Evaluation of thermal and oxidative stability of three generations of phenolic based novel dendritic fuel and lubricant additives

Article abstract of DOI:10.1016/j.reactfunctpolym.2019.06.009

Antioxidants, particularly those designed for use in hydrocarbon media, suffer from a variety of limitations including high volatility and poor solubility. Using 2,2-bis(hydroxymethyl)propionic acid as the branching unit, a series of novel dendrons featuring 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionic ester chain ends have been synthesised to provide improved solubility of such hindered phenolic antioxidants. The thermal stability, assessed by thermogravimetric analysis, revealed that all the functionalised dendrons have enhanced thermal stability when compared to commercial antioxidants (BHT, Irganox L135 and Irganox L57). Antioxidant ability was evaluated using pressurised differential scanning calorimetry and when blended with a lubricant base oil, at 0.5% w/w, an increase in antioxidant performance was observed when compared to the commercial antioxidants.

Full text of DOI:10.1016/j.reactfunctpolym.2019.06.009

ReactiveandFunctionalPolymers142(2019)119–127
Contents lists available at ScienceDirect
Reactive and Functional Polymers
journal homepage: www.elsevier.com/locate/react
Evaluation of thermal and oxidative stability of three generations of
phenolic based novel dendritic fuel and lubricant additives
Clare L. Higginsa, Sorin V. Filipb, Ashfaq Afsara, Wayne Hayesa,⁎
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
Keywords:
Antioxidants
Dendritic
Oxidation stability
Thermal stability
Oxidation induction time
Antioxidants, particularly those designed for use in hydrocarbon media, suffer from a variety of limitations
including high volatility and poor solubility. Using 2,2-bis(hydroxymethyl)propionic acid as the branching unit,
a series of novel dendrons featuring 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionic ester chain ends have been
synthesised to provide improved solubility of such hindered phenolic antioxidants. The thermal stability, as-
sessed by thermogravimetric analysis, revealed that all the functionalised dendrons have enhanced thermal
stability when compared to commercial antioxidants (BHT, Irganox L135 and Irganox L57). Antioxidant ability
was evaluated using pressurised differential scanning calorimetry and when blended with a lubricant base oil, at
0.5% w/w, an increase in antioxidant performance was observed when compared to the commercial anti-
oxidants.
1. Introduction
in the oxidative process, are scavenged efficiently. Additives such as
antioxidants are introduced into hydrocarbon materials to extend their
With a constant effort to reduce worldwide automotive emissions
and meet the ever-tightening environmental legislation controlling
Original Equipment Manufacturers (OEMs), innovative hydrocarbon-
based fluid formulations are required to assure vehicle performance and
efficiency at the design parameters [1–3]. The requirement for en-
hanced fuel efficiency and lower tailpipe CO2 emissions is a growing
demand, which can be achieved through the use of fuel and lubricant
additives designed to protect the engine from deterioration. The oxi-
dative degradation of organic materials has been studied for many years
and materials derived from petroleum, such as fuels and lubricants, are
particularly susceptible as a result of increasingly harsher conditions
within a combustion engine. [4–7] High temperatures and pressures in
the presence of air and metal surfaces or contaminants contribute to the
acceleration of oxidative degradation. [8,9] The oxidation process of
liquid hydrocarbons was first proposed by Bolland and Gee in 1946 who
described a free radical pathway. [10,11] Since then, the mechanism of
oxidation has been investigated extensively and these studies high-
lighted a complex process where by-products are formed such as acids,
alcohols, aldehydes, ketones and higher molecular weight hydro-
carbons. [8,9,12–16] Collectively, these by-products cause discoloura-
tion, carbonaceous deposition, increased oil viscosity and eventually
can lead to physical failure of the combustion engine. The rate of this
detrimental process can be decreased if alkyl peroxy radicals, produced
lifetime. A class of compounds described as hindered phenolics have
been studied comprehensively for their radical scavenging ability in
petroleum based-products. [17–22] It has been found that phenolic
antioxidants act by interrupting the reported [10] radical pathway by
providing a more labile hydrogen atom when compared to the hydro-
carbon species. This scavenging mechanism is highlighted in Fig. 1 for
the antioxidant butylated hydroxytoluene (BHT) and shows that phe-
nolic antioxidants can directly remove a number of radicals from the
oxidation chain reaction. This is achieved by both liberation of a hy-
drogen radical and through radical coupling processes, consequently
inhibiting the oxidation pathway [9]. Numerous examples of both
synthetic and naturally occurring sterically hindered phenols have been
reported to date (see Fig. 2) and structure-activity relationships of these
materials have been determined. [23–28] Antioxidants are, however,
eventually consumed either through chemical loss, from their anti-
oxidant action, or physical loss. [16,29,30] Physical loss is influenced
by factors such as volatilisation and precipitation out of the hydro-
carbon matrix and such decreases are often compensated by the addi-
tion of excess antioxidant at the outset. [16,29–31] This simplistic so-
lution does, however, have its own disadvantages since many of the
antioxidants used often exhibit limited solubility in hydrocarbons and
are also relatively expensive. [30,32] In an attempt to overcome the
solubility and efficacy issues that face these hydrocarbon additives, a
⁎ Corresponding author.
E-mail address: w.c.hayes@reading.ac.uk (W. Hayes).
https://doi.org/10.1016/j.reactfunctpolym.2019.06.009
Received 3 May 2019; Received in revised form 7 June 2019; Accepted 10 June 2019
Availableonline11June2019
1381-5148/©2019ElsevierB.V.Allrightsreserved.

