Short Communication

Solvated liquid-lignin fractions from a Kraft black liquor

Julian Velez, Mark C. Thies

Department of Chemical and Biomolecular Engineering, Clemson University, Clemson, SC 29634-0909, USA
h i g h l i g h t s
Liqueed, lignin-rich phases were
produced via CO
2
acidication at
115 C and 6.2 bar.
Liquid-lignin fractions were produced
within narrow (0.5 units) bands of
pH.
Fractions are highly hydrated and
thus have surprisingly low softening
points.
Selected fractions have reduced
sodium content, indicating selectivity
occurs.
g r a p h i c a l a b s t r a c t
a r t i c l e i n f o
Article history:
Received 4 July 2013
Received in revised form 14 August 2013
Accepted 17 August 2013
Available online 29 August 2013
Keywords:
Lignin
Hydration
Fractionation
Precipitation
CO
2
acidication
a b s t r a c t
A softwood Kraft black liquor was acidied with carbon dioxide at 115 C and 6.2 bar over a pH range of
13.69.5, resulting in the precipitation of liqueed-lignin fractions as a separate phase. Seven such
liquid-lignin fractions were produced, with each fraction being phase-separated within a narrow pH
band of 0.5 units. The fractions were found to be highly hydrated phases, containing 32.348.2 wt.%
water; as a result, their measured melting points were quite low, 90.7110.5 C. In contrast, no melting
point was detected up to 375 C for any of the lignin fractions after drying. Signicant reductions in met-
als content were observed for the lignin fractions compared to the original black-liquor feed.
2013 Elsevier Ltd. All rights reserved.
1. Introduction
Lignin separated from Kraft black liquor has the potential to be-
come an inexpensive and renewable platform for the production of
aromatic chemicals, bio-based materials, and clean biofuels (Doh-
erty et al., 2011; Holladay et al., 2007). However, the heterogeneity
of lignin presents a challenge for obtaining a more fundamental
understanding of the chemical structure of this material. The frac-
tionation of lignin has two potential benets: (i) chemical struc-
ture-vs.-bulk property relationships can be obtained, and (ii) the
fractions themselves can have properties useful for various appli-
cations (Doherty et al., 2011; Holladay et al., 2007).
Processes have been developed for the separation and purica-
tion of lignin fromthe black-liquor streamof a paper mill (Lake and
Blachburn, 2011; Ohman et al., 2009). Of particular interest to us is
a recently discovered process (Lake and Blackburn, 2011) known as
Sequential Liquid Lignin Recovery and Purication (SLRP). With
SLRP, environmentally benign CO
2
is injected into the Kraft black
liquor at a unique set of temperatures and pressures, acidifying
the black liquor and resulting in the formation of a liquid-lignin
phase. The precipitated lignin is then further acidied with H
2
SO
4
to produce a low-ash (12 wt.%) lignin.
The subject of this study was the fractionation of the liquid-lig-
nin phase described above, followed by selected characterization of
the fractions obtained. To accomplish the fractionation, we used
CO
2
acidication to induce the phase separation of liquid-lignin
fractions from the black-liquor phase. Fresh Kraft black liquor
0960-8524/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2013.08.097

Corresponding author. Tel.: +1 864 656 5424; fax: +1 864 656 0784.
E-mail address: mcths@clemson.edu (M.C. Thies).
Bioresource Technology 148 (2013) 586590
Contents lists available at ScienceDirect
Bioresource Technology
j our nal homepage: www. el sevi er . com/ l ocat e/ bi or t ech
was used as the feed only for the 1st fractionation. The partially
spent black liquor from the 1st then served as the feed to the
2nd fractionation, the spent black liquor from the 2nd served as
the feed to the 3rd fractionation, etc. (see Suppl. Fig. 1). Each frac-
tion was collected over a narrow range of 0.5 pH units. The pro-
cessability of the liquid-lignin fractions being isolated was of
particular interest in this study, so a technique was developed to
measure the softening point of each fraction in its hydrated state
at process operating conditions. The water, ash, and metals con-
tent, as well as the yield of the various fractions, were also
determined.
Previously reported work on the pH-induced precipitation of
lignin from black liquor has been limited. Nagy et al. (2010) used
CO
2
at 15 bar to acidify a Kraft black liquor down to a pH of 10.5
and 9.5, respectively, and obtained two solid lignin fractions. Gou-
veia et al. (2012) and Garca et al. (2009) both used H
2
SO
4
to pre-
cipitate out solid lignin fractions from a Kraft black liquor down to
a pH of 2.0 and 0.72, respectively.
