Synthesis, characterisation, and performance evaluation of tri-armed phenolic antioxidants

Article abstract of DOI:10.1016/j.tetlet.2020.152127

In this study, a series of core units (glycerol, triethanolamine and triisopropanolamine derivatives) were investigated for their use in tri-armed phenolic antioxidants. The antioxidant ability of these tri-armed phenolic compounds featuring different core units were then evaluated in a hydrocarbon lubricant using differential scanning calorimetry (DSC) and compared to the commercially available antioxidants Irganox L135 and Irganox L57. An impressive oxidation induction time of ca. 9–12 min was observed for the glycerol based antioxidants when compared to the commercial antioxidants (ca. 4–6 min), whereas in contrast in the case of triethanolamine and triisopropanolamine derived antioxidants, a solubilising unit was incorporated in order to provide appropriate solubility within the hydrocarbon medium and revealed an excellent oxidation induction time of ca. 11–12 min.

Full text of DOI:10.1016/j.tetlet.2020.152127

Tetrahedron Letters 61 (2020) 152127
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis, characterisation, and performance evaluation of tri-armed
phenolic antioxidants
Clare L. Higgins a, Sorin V. Filip b, Ashfaq Afsar a, Wayne Hayes a,
⇑
a Department of Chemistry, University of Reading, Whiteknights, Reading RG6 6AD, UK
b BP Formulated Products Technology, Research & Innovation, Pangbourne, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, a series of core units (glycerol, triethanolamine and triisopropanolamine derivatives) were
investigated for their use in tri-armed phenolic antioxidants. The antioxidant ability of these tri-armed
phenolic compounds featuring different core units were then evaluated in a hydrocarbon lubricant using
differential scanning calorimetry (DSC) and compared to the commercially available antioxidants Irganox
L135 and Irganox L57. An impressive oxidation induction time of ca. 9–12 min was observed for the glyc-
erol based antioxidants when compared to the commercial antioxidants (ca. 4–6 min), whereas in con-
trast in the case of triethanolamine and triisopropanolamine derived antioxidants, a solubilising unit
was incorporated in order to provide appropriate solubility within the hydrocarbon medium and revealed
an excellent oxidation induction time of ca. 11–12 min.
Received 20 April 2020
Revised 2 June 2020
Accepted 5 June 2020
Available online 9 June 2020
Keywords:
Glycerol
Triethanolamine
Triisopropanolamine
Tri-armed phenolic antioxidants
Oxidative stability studies
Ó 2020 Elsevier Ltd. All rights reserved.
Introduction
biodiesel production is still a major question that needs to be
addressed [10–12]. Glycerol has also been used for a number of
The requirement for enhanced fuel efficiency and lower tailpipe
CO2 emissions have led to the introduction of alternative fuels from
renewable resources for the partial substitution of conventional
petroleum-derived fuels [1]. Biodiesel, which consist of saturated
and unsaturated long-chain fatty acid alkyl esters along with glyc-
erol fractions are both products of trans-esterification reactions
that can be manufactured from various oil feedstocks; for example,
oils from rapeseed, sunflowers, soybean, canola and palm, waste
cooking oils and sewage sludge [2–5]. Biodiesel has great potential
as an alternative fuel for compression ignition engines because its
chemical composition is comparable to diesel fuel but with an easy
to handle higher boiling point and lower carbonaceous emissions
[4,6–8]. In the light of the increased demand for biodiesel, espe-
cially in Europe and the Americas, there is a potential for overpro-
duction of the glycerol by-product. Biodiesel production generates
about 10% (w/w) of crude glycerol and, with the world biodiesel
market continuously increasing, it has been predicted that by
2020 more than 4 billion litres of glycerol will be produced per
annum [9]. This surplus has prompted the chemical industry to
seek application of glycerol as a low-cost feedstock in areas such
as fuels, chemicals, the automotive industry, pharmaceuticals and
detergents. However, finding new uses for the excess glycerol from
years as a monomer in the synthesis of dendritic macromolecules
targeted for deployment in biological applications such as drug
delivery systems and tissue engineering [13–15]. Only a few exam-
ples have been reported where glycerol is transformed into radical-
trapping fuel additives [16]. Such an application is particularly
attractive as it enables the glycerol by-product generated by plants
grown for biodiesel production to be used to improve the resis-
tance of this fuel stock to auto-oxidation. The oxidation process
of biodiesel can be particularly detrimental because it leads to
the formation of hydroperoxides which produces insoluble gums
and sediments that can plug fuel filters or make deposits on the
fuel injector leading to ineffective engine operation [17]. In addi-
tion, the final products of oxidation can also increase the viscosity
of the fuel that in turn leads to poor fuel atomization [17]. Another
notable class of dendritic macromolecules are poly(amidoamines)
(PAMAM) dendrimers and were first reported by Tomalia and co-
workers in the 1980s [18]. PAMAM dendrimers are particularly
unique because they can be designed to mimic the structural archi-
tecture of globular proteins and hence a range of biomedical appli-
cations have been proposed for their use such as drug delivery,
molecular encapsulation and gene therapy [19,20]. PAMAM den-
drimers typically consist of an ethylenediamine core, however,
the more flexible triethanolamine core has been recently reported
[21,22]. Ottaviani and co-workers [23] reported a series of PAMAM
dendrimers with a triethanolamine core which revealed interest-
⇑
Corresponding author.
E-mail address: w.c.hayes@reading.ac.uk (W. Hayes).
https://doi.org/10.1016/j.tetlet.2020.152127
0040-4039/Ó 2020 Elsevier Ltd. All rights reserved.

