See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/317136905

Removal of naphthenic acids from high acid crude via esterification with
methanol

Article  in  Fuel Processing Technology · October 2017

DOI: 10.1016/j.fuproc.2017.05.015

CITATIONS READS

4 133

4 authors:

Muhammad Kashif Khan Asim Riaz

Sungkyunkwan University Sungkyunkwan University
8 PUBLICATIONS   23 CITATIONS    11 PUBLICATIONS   70 CITATIONS   

SEE PROFILE SEE PROFILE

Minhoe Yi Jaehoon Kim

Korea Advanced Institute of Science and Technology Sungkyunkwan University
5 PUBLICATIONS   23 CITATIONS    168 PUBLICATIONS   2,519 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

crude oil upgrading View project

Lignin depolymerization View project

All content following this page was uploaded by Muhammad Kashif Khan on 14 October 2017.

The user has requested enhancement of the downloaded file.

 Fuel Processing Technology 165 (2017) 123–130

Contents lists available at ScienceDirect

Fuel Processing Technology

journal homepage: www.elsevier.com/locate/fuproc

Removal of naphthenic acids from high acid crude via esterification

with methanol
Muhammad Kashif Khan a, Asim Riaz a, Minhoe Yi b, Jaehoon Kim a,c,⁎
a
School of Mechanical Engineering, Sungkyunkwan University, 2066, Seobu-Ro, Jangan-Gu, Suwon, Gyeong Gi-Do, 16419, Republic of Korea
b
SK Innovation, 325, Exporo, Yuseong-gu, Daejeon 34124, Republic of Korea
c
SKKU Advanced Institute of Nano Technology (SAINT), 2066, Seobu-Ro, Jangan-Gu, Suwon, Gyeong Gi-Do, 16419, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Refining of high acid crudes (HACs) is seriously limited by extensive corrosion of refinery equipment due to the
Received 22 December 2016 presence of intrinsic naphthenic acids (NAs). Herein, in the absence of a catalyst or external hydrogen, effective
Received in revised form 15 May 2017 removal of the NAs contained in HACs was achieved using methanol at moderate temperature and moderate
Accepted 15 May 2017
pressure. Several process variables, including the temperature, crude concentration, and reaction time are ex-
Available online xxxx
plored to optimize the removal of NAs from HACs. At 250 °C, 6.4 MPa, and 33.3 wt% crude, a very low total
Keywords:
acid number (TAN) of 0.08 mg-KOH/g-oil (96.9% reduction efficiency) with a high oil yield (95 wt%) is achieved.
High acid crudes During the reduction of the TAN using methanol, some fraction of the resins and asphaltenes in the crude oil crack
Deacidification to form low-molecular-weight aromatic compounds. Compared to deacidification using methanol, pyrolysis
Naphthenic acids (without methanol) results in a much lower TAN reduction efficiency (38.6%). The major deacidification mech-
Methanol anism using methanol is esterification, while that of pyrolysis (without methanol) is decarboxylation.
Total acid number © 2017 Elsevier B.V. All rights reserved.
Unconventional crude oils

1. Introduction NAs are naturally-existing, complex aliphatic, monocyclic, and polycy-

In recent years, technological advancements and population growth
have led to a steep increase in the global energy demand. Current con-
ventional crude oil reserves are not sufficient to sustain the increasing
energy demand because of the rapid depletion of oil resources [1]. Un-
conventional crudes (heavy crudes, high acid crudes, bitumen, oil/tar
sand, and shale oil), which account for N70% of the total crude oil re-
serves, are considered to be the most practical alternatives to conven-
tional crude oils. Despite their abundance, unconventional crude oils
possess several operational and environmental drawbacks that make
safe and cost-effective utilization of these crudes in existing refineries
difficult [2]. For example, unconventional crude oils are highly viscous
(up to 106 cP), dense (~1050 kg m−3), highly acidic (total acid number
(TAN) N 0.5 mg KOH/g-oil), and enriched with metallic (e.g., Ni, Ca, Fe,
Al, V) and non-metallic (e.g., Si, N, S, clays, minerals) impurities, and
they contain high amounts of polar fractions (e.g., asphaltenes and
resins) [3].
High acid crudes (HACs), which are much cheaper than convention-
al crude oil and are consumed in quantities of N9 million barrels/day
nowadays [4], contain a substantial amount of naphthenic acids (NAs).
⁎ Corresponding author at: School of Mechanical Engineering and SKKU Advanced
Institute of Nano Technology (SAINT), Sungkyunkwan University, 2066, Seobu-Ro,
Jangan-Gu, Suwon, Gyeong Gi-Do 16419, Republic of Korea.
E-mail address: jaehoonkim@skku.edu (J. Kim).
clic carboxylic acids. The inherent acidity of HACs engenders severe en-
vironmental and economic problems. Direct processing of HACs in
existing refineries causes serious adverse effects on various units such
as the desalter, distillation unit, pipelines, and catalytic reactors. These
effects include erosion, corrosion, and catalyst deactivation, which
cause serious operational problems and lead to reduced refinery mar-
gins [5]. Moreover, NAs, which are amphiphilic in nature, facilitate the
formation of recalcitrant petroleum emulsions during recovery of the
crude oil from unconventional oil fields using steam as well as during
the desalting operation [6]. For example, the water used for the process-
ing of oil sand is highly toxic because of exposure to the hazardous NAs,
and this leads to severe environmental, aquatic, and public health con-
cerns [7,8]. Therefore, from the economic and environmental perspec-
tives, it is highly desirable to develop an effective approach for the
removal of NAs from HACs.
There are two main practical approaches for reducing the TAN of
HACs. One is the blending of HACs with conventional crude oils with
low acidity to diminish the corrosive and toxic impact of NAs. However,
the content of HACs in the blended oil is typically very low (around 2–
3%) and the NAs are not completely removed from the crude oil mix-
ture. Therefore, some low-boiling-point NAs may be present in the
light fraction emitted from the distillation unit, and high-boiling-point
NAs may accumulate in the heavy fractions or residues. Caustic washing
is another practical treatment method for neutralizing acidic HACs, but
the production of a huge amount of wastewater is a major limitation. In