2. Methods
2.1. Materials
A softwood Kraft black liquor was obtained from a pulp and pa-
per mill. For safety reasons, it was pre-treated prior to our experi-
ments in a controlled-cavitation reactor (Hydrodynamics, Inc.,
Shockwave Power Reactor) to oxidize the sodium sulde and thus
avoid the release of hazardous H
2
S gas when lowering the pH of
the black liquor (Frederick et al., 2001). Industrial-grade CO
2
with
a purity of 99.5% (CAS 124-38-9) was supplied by Airgas. Nitric acid
and hydrogen peroxide were used to prepare black-liquor and li-
quid-lignin samples for metal-content measurements. Certied-
ACS-Plus-grade nitric acid (CAS 7697-37-2, 70 wt.% in water) and
ACS-grade hydrogen peroxide (CAS 7722-84-1, 30 wt.% in water)
were both obtained from Fischer Scientic.
2.2. Precipitation of liquid-lignin fractions from Kraft black liquor via
CO
2
acidication
The experimental apparatus consisted of a 2-L pressure reactor
and associated components (Parr Instrument Company, model
4541). The original 1-L reactor vessel was replaced with a 2-L ves-
sel having a 45-angle conical bottom to facilitate the collection of
liquid-lignin fractions. A semi-batch operating mode was used,
with CO
2
being bubbled into the black-liquor charge in order to
bring about acidication and the resultant formation of a liquid-
lignin phase. For a typical run, approximately 2200 g of fresh black
liquor was charged to the reactor. The reactor was both purged and
pressurized with nitrogen and heated up to 115 2 C under agita-
tion at 60 rpm with a helical ribbon impeller. As temperature was
reached, the reactor pressure was adjusted to 6.2 0.1 bar (i.e.,
about 3 bar above the vapor pressure of water) to ensure that the
aqueous black-liquor phase was always in the liquid state. The
addition of CO
2
then commenced at a ow rate of 250 std mL/
min via a sparger immersed in the black liquor. Prior to the frac-
tionation experiments, a titration curve had been constructed to
determine the volume of CO
2
needed to produce pH decrements
of 0.5 over a pH range of 13.69.5. Based on this titration curve,
the CO
2
ow was stopped after the required volume of CO
2
had
been fed, and the pH of the black liquor was conrmed by sampling
via a dip tube. The agitation was then stopped to allow the liquid-
lignin phase to settle out by gravity for 2 h, the reactor was cooled
down to 65 C, the pressure was released, and the reactor was
opened. The spent black-liquor phase was then decanted off, and
the solidied liquid-lignin phase that had separated out in the
bottom of the reactor was collected for characterization. After this
1st pH decrement from 13.6 to 12.8 was completed, the spent
black liquor was returned to the reactor for the next CO
2
addition
and the next pH decrement from 12.8 to 12.1 pH. This procedure
was repeated in 0.5 pH decrements until a pH of 9.5 had been
reached and seven narrow-pH lignin fractions had been recov-
ered (see Suppl. Fig. 1).
CO
2
ow rates to the reactor were controlled with a mass ow
controller (Brooks Instrument, model 5850C), with the estimated
uncertainty in the ow rate being 2 mL/min. The reactor was
heated via heating tapes (BriskHeat model BIH101080L, 2.5 cm
wide) wrapped around the outside of the reactor. Operating tem-
peratures were measured with a J-type thermocouple in the bot-
tom and a K-type thermocouple in the top of the reactor, with
temperatures being controlled to within 2 C via a temperature
controller (Omega, model CN8500). The reactor pressure was mea-
sured to within 0.1 bar with a Bourdon tube pressure gauge (Nos-
hok, model 401, 020 bar).
2.3. Softening points of liquid-lignin fractions
Softening-point measurements were made inside a windowed,
high-pressure, variable-volume view cell (DB Robinson & Assoc.
PVT view cell, 120 cm
3
volume, model No. 150-10-155, see
Fig. 1) located inside a forced-convection oven. The use of such a
cell enabled the measurement of softening points under pressure
so that the liquid-lignin samples did not de-volatilize (i.e., lose
water) at the elevated temperatures of operation. The softening-
point apparatus itself was fashioned in a manner consistent with
ASTM D6090-99, using the cup-and-ball assembly from a Mettler
dropping-point cell (model No. FP83HT). The support for the
assembly contained a 0.5-mL water reservoir to ensure a water-
saturated environment. For a typical softening-point measure-
ment, 0.5 g of a solidied liquid-lignin fraction taken directly
from the pressure reactor was packed into the Mettler cup. The
ball-and-cup assembly and support were then placed on top of
the piston in the view cell, and the top end cap was installed to seal
the cell. The process uid side was then ushed with two cell vol-
umes of water-saturated nitrogen to ensure an inert environment.