2
C.L. Higgins et al. / Tetrahedron Letters 61 (2020) 152127
ing copper (II) binding characteristics. The transition metal binding
ability of the triethanolamine core has been known for a number of
years [24–26] and could be particularly relevant in the design of
oxidation inhibitors as traces of transition metal ions play a signif-
icant role in the catalysis of the oxidation processes [27]. It is fea-
sible that the oxidative stability of a material could be enhanced by
introducing an antioxidant with both radical scavenging and metal
chelation properties. In this paper, we report the synthesis of a ser-
ies of tri-armed sterically hindered phenolic-based antioxidants
whereby glycerol and triethanolamine derivatives were utilised
as the core unit. Triisopropanolamine was also used as a central
core monomer as it was proposed that the methyl moieties of
the core would provide additional solubility to the antioxidant.
Solubilising alkyl chains were introduced to the core unit to
improve dispersion within a hydrocarbon medium. Promisingly,
in comparison to the industrial antioxidants Irganox L135 and Irga-
nox L57, these new tri-armed materials exhibited enhanced stabil-
isation properties when blended into a lubricant base oil and
subjected to accelerated oxidative conditions.
additional stability to both the ester functionalities and the nitro-
gen core in addition to improving the solubility of the final racemic
triphenol 12. The desired tri-armed derivative was achieved suc-
cessfully via the reaction of triisopropanolamine 10 with the 2
using a DCC mediated coupling reaction (Fig. 1, Scheme S2).
Although the triethanolamine and triisopropanolamine deriva-
tives 11 and 12 did exhibit excellent antioxidant capabilities, dis-
appointingly their physical properties such as solubility in
hydrocarbon media was not yet optimised. Hence, in an attempt
to overcome the solubility issues encountered with 11 and 12, an
alternative synthetic approach was targeted to allow incorporation
of a solubilising alkyl chain. The synthesis was achieved by react-
ing diethanolamine 13 with 2-ethylhexyl bromide 14 to afford
the diol 15. A DCC mediated esterification was then utilised to cou-
ple 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionic acid 2 to the
core to yield the racemic diphenol 16 as a yellow oil in 73% yield
(Fig. 2, Scheme S3). For comparison against 12, the synthesis of
intermediate diol 18 was achieved by first reacting 1,10-azanediyl-
bis(propan-2-ol) 17 with 14 followed by functionalisation of 18
with a solubilising alkyl chain 2-ethylhexyl bromide 14 and 3-
(3,5-di-tert-butyl-4-hydroxyphenyl)propionic acid 2 to afford the
racemic diphenol 19 as a yellow oil in 62% yield (Fig. 2, Scheme S4).
Results and discussion
Synthesis
Oxidative stability studies
Initial synthesis began by exhaustively coupling all three hydro-
xyl moieties of glycerol 1 with 3-(3,5-di-tert-butyl-4-hydrox-
yphenyl)propionic acid 2, using N,N0-dicyclohexylcarbodiimide
(DCC) mediated esterification [28,29] to generate the first genera-
tion triphenol 3 as a yellow oil in 70% yield (Scheme 1) (see ESI for
the spectroscopic data of the compounds described herein). The
difference in reactivity of the primary and secondary hydroxyl
moieties of 1 was then exploited to introduce 2-ethylhexanoic acid
4 as a solubilising alkyl chain with the aim to improve dissolution
within a hydrocarbon medium. To generate the racemic diphenol
5, the primary alcohol moieties of 1 were first reacted with two
equivalents of 2 to generate diester 6 with one free hydroxyl moi-
ety, followed by another DCC mediated esterification of the hydro-
xyl moiety of 6 with acid 4 (Scheme 2).
The final reaction (Scheme 3) of the glycerol series involved
incorporation of two alkyl moieties onto the glycerol monomer 1.
In this case, the primary hydroxyl moieties were first reacted with
the acid 4 to yield the diester 7, followed by reaction of the remain-
ing secondary hydroxyl moiety with 2 to yield the racemic
monophenol 8 as a colourless oil in 75% yield.
To assess the antioxidant potential, mono-, di- and tri-phenol
glycerol derivatives 3, 5 and 8 were blended into a synthetic lubri-
cant base oil - DurasynÒ 164 (a polyalphaolefin, hydrogenated
hydrocarbon base oil composed of dec-1-ene trimers typically used
in lubricating oils). Typical lubricant commercial antioxidants Irga-
nox L135 (phenolic antioxidant) and Irganox L57 (aromatic amine
antioxidant) were used as a direct comparison and samples were
prepared by blending of 0.5% w/w of each antioxidant in 50 mL
of the lubricant base oil. The blends were analysed using pres-
surised differential scanning calorimetry (PDSC) to monitor the
heat effects associated with phase transitions and chemical reac-
tions as a function of temperature. Oxidation induction time
(OIT) and oxidation onset temperature (OOT) were used to investi-
gate the effect of antioxidants on the stability of an oil sample. OIT
revealed that the presence of glycerol derivatives 3, 5 and 8 in the
base oil had resulted in a significant increase in the stability of the
sample as shown in Fig. 3.
The induction time was increased from <3 min for the
unblended base oil to ca. 10–12 min for the blended samples. In
addition, the glycerol derivatives 3, 5 and 8 exhibited superior per-
formance to both commercial antioxidants, Irganox L135 and Irga-
nox L57. The OOT results for each oil blend are presented in Fig. 4
where again, a significant increase in temperature was observed
when 3, 5 and 8 were incorporated into the blend when compared
to the base oil in isolation (ca. 245–249 °C).
The second series featuring different core units saw the use of
triethanolamine 9 and triisopropanolamine 10 which provided a
nitrogen at the core of the macromolecule. Triethanolamine 9
was first reacted with 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propi-
onic acid 2 in a DCC mediated esterification to afford the triphenol
11 as a waxy solid in 65% yield (Fig. 1, Scheme S1). Triiso-
propanolamine 10 was also used as the central core unit as it
was proposed that the methyl moieties of the core would provide
The structure-activity relationships were also investigated by
comparing the triphenol 3 and the diphenol 5. It was expected that
OH
OH
DCC, DPTS,
CH2Cl2,
OH
R.T, 24 hours
HO
OH
HO
OH
70%
O
O
O
3
O
O
OH
O
O
1
2
Scheme 1. Synthesis of the triphenol 3 from the reaction between glycerol 1 and 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionic acid 2.

