Synthesis of renewable bisphenols from creosol

Article abstract of DOI:10.1002/cssc.201100402

A series of renewable bisphenols has been synthesized from creosol (2-methoxy-4-methylphenol) through stoichiometric condensation with short-chain aldehydes. Creosol can be readily produced from lignin, potentially allowing for the large scale synthesis o

Full text of DOI:10.1002/cssc.201100402

DOI: 10.1002/cssc.201100402
Synthesis of Renewable Bisphenols from Creosol
Heather A. Meylemans, Thomas J. Groshens, and Benjamin G. Harvey*[a]
Dedicated to Dr. Albert R. Harvey on the occasion of his retirement.
A series of renewable bisphenols has been synthesized from
creosol (2-methoxy-4-methylphenol) through stoichiometric
condensation with short-chain aldehydes. Creosol can be readi-
ly produced from lignin, potentially allowing for the large scale
synthesis of bisphenol A replacements from abundant waste
biomass. The renewable bisphenols were isolated in good
yields and purities without resorting to solvent-intense purifi-
cation methods. Zinc acetate was shown to be a selective cata-
lyst for the ortho-coupling of formaldehyde, but was unreac-
tive when more sterically demanding aldehydes were used.
Dilute HCl and HBr solutions were shown to be effective cata-
lysts for the selective coupling of aldehydes in the position
meta to the hydroxyl group. The acid solutions could be recy-
cled and reused multiple times without decrease in activity or
yield.
Introduction
A myriad of approaches have been developed for the efficient
conversion of biomass to custom chemicals and fuels.[1] In par-
allel with these efforts, routes to renewable and sustainable
polymer systems have been investigated and, in some cases,
commercialized.[2] Although triglycerides and cellulose have
been examined as the preferred feedstocks for many of these
polymers, well-defined polymers from lignin have been largely
unexplored, which is probably attributable to the difficulty of
isolating pure, well-defined monomers on a large scale.[3] Given
the clear benefits of highly aromatic polymer systems, includ-
ing good thermal and mechanical properties, the study of
methods for the efficient utilization of lignin has the potential
to yield industrially relevant quantities of renewable polymers
that meet the demanding requirements of conventional aro-
matic-based resins.[4] The utilization of renewable polyphenols
as precursors to epoxy resins, polycarbonates, and high-tem-
perature thermosetting resins, such as cyanate esters, provides
an opportunity to develop full-performance materials while re-
ducing the use of petroleum-based feedstocks. This approach
diminishes the overall environmental impact of resin produc-
tion, due to its use of a sustainable source of phenols. As a
tangential benefit, renewable phenols may have significantly
lower toxicities than typical precursors such as bisphenol A
(BPA).[5]
Creosol (1),a reduced form of vanillin (Scheme 1), is a com-
pelling feedstock for the synthesis of bisphenols. The large
scale isolation of vanillin from lignin is a demonstrated com-
mercial process,[6] and the straightforward hydrogenation of
vanillin has been shown to produce 1 in 99% yield.[7] The con-
densation of 1 with short-chain aldehydes has the potential to
produce bisphenols that can be converted to resins with tuna-
ble physical characteristics, which include melting point, hy-
drophobicity, and toughness. The methyl group in 1 effectively
blocks the position para to the hydroxyl group, while the me-
thoxy and phenol groups direct coupling to the 5- and 6-posi-
tions, respectively (Scheme 1). Thus, although renewable phe-
nols often have additional functional groups on the aromatic
ring, the steric hindrance and directing effects of these groups
can be utilized to control product distribution in some cases.
In contrast, aldehyde and ketone condensation reactions using
simple phenol must be run at a high phenol/(aldehyde or
ketone) ratio to reduce side products, which include o,p-bis-
phenols and novolac-type resins (Scheme 2).[8] As an example
Scheme 2. Synthesis of BPA.
