the developed plates.




     1.  Ultraviolet light




     2.  Ultraviolet light with the plate in ammonia vapors




     3.  Sprayed with 30% aqueous sulfuric odd, heated for 5 to 10 min




         at 105° in an oven and finally viewed under ultraviolet light.




     4.  Exposure  to iodine vapors.




     5.  Sprayed with bromothymol blue solution  (50 mj* bromothymol .blue,




         1.25 g boric acid, 8 ml  1 N  sodium  hydroxide and 112 ml water, or




         40 mg  bromothymol blue and 100 ml 0.01  N sodium hydroxide)




     6.  Sprayed with £-nitroaniline  solution  ((a)£-nitroaniline, 0.3%  in




         8%  (v/v)  HC1  (25 ml) and NaN02  (5%  w/v)(1.5 ml) mixed  immediately




         before spraying,   (b)  Na CO. (20% v/v)).   The two  solutions were




         used successively.




     Authentic  B-sitosterol  (Aldrich  Chemical  Co.,  Inc., Milwaukee, V.'isconsin)




was  chromatographed  by  one-dimensional  thin-layer chromatography simultaneously




with the "yellow band"  in  each  of the 14  solvent systems  listed above.  The




spots were located by  exposure  to iodine  vapor.  The separation by  every




solvent  systea  showed  a compound  in the  "yellow band"  which migrated  the

-------
                                  -125-
saire distance as the authentic B-sitosterol.




3.  Separation by Gas-Liquid Chromatography




     Gas-liquid chromatography was used to test the purity of the samples




separated by column chromatography and thin-layer chromatography and to




collect pure samples.  A Hewlett-Packard 5751B Research Chromatograph




(Hewlett-Packard Company, Palo Alto, California) were used.  Both instruments




were equipped with flame ionization detectors and used helium as the carrier




gas.  Stainless steel columns (1/8 in O.D.) were filled with the packing




material of choice.  A glass wool plug was inserted in the end of the




column and the material was packed to a uniform tightness by vibrating the




column with a small vibrating tool.  Four different column packings were




used in an attempt to find a column packing which would give good resolution.




Details of the four columns are listed in Table 7.




     The columns were put on the instrument for conditioning for 12 to 18




hr at a temperature of 10° above the maximum temperature used for resolution.




Various temperatures, flow rates, sensitivities and chart speeds were tested.




Both temperature programming and isothermal methods were used.




     Most of the samples were directly injected into the gas chromatograph




in iv-hexane solution.  However, because some of the compounds were thought




to contain hydroxyl groups the trimethylsilyl ethers of several of the




fractions were prepared in an attempt to improve resolution.  These




trimethylsilyl ethers were made by reacting dry samples for 5 to 10 min at




room temperature with hexamethyldisilazane and trimethylchlorosilane in




pyridine (2:1:10 v/v/v).  The pyridine solution was injected directly into




the gas chromatograph.




     The "yellow band" from column chromatography was investigated.  Four




different columns were used (Table 7).  The experimental conditions are

-------
                                  -126-
shown in Table 8.




     Two bands from the thin-layer chromatographic separation [developer,




diethyl ether-n_-hexane (1:4 v/v)] of the "yellow band" were tested for




purity by gas-liquid chromatography.  The two bands were readily observed




on the thin-layer plates by ultraviolet light.  The faster moving band




showed a yellow-blue color and the slower moving band showed a bright-blue




color.  These two bands were separately scraped from the thin-layer plates,




and the organic compounds extracted from the solid Silica Gel G with




r\-hexane.  The compounds were separated by gas-liquid chromatography on




the following column packings:  SE-32; US W-98; OV-17 (Table 10).  Both




silylated and non-sllylated samples were injected under the general




conditions outlined in Table 8.




     An authentic sample of 6-sitosterol was subjected to gas-liquid




chromatography using the columns and the conditions shown in Table 8.  The




retention times were compared with those for the peaks in the unknown.




     A "peak enhancement" study was conducted by adding a small quantity




of authentic 6-sitosterol to the n-hexane solution of the "yellow band"




and injecting the solution into the gas chromatograph under the conditions




outlined in Table 8.




     A known sample of stigmasterol (Aldrich Chemical Co., Milwaukee, Wisconsin)




was chromatographed on SE-52 and OV-17 columns under the conditions outlined




in Table 8.




A.  Identification by Gas-Liquid Chromatography Combined With Rapid-Scan




    Mass Spectrometry




     Gas-liquid chromatography in combination with mass spectrometry was




used to separate and identify a range of compounds present in the n-hexane-




soluble fraction of Douglas-fir bark.  The following operating conditions

-------
                                  -127-
were used;




Gas-Liquid Chroma t ography




     Instrument




     Detector




     Detector temperature




     Injection sample




     Injection port temperature




     Column




     Column temperature




     Carrier gas flow rate




Mass Spectrometry




     Filament current




     Electron voltage




     Analyzer pressure




     Multiplier voltage




     Scanning speed
                                    F & M 810 (Hewlett-Packard)




                                    Hydrogen flame ionization




                                    248°




                                    4 yl of the sllylated mixture




                                    270°




                                    UC W-98




                                    Isothermal at 265°




                                    30 ml/min of helium








                                    70 eV source, 40 MA




                                    20 eV and 70 eV




                                    1 x 10~6 mm Hg
                                    3.00 KV




                                    6.5 seconds from m/e_ 24 to 500




     The mass spectrometer used was an Atlas CH-4, Nier-type (nine inch, 60




degree sector) single-focusing instrument.  The trimethylsilyl ethers of the




fractions of interest were separated by gas-liquid chromatography and a




portion of the effluent was passed directly into the dual ion source of the




mass spectrometer.  The gas-liquid chromatograph was fitted with a 5:1 (EC-1




valve:flame) splitter.  The mass spectrometer was equipped with an EC-1




throttle valve which was adjusted to permit approximately 10% of the remaining




column effluent to enter the ionization chamber, the remaining effluent being




vented into the air through a heated tube.  The effluent was further split




in the ion source with 50% going to the 70 eV source.  The 20 eV source




operated at less than the ionization potential of the carrier gas (helium),

-------
                                  -128-
but above that of organic compounds, and was therefore used as a continuous




total ionization readout without any contribution from ionized helium.  The




70 eV source provided the ionization used to obtain the mass spectra.  The




spectra were recorded on a Honeywell 1508 oscillograph.




111.  Experimental:  Benzene-Soluble Wax (78, 79)




A.  Bark Collection and Solvent Extraction




     Whole bark was stripped from the bottom end of a freshly-felled, 58




year old Douglas-fir at Black Rock, Oregon.  The bark was air-dried  (moisture




content 9.3%), ground to pass a 1/2-inch screen, and a portion (768.2 g dry




cut) was packed into a stainless-steel mesh basket and extracted in a large




Soxhlet extractor with n-hexane (8.0 liters) for 52 hr.  The extract was




filtered while hot through glass wool and the ii-hexane evaporated on a




rotary evaporator at 30° or less leaving a cream-colored, "wax-like" solid;




yield, 30.0 g.  This n-hexane "wax" was stored for future investigations.




     The bark residue from the n-hexane extraction was air-dried, re-packed




into the stainless-steel mesh basket and extracted with benzene  (8.0 liters)




in the Soxhlet extractor for 52 hr.  The extract was filtered, and




concentrated as above.  The final traces of benzene were removed by  freeze-




drying, leaving a light-brown,  "wax-like" solid; yield 20.0 g.




B.  Saponification and Separation of Acids and Neutrals




     An aliquot  (10.0 g) of the benzene "wax" was saponified by  gently




refluxing for 3.75 hr in 10% NaOH (250.0 ml).  The mixture was cooled in




an ice bath and acidified with glacial acetic acid resulting in  a




precipitate.  The organic compounds were recovered in two ways.  In  the




first procedure the mixture was extracted with benzene without first




collecting the precipitate.  In the second procedure the precipitate was




recovered on a filter, and then only the precipitate was extracted with benzene.

-------
                                  -129-
     The benzene solution, containing both acids and neutrals, from both



procedures was extracted in a beaker with 5% NaOH with warming and stirring.



The aqueous layer was recovered and washed several times with fresh benzene.



The benzene solution including washings, containing the neutrals, were



washed first with dilute HOAc, then with water and dried with anhydrous



Na-SO..
  24


     The aqueous 5% NaOH fraction containing the acids was cooled in an ice



bath, acidified with glacial acetic acid and extracted with benzene.  A



little diethyl ether was added to the benzene to break emulsions.  The benzene



solution was dried over anhydrous Na~SO  and the solvent removed on a rotary



evaporator at 30° or less leaving a tan-colored solid.



C.  Preparation of Methyl Esters



     The methanol used was reagent grade which had been doubly distilled



through a short Vigreax column and checked by GLC with SE-30 as stationary



phase.



     The acids Isolated above were dissolved in MeOH  (400.0 ml) containing



5-6% gaseous HC1 by weight.  The brown solution was refluxed for 2-4 hr,



cooled under tap water, concentrated to approximately 20 ml on a rotary



evaporator and diluted with water (50-55 ml), resulting in a white precipitate.



The mixture was extracted 3 times with 100 ml portions of doubly distilled



n-hexane.  The rv-hexane extract was washed with 1% NaHCO,, with water, and



then dried with anhydrous Na~SO,.  The ii-hexane extract was removed on a



rotary evaporator leaving a tan-colored'solid.



D.  Thin-Layer Chromatography (TLC)



     The methyl esters isolated above were dissolved  in benzene, diluted to



25 ml and aliquots were subjected to TLC on layers of pre-washed Silica Gel



G using n-hexane-diethyl ether (70:30 v/v) as solvent and a spray reagent of

-------
                                  -130-
0.2% 2',7'-dichloroFluorescein as indicator.  Authentic standard methyl esters




were developed simultaneously with the unknown compounds.




     Three major spots or "families" were evident.  Spot No. 1 (R 0.74)




migrated essentially the same distance as authentic methyl lignocerate (R 0.76)




indicating that it was composed of non-oxygenated fatty acid methyl esters.