C.L. Higgins et al. / Tetrahedron Letters 61 (2020) 152127
3
OH
DCC, DPTS,
HO
OH
CH2Cl2,
OH
1
OH
R.T, 24 hours
HO
OH
O
O
45%
O
O
2
6
O
OH
2-ethylhexanoic acid 4,
DCC, DPTS, CH2Cl2,
R.T, 24 hours
76%
HO
OH
O
5
O
O
O
O
O
Scheme 2. Synthesis of the diester 6, followed by reaction of the remaining secondary hydroxyl to yield the diphenol 5.
DCC, DPTS,
O
CH2Cl2,
OH
OH
1
R.T, 24 hours
O
O
HO
OH
HO
O
O
30%
4
7
OH
OH
DCC, DPTS,
CH2Cl2,
R.T, 24 hours
O
OH
75%
O
O
O
2
O
O
O
8
Scheme 3. Synthesis of the diester 7 followed by reaction of the remaining secondary hydroxyl moiety to yield the monophenol 8.
OH
HO
OH
O
O
O
O
O
O
OH
O
O
N
N
O
O
HO
O
HO
O
11
12
Fig. 1. Structures of triphenol 11 and triphenol 12 synthesised using the methodology described in Schemes S1 and S2, respectively.
the triphenol 3 would provide the greatest oxidative stability to
the lubricant base oil in comparison to the diphenol 5 as a result
of the extra active phenolic functionality. This, however, was not
observed and instead the diphenol 5 provided excellent stability
with an oxidation induction time of ca. 12 min in comparison to
ca. 10 min for the triphenol 3. This induction time was also much
greater than both of the commercially available antioxidants Irga-
nox L135 and L57 and it was proposed that the diphenol 5 had the
most effective balance between active functionalities and solubil-
ity. Oxidative stability analysis of triethanolamine and triiso-
propanolamine derivatives 11 and 12 proved more challenging as
the solubility in the lubricant base oil remained an issue and a
longer heating time (ca. 2 h compared to 30 min) and higher tem-
perature (ca. 70 °C compared to 50 °C) was required in order to
produce the blend. However, unexpectedly excellent OIT of
ca. 11–14 min and OOT of ca. 245–251 °C were observed for 11
and 12 revealing enhanced oxidative stability in comparison to
Irganox L135 and Irganox L57 (Fig. 5).
The above results revealed that these compounds did have very
promising antioxidant capabilities but disappointingly solubility