[a] Dr. H. A. Meylemans, Dr. T. J. Groshens, Dr. B. G. Harvey
Research Department, Chemistry Division, US Navy
Naval Air Warfare Center Weapons Division (NAWCWD)
1900 N. Knox Rd. Stop 6303, China Lake, CA 93555 (USA)
Fax: (+1)760-939-1617
E-mail: benjamin.g.harvey@navy.mil
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201100402.
Scheme 1. Isolation of 1 from lignin.
206
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 206 – 210

Synthesis of Renewable Bisphenols from Creosol
of the low selectivity of conventional substituted phenols in al-
dehyde condensation reactions, a total of six isomers were re-
ported when a 3-methylphenol/formaldehyde ratio of 50:1 was
allowed to react.[9] In the special case of renewable materials,
atom economy is at a premium, making a stoichiometric pro-
cess considerably more favorable. In this report, we describe
the efficient, selective, and stoichiometric synthesis of bisphe-
nols from 1.
Results and Discussion
Synthesis of bisphenols
Scheme 3. Proposed mechanism for the Zn(Ac)2-mediated coupling of 1
with formaldehyde.
As an initial entry into this chemistry, we sought to condense
one equivalent of an aldehyde or ketone with two equivalents
of 1. In an effort to reduce the environmental impact of the
process, we first evaluated Nafion as a heterogeneous acid cat-
alyst for the condensation reactions. Surprisingly, Nafion was
ineffective at catalyzing the condensation reaction between 1
and formaldehyde, acetaldehyde, propionaldehyde, and ace-
tone, respectively. Regardless of the aldehyde or ketone used,
no reaction was observed even at temperatures in excess of
1008C. Next, a Lewis acid catalyst, Zn(Ac)2, was used at very
moderate loadings (0.8 mol%), and the ortho-coupled product,
6,6’-methylenebis(2-methoxy-4-methylphenol) (2),[10] was isolat-
ed in 41% yield (Table 1). ZnII has been reported to be selective
for ortho-condensation products based on a chelating effect[11]
(Scheme 3). In agreement with the report, 2 was the only ob-
served product. Although it has been suggested that other
simple aldehydes, such as acetaldehyde, can be effectively con-
densed with 1 by using this method,[10] no reaction was ob-
served with either acetaldehyde or propionaldehyde, even
after 72 h of heating to reflux and using up to ten times the
standard catalyst loading. This difference in reactivity is attrib-
uted to the electron-donating effect of the alkyl substituents,
which could potentially be overcome with a stronger Lewis
acid.
Turning to Brønsted acid catalysts, we explored the use of
biphasic systems consisting of a mineral acid, an aldehyde, and
a stoichiometric amount of 1 (Scheme 4).[12] HCl (2.5m) was
Scheme 4. Brønsted acid-catalyzed coupling of 1 with aldehydes.
determined to be the most effective catalyst for the conden-
sation of 1 with formaldehyde.[13] In contrast to the zinc-
catalyzed reaction, formaldehyde coupled almost exclusively at
the position meta to the hydroxyl group with a selectivity
of 97%.[14] As opposed to the acid-catalyzed condensation of
simple phenol with formaldehyde, the steric constraints of
1 along with the para- and ortho-directing effects of the
methoxy and methyl groups, respectively, resulted in the prod-
uct specificity. In a similar manner, the stoichiometric conden-
sation product 4 was prepared in 68% yield from 1 and acet-
aldehyde.
Table 1. Synthesis of bisphenols from 1.
Product
Catalyst
Zn(Ac)2
Yield [%]
41[a]
2.5m HCl
2.5m HCl
63[b]
Propionaldehyde proved to be a considerably more de-
manding substrate than acetaldehyde[15] and could not be
condensed even by using concentrated HCl. H2SO4 (5m) also
produced no reaction, whereas 9m and concentrated H2SO4
produced black solutions with multiple products. Although the
dark color suggested the presence of quinones, concentrated
H2SO4 has been reported to convert aryl methyl ethers to phe-
nols, which may further complicate the product distribution.[16]
Interestingly, HBr (6m) cleanly produced bisphenol 5 in 67%
yield at room temperature. This is probably the result of the
enhanced reactivity of the intermediate carbocation due to the
greater polarizability of bromide compared to chloride.