Spot No. 2 (R 0.54) migrated similarily to authentic dimethyloctadecanedioate




(Rf0.54) indicating that it was composed of dicarboxylic fatty acid dimethyl




esters.




     Spot No. 3 (Rf0.17) was not readily identified by comparison with known




standards but TLC and gas-liquid chromatographic  (GLC) data, as well as




properties of a derivative, leads to the belief that it contains hydroxy-acids.




     To  isolate the methyl ester "families" for further resolution by gas-




liquid chromatography the methyl ester solution was applied as closely spaced




spots  to each of five TLC plates and developed as above.  A narrow band on




each plate edge was sprayed with 2' ,7'-dichloro-Fluorescein, and the whole




plate  briefly exposed to iodine vapors.  As soon  as the iodine color disappeared




the spots were scraped from the plates and  the organic compounds eluted by




repeated washings with benzene.  The eluents were concentrated to 1 ml or




less for GLC.




E.  TLC  of the Neutral Fraction




     The neutral fraction of benzene "wax"  was subjected to TLC under the




same condition as described above.  No major spots resulted, spot A (R 0.30)




and spot B (R 0.07).  Spot B migrated the same distance as authentic 6-




sitosterol.  In addition,, the reaction of the following sprays used for the




detection of steroids were noted:   85% phosphoric acid - H^O (1:1 v/v);




Liebermann-Burchard Reagent; Stahl  SbCl_-HOAc (1:1 v/v).




F.  GLC  of the Fatty Acid Methyl Esters

-------
                                  -131-
     All GLC was performed on a Hewlett-Packard Model 5751B equipped with




dual flame ionization detectors, a Model 17503A input module, a Model 7127A




strip chart recorder, and a Series 300 DISK Integrator.




     Column packings were prepared by coating Gas-Chrom Q (100/120 mesh) with




the desired liquid phase and calculating the loading (w/w) by the filtrate




measurement technique.  All columns were made from 1/8 in O.D. stainless




steel tubing.  Specific columns used were:  (a) 5 ft, 4.75% Sillcone SE-30;




(b) 2.5 ft, 8.25% Apiezon-L; (c) 3 ft, 7.4% Reoplex-400;  (d) 6 ft, 11.1%




EGGS-X.  Various conditions and temperatures were used as specified in the




results.  Carrier gas (helium) flow rates were approximately 30 ml/min.




     The methyl esters were injected into the GC in ri-hexane or benzene




solutions.  Identification of GLC peaks were made by comparing retention




times of the "unknown" peaks with a standard plot of chain-length vs. log




retention time prepared from authentic standard compounds.  Standard compounds




were also added to the unknown samples for "peak enhancement" studies.




     Peaks due to unsaturated compounds were located by comparing GLC's of




bromine-treated samples with untreated samples.




     GLC of standard mixtures of saturated and unsaturated methyl esters and




of saturated diesters showed that peak areas very closely reflected the




weight composition ratios of each series.  In this way the ratios of the




compounds of the two spots from TLC were calculated by comparing peak areas.




G.  GLC Combined with Rapid-Scan Mass Spectrometry




     Mass spectra (MS) of the compounds'of Spot No. 2  (TLC) were obtained




by tanden GLC-MS.  GLC was performed on an F&M 810 (Hewlett-Packard) with




the SE-30 solumn under conditions close to those already described.  One-




fifth of the column effluent was fed to the GLC flame ionization detector,




while the remainder was passed to the heated EC-1 inlet of an Atlas MAT CH-4

-------
                                  -132-
Nier type mass spectrometer.  The 20eV source provided constant monitoring




of the organic compounds reaching the MS, while the 70eV source provided




ionization for the spectra obtained by rapid magnetic scan.




H.  Identification of the Alcohols




     Identification of the alcohols of Spot A (section I1I-E) was made by




co-GLC of the free alcohols (4.75% SE-30 and 7% Carbowax 20M) with authentic




compounds, and by co-GLC (SE-30) of the acetates (Ac_0 in pyridine with toluene-




sulfonic acid), methyl ethers (CH I and Ag 0) , and TMSi ethers (HMDS and TMcS




in pyridine) with the respective derivatives made from authentic compounds.




     The €„„ and C0, alcohols from benzene wax were collected from a




preparative SE-30 solumn and identity was further verified by IR and m.m.p.




1.  1-Docosanol




     1-Docosanol from benzene wax had m.p. 69.5-71°, which was not depressed




by mixture with approximately equal amounts of authentic behenyl alcohol; IR




(KBr), cm"1: 3340 (broad) with shoulder ca 3230 (-OH stretch, H-bonded) , 2910




and 2840  (alkyl), 1470 and  1460  (alkyl) , 1056 (primary -OH), 723 and 712




(-(CH,,) -) , and identical to spectrum of authentic behenyl alcohol.  Acetate
     i n


IR  (KBr), cm'1: 2915 and 2850 (alkyl stretch), 1745  (ester carbonyl) , 1465
 (alkyl),  1365  (-CHj) ,  1235  (C-O-C for acetate), 1040 (C-O-C) , 715  (-(CH2)n-) ,




 and identical  to spectrum of acetate prepared from authentic alcohol.




 2.  1-Tetracosanol




     1-Tetracosanol  from benzene wax had m.p. 76.0-76.5°, identical  to that




 of authentic lignoceryl alcohol; IR  (KBr K... cm"  : 3330  (broad) with shoulder
                                          ~"V- j



 C£ 3230  (-OH stretch,  H-bonded), 2910 and 2840  (alkyl), 1470 and 1460 (alkyl),




 1056 (primary  -OH),  722 and 712  (-(CH^-) , and identical to a spectrum of




 authentic lignoceryl alcohol.

-------
                                  -133-
I.   Identification of the Hydroxy Acids



     Hydroxy methyl esters and the methoxy methyl esters were collected



from a preparative SE-30 column for further analysis.



1.   Methyl 16-hydroxyhexadecanoate



     Methyl 16-hydroxyhexadecanoate from benzene wax had m.p. 54-55° (benzene)



(lit. 54.5-55.5°); IR (KBr), cm'1: 3400 with shoulder c£ 3320 (hydroxyl),



2920 and 2850 (alkyl), 1740 (ester carbonyl), 1470 and 1463 (alkyl), 1435



(-CH , methyl ester), 1362, 1340, series of small bands 1190-1300 (-ClU-



wag and twist), 1160 (methyl ester C-O-C), 1060 and 1045 (C-0), 880 (-CH -COOCH ),



730 and 720 (-(CH_) -rock).  (Found: C, 71.5; H, 12.0.  Calc. for C.,H,.0,:
                 i n                                               17 34 3


C,  71.3; H, 12.0)



     Treatment with HI and red P followed by Zn and HC1, and methylation



(MeOH - HC1) gave one product whose retention time (SE-30 and EGSS-X) was



the same as that of authentic methyl hexadecanoate, and the MS of the product



showed the ion fragments expected for methyl hexadecanoate.



     Oxidation (CrO- in HOAc) gave a product whose methyl ester had the same



TLC R , the same R  (SE-30 and EGSS-X), and the same MS as authentic dimethyl



hexadecanedioate.


                                                   — 1
     Methyl 16-methoxyhexadecanoate gave IR (KBr)cra   : 2995 (weak), 2915, 2855



and 2815 (weak) (alkane), 1740 (ester carbonyl), 1475 (alkyl), 1440, 1380, 8



sharp bands of varying intensity 1345-1203, 1170 (methyl ester C-O-C), 1110



implies ether C-O-C), 940, 880, and 715 ((CH.) ).  GLC-MS, m/e: 300 (M+), 285
                                            / n


(M-CH3), 270 (M-30), 269 (M-31), 268 (M-32), 253 (M-47), 241  (M-59), 236 (M-64),



208  (M-92); also the following fragments (which were common to all homologs):



143, 129, 115, 101, 87, 74 (methyl ester), 55, and 45 (CH OCH +, base peak).



2.   Methyl 18-hydroxyoctadecanoate



     White crystals from hexane had m.p. 62-63° (lit. 62.5°).  TR (KBr) was

-------
                                  -134-
identical to a published spectrum:  3390 cm   with shoulder at ca 3330 (-OH),



2910 and 2840 (alkane), 1740 (ester carbonyl), 1475 and 1468 (alkane), 1440



(-CH , methyl ester), 1370 and 1355, 1330.  Series of small bands 1180-1310



(-CH2 - wag and twist), 1170 (methyl ester C-O-C), 1055 (C-0), 883 (CH  - COOCH-),



730 and 720 (-(CH~)  - rock).  The MS of the methoxy methyl ester was
                 i n


identical to a published spectrum of methyl 18-raethoxyoctadecanoate:   m/e



328, 313, 298, 297, 296, 281, 269, 264, 236 (assignments as for C 6 homolog);



m/e 143 and lower was identical to that for the C., homolog.



3.  Methyl 20-hydroxyeicosanoate



     White crystals from hexane had m.p. 69.0-69.5° (lit. 68-68.5°);  IR (KBr),



cm"1: ca 3330 (broad, -OH, H-bonded), 2910 and 2840 (alkyl), 1740 (ester



carbonyl), 1475 and 1465 (alkyl), 1438  (-CH , methyl ester), 1385, series of



8 small bands 1310-1185 (-CH- - wag and twist), 1175 (methyl ester C-O-C),



1055 (C-0), 880 (-CH0-COOCH,), 728 and  715 (-(CH.) -).
                    /      J                    / n


     Reduction of the hydroxyl group gave a product whose methyl ester had



the same R  as authentic methyl eicosanoate (SE-30, EGSS-X); the MS contained



ion fragments expected for methyl eicosanoate.  Oxidation  (CrO., in HOAc)



followed by methylation gave a product  whose TLC R , R  (SE-30, EGSS-X), and



MS were identical to those of dimethyl  eicosandeioate isolated previously



from benzene wax.



     The methoxy methyl ester gave a MS similar to that of  the lower homologs:



m/e 356, 341, 326, 325, 324, 309, 297,  292, 264  (assignments as for C,£
                                                                     16


homolog); m/e 143 and lower was identical to lower homologs.



