Content uploaded by Antje Potthast
Author content
All content in this area was uploaded by Antje Potthast on Feb 05, 2021
Content may be subject to copyright.
Tools for Bark Biorefineries: Studies toward Improved
Characterization of Lipophilic Lignocellulosic Extractives by
Combining Supercritical Fluid and Gas Chromatography
Stefano Barbini, Dev Sriranganadane, Sebastian Espana Orozco, Armig Kabrelian, Katarina Karlstrom,
Thomas Rosenau, and Antje Potthast*
Cite This: ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
Read Online
ACCESS Metrics & More Article Recommendations *
sıSupporting Information
ABSTRACT: The bark of trees contains an interesting mixture of
bioactive compounds, or so-called extractives. The use of
supercritical carbon dioxide (sc-CO2) eliminates both the need
for organic solvents as extractants and the danger that solvent
traces might compromise the purity of the extracts. Unfortunately,
the complexity and natural variability of extracts’composition
render any utilization attempts rather challenging. Thus, in order
to implement exploitation concepts in a meaningful way,
appropriate analytical techniques for characterizing extracts must
be available beforehand. In our work, we explored gas
chromatography coupled to both mass spectrometry and a flame
ionization detector (GC-MS/FID), in combination with ultra-
performance convergence chromatography and quadrupole time-
of-flight mass spectrometry (UPC2-QTof-MS), for the characterization of bark extracts from pine (Pinus sylvestris L.) in both
qualitative and quantitative terms. Although the conventional GC-MS/FID approach is a robust method for overall quantification of
extractives, it fails to provide ample information about native sterol esters and triglycerides. These data are provided by a new,
complementary analytical technique based on supercritical carbon dioxide, as the chromatographic eluant, coupled to a high-
resolution mass spectrometer. The combination of both techniques and the use of sc-CO2as both an extraction solvent and eluant
made this combined tool especially powerful. The most prominent triglycerides in the extract were identified qualitatively and
quantitatively, and the dominating sterol esters were identified qualitatively, by UPC2-QTof-MS.
KEYWORDS: Bark, Biorefinery, CO2supercritical extraction, Nonpolar extractives, Quadrupole time-of-flight mass spectrometer,
Gas chromatography, Softwood, Pinus sylvestris, Ultraperformance convergence chromatography
■INTRODUCTION
Lignocellulosic extractives have been used for millennia and
have always been important in human history. The first utilized
extractive was amber,
1
a fossilized tree resin, and words such as
“electricity”and “electronic”are derived from the Ancient
Greek word for amber, ηλεκτρον. This natural pitch, also
known as rosin, was used in ancient medicine and for sealing in
the naval industry.
2
In current discussions about biorefinery,
lignocellulosic extractives are once again receiving attention,
and they play an essential role due to their broad spectrum of
applications.
3
They can be used as an adhesive for wood chips
and wood fibers,
4
providing additional economic value. Highly
toxic synthetic wood preservatives, such as chromated copper
arsenate (CCA) or creosote, can be partially replaced with
nontoxic extractives.
5
The overall content of extractives varies
greatly between tree species and tissues, ranging from 21% for
Picea abies L. Karst.,
6
of which roughly 5% is extracted by
nonpolar solvents as dichloromethane (DCM) or petroleum
ether,
7,8
to 40% for Schinopsis lorentzii Engl. based on the dry
mass of wood or bark.
4
The bark obtained from softwood
trees, such as Scots pine (Pinus sylvestris L.) and Norway
spruce (Picea abies L.), is still a largely underused byproduct
from European pulp mills and round wood production.
Moreover, since the production of round wood in the EU-27
continues to increasein 2018 it was 21% higher than in
2000
9
the total bark availability increases as well. Currently,
bark is mainly incinerated for heat and energy, but it could be
extracted to obtain nutraceuticals (antioxidants, products with
cholesterol-lowering properties), cosmetics (fragrances), and
Received: October 29, 2020
Revised: December 6, 2020
Published: December 29, 2020
Research Article
pubs.acs.org/journal/ascecg
© 2020 American Chemical Society 1323
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
This is an open access article published under a Creative Commons Attribution (CC-BY)
License, which permits unrestricted use, distribution and reproduction in any medium,
provided the author and source are cited.
Downloaded via 213.47.149.221 on February 5, 2021 at 13:04:06 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
pharmaceuticals (anti-inflammatory drugs), provided that
appropriate extraction methods and utilization concepts
become available. Arshadi et al.
10
pointed out that different
assortments of softwood bark derived from Scots Pine and
Norway Spruce could be used as an industrial feedstock for tall
oil production when more quantitative information is
provided. Well-known is also the presence of a high
concentration of polar extractives in bark as tannins,
commonly extracted by hot water or polar organic solvents
(EtOH, MeOH), which are sold as nutraceutical pills (i.e.,
Picnogenol). Overall, these classes of molecules could best be
used in different commercial segments for valorization if
detailed information on their qualities and quantities is
available.
11
Supercritical CO2(scCO2) is already used as a
green solvent for decaffeination of coffee, which was the first
scCO2process to be commercialized.
12
It has replaced
nonpolar organic solvents, which proved to be harmful for
the environment. On the contrary, CO2, which possesses a low
critical pressure and temperature (7.4 MPa, 31 °C), is
nontoxic, cheap, and nonflammable. It can be recycled
indefinitely during extractions. Since it is gaseous at ambient
conditions, it does not leave any traces, allowing the
classification as food-grade. The most lipophilic classes of
compounds contained in bark, such as fatty acids, resin acids,
acylglycerides, and sterol esters, can be extracted by scCO2.
13
Moreover, polar organic solvent in scCO2, such as EtOH
(“entrainer”), offers an increasing solubilization of more polar
compounds, such as lignans, triterpenoids, and polyphenols.
Compound classes similar to extractives found in lignocellu-
loses are already being extracted with success, such as fatty
acids from pomegranate seed oil, yacon leaves, or spent
barley.
14
Economic analysis of supercritical fluid extraction
(SFE) processes have to be evaluated for future commercial
applications. Capital costs are usually not linear with increasing
pressure of extraction, and although operation costs and
regulations are generally less expensive and regulated than
handling flammable organic solvents, continuous, batch, or
semicontinuous approaches have to be considered for larger
units.
15
Generally, barks tend to have a much higher amount of
extractives than their respective wood counterparts,
6
making it
ideal to obtain extractive fractions. Extractives are divided into
lipophilic and hydrophilic compounds.
16
Resin acids, free fatty
acids, acylglycerols, sterol esters, free sterols, and mono- and
diterpenoids form the lipophilic part, whereas sugar, salts,
simple phenols, and polyphenols (such as lignans, flavonoids,
and tannins) are hydrophilic compounds. The complexity of
extractives obtained from wood and bark inevitably renders
their analysis a challenging task.
17
Chromatographic techni-
ques yield more in-depth characterization; the suitability of gas
chromatography (GC) for analysis of volatile and low
molecular mass compounds is obvious. Coupling of GC with
mass spectrometry (MS) or a flame ionization detector (FID)
is a standard approach to the analysis of lignocellulosic
extractives.
18,19
GC-MS is more useful for qualitative character-
ization, while FID is better suited for compound quantification,
although both identification and quantification can be derived
from GC-MS analysis if standards are used.
19,20
Unfortunately,
GC techniques are not suitable for the analysis of thermolabile
and nonvolatile compounds.
19,20
Recently, high-performance (ultraperformance) liquid chro-
matography (HPLC/UPLC) was used in combination with
MS for an overall identification of compounds.
21
Various
studies have used soft ionization methods to quantify phenolic
compounds, such as apigenin, caffeic acid, myricetin, quercetin,
or rutinin, from the methanolic extract of bark, flowers, or
fruit.
22,23
In this study, bark extractives from Scots pine, which
is of great relevance to bark biorefinery strategies, were
obtained by extraction with scCO2, a sustainable and
environmentally friendly method to access mid-polar to
nonpolar compounds. We advanced the characterization of
lipophilic extractives by applying UPC2-QTof-ESI-MS. To
date, UPC2-QTof-MS has been applied in biorefinery analyses
of the monomers and oligomers present in degraded
lignin,
24,25
metabolite profiling in the leaves of yerba mate
(Ilex paraguariensis A. St. Hil.) and white willow bark (Salix
alba L.),
26
and quantification of different phenolic and
glycosidic constituents in Liquidambar L. trees or Verbena
officinalis L. extracts.