http://dx.doi.org/10.1016/j.fuproc.2017.05.015
0378-3820/© 2017 Elsevier B.V. All rights reserved.
124 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130
addition to the above practical methods, various approaches have been
proposed to reduce the TAN; for example, treatment using ionic liquids
[9], catalytic esterification [10,11], catalytic decarboxylation [12–14],
treatment with ammonia solution [15], microwave irradiation [16],
physical adsorption [17], and solvent extraction [18]. Nevertheless,
these processes have certain drawbacks that should be addressed care-
fully prior to practical implementation. For example, although the TAN
can be reduced by the catalytic, ionic liquid, or ammonia route, the
use of expensive chemicals, heterogeneous catalysts, and hydrogen
makes these methods economically unfavorable. Catalytic routes have
been explored for direct reduction of the TAN of HACs using alkali and
alkaline earth metal oxides [12,19] and hydrotalcite [11], but a high-
temperature operation (N300 °C) was necessary for effective TAN re-
duction, and loss of catalyst activity caused by coking was observed. In
addition, metallic and non-metallic impurities present in resins and
asphaltenes tend to poison catalysts, thereby reducing the TAN reduc-
tion activity over time. Solvent extraction and ionic liquids require a
large amount of solvent and suffer from solvent recovery problems.
Thermal decomposition and hydrogenation are very energy-intensive
because of the high-temperature operation (400–500 °C). Therefore, it
is imperative to develop a simple, effective, environmentally-friendly,
and economically-viable process for reducing the TAN of HACs for ulti-
mate utilization of low-cost HACs in existing petroleum facilities.
In this study, we develop a new approach for effective deacidifica-
tion of HACs using methanol in a relatively moderate-temperature
and moderate-pressure system without using molecular hydrogen and
external catalysts. Methanol under high pressure and high temperature
conditions, i.e., supercritical methanol (scMeOH; Tc = 239 °C, Pc =
8.1 MPa), was previously evaluated for reducing the TAN of “neat” NA
mixtures [20]. The major mechanism of TAN reduction using scMeOH
is esterification of the carboxylic acid group in the NAs to generate the
corresponding methyl esters. Although effective TAN reduction was
achieved with scMeOH, the high-temperature (350–400 °C) and high-
pressure (N 10 MPa) operation would be economically unfavorable on
the industrial scale [20]. The low reaction temperature of 300 °C result-
ed in a low TAN reduction efficiency of b 60% even after 3 h [21]. In con-
trast, direct reduction of the TAN of “HACs” using methanol under
subcritical conditions has not previously been explored in detail. Previ-
ously, we showed that scMeOH at 400 °C and 30 MPa could effectively
reduce the TAN of HACs, but a considerable loss of crude oil resulted
during the TAN reduction by the formation of coke (15–20 wt%) [21];
this is undesirable in terms of processing productivity and economics.
Therefore, it is highly desirable to develop a technique that could
achieve high degree of TAN reduction of HACs, and the same time, sup-
press coke formation under reduced temperatures and reduced pres-
sures. In this study, we achieved high amount of liquid crude oil
recovery (95–98 wt%) with negligible coke formation and gasification
(≤ 0.3 wt%) along with high degree of TAN reduction. The other novel
aspects of current study is the non-catalytic moderate-temperature
and moderate-pressure TAN reduction approach could allow us to use
HACs up to 50% with conventional crude oil, which has never been re-
ported in literature. Previously, much higher TAN reduction efficiency
has never been achieved at such a moderate temperature and pressure
(≤250 °C, ≤6.5 MPa) condition without using catalysts [10,11,19]. Final-
ly, we present quantitative TAN reduction mechanisms in methanol in
terms of decarboxylation and esterification with complete decarboxyl-
ation calculations based on CO2 production from reaction mixtures.
Herein, utilization of methanol at relatively moderate temperatures
(≤ 250 °C) and moderate pressures (≤ 6.5 MPa) is explored for effective