4.  Methyl 22-hydroxydocosanoate



     White crystals from hexane had m.p. 74.5-75.5° (lit.  73.5-74.5°).  IR



(KBr) was similar to that of the  lower  homologs; there was  a series of 9



small bands 1315-1185 cm"  (-CH2  - wag  and twist).

-------
                                  -135-
     Reduction of the hydroxyl group gave a product whose methyl ester had




the same R  (SE-30, EGSS-X) as authentic methyl docosanoate; the MS contained




ion fragments expected for methyl docosanoate.  Oxidation (CrO. In HOAc)




and methylation gave a product whose TLC R , R  (SE-30. EGSS-X) and MS




were identical to those of authentic dimethyl docosanedioate.




     The methoxy methyl ester gave a MS similar to those for the lower




homologs: m/e 384, 369, 354, 353, 352, 337, 325, 320, 292; fragmentation




m/e 143 and below was identical to that for lower homologs.




5.  Methyl 24-hydroxytetracosanoate




     White crystals from hexane had m.p. 80-81° (lit. 80-81°).  IR (KBr) was




similar to that of the lower homologs; there was a series of 10 small bands




1310-1180 cm"1 (-CH  - wag and twist).









IV.  Results and Discussion




     Since a reasonably representative sample of Douglas-fir bark was desired,




small amounts were taken from a number of trees in the area of Corvallis,




Oregon.  Because of the number of trees sampled and because these samples




were thoroughly mixed into a larger collection, no effort was made to




ascertain the diameter or age of the  trees.




     The n-hexane extraction closely  followed the method of Kurth and Kiefer




(71).  These authors reported the n-hexane-soluble material to be a light




colored "wax," whereas other solvents such as benzene extracted dark colored




materials.  The light color indicated a'less complex fraction which was more




desirable for an initial study.




     Column chromatography on Silica Gel G was used to provide a coarse




separation of the ii-hexane-soluble compounds.  A number of distinct bands




were evident as the column developed.

-------
                                  -136-
     The following bands labelled "light bluish green" and "bright blue"




(designated in combination as the "yellow band" was eluted from the column




and tested for purity by thin-layer chromatography.  The better resolving




power of the thin-layer technique showed that the "yellow band" was a




mixture of several chemical compounds.  Some thin-layer plates showed as




many as 11 distinct spots plus a sizable spot which remained at the origin.




The spot at the origin contained an undetermined number of compounds.




     A solution of the "yellow band" showed a positive Lieberman-Burchard




test (73, 74) indicative of the presence of sterols.




     A quantity of the "yellow band" was accumulated from several columns




and the mixture was further separated by thin-layer and gas-liquid chromato-




graphy.




     The extensive thin-layer chromatographic study was conducted to find




out how many compounds were in the "yellow band" and to find a solvent




system which gave good separation which could be used as a method to collect




pure compounds.




     The 14 solvent systems employed are widely used for the separation




of lipids as well as for sterols and terpenes.  The solvent systems composed




of chloroform-carbon tetrachloride (6:1 v/v), diethyl ether-n-hexane (1:4




v/v), n-hexane-diethyl ether-acetic acid (70:30:1 v/v/v) were the best.  Of




these three, chloroform-carbon tetrachloride (6:1 v/v) produced more spots




than the other two, but each spot had considerable tailing.  The diethyl




ether-n_~hexane (1:4 v/v) system gave very good separation and although the




chromatograro showed fewer spots than the solvent system of chloroform-carbon




tetrachloride (6:1 v/v) the amount of tailing was reduced.  The addition of




acetic acid to the ether and n-hexane solvent systems [n-hexane-diethyl




ether-acetic acid (70:30:1 v/v/v), and ji-nexane-diethyl ether-acetic acid

-------
                                  -137-
(85:15:1 v/v/v)] did not show improvement.  Solvent systems with higher




polarity such as benzene-methanol-acetic acid (45:8:4 v/v/v) and chlorofora-




diethyl ether-formic acid (5:4:1 v/v/v) showed poor separation and fewer




spots could be observed than with the less polar solvent systems.




     Exposure of the thin-layer plates to iodine vapor proved to be the




best method of detecting the spots.  It was the easiest to use and showed




more spots than the other methods tried.  Simple exposure to ultraviolet




light was the second best method of spot detection.  The combination of




ammonia vapors and ultraviolet light did not show any improvement over the




use of ultraviolet light alone.  The other methods, 30% aqueous sulfuric




acid, bromothymol blue spray, and p_-nitroaniline spray showed no improvement




over the iodine vapor method and were more difficult to use.




     Four columns (Tables 7 and 8) were used in an attempt to completely




resolve the compounds in the "yellow band" from the column chromatograph.




     The Hi-EFF 8 BP column did not give good resolution.  Even with




temperature programming, it showed only five peaks.  The other three columns




showed no great difference in resolving power.  Each of these columns showed




as many as 16 peaks and the UC W-98 column showed 19 peaks.




     Separation of the "yellow band" from the column chromatograph by




thin-layer chromatography prior to injection into the gas chromatograph




yielded improved separation.  The "yellow-blue band" (high R ) from the




thin-layer plates was shown to contain two compounds by gas-liquid chromato-




graphy.




     However, there were no peaks by gas-liquid chromatography which could




be attributed to the "bright-blue band" (low Rf) from the thin-layer plates.




It may be that the "bright-blue band" contained high-boiling compounds




which did not elute under the conditions used.

-------
                                  -138-
     The results of gas-liquid chronatop.raphy showed that at least 20




compounds were jammed into the "yellow band" from column chromatography.




Nineteen of these were resolved by the DC W-98 column and there was a




minimum of one in the "bright-blue band" from the thin-layer plates which




dlil not show from the DC W-98 column making a minimum of 20 compounds.




These materials must be similar in chemical nature and structure.  This




further increases the difficulty of separation and purification.  The




identification of every compound shown by gas-liquid chromatography is




often impossible, mainly because of the time and difficulties involved




and the ninute amounts of material which can be detected.




     However, the combination of thin-layer chromatography and gas-liquid




chromatography produced evidence fcr the identity of some of the compounds.




It is concluded  from the thin-layer and gas-liquid chromatographic evidence




that the  n-hexane-soluble fraction from Douglas-fir bark contains campesterol




and fl-sitosterol.




     A  portion of the gas effluent from the gas-liquid chromatographic




column  was passed directly  into the ion-source of the mass spectrometer.




The remaining portion of the  gas effluent from the column was passed  -.hrough




a flarat ioniz.\tion  detector which produced an electrical signal which was




recorded  on  an X-Y  strip recorder.  In  this way when a peak  appeared  on




the recorder, a  portion of  the compounds causing  the peak was also in the




nans spectrometer  and a rapid scan by  the mass spectrometer  produced  a




fragmentation spectra  for  that compound.  Thus, the gas-liquid  chromatograph




separated the mixture  into  pure compounds and the mass  spectrometer  yielded




information  as  to  the  identity of  each  pure  compound.




     An aliquot  of  the  trimethylsilyl  ethers of  the known campesterol and




K-sitosterol mixture was  injected into the  gas chromatograph.   As the peaks

-------
appeared on the recorder, a scan was made by the mass spectrometer.  In this

way standard mass spectra were obtained for known B-sitosterol and campesterol

for reference purposes.

     In the same way, the trimethylsilyl ethers of the "yellow band" were

injected into the gas chromatograph.  The gas-liquid chromatographic

spectrum was scanned by mass spectrometry.  In the mass spectrum of the

trimethylsilyl ether of B-sitosterol, characteristic peaks were found at

m/e 486, 396, 381, 357, 255 and 129,  The base peak at m/e 129 s which is

a characteristic fragment for a 5-en-3-ol steroid trimethylsilyl ether was

very obvious.  In the fragmentation reaction the triraethylsilyl ether

derivative is split into two parts, one part which is charged and the other

which is neutral.  The charge can reside on the m/e 129 fragment or the

m/e M-129 fragment (M is the molecular ion mass).  Therefore, both fragments

are recorded, but usually the intensities are different.

     The mass spectra were not as conclusive for campesterol as might be

expected from the retention times.  However, peaks at m/e 129 and 255, which

are characteristic for sterols, were observed.  These spectra  indicated

mixtures of terpenes and sterols.

     The mass spectral data support the gas-liquid chromatograhic retention

data as to the presence of campesterol.  The ratio of campesterol to

B-sitosterol is about 1 to 8 as estimated by areas under the gas-liquid

chromatographic curves.

     The identification of campesterol and B-sitosterol in the n-hexane-soluble
                                       (
fraction of Douglas-fir bark has proven very interesting because it leads to

a search for other sterols and the part which they might play  in bark.  The
          I
physiological role of these materials is largely unknown, although Rowe  (70)

has suggested that B-sitosterol may have a function in cell-wall permeability.

-------
                                  -140-
     The mass spectra also showed peaks at m/c 93 and m/e 121 whicli are very




characteristic of terpenes and it is concluded that this particular fraction




of the bark contains terpene compounds.




     The acids recovered from saponified "benzene wax" could be resolved




into three major families by TLC of their methyl esters.  Spot 1 (R  0.74)




migrated nearly the same distance as authentic methyl lignocerate (Rf 0.76)




suggesting that it was composed of methyl esters of non-oxygenated fatty




acids.  Spot 2 (R  0.54) migrated similarly to authentic dimethyl octadecane-




dioate  (R  0.57) suggesting that it was composed of dimethyl esters of




dicarboxylic acids.  Spot 3 (Rf 0.17) was not readily identified by standards




in our possession, but further investigation suggests that it contains the




methyl esters of the hydroxy acids observed by Kurth.




     TLC of the neutral compounds from saponified  "benzene wax" yielded two




major components, spot A  (R, 0.30) and spot B (Rf  0.17).  The latter spot




reacted positively to steroid-detection reagents and migrated the same




distance as authentic 3-sltosterol.  These spots could be the aliphatic




alcohols and "phytosterols" observed by Kurth (72).




