Abstract Pulpwood is the primary product goal of limited commercial planting of eucalypts in Florida. Pulp yield and properties of four species for three product goals were investigated. NSSC pulping of three eucalypt species and a representa­ tive native hardwood mix showed substantially higher yields of pulp from the eucalypts with acceptable strength for corrugating medium. Yield of dissolving pulp from Eucalyptus tereticornis wood was slightly lower than from sweetgum. Keywords Eucalyptus grandis, E. robusta, and E. camaldulensis wood gave screened yields Eucalyptus of kraft bleached grade equal to those of representative native hardwood con­ NSSC pulping trols at comparable kappa numbers. Strength of handsheets was inferior while Kraft pulping brightness was superior to controls. Eucalyptus tereticornis gave substantially Yield lower screened yields of kraft bleached grade pulp at equal or higher kappa Mechanical properties numbers than the representative hardwood controls. Brightness

Yield and properties of pulp from eucalypt wood grown in Florida

E. C. Franklin

Eucalypts were first planted in Florida ic improvement has been on growth Samples were taken from 7- to 10-year­ before the turn of this century. Home­ rate, survival, and cold hardiness. Cold old plantation trees, 5 to 10 in. (12. 7 steaders planted seedlings to quickly hardiness is of particular importance to 25.4 cm) diameter, growing under obtain shade and shelter from the in­ because subtropical Gulf Coastal areas typical commercial conditions in central tense seasonal winds. A few small stands have more severe frosts than compa­ and southern Florida. volunteered from these early introduc­ rable areas of Australia. Family plus tions, but natural reproduction by eu­ individual tree selection has been prac­ RESULTS AND DISCUSSION calypts has been extremely limited. In ticed for one to three generations in the early 1960's efforts to introduce those species proven most adaptable to NSSC for Corrugating Medium exotic hardwood for timber production Florida: Eucalyptus grandis Hill ex in southern Florida were intensified. Maiden,E. robusta Sm.,E. camaldulen­ Three species of eucalypt trees, ap­ The Florida Division of Forestry made sis Dehnh, and E. tereticornis Sm. proximately 10 years old, (E. camal­ extensive introductions and started de­ Progeny tested seed orchards now sup­ dulensis, E. robusta, andE. tereticornis) veloping commercial nursery tech­ port commercial production of contain­ were pulped by the neutral sulfite semi­ niques. The privately endowed Florida erized planting stock. Because pulp­ chemical process (NSSC) (Table I) . Pulps Forest Foundation was founded with wood was the primary product goal, it were converted to corrugating medium the mission to develop a hardwood tim­ was necessary to investigate the pulp­ with results generally equal to, or ber resource, particularly in southern ing properties of the various species better than, with native hardwoods Florida. Replicated trials were estab­ which are currently available for plant­ presently used. Under the same pulp­ lished with 67 species and 156 seed ing and further to characterize the prop­ ing conditions eucalypts gave 4-10% sources of eucalypts as well as other erties of products made from this pulp. higher screened yields (o. d. wood basis) exotic species. Three or four species showed promise for fiber production, Table I. Neutral Sulfite Semichemical Pulping of Eucalyptus tereticornis, and in 1967 the Southeastern Forest E. robusta, and E. camaldu/ensis, Age Approximately Ten Years, Experiment Station, Forest Service, for Corrugating Medium USDA, absorbed the Foundation's re­ Furnish search program. In 1971, the Station entered into cooperative research agree­ E. ments with several forest industries and E. carnal- Native landowners* to develop a commercial tereti- E. du/en- hard- planting program for eucalypt pulpwood Property cornis robusta sis wood production in Gulf Coastal areas of the Wood United States. Density, lb/ft3 36 28 33 29 Primary selection emphasis in genet- Moisture, % o.d. wood 30 28 32 44 Pulp *Cooperators are Buckeye Cellulose Corp., Perry, Fla.; Container Corp. of America, Fernan­ Kappa number, ml 38 37 38 36 dina Beach, Fla.; Hudson Pulp and Paper Corp., Screened yield, % o.d. wood 75 74 74 67 Palatka, Fla.; International Paper Co., Bainbridge, Burst index8 36 48 45 40 8 3 Ga.; ITI' Rayo~er, Fernandina Beach, Fla.; Lykes Breaking length , m X 10- 6.5 8. 7 7.2 6.8 Bros., Inc., Tampa, Fla.; St. Regis Paper Co., CMTa,b, lb 19.5 19.5 22.0 18.5 Jacksonville, Fla. Chlorine number 23.9 25.2 E. C. Frenklln, Program Manager, Loblolly Pine Residual total alkali 5.5 6.4 Management R&D Program, Southeastern Forest Experiment Station, USDA, Charleston, S. C. 8 CSF = 400 ml. 29403. bconcora Medium Test for Crush Resistance.