4
C.L. Higgins et al. / Tetrahedron Letters 61 (2020) 152127
OH
OH
HO
O
HO
O
O
O
O
O
N
N
O
O
19
Fig. 2. Structures of diisopropanolamine 16 and diphenol 19 synthesised using the methodology described in Schemes S3 and S4, respectively.
16
Fig. 3. Average oxidation induction time analysis of the glycerol series 3, 5 and 8 (tested in duplicate).
Fig. 4. Average oxidation onset temperature analysis of the glycerol series 3, 5 and 8 (tested in duplicate).
was an issue. To further refine the structure of triethanolamine and
triisopropanolamine derivatives 11 and 12, an additional alkyl
chain was incorporated for enhanced solubility. Oxidative stability
analysis of 16 and 19 were then carried out that revealed an oxida-
tion induction time of ca. 11–12 min and oxidation onset temper-
ature of ca. 245–249 °C (Fig. 6), performing in the same region as
triethanolamine (11), triisopropanolamine (12) and glycerol (3, 5
and 8) derivatives.
These results highlight the importance of balancing functional-
ity with solubility where replacement of one active phenolic moi-
ety for a single solubilising alkyl chain had little to no effect on the
OIT and OOT parameters. In addition, whilst the core monomers
themselves do not necessarily contribute to the antioxidant capa-
bilities of these compounds, they do afford higher molecular
weight additives which possess a larger number of antioxidant
units that are in turn are less volatile (cf. commercially available

C.L. Higgins et al. / Tetrahedron Letters 61 (2020) 152127
5
Fig. 5. Average oxidation induction time analysis of triphenols 11 and 12 (tested in duplicate).
Fig. 6. Average oxidation induction time analysis of diphenols 16 and 19 (tested in duplicate).
antioxidants) and thus residence times within the hydrocarbon
lubricant matrix are enhanced.
of ca. 11–12 min. These results confirmed that the central core unit
does not necessarily contribute to the antioxidant capabilities and
the solubility of an additive is just as important as antioxidant
functionality when considering the design of new tri-armed
antioxidants.
Conclusion
In summary, a series of alternative core units (glycerol, tri-
ethanolamine and triisopropanolamine derivatives) were investi-
gated for their use in the synthesis of tri-armed phenolic
antioxidants. Oxidation induction time (OIT) revealed that the
presence of glycerol based derivatives 3, 5 and 8 in the base oil
resulted in a significant increase in the stability of the sample from
<3 min for the unblended base oil to ca. 9–12 min for the blended
samples. In addition, 3, 5 and 8 showed superior performance to
both commercial antioxidants, Irganox L135 and Irganox L57. A
series of nitrogen core units were also investigated and the tri-
ethanolamine and triisopropanolamine derivatives 11 and 12
revealed excellent oxidation induction times of ca. 11–14 min,
however, solubility in the lubricious base oil proved to be an issue.
As an alternative, 16 and 19 were synthesised that incorporated a
solubilising alkyl chain and revealed oxidation times in the region
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgements
The authors acknowledge the UK Biotechnology and Biological
Sciences Research Council (BBSRC) and BP p.l.c. for financial sup-
port. Use of the Chemical Analysis Facility (CAF) at the University
of Reading and Analytical department at the BP Technology Centre,
Pangbourne are gratefully acknowledged.