     CLC resolved the compounds of TLC spot 1 into a number of  fatty acid




methyl  esters  (Table 9),  identifications of which  were based on studies with




four GLC liquid phases.  The ester of greatest abundance was methyl lignocerate,




but substantial amounts of several other esters were present.   Several of  the




GLC components of spot 1 remain unidentified.




     The dimethyl esters  identified  in' TLC spot 2  are listed, with their




relative abundances, in Table 10.   Initial identifications were based on GLC




studies made with three liquid phases:  SE-30, Apiezon-L, and Reoplex-400.




Besides the C  „ compound  reported by Kurth (72) we found substantial amounts




of other dimethyl esters.

-------
                                  -141-
     Verification of the identities of the spot 2 dimethyl esters was obtained



by tandem GLC-MS.  We examined the high ends of the mass spectra of authentic



dimethyl esters and of the GLC peaks in spot 2 for those m/e ion peaks



characteristic of dimethyl esters of dicarboxylic acids:  M, M-31, M-64, M-73,



M-92, and M-105.  We positively identified all of these m/e peaks for each of



our authentic compounds except for £„., where the peak believed to be M could



not be accuratley assigned an m/e.  For the C , and C__ dimethyl esters of



spot 2, we identified all the above ions; for C~?, all except M; for C ,,



only M-31 and M-73.  The spectra of the C g   and Clg , dimethyl esters were



of mixtures of the two, since resolution was incomplete at the mass spectro-



meter.  However, M, M-31, M-64 and M-73 were identified for the C g _ dimethyl



ester, and M = 340 m/c and (M-CH OH) - 308 m/e were identified, which indicates



the presence of a C10 .. dimethyl ester.  The MS taken at various points on
                   lo'. i


the total-ionization peak showed that the unsaturated diester emerged from



the GLC first.  This is consistent with the GLC studies.  No obvious signs



of branching were noted in any of the spectra.



     Gas chromatopraphic analysis (SE-30) of the TLC spot thought to contain



fatty alcohols showed two major components which were identified as 1-



docosanol and 1-tetracosanol; smaller amounts of 1-hexadecanol, 1-octadecanol



and 1-eicosanol were also present (Table 11).  Earlier, Kurth  (75) isolated



1-tetracosanol from the neutral fraction of saponified benzene wax.



     Identification was established by co-GLC of the free alcohols of benzene



wax with authentic compounds  (SE-30 and'Carbowax 20M) and by co-GLC (SE-30)



of the alcohol acetates, methyl ethers, and TMSi ethers with the respective



derivatives prepared from authentic compounds.  The IR spectra (KBr) of C~_



and C_, free alcohol and alcohol acetate peaks collected were identical to



those of the authentic compounds.  In particular, the -OH stretch bands at

-------
                                  -142-
££ 3330 cm   present in spectra of the free alcohols were absent in spectra
of the acetates; in the latter, the ester carbonyl absorption at 1745 cm
appear.  Melting points of the C._ and C_, alcohols from benzene wax agreed
with those of authentic compounds, and mixed m.p.'s with respective authentic
compounds showed no depression.
     GLC (SE-30) of TLC spot 3, thought to contain methyl esters of hydroxy
acids, showed it to contain a family of six compounds for which a plot of
chain length (arbitrary integer) vs. log R  was a straight line.  Five of
these compounds were isolated by preparative GLC and identified as the methyl
esters of w-hydroxy acids C.., - C_,  (even); not enough of the sixth compound
(C0,?) was present to conveniently collect for positive identification.
  Zo
Kurth reported  the presence of C , , C__ and C   hydroxy acids in benzene
wax, but though he noted the unusual melting characteristics of the free
hydroxy behenic acid, he did not locate the hydroxyl group (76).
     Also present in the chromatogram, but unidentified, were two peaks
emerging before w-hydroxy C.,, and a peak  (peak "2a") only partially resolved
from w-hydroxy  C,0  (Table 12).  The Eiethyl ether  of peak 2a emerged after
                lo
w-hydroxy C,_ on EGSS-X, and contained at  least two major components.  The
           lo
GLC behavior of peak 2a8 as well as  the fact that it was liquid at room
temperature, suggest that it may contain unsaturated C,0 w-hydroxy methyl
                                                      lo
esters.  Kurth  claims to have  isolated a fraction rich in hydroxy unsaturated
acids from the  cork of Douglas fir bark.
     Treatment  of the mixed w-hydroxy esters from benzene wax with alkali in
a manner similar to saponification of the  wax, followed by methylation did
not produce any detectable dicarboxylic esters hence we are confident that
the dicarboxylic acids identified earlier  by us are not artifacts of the
saponification  procedure.  That w-hydroxy  acid and dicarboxylic acid families

-------
                                  -143-
of similar composition were found in benzene wax may be of biosynthetic




significance.




     Though not as widespread or abundant as the hydroxy acids In general,




u>-hydroxy acids as constituents of plant waxes and extracts have been known




for some time.  The inter-esterification of the free hydroxy acids to form




"estolides" occurs readily, especially in acid medium or upon heating above




the melting point, and may partly explain the variable melting point observed




by Kurth (76) for his hydroxy-behenic acid.  Murray and Schoenfeld considered




estolide formation upon acidification of a saponification mixture of w-hydroxy




acid salts enough of a problem to avoid isolating the free acids during their




qualitative and quantitative analysis of the co-hydroxy acids of carnauba wax.




Estolide formation could partly explain our inability to totally account for




the weight of benzene wax after isolation of neutrals, methyl esters, and




"phenolic acids"; and it could explain the somewhat unreproduceable relative




abundances of w-hydroxy esters between complete isolations of the esters from




the wax.  The relative abundances shown in Table 12 should be considered only




tentative and typical of our Isolation procedure.




     Identity of the w-hydroxy acids was established by analyzing individual




hydroxy methyl esters and the corresponding methoxy methyl esters collected




by preparative GLC.  The IR spectra (CC1.) contained a weak, sharp band at




3620 cm  .indicating a hydroxyl group; this band was absent in the IR




spectra of the corresponding co-hydroxy methyl esters.  The IR spectrum (KBr)




of methyl 18-hydroxyoctadecanoate was identical to published spectrum, and




the spectra of the other homologs were similar.  The most conspicuous




difference between homologs was the addition of one progression band in the




region between 1180 and 1315 cm   for each increase of two methylene units.




     The lengths of the hydrocarbon chain of the C..,, C^n and C_9 homologs

-------
                                  -144-
were established by reduction of the hydroxyl group and identification of



the resulting fatty acids (as methyl esters) by GLC (two columns) and by



GLC-MS.  Hydroxyl group positions for the C.,, C . and C__ homologs were



established by oxidation of the hydroxyl group and identification of the



resulting dicarboxylic acids (as methyl esters) by GLC (two columns) and



by GLC-MS.



     The mass spectra of methyl 18-methoxyoctadecanoate agreed with a



published spectrum, and the spectra for the C.,, C   and C00 homologs were
                                             lo   zu      if.


similar.  In all cases, the portions of the spectra m/e 143 and below were



identical.  The peak m/e 45, the strongest in the spectra, is mainly due to



the fragment CH_OCH_ and indicates the methoxy is in the terminal (to)



position.  The high m/e^ portions of the spectra all shoved significant peaks



assigned as M+, (M-CH3), M-30, M-31, M-32, M-47, M-59, M-64 and M-92.



     Finally, the melting points of the C-,-C_, to-hydroxy esters agreed



with values reported in the literature.

-------
                                  -145-
                           AMMONIATION OF BARK




I.  Historical Review (80)




     The use of ground bark for mulching, soil conditioning,  and ornamental




purposes in home gardens and elsewhere has greatly increased  in recent years




(1, 2, 3, 81, 82).  A major consideration in these applications is to




ensure that the addition of bark does not lower the nitrogen  content of the




soil.  Bark contains very little nitrogen and the microorganisms which bring




about its decomposition tend to compete with plant roots for  the available




nigrogen (83, 84).  According to Bollen (85) the fortification of bark with




nitrogen is vital to the successful use of bark on or in the  soil.  Treatments




include chemical processing, mechanical mixing with nitrogen  fertilizers,




aging, composting and combinations of treatments.  The work herein reported




is concerned with fortification by chemical processing and with gaseous




ammonia in particular.




     There have been several reports on the subject (3, 83, 84, 85, 86, 87).




Aspitarte  (86) and Bollen and Glennie (83) exposed Douglas-fir bark to




anydrous ammonia in a closed screw conveyor system.  Bollen and Glennie (84)




also sprayed ground bark with aqueous ammonia, nitric acid, an aqueous solution




of urea, and an aqueous solution of urea and phosphoric acid.  Simpler




procedures have involved covering a pile of bark with a polyethylene sheet




and passing gaseous ammonia through it from a perforated pipe or hose  (85, 88).




     Our objectives in the present study included a determination of the




amount of nitrogen retained by Douglas-fir bark after treatment with gaseous




ammonia under controlled conditions.  We also wished to ascertain how readily




this nitrogen was extracted with selected solvents and if any nitrogen




remained after exhaustive extraction.  We investigated the extracts to




determine if the ammoniation altered the chemical compounds which are extracted

-------
                                  -146-
by organic solvents and water.  Since these compounds are most likely to be




leached into the soils we particularly wanted to determine if any known




harmful products resulted.




II.  Experimental (80)




A.  Sample Preparation




     Samples of bark from Douglas-fir [Pseudotsuga menziesii (Mirb.) Franco]




were collected near Corvalllss Oregon.  They were thoroughly mixed, ground




in a ball mill to a size suitable for research investigations and air dried;




moisture content, 15.8%  (oven at 110°C overnight); nitrogen content, 0,46%




(Kealdahl).




B-  Bark Extraction




     An aliquot  (21.04 g) of  the bark was successively extracted in a Soxhlet




apparatus  (50 exchanges  of solvent) with ii-hexane, benzene, diethyl ether,




95% ethanol and hot water.  An ultraviolet spectrum of each extract was




obtained over a range of 230 nm to 350 nm.  Each extract was concentrated




under vacuum on a rotary evaporator, dried in an oven at 110°C overnight, and




the solids weighed.