68[c]
6m HBr
67[d]
[a] 0.8 mol% catalyst, 16 h, reflux. [b] 3h, reflux. [c] 4h, reflux, phenol/al-
dehyde ratio of 1.5. [d] 16 h, ambient.
ChemSusChem 2012, 5, 206 – 210
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
207

H. A. Meylemans, T. J. Groshens, B. G. Harvey et al.
Table 2. Selectivity of aldehyde coupling reactions using 1.
Aldehyde
meta,meta Isomer [%]
meta,ortho Isomer [%]
CH2O
CH3CHO
CH3CH2CHO
97
96
100
3
4
–
The meta,meta isomers were almost exclusively isolated, re-
gardless of the aldehyde (Table 2). Significant differences in the
selectivity were not observed between formaldehyde and acet-
aldehyde, but when propionaldehyde was condensed, no trace
of the meta,ortho product was found. This suggests that, in ad-
dition to electronic and steric effects inherent to the phenol,
the alkyl group on the aldehyde plays a role in the selectivity
of the reaction.
Figure 1. Solid-state structure of 2.
The biphasic approach to bisphenol synthesis has a number
of advantages over conventional techniques. The initial reac-
tion requires no organic solvent and the crude product can be
isolated by using a simple decantation followed by a water
wash. Although subsequent steps utilize solvents for purifica-
tion, the phenols can also be purified by sublimation, which re-
sults in a solvent-free process. In addition, the procedure is
performed at moderate temperatures and does not require
protection from atmospheric conditions. Perhaps one of the
most attractive aspects of this method is the potential to reuse
the acid solutions. To examine the feasibility of recycling the
acid solutions, 4 was prepared according to the standard con-
ditions with HCl (2.5m). The acid solution was then reused in
two subsequent reactions without any purification. The yields
actually increased slightly in later reactions, perhaps due to
residual reagents that were solubilized in the aqueous phase.
Although the use of mineral acids may be considered detri-
mental for a sustainable process, the ability to reuse the
solutions greatly reduces the environmental impact of this
approach.
Figure 2. Solid-state structure of 4.
water even after extended drying in a vacuum oven. This
excess water was confirmed by elemental analysis, and leads
to the conclusion that the oil is a water adduct of 5 with a
stoichiometry of 5·1.5H2O. In the presence of dimethylsulfox-
ide (DMSO), 5 formed a solvent adduct that readily crystallized,
allowing for a structural comparison. The adduct is stabilized
by a strong hydrogen bonding interaction between the
phenol and DMSO.
One of the goals of this research was the development of a
method that allowed for the selective modification of both the
hydrophobicitiesand melting points of the phenols. Daughter
resins (e.g., cyanate esters) are expected to inherit many of the
properties of the parent phenols, and thus, this work will allow
the resins to be custom-tailored based on the choice of alde-
hyde and the condensation catalyst.
Structures and properties of bisphenols
To confirm the structures and evaluate the suitability of various
phenols as precursors to thermoplastics and thermosetting
resins, the X-ray structures of the bisphenols were deter-
mined.[17] Compound 2, in addition to exhibiting intermolecu-
lar hydrogen bonding, has two intramolecular hydrogen bonds
of 1.83(2) and 2.24(2) ꢁ between the two phenol units and be-
tween a phenol and a methoxy oxygen, respectively. This inter-
action is possible due to the close proximity of the phenols re-
sulting from the ortho coupling and gives rise to a symmetrical
molecule in the solid state (Figure 1). The bisphenols substitut-
ed at the meta position also participate in intramolecular hy-
drogen bonding, but the aromatic substitution precludes the
strong inter-ring hydrogen bonding observed in 2 and results
in rotated aromatic groups; as an example, the structure of 4
is shown in Figure 2.[18] In contrast to the other phenols, pure
5 was isolated as an oil at room temperature. Although
13C NMR spectroscopy and GC–MS confirmed the purity of this
oil, 1H NMR spectroscopy revealed the presence of residual
Some insight into the effect of aldehyde choice can be
gained by comparing the melting points of the bisphenols.