Finally, the piston height was adjusted both to pressurize the cell
to 1.5 bar and to align the ball-and-cup assembly with the laser/
photoresistor setup.
For softening-point measurements, the oven temperature con-
troller was set to provide a heating rate inside the cell of 0.1
0.3 C/min from ambient. When the lignin sample softened and
dripped down from the cup, blocking the path of the laser beam
to the photoresistor, the temperature at the change in resistance
was recorded as the softening point via a digital multimeter inter-
faced to a PC. The softening-point temperature was measured to
within 2 C with a calibrated RTD located inside the view cell.
Typically, the cell pressure was 6 bar when the softening-point
temperature was reached.
To evaluate the effect of the hydrating water as a plasticizer in
the liquid-lignin fractions, the fractions were rst dried in nitrogen
at 105 C for 24 h; then their softening points were measured in a
Mettler FP83HT dropping point cell, using an adaptation of ASTM
D6090-99.
2.4. Characterization of the liquid-lignin fractions
Inductively coupled plasma atomic emission spectroscopy (ICP-
AES) was used for the determination of sodium, potassium, and
sulfur in both spent black liquors and liquid-lignin fractions, using
a Spectro Analytical Instruments spectrometer, model ARCOS.
Analyses were carried out by our Agricultural Services Laboratory
at Clemson. The solids content of the black liquor and of the
J. Velez, M.C. Thies / Bioresource Technology 148 (2013) 586590 587
liquid-lignin fractions was measured using test method T 650 om-
05 of the Technical Association of the Pulp and Paper Industry
(TAPPI). Ash contents of the same were determined gravimetrically
after burning the samples at 575 C for 24 h.
3. Results and discussion
3.1. Precipitation behavior and bulk properties of liquid-lignin
fractions
The pH at the end of collection of a given fraction, the CO
2
con-
sumption, the liquid-lignin yield, the solids and water content, the
percent ash, and the softening point for each fraction are given in
Table 1. Mass balances (i.e., the mass of black liquor feed + the
mass of CO
2
consumed = the mass of liquid lignin produced + the
mass of spent black liquor remaining) were performed on each of
the fractions collected, and good closure (i.e., to within
95.5 1.2%) was obtained. Note that more than 90% of the lignin
precipitation occurred in the 11.610.0 pH range. Because the pK
a
s
of guaiacyl and syringyl phenols are in the range of 911 (Ragnar
et al., 2000), the phase separation of lignin via CO
2
acidication
would be expected to be primarily due to the protonation of those
types of phenolic groups in lignin.
Upon cooling, fractions 12, 57 exhibited a at, solid surface
with a mirror-like nish in the bottom of the pressure reactor after
decanting off the spent black-liquor phase. We thus concluded that
these fractions phase-separated as true liquid lignins at 115 C. On
the other hand, fractions 3 and 4 each separated out at 115 C as a
higher-viscosity liquid or solid phase that trapped some of the
black liquor, forming a mud-like mixture of the lignin fraction
and spent black liquor. These muds were vacuum-ltered using
Whatman paper grade 541 to separate the solidied liquid lignin
from the partially spent black liquor. Herein we denote fractions
3 and 4 as the lignins isolated after the black liquor had been re-
moved by ltering.
Table 1 shows how the liquid-lignin fractions all contain signif-
icant amounts of water, ranging from 32.3% to 48.2% by weight.
Thus, the lignin species that phase-separate from the black liquor
to form a lignin-rich phase during CO
2
acidication do so in strong
association with water. It is this solvating water that results in
the lignin phase being present in the liquid state at our operating
temperature and pressure of 115 C and 6.2 bar. Under ambient
conditions, this lignin would simply precipitate out as a solid. To
our knowledge, this phase behavior has not previously been re-
ported in the literature. A related phenomenon reported by Bert-
hold et al. (1996) is the adsorption of moisture by solid lignin
from ambient air, in which moisture levels in the lignin of up to
20 wt.% were reached at 98% relative humidity. Both the proton-
ated and sodium forms of the hydroxyl and carboxylic-acid groups
in lignin were said to be responsible for the adsorption behavior.
3.2. Softening points of liquid-lignin fractions
The softening points of the seven liquid-lignin fractions, as mea-
sured in the PVT view cell (see Fig. 1), are given in Table 1 and were
reproducible to within 0.5 C. The range of softening points ob-
served explains the differences in appearance noted above among
our various fractions. In particular, note that fractions 12 and 57
possess softening points well below the acidication operating
temperature of 115 C; thus, when these fractions precipitated,
they did so as liquids that coalesced to form a uniform liquid-lignin
Fig. 1. PVT view cell for the measurement of Mettler softening points of liquid-lignin fractions. The elevated pressures inside the cell prevent water loss from the lignin
fractions.