27,28
In this study, UPC2-QTof-ESI-MS
technique is tested for its compatibility with GC-MS/FID and
for the extent to which it enables qualitative and quantitative
determination of compounds that would be difficult to analyze
otherwise. In addition, strategies for automatic identification of
extract constituents using MZmine 2
29
are presented and
discussed.
■EXPERIMENTAL SECTION
Material and Sample Preparation. Bark from Scots pine (Pinus
sylvestris L.) was obtained from Svenska Cellulosa AB (Sundsvall,
Sweden). The trees were harvested in February 2018 and debarked
on-site after high-pressure hot water treatment intended to defreeze
the logs. The bark had a water content of 54 ±2% determined by
drying it at 105 ±1°C to constant weight (n= 10). For extraction,
the bark was dried at 50 °C for 48 h until a final water content of 5 ±
1% was reached. Then, it was milled using a Granulator Retsch SM1
fitted with a 6.0 mm sieve. To narrow the particle distribution,
screening was applied using tower sieving (Retsch AS 200) involving
five sieves with decreasing diameters (4, 2, 0.8, 0.4, and 0.2 mm). The
amplitude of shaking was set to 50, and the time was set to 10 min for
aliquots of bark of about 200 g. No other milling step was applied in
order to avoid degradation of heat-sensitive compounds. The sample
was stored in sealed plastic bags (polyethylene, PE) at −80 °C until
further use.
Supercritical Fluid Extraction. Supercritical extraction of pine
bark was carried out with carbon dioxide (>99.5% food grade, Biogon
C, Linde AG, Austria) using a high-pressure SF-1 supercritical fluid
system (Separex, Champigneulles, France) equipped with three
subunits: a preheater, autoclave, and separator. Samples of pine
bark (approximately 12 g per batch, n= 3) were individually extracted
in dynamic mode at 30 MPa and 40 °C(ρ(CO2), 910 kg m−3) using a
cylindrical extraction vessel (i.d., 35 mm; height, 50 mm). The
particle diameter (DP) of pine bark was between 0.2 and 0.4 mm. To
enhance the solubility of polar compounds (e.g., polyphenols), 1%
(weight basis) of ethanol 96% (v v−1;ρ20 °C, 808.5 kg m−3; Merck)
was added as a cosolvent using an HPLC pump (Lab Alliance). The
total mass flow rate (CO2+ EtOH) and solvent to sample ratio were
set to 45 g min−1and 84 g of CO2g−1of oven-dried bark, respectively.
To avoid loss of extract during depressurization, a stainless steel tube
was connected to the outlet valve exiting from the separator and
immersed in an Erlenmeyer flask filled with 100 mL of cold EtOH
(ice bath). The bark extract was dried under reduced pressure and
stored at −80 °C in brown vials until further use.
Accelerated Solvent Extraction. To determine the complete-
ness of the scCO2extraction, pressurized liquid extractions (n=3)
were carried out using the same batch of bark (particle diameter, 0.2 <
Dp< 0.4 mm) by accelerated solvent extraction (ASE) in a 34 mL
extraction cell. The solvent, which has to be comparable in the
polarity range with scCO2, was 9:1 n-hexane/acetone (v v−1).
30
An
ASE 350 (Dionex, Sunnyvale, CA, USA)
31
was used on the basis of
conditions used by Willfor et al.:
32
120 °C (temperature), 6 min
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1324
(heating time), four cycles, 60 s (purge time), 40% (rinse volume),
and 15 min (static time).
Extraction Efficiency. The extraction yield obtained by scCO2
(+1 wt % of EtOH as cosolvent) is equal to 49.89 ±1.00 mg g−1
oven-dried bark as it is reported in Table S2 (Supporting
Information), while the extraction yield achieved by ASE is 51.15 ±
0.35 mg g−1oven-dried bark. Therefore, the efficiency of the former
was 97.5%.
Gas Chromatography: GC-MS/FID Analysis. Extracts were
analyzed with GC as their trimethylsilyl derivatives. 200 μLof
silylating agent (9:1 (v v−1)ofN,N-bis(trimethylsilyl)-
trifluoroacetamide (BSTFA, ≥99%, Sigma-Aldrich) and trimethyl-
chlorosilane (TMCS, ≥99%, Sigma-Aldrich) were added to each vial,
which contained 1.5−5.0 mg of homogenized crude extract (yellowish
and viscous oil) and four internal standards: n-heptadecanoic acid
(98%, Alfa Aesar, CAS Reg. No. 506-12-7); cholesterol (>99%,
Sigma-Aldrich, CAS Reg. No. 57-88-5); cholesteryl palmitate (>98%
HPLC, Sigma-Aldrich, CAS Reg. No. 601-34-3); and glyceryl
triheptadecanoate (>99%, Sigma-Aldrich CAS Reg. No. 2438-40-6).
The vials were vortexed and heated at 70 °C for 1 h. Then, the
samples were transferred to a glass insert and injected into the GC-
MS/FID system. Synchronized recording of FID and MS spectra was
performed by a splitter mounted after the column. About one-third of
the flow was sent to the MS, and the remaining two-thirds to the FID.
The MS trace was used for identification, and the FID trace for
quantification. An Agilent 5975C apparatus with MSD inert XL TAD
and FID was used. The MS detector was operated in the electron-
impact (EI) mode at 70 eV using a temperature of 280 °C. The mass
scanning range was set to 29−1050 amu, and the solvent cutting time
was 4 min. The FID was operated at 400 °C, with H2flow of 30 mL
min−1, air flow of 400 mL min−1, and makeup flow (combined) of 25
mL min−1. The device was fitted with a DB-5HT capillary column (30
m×250 μm×0.1 μm, Agilent J&W). The column temperature
program was set as follows: initial T=60°C isothermal for 5 min,
ramp to 380 °C (rate, 10 °C min−1), and maintain at 380 °C for 8
min. Helium was used as a carrier gas, with a gas flow of 2.5 mL
min−1. Injection (1.0 μL) was performed by an autosampler in a cold
multimode inlet (MMI), which was kept at 65 °C for 6 s, increased to
380 °C at 500 °C min−1, and then held for 5 min (cold split
injection). The split ratio was set to 15:1 (split flow, 37.5 mL min−1).
The total analysis time was 45 min. The MS data were collected with
an Enhanced ChemStation (MSD Chemstation F.01.01.2317),
deconvoluted, and then evaluated using the MassHunter Workstation
software (Unknowns Analysis). The compounds were identified by
comparison with the Wiley10 and NIST11 mass spectral library. The
resolution of the retention time window size factor was set to 100 for
deconvolution, while the minimum match factor was set to 50. A
solution in hexane (10 μgmL
−1) containing a series of saturated
alkane standards from C7H16 to C40H82 (1000 μgmL
−1in hexane,
Merck) was injected using the same GC method. An RTC format file
was generated from n-decane (C10H22)ton-tetracontane (C40H82) for
each standard. Comparison of the calculated and library-reported
retention indices was performed for each compound, and a maximum
difference of ±15 was considered acceptable. For quantification of the
extract, the FID chromatogram was exported in CSV format and
evaluated using R.
33
Baseline correction was applied using the “rolling
ball”algorithm developed by Kneen and Annegarn on Pascal
34
and
translated into R by Liland et al.
35
The width of the local window for
minimization/maximization was set to 700 (wm), while the width of
the local window for smoothing was set to 175 (ws). The baseline
correction was checked visually for each spectrum. The integration
method was applied using trapezoidal integration within R. The
standard deviation (σ) of the quantification data was calculated on the
basis of three replicates, while the precision of the method was
evaluated using relative standard deviation (RSD%).