Compound 2 has the lowest melting point (123–1258C), fol-
lowed by 3 (131–1348C), and 4 (143–1468C). The increase in
melting point from 3 to 4 is surprising compared to the melt-
ing points of the related bisphenol F and bisphenol E, which
are 162–164 and 123–1278C, respectively (Figure 3). The 1,1-di-
Figure 3. Structures of conventional bisphenols.
208
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 206 – 210

Synthesis of Renewable Bisphenols from Creosol
105. elemental analysis calcd. (%): C 70.81, H 6.99; found: C 70.84,
H 7.13.
phenylethane framework typically results in depressed melting
points compared to methylene or 2,2-diphenylpropane linkag-
es.[19] Despite this discrepancy, the melting point exhibited a
drastic decrease upon extension of the reaction from acetalde-
hyde to propionaldehyde, and 5 was isolated as a viscous oil.
For applications as precursors to thermosetting composites,
lower melting resins should allow simplified and lower-cost
fabrication methods.
5,5’-Methylenebis(2-methoxy-4-methylphenol) (3)
Compound 1 (5.03 g, 36.4 mmol) and 37% formaldehyde (1.47 g,
18.1 mmol) were diluted with deionized H2O (40 mL). Concentrated
HCl (10 mL) was slowly added, and the reaction was heated to
reflux under N2 for 3 h. A precipitate formed, the solution was dec-
anted, and the solid was washed with 10% ethanol solution.
The solid was dissolved in ether and precipitated with heptane.
Compound 3 was isolated as a crystalline white solid (3.29 g, 63%).
Crystals suitable for an X-ray diffraction study were obtained from
the slow evaporation of an ether solution at room temperature.
Conclusions
A series of bisphenols has been synthesized from creosol, a re-
newable phenol that can be produced from abundant waste
lignin. The functional groups on the ring and choice of catalyst
allow for control of the condensation productsand these reac-
tions can be conducted with stoichiometric amounts of the
phenol. The ability to recycle the acid catalysts, perform the re-
action in the absence of organic solvents, and isolate the prod-
ucts without resorting to solvent-intense purification methods
make these phenols intriguing renewable and sustainable
precursors to a variety of polymeric materials. Further work on
the conversion of these bisphenols to resins and polymers is
ongoing.
1
m.p. 131–1348C; H NMR ([D6]DMSO): d=2.08 (s, 6H), 3.56 (s, 2H),
3.71 (s, 6H), 6.30 (s, 2H), 6.72 (s, 2H), 8.54 ppm (s, 2H); 13C NMR
([D6]DMSO): d=19.0, 35.0, 56.2, 115.0, 117.0, 126.3, 131.3, 144.8,
146.6 ppm; MS: m/z=288, 273, 257, 241, 227, 213, 195, 181, 165,
150; elemental analysis calcd. (%): C 70.81, H 6.99; found: C 70.66,
H 7.16.
5,5’-(Ethane-1,1-diyl)bis(2-methoxy-4-methylphenol) (4)
Compound
1 (5.1 g, 37.0 mmol) and acetaldehyde (1.06 g,
24.1 mmol) were diluted with deionized H2O (40 mL). Concentrated
HCl (10 mL) was slowly added, and the mixture was heated to
reflux under N2 for 4 h. The supernatant was carefully decanted
from the resulting dense oil. Compound 4 was obtained as a white
solid after extraction into ether and precipitation with heptane
(3.74 g, 68%). Crystals suitable for an X-ray diffraction study were
obtained from the slow evaporation of an ether solution at room
Experimental Section
General
2-Methoxy-4-methylphenol (1), acetaldehyde, propionaldehyde,
formaldehyde (37%), Zn(Ac)2·2H2O, and concentrated HBr (48%)
were purchased from Aldrich and used as received. Concentrated
HCl and H2SO4 were purchased from Fisher Scientific and used as
received. NMR spectra were collected by using a Bruker Avance II
300 MHz NMR spectrometer. Samples of bisphenols were prepared
in [D6]DMSO, and spectra were referenced to the solvent peaks
(d=2.50 and 39.5 ppm for 1H and 13C NMR spectra, respectively).