reduction of the TAN of HACs. A TAN below 0.08 mg KOH/g-oil and a
TAN reduction efficiency of 96.9% could be achieved under conditions
of 250 °C, 6.4 MPa, and 90 min reaction time. The degree of TAN reduc-
tion observed in this study is quite comparable to those of catalytic TAN
reduction, as listed in Table S1. The TAN of the deacidified oil is adequate
for current refineries without modification. The effect of the process pa-
rameters (temperature, reaction time, and concentration) on reduction
of the TAN is discussed in depth. The liquid and gaseous products are an-
alyzed in detail using gas chromatography-time of flight/mass spec-
trometry (GC-TOF/MS) and refinery gas analyzer-gas chromatography
(RGA-GC) to gain insight into the mechanism of TAN reduction. The
content of saturates, aromatics, resins, and asphaltene (SARA) in the
raw crude mixture and deacidified oil are analyzed to understand the
structural transformation of the heavy fractions of HAC during deacidi-
fication. The products from the moderate-temperature methanol reac-
tion are compared with those from pyrolysis and supercritical
methanol treatment for comparison of the TAN reduction efficiency
and to understand the main deacidification mechanisms.
2. Experimental
2.1. Materials
Unconventional HAC (Laguna, Venezuela) and conventional crude
(Kuwait) were provided by SK Innovation (South Korea). HPLC grade
dichloromethane (DCM), methanol, and 2-propanol were purchased
from Honeywell Burdick & Jackson® (USA). For Karl Fischer titration,
0.1 N KOH in 2-propanol, Hydranal® composite 5.0, and toluene
(99.5%) were purchased from Sigma-Aldrich (USA). Distilled and deion-
ized water was prepared by using an AQUAMax™-Basic 360 water pu-
rification system (Younglin Instrument Co., Ltd., South Korea). High
purity He (99.99%), H2 (99.99%), and N2 (99.99%) were obtained from
JC Gas Company (South Korea).
2.2. Reaction in methanol
A mixture of HAC from Laguna (TAN = 5.05 mg KOH/g-oil) and con-
ventional crude oil (Kuwait, TAN = 0.09 mg KOH/g-oil) was prepared
by mixing equal amounts of the two crudes on a weight basis. The
TAN of the oil mixture was 2.57 mg KOH/g-oil. A schematic of the reac-
tion and separation protocols is shown in Fig. S1. The reaction between
the crude oil mixture and methanol was conducted in a custom-built
SUS 316 batch reactor with an inner volume of 140 mL. The reactor
was equipped with a magnetically-driven stirrer and all reactions
were carried out at a stirring rate of 700 rpm. The detailed reactor as-
sembly and experimental protocol are explained elsewhere [21]. After
completion of the reaction, a wet gas meter (W-NK-2 type, Shinagawa
Corporation, Japan) was used to measure the volume of gas produced
and the gaseous products were stored in 1 L Tedlar® sampling bag for
compositional analysis. After the reaction, the reaction mixture was col-
lected by rinsing with DCM. The solid residue, if formed during the de-
acidification, was separated from the liquid oil product by filtration of
the mixture through a Whatman filter paper. The deacidified crude oil
was collected by vacuum evaporation of DCM and methanol at 40 °C
and −0.08 MPa for 30 min.
The formulas used to calculate the product yields are as follows:
Weight of oil
Oil yield ðwt%Þ ¼  100% ð1Þ
Weight of crude oil
Weight of solid residue
Solid residue yield ðwt%Þ ¼  100% ð2Þ
Weight of crude oil
Weight of gas produced
Gas yield ðwt%Þ ¼  100% ð3Þ
Weight of crude oil
2.3. Product analysis
The gases collected after the reaction were analyzed using GC to
quantify all of the gaseous products, which included H2, CO2, CO, CH4,
C2H6, C2H4, and C3–C6 + compounds. A Clarus 580 GC-Model Arnel
1115PPC Refinery Gas Analyzer (RGA) (PerkinElmer, USA) was used
 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130
for GC analysis. The detailed specifications of the RGA-GC are provided
in a previous paper [22]. The TANs of the crude oils before and after
the reaction were measured using the ASTM D664-11a potentiometric
method with a Metrohm 877 Titrino Plus (Metrohm Ltd., Switzerland)
instrument. A mixture of toluene (50%), 2-propanol (49.5%), and deion-
ized water (0.5%) was used as the solvent for the TAN measurement.
The TAN and TAN reduction was calculated using the following equa-
tions:
 