UPC2-ESI-QTof-MS Analysis. All MS experiments in positive and
negative modes were performed with an Acquity UPC2instrument
(Waters, Milford, MA, USA). The final method for extractive analysis
was performed with a Torus 2-PIC column (3.0 mm i.d. ×50.0 mm
length, 1.7 μm particle size, Waters) equipped with a precolumn of
the same chemistry (VanGuard, 2.1 mm i.d. ×5.0 mm length) under
the following conditions: flow rate of 1.2 mL min−1, injection volume
of 1.0 μL, column temperature of 45 °C, automated back-pressure
regulator (ABPR) pressure of 2000 psi (137.9 bar), and autosampler
set at 20 °C. In the positive mode, carbon dioxide (>99.995%, Linde
AG, Austria) was used as solvent “A”and solvent “B”(modifier) was
composed of a methanol/acetonitrile mixture (1:1, v v−1, LC-MS
grade, Fisher Chemicals) containing 25 mM ammonium formate (pH
= 5.7, >99% LC-MS grade, VWR). The conditions were set as follows:
0 min, 1%; 10 min, 20%; 11.5 min, 40%; 12 min, 1%; 15 min, 1%
(curve 6). In the negative mode, solvent B was composed of
ammonium hydroxide methanol (25 mM, pH = 10.8, ACS 28−30%
NH3basis, Sigma-Aldrich), and the gradient was the same as in the
positive mode. The injector needle was subjected to strong (2-
propanol, 600 μL) and weak (methanol, 200 μL) washing steps after
each injection. The UPC2instrument was combined with the mass
spectrometer using a commercial interface kit (Waters) composed of
two T-pieces that enabled the back-pressure control (ABPR) and
mixing of column effluent with a makeup solvent delivered by the
isocratic solvent manager (Waters). A decreasing flow rate of
methanol/2-propanol/water mixture (7:2:1 (vol), LC-MS grade,
Fisher Chemicals) was set on the basis of the increasing percentages
of the modifier in the chromatography run: 0 min, 0.3 mL min−1;10
min, 0.2 mL min−1; 11.5 min, 0.1 mL min−1; 12 min, 0.3 mL min−1;
15 min, 0.3 mL min−1. The hybrid QTof-MS, XEVO G2-XS QTof
(Waters), was used in the resolution mode with both positive-ion and
negative-ion ESI modes. It was used in the mass range of 50−2000 m/
zwith the following tuning parameters: capillary voltages of 3.0 and
2.5 kV for the positive-ion (+) and negative-ion (−) modes,
respectively; sampling cone of 50 V (+)/20 V (−); source offset of
90 V; source temperature of 150 °C; desolvation temperature of 500
°C; cone gas (N2)flow of 48 L h−1; and desolvation gas flow (N2)of
1020 L h−1(+)/1018 L h−1(−). The lock mass for automated mass
correction was based on the monoisotopic mass of leucine-
enkephaline (Waters): positive [M + H]+= 556.2771 Da and
negative [M −H]−= 554.2615 Da. A scan time of 0.50 s, interval of
20 s, mass window of ±0.5 Da, infusion flow rate of 5.0 μL min−1, and
capillary voltage of 2.20 kV (+)/2.55 kV (−) were set as LockSpray.
Argon (5.0, Linde AG, Austria) was used as collision gas for MS/MS
experiments, which were performed on a transfer cell with a ramp of
collision energy from 25 to 65 eV.
Data Mining with ESI-QTof-MS. Raw data were analyzed using
MZmine 2 software
29
(version 2.51, downloaded from http://
mzmine.sourceforge.net), which applies four stages of processing:
peak detection (the noise level was set to 5.02e2 for the positive mode
and 9.0e2 for the negative mode), ADAP
36
chromatogram building
(Automated Data Analysis Pipeline algorithm module for chromato-
gram deconvolution), isotope peak grouping, and peak identification
(with a homemade library containing 4,700 known extractive
compounds). All detected peaks were identified using filtering with
a mass shift error not exceeding ±5 ppm or ±0.0075 Da. To verify
different identified peaks, we decided to evaluate the retention time
within the same class of compound. Three main molecular features
were found to be correlated with the retention time: increasing
polarity (e.g., hydroxyl groups), increasing number of carbon atoms
(CNs), and increasing number of double bonds (DBs). Each detected
peak was subjected to manual validation by evaluation of the MS/MS
fragmentation pattern using MassLynx software (version 4.2, Waters).
■RESULTS AND DISCUSSION
To obtain extractives from bark, we applied scCO2extraction,
which is a green, almost-solvent-free way to access extractives.
Coupling with MS should provide the required identification
power in a UPC2-QTof-ESI-MS approach. To obtain a
benchmark comparison, the GC-MS/FID method, which is
commonly used for analysis of lignocellulosic extractives, was
used. For bark extractives, which are a complex mixture of
different families of molecules, both techniques give different
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1325
analytical responses. To understand how both techniques can
provide complementary support for deciphering the composi-
tion of complex samples, the two analytical systems were
compared and, later, combined to achieve a better under-
standing of lipophilic bark extractives.
GC-MS/FID Analysis. Classes of Molecules. GC-MS/FID
provides a general overview of the samples, as all of the
compound classes can be addressed quantitatively. The high-
temperature version also permits analysis of sterol esters (SEs)
and triglycerides (TGs). The FID/MS splitter enables
simultaneous quantification and identification. The retention
times can be matched accordingly, and the FID trace is used
for more reliable quantification. The compound classes were
quantified on the basis of four internal standards. This method,
which was initially developed by Ekman and Holmbom
18
and
later improved by Orså and Holmbom,
19
relies on the concept
that the discrimination in derivatization, injection, and FID
response is the same for all members of a class of compounds.
The FID chromatograms of each extract were divided into
eight main regions, as outlined in Table S1 (Supporting
Information) after identification based on the MS trace. The
eight intervals of integration (cf. Figure 1) are not intended to
indicate a neat separation between each class of compound.
The relative response factors (RRFs) for each identified
analyte were calculated relative to its internal standard using
the method described by de Saint Laumer et al.
37
The most
abundant compound (analyzed by FID) in each interval of
integration was chosen as reference compound for the
response factor calculation (Table S2, Supporting Informa-
tion). The results of the quantification of the pine bark extract
by GC-MS/FID are given in Table S3 (Supporting
Information). On average, 57% of the pine bark extract
(based on weight), which was effectively eluted from the high-
temperature capillary column, was qualitatively assessed by MS
(% identified, MS/FID; Table S3, Supporting Information).
Quantification based on the FID signal resulted in description
of 76% (based on weight) of the extract (% detected, FID/
wextract;Table S3, Supporting Information). Qualitative analysis
of free fatty acids and resin acids reached 84% and 92% of the
identified peak area, respectively. In total, 60% of the low
molar mass components were identified, while 71% of
phytosterols were assigned. All of these data were based on
the overall area detected by FID (76%). From lignans to later-
eluting compounds, such as triterpenoids, diglycerides, sterol
esters, and triglycerides, a qualitative evaluation of MS
produced unacceptable results in the case of lignans (6%) as
expected, while in the case of diglycerides, sterol esters, and
triglycerides, no structures were identified.
Individual Compounds. As shown in Table S4 (Supporting
Information), individual compounds found in the extract were
validated using match factors and the retention index. Figure
S1 (Supporting Information) provides the MS and FID traces
for pine extract from between 35 and 45 min (T> 360 °C). It
is clear that the FID still allows for ionization of the two
standards and relative compounds of the same family, and
therefore, quantification was possible. The MS signal, however,
does not give sufficient structural information. Improvements
for substances with a high boiling point are possible if one uses
a significantly shorter column (15 m) or cold on-column
injection to reduce thermal stress and consequently produce
good ionization.
38−42
Another option, which was carried out
by Gutierrez et al.,
43
is to perform prefractionation via solid-
phase extraction in order to concentrate SEs and TGs. This
operation will result in an enriched fraction. However, it is a
laborious procedure.
The limit for TGs and SEs in GC-MS analysis has been
stated in the literature.
44
Highly unsaturated and saturated oils
that require very high temperatures (up to 380 °C) may
decompose thermally, leading to free fatty acids and
diglyceride-like compounds, likely enol esters.
45
To avoid
large molar masses, a three-step approach is usually applied:
isolation by column chromatography or preparative TLC,
saponification of esters in the presence of bases (NaOH or
KOH), and then silylation or methylation of free components
to form more volatile compounds.
7,46,47
This addresses the
overall composition of individual sterols (from sterol esters)
and fatty acids (from TGs and SEs) as free compounds, but
fails to report them in their native form.
According to the GC-FID trace for the pine bark extracted
with scCO2, there are 0.96 ±0.02 mg g−1of sterol esters and
2.95 ±0.14 mg g−1of triglycerides (Table S3, Supporting
Information). Unfortunately, due to the low volatility of these
compounds, assignment of their structures by MS was not
possible. To the best of our knowledge, there is no available
literature on the native structures of SEs and TGs from the
wood or bark of Pinus sylvestris. Anas et al.