588 J. Velez, M.C. Thies / Bioresource Technology 148 (2013) 586590
phase, which then solidied upon cooling to give a mirror-like n-
ish. However, because the softening points of fractions 3 and 4 are
close to the operating temperature of 115 C, they precipitated
from solution either as viscous liquids or even as solids, and thus
did not easily separate from the partially spent black liquor. As a
check that the water-saturated environment inside the PVT cell
did indeed prevent the loss of solvated water from the lignin
fractions, the water content of the fractions was re-checked after
the softening-point measurements. Less than a 2 wt.% deviation
from the values reported in Table 1 was obtained.
The softening points of the lignin fractions, dried to remove all
solvating water, were also determined. In a compelling illustra-
tion of the plasticizing effect of water, no softening points were ob-
served for temperatures up to 375 C. This plasticizing effect has
also been observed in clean wood lignin, where the glass-transition
temperature of lignin decreased from 205 C dry (Back and Salmn,
1982) to 100 C in the water-saturated state (Salmn, 1984).
3.3. Elemental analyses of lignin fractions
Elemental-analysis results for the sodium, potassium, and sul-
fur content of the liquid-lignin fractions are presented in Table 2.
Mass balances for the above inorganics in the spent black liquor
and liquid-lignin phases closed to within 90%, so the results are
reliable. We dene for the inorganics a distribution ratio k
i
, which
is the concentration of component i in the liquid-lignin fraction
versus that same component in the accompanying spent, black-li-
quor phase. The k
i
< 1 for the metals means that a lower concentra-
tion of these elements is found in the liquid-lignin phase vs. the
spent, black-liquor phase. The lower metal contents observed for
fractions 3 and 4 are consistent with their lower ash contents
(see Table 1).
In Fig. 2, the tendency for a given liquid-lignin fraction to ex-
clude inorganics is shown in terms of the selectivity (b), dened
here as k
i
for the inorganic content divided by k
i
for the organic
content. In contrast to the denition used for liquidliquid extrac-
tions (Seader et al., 2011), here we have dened the selectivity
such that lower is better, with lower values meaning that the inor-
ganics are being preferentially excluded from the liquid-lignin
phase. Fig. 2 shows that the selectivities for sodium vs. organics,
potassium vs. organics, and sulfur vs. organics overlap for a given
Table 1
Final pH, CO
2
consumption, yield, solids and water content, ash analysis, and softening point for the liquid-lignin fractions generated via CO
2
acidication at 115 C and 6.2 bar.
Fraction no. Final pH
achieved
std L CO
2
used/100 g black
liquor
g liquid lignin/100 g black liquor
feed
Solids/water
(wt.%)
Ash
b
(wt
%)
Softening point
(C)
Feed
a
13.6 0.00 0.00 42.0/58.0 47.4
1 12.8 0.79 0.10 62.1/37.9 31.5 107.1
2 12.1 0.09 0.06 67.7/32.3 28.1 103.5
3 11.6 0.13 0.37 62.1/37.9 22.1 110.5
4 11.1 0.12 4.15 51.8/48.2 22.2 110.3
5 10.6 0.21 4.61 57.1/42.9 27.8 101.1
6 10.0 0.39 2.60 60.8/39.2 25.0 100.4
7 9.5 0.37 0.58 58.1/41.9 27.6 90.7
Unfractionated liquid
lignin
9.5 2.10 12.47 68.2/31.8 25.4 105.2
a
Feed refers to the softwood Kraft black liquor charged for the rst fractionation.
b
Ash content is on a dry basis.
Table 2
Weight percent of sodium, potassium and sulfur in the liquid-lignin fractions; distribution ratio (k
i
) of these elements between the liquid-lignin and the spent black-liquor phases.
Fraction # Final pH achieved Sodium (wt.%)
a
k
Na
b
Potassium (wt.%) k
K
b
Sulfur (wt.%) k
S
b
Feed 13.6 19.9 5.5 2.9
1 12.8 11.64 0.59 1.60 0.56 3.11 1.36
2 12.1 9.43 0.48 1.43 0.52 2.78 1.23
3 11.6 6.04 0.33 0.79 0.31 1.74 0.81
4 11.1 4.17 0.20 0.59 0.19 1.22 0.51
5 10.6 8.57 0.37 1.23 0.38 2.36 0.95
6 10.0 7.61 0.34 1.12 0.36 2.26 0.92
7 9.5 9.25 0.39 1.30 0.40 2.73 1.02
a
Weight percent of elemental Na/K/S in the liquid-lignin fractions on a dry basis.
b
k
i
is the distribution ratio of component i in the liquid-lignin (LL) fraction vs. that in the accompanying spent black liquor (SBL) phase: k
i
= x
i,LL
/x
i,SBL
.