mg KOH
TAN
g−oil
ðNormality of KOH  CF  ðVol:of KOH used–Vol:of blank titration Þ
¼
Weight of crude oil in grams
Initial TAN −Final TAN
TAN reduction ¼  100
Initial TAN
where the normality of KOH was 0.1 N; CF = conversion factor =
56.1 mg mmol−1. The error range of the TAN values was within ±5%.
The deacidified crude oil was analyzed using an Agilent Technologies
7890A GC instrument with a PegasusHT TOF/MS detector (LECO Corpo-
ration, USA). The GC was equipped with a 7860 series auto-injector
(Agilent Technologies) and an Rxi®-5 ms non-polar column (30 m ×
0.25 mm × 0.25 μm, RESTEK, USA). A DCM solution of the crude oil
(2.0 μL) was injected into the GC in a 1:50 split mode (injector temper-
ature: 280 °C). The temperature of the column was maintained at 40 °C
for 2 min, increased to 300 °C at a rate of 3 °C min−1, and held at this
temperature for 5 min. The chemical compounds were identified
using the National Institute of Standards and Technology (NIST) MS
Search 2.0. For the TOF/MS analysis, the temperature of the transfer
line was fixed at 250 °C and the temperature of the ion source was
fixed at 230 °C with a molecular weight range of 35–800 Da. A model
IATROSCAN MK-6S thin film chromatography instrument equipped
with a flame ionized detector (FID) (LSI Medience Corporation, Japan)
was used to assay the saturates, aromatics, resins, and asphaltene
(SARA) fractions of the HAC and the deacidified oil (column tempera-
ture: 25 °C; FID temperature: 600–700 °C). The reproducibility and reli-
ability of the equipment and separation protocol were verified from five
replicate analyses.
3. Results and discussion
3.1. Effect of reaction temperature
To investigate the effect of the reaction temperature on reduction of
the TAN, the crude oil mixture was treated with methanol at tempera-
tures in the range of 150–300 °C for a fixed reaction time of 90 min
and a fixed crude concentration of 33.3 wt%. At the end of the reaction
period, the pressure of the reactor was in the range of 5.2–6.4 MPa.
The reaction pressure at different temperatures was controlled by
charging the calculated amount of methanol using Peng-Robinson
equation of state. In the low temperature ranges of 150–200 °C, 90–
60% of methanol was calculated to be in the liquid phase, while in the
moderate temperature ranges of 250–300 °C, methanol was in the su-
perheated state. As shown in Fig. 1a, the TAN declined significantly
from 2.42 to 0.08 mg-KOH/g-oil with an increase of the reaction tem-
perature from 150 to 250 °C. The degree of TAN reduction at 250 °C
was as high as 96.9%, and the TAN value at 250 °C was significantly
lower than that required for crude distillation units (0.50 mg-KOH/g-
oil) [4]. This indicates that methanol at 250 °C is very effective for
converting the NAs in the crude oil mixture to non-acidic species.
As reported previously, the dominant reaction pathway for the de-
acidification of “neat” NA in scMeOH (350–400 °C, 10–30 MPa) is ester-
ification of the NAs with methanol [20,21]. To investigate the possibility
of forming methyl esters by reaction of the NAs in the HAC mixture with
125
methanol under moderate-temperature and moderate-pressure condi-
tions (250 °C, 6.4 MPa), the deacidified oil produced by reaction at 250
°C for 90 min was analyzed using GC-TOF/MS. Various types of methyl
esters were present in the deacidified oil (Fig. 2; Table 1). Thus, moder-
ate-temperature and moderate-pressure methanol was very effective
for inducing esterification of the NAs contained in the HAC mixture. In-
creasing the temperature decreases the intermolecular attractive forces
between the methanol molecules, which allows methanol to interact
more effectively with the carboxylic acids in the NAs. The degree of hy-
drogen bonding between the methanol molecules is reduced by an
order of magnitude (from 0.7 to 0.08) with an increase of the tempera-
ture from 150 to 250 °C at 6 MPa [23]. The weaker hydrogen bonding in-
creases the tendency of methanol to produce protons, which ultimately
ð4Þ increases the esterification activity. In addition, other thermodynamic
parameters can facilitate reduction of the TAN; for example, at 150 °C
and 6.0 MPa, the density and internal energy of methanol are 0.66 g
ð5Þ
cm−3 and 8.7 kJ mol−1, respectively. As the temperature increases to
250 °C at 6.0 MPa, the vapor density of methanol decreases by an
order of magnitude to 0.06 g cm−3, while the internal energy increases
almost five-fold to 37.6 kJ mol−1 [24]. This allows the methanol mole-
cules to easily penetrate into the crude oil phase, and enhances the col-
lision strength of the molecules, which can result in high esterification
activity.
Further increasing the temperature to 300 °C (Fig. 1a) and 350 °C
(Fig. S2, Case 1) resulted in an increase of the TAN of the deacidified
oils to 0.20 and 0.45 mg-KOH/g-oil, respectively. This may be because
for an identical methanol-to-crude mixture ratio, at higher temperature,
more methanol is vaporized and less methanol is available in the liquid
phase to trigger the esterification. In addition, more methanol can be
self-decomposed at higher temperature, which reduces the esterifica-
tion ability of methanol.
Another plausible mechanism for reduction of the TAN of the HAC
mixture is thermal decarboxylation of the NAs. Fig. 1b shows the com-
position of the gases produced at various temperatures. As the temper-
ature increased from 150 to 250 °C, the amount of CO2 produced
increased from 1 × 10−3 mmol to 5 × 10−3 mmol. The amount of CO2
produced at 300 °C remained similar to that produced at 250 °C. There
are several possible reasons for the increase in the generation of CO2
with temperature. First, thermal decarboxylation of the carboxylic acid
group in the NAs produces CO2 via cleavage of the aryl-CO2 bond [25,
26]. Another possibility for the production of CO2 is self-decomposition
of methanol and decomposition of other oxygenated compounds in the
crude oil, apart from the NAs. To verify the possibility of CO2 generation
via methanol self-decomposition, pure methanol was heated to 250 °C
at 5.5 MPa for 90 min and the gas phase was analyzed, as shown in
Fig. 1b (Blank MeOH). Even though the gas yield was very low
(0.02 wt%), the CO2 content in the gaseous products constituted almost
half of the generated gas (46.3 mol%). This suggests that during the HAC
reaction in methanol, self-decomposition of methanol may be one of the
main routes for the production of CO2. The self-decomposition of meth-
anol can also contribute to increasing the H2 content with increasing
temperature.
The yields of deacidified crude oil, solid residue, and gases after
treatment of the crude oil mixture with methanol at different tempera-
tures are shown in Fig. 1c. N95 wt% of the oil could be recovered even at
the reaction temperature of 250 °C. The yields of solid reside and gases
were very low; for example, the solid residue yield was b 0.2 wt% and
the gas yield was b0.08 wt%. This indicates that the moderate-tempera-
ture and moderate-pressure operation effectively suppresses coke for-
mation as well as gasification, which could result in a deacidified oil
yield of up to 99 wt% (theoretical oil + gas + solid residue yields =
100 wt%). The loss of some oil fractions during handling and separation
lowered the experimental oil yield to 95 wt%. Further, increasing the
temperature to 300 and 350 °C (Fig. S2, Case 1) resulted in an increase
in the solid residue yield to 1.1 and 6.3 wt%, respectively. Thus, at elevat-
ed temperature, the recoverable deacidified oil decreased.
126 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130