7
showed that the
SEs and TGs contained in inner and outer bark extracted from
Scots pine with petroleum ether were predominantly
unsaturated. Sterol esters were composed of β-sitosterol,
campesterol, and β-stigmastanol combined with different
unsaturated fatty acids. For both groups, oleic, linoleic, and
pinoleic acids were the principal components, and the other
components were saturated acids, such as arachidic, behenic,
Figure 1. FID chromatogram of pine bark extract (where, for example, 4E+6 represents 4 ×106): LMW, low molecular weight compound; FA, free
fatty acids; RA, resin acids; ST, free sterols; LG, lignans; DG, diglycerides; SE, sterol esters; TG, triglycerides; IST1, n-heptadecanoic acid; IST2,
cholesterol; IST3, cholesteryl palmitate; IST4, glyceryl triheptadecanoate. The red line shows the gradient of temperature.
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1326
and lignoceric acid. Coniferonic acid was also detected in pine
bark. Any quantitative data obtained from the literature cannot
be compared since, as pointed out by Nisula,
48
“the natural
variability is so large that it would be impossible to collect and
study an adequate number of samples for reliable statistics.”
Moreover, there are simply too many factors that influence the
amount of lignocellulosic extractives: the origin of the sample,
harvesting season, storing conditions, sample preparation,
extraction method, analysis method, and data evaluation.
Nevertheless, the chemical composition of extractives should
be similar when all of these factors are taken into
consideration. In our case, the qualitative composition of the
extract was in agreement with that of Anas et al.
7
UPC2-QTof-ESI-MS Analysis. As was obvious from the
discussion about GC-MS/FID, the most problematic region is
comprised of the less volatile compounds. Hence, we wanted
to explore the potential of UPC2-QTof-ESI-MS for bark
analysis in comparison to GC-MS/FID, focusing on the speed
and qualitative analysis of TGs and SEs. In addition, we
wanted to demonstrate the capability of UPC2-QTof-ESI-MS
to identify TGs and SEs in their native states. Since the original
extraction step is based on scCO2, we chose to use UPC2with
the same solvent, adopting an approach that combines
chromatographic separation with mass spectrometry. Since
the UPC2method is an innovative method for qualitative and
quantitative analysis of extractives, we chose glyceryl tris-
(heptadecanoate), cholesterol, and n-heptadecanoic acid as
representative internal standards to address the matrix effect,
and glyceryl trilinoleate, stigmasterol, and n-octadecanoic acid
to calibrate unsaturated TG, sterols, and fatty acids. The third
Figure 2. Base peak ion chromatogram of pine extract obtained in the positive mode using UPC2-QTof-ESI-MS as well as classification of detected
peaks.
Figure 3. Base peak ion chromatogram of pine extract obtained in the negative mode using UPC2-ESI-QTof-MS as well as peak assignment. Fatty
acids: C16:0 (n-hexadecanoic acid), C18:0 (n-octadecanoic acid), C18:1 (octadecenoic acid), C18:2 (octadecadienoic acid), C18:3
(octadecatrienoic acid), C20:2 (icosadienoic acid), C22:0 (icosanoic acid), C24:0 (tetracosanoic acid), and C26:0 (hexacosanoic acid). Resin
acids and derivatives: RA (resin acids), deRA (dehydrogenated resin acids), OH-RA (hydroxy resin acids), OH-RA-deRA (dimer: hydroxy resin
acid + dehydrogenated resin acid), OH-deRA (hydroxy dehydrogenated resin acids).
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1327
internal standard previously used in GC, cholesteryl palmitate
(ISTD3; cf. Figure 1), was not included in the UPC2analysis
because it requires the corresponding sterol ester contained in
the bark. Quantification by single phytosterols’ester standards
is limited by the commercial availability of those compounds,
but synthesis is available from literature.
49,50
To identify TGs
and SEs in their native forms, a homemade library using
MZmine 2 was applied to screen compounds found in pine
bark that are already known in literature
7,51−54
as well as
possible polyphenols present in food.
55−57
In total, 4,700
compounds were collected in a single library, and how the
compounds were ionized in both the positive and negative
modes, forming various different adducts, was studied on the
basis of both previous works
26,54,58
and available standards.
Application of MZmine 2 and the generated database resulted
in 472 identified substances out of 2,164 detected peaks in the
positive mode, which were grouped into 11 main classes of
molecules (Figure 2). Likewise, 60 substance peaks were
identified out of 272 detected peaks in the negative mode,
which were assembled into six groups of molecules (Figure 3).
After a review of the literature presenting analyses of
metabolites using UPC2,
59
it was clear that most of the
compounds (>90%) were subordinated to a simple rule, as
described in the materials section, which was applied to verify
the different classes of detected “extractives.”For the present
study, we have concentrated our attention on SEs and TGs. In
future studies, this will be extended toward other classes of
extractives found in softwood trees.
Sterol Esters. Upon injection of the pine bark extract into
the UPC2system and data mining, 14 different native sterol
esters were assigned on the basis of their molecular formula
(Table S5, Supporting Information). All of them were eluted
following the generic rule of order of elution, which is based on
the number of carbon and double bonds. The equivalent
carbon number, as in RP-HPLC, can be used as a parameter
with some adjustments. In RP-HPLC, since the stationary
phase is nonpolar, the components with fewer double bonds
(higher equivalent carbon number) will elute later because
they are more lipophilic and hence are retained longer by the
stationary phase.
44
However, UPC2based on a polar phase
shows the opposite behavior; compounds with fewer double
bonds will elute early. Other characteristics can influence the
retention time within the same family of compounds, such as
molecule conformation, cis−trans geometry of double bonds,
or aggregate formation, but these particular properties were
not targeted in our investigation. A bigger obstacle than
anticipating the retention time of sterol esters was highlighted
by Caboni et al.
60
and Hailat et al.:
61
sterol esters are difficult
to ionize due to their nonpolar character. Therefore,
ammonium acetate or ammonium formate are used to facilitate
their detection. Nevertheless, in our case, the use of
ammonium formate as an additive did not form ammonium
adducts ([M + NH4]+are the most intense ones), but sodium
adducts were detected ([M + Na]+). Moreover, APCI-MS
usually fragments the molecules that are already in the source
during ionization before the mass analyzer (in-source
fragmentation). Hence, only the fragment of the molecule
formed by loss of the corresponding attached fatty acid is
detected ([M −RCOOH + H]+). A similar issue was also
encountered when we used ESI instead of APCI, aided by the
fact that a very high capillary voltage (3.0 kV) was applied in
the positive mode in order to ionize as many molecules as
possible. In-source fragmentation of sterol esters ([M −
RCOOH + H]+) led to overlaying of detected masses, which
could be assigned to sterols as the [M −H2O+H]
+adduct.
This problem was solved by injecting some available standards
of sterols and sterol esters and comparing their retention times.
More hydrophobic sterol esters eluted earlier than free sterols.
Hence, all of the sterols falsely positive detected as [M −H2O
+H]
+adducts eluted before the retention time of the
cholesterol standard. First eluted free sterols were considered
as the results of in-source fragmentation of sterol esters as [M
−RCOOH + H]+adduct. Among the 14 native sterol esters
detected by UPC2-ESI-MS, only the six most intense ones were
chosen to be clearly identified. Identification of the correct
sterol ester among different constitutional isomers was
performed by MS/MS. Breaking the weakest chemical bond
of each sterol ester, which is the ester linkage, forms the
corresponding fragments [M −RCOOH + H]+and [RCOOH
+Na]
+, from which it is possible to find the correct
composition of sterol and fatty acid. According to the
chromatogram depicted in Figure 4a, where three of the six
identified sterol esters are shown, the most intense sterol esters
elute between 1.71 and 1.93 min. The retention time of this
class of molecules, the ionization pattern, and the fragments
formed by MS/MS were also found for single injection of
cholesteryl palmitate, which eluted at 1.00 min and formed [M
+ Na]+as the main adduct.
Here, the concept of the assignment is explained. Starting
from the most intense sterol ester, at 1.79 min, a peak of m/z
697.5907 (C47H78O2) [M + Na]+was observed, correspond-
ing, in theory, to three different types of sterol esters: β-
sitosteryl octadecatrienoate (sterol ester S,Ln: β-sitosterol +
octadecatrienoic acid C18:3), stigmasteryl octadecadienoate
(sterol ester Ste,L: stigmasterol + octadecadienoic acid C18:2),
and stigmastanyl octadecatetraenoate (sterol ester Sta, Co:
stigmastanol + octadecatetraenoic acid C18:4).