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
Fig. 1. Proposed mechanism of radical scavenging of BHT.
Fig. 2. Examples of synthetic and natural phenolic antioxidants: BHT, 4,4′-methylenebis(2,6-di-tert-butylphenol) and α-tocopherol from vitamin E.
series of phenolic-based branched oligomers with an increasing number
of antioxidant end group units were prepared in this study. A branched
alkyl chain was utilised to aid the solubility of these phenolic-based
branched oligomers in hydrocarbon media such as base oils. The anti-
oxidant potencies were then evaluated using a series of thermal and
oxidative tests including thermogravimetric analysis (TGA) and pres-
surised differential scanning calorimetry (PDSC).
solvent was removed in vacuo and the resulting residue was dissolved in
dichloromethane (200 mL) and extracted with two portions of water
(40 mL). The organic phase was dried over magnesium sulfate (MgSO4),
filtered and then evaporated to yield 2 as a white waxy solid (10.32 g,
80%); 1H NMR (400 MHz/CDCl3)/ppm, δ = 1.21 (s, 3H, eCH3), 1.42
(s, 3H, eCH3), 1.45 (s, 3H, eCH3), 3.68 (d, 2H, J = 12.0 Hz, eCH2O,
equatorial), 4.20 (d, 2H, J = 12.0 Hz, eCH2O, axial); 13C NMR
(100 MHz/CDCl3)/ppm, δ = 18.4, 21.8, 25.4, 41.7, 65.9, 98.4, 180.0;
Found [M + H]+ (C8H14O4) m/z = 175.0964 (Calc. 175.0965); IR
(ATR) v/cm−1: 2984, 1719, 1072, 825, 717.
2. Experimental
2.1. General procedure for purification and characterisation
2.2.2. Preparation of the first generation acetonide (3) and general
esterification procedure
Reagents and solvents were purchased from Sigma Aldrich and used
without further purification except for 3-(3,5-di-tert-butyl-4-hydroxy-
phenyl)-propionic acid which was purchased from Alfa Aesar. All of the
solvents used were dried and freshly distilled prior to use.
Tetrahydrofuran (THF) was distilled under a nitrogen atmosphere from
sodium and benzophenone. Dichloromethane was distilled under a ni-
trogen atmosphere from calcium hydride. Thin layer chromatography
(TLC) was performed on aluminium sheets coated with Merck silica gel
60 F24. Spots were visualised under ultra-violet light (254 nm) with
potassium permanganate as the visualising agent. Column chromato-
graphy was performed using Merck silica gel 60 (40–63 μm particle
size) and a mobile phase as specified. Melting points were recorded
using a Stuart MP10 melting point apparatus. 1H NMR and 13C NMR
spectra were recorded using either CDCl3 or DMSO‑d6 as solvent on
either a Bruker Nanobay 400 or Bruker DPX 400 operating at 400 MHz
for 1H NMR or at 100 MHz for 13C NMR. Infrared (IR) spectroscopic
analysis was carried out using a Perkin Elmer 100 FT-IR instrument
with a diamond ATR sampling attachment with samples either as solids
or oils. Mass spectrometry analysis was carried out on a Thermo-Fisher
Scientific Orbitrap XL LC-MS. Samples were prepared as methanol so-
lutions (1 mg/mL) and were ionised using electrospray ionisation (ESI)
and the parent mass ions were quoted.
2-Ethylhexan-1-ol (7.30 mL, 47.0 mmol), 2,2,5-trimethyl-1,3-di-
oxane-5-carboxylic acid (2) (9.00 g, 51.7 mmol) and DPTS (60%) were
dissolved in dry dichloromethane (40 mL). The solution was stirred at
room temperature for 30 min. To the solution, N,N′-dicyclohex-
ylcarbodiimide (DCC) (12.60 g, 61.1 mmol) dissolved in dry di-
chloromethane (40 mL) was added over 15 min. The reaction was left
overnight at room temperature under a nitrogen atmosphere. The re-
action mixture was filtered to remove the white N,N′-dicyclohexylurea
(DCU) precipitate and the filtrate was concentrated. The crude product
was dissolved in dichloromethane and washed sequentially with 0.5 M
HCl and saturated NaHCO3. The organic phase was dried over MgSO4,
filtered and the solvent was removed in vacuo to yield a pale-yellow oil.
Hexane was added to the crude product and the resulting white pre-
cipitate was filtered off. The solvent was once again removed in vacuo
and the resulting oil was purified by flash column chromatography on
silica eluting with hexane/ethyl acetate (95:5) (Rf = 0.23) to afford 3 as
a thin colourless oil (9.50 g, 71%); 1H NMR (400 MHz/CDCl3)/ppm,
δ = 0.90 (s, 3H, eCH3), 1.21 (s, 3H, eCH3), 1.30 (m, 6H, eCH2), 1.39
(s, 3H, eCH3), 1.43 (s, 3H, eCH3), 1.61 (m, 1H, eCH) 3.63 (d, 2H,
J = 12.0 Hz, eCH2O), 4.07 (m, 2H, eCH2), 4.17 (d, 2H, J = 12.0 Hz,
eCH2O); 13C NMR (100 MHz/CDCl3)/ppm, δ = 11.0, 14.0, 18.7, 23.0,
23.8, 24.2, 28.9, 30.4, 38.8, 41.9, 66.0, 67.0, 98.0, 174.3; Found
[M + H]+ (C16H30O4) m/z = 287.2222 (Calc. 287.2223); IR (ATR) v/
cm−1: 2934, 1735, 1158, 1080.