Fig. 1. Effect of reaction temperature on deacidification of the crude mixture at 5.2–6.4 MPa, using 33.3 wt% crude for 90 min. (a) TAN values and the degree of TAN reduction, (b) gas
composition, (c) product yields, and (d) SARA analysis.

In addition to reducing the TAN, the HAC reaction with methanol re-
duced the amount of high-molecular-weight resin and asphaltene to
some extent (Fig. 1d). Increasing the reaction temperature from 150
to 250 °C caused the resin + asphaltene content to decrease from
43.3% to 33.3 area%, while the aromatic content increased from 38.0 to
50.5 area%, indicating some degree of cracking during the deacidifica-
tion process.
As discussed in the Introduction, effective reduction of the TAN of
HACs via removal of the “neat” NAs using scMeOH was reported [20,
21]. To investigate the feasibility of using scMeOH for deacidification
of HACs, three different types of HACs, i.e., Laguna (TAN = 5.05 mg-
KOH/g oil), Bachaquero-13 (Venezuela, TAN = 4.2 mg-KOH/g oil), and
Rubiales (Colombia, TAN = 0.55 mg-KOH/g oil) were treated in
scMeOH at 400 °C and 30 MPa for 1 h (Fig. S2; Cases 2–4). Although
effective conversion of the NAs in the HACs was observed with a high
TAN reduction efficiency of 92.0–94.0%, a substantial amount of solid
residue was formed (15–20 wt%). This clearly indicates that moder-
ate-temperature and moderate-pressure methanol is very effective in
retarding coke formation while reducing the TAN of the HAC mixture.
3.2. Effect of reaction time
The effect of the reaction time on reduction of the TAN at a fixed
temperature of 250 °C and a fixed concentration of 33.3 wt% was ana-
lyzed (Fig. 3a). As the reaction time increased from 30 to 90 min, the
acidity decreased from 1.38 to 0.08 mg-KOH/g-oil, indicating that an ex-
tended reaction time is required for effective deacidification of the HAC
mixture. The product yields (Fig. 3b) show that reaction times of 30–

Fig. 2. GC-TOF/MS chromatogram of the deacidified oil. Reaction conditions: 250 °C, 6.4 MPa, 33.3 wt% crude, 90 min.
 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130 127

Table 1 × 10−3 mmol to 5.1 × 10−3 mmol) with increasing reaction time (30
Classification of esters obtained after deacidification of crude oil mixture in methanol at to 90 min) suggests an increase in the decarboxylation of the NAs
250 °C and 6.4 MPa using 33.3 wt% crude; reaction time: 90 min; technique: GC-TOF/MS.
with extended reaction time. To examine the contribution of decarbox-
Peak Name of compound Retention time Area ylation to removal of the NAs under the assumption that the CO2 pro-
# (min:sec) % duced originates from decarboxylation, a series of calculations was
Methyl ester carried out as listed in Table S2, and the results are shown in Fig. 3d.
1,2 3,6-Octadecadiynoic acid, methyl ester 20:46.6 0.11 The calculations were based on the amount of carboxylic acid groups
3 Octadecanoic acid, 4-methoxy-, methyl ester 22:58.5 0.11
(which was measured by titration) and the amount of CO2 produced
4 1H–Indene-4-carboxylic acid, 36:17.7 0.08
2,3-dihydro-1,1-dimethyl-, methyl ester during reaction of the crude mixture with methanol at 250 °C (as mea-
5 9-Octadecen-12-ynoic acid, methyl ester 37:24.9 0.16 sured by RGA-GC). Note that the CO2 produced from the self-decompo-
6 6-Methyl-11-oxo-undeca-5,9-dienoic acid, 37:39.3 0.12 sition of methanol, although a very small volume of gas was produced at
methyl ester 250 °C (0.02–0.06 mg CO2/g methanol depending on reaction time), is
7 9,15-Octadecadienoic acid, methyl ester 38:33.7 0.15
8 5,8,11,14-Eicosatetraynoic acid, methyl ester 39:21.7 0.21
not included in the calculation. As the reaction time increased from 30
9 Cyclopropanetetradecanoic acid, 2-octyl-, methyl 51:42.2 0.13 to 90 min, the contribution of decarboxylation to reduction of the TAN
ester increased slightly from 6.3% to 8.6%, indicating the minor contribution
Sub-total 1.07 of thermal decarboxylation to reduction of the TAN. Therefore, other
TAN reduction mechanisms, including esterification, constitute the
major reaction pathways for reduction of the TAN at the moderate reac-
90 min did not cause significant coke formation and gasification; the gas tion temperature of 250 °C. Enhanced cracking at long reaction times is
and the solid residue yields were extremely low (0.003–0.08 wt% and responsible for the increase (from 59.6 to 94.2 mol%) in the C1 − C6 frac-
0.1–0.2 wt%, respectively). tion in the gas product.
Hydrogen was not observed for reaction times of b 60 min, while at
90 min, 2.96 mol% of H2 was produced (Fig. 3c). In the early stage of 3.3. Effect of crude concentration
the reaction, H2 produced from methanol might be completely con-
sumed for cleavage of the thermally labile bond in the resins and To investigate the effect of the crude oil concentration on reduction
asphaltenes. The residual H2 present after 90 min implies that hydrogen of the TAN, the concentration of the crude oil mixture in methanol was
is continuously produced from the crude oil itself and by self-decompo- varied from 20.0 to 50.0 wt% under the following conditions: 250 °C,
sition of methanol. The increased generation of CO2 (from 1.8 5.9–6.4 MPa, and 90 min; the results are shown in Fig. 4. The results