Reprinted from Tappi, The Journal of the Technical Association of the Pulp and Paper Industry, Vol. 60, No. 6, June 1977. Copyright, 1977 by Tappi and reprinted by permission of the copyright owner. a FOREST ENGINEERING than native hardwood controls at com­ yield and pulping capacity at higher in lower power requirements than those parable kappa numbers. Burst and ten­ machine speeds. from the pine. Eucalypt samples had sile tests at the same freeness (CSF) substantially higher viscose viscosity levels were higher for E. robusta and Kraft Process for Dissolving Pulp and did not swell as well as pine when E. camaldulensis than for native hard­ Average results of three kraft runs mixed with caustic, and therefore had wood, but lower for E. tereticornis. of E. tereticornis for dissolving pulp lower filtration values (Table II). Slight­ Crush resistance, as measured by the gave screened yields lower (32 vs. 35%) ly modified processing conditions may Concora Medium Test (CMT) was higher than comparable runs with sweetgum be needed for eucalypt pulps. for all eucalypt species than for na­ (Liquidambar styraciflua L.), even Kraft Process for Bleached tive hardwood. Eucalyptus tereticornis though E. tereticornis had slightly Grade Paper Pulp and E. camaldulensis pulps appeared higher density (Table II). Brightness much rougher than those from the hard­ and alpha cellulose content did not Several trials using conventional wood andE. robusta and required more differ significantly from those of sweet- kraft pulping processes have been done refining to reach the same freeness gum pulp. With the kraft process, E. by four companies on one or more of levels. tereticornis pulp had substantially higher four eucalypt species (Table III). Eucalypt Yields by NSSC pulping were gen­ lignin content than that from sweet- wood of all species had equal or higher erally higher for eucalypts, even though gum (Table II). densities than native hardwood con- the permanganate. and chlorine num­ To facilitate filtration testing, eucalypt trols used by mills in the southeastern bers were comparable. Higher yields pulp was blended with a standard pine Gulf Coast region. Eucalyptus tereti­ from eucalypts are reasonable, con­ pulp (V-68) in equal proportions. This cornis andcamaldulensis had the highest 3 sidering that they have lower lignin was done to avoid problems with press- densities, averaging about 32 lb/ft , and higher holocellulose contents than ing short-fibered pulps. The blend was whileE. grandis andE. robusta averaged 3 native hardwoods. tested against a pure sample of pine 28 lb/ft , a value similar to that of the Eucalypt pulp, even after refining in standard V-68 pulp. Eucalypt samples native hardwood. Moisture contents a small disk refiner, has substantially shredded quite easily, which resulted were generally higher than the two more shives than the hardwood con­ trols. Eucalypt pulps also required more Table 11. Kraft Pulping of Eucalyptus tereticornis Sm. Wood beating energy to reach the same free­ for Dissolving Pulp by Buckeye Cellulose Corp. ness levels as the hardwood. After reaching the same freeness levels, Furnish the appearance was the same as the Property E. tereticornis Sweetgum 8 native hardwood with only a slightly darker color. Total residual alkali analy­ Wood sis indicated that E. tereticornis had Density, lb/ft3 31 29 slightly higher chemical demand than Moisture content,% o.d. wood 54 46 the E. camaldulensis, but results were Pulp not conclusive. The higher average den­ Kappa numberb, ml 9 ? sity of the eucalypts indicated that Screened yield, % o.d. wood 32 35 their use could increase the pulping Brightness, % 92 94 Alpha cellulose, % 96 94 capacity of a mill. Lignin, % 0.35 0.24 Eucalypt pulps had higher initial ten­ Viscose