The products were further analyzed by using an Agilent 6890-GC
system with a Restek RTX-5MS 30-meter column. The GC inlet tem-
perature was 2508C, and the column oven temperature program
began at 408C for three minutes and increased to 3508C at
108CminÀ1. An Agilent mass selective detector (MSD) 5973 system
was used to identify various products. Elemental analysis was per-
formed by Atlantic Microlabs Inc. Norcross, GA.
1
temperature. m.p. 143–1468C; H NMR ([D6]DMSO): d=1.30 (d, J=
7 Hz, 3H), 2.06 (s, 6H), 3.69 (s, 6H), 4.05 (t, J=7 Hz, 1H), 6.48 (s,
2H), 6.66 (s, 2H), 8.57 ppm (s, 2H); 13C NMR ([D6]DMSO): d=18.5,
21.3, 36.1, 56.0, 114.6, 115.0, 125.7, 137.0, 144.6, 145.5 ppm; MS: m/
z=303, 287, 269, 240, 211, 195, 164, 145, 128, 105; elemental anal-
ysis calcd.(%): C 71.50, H 7.33; found: C 71.58, H 7.46.
5,5’-(Propane-1,1-diyl)bis(2-methoxy-4-methylphenol) (5)
Compound 1 (5.02 g, 36.4 mmol) and propionaldehyde (1.04 g,
17.9 mmol) were diluted with deionized H2O (10 mL). 48% aqueous
HBr (20 mL) was slowly added, and the reaction was stirred
at room temperature overnight. The supernatant was carefully de-
canted from the resultant oil, and the product was washed with
water. Work up by using the method described above yielded 5 as
a viscous tan oil (3.82 g, 67%). The product formed a solvent
adduct with DMSO that crystallized from ether solutions upon
6,6’-Methylenebis(2-methoxy-4-methylphenol) (2)
1
standing at room temperature. H NMR ([D6]DMSO): d=0.85 (t, J=
Compound
1 (5 g, 36.2 mmol), 37% formaldehyde (1.56 g,
7 Hz, 3H), 1.73 (t, J=7 Hz, 2H), 2.10 (s, 6H), 3.69 (s, 6H), 3.81 (t, J=
7 Hz, 1H), 6.52 (s, 2H), 6.65 (s, 2H), 8.55 ppm (s, 2H); 13C NMR
([D6]DMSO): d=13.1, 18.9, 28.9, 43.3, 56.1, 114.9, 115.0, 126.3,
135.5, 144.5, 145.6 ppm; MS: m/z=316, 287, 257, 240, 211, 195,
167, 151, 131, 115; elemental analysis calcd. for 5·1.5H2O: C 66.45,
H 7.92; found: C 66.33, H 7.78.
19.2 mmol), and Zn(ac)2·2H2O (70 mg, 3.2ꢂ10À4 mol) were heated
to reflux overnight under N2. The resulting oil was washed with
10% ethanol, and extracted into ether. The ether was removed on
a rotary evaporator and the resulting oil was heated to 1008C
overnight under reduced pressure (133.3 Pa). The resulting solid
was dissolved in ether and precipitated with heptane. The light-tan
solid was collected by filtration, washed with excess heptane, and
dried to yield 2.13 g (41%) of 2. m.p. 123–1258C; 1H NMR
([D6]DMSO): d=2.11 (s, 6H), 3.71 (s, 2H), 3.74 (s, 6H), 6.35 (d, J=
2 Hz, 2H), 6.58 (d, J=2 Hz, 2H), 8.20 ppm (s, 2H); 13C NMR
([D6]DMSO): d=21.1, 29.0, 56.2, 110.7, 122.7, 127.4, 127.7, 141.9,
147.5 ppm; MS: m/z=288, 271, 255, 239, 212, 195, 165, 138, 121,
Acknowledgements
The authors would like to thank Dr. Michael Wright and Dr. Matt
Davis for helpful discussions, Ms. Roxanne Quintana for GC–MS
ChemSusChem 2012, 5, 206 – 210
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
209

H. A. Meylemans, T. J. Groshens, B. G. Harvey et al.
[11] D. A. Fraser, R. W. Hall, P. A. Jenkins, A. L. J. Raum, J. Appl. Chem. 1957, 7,
689–700.
analysis, and The Strategic Environmental Research and Develop-
ment Program for funding this work (Project WP-1759).