6
C.L. Higgins et al. / Tetrahedron Letters 61 (2020) 152127
[13] R. Haag, F. Vogtle, Angew. Chem. Int. Ed. Engl. 43 (2004) 272–273.
Appendix A. Supplementary data
[14] H. Frey, R. Haag, Rev. Mol. Biotechnol. 90 (2002) 257–267.
[15] M.M.K. Boysen, K. Elsner, O. Sperling, T.K. Lindhorst, Eur. J. Org. Chem. 2003
(2003) 4376–4386.
[16] Z. Namazifar, F. Saadati, A.A. Miranbeigi, New J. Chem. 43 (2019) 10038–
10044.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2020.152127.
[17] K. Varatharajan, D.S. Pushparani, Renew. Sustain. Energy Rev. 82 (2018) 2017–
2028.
References
[18] D.A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin, J. Roeck, J. Ryder,
P. Smith, Polym. J. 17 (1985) 117–132.
[19] D.A. Tomalia, Mater. Today 8 (2005) 34–46.
[1] G. Karavalakis, S. Stournas, Energy Fuels 24 (2010) 3682–3686.
[2] M. Jerzykiewicz, I. Cwielag, W. Jerzykiewicz, J. Chem. Technol. Biotechnol. 84
[20] A.W. Bosman, H.M. Janssen, E.W. Meijer, Chem. Rev. 99 (1999) 1665–1688.
[21] J. Wu, J. Zhou, F. Qu, P. Bao, Y. Zhang, L. Peng, Chem. Commun. 2005 (2005)
313–315.
(2009) 1196–1201.
[3] J.R. Dodson, T. Avellar, J. Athayde, C.J.A. Mota, Pure Appl. Chem. 86 (2014) 905–
911.
[22] X. Liu, J. Wu, M. Yammine, J. Zhou, P. Posocco, S. Viel, C. Liu, F. Ziarelli, M.
Fermeglia, S. Pricl, G. Victorero, C. Nguyen, P. Erbacher, J.-P. Behr, L. Peng,
Bioconjug. Chem. 22 (2011) 2461–2473.
[4] G. Knothe, Fuel Process. Technol. 86 (2005) 1059–1070.
[5] P. Lv, Y. Cheng, L. Yang, Z. Yuan, H. Li, W. Luo, Fuel Process. Technol. 110 (2013)
61–64.
[23] M.F. Ottaviani, M. Changiotti, A. Fattori, C. Coppola, P. Posocco, E. Laurini, X.
Liu, C. Liu, M. Fermeglia, L. Peng, S. Pricl, Phys. Chem. Chem. Phys. 16 (2014)
685–694.
[6] K. Ryu, J. Mech. Sci. Technol. 23 (2009) 3105–3113.
[7] A.S. Ramadhas, A. Muraleedharan, S. Jayaraj, Renew. Energy 30 (2005) 1789–
1800.
[24] K.H. Whitmire, J.C. Hutchison, A. Gardberg, C. Edwards, Inorganica Chim. Acta
294 (1999) 153–162.
[8] N. Usta, Energy Convers. Manage. 46 (2005) 2373–2386.
[9] S. Nomanbhay, R. Hussein, M.Y. Ong, Green Chem. Lett. Rev. 11 (2018) 135–
[25] A. Karadag, V.T. Yilmaz, C. Thoene, Polyhedron 20 (2001) 635–641.
[26] B. Sen, R.L. Dotson, J. Inorg. Nucl. Chem. 32 (1970) 2707–2716.
[27] T. Colclough, Ind. Eng. Chem. Res. 26 (1987) 1888–1895.
[28] B. Neises, W. Steglich, Angew. Chem. Int. Ed. 17 (1978) 522–524.
[29] C.L. Higgins, S.V. Fillip, A. Afsar, W. Hayes, React. Funct. Polym. 142 (2019)
119–127.
157.
[10] M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi, C. Della Pina, Angew. Chem. Int.
Ed. 46 (2007) 4434–4440.
[11] A. Behr, J. Eilting, K. Irawadi, J. Leschinski, F. Lindner, Green Chem. 10 (2008)
13–30.
[12] Y. Gu, A. Azzouzi, Y. Pouilloux, F. Jerome, J. Barrault, Green Chem. 10 (2008)
164–167.