Figure 4. (a) Base peak ion chromatogram of the three most intense
sterol esters according to peak height in pine bark extract analyzed in
the positive mode by UPC2-QTof-ESI-MS. The double bonds
associated with each esterified fatty acid were not studied in detail,
and no information about the position or cis/trans geometry was
obtained. (b) Intensity ion count of each parent peak (725.6226;
697.5907; 699.6058) fragmented by MS/MS experiment. The
common ion, 397, indicates that the β-sitosteryl unit is a [M −
H2O+H]
+adduct, while the 329.2302, 301.2137, and 303.2295 ions
indicate that the C20:3, C18:3, and C18:2 fatty acids, respectively, are
[M + Na]+adducts. The colors indicate each sterol ester.
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1328
As highlighted in Figure S2 (Supporting Information), all of
them show the same monoisotopic mass and, hence, the same
adduct mass with sodium. Assisted by MS/MS, it was possible
to confirm that the structure of this sterol ester was β-sitosteryl
octadecatrienoate due to the fragment at m/z397.3834 at the
same retention time. This fragment was associated with the
parent peak m/z697.5907, which corresponds only to β-
sitosterol that has lost octadecatrienoic acid [M −RCOOH +
H]+(Table S6, Supporting Information). The nature of the
fatty acids was also proved by the presence of the m/z
301.2137 [M + Na]+adduct formed by cleavage. The same
principle was applied to the next five sterol esters, which were
detected and identified (Table S7, Supporting Information).
Triglycerides. Twenty native TGs were assigned on the basis
of the molecular formula, with elution between 2.09 and 2.72
min (Table S8, Supporting Information). The order of elution
corresponded to the equivalent carbon number described for
the SEs. The majority of TGs within the acylglyceride family
(mono-, di-, and triacylglycerides) eluted before DGs (Table
S9) and MGs (Table S10), as shown in Figure S3.
The major driver for this elution pattern is the number of
free hydroxyl groups (none for TGs, one for DGs, and two for
MGs) on the glycerol backbone. The same behavior was also
found by Li
sa and Holcapek,
58
who were also able to separate
1,3-diglyceride from later-eluting 1,2-diglyceride and 1-
monoglyceride and from later-eluting 2-monoglyceride. The
five most intense TGs according to peak height are given in
Figure 5.
The five most intense TGs eluted as follows: 2.28 min (m/z
900.8051), 2.34 min (m/z898.7904), 2.38 min (m/z
896.7733), 2.45 min (m/z894.7564), and 2.48 min (m/z
924.8031). All five TG peaks were detected as both the [M +
NH4]+adduct, which was chosen as a precursor ion for MS/
MS evaluation, and the [M + Na]+adduct. For quantification
purposes, both adduct intensities were combined. For
calibration, glyceryl trilinoleate (TG L,L,L) was used as a
standard, and each concentration was corrected using the
matrix effect obtained from the internal standard glyceryl
tris(heptadecanoate) (Ma,Ma,Ma), which was spiked to the
pine bark as a matrix (Table S11, Supporting Information).
The triglycerides eluted according to the number of double
bonds and the number of carbons. Thus, according to the
equivalent carbon number. This approach was used in several
studies
62,63
for qualitative analysis of different TGs. MS/MS
fragments originating from the NH4+adduct of the precursor,
which was easier to fragment than sodium adduct, were
assigned in line with Lee et al.
52
From the relative intensity of
each single diglyceride-like fragment, confirmation of the
regioselectivity was performed for some variants, as described
by Laakso and Voutilainen
64
and later by Leskinen et al.
65
The
fatty acid esterified at sn-2 requires more energy to be
fragmented than fatty acids esterified at sn-1 and sn-3 due to
the lower steric hindrance at the latter two positions.
Therefore, the fragments obtained by losing fatty acids from
the sn-1 or sn-3 positions would have a higher relative intensity
than fragments generated by the loss of fatty acids from the sn-
2 position. Lee et al.
52
indicated that, on the basis of the
product ion scan data, fatty acids with higher numbers of
carbons and double bonds were more easily fragmented than
fatty acids with lower numbers of carbons and double bonds at
the sn-1/-3 position. Based on previous studies
66−71
on the
assignment of TGs, we report the five most abundant peaks
corresponding to TGs with the relative fragments formed by
the loss of one fatty acid chain, [M −RCOOH]+.
As shown in Table S12 (Supporting Information), the first
two peaks (2.28 and 2.34 min) were assigned to single
triglycerides based on the fragments generated by MS/MS,
TG(O,L,O), and TG(L,L,O) respectively. The last three peaks
(2.38, 2.45, and 2.48 min) were assigned to a mixture of two,
three and four isomers, respectively. The separation of isomeric
triglycerides which have the exact same elemental composition,
but represent different compounds, can be done using ion
mobility spectrometry (IMS).
72
The TG calibration curve was
obtained just from one triglyceride TG(L,L,L) and applied to
all triglycerides.
Fatty Acids and β-Sitosterol. Semiquantification of the
most abundant fatty acids and β-sitosterol detected was
performed by UPC2-QTof-MS. Fatty acids were separated with
UPC2using methanol and ammonia as additives, and detected
as the deprotonated adduct [M −H]−in the negative mode
(Table S13, Supporting Information). No free sterols, such as
stigmasterol and campesterol, which were detected by GC-MS,
were found in pine bark extract. However, β-sitosterol was
found as the sodium adduct [M + Na]+at 3.63 min in the
positive mode. The results of UPC2quantification of fatty acids
and β-sitosterol (blue bars, Figure S4, Supporting Information)
was less precise than that of GC-FID (yellow bars, Figure S4,
Supporting Information). The increased variability in the
measurements (ion counts) may have occurred because of the
evaporation during the sampling in the same vial (three times
for the positive mode and three times for the negative mode in
a total of 40 h, with 8 h in between). Most of the fatty acids
showed increased peak intensity over the replicates (data not
shown), which resulted in increased variation. This variation
between replicates was even higher for highly concentrated
analytes due to ion suppression phenomena.
73
The most
concentrated fatty acids, C18:2 and C18:1, which are shown in
Figure S4 (Supporting Information), were heavily affected.
This variation was observed only for the negative mode and
could be solved by using different vials for each replicate.
Figure 5. (a) Base peak ion chromatogram of the five most abundant
TGs according to peak height in pine bark extract analyzed in the
positive mode with UPC2-QTof-ESI-MS. The masses are presented as
[M + NH4]+, which was more intense than [M + Na]+. (b) Intensity
ion count of each parent peak (900.8; 898.8; 896.7; 894.7; 924.8)
fragmented by MS/MS experiment. The ions with m/zof 597.4887−
629.5482 represent the diglyceride-like fragment [M−RCOOH]+.
The colors indicate each TG.
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1329
All of the fatty acids and β-sitosterol were quantified using
octadecanoic acid and stigmasterol, respectively, as standards.
Heptadecanoic acid and cholesterol were used to evaluate the
matrix effect for fatty acids and β-sitosterol, respectively (Table
S11, Supporting Information).
■CONCLUSIONS
Two different techniques, GC-MS/FID and UPC2-ESI-QTof-
MS, and their results were presented for the analysis of
extractives obtained from pine bark. GC-MS/FID was shown
to be a reliable method for quantification of different families
of compounds, thanks to its ability to operate at a high
temperature. However, later-eluting compounds were not
qualitatively analyzable by the GC-MS method due to several
reasons. Different GC-MS methods need to be employed in
order to reduce thermal stress of both eluting compounds and
stationary phase, which results in a better identification.
74,75
Moreover, the Wiley10/NIST11 database alone is not able to
provide a complete identification of all classes of extractives
that could be found in lignocellulosic materials. This requires
in-house solutions instead. The use of UPC2coupled to a high-
resolution mass detector (HR-MS) offers a complementary
tool for the characterization of lignocellulosic extractives.
Indeed, the soft ionization technique (electrospray ionization)
enables the creation of dedicated homemade libraries of
lignocellulosic extractives as well as data mining with open-
source software (i.e., MZmine 2). Applying the equivalent
carbon number (ECN) rule together with a mass shift error
not exceeding ±5 ppm or ±0.0075 Da, 14 native SEs and 20
native TGs were detected. Among them, six SEs and five TGs
were accurately identified using MS/MS techniques. In
conclusion, supercritical carbon dioxide possesses unique
characteristics, such as easy application due to its low critical
point (31 °Cand7.4MPa),lowtoxicity,andlow
environmental impact compared to other nonpolar organic
solvents, such n-hexane and n-heptane. In the near future, it
could become an important industrial solvent for extraction of
bark with concomitant analysis and preparative isolation/
purification of interesting high-value compounds, such as
polyphenols, lignans, triterpenoids, TGs, SEs, and sterols
contained in the bark of the most exploited softwood trees.