2.2. Synthesis
2.2.1. 2,2,5-trimethyl-1,3-dioxane-5-carboxylic acid (2)
2,2-Bis(hydroxymethyl)propanoic acid (bis(MPA))
1
(10.00 g,
2.2.3. Preparation of the first generation hydroxyl linker (4) and general
procedure for removal of acetonide protecting group
First generation acetonide 3 (2.5 g, 8.73 mmol) was dissolved in
methanol (30 mL) and DOWEX 5W-X8 resin (ca. 2 g) was added. The
solution was stirred at 50 °C and monitored by TLC analysis, using
74.6 mmol), 2,2-dimethoxypropane (13.8 mL, 111.8 mmol) and p-to-
luene sulfonic acid (p-TsOH) (0.71 g, 3.7 mmol) were dissolved in
acetone (50 mL). The reaction was stirred at room temperature for 2 h.
The catalyst was neutralised with 1.0 mL of NH3/EtOH (50:50). The
120

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
hexane/ethyl acetate (80:20) as the eluent, until the deprotection was
complete. The resin was filtered off and the filtrate was concentrated in
vacuo to yield 4 as a colourless viscous oil (2.02 g, 94%); 1H NMR
(400 MHz/CDCl3)/ppm, δ = 0.90 (m, 6H, eCH3), 1.08 (s, 3H, eCH3),
1.36 (m, 8H, eCH2), 1.60 (m, 1H, eCH) 3.08 (t, 2H, J = 16.0 Hz,
eOH), 3.72 (m, 2H, eCH2O), 3.88 (m, 2H, eCH2O), 4.09 (m, 2H,
eCH2); 13C NMR (100 MHz/CDCl3)/ppm, δ =11.0, 14.0, 17.2, 22.9,
23.8, 28.9, 38.7, 49.2, 67.3, 68.0, 176.1; Found [M + H]+ (C13H27O4)
m/z = 247.1909 (Calc. 247.1910); IR (ATR) v/cm−1: 3458, 2694,
1722, 1042.
eCH2), 3.99 (m, 2H, eCH2), 4.11 (m, 12H, eCH2), 4.65 (t, 8H,
J = 12 Hz, eOH); 13C NMR (100 MHz/DMSO‑d6)/ppm, δ = 10.7, 13.8,
16.7, 16.9, 17.1, 22.3, 23.2, 28.2, 29.7, 38.0, 46.2, 46.3, 50.2, 63.6,
64.5, 65.8, 171.8, 172.0, 174.0; Found [M + H]+ (C43H75O22) m/
z = 943.4751 (Calc. 943.4745); IR (ATR) v/cm−1: 3278, 2934, 1727,
1119, 1043.
2.2.8. Preparation of first generation diphenol (10)
3-(3,5-Di-tert-butyl-4-hydroxyphenyl)propanoic acid
9
(3.40 g,
12.17 mmol), first generation hydroxyl linker 4 (1.00 g, 4.059 mmol),
DPTS (60%) and DCC (2.51 g, 12.17 mmol) were allowed to react ac-
cording to the general esterification procedure. The crude product was
purified by flash column chromatography on silica eluting with
hexane/ethyl acetate (90:10) (Rf = 0.38) to afford 10 as a viscous
colourless oil (2.34 g, 75%); 1H NMR (400 MHz/CDCl3)/ppm, δ = 0.88
(m, 6H, eCH3), 1.16 (s, 3H, eCH3), 1.33 (m, 8H, eCH2), 1.43 (s, 36H,
eCH3), 1.61 (m, 1H, eCH), 2.60 (t, 4H, J = 16 Hz, eCH2), 2.85 (t, 4H,
J = 16.0 Hz, eCH2), 4.04 (m, 2H, eCH2), 4.23 (s, 4H, eCH2), 5.08 (s,
2H, eOH), 6.98 (s, 4H, -ArCH); 13C NMR (100 MHz/CDCl3)/ppm, δ
=11.0, 14.0, 17.7, 22.9, 23.7, 28.9, 30.3, 30.9, 34.3, 36.2, 38.7, 46.4,
65.5, 67.3, 124.7, 130.8, 135.9, 152.2, 172.7, 172.9; Found [M + H]+
(C47H75O8) m/z = 767.5462 (Calc. 767.5462); IR (ATR) v/cm−1: 3644,
2957, 1734, 1435, 1135, 756.
2.2.4. Preparation of the second generation acetonide (5)
2,2,5-Trimethyl-1,3-dioxane-5-carboxylic
acid
2
(4.88 g,
28.01 mmol), first generation hydroxyl linker 4 (3.00 g, 12.18 mmol),
DPTS (60%) and DCC (5.78 g, 28.01 mmol) were allowed to react fol-
lowing the general esterification procedure. The crude product was
purified by flash column chromatography on silica eluting with
hexane/ethyl acetate (90:10) (Rf = 0.05) increasing polarity to (80:20)
to afford 5 as a colourless oil (4.60 g, 70%); 1H NMR (400 MHz/CDCl3)/
ppm, δ = 0.89 (m, 6H, CH3), 1.16 (s, 6H, CH3), 1.29–1.42 (m, 23H, CH2
and CH3), 1.59 (m, 1H, CH), 3.63 (d, 4H, J = 12 Hz, CH2), 4.05 (m, 2H,
CH2), 4.16 (d, 4H, J = 12 Hz, CH2), 4.33 (s, 4H, CH2); 13C NMR
(100 MHz/CDCl3)/ppm, δ = 10.9, 14.0, 17.8, 18.5, 22.4, 23.7, 24.8,
28.9, 30.3, 38.7, 42.0, 46.8, 65.3, 65.9, 67.6, 98.1, 172.6, 173.5; Found
[M + Na]+ (C29H50O10Na) m/z = 581.3295 (Calc. 581.3296); IR
(ATR) v/cm−1: 2966, 1734, 1079, 831.