Fig. 3. Effect of reaction time on deacidification of the crude mixture at 250 °C and 5.7–6.4 MPa using 33.3 wt% crude; (a) TAN values and the degree of TAN reduction, (b) product yields, (c)
gas composition, and (d) contribution of thermal decomposition to NA removal.
128 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130

obtained from slow pyrolysis of the HAC mixture (in the absence of
methanol), which was performed at 250 °C for 90 min under a N2 pres-
sure of 1 MPa, are also presented in the figure for comparison. During
pyrolysis, the TAN of the HAC mixture declined slightly from 2.57 to
1.58 mg-KOH/g-oil with only 38.6% reduction efficiency (Fig. 4a). The
main mechanism of TAN reduction under the pyrolysis conditions was
proposed to be thermal decarboxylation of the NAs via cleavage of the
aryl-CO2 bond, as discussed in the previous section. The CO2 detected
in the gas phase during pyrolysis (Fig. 4c) may be generated from ther-
mal decarboxylation of the NAs. Catalytic-based reduction of the TAN of
crude oil has been proposed in order to overcome the low TAN reduc-
tion efficiency associated with pyrolysis; for example, increasing the
loading of calcined MgO from 7 to 40 wt% in the presence of dodecane
increased the TAN reduction efficiency from 45 to 98.0% at 250 °C [27].
Similarly, the TAN of HAC was reduced from 2.7 (feedstock) to 0.5
(deacidified oil) mg KOH/g-oil (with 81.5% reduction efficiency) using
Mg\\Al hydrotalcite/γ-Al2O3 catalyst-to-oil ratios of 0.17–0.20 at 320
°C [28]. Therefore, thermal decarboxylation alone is not sufficient for ef-
fective reduction of the TAN of the HAC mixture at the moderate tem-
perature of 250 °C. In contrast, significant reduction of the TAN was
achieved in the presence of methanol (Fig. 4a). When the crude oil con-
centration was in the range of 20.0–33.3 wt%, the acidity decreased to
0.18–0.08 mg-KOH/g-oil with 93.0–96.9% reduction efficiency. For a
TAN reduction of 20.0–33.3 wt%, thermal decarboxylation accounts for
only 8.6–13.3%. (Table S3; Fig. 4b). Further increasing the crude mixture
concentration to 50.0 wt% resulted in an increase in the contribution of
decarbonylation (to 27.2%) to reduction of the TAN. In the case of slow
pyrolysis, the contribution of decarboxylation increased to 68.1%. At
this stage, it is not clear what accounts for the remaining 31.9% TAN re-
duction, but other conversion mechanisms, including decarbonylation
and hydrodeoxygenation, may be operative. Thus, the presence of
methanol changes the dominant reaction pathway from thermal decar-
boxylation to esterification. Consequently, increasing the concentration
of the HAC mixture to 50 wt% resulted in a lower TAN reduction efficien-
cy than that of the reaction with 33.3 wt% HAC (Fig. 4a), indicating less
efficient TAN reduction under methanol-deficient conditions.
Fig. 4c shows the composition of the gases produced using different
concentrations of the crude oil mixture. The amount of H2 produced in-
creased with increasing crude oil concentration, indicating that pyroly-
sis of the crude oil mixture was the main H2 production pathway.
Increasing the methanol content from 50 to 80 wt% caused the amount
of CH4 produced to increase from 0.05 to 0.09 mmol. Because the self-
decomposition of methanol at 250 °C did not produce CH4, cracking of
the hydrocarbons in the crude oil mixture with the aid of methanol
may responsible for CH4 generation.
Fig. 4d shows the product yield as a function of the crude mixture
concentration. The recovery of deacidified crude oil was high in the con-
centration range of 95.0–98.5 wt%. In the case of pyrolysis, the oil recov-
ery was higher (98.5 wt%). Considering the extremely low solid residue
yields (0.1–0.2 wt%) and low gas yields (0.07–0.12 wt%) obtained in the
presence of methanol, the lower liquid yield compared to that achieved
with pyrolysis is due to loss during product separation.