[12] For similar procedures, see for example: a) A. G. Sverker Hoegberg, J.
Org. Chem. 1980, 45, 4498–4500; b) Y. Zhou, C. Jiang, Y. Zhang, Z.
Liang, W. Liu, L. Wang, C. Luo, T. Zhong, Y. Sun, L. Zhao, X. Xie, H. Jiang,
N. Zhou, D. Liu, H. Liu, J. Med. Chem. 2010, 53, 5449–5466.
[13] A. Arcoleo, T. Garofana, Annali di Chimica 1957, 47, 1142–1162.
[14] The selectivity of the m,m isomer is based on GC–MS data—see the
Supporting Information.
[15] Compound 5 has been reported as a condensation product by using
dilute H2SO4 as a catalyst: S. Yasuda, K. Ota, Mokuzai Gakkaishi 1986,
32, 51–58. In our laboratory, dilute H2SO4 did not promote the conden-
sation of 1 and propionaldehyde.
Keywords: aldehydes · biomass · bisphenols · catalysis ·
sustainable chemistry
[1] a) M. J. Climent, A. Corma, S. Iborra, Green Chem. 2011, 13, 520–540;
b) R. A. Sheldon, Catal. Today 2011, 167, 3–13; c) A. J. Ragauskas,
C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney, C. A. Eckert, W. J.
Frederick, J. P. Hallett, D. J. Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R.
Templer, T. Tschaplinski, Science 2006, 311, 484–489.
[2] Renewable and Sustainable Polymers: ACS Symposium Series 1063 (Eds.:
G. F. Payne, P. B. Smith), American Chemical Society, Washington DC,
2011, pp. 1–212.
[3] A. Gandini, Macromolecules 2008, 41, 9491–9504.
[4] a) J. J. Lindberg, T. A. Kuusela, K. Levon, ACS Symp. Ser. 1989, 397, 190–
204; b) H. Hatakeyama, S. Hirose, T. Hatakeyama, ACS Symp. Ser. 1989,
397, 205–218.
[16] R. L. Burwell, Chem. Rev. 1954, 54, 615–683.
[17] CCDC 831799 (2), 831800 (3), 831801 (4), and 831802 (5) contain the
supplementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre through www.ccdc.cam.ac.uk/data request/cif. A cif file that
contains the structural information for each compound is available in
the Supporting Information.
[5] H. Nouailhas, C. Aouf, C. Le Guerneve, S. Caillol, B. Boutevin, H. Fulcrand,
J. Polym. Sci. Part A 2011, 49, 2261–2270.
[18] The angles between the rings are 78.18, 81.49, and 72.348 for 3, 4, and
5, respectively.
[6] M. B. Hocking, J. Chem. Edu. 1997, 74, 1055–1059.
[7] Q. Wang, Y. Yang, Y. Li, W. Yu, Z. J. Hou, Tetrahedron 2006, 62, 6107–
6112.
[19] L. R. Cambrea, M. C. Davis, T. J. Groshens, A. J. Guenthner, K. R. Lamison,
J. M. Mabry, J. Polym. Sci. Part A 2010, 48, 4547–4554.
[8] G. D. Yadav, N. Kirthivasan, Appl. Catal. A: Gen. 1997, 154, 29–53.
[9] L. E. Bogan, S. K. Wok, Macromolecules 1992, 25, 161–165.
[10] H. C. Bailey, GB 845608, 1960.
Received: July 27, 2011
Published online on December 12, 2011
210
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 206 – 210