2.2.9. Preparation of second generation tetraphenol (11)
3-(3,5-Di-tert-butyl-4-hydroxyphenyl)propanoic acid
9
(4.36 g,
15.67 mmol), second generation hydroxyl linker 6 (1.50 g, 3.13 mmol),
DPTS (60%) and DCC (3.23 g, 15.67 mmol) were allowed to react ac-
cording to the general esterification method. The crude product was
purified by flash column chromatography on silica eluting with
hexane/ethyl acetate (90:10) (Rf = 0.13) to afford 11 as a colourless
glassy solid (3.39 g, 72%); Mp (42e45 °C); 1H NMR (400 MHz/CDCl3)/
ppm, δ = 0.87 (m, 6H, eCH3), 1.13 (s, 6H, eCH3), 1.22 (s, 3H, eCH3),
1.31 (m, 8H, eCH2), 1.42 (s, 72H, eCH3), 1.57 (m, 1H, eCH), 2.59 (t,
2H, J = 16 Hz, eCH2), 2.83 (t, 2H, J = 16 Hz, eCH2), 4.01 (m, 2H,
eCH2), 4.18 (s (br), 8H, eCH2), 4.25 (s (br), 4H, eCH2), 5.06 (s, 4H,
eOH), 6.98 (s, 8H, -ArCH); 13C NMR (100 MHz/CDCl3)/ppm, δ = 11.0,
14.0, 17.7, 18.2, 22.9, 23.7, 28.9, 30.3, 30.9, 34.3, 36.2, 38.7, 46.4,
65.1, 65.7, 67.7, 124.7, 130.8, 135.9, 152.2, 172.0, 172.6; Found
2.2.5. Preparation of the second generation hydroxyl linker (6)
The second generation acetonide 5 (3.9 g, 7.0 mmol) was dissolved
in methanol (40 mL). Using the general procedure for removal of the
acetonide protective group described above, 6 was obtained as a waxy
solid (2.32 g, 70%); Mp (38–41 °C); 1H NMR (400 MHz/CDCl3)/ppm,
δ = 0.90 (m, 6H, eCH3), 1.05 (s, 6H, eCH3), 1.31 (m, 11H, eCH3,
eCH2), 1.59 (m, 1H, eCH), 3.22 (m, 4H, eOH), 3.71 (m, 4H, eCH2),
3.83 (m, 4H, eCH2), 4.07 (m, 2H, eCH2), 4.27 (d, 2H, J = 12 Hz,
eCH2), 4.45 (d, 2H, J = 12 Hz, eCH2); 13C NMR (100 MHz/CDCl3)/
ppm, δ = 10.9,14.0, 17.1, 18.2, 22.9, 23.7, 28.9, 30.4, 38.7, 46.5, 49.7,
64.8, 67.8, 68.1, 173.1, 175.2; Found [M + H]+ (C23H43O10) m/
z = 479.2836 (Calc. 479.2851); IR (ATR) v/cm−1: 3284, 2940, 1733,
1240, 1115, 1044.
[M + Na]+ (C91H138O18
) m/z = 1541.9775 (Calc. 1541.9883); IR
(ATR) v/cm−1: 3646, 2958, 1739, 1435, 1121.
2.2.6. Preparation of the third generation acetonide (7)
2,2,5-Trimethyl-1,3-dioxane-5-carboxylic
acid
2
(9.43 g,
2.2.10. Preparation of third generation octaphenol (12)
3-(3,5-Di-tert-butyl-4-hydroxyphenyl)propanoic acid
54.16 mmol), second generation hydroxyl linker 6 (4.32 g, 9.03 mmol),
DPTS (60%) and DCC (11.18 g, 54.16 mmol) were allowed to react
according to the general esterification procedure. The crude product
was purified by flash column chromatography on silica eluting with
hexane/ethyl acetate (70:30) increasing polarity to (50:50) (Rf = 0.28)
to afford 7 as a white waxy solid (5.95 g, 60%); 1H NMR (400 MHz/
CDCl3)/ppm, δ = 0.89 (t, 6H, J = 12 Hz, eCH3), 1.15 (s, 12H, eCH3),
1.27 (m, 16H, eCH3), 1.35 (s, 13H, eCH3 and eCH2), 1.41 (s, 12H,
eCH3), 1.60 (m, 1H, eCH), 3.62 (d, 8H, J = 12 Hz, eCH2), 4.03 (m,
2H, eCH2), 4.15 (d, 8H, J = 12 Hz, eCH2), 4.26 (m, 4H, eCH3), 4.31
(m, 8H, eCH3); 13C NMR (100 MHz/CDCl3)/ppm, δ =10.9, 14.1, 17.7,
18.5, 22.2, 22.9, 23.7, 25.1, 28.9, 30.3, 38.7, 42.0, 46.7, 46.8, 64.9,
66.0, 67.9, 98.1, 171.9, 172.1, 173.5; Found [M + Na]+
(C55H90O22Na) m/z = 1125.5800 (Calc. 1125.5824); IR (ATR) v/
cm−1:2937, 1723, 1079, 830.