C,  Treatment with Gaseous Ammonia




     An aliquot  (233.26  g) of the air-dried bark was loosely packed  into a




glass tube to provide a  sample 4.6 cm in diameter and 100.0 cm In  length.




Typical conditions for ammonia treatment were:  tank gauge pressure, 10 psl,




time of gaseous ammonia  flowB 2 hr;  total time in ammonia atmosphere, 17 hr;




maximum temperature  in sample, 68°C; nitrogen content of air-dried bark after




treatment9 4.08%.




     An aliquot  (25.00 g) of  the treated bark after air drying was successively




extracted with n-haxane, benzene, diethyl ether, 95% ethanol and hot water




and the extracts investigated as described above for those from the original




bark.  The nitrogen content of the bark residue after extraction was 2.98%.

-------
                                  -147-
III.  Results and Discussion (80)

     The sample of Douglas-fir bark used in the present treatment contained

0.14% nitrogen.  This is similar to the 0.11% and 0.16% reported by Bollen

and Glennie (86).  After treatment with gaseous ammonia the nitrogen content

Increased to 4.08% which is in the range of 3.71% and 4.27% obtained by

Bollen and Glennie (83) for bark ammoniated under selected conditions.

This amount of nitrogen has been shown (83, 84) to be more than adequate

for the prevention of nitrogen starvation in soil.

     Ammoniation of finely ground bark (passed a screen of 32 meshes to the

inch) resulted in a nitrogen content to 2.34%.  The lower nitrogen retention

was attributed to a more solid packing of the fine particles in the treating

tube so that the ammonia gas may not have diffused freely through the sample.

A loose packing of the bark or an increased treating time would be expected

to increase the nitrogen content.  This illustrates the flexibility of

ammonlation with gaseous nitrogen.  By careful control of treating conditions,

it is possible to tailor the nitrogen content of bark to the desired content.

Bollen and Glennie (83) also demonstrated this control.  They obtained

nitrogen contents ranging from 1.68% to 4.27% by changing the treating

conditions.

     The amount of solids extracted in each of the organic solvents was not

significantly altered by treatment with gaseous ammonia (Table 13).  However,

the hot water extraction of ammoniated bark resulted in an increased solid
                                        t
dissolution of 3.5 times for coarse bark and 3.2 times for finely ground

bark when compared to untreated bark.  Ammoniation is known to Increase the

pH of bark (83, 84) which could bring about hydrolysis of some of the

polyphenolic compounds so that they would become water soluble.  The data

(Table 13) indicate that there would be a more Immediate release of leachable

-------
                                  -148-
solids Into the soil from ammoniated bark than from untreated  bark.   However,




there appears to be no evidence of  toxic materials from ammoniated bark.




Bollen and Glennie (83, 84) and Bollen and Lu (87) report that tannins,  resins,




and other readily extracted compounds from treated and untreated bark have




no toxic effects on plants when added to the soil.  The conditions of




extraction used in the present study are considered to be exhaustive and  it




appears that the early rate of leaching would not be continuous but  would




level off to a slow dissolution of  the bark due to microbial action.




     Table 13 also shows that grinding the bark to a small screen size has




little or no effect upon the amounts of solids extracted by each solvent.




Therefore, whether the bark is used as chunk-sized decorative bark,  or ground




and used as a mulch has little effect upon the immediate dissolution of  solids.




However, the immediate leaching of  solids should not be confused with microbial




decomposition for in the latter case bark size has a considerable effect.




According to Bollen (85) the finer the bark particles the more rapidly it is




decomposed.




     The bark residue from the exhaustive extraction with organic solvents




and hot water analyzed for 2.98% nitrogen.  These results support the findings




of Bollen and Glennie  (83, 84) that a part of the nitrogen in ammoniated bark




is immediately available but that a substantial part is slowly released only




through normal decomposition.  In many soil uses this dual effect is very




desirable and beneficial.




     The ultraviolet spectra of the extracts showed that little if any chemical




changes occurred in the extractable materials because of ammoniation (Table




14).  Only the major peaks and shoulders are reported, but in general there




were little differences in the spectra.  Although the wavelength range




included only the ultraviolet region, it is expected that major changes in

-------
                                  -149-
che leachable compounds' would have been detected.  Indications are that




chemical changes which might have occurred took place in those compounds




which were not soluble.

-------
                                  -150-
                              BIBLIOGRAPHY

 1.   Harkin,  J.  M.  and J.  Wo  Rowe.   Bark  and  its possible uses.  USDA Forest
          Service,  Research Not* FP-.-0910, October 1969.

 2.   Bark Committee of the Forest Products Research Society,  Marketing and
          selling bark products. Ed.  by  Jean Mater, R. Martin, D. Williams,
          R.  Sarles, F. Lamb, and R. Allison, Forest Products Research
          Society,  Madison, WI.   1969.

 3.   Hall, J. Alfred.  Utilization  of  Douglas-fir bark.  Pacific Northwest
          Forest and Range Experiment  Station,  Forest Service, U. S.
          Department of Agriculture, Portland,  OR.  1971.

 4.   Grilles, S. J.  Structure and  development  of the bark of Douglas-fir
          Pseudotsuga menziesii (Mlrb.) Franco.  Doctoral thesis.  Corvallis,
          Oregon State University.   1956. 67 numb, leaves.

 5.   Grilles, S. J. and F. H. Smith.   The secondary phloem of Douglas-fir.
          Forest Science 5:377-388. 1959.

 6.   Chang, Y.  Bark structure of North American conifers.   (U. S. Dept. of
          Agriculture, Technology Bulletin 1095).  1954.  86 p.

 7.   Ross, W. D. and R. L. Krahmer. Some sources of variation in structural
          characteristics of  Douglas-fir  bark.  Wood and Fiber 3:35-46.  1971.

 8.   Kiefer,  H.  J.  and E.  F.  Kurth. The  chemical composition of the bast
          fibers of Douglas-fir bark.  Tappi  36:14-19.  1953.

 9.   Segal1,  G.  H.  and C.  B.  Purves.   Chemical  composition of wood barks.
          Pulp and  Paper Magazine of Canada 47:149-161.  1946.

10.   Kurth, E. F.  The chemical composition of  bark.  Chemical Reviews
          40:33-49.  1947.

11.   Aspinall, G. 0.  Structural chemistry of the hemicelluloses.  In:
          Advances  in carbohydrate  chemistry.  Vol. 14.  Ed. by M. L. Wolfrom.
          New York, Academic  Press, Inc.  1959. p. 429-433.

12.   Timell,  T.  E., C. P.  Glaudemans and  J. K.  Gillham.  Recent studies on
          the polysaccharices of white birch  and other hardwoods.  Tappi
          42:623-634.  1959.

13.   Hamilton, J. K. and N. S. Thompson.  A comparison of the carbohydrates
          of  hardwoods and softwoods.  Tappi  42:752-760.  1959.

14.   Timell,  T.  E.   Wood hemicelluloses.  In:  Advances in carbohydrate
          chemistry.  Vol. 19.  Ed. by M. L.  Wolfrom.  New York, Academic
          Press, Inc.  1964.   p. 247-302.

-------
                                  -151-
15.   Whistler,  R.  L.  and E.  L.  Richards.   Hemicelluloses.   In:  The
          Carbohydrates Chemistry and  Biochemistry, Vol.  IIA, 2nd edition.
          Ed.  by W.  Pigman and  D.  Horton.  New York, Academic Press,  Inc.
          1970.  469  p.

16.   Graham,  H. M. and E.  F. Kurth.  Constituents  of extractives from Douglas-
          fir.   Industrial Engineering Chemistry 41:409-414.  1949.

17.   Holmes,  G. W. and E.  F. Kurth.  The  chemical  composition of the  newly
          formed inner bark  of  Douglas-fir.   Tappi 44:893-898.  1961.

18.   Cleland,  J. E.  and W. R. Fetzer.   Determination of  total solids  in
          glucose syrup.  Industrial and  Engineering Chemistry.  Analytical
          Edition 13:858-660.  1941.

19.   Timell,  T. E.,  C. P.  Glaudemans and  A.  C. Currie.   Spectrophotometric
          method for determination of  sugars. Analytical Chemistry 28:
          1916-1920.   1956.

20.   Timell,  T. E.  Isolation of polysaccharides from  the bark  of gymnosperms.
          Svensk papperstidning 64:651-661.   1961.

21.   Browning,  B.  L.   Methods of wood  chemistry.   Vol.  I.   New  York,
          Interscience Publishers.  1967.  384 p.

22.   Laver, M.  L., D. F. Root,  F. Shafizadeh and J. C. Lowe.  An improved
          method for the analysis of the  carbohydrates of wood  pulps  through
          refined conditions of hydrolysis,  neutralization, and monosaccharide
          separation.  Tappi 50:618-622.   1967.

23.   Black, R.  J., R. LeShange  and G.  Zweig. Paper chromatography, a
          laboratory manual. New York, Academic Press,  Inc.  1952.   195 p.

24.   Clark, J.  M., Jr.  Experimental biochemistry. San  Francisco, CA, W.
          H.  Freeman and Co.  1964.  228  p.

25.   Hough, L.  and J. K. N.  Jones.  Chromatography on  paper.  In:  Methods
          in carbohydrate chemistry.  Vol. I. Ed. by  R. L. Whistler, M.
          L.  Wolfrom.  New York, Academic Press.   1962.   589 p.

26.   Piper, C.  V. and L. J.  Bernardin. A spectrophotometric method for
          the chromatographic analysis of sugars.  Tappi 41:16-18.  1958.

27.   Miles Laboratories, Inc.  HT-1000. 'Product  Information.   Elkhart,
          Indiana.  No date.

28.   Miles Laboratories, Inc.  Diazyme, an amyloglucosidase enzyme for
          conversion of starch  and/or  dextrins to  glucose. Technical
          Bulletin No. 9-245.  Elkhart, Indiana.   1963.

29.   Wolfrom, M. L.  and D. L. Patin.  Carbohydrates of  the coffee bean.
          IV.  An arabinogalactan.  Journal  of Organic Chemistry 30:4060-
          4063.  1965.