■ASSOCIATED CONTENT
*
sıSupporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acssuschemeng.0c07914.
Details for quantification and characterization of various
compounds by GC-MS/FID and UPC2-qToF MS and
method validation for UPC2-ESI-QTof-MS (PDF)
■AUTHOR INFORMATION
Corresponding Author
Antje Potthast −Department of Chemistry, Institute of
Chemistry of Renewable Resources, University of Natural
Resources and Life Sciences (BOKU, Vienna), A-3430 Tulln,
Austria; orcid.org/0000-0003-1981-2271;
Email: antje.potthast@boku.ac.at
Authors
Stefano Barbini −Department of Chemistry, Institute of
Chemistry of Renewable Resources, University of Natural
Resources and Life Sciences (BOKU, Vienna), A-3430 Tulln,
Austria
Dev Sriranganadane −Department of Chemistry, Institute of
Chemistry of Renewable Resources, University of Natural
Resources and Life Sciences (BOKU, Vienna), A-3430 Tulln,
Austria
Sebastian Espana Orozco −Competence Center Holz GmbH,
Wood Kplus, A-4040 Linz, Austria
Armig Kabrelian −Department of Chemistry, Institute of
Chemistry of Renewable Resources, University of Natural
Resources and Life Sciences (BOKU, Vienna), A-3430 Tulln,
Austria
Katarina Karlstrom−SCA Forest Products AB, 851 21
Sundsvall, Sweden
Thomas Rosenau −Department of Chemistry, Institute of
Chemistry of Renewable Resources, University of Natural
Resources and Life Sciences (BOKU, Vienna), A-3430 Tulln,
Austria; orcid.org/0000-0002-6636-9260
Complete contact information is available at:
https://pubs.acs.org/10.1021/acssuschemeng.0c07914
Notes
The authors declare no competing financial interest.
■ACKNOWLEDGMENTS
We gratefully acknowledge Svenska Cellulosa AB (Sundsvall,
Sweden) for providing bark and Wood Kplus (Linz, Austria)
for access to milling and screening equipment. This work was
supported by the County of Lower Austria within the Austrian
Biorefinery Center Tulln (ABCT), BOKU University, and
Svenska Cellulosa AB (Sundsvall, Sweden) as participating
industry partner and the BOKU doctoral school ABCM. We
are grateful to Professor Stefan Willfor, Abo Academy Turku,
Finland, for fruitful discussions.
■REFERENCES
(1) Hillis, D. W. E. Forever Amber a Story of the Secondary Wood
Components. Wood Sci. Technol. 1986,20 (3), 203−227.
(2) Silvestre, A. J. D.; Gandini, A. Rosin: Major Sources, Properties
and Applications. In Monomers, Polymers and Composites from
Renewable Resources; Belgacem, M. N., Gandini, A., Eds.; Elsevier:
Amsterdam, 2008; Chapter 4, pp 67−88, DOI: 10.1016/B978-0-08-
045316-3.00004-1.
(3) Zinkel, D. F. Naval Stores. In Natural Products of Woody Plants:
Chemicals Extraneous to the Lignocellulosic Cell Wall; Rowe, J. W., Ed.;
Springer: Berlin, Heidelberg, 1989; pp 953−978. DOI: 10.1007/978-
3-642-74075-6_26.
(4) Roffael, E. Significance of Wood Extractives for Wood Bonding.
Appl. Microbiol. Biotechnol. 2016,100 (4), 1589−1596.
(5) Singh, T.; Singh, A. P. A Review on Natural Products as Wood
Protectant. Wood Sci. Technol. 2012,46 (5), 851−870.
(6) Fengel, D.; Wegener, G. Constituents of Bark. Wood: Chemistry,
ultrastructure, reactions; De Gruyter: Berlin, 1983; pp 240−267;
Chapter 9, DOI: 10.1515/9783110839654.240.
(7) Anas, E.; Ekman, R.; Holmbom, B. Composition of Nonpolar
Extractives in Bark of Norway Spruce and Scots Pine. J. Wood Chem.
Technol. 1983,3(2), 119−130.
(8) Norin, T.; Winell, B.; Enzell, C. R.; Nilsson, J. L. G.; Svensson, S.
Extractives from the Bark of Common Spruce, Picea Abies L. Karst.
Acta Chem. Scand. 1972,26, 2289−2296.
(9) Eurostat. Forestry Activities. In Agriculture, forestry and fishery
statistics; Cook, E., Ed.; Belgium, 2018; Chapter 6, pp 89−106.
DOI: 10.2785/340432.
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1330
(10) Arshadi, M.; Eriksson, D.; Isacsson, P.; Bergsten, U. Bark
Assortments of Scots Pine and Norway Spruce as Industrial Feedstock
for Tall Oil Production. Forests 2018,9(6), 332.
(11) Sartori, C.; da Silva Mota, G.; Ferreira, J.; Miranda, I.; Mori, F.
A.; Pereira, H. Chemical Characterization of the Bark of Eucalyptus
Urophylla Hybrids in View of Their Valorization in Biorefineries.
Holzforschung 2016,70 (9), 819−828.
(12) Zabot, G. L. Chapter 11 - Decaffeination Using Supercritical
Carbon Dioxide. In Green Sustainable Process for Chemical and
Environmental Engineering and Science; Inamuddin Asiri, A. M., Isloor,
A. M., Eds.; Elsevier, 2020; pp 255−278, DOI: 10.1016/B978-0-12-
817388-6.00011-8.
(13) Attard, T. M.; Bukhanko, N.; Eriksson, D.; Arshadi, M.; Geladi,
P.; Bergsten, U.; Budarin, V. L.; Clark, J. H.; Hunt, A. J. Supercritical
Extraction of Waxes and Lipids from Biomass: A Valuable First Step
towards an Integrated Biorefinery. J. Cleaner Prod. 2018,177, 684−
698.
(14) Uwineza, P. A.; Waskiewicz, A. Recent Advances in
Supercritical Fluid Extraction of Natural Bioactive Compounds from
Natural Plant Materials. Molecules 2020,25 (17), 3847.
(15) de Melo, M. M. R.; Silvestre, A. J. D.; Silva, C. M. Supercritical
Fluid Extraction of Vegetable Matrices: Applications, Trends and
Future Perspectives of a Convincing Green Technology. J. Supercrit.
Fluids 2014,92, 115−176.
(16) Vercoe, D.; Stack, K.; Blackman, A.; Yates, B.; Richardson, D.
An Innovative Approach Characterising the Interactions Leading to
Pitch Deposition. J. Wood Chem. Technol. 2005,24 (2), 115−137.
(17) Holmbom, B. Extractives. In Analytical Methods in Wood
Chemistry, Pulping, and Papermaking; Sjostrom, E., Alen, R., Eds.;
Springer Series in Wood Science; Springer: Heidelberg, 1999; pp
125−148, DOI: 10.1007/978-3-662-03898-7_5.
(18) Ekman, R.; Holmbom, B. Analysis by Gas Chromatography of
the Wood Extractives in Pulp and Water Samples from Mechanical
Pulping of Spruce. Nord. Pulp Pap. Res. J. 1989,4(1), 16−24.
(19) Orså, F.; Holmbom, B. A Convenient Method for the
Determination of Wood Extractives in Paper Making Process Waters
and Effluents. J. Pulp Pap. Sci. 1994,20, 361−366.
(20) Valto, P.; Knuutinen, J.; Alen, R. Overview of Analytical
Procedures for Fatty and Resin Acids in the Papermaking Process.
BioResources 2012,7(4), 6041−6076.
(21) Yang, G.; Liang, K.; Zhou, Z.; Wang, X.; Huang, G. UPLC-ESI-
MS/MS-Based Widely Targeted Metabolomics Analysis of Wood
Metabolites in Teak (Tectona Grandis). Molecules 2020,25 (9), 2189.
(22) Sinan, K. I.; Saftic, L.; PerSuric,Z
.; Pavelic, S. K.; Etienne, O.