9
(1.48 g,
5.30 mmol), third generation hydroxyl linker 8 (0.50 g, 0.53 mmol),
DPTS (60%) and DCC (1.09 g, 5.30 mmol) were allowed to react ac-
cording to the general esterification method with the exception of di-
methylacetamide which was used as the solvent (30 mL). The crude
reaction was purified by precipitation into water followed by the gen-
eral washing procedure and flash column chromatography on silica
eluting with chloroform/methanol (99.5:0.5) to afford 12 as a white
waxy solid (0.24 g, 15%); 1H NMR (400 MHz/CDCl3)/ppm, δ = 0.85
(m, 6H, eCH3), 1.13 (s, 12H, eCH3), 1.21 (s, 6H, eCH3), 1.26 (s, 11H,
eCH2), 1.41 (s, 142H, tert-butyl eCH3), 1.54 (m, 1H, eCH), 2.59 (t, 2H,
J = 16 Hz, eCH2), 2.83 (t, 2H, J = 16 Hz, eCH2), 4.18 (m, 2H, eCH2),
4.22 (m, 28H, eCH2), 5.06 (s, 8H, eOH), 6.97 (s, 16H, -ArCH). 13C
NMR (100 MHz/CDCl3)/ppm, δ = 10.9, 14.1, 17.7, 22.9, 30.3, 30.8,
34.3, 36.1, 46.4, 65.0, 124.7, 130.8, 135.9, 152.2, 171.9, 172.5. Found
[M + Na]+ (C179H266O38) m/z = 3046.8690 (Calc. 3046.8882); IR
(ATR) v/cm−1: 3640, 2954, 1736, 1435, 1120.
2.2.7. Preparation of the third generation hydroxyl linker (8)
The third generation acetonide 7 (3.83 g, 3.5 mmol) was dissolved
in methanol (40 mL). Using the general procedure for removal of the
acetonide protective group described above, 8 was obtained as a white
waxy solid (3.03 g, 79%); 1H NMR (400 MHz/DMSO‑d6)/ppm, δ = 0.85
(m, 6H, eCH3), 1.01 (s, 12H, eCH3), 1.17 (s, 6H, eCH3, eCH2), 1.20 (s,
3H, eCH3), 1.26 (m, 8H, eCH2), 1.56 (m, 1H, eCH), 3.46 (m, 16H,
2.3. General procedure for thermal and oxidative analysis
Thermogravimetric analysis (TGA) was performed using a TA in-
strument TGA 2950. TGA was carried out under a nitrogen atmosphere
from ambient temperature to 500 °C at a rate of 10 °C/min using a
121

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
sample of approximately 10 mg. Pressurised differential scanning ca-
lorimetry (PDSC) was carried out at the BP Technology Centre,
Pangbourne. Oxidation induction time (OIT) was performed using a TA
instrument Q10 (0010-0141) or Q20 (0020P-0137). The industry
standard CEC L 085-99 method was followed whereby 2 mg of sample
was added to an aluminium crucible. The cell was pressurised to 100 psi
with cylinder air and the temperature was raised to 50 °C and held
isothermally for 5 min. The temperature was then ramped at 20 °C/min
to 210 °C and held isothermally until the oxidation of the sample was
induced. The time of onset of the exotherm minus the time taken to
reach 210 °C was reported. PDSC oxidation onset temperature (OOT)
was performed using a TA instrument 2910. An in-house method was
used whereby 0.5 mg of sample was added to an aluminium crucible.
The cell was pressurised to 500 psi with cylinder air with a flow of
60 mL/min. The temperature was raised to 50 °C and was allowed to
stabilise before heating at 50 °C/min to 350 °C. The temperature at
which the oxidation exotherm occurred was reported.
3. Results and discussion
3.1. Synthesis of dendritic antioxidants
The synthetic route used to produce the first generation hydroxyl
linker 4 is shown in Scheme 1, where the first step involved protection
of the 1,3-diol moiety of bis(MPA) 1 with an acetonide group and was
achieved by reacting 1 with 2,2-dimethoxypropane (DMP) in the pre-
sence of catalytic amount of p-toluene sulfonic acid (TsOH). Ester-
ification of 2 using N,N′-dicyclohexylcarbodiimide (DCC) as the cou-
pling agent and 4-(dimethylamino)pyridinium-4-toluenesulfonate
(DPTS) as an esterification catalyst were employed for the synthesis of 3
and all subsequent esterifications. [33] To yield the first generation
hydroxyl linker 4, the acetonide group was removed by stirring the
protected diol 3 in methanol in the presence of an acidic Dowex resin.
Higher generation hydroxyl terminated polyesters were synthesised
using the same protection and deprotection strategy as outlined in
Scheme 1. A DCC mediated coupling between 4 and 2 was employed in
the synthesis of the diacetonide triester 5 and this was deprotected to
yield the second generation linker 6. The third generation hydroxyl
linker 8 was synthesised according to the synthetic protocol shown in
Scheme 1, where a DCC mediated coupling was first employed in the
Scheme 1. Synthesis of the first generation hydroxyl linker 4 based upon bis(MPA). Second and third generation polyester hydroxyl linker 6 and 8 were synthesised
using the same protection, esterification and deprotection strategy.
122

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
Scheme 2. Attachment of antioxidant functionality to the first generation hydroxyl linker.