-------
                                  -152-
30.   Abdel-Ahker, M., J.  K,  Hamilton and F.  Smith.   The  reduction  of  sugars
          with sodium borohydride.   Journal  of  the  American  Chemical  Society
          73:46-91-4692.   1961.

31.   Borchardt, L. G. and C. V.  Piper.   A gas chromatographic  method  for
          carbohydrates as alditol-acetates.  Tappi 53:257-260.   1970.

32.   Albersheim, P.,  D. J. Nevins,  P. D. English, and A. Karr.  A  method
          for the analysis of sugars in plant cell-wall  polysaccharides  by
          gas-liquid  chromatography.  Carbohydrate  Research  5:300-345.   1967.

33.   Sloneker, J. H.   Gas-liquid chromatography of  alditol acetates.   In:
          Methods in  Carbohydrate Chemistry.  VI.   General Carbohydrate
          Methods.  Ed. by R. L. Whistler and J. N. BeMiller.   New York,
          Academic Press.  1972.  603 p.

34.   Paech, K. and M. V.  Tracey.  Moderne Methoden  Per Pflanzen-analyse
          (modern methods of plant analysis).  1:468-473.  1956.

35.   Wolfrom, M. L. and D. L. Patin.  Carbohydrates of the coffee  bean.
          IV.  An arabinogalactan.  Journal  of  Organic Chemistry 30:4060-
          4063.  1965.

36.   Vogel, I. A.  Textbook of practical organic chemistry including
          qualitative organic analysis.  Third  ed., New  York,  Longmans,
          Green and Co.  1956.  1188 p.

37.   Clark, John N.  Experimental biochemistry.  San Francisco,  CA, W.  H.
          Freeman and Co.  1964.  228 p.

38.   Pische, S.  Color reactions of hexuronic acids.  In:  Methods of
          Carbohydrate Chemistry, Vol 1.  Ed. by R. L. Whistler and M.  L.
          Wolfrom, New York, Academic Press.  1962.  589 p.

39.   Jayme, G. and H. Knolle.  Paper chromatography of sugar mixtures on
          glass-fiber papers.  Angewande Chemie 68:243-246.   1956.

40.   Partridge, S. M.  Aniline hydrogen phthalate as a spraying reagent  for
          chromatography of sugars.  Nature  164:443.  1949.

41.   Wolfrom, M. L.,  M. L. Laver and D. L. Patin.   Carbohydrates of the
          coffee bean.  II.  Isolation and characterization  of a mannan.
          Journal of  Organic Chemistry 26:4533-4535.  1961.

42.   Falconer, E. L.  and G. A. Adams.  The aldobiouronic acids of
          hemicellulose B of oat hulls.  Canadian Journal of Chemistry 34:
          338-344.  1956.

43.   Acid Insoluble Lignin in Wood Pulp.  Tappi Standard T222 M-24.

44.   Browning, B. L.  and L. 0. Bublitz.  The ioslation of holocellulose
          from wood.   Tappi 36:452-458.  1953.

-------
                                  -153-
45.   Goldschmid, Otto.   Ultraviolet spectra.   In:   Lignins,  Occurrence,
          Formation, Structure and Reactions.   Ed.  by K.  V.  Sarkanen and
          C. H. Ludwig.   New York, Wiley-Interscience.   1971.   916 p.

46.   Wolfrom, M. L. and  J.  V. Karabinos.   The  identification of aldose
          sugars by their mercaptal acetates.   Journal of the American
          Chemical Society 67:500-501.   1945.

47.   Bates, F. J. and Associates.   Polarimetry, saccharimetry and the
          sugars.  National Bureau of Standards Circular C440.  1942.   p.  488.

48.   Whistler, R. L. and J. N. BeMiller.   a-D-xylose.  I.  Isolation from
          corn cobs or xylan.  In:  Methods in Carbohydrate  Cehmistry, Vol.
          1.  Ed. by R.  L.  Whistler and M. L.  Wolfrom.  New  York, Academic
          Press.  1962.   589 p.

49.   Isbell, H. S. and H. L. Frush.  d-D-Mannose from ivory-nut meal.   In:
          Methods in Carbohydrate Chemistry, Vol.  1.   Ed. by R. L. Whistler
          and M. L. Wolfrom.  New York, Academic Press.  1962.  589 p.
                                              •
50.   Fales, H. A. and F. Kenny.  Inorganic quantitative analysis.  New York,
          Appleton-Century-Crofts, Inc.  1939.  717 p.

51.   Fisher, R. B.  Quantitative chemical analysis.  Second  ed.  Philadelphia
          and London, W.B., Saunders Co.   1964.  501 p.

52.   Hodge, J. E. and B. T. Hofreiter.   Determination of reducing sugars
          and carbohydrates.  In:   Methods in  Carbohydrate Chemistry, Vol.
          I.  Ed. by R.  L.  Whistler and M. L.  Wolfrom.  New  York, Academic
          Press.  1962.   589 p.

53.   Sawardeker, J. S.,  J.  H. Soneker and A.  Jeanes.   Quantitative deter-
          mination of monoaaccharides as their alditol acetates by gas
          liquid chromotography.  Analytical Chemistry 12:1602-1603.  1965.

54.   Langlois, David L.   myo-Inositol from corn steep water  or calcium
          phytaten.  In:  Methods in Carbohydrate Chemistry, Vol. II.   Ed.
          by R. L. Whistler and M. L. Wolfrom.  New York, Academic Press.
          1963.  572 p.

55.   Beelik, A., R. J. Conca, J. K. Hamilton and E. V. Partlow.  Selective
          extraction of  hemicelluloses from softwoods.  Tappi 50:78-81.
          1967.

56.   Technical Association of the Pulp and Paper Industry.  Ash in pulp.
          Tappi Standards and Suggested Methods.  T211 M-58.  New York.
          1958.

57.   Willard, H. H., L.  L.  Merritt, Jr. and J. A.  Dean.  Instrumental
          methods of analysis.  Fourth edition.  Princeton,  N.J., Van
          Nostrand Co.,  Inc.  1965.  784 p.

-------
                                  -154-
58.   Browning.  B.  L.  Methods of wood  chemistry,  Vol.  II.   New  York,
          Interscience Publishers.   1967.   p.  384-882.

59.   Song, M. J.  The structure of  a xylan from Tilia  americana L.  Master's
          thesis.   Syracuse, N.Y.,  State University College of  Forestry
          at Syracuse University.  1970.  97 numb,  leaves.

60.   Technical Association of the Pulp and Paper  Industry.   Cupriethylene-
          diamine disperse viscosity of pulp.   Tappi Standards  and
          Suggested Methods.  T230 SU-66.   New York.  1966.

61.   Sears, K. D., A. Beelik, R. L. Casebier,  R.  J. Engen,  J. K. Hamilton
          and H. L. Hergert.  Southern pine prehydrolyzates: characterization
          of polysaccharides and lignin fragments.   Journal of  Polymer Science
          Part C:425-443.  1971.

62.   Lai, Y. Z.  Kinetics of degradation of carbohydrates.   Doctoral  thesis.
          Seattle, University of Washington.  1968.  97 numb,  leaves.

63.   Lai, Y. Z.' and K. V. Sarkanen.  Kinetic study on  the alkaline  degrada-
          of amylose.  Journal of Polymer Science,  Part C,  28:15-26.   1969.

64.   Barnett, A. J. G. and G. A. Tawab.  A rapid  method for the determination
          of lactose in milk and cheese.  Journal of Scientific Food
          Agriculture 8:437-441.  1957.

65.   Dubois, M. , K. Gilles, J. K. Hamilton, P. A. Rebers and F. Smith.
          Colorimetric method for determination of sugars and related
          substances.  Analytical Chemistry 28:350-356.  1956.

66.   Samuelson, 0. and A. Wennerblom.   Degradation of  cellulose by  alkali
          cooking.  I.  Formation of carboxyl groups.   Svensk Papperstidning
          57:827-830.

67.   Boggs, L.» L. S. Cuendet, I. Ehrenthal, R. Koch and F. Smith.   Separation
          and identification of sugars using paper chromatography.   Nature
          166:520-521.  1950.

68.   Hough, L., J. K. N. Jones and W.  H. Wadman.   Quantitative  analysis of
          mixtures of sugars by the method of partition chromatography.
          Part V.  Improved methods for the separation and dectection of
          the sugars and their methylated derivatives on the paper  chromato-
          gram.  Journal of the Chemical Society 1702-1706.  1950.
                                       i
69.   Partridge» S. M.  Aniline hydrogen phthalate as a spraying reagent
          for chromatography of sugars.  Nature 164:443.  1949.

70.   Wolfrom, M. L. and D. L. Patin.  Isolation and characterization  of
          cellulose in the coffee bean.  Agriculture and Food Chemistry
          12:376-377.  1964.

71.   Kurth, E. F. and H. J. Kiefer.  Wax from Douglas-fir bark.  Tappi
          33:183-186.  1950.

-------
                                  -155-
72,   Kurth,  E.  F.   The composition of  the  wax in  Douglas-fir bark.  Journal
          of the American Chemical Society 72:1685-1686.   1950.

73.   Fieser, L. F.   Experiments in organic chemistry.   Third ed.   Boston,
          D. C. Heath and Co.   1957.   353  p.
        a
74.   Fieser, L. F.  and Mary Fieser.  Natural  products  related  to  phenantherene.
          Third ed.  New York,  Reinhold Publ. Co.  1949.   704  p.

75.   Kurth,  E.  F.   The chemical composition of conifer bark waxes and
          corks.  Tappi 50:253-257.   1967.

76.   Hergert, H. L. and E. F.  Kurth.   The  chemical nature of the  cork  from
          Douglas-fir bark.  Tappi 35:59-66.   1952.

77.   Kiefer, H. J.  and E. F. Kurth.  The chemical composition  of  the bast
          fibers of Douglas-fir bark.   Tappi  36:14-19.  1953.