Fig. 4. Effect of crude oil concentration on deacidification of the crude mixture at 250 °C and 5.9–6.4 MPa (initial N2 pressure in case of pyrolysis was 1 MPa); reaction time: 90 min: (a) TAN
values and the degree of TAN reduction, (b) contribution of thermal decomposition to NA removal, (c) gas composition, and (d) product yields.
 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130
3.4. Deacidification of neat HAC
As discussed in the Introduction, only a small fraction of the HAC (2–
3%) is blended with conventional crude oil for use in existing refinery
units. Therefore, reduction of the TAN of the crude oil mixture using
methanol in the absence of an external catalyst and hydrogen, and
blending at a 1:1 ratio of HAC-to-conventional crude oil is very promis-
ing for expanding the utilization of HACs. To examine the feasibility of
reducing the TAN of the “neat HAC” (without blending with convention-
al crude oil), Laguna crude was treated in methanol under the following
conditions: 250 °C, 6.0–6.4 MPa, 33.3 wt%, 90 min (Fig. 5). The TAN of
Laguna crude was reduced from 5.05 to 0.59 mg-KOH/g-oil with a
high reduction efficiency of 88.3%. The yields of deacidified oil from
the pure HAC and the crude mixture were 97.5 and 95.1 wt%, respec-
tively. As in the case of the crude oil mixture, the yields of solid residue
(0.11 wt%) and gas (0.05 wt%) were negligible when the HAC was used.
The composition of the gases produced in both the cases was very sim-
ilar (Fig. S3a). The decarboxylation tendency for both feedstocks was
also estimated on the basis of the CO2 produced during the course of
the reaction with methanol at 250 °C (Table S4; Fig. 5b). The extents
of TAN reduction via decarboxylation were 2.7% and 8.6% for the Laguna
Fig. 5. Comparison of deacidification of Laguna crude and the crude oil mixture (Laguna:
conventional oil = 1:1) in methanol at 250 °C, 6.0–6.4 MPa, using 33.3 wt% crude,
reaction time: 90 min. (a) TAN values and the degree of TAN reduction, (b) contribution
of thermal decomposition to NA removal.
129
crude and the crude oil mixture, respectively. During reduction of the
TAN of Laguna crude in moderate-temperature methanol, a slight in-
crease in the saturates + aromatics (from 53.6 to 59.9 area%) fraction
was observed (Fig. S3b), as in the case of the crude oil mixture. There-
fore, moderate -temperature and moderate -pressure methanol is very
promising for the effective removal of NAs in HACs with a high yield
of deacidified oil due to the retardation of byproduct formation.
4. Conclusions
A technique employing methanol at moderate-temperature and
moderate-pressure was developed for the effective removal of naph-
thenic acids from high acid crudes. Significant reduction of the total
acid number of the crude mixture from 2.57 to 0.08 mg-KOH/g-oil
was achieved under conditions of 250 °C, 6.4 MPa, 33.3 wt%, and
90 min, with a TAN reduction efficiency of 96.9%, making the process
suitable for existing refineries. The major mechanism of deacidification
in the presence of methanol was esterification of the carboxylic acid
group with methanol. During the moderate-temperature TAN reduction
process, the formation of coke and gases was effectively suppressed,
resulting in high yields of deacidified oils (N 95 wt%). A slight reduction
of the resins and asphaltenes was observed, suggesting that moderate-
temperature methanol is effective for cracking the high-molecular-
weight species to some degree. The dominant reaction pathway for re-
duction of the TAN via pyrolysis of the crude mixture is decarboxylation,
whereas esterification is considered to be the major deacidification
mechanism when methanol is used at 250 °C. The deacidification of
HACs with methanol at moderate temperatures and pressures is a pro-
spectively economically-viable option for extensive utilization of HACs
in existing petroleum refineries.
Acknowledgements
The authors acknowledge financial support from the Energy Effi-
ciency & Resources Core Technology Program of the Korea Institute of
Energy Technology Evaluation and Planning (KETEP) through a grant
from the financial resources of the Ministry of Trade, Industry & Energy,
Republic of Korea (No. 20152010103120). Additional supports from the
New & Renewable Energy Core Technology Program of the Korea Insti-
tute of Energy Technology Evaluation and Planning (KETEP) through a
grant from the financial resources of the Ministry of Trade, Industry &
Energy, Republic of Korea (No. 20143030090940) and a National Re-
search Foundation of Korea (NRF) grant funded by the Korean govern-
ment (MSIP) (grant number: NRF-2016R1A2B3008800) are also
appreciated.
Appendix A. Supplementary data
Supplementary material.
References
[1] N.A. Owen, O.R. Inderwildi, D.A. King, The status of conventional world oil
reserves—hype or cause for concern? Energy Policy 38 (2010) 4743–4749.