Fig. 3. Structures of the second generation (11) and third generation (12) antioxidants, respectively.
synthesis of 7 followed by deprotection of the acetonide protecting
groups in order to afford the desired polyol 8 in 79% yield. The anti-
oxidant functionality was then introduced in a divergent approach
utilising the proven DCC mediated coupling route (Scheme 2) to obtain
the desired diphenol 10. The sterically hindered phenol 3-(3,5-di-tert-
butyl-4-hydroxy-phenyl)-propionic acid 9 was chosen as the terminal
unit to provide antioxidant functionality onto the hydroxyl linkers of
the bis(MPA)-based ester dendrons for several reasons. The acid moiety
of 9 render it suitable for DCC esterification which had proved effective
in this synthetic study and the ethyl linker between the aromatic ring
and the acidic functionality serves to aid the solubility of the branched
antioxidants in a hydrocarbon-based medium. The second and third
generation polyester hydroxyl linkers (Fig. 3) were subjected to the
same synthetic procedure as described for the first generation ester that
afforded the tetra- 11 and octa- 12 phenolic esters in 72% and 15%
yields, respectively. The third generation octaphenol 12, however, re-
quired the use of an alternative solvent, dimethylacetamide, to over-
come the insolubility of the third generation hydroxyl linker. In addi-
tion, purification of 12 also proved to be more challenging with initial
precipitation into water to remove the dimethylacetamide, followed by
dissolution in dichloromethane and multiple washings with sodium
hydroxide to remove the excess 9 that had been used. Flash column
chromatography was then employed twice to remove traces of the N,N′-
dicyclohexylurea (DCU) by-product from the esterification process. The
lengthy purification process suggested that impurities and solvent were
easily trapped within the large structure of the compound 12 and such
phenomenon within dendritic structures has been reported previously
by Meijer and co-workers. [34,35] The low yield of the octaphenol 12
can be attributed to the steric congestion of the hydroxyl end groups in
the third generation hydroxyl linker 8, which inhibits the desired
coupling. For spectroscopic data of these branched materials, please
refer to the Supporting Information (SI).
3.2. Thermal stability studies
TGA was used to investigate the thermal stability characteristics of
the synthesised antioxidants 10–12. The thermal stability of the di-
phenol 10 was compared to three commercial antioxidants, BHT,
Irganox L135 and Irganox L57 (Fig. 4). As shown in Fig. 5, the thermal
stability of 10 was significantly higher when compared to BHT and
examination of molecular weight alone revealed that BHT
(Mw = 220.36) was nearly 4 times smaller than 9 (Mw = 767.10) in-
dicating that it was much more susceptible to volatilisation, hence
complete consumption was seen at ca. 150 °C. The first generation di-
phenol 10 also revealed a thermal stability of ca. 100 °C higher than
both Irganox L135 (Average Mw = 390.61) and L57 (Average
123

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
analysed using PDSC to monitor the heat effects associated with phase
transitions and chemical reactions as a function of temperature.
Oxidation induction time (OIT) and oxidation onset temperature (OOT)
were used to investigate the effect of antioxidants on the stability of an
oil sample. OIT revealed that the presence of 10 and 11 in the base oil
had resulted in significant increase in the stability of the sample (ca.
229%) as shown in Fig. 7. The induction time was increased from <
3 min for the unblended base oil to ca. 12 min for the blended samples.
In addition, 10 and 11 showed superior performance to both com-
mercial antioxidants, Irganox L135 and Irganox L57. To further in-
vestigate the properties of these antioxidants, a normalisation test was
carried out where additional blends were generated with respect to the
number of moles of Irganox L135 which possesses only one active
phenolic group. The number of moles of the first and second generation
were then either halved or quartered corresponding to their respective
2 and 4 active phenolic groups (Table 1).
The results from these blends shown in Fig. 8 were very promising
and revealed that even though there was a slight drop in induction
time, the dendrons 10 and 11 still performed much better than Irganox
L135. The OOT results for each oil blend are presented in Fig. 9 where
again, a significant increase in temperature was observed when 10 and
11 were incorporated into the blend when compared to the base oil in
isolation. The above results indicated that the onset of oxidation of a
hydrocarbon matrix can be delayed using 10 and 11 and that these
additives not only provide increased oxidative stability but also en-
hances the performance of the bulk matrix at higher temperatures.
Fig. 4. Commercial antioxidants 13) Irganox L135 and 14) Irganox L57.
Mw = 337.55). Again, these results confirmed that the higher mole-
cular weight of 10 was contributing to the observed enhanced thermal
stability properties. The thermal stability of polyphenols 11 and 12
were also analysed and revealed a one-step degradation, other than loss
of residual entrapped solvent (ca. 10%) at ca. 100 °C as shown in Fig. 6.
A slight increase in the stability was observed between each generation,
most likely as a result of increased bulkiness. [30] TGA results con-
firmed that the design of these novel dendritic compounds had suc-
cessfully resulted in increasing the bulkiness, thus reducing the effect of
physical loss of the antioxidant through volatilisation. It is not un-
reasonable to thus assume that these new branched antioxidants will be
present in hydrocarbon media for an increased period of time at ele-
vated temperatures in relation to the lower molecular weight systems
and will consequently provide prolonged stability to the bulk phase.