78.   Loveland,  P.  M. and M. L.  Laver.   Monocarboxylic  and Dicarboxylic acids
          in the benzene extract of Pseudotsuga menziesii bark.   Phytochemistry
          11:430-432.  1972.

79.   Loveland,  P.  M. and M. L.  Laver.   u-Hydroxy  fatty acids and  fatty
          alcohols from Pseudotsuga menziesii bark. Phytochemistry 11:
          3080-3081.  1972.

80.   Laver,  M.  L.,  J. V. Zerrudo and Harvey Aft.   Treatment  of Douglas-fir
          bark with gaseous ammonia.   Forest  Products  Journal  22:82-83.
          1972.

81.   Mater,  Jean, ed.  Marketing bark agricultural and horticultural o
          products.  Forest Products Research Society, Madison, WI.  1970.

82.   Mater,  Jean.   Utilization of bark in  highway landscaping. Forest
          Products Journal 21:17-20.   1971.
                          L
83.   Bollen, W. B.  and D. W. Glennie.   Sawdust, bark and other wood wastes
          for soil conditioning and mulching.  Forest  Products Journal
          11:38-46.  1961.

84.   Bollen, W. B.  and D. W. Glennie.   Fortified  bark for mulching and soil
          conditioning.  Forest Products Journal  13:209-215.   1963.

85.   Bollen, W. B.   Properties of tree barks in relation to  their agricul-
          tural utilization.  Research Paper PNW-77, U. S. Dept.  of Agric.,
          For.  Serv., Pacific NW For.  and  Range Exp. Sta., Portland, Oregon.
          1969.

86.   Aspitarte, T.  R.  Availability of nitrogen in ammoniated  bark used as
          a soil amendment.  Doctoral  thesis, Oregon State University,
          Corvallis, Oregon.  1959.

-------
                                  -156-
87.  Bollen, W. B. and K.  C.  Lu,   Douglas-fir bark  tannin decomposition in
          two forest soils.   Research Paper PNW-85, U, S. Dept. of Agric.,
          For. Serv.j Pacific NW  For.  and  Range  Exp. Sta., Portland, Oregon,
          1969.

88.  Ticknor, R. L. and J. L. Ricard.   Oregon State University, North
          Willamette Exp.  Sta., Aurora, Oregon.   Personal communication.
          1971.

-------
                                 -157-
              Table  1.  Raw material  for pelleting trials
     Pure  Barks




 1.   Douglas  fir




 2.   western  hemlock




 3.   western  larch




 4.   ponderosa  pine




 5.   lodgepole  pine




 6.   sugar pine




 7.   western  white pine




 8.   white fir




 9.   noble fir




10.   silver fir




11.   grand fir




12.   western  redcedar




13.   Sitka spruce




14.   Engelmann  spruce




15.   red alder
     Mixtures




1 + 2




2 + 8




1 + 2 + 3 + 4 + 7




Douglas fir bark + shavings




Douglas fir bark + hog fuel




Douglas fir bark + hog fuel + shavings




Douglas fir bark + sawdust




Douglas fir bark + sanderdust




Douglas fir bark + fertilizer




western hemlock bark + fertilizer




chemically extracted Douglas fir bark




western redcedar bark plus woody material




mixed northern hardwood bark from Maine




mixed eastern softwood bark from Maine




mixed eastern softwood bark + sawdust




southern oak bark + woody material from




  Tennessee

-------
                                  -158-
  Table 2.   Bulk densities and compression factors for some pelleted barks,
Bark species
Douglas fir
Douglas fir bast fibers
Douglas fir fines
Western hemlock
Western hemlock
Western hemlock
Ponderosa pine
Ponderosa pine
Lodgepole pine
Lodgepole pine
Silver fir
Silver fir
Silver fir
White fir
White fir
White fir
Grand fir
Grand fir
Red alder
Red alder
Pellet
diameter
(inches)
1/2
1/2
3/16
1/2
1/4
3/16
1/2
1/4
1/4
3/16
1/2
1/4
3/16
1/2
1/4
3/16
1/2
1/4
1/2
3/16
Bulk density
Before
pelleting
14.2
13.0
13.4
14.9
14.9
14.9
12.6
12.6
11.1
11.1
14.0
14.0
14.0
14.7
14.7
14.7
14.9
14.9
22.3
22.3
(Ib/cu ft)
After
pelleting
39.3
49.4
45.7
45.2
43.0
49.6
42.3
38.4
37.6
42.0
45.8
44.9
46.0
38.9
40.3
35.6
37.4
30.4
46.7
46.2
Compres-
sion
factor
2.8
3.8
3.4
3.0
2.9
3.3
3.4
3.0
3.4
3.8
3.8
3.2
3.3
2.6
2.7
2.4
2.5
2.0
2.1
2.1
Mixed eastern softwoods
1/2
15.9
47.5
3.0

-------
                                  -159-
        Table  3.   Partial  list,  recipients  of  bark pellet samples
     Company




 1.   Dant  and Russell,  Inc.




 2.   Sigma Plastics




 3.   Grant & Roth Plastics




 4.   Colorado State University




 5.   Small Business Administration




 6.   California Pellet  Mill  Company




 7.   Bohemia, Inc.




 8.   Fruit Growers' Supply Company




 9.   Sequoia Forest Products




10.   DiGiorgio Corporation




11.   MacMillan Bloedel




12.   Architectural Short Course Display




13.   MEG Company




14.   Publishers Paper Company




15.   Western Wood Fibre




16.   U. S. Plywood-Champion  Papers




17.   U. S. Forest Service




18.   Oregon Timber Products  Company




19.   Reydco Trading Company




20.   University of Maine




21.   Borden Company




22.   Ellingson Timber Company




23,   U.S. Forest Products Laboratory




24.   Oregon Lumber Export Company
   Location




Portland, Oregon




Portland, Oregon




Hillsboro, Oregon




Ft. Collins, Colorado




Portland, Oregon




Portland, Oregon




Eugene, Oregon




Los Angeles, California




Dinuba, California




San Francisco, California




Vancouver, B. C., Canada




Portland, Oregon




Neodesha, Kansas




Portland, Oregon




Eugene, Oregon




Eugene, Oregon




Washington, D. C.




Albany, Oregon




Eugene, Oregon




Farmington, Maine




Springfield, Oregon




Baker, Oregon




Madison, Wisconsin




Portland, Oregon

-------
                                  -160-
25.   Pacific Keynon Corporation

26.   Oxford Paper Company

27.   National Distillers Products Company

28.   Canadian Car (Pacific)

29.   H. J. Stoll & Sons

30.   W. R. Grace 5. Company

31.   Princeton Turf Farms

32.   Potlatch Forests, Inc.

33.   University of Washington

34.   FAO

35.   Agricultural Engineering Dept. , OSU

36.   Bear River Lumber Company

37.   Hercules, Inc.

38.   Agricultural Chemistry Dept., OSU

39.   B. P. John Furniture Company

40.   Crown-Zellerbach Corporation

41.   Koppers Company

42.   American Forest Products Company

43.   Peco, Inc.

44.   Head Groundskeeper, OSU

45.   NASA

46.   Chemistry Dept. Brigham Young Univ.

47.   George D. Ward & Associates

48.   Weyerhaeuser Company

49.   Institute for Wood Biology &
          Wood Preservation

50.   Collier Carbon & Chemical Company
Los Angeles, California

Rumford, Maine

Memphis, Tennessee

Vancouver, B. C., Canada

Portland, Oregon

Halsey, Oregon

Cranbury, New Jersey

Lewiston, Idaho

Seattle, Washington

Rome, Italy

Corvallis , Oregon

Logan, Utah

San Francisco, California

Corvallis, Oregon

Portland, Oregon

Portland, Oregon

Los Angeles, California

Martell, California

Portland, Oregon

Corvallis, Oregon

Moffett Field, California

Provo, Utah

Portland, Oregon

Longview, Washington


Hamburg, West Germany

Salem, Oregon

-------
                                  -ibi-
51.   Unalit S.  A.




52.   Wood Products Service Company




53.   Department of Forestry




54,   Ekono




55.   Monsanto Company        "




56.   University of Denver




57.   Incentive, AB




58.   Agricultural  Service Corporation




59.   Canadian Forestry Service
St. Usage, France




Memphis, Tennessee




Salem, Oregon




Helsinki, Finland




Eugene, Oregon




Denver, Colorado




Stockholm, Sweden




Salem. Oregon




Victoria, B. C., Canada

-------
                                  -162-
       Table 4.  Composition of bast fibers and Douglas-fir wood.





                                            Fiberb                 Wooda





Kther soluble                                2.92                   1.32




Alcohol soluble                              8.65                   5.46




Hot water soluble                            2.58                   2.82




   Sum of three extractives                 14.15                   9.60
Values based on oven-dry extractive-free material




Ash                                          0.60                   0.17




Lignin                                      44.80                  30.15




Holocellulose                               54.58b                 71.40




I'entosans                                    8.62                  10.11




Methoxyl group                               3.89               '4.75




Acetyl group                                 2.39                   0.59




Uronic acid anhydride                        4.62                   2.80




Methoxyl on lignin                           7.16                  15.20







 Values in percent of oven-dry bark.




 Corrected for lignin content.

-------
                                  -163-
Table 5.  Mixtures of alditol acetates for determination of an "Instrument
          K Factor"
Sugar
Myo-inositol hexaacetate
Glucitol hexaacetate
Galactitol hexaacetate
Mannitol hexaacetate
Acabinitol pentaacetate
Xylitol pentaacetate
Weight In mg
10.0
8.0
0.3
1.4
0.4
1.1
10.0
12.0
6.5
2.1
0.6
1.65
10. 0
16.0
0.6
2.8
0.8
2.2
10.0
20.0
7.5
3.5
1.0
2.75
10.0
24.0
0.9
6.2
1.2
3.3

-------
                                  -164-
  Table 6.   Extractives content  of  the bast  fibers  of  Douglas-fir  bark.
Solvent
u-Hexane
Itenzene
Diethyl ether
Hot water
Ethanol
Yield %a
1.59
0.95
0.38
9.61
0.96
Nature
Light yellow "wax"
lirown "wax"
Dihydroquercetin
Tannin, carbohydrate, etc.
Phlobaphine
o
 Percentages based on oven-dry weight of bast fibers.