[2] R. Santos, W. Loh, A. Bannwart, O. Trevisan, An overview of heavy oil properties and
its recovery and transportation methods, Braz. J. Chem. Eng. 31 (2014) 571–590.
[3] L. He, F. Lin, X. Li, H. Sui, Z. Xu, Interfacial sciences in unconventional petroleum pro-
duction: from fundamentals to applications, Chem. Soc. Rev. 44 (2015) 5446–5494.
[4] J.G. Speight, High Acid Crudes, Gulf Professional Publishing, United States of Ameri-
ca, 2014.
[5] X. Wu, H. Jing, Y. Zheng, Z. Yao, W. Ke, Erosion–corrosion of various oil-refining ma-
terials in naphthenic acid, Wear 256 (2004) 133–144.
[6] R. Varadaraj, C. Brons, Molecular origins of heavy crude oil interfacial activity part 2:
fundamental interfacial properties of model naphthenic acids and naphthenic acids
separated from heavy crude oils, Energy Fuel 21 (2007) 199–204.
[7] J.P. Swigert, C. Lee, D.C. Wong, R. White, A.G. Scarlett, C.E. West, S.J. Rowland, Aquatic
hazard assessment of a commercial sample of naphthenic acids, Chemosphere 124
(2015) 1–9.
[8] L.D. Brown, A.C. Ulrich, Oil sands naphthenic acids: a review of properties, measure-
ment, and treatment, Chemosphere 127 (2015) 276–290.
130 M.K. Khan et al. / Fuel Processing Technology 165 (2017) 123–130
[9] R. Martínez-Palou, R. Luque, Applications of ionic liquids in the removal of contam-
inants from refinery feedstocks: an industrial perspective, Energy Environ. Sci. 7
(2014) 2414–2447.
[10] Y.-z. Wang, J.-y. Li, X.-y. Sun, H.-l. Duan, C.-m. Song, M.-m. Zhang, Y.-p. Liu, Removal
of naphthenic acids from crude oils by fixed-bed catalytic esterification, Fuel 116
(2014) 723–728.
[11] X. Li, J. Zhu, Q. Liu, B. Wu, The removal of naphthenic acids from dewaxed VGO via
esterification catalyzed by mg–al hydrotalcite, Fuel Process. Technol. 111 (2013)
68–77.
[12] A. Zhang, Q. Ma, K. Wang, X. Liu, P. Shuler, Y. Tang, Naphthenic acid removal from
crude oil through catalytic decarboxylation on magnesium oxide, Appl. Catal. A
Gen. 303 (2006) 103–109.
[13] X. Fu, Z. Dai, S. Tian, J. Long, S. Hou, X. Wang, Catalytic decarboxylation of petroleum
acids from high acid crude oils over solid acid catalysts, Energy Fuel 22 (2008)
1923–1929.
[14] D.J. Darensbourg, M.W. Holtcamp, E.M. Longridge, B. Khandelwal, K.K. Klausmeyer,
J.H. Reibenspies, Role of the Metal Center in the Homogeneous Catalytic Decarbox-
ylation of Select Carboxylic Acids. Copper (I) and Zinc (II) Derivatives of
Cyanoacetate, J. Am. Oil. Chem. Soc. 117 (1995) 318–328.
[15] Y. Wang, Z. Chu, B. Qiu, C. Liu, Y. Zhang, Removal of naphthenic acids from a vacuum
fraction oil with an ammonia solution of ethylene glycol, Fuel 85 (2006) 2489–2493.
[16] M. Huang, S. Zhao, P. Li, D. Huisingh, Removal of naphthenic acid by microwave, J.
Clean. Prod. 14 (2006) 736–739.
[17] J.P. Silva, A.L.H. Costa, S.S.X. Chiaro, B.E.P.C. Delgado, M.A.G. de Figueiredo, L.F. Senna,
Carboxylic acid removal from model petroleum fractions by a commercial clay ad-
sorbent, Fuel Process. Technol. 112 (2013) 57–63.
[18] S.N. Shah, L.K. Chellappan, G. Gonfa, M.I.A. Mutalib, R.B.M. Pilus, M.A. Bustam, Ex-
traction of naphthenic acid from highly acidic oil using phenolate based ionic liq-
uids, Chem. Eng. J. 284 (2016) 487–493.
View publication stats
[19] L. Ding, P. Rahimi, R. Hawkins, S. Bhatt, Y. Shi, Naphthenic acid removal from heavy
oils on alkaline earth-metal oxides and ZnO catalysts, Appl. Catal. A Gen. 371 (2009)
121–130.
[20] P.C. Mandal, M. Sasaki, M. Goto, Non-catalytic reduction of total acid number (TAN)
of naphthenic acids (NAs) using supercritical methanol, Fuel Process. Technol. 106
(2013) 641–644.
[21] M.K. Khan, R. Insyani, J. Lee, M. Yi, J.W. Lee, J. Kim, A non-catalytic, supercritical
methanol route for effective deacidification of naphthenic acids, Fuel 182 (2016)
650–659.
[22] R.F. Susanti, L.W. Dianningrum, T. Yum, Y. Kim, B.G. Lee, J. Kim, High-yield hydrogen
production from glucose by supercritical water gasification without added catalyst,
Int. J. Hydrog. Energy 37 (2012) 11677–11690.
[23] M.M. Hoffmann, M.S. Conradi, Are There hydrogen bonds in supercritical methanol
and ethanol? J. Phys. Chem. B 102 (1998) 263–271.
[24] E. Lemmon, M. McLinden, D. Friend, Thermophysical Properties of Fluid Systems,
NIST Chemistry Webbook, NIST Standard Reference Database 69, 2005.
[25] K. Chuchev, J.J. BelBruno, Mechanisms of decarboxylation of ortho-substituted
benzoic acids, J. Mol. Struc: Theochem. 807 (2007) 1–9.
[26] P. Ruelle, Theoretical study of the mechanism of thermal decarboxylation of salicylic
and p-aminobenzoic acids; models for aqueous solution, J.C.S. Perkin 2 (1986)
1953–1959.
[27] H.-Y. Oh, J.-H. Park, Y.-W. Rhee, J.-N. Kim, Decarboxylation of naphthenic acid using
alkaline earth metal oxide, J. Ind. Eng. Chem. 17 (2011) 788–793.
[28] Y.-z. Wang, H.-l. Duan, C.-m. Song, X.-t. Han, X.-r. Ma, Removal of naphthenic acids
from crude oils by catalytic decomposition using mg–al hydrotalcite/γ-al 2 O 3 as a
catalyst, Fuel 134 (2014) 499–504.