3.3. Oxidative stability studies
To further assess the antioxidant potential, 10–12 were blended into
a synthetic lubricant base oil - Durasyn 164 (a polyalphaolefin, hy-
drogenated hydrocarbon base oil composed of dec-1-ene trimers typi-
cally used in lubricating oils). At this stage it was found that both the
first generation 10 and second generation 11 were soluble in the hy-
drocarbon, however, the third generation 12 was insoluble and hence
analysis of this polyphenol 12 proved to be unfeasible. The commercial
antioxidants Irganox L135 and Irganox L57 were once again used as a
comparison and samples were prepared by blending of 0.5% w/w of
each antioxidant in 50 mL of lubricant base oil. The blends were
4. Conclusions
In summary, we report a divergent synthetic approach to develop a
series of novel antioxidant terminated polyester dendrons (3 different
generations) using 2,2-bis(hydroxymethyl)propionic acid (bis(MPA)) as
the branching unit to enhance solubility and sterically hindered phenol
3-(3,5-di-tert-butyl-4-hydroxy-phenyl)-propionic acid to provide anti-
oxidant functionality. The antioxidant dendrons 10 and 11 (first and
second generations, respectively) were scaled up successfully to 100 g
and the thermal stability, assessed by thermogravimetric analysis,
Fig. 5. Thermogravimetric analysis of 10 when compared to BHT, Irganox L135 and Irganox L57.
124

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
Fig. 6. Thermogravimetric analysis of polyphenol dendrons 10, 11 and 12.
14
12
10
8
12.14
11.78
5.92
6
3.58
4
2
0
Base Oil
Irganox L135
Irganox L57
10
11
Sample
Fig. 7. Average Oxidation Induction Time of 0.5% w/w antioxidant-base oil samples run in duplicate.
revealed that all the functionalised dendrons (10, 11 and 12) have
enhanced (ca. 100 °C higher) thermal stability when compared to BHT,
Irganox L135 and Irganox L57. Antioxidant ability was evaluated using
pressurised differential scanning calorimetry and when blended with a
lubricant base oil, at 0.5% w/w, an increase in antioxidant performance
was observed for dendrons 10 and 11 when compared to BHT, Irganox
L135 and most promisingly the high temperature antioxidant Irganox
L57. These results confirmed that the onset of oxidation of a hydro-
carbon matrix can be delayed using the dendritic antioxidants. This
study revealed that encouragingly these antioxidant dendritic additives
provide not only increased oxidative stability of hydrocarbon matrices
Table 1
Calculations to determine amount of 10 and 11 required when compared to
Irganox L135.
Sample
Mw
Number of phenolic
functionalities
Number of
mmols
Mass required
for the oil blend
Irganox L135 390.61
10
11
1
2
4
0.64
0.32
0.16
0.2500 g
0.2455 g
0.2410 g
767.09
1506.08
125

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
14
12
10
8
12.14
11.78
10.23
9.84
6
3.58 3.58
4
2
0
Base Oil
Irganox L135
AO 10
AO 11
Sample
Average Induction Time (w/w%)
Average Induction Time (mol%)
Fig. 8. Average Oxidation induction time comparison of w/w and mol% oil blends run in duplicate.
260
249.60
249.43
249.07
250
240
230
220
210
200
237.66
223.11
Base Oil
Irganox L135
Irganox L57
10
11
Sample
Fig. 9. Average Oxidation Onset Temperature of 0.5% w/w antioxidant-base oil samples run in duplicate.
for longer time periods but also serve to enhance their performance at
higher temperatures.
References
[1] Deloitte, Global Automotive Consumer Study Exploring consumers' Mobility
Choices and Transportation Decisions, (2014).
[2] PriceWaterhouseCoopers, The Automotive Industry and Climate Change:
Framework and Dynamics of the CO2 Revolution, (2007).
[3] KPMG International, The Transformation of the Automotive Industry: The
Environmental Regulation Effect, (2010).
[4] E.H. Farmer, A. Sundralingam, The course of autoxidation reactions in poly-
isoprenes and allied compounds. Part I. The structure and reactive tendencies of the
peroxides of simple olefins, J. Chem. Soc. (1942) 121–139.
[5] E.H. Farmer, A. Sundralingam, The course of autoxidation reactions in poly-
isoprenes and allied compounds. Part VI. The peroxidation of rubber, J. Chem. Soc.
(1943) 125–133.
[6] J.W. Robinson, Y. Zhou, J. Qu, J.T. Bays, L. Cosimbescu, Highly branched poly-
ethylenes as lubricant viscosity and friction modifiers, React. Funct. Polym. 109
(2016) 52–55.
[7] C.E. Frank, Hydrocarbon autoxidation, Chem. Rev. 46 (1950) 155–169.
[8] R.M. Mortier, S.T. Orszulik, Chemistry and Technology of Lubricants, Springer,
Netherlands, 1997.
[9] T. Mang, W. Dresel, Lubricants and Lubrication, 2nd ed., Wiley-VCH, Weinheim,
2007.
[10] J.L. Bolland, G. Gee, Kinetic studies in the chemistry of rubber and related mate-
rials. II. The kinetics of oxidation of unconjugated olefins, Trans. Faraday Soc. 42
(1946) 236–243.
Acknowledgements
The authors acknowledge the UK Biotechnology and Biological
Sciences Research Council (BBSRC) and BP p.l.c. for financial support.
Use of the Chemical Analysis Facility (CAF) at the University of Reading
and Analytical department at the BP Technology Centre, Pangbourne
are gratefully acknowledged.
Data availability
The raw/processed data required to reproduce these findings cannot
be shared at this time due to technical or time limitations. Processed
data can be found in the associated Supporting Information file.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.reactfunctpolym.2019.06.009.