Fig. 2. The selectivity of the liquid-lignin fractions for excluding the inorganics
sodium, sulfur, and potassium from the liquid-lignin (LL) phase vs. the spent
black-liquor (SBL) phase.
J. Velez, M.C. Thies / Bioresource Technology 148 (2013) 586590 589
pH, indicating that there may be a common structural property in
the lignin fractions responsible for the observed behavior.
4. Conclusions
If a Kraft black liquor is acidied with CO
2
at elevated temper-
atures and pressures and maintained in the liquid state, the lignin
that precipitates from the black-liquor solution does so as a lignin-
rich, highly hydrated liquid phase. The properties of this liquid-
lignin phase were characterized by fractionating it into cuts that
precipitated over narrow bands (0.5 units each) of pH. The soft-
ening points of the hydrated fractions all were found to be
<115 C, while no softening point was observed for dried fractions
tested up to 375 C. A signicantly lower metals content was ob-
tained for fraction 4 (collection pH = 11.611.1), which also had
the highest level of hydration (48.2 wt.% water).
Acknowledgements
This material is based upon work supported by the National Sci-
ence Foundation (Award No. CBET-1236759) and the Department
of Energy (DE-SC0003312). The authors wish to thank Drs. Michael
A. Lake and John C. Blackburn of Liquid Lignin Company for their
helpful discussions.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.biortech.2013.
08.097.
References
Back, E.L., Salmn, L., 1982. Glass transitions of wood components hold implications
for molding and pulping processes. Tappi 65, 107110.
Berthold, J., Rinaudo, M., Salmn, L., 1996. Association of water to polar groups;
estimations by an adsorption model for ligno-cellulosic materials. Colloids Surf.
A 112, 117129.
Doherty, W.O.S., Mousavioun, P., Fellows, C.M., 2011. Value-adding to cellulosic
ethanol: lignin polymers. Ind. Crops Prod. 33, 259276.
Frederick, J., Armstead, D., Lien, S., Schmidl, W., Kazem, B., 2001. Economic benets
of utilizing controlled cavitation technology for black liquor oxidation and
heating. Proceedings of the 2001 International Chemical Recovery Conference.
Whistler, BC, Canada, June 1114, 2001. Pulp and Paper Technical Association of
Canada: Montreal, pp 413420.
Garca, A., Toledano, A., Serrano, L., Egs, I., Gonzlez, M., Marn, F., Labidi, J., 2009.
Characterization of lignins obtained by selective precipitation. Sep. Purif.
Technol. 68, 193198.
Gouveia, S., Fernandez-Costas, C., Sanroman, M.A., Moldes, D., 2012. Enzymatic
polymerisation and effect of fractionation of dissolved lignin from Eucalyptus
globulus Kraft liquor. Bioresour. Technol. 121, 131138.
Holladay, J.E., White, J.F., Bozell, J.J., Johnson, D. Top Value-Added Chemicals from
Biomass. Volume II Results of Screening for Potential Candidates from
Biorenery Lignin. Report PNNL-16983; National Technical Information Service,
U.S. Department of Commerce: Springeld, VA, 2007.
Lake, M.A., Blackburn, J.C. Process for recovering lignin. Int. Pat. Appl. PCT/US2010/
049773, 2011.
Nagy, M., Kosa, M., Theliander, H., Ragauskas, A.J., 2010. Characterization of CO
2
precipitated Kraft lignin to promote its utilization. Green Chem. 12, 3134.
Ohman, F., Theliander, H., Tomani, P., Axegard, P. Method for separating lignin from
black liquor, a lignin product, and use of a lignin product for the production of
fuels or materials. PCT/SE2008/000142, 2009.
Ragnar, M., Lindgren, C.T., Nilvebrant, N., 2000. PKa-values of guaiacyl and syringyl
phenols related to lignin. J. Wood Chem. Technol. 20, 277305.
Salmn, L., 1984. Viscoelastic properties of in situ lignin under water-saturated
conditions. J. Mater. Sci. 19, 30903096.
Seader, J.D., Henley, E.J., Roper, D.K., 2011. Thermodynamics of Separation
Operations. Separation Process Principles, 3rd ed. John Wiley & Sons Inc.,
Hoboken, NJ, USA.
590 J. Velez, M.C. Thies / Bioresource Technology 148 (2013) 586590