-------
                   Table 7.  Columns used in gas-liquid chromatography.
Col.
No.
1
2
3
A
Length
ft.
12
6
6
6
O.D.
in.
1/8
1/8
1/8
1/8
Liquid
coating
SE-52
OV-17
US W-98
Hi-EFF 8 BP
Solid
Concentration support
10 wt.
3 wt.
10 wt.
3 wt.
%
11
/5
%
%
Gas
Gas
Chrom.
Chrom.
W.
Q.
Silicone Gum
Gas
Chrom.
Q-
Mesh
60/80
100/120
80/100
100/120
Instrument
Aerograph 200
Hewlett-Packard
Hewlett-Packard
Hewlett-Packard
Outside diameter

-------
Table 8.  Conditions for separation of  the "Yellow Band" by gas-liquid chromatography.
Injection
Detector port
Sample temp. °C temp. °C Column
Silylated
Mixture 265 277 SE-52
Mixture 265 277 SE-52
Mixture 270 275 UC W-98
Mixture 255 235 UC W-98
Mixture 278 272 OV-17
Mixture 260 245 OV-17
Mixture 275 270 Hi-Eff 8 BP

Flow
rate Range Attenuation
Column temp. °C ml/min setting setting Instrument
Isothermal 255 60 1
Isothermal 255 60 1
Isothermal 275 28 102
Program 28 10'
148-^300
2°/min
Isothermal 255 30 1Q2
Program 30 10"
170^300
2°/min
7
Program 30 10"
30°/min
4 Aerograph
4 Aerograph
16 Hewlett-Packard
16 Hewlett-Packard
16 Hewlett-Packard
16 Hewlett-Packard
16 Hewlett Packard


-------
                                  -167-
    Table 9.   Monocarboxylic  acids   in Douglas-fir  bark  "benzene  wax".
Acid2
I'almLtic
(UnLdent. sat'd.)
(C1Q unsat'd.)
lo
Stearic
(Unident. unsat'd.)
1 1
it
if
Arachidic
ii-Heneicosanoic
Uehenic
n-Tricosanoic
Lignoceric
n-Pentacosanoic
Cerotic
(N-Heptacosanoic)
(Montanic)
Retention time
SK-30, 215°
1 .3
1.6
2.2
2.4
3.1
3.8
4.53
4.53
4.5
6 . 4
8.9
12.1
17.0
23.0
31.7
	 4
59.3
Percent
relative
abundance
0.9
0.3
3.1
0.4
0.8
trace
0.4
3.3
2.5
tract
22.3
0.7
51.7
0.5
12.7
trace
0.3
99.9
 Identified by GLC of methyl esters in TLC spot 1.

2
 Named compounds in parentheses, "(	)",  denote tentative identifications


 These peaks were resolved from methyl arachidate at 195°.

4
 Observed in a temperature-programmed run  on SE-30,  and isothermally  on
 Apiezon-L.

-------
                              -168-
Table 10.  Dicarboxylic acids  in Douglas-fir hark "benzene wax".
                                                  Percent
                                                 relative
      Acid	abundance

      Hexadecanedioic                              36.3

      Octadec-?-enedioic3                          24.8^

      Octadecanedioic                              14.6

      Eicosanedioic                                14.3

      Docosanedioic                                 8.7

      Tetracosanedioic                              1.3

                                                  100.0


       Identified by GLC of dimethyl esters in TLC spot 2
       and verified by tandem GLC-MS.
      2
       Estimated from SE-30 data.

       Location of double bond not established.
      4
       Includes small, unresolved shoulder.

-------
  Table 11.   Aliphatic alcohols from benzene  wax.
Acid
l-hexadecanol
1-octadecanol
1-eicosanol
1-docosanol
1-tetracosanol
Percent
relative .
abundance
trace
4.0
3.6
45. 0
47.5
100.1
Based on peak areas from GLC of acetate derivatives.

-------
                                  -170-
Table 12.  GLC of TLC spot 3 containing methyl esters  of  oHiydroxy  acids.
GLC
peak
(SE-30)
00
0
1
2a
2
3
-
4
5
6
Acid
unident .
unident .
16-hydroxyhexadecanoic
unident .
18-hydroxyoctadecanoic
20-hydroxyeicosanoic
unident.
22-hydroxydocosanoic
24-hydroxytetracosanoic
(26-hydroxyhexaconsanoic)
«tl
(min)
0.8
2.7
9.5
14.1
15.1
21.4
22.9
27.8
33.9
41.1
Percent
relative
abundance
1.7
7.8
25.6
7.1
7.1
18.0
2.0
23.6
5.9
1.3
J
 For temp, programed run 180-248°.

 ?
 "Calc. from peak areas, uncorrected for detector response.


 Tentative identification.

-------
                                 -171-
             Table  13.   Solvent  extraction  yields  from bark
Solvent
n-Hexane
Benzene
Diethyl ether
Kthanol
Hot water

Untreated
bark
4.67
2.70
1.81
6.38
2.20
Percent Yield-L
NH -treated
bark2
4.63
2.95
2.21
7.13
7.60

NH -treated
bark
(32 mesh)
4.16
3.92
1.19
6.06
7.00
 Percentages  based  on dry  weight.

?
"NH   represents  gaseous  ammonia.


 Passed  a screen of 32 meshes  to  the  inch.

-------
                                  -172-
           Table 14,  Ultraviolet absorption of bark extracts.

Extraction solvent                           Major              Shoulder
 and bark sample    	peak nm	nm

ri-Hexane
  Untreated bark  ,                            264                  228
  NHj-treated bark          „                 264                  330
  NH3-treated bark (32 mesh)                  264       -           229

Benzene
  Untreated hark                              262                  	
  NH3~treated bark                            263                  	
  NH3-treated bark  (32 mesh)                  265                  	

Diethyl ether
  Untreated bark                              254                  	
  NH3-treated bark                            257                  	
  NH3-treated bark  (32 mesh)                  258                  	

Ethanol
  Untreated bark                              261                  	
  NH3-treated bark                            265                  	
  NH3-treated bark  (32 mesh)                  264

Hot water
  Untreated bark
  NH3-treated bark                            249                  	


 NH_ represents gaseous ammonia.
2
 Passed a screen of 32 meshes to the inch.

-------
                  Figure 1
1.  Douglas fir bark from veneer logs.
2.  Through hanonermill containing 1/4-inch screen.
3.  Reground to pass 20 mesh screen.
1.  Pelleted Sitka spruce bark, 1/2-inch die.
2.  Pelleted red alder bark, I/4-inch die.
3.  Pelleted Douglas fir bark with fertilizer added, 3/16-inch die,
4.  Pelleted Douglas fir bark. 1/lQ-inch die.

                                 NOT REPRODUCIBLE
                     -173-

-------
                             Figure 2
      A
                  *
           ¥#**'•
         • wr"
         '*
 
-------
                         Figure 3
Carbohydrate fraction from Douglas  fir  inner  bark.  Left side:
Inner bark starting material,  fibrous in  nature.  Right side:
Carbohydrate (holocellulose)  fraction,  40-50% yield, pure
white, still fibrous in nature.
 Carbohydrate  (holocellulose) fraction from inner bark at a higher
 magnification than above.  It shows the white, fibrous nature of
 the material.                                      	--
                                        NOT REPRODUCIBLE
                             -175-

-------
          Figure 4
                    NOT  REPRODUCIBLE
Anatomy  of Douglas fir  Bark
          -176-

-------
    Douglas-fir inner bark (100.0 g dry weight)
                  Extraction with
                  ethanol-water (A:1 v/v)
    r
Kxtract 15.4 g soJids   Residue 84.6 g
                             Extraction with
                             benzene-ethanol (2:1 v/v)
   Extract 4.4 g solids
Residue 80.2 g
                                   Extraction with hot water
      Extract 11.1 g solids
         Extract 2.8 g solids
      Residue 69.1 g
            Extract 22.0 g solids
                                         Extraction with 0.5%
                                         aqueous ammonium oxalate
            Residue 66.3 g
                                               Acidified sodium
                                               chlorite treatment
                  Residue 44.3 g
                                                     Acidified sodium
                                                     chlorite treatment
               Extract 14.3 g solids
                        Residue 30.6 g
                         (Holocellulose)
Figure 5.  Isolation of the holocellulose fraction of Douglas-fir inner
           bark, (g/100 g dry-weight basis).
                            -177-

-------
                  Hot-water-soluble  solids  from Douglas-fir  inner
                  bark (11.1 g from  100,0 g of  original  Inner  bark)
                                      Extraction with water
                                      at room temperature
        Water solubles 7.8 g
                       Precipitation with 70%
                       ethanol then centrifuge
                    Residue 3,3 g
                      Fraction A
70% ethanol solubles
4.9 g, "Fraction C"
Residue 2.9 g
 "Traction B"
                               HT-1000 then Diazyme
                               L 30 then dialysis
          Dialyzables
                 Dialyzables
       Non-dialyzables 1.1 g
              Chymotrypsin then
              trypsin then dialysis
       Precipitate    Non-dialyzables 0.16 g
                         "Fraction D"
Figure 6.  Purification of the polysaccharides extracted by hot water.
                            -178-

-------
               Hoiocellulose  (100.0  g, dry weight)
           I'	
           |
           i
       Extract

            Methanol to 70.0%


     Precipitate
     Hemicellulose A 7.0 g
                                  20%  Ba(OH)2,  then  KOH  to  10.0%
Residue A 84.6
         1.0% NaOH
              Extract

                   Methanol to 70.0%
            Precipitate
            Hemicellulose B 1.7 g
                     Extract

                          Methanol to 70.0%
                   Precipitate
                   Hemicellulose C 2.9 g

       Residue B 78.2 g
                                                              15.0% NaOH
              Residue C 62.6
Figure 7.  Fractionation of the holocellulose from Douglas-fir inner bark
           into its component hemicellulosea.
                            -179-

-------