viscosity, sec 74c 53d sile and CMT values and maintained this Viscose filtration, g 806c 900d advantage over the hardwood during the beater runs. Eucalyptus tereticornis 8 Liquidambar styracif/ua L. proved to yield the poorest pulps, falling bTAPPI 40 ml. below the native hardwood in tensile c50% eucalypt blended with 50% pine standard V-68. d!OO%pinestandard V-68. strength and never developing compa­ rable burst strength. All curves indi­ Table Ill. Pulping of Wood from Four Species of Eucalypts, With and cated that eucalypt pulps developed Without Bark, for Bleached Grade Kraft Pulp by Four Companies equal strength at higher freeness levels to those of native hardwood. This higher Pulp Pulp freeness may offset the slightly higher Wood moisture, yield, Kappa energy required to meet the same free­ density, %0.d. %0.d. number, Company Species lb/ft3 wood wood ml ness mentioned earlier. Bulk of the eu­ calypt pulps was generally higher than Buckeye E. tereticornis 31 54 43 14 that of native hardwood. In addition, no Cellulose Corp. Control8 29 46 46 11 increase in water drop was noted when Hudson E. camaldulensis 33 55 45 12 the pulps were aged on a hot plate. These Pulp and Dehnh. 32 56 43 16 two facts indicate that eucalypt pulps Paper Corp. E. camaldulensisb should be more absorbent than pulps E. grandis Hill from native hardwoods. ex Maiden 28 50 45 12 Results of this trial indicated that E. robusta Sm. 28 48 48 13 all three species of eucalypts tested E. tereticornis 34 42 39 13 produced NSSC pulps generally equal E. tereticornisb 33 49 42 20 or superior to native hardwood furnish. Control8 46 47 11 E. robusta andE. camaldulensis yielded International E. tereticornis 34 49 43 12 stronger pulps with lower chemical Paper Co. consumption than did E. tereticornis. ITT Rayonier E. tereticornis 31 50 39 13 The generally superior pulping proper­ E. tereticornis 30 51 36 15 Control 12 ties of the eucalypts compared to native 48 hardwood could be expected to increase 8 Representative hardwood furnish. overall production through increased bWith bark.

66 Vol. 60, No. 6 June 1977 I Tappi Table IV. Properties of Bleached Grade Kraft Pulp from Four Species properties ofhandsheets were variable, of Eucalypts Grown in Southern Florida8 with burst factor being generally lower and tear factor being equal to or higher Burst, Tear, Brightness, Opacity, than native hardwood furnish. Bright­ Mill species % units % % ness of bleached pulp was superior to that of native hardwood furnish as was Hudson Pulp and opacity of handsheets. Costs for bleach­ Paper Co. ing were somewhat higher for E. tere­ E. camaldulensis Dehnh 70 93 E. camaldulensisb 66 97 ticornis than for native hardwoods. E. grandis Hill ex Maiden 62 70 87 When bark was included with chips E. robusta Sm. 59 85 86 fromE. camaldulensis andE. tereticornis, E. tereticomis Sm. 37 82 82 screened yields of bleachable-grade E. tereticomisb 27 78 81 kraft pulp were lower at substantially Controlc 59 96 79 higher kappa numbers, reflecting in­ International Paper Co. creased chemical consumption. Tear E. tereticomis 148 100 88 strength was equal to or slightly high­ ITT Rayonier er, while burst strength was equal to E. tereticomis 76 130 92 82 E. tereticomisb 74 130 93 82 or slightly lower than those of samples Control 100 171 91 72 without bark. Changes in brightness and opacity were insignificant for E. 8 Results are at 300 ml CSF except for Hudson E. tereticornis which is at 400 ml CSF. tereticornis with bark. bWith bark. While the general conclusions seem cRepresentative h?.