[11] E.T. Denisov, I.B. Afanas'ev, Oxidation and Antioxidants in Organic Chemistry and
126

C.L. Higgins, et al.
ReactiveandFunctionalPolymers142(2019)119–127
[24] Y. Ohkatsu, T. Nishiyama, Phenolic antioxidants-effect of ortho-substituents,
Polym. Degrad. Stab. 67 (2000) 313–318.
[25] Y.L. Kobzar, I.M. Tkachenko, V.N. Bliznyuk, V.V. Shevchenko, Fluorinated poly-
benzoxazines as advanced phenolic resins for leading-edge applications, React.
Funct. Polym. 133 (2018) 71–92.
Biology, Taylor & Francis, Florida, 2005.
[12] A.C. de Leon, Q. Chen, N.B. Palaganas, J.O. Palaganas, J. Manapat, R.C. Advincula,
High performance polymer nanocomposites for additive manufacturing applica-
tions, React. Funct. Polym. 103 (2016) 141–155.
[13] F.R. Mayo, Free radical autoxidations of hydrocarbons, Accounts Chem. Res. 1
(1968) 193–201.
[26] T. Kajiyama, Y. Ohkatsu, Effect of para-substituents of phenolic antioxidants,
[14] G. Karavalakis, S. Stournas, D. Karonis, Evaluation of the oxidation stability of
diesel/biodiesel blends, Fuel. 89 (2010) 2483–2489.
[15] S. Schober, M. Mittelbach, The impact of antioxidants on biodiesel oxidation sta-
Polym. Degrad. Stab. 71 (2001) 445–452.
[27] T. Kajiyama, Y. Ohkatsu, Effect of meta-substituents of phenolic antioxidant-
s—proposal of secondary substituent effect, Polym. Degrad. Stab. 75 (2002)
535–542.
bility, Eur. J. Lipid Sci. Technol. 106 (2004) 382–389.
[16] H. Zweifel, Stabilisation of Polymeric Materials, Springer-Verlag, Berlin, 1998.
[17] K.U. Ingold, Inhibition of the autoxidation of organic substances in the liquid phase,
Chem. Rev. 61 (1961) 563–589.
[28] A. Torres de Pinedo, P. Peñalver, J.C. Morales, Synthesis and evaluation of new
phenolic-based antioxidants: structure–activity relationship, Food Chem. 103
(2007) 55–61.
[18] R.O. Dunn, Effect of antioxidants on the oxidative stability of methyl soyate (bio-
[29] N.C. Billingham, P. Garcia-Trabajo, The possibilities of microencapsulated anti-
oxidants for polymers. I: a theoretical analysis, Polym. Degrad. Stab. 48 (1995)
419–426.
[30] H. Bergenudd, P. Eriksson, C. DeArmitt, B. Stenberg, E. Malmström, M. Jonsson,
Synthesis and evaluation of hyperbranched phenolic antioxidants of three different
generations, Polym. Degrad. Stab. 76 (2002) 503–509.
diesel), Fuel Process. Technol. 86 (2005) 1071–1085.
[19] A. Dario Arrua, J.F. Basbus, M.C. Strumia, C.I. Alvarez Igarzabal, M.A. Nazareno,
Synthesis of macroporous polymers with radical scavenging properties by im-
mobilization of polyphenolic compounds, React. Funct. Polym. 72 (2012) 807–813.
[20] W.W. Focke, I. van der Westhuizen, A.B.L. Grobler, K.T. Nshoane, J.K. Reddy,
A.S. Luyt, The effect of synthetic antioxidants on the oxidative stability of biodiesel,
Fuel. 94 (2012) 227–233.
[31] E. Malmström, M. Jonsson, A. Hult, Hyperbranched aliphatic polyesters,
Macromolecules 28 (1995) 1698–1703.
[21] D. Lomonaco, F.J.N. Maia, C.S. Clemente, J.P.F. Mota, A.E. Costa, S.E. Mazzetto,
Thermal studies of new biodiesel antioxidants synthesized from a natural occurring
phenolic lipid, Fuel. 97 (2012) 552–559.
[22] N.A. Santos, A.M.T.M. Cordeiro, S.S. Damasceno, R.T. Aguiar, R. Rosenhaim,
J.R. Carvalho Filho, I.M.G. Santos, A.S. Maia, A.G. Souza, Commercial antioxidants
and thermal stability evaluations, Fuel. 97 (2012) 638–643.
[23] Z. Balzadeh, H. Arabi, Novel phenolic antioxidant-functionalized dendritic poly-
ethylene: synthesis by tailor-made nickel(II) α-diimine-catalyzed copolymerization
and its characteristics as non-releasing additive, React. Funct. Polym. 111 (2017)
68–78.
[32] J.C.J. Bart, Additives in Polymers: Industrial Analysis and Applications, John Wiley
& Sons Ltd, Chichester, 2005.
[33] B. Neises, W. Steglich, Simple method for the esterification of carboxylic acids,
Angew. Chem. Int. Ed. (1978) 522–524.
[34] J.F. Jansen, E.M.M. de Brabander-van den Berg, E.W. Meijer, Encapsulation of guest
molecules into a dendritic box, Science 266 (80) (1994) 1226–1229.
[35] F.G.A. Jansen, E.W. Meijer, E.M.M. de Brabander-van den Berg, The dendritic box:
shape-selective liberation of encapsulated guests, J. Am. Chem. Soc. 117 (1995)
4417–4418.
127