K.; Picot-Allain, M. C. N.; Mahomoodally, M. F.; Zengin, G. A
Comparative Study of the Chemical Composition, Biological and
Multivariate Analysis of Crotalaria Retusa L. Stem Barks, Fruits, and
Flowers Obtained via Different Extraction Protocols. S. Afr. J. Bot.
2020,128, 101−108.
(23) Bouhafsoun, A.; Yilmaz, M. A.; Boukeloua, A.; Temel, H.;
Harche, M. K. Simultaneous Quantification of Phenolic Acids and
Flavonoids in Chamaerops Humilis L. Using LC-ESI-MS/MS. Cienc.
Tecnol. Aliment. 2018,38, 242−247.
(24) Prothmann, J.; Sun, M.; Spegel, P.; Sandahl, M.; Turner, C.
Ultra-High-Performance Supercritical Fluid Chromatography with
Quadrupole-Time-of-Flight Mass Spectrometry (UHPSFC/QTOF-
MS) for Analysis of Lignin-Derived Monomeric Compounds in
Processed Lignin Samples. Anal. Bioanal. Chem. 2017,409 (30),
7049−7061.
(25) Prothmann, J.; Spegel, P.; Sandahl, M.; Turner, C.
Identification of Lignin Oligomers in Kraft Lignin Using Ultra-
High-Performance Liquid Chromatography/High-Resolution Multi-
ple-Stage Tandem Mass Spectrometry (UHPLC/HRMSn). Anal.
Bioanal. Chem. 2018,410 (29), 7803−7814.
(26) Grand-Guillaume Perrenoud, A.; Guillarme, D.; Boccard, J.;
Veuthey, J.-L.; Barron, D.; Moco, S. Ultra-High Performance
Supercritical Fluid Chromatography Coupled with Quadrupole-
Time-of-Flight Mass Spectrometry as a Performing Tool for Bioactive
Analysis. J. Chromatogr. A 2016,1450, 101−111.
(27) Scheuba, J.; Wronski, V.-K.; Rollinger, J. M.; Grienke, U. Fast
and Green - CO2 Based Extraction, Isolation, and Quantification of
Phenolic Styrax Constituents. Planta Med. 2017,83 (12−13), 1068−
1075.
(28) Gibitz-Eisath, N.; Eichberger, M.; Gruber, R.; Seger, C.; Sturm,
S.; Stuppner, H. Towards Eco-Friendly Secondary Plant Metabolite
Quantitation: Ultra High Performance Supercritical Fluid Chroma-
tography Applied to Common Vervain (Verbena Officinalis L.). J. Sep.
Sci. 2020,43 (4), 829−838.
(29) Pluskal, T.; Castillo, S.; Villar-Briones, A.; OreSic, M. MZmine
2: Modular Framework for Processing, Visualizing, and Analyzing
Mass Spectrometry-Based Molecular Profile Data. BMC Bioinf. 2010,
11 (1), 395.
(30) Arshadi, M.; Hunt, A. J.; Clark, J. H. Supercritical Fluid
Extraction (SFE) as an Effective Tool in Reducing Auto-Oxidation of
Dried Pine Sawdust for Power Generation. RSC Adv. 2012,2(5),
1806−1809.
(31) Richter, B. E.; Jones, B. A.; Ezzell, J. L.; Porter, N. L.; Avdalovic,
N.; Pohl, C. Accelerated Solvent Extraction: A Technique for Sample
Preparation. Anal. Chem. 1996,68 (6), 1033−1039.
(32) Willfor, S.; Hemming, J.; Leppanen, A. S. Analysis of Extractives
in Different Pulps - Method Development, Evaluation, and
Recommendations; Analytical tools with applications for wood and
pulping chemistry, Report No. B1 of the EU COST E41 action;
Laboratory of Wood and Paper Chemistry, Åbo Akademi University:
Turku, Finland, 2004−2009.
(33) Core team R. R: A Language and Environment for Statistical
Computing; R Foundation for Statistical Computing: Vienna, Austria,
2019.
(34) Kneen, M. A.; Annegarn, H. J. Algorithm for Fitting XRF, SEM
and PIXE X-Ray Spectra Backgrounds. Nucl. Instrum. Methods Phys.
Res., Sect. B 1996,109−110, 209−213.
(35) Liland, K. H.; Almøy, T.; Mevik, B. H. Optimal Choice of
Baseline Correction for Multivariate Calibration of Spectra. Appl.
Spectrosc. 2010,64 (9), 1007−1016.
(36) Myers, O. D.; Sumner, S. J.; Li, S.; Barnes, S.; Du, X. One Step
Forward for Reducing False Positive and False Negative Compound
Identifications from Mass Spectrometry Metabolomics Data: New
Algorithms for Constructing Extracted Ion Chromatograms and
Detecting Chromatographic Peaks. Anal. Chem. 2017,89 (17), 8696−
8703.
(37) de Saint Laumer, J.-Y.; Leocata, S.; Tissot, E.; Baroux, L.;
Kampf, D. M.; Merle, P.; Boschung, A.; Seyfried, M.; Chaintreau, A.
Prediction of Response Factors for Gas Chromatography with Flame
Ionization Detection: Algorithm Improvement, Extension to Silylated
Compounds, and Application to the Quantification of Metabolites. J.
Sep. Sci. 2015,38 (18), 3209−3217.
(38) Hinshaw, J. V., Jr.; Seferovic, W. Analysis of Triglycerides by
Capillary Gas Chromatography with Programmed-Temperature
Injection. J. High Resolut. Chromatogr. 1986,9(12), 731−736.
(39) Collomb, M.; Spahni, M. S.; Buehler, T. Analysis of
Triglycerides. I. Optimization and Validation of a Rapid GC Method
and Application to Different Vegetable and Animal Fats. Trav. Chim.
Aliment. Hyg. 1998,89,59−74.
(40) Marques, G.; Rencoret, J.; Gutierrez, A.; del Río, J. C.
Lipophilic Compounds from Maize Fiber and Rice Husk Residues -
An Abundant and Inexpensive Source of Valuable Phytochemicals.
Ind. Crops Prod. 2020,146, 112203.
(41) Gominho, J.; Costa, R. A.; Lourenco, A.; Quilho, T.; Pereira, H.
Eucalyptus globulus Stumps Bark: Chemical and Anatomical
Characterization Under a Valorisation Perspective. Waste Biomass
Valorization 2020; DOI: 10.1007/s12649-020-01098-y.
(42) Freire, C. S. R.; Coelho, D. S. C.; Santos, N. M.; Silvestre, A. J.
D.; Pascoal Neto, C. Identification of Δ7 Phytosterols and Phytosteryl
Glucosides in the Wood and Bark of Several Acacia Specie-
sphytosterols and Phytosteryl Glucosides in the Wood and Bark of
Several Acacia Species. Lipids 2005,40 (3), 317−322.
(43) Gutierrez, A.; del Río, J. C.; Gonzalez-Vila, F. J.; Martín, F.
Analysis of Lipophilic Extractives from Wood and Pitch Deposits by
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1331
Solid-Phase Extraction and Gas Chromatography. J. Chromatogr. A
1998,823 (1), 449−455.
(44) Buchgraber, M.; Ulberth, F.; Emons, H.; Anklam, E.
Triacylglycerol Profiling by Using Chromatographic Techniques.
Eur. J. Lipid Sci. Technol. 2004,106 (9), 621−648.
(45) Grob, K. Degradation of Triglycerides in Gas Chromatographic
Capillaries: Studies by Reversing the Column. J. Chromatogr. A 1981,
205 (2), 289−296.
(46) Saranpaa, P.; Nyberg, H. Lipids and Sterols of Pinus Sylvestris
L. Sapwood and Heartwood. Trees 1987,1(2), 82−87.
(47) Vikstrom, F.; Holmbom, B.; Hamunen, A. Sterols and
Triterpenyl Alcohols in Common Pulpwoods and Black Liquor
Soaps. Holz Als Roh- Werkst. 2005,63 (4), 303−308.
(48) Nisula, L. Wood Extractives in Conifers: A Study of Stemwood and
Knots of Industrially Important Species. Ph.D. Dissertation, Åbo
Akademi University, Turku, Finland, 2018.
(49) Teng, J. I.; McGehee, M. F.; Smith, L. L. Sterol Metabolism
XLVI. Synthesis of Oxidized Cholesterol Fatty Acyl Esters. J. Steroid
Biochem. 1981,14 (6), 569−573.
(50) Pinter, K. G.; Hamilton, J. G.; Muldrey, J. E. A Method for
Rapid Microsynthesis of Radioactive Cholesterol Esters. J. Lipid Res.