«iwood furnish. to be well supported by the data, comparisons between tests by differ­ controls although preprocess handling low to substantially above comparable ent companies must be tempered with procedures can cause this parameter controls. Inclusion of bark with the caution because of differences in con­ to vary widely. E. tereticornis wood did not substan­ trol lots, processing details, and product Pulp yields of control samples were tially affect burst or tear strength in standards. very consistent, averaging 4 7%. De­ these trials. barked E. robusta was the only wood Brightness of eucalypt pulps was uni­ Received for review Feb. 14, 1977. with pulp yields exceeding the controls. formly better than controls at equal or Accepted March 7, 1977. DebarkedE. camaldulensis andE. gran­ slightly higher bleaching costs under Based on a paper published in TAPP! Conference dis were second to E. robusta, with yields comparable conditions. In one trial, Papers - 1977 Forest Biology/Wood Chemistry, a couple of points below the controls opacity of handsheets from E. tereti­ a copyrighted publication of the Technical AMD­ ciation of the Pulp and Paper Industry, Inc., at comparable kappa numbers. Pulp cornis pulps, with and without bark, Atlanta, Ga. yields of debarked E. tereticornis were was 10 points higher than that of the somewhat variable, averaging 5-8 points controls (Table IV). below the controls. When bark was not removed, E. camaldulensis showed a 4- SUMMARY point reduction below the controls while Neutral sulfite semichemical (NSSC) E. tereticornis showed a 2-point increase process gave pulp yields from E. tere­ in one case and a 12-point reduction ticornis, E. robusta, andE. camaldulensis in another. In both, reduction below substantially higher than those from the yield of bark-free wood was 2 to native hardwood furnish at comparable 3 points. One consistent result of pulp­ kappa numbers. Strength properties of ing with bark was a higher consump­ test sheets indicated slight inferiority tion of chemicals and thus a less com­ for E. tereticornis and slight to moderate plete cook under conditions similar to superiority for E. robusta andE. camal­ that of bark-free wood. This was re­ dulensis. Preliminary results showed flected by the consistently high kappa yields of dissolving pulp from E. tereti­ numbers for runs which included bark. cornis to be slightly lower than those The advantage of including bark was from native sweetgum. Brightness and a general increase in tear factor through alpha cellulose levels were comparable. the addition of the longer bark fiber The easily shreddedE. tereticornis pulp to the pulp. Of the important pulp had lower filtration values than the properties, burst and tear strength controls. were most variable (Table IV). In gen­ DebarkedE. grandis, E. robusta, and eral, theE. camaldulensis pulp without E. camaldulensis wood gave screened bark developed very high burst without yields of kraft bleached paper grade pulp substantial loss in tear. When bark equal to those of representative native was included, the E. camaldulensis hardwood furnish at comparable kappa pulp exceeded the controls in both burst numbers. Strength properties of hand­ and tear strength. Eucalyptus robusta sheets were generally inferior to those and E. grandis pulps developed burst of native hardwood controls. Brightness values in excess of the controls at 300 of bleached pulp was equal or superior ml CSF, but tear values were 11 and 26 to native hardwood controls at equal or points below the controls, respectively. slightly higher processing costs. Thus tear strength of these two pulps Debarked E. tereticornis wood gave was deficient in this trial. Burst values substantially lower screened yields of for E. tereticornis were consistently kraft bleached paper grade pulp at well below the controls while corre­ equal or higher kappa numbers than sponding tear values were slightly be- native hardwood furnish. Strength

Tappi / June 1977 Vol. 60, No. 6 67