1964,5, 273−274.
(51) Norin, T.; Winell, B.; Enzell, C. R.; Nilsson, J. L. G.; Svensson,
S. Extractives from the Bark of Scots Pine, Pinus Silvestris L. Acta
Chem. Scand. 1972,26, 2297−2304.
(52) Lee, J. W.; Nagai, T.; Gotoh, N.; Fukusaki, E.; Bamba, T.
Profiling of Regioisomeric Triacylglycerols in Edible Oils by
Supercritical Fluid Chromatography/Tandem Mass Spectrometry. J.
Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2014,966, 193−199.
(53) Jablonsky, M.; Nosalova, J.; Sladkova, A.; Haz, A.; Kreps, F.;
Valka, J.; Miertus, S.; Frecer, V.; Ondrejovic, M.; Sima, J.; Surina, I.
Valorisation of Softwood Bark through Extraction of Utilizable
Chemicals. A Review. Biotechnol. Adv. 2017,35 (6), 726−750.
(54) Smeds, A. I.; Eklund, P. C.; Willfor, S. M. Characterization of
High-Molar-Mass Fractions in a Scots Pine (Pinus Sylvestris L.)
Knotwood Ethanol Extract. Holzforschung 2018,72 (3), 201−213.
(55) Neveu, V.; Perez-Jimenez, J.; Vos, F.; Crespy, V.; du Chaffaut,
L.; Mennen, L.; Knox, C.; Eisner, R.; Cruz, J.; Wishart, D.; Scalbert, A.
Phenol-Explorer: An Online Comprehensive Database on Polyphenol
Contents in Foods. Database 2010,2010, No. bap024.
(56) Rothwell, J. A.; Urpi-Sarda, M.; Boto-Ordonez, M.; Knox, C.;
Llorach, R.; Eisner, R.; Cruz, J.; Neveu, V.; Wishart, D.; Manach, C.;
Andres-Lacueva, C.; Scalbert, A. Phenol-Explorer 2.0: A Major
Update of the Phenol-Explorer Database Integrating Data on
Polyphenol Metabolism and Pharmacokinetics in Humans and
Experimental Animals. Database 2012,2012, No. bas031.
(57) Rothwell, J. A.; Perez-Jimenez, J.; Neveu, V.; Medina-Remon,
A.; M’Hiri, N.; García-Lobato, P.; Manach, C.; Knox, C.; Eisner, R.;
Wishart, D. S.; Scalbert, A. Phenol-Explorer 3.0: A Major Update of
the Phenol-Explorer Database to Incorporate Data on the Effects of
Food Processing on Polyphenol Content. Database 2013,2013,
No. bat070.
(58) Lísa, M.; Holcapek, M. High-Throughput and Comprehensive
Lipidomic Analysis Using Ultrahigh-Performance Supercritical Fluid
Chromatography-Mass Spectrometry. Anal. Chem. 2015,87 (14),
7187−7195.
(59) Pilarova, V.; Plachka, K.; Khalikova, M. A.; Svec, F.; Novakova,
L. Recent Developments in Supercritical Fluid Chromatography -
Mass Spectrometry: Is It a Viable Option for Analysis of Complex
Samples? TrAC, Trends Anal. Chem. 2019,112, 212−225.
(60) Caboni, M. F.; Iafelice, G.; Pelillo, M.; Marconi, E. Analysis of
Fatty Acid Steryl Esters in Tetraploid and Hexaploid Wheats:
Identification and Comparison between Chromatographic Methods. J.
Agric. Food Chem. 2005,53 (19), 7465−7472.
(61) Hailat, I.; Helleur, R. Identification of Fatty Acid Steryl Esters
in Margarine and Corn Using Direct Flow Injection ESI-MSn Ion
Trap-Mass Spectrometry. Int. J. Mass Spectrom. 2014,362,24−31.
(62) Gao, B.; Luo, Y.; Lu, W.; Liu, J.; Zhang, Y.; Yu, L. L.
Triacylglycerol Compositions of Sunflower, Corn and Soybean Oils
Examined with Supercritical CO2 Ultra-Performance Convergence
Chromatography Combined with Quadrupole Time-of-Flight Mass
Spectrometry. Food Chem. 2017,218, 569−574.
(63) Zhou, Q.; Gao, B.; Zhang, X.; Xu, Y.; Shi, H.; Yu, L. L.
Chemical Profiling of Triacylglycerols and Diacylglycerols in Cow
Milk Fat by Ultra-performance Convergence Chromatography
Combined with a Quadrupole Time-of-Flight Mass Spectrometry.
Food Chem. 2014,143, 199−204.
(64) Laakso, P.; Voutilainen, P. Analysis of Triacylglycerols by
Silver-Ion High-Performance Liquid Chromatography-Atmospheric
Pressure Chemical Ionization Mass Spectrometry. Lipids 1996,31
(12), 1311−1322.
(65) Leskinen, H.; Suomela, J.-P.; Kallio, H. Quantification of
Triacylglycerol Regioisomers in Oils and Fat Using Different Mass
Spectrometric and Liquid Chromatographic Methods. Rapid
Commun. Mass Spectrom. 2007,21 (14), 2361−2373.
(66) Amaral, J. S.; Cunha, S. C.; Santos, A.; Alves, M. R.; Seabra, R.
M.; Oliveira, B. P. P. Influence of Cultivar and Environmental
Conditions on the Triacylglycerol Profile of Hazelnut (Corylus
Avellana L.). J. Agric. Food Chem. 2006,54 (2), 449−456.
(67) Bird, S. S.; Marur, V. R.; Sniatynski, M. J.; Greenberg, H. K.;
Kristal, B. S. Serum Lipidomics Profiling Using LC-MS and High-
Energy Collisional Dissociation Fragmentation: Focus on Triglyceride
Detection and Characterization. Anal. Chem. 2011,83 (17), 6648−
6657.
(68) Hsu, F.-F.; Turk, J. Electrospray Ionization Multiple-Stage
Linear Ion-Trap Mass Spectrometry for Structural Elucidation of
Triacylglycerols: Assignment of Fatty Acyl Groups on the Glycerol
Backbone and Location of Double Bonds. J. Am. Soc. Mass Spectrom.
2010,21 (4), 657−669.
(69) Ikeda, K.; Oike, Y.; Shimizu, T.; Taguchi, R. Global Analysis of
Triacylglycerols Including Oxidized Molecular Species by Reverse-
Phase High Resolution LC/ESI-QTOF MS/MS. J. Chromatogr. B:
Anal. Technol. Biomed. Life Sci. 2009,877 (25), 2639−2647.
(70) Kalo, P. J.; Kemppinen, A. Regiospecific Analysis of TAGs
Using Chromatography, MS, and Chromatography-MS. Eur. J. Lipid
Sci. Technol. 2012,114 (4), 399−411.
(71) Lísa, M.; Velínska, H.; Holcapek, M. Regioisomeric Character-
ization of Triacylglycerols Using Silver-Ion HPLC/MS and Random-
ization Synthesis of Standards. Anal. Chem. 2009,81 (10), 3903−
3910.
(72) Bowman, A. P.; Abzalimov, R. R.; Shvartsburg, A. A. Broad
Separation of Isomeric Lipids by High-Resolution Differential Ion
Mobility Spectrometry with Tandem Mass Spectrometry. J. Am. Soc.
Mass Spectrom. 2017,28 (8), 1552−1561.
(73) Jessome, L. L.; Volmer, D. A. Ion Suppression: A Major
Concern in Mass Spectrometry. LCGC N. Am. 2006,24, 498−510.
(74) Christie, W. W.; Han, X. Molecular Species of Triacylglycerols,
Diacylglycerols Derived from Complex Lipids, and Related Lipids. In
Lipid Analysis, 4th ed.; Christie, W. W., Han, X., Eds.; Woodhead,
2012; Chapter 10, pp 215−244, DOI: 10.1533/9780857097866.215.
(75) Cruz-Hernandez, C.; Destaillats, F. Analysis of Lipids by Gas
Chromatography. In Gas Chromatography; Poole, C. F., Ed.; Elsevier:
Amsterdam, 2012; Chapter 23, pp 529−544, DOI: 10.1016/B978-0-
12-385540-4.00023-7.
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article
https://dx.doi.org/10.1021/acssuschemeng.0c07914
ACS Sustainable Chem. Eng. 2021, 9, 1323−1332
1332
