PETROTECH 2019

Guidelines & Instructions to Authors for Full Manuscript

Guidelines:

• Papers are strictly limited to 6 pages (A4 size) including figures / tables /
graphics. All papers must be submitted in English language only.
• Use Arial font throughout.
• Lines must be single-spaced (plus additional space between paragraphs)
• Align text to both left and right margins (justify). Do not use columns.
• First page of the paper must indicate the ID number on TOP RIGHT corner
• The Title of the paper should be in bold and in sentence case (Font size=14
point)
• The next item of the paper should be the author's name (the paper presenter),
followed by co-authors. (Font size=12 point)
o Name of the presenting author should be underlined, with whom all
the future correspondence will be made.
• This should be followed by complete official address. (Font size=10 point)
• This should be followed by the e-mail of the Presenting author (Font size=12
point and Italics). Providing e-mail contact is mandatory.
• Main Text:
o All headings 12 point bold and rest of the text 10 points normal
Please Note

• The papers must be submitted online at www.petrotech.in and file size should
not be more than 20 Mb.
• Company logo to be avoided.
• The papers should be in PDF format only. Power Point will not be accepted.
• Tables and figures must be included in the same file as the text. Figures
must be inserted (and NOT copy/paste) into the document in JPEG or Tagged
Image File (TIF) format. All text, figures and tables should be formatted to A4
page size (210x297mm).
• Please see a SAMPLE PAPER enclosed

Contacts

Dr. Anand Gupta

Head Technical Committee
Petrotech-2019
E-Mail: techpaper@petrotech.in
 SAMPLE PAPER
ID: AXXXX

Understanding relationships between NMR derived compositional

parameters and lubricity of diesel fuels
Name of Author 1 , Name of Author 2 , Name of Author 3
Organisation Name and Address of Authors ………………………………………

Presenting Author, E-mail:………………………

Abstract
In this paper, attempts have been made to correlate the composition of diesel samples with their wear
characteristics (lubricity behaviour) measured using High Frequency Reciprocating Rig (HFRR) with the
help of various statistical techniques. Different varieties of around 35 diesel samples (with varying
aromatic and sulphur content) have been taken up for establishing structure-lubricity relationships in this
preliminary study.

Introduction
The transportation fuels such as gasoline, aviation fuel (or Kerosene) and diesel must possess a
modicum of lubricating ability in order to protect pumps and related fuel supply equipment from wear and
other tribological problems. However, transportation fuels are inherently poor lubricants, from the point of
view of both hydrodynamic and boundary lubrication (1). Very recently, it has been suggested that
similar problems may be imminent in gasoline fuel applications (2).

Experimental Details
Samples: 32 diesel samples have been collected from various sources. The samples were quite diverse
in their hydrocarbon composition and sulfur content.

Measurements of Lubricity: A 2 ml test sample of the fuel is placed in the test reservoir of the HFRR
test rig and adjusted to a temperature of 600C. When the temperature has stabilized, a vibrator arm
holding a non rotating steel ball and holding a 200 gm mass is lowered until it contacts a test disk with a 1
mm stroke at a frequency of 50 Hz for 75 minutes.

Statistical Analysis: The statistical analysis and ANN modeling have been performed using the
commercial packages “STATISTICA”.

Results and Discussion

The high-resolution proton nuclear magnetic resonance (1H-NMR) spectroscopy allows the rapid and
quantitative determination of the composition of fuel samples. Besides, NMR spectra of the fuel samples
provide a wealth of structural information, where large number of signals due to the presence of various
structural moieties (aromatics (mono, di- and tri-aromatics), naphthenes, paraffins etc) appear. The
spectra are also rich with other subtle structural features and in fact provide a sort of fingerprints of the
samples. It is even possible to estimate various physical or specification properties of the fuels from such
high content spectra or fingerprints. Recently, Kapur et al. (11) have demonstrated the estimation of
seven specification properties of diesel samples (density, viscosity, cetane number, sulfur content etc.)
from a single 1H-NMR spectrum.

Table 1 gives the hydrocarbon composition of a representative set of diesel samples used in this study
along with sulfur content and wear results. The composition has been determined using 1H-NMR base
method previously developed in our laboratory (10). Figure 1 shows the representative 1H-NMR spectra
of two of the diesel samples used in this study. The spectra also give total aromatic content, sulfur
content and wear scar results. The samples have significant differences in their spectra particularly in the
region of aromatic species. The sample at the top has very low abundance of aromatics as well as sulfur
content compared to the bottom sample. However, the top sample shows much better lubricity
characteristics compared to the bottom sample. Similar observation was made for number of samples
associated with very low sulfur and aromatic contents, still demonstrated better lubricity behavior
compared to many other samples. This shows that reduced sulfur and aromatic content is not directly
related to poor lubricity characteristics. This is graphically shown for a set of sample in figure 2, where
many low sulfur diesel samples have similar or even better lubricity compared to other samples with much
higher sulfur content. Though the above observations are contradictory to the general belief, however, in
agreement with few literature studies. As fuels are more severely refined to remove polar species, the
relative importance of other factors such as viscosity and bulk hydrocarbon composition may be
enhanced.

The effect of bulk hydrocarbon composition and structural parameters (Table 1) on the lubricity behavior
of the present set of diesel samples have been tried to be examined using the Statistical techniques. By
hierarchical cluster analysis of the complete data set, the similarities between the various variables
(compositional parameters and wear) have been calculated. Figure 3 depicts the results in a
dendogram. The tree is printed with a unique ordering in which every branch is lined up so that similar
objects are closets to each other. The compositional parameters related to aromatics (Di, Mono, total
aromatics and CA content) are joined first. The group shows next linkage with the sulfur content. The next
group is formed by saturated cyclic rings (i.e. CN content), which is linked to the wear (HFRR). The last
group is formed by paraffinic parameters (paraffins and CP content), which is also linked to HFRR
values. The saturated compositional parameters have been shown to qualitatively joined first with the
wear values, followed by the sulfur and the aromatic parameters. Linear regression analysis between
wear and the single compositional parameters give regression coefficient (r) between 0.1 to 0.4. This
means that only weak linear relations exist between the single compositional parameter and wear values,
as expected. So, a common factor analysis was used to represent more clearly the complex relationships
between the single parameters amongst each other and regarding to a property (Wear). Common factors
are unobserved variables that are postulated to account for the inter-correlations among observed
variables.

The results obtained with the present approach of quantitatively correlating lubricity of diesel samples with
their structural group concentration as perceived through NMR spectra using ANN models is highly
encouraging. The work is being planned to extend the present study by increasing number of samples
particularly those with low aromatic content.

Conclusions
Based upon the present studies, it is evident that lubricity behavior of diesel samples is a complex
function of their hydrocarbon composition. It has been found that some low sulfur diesels can have
lubricity as high as some high sulfur diesels, suggesting that lubricity is not directly related to the sulfur
content. The ANN model enables prediction of lubricity behavior of diesel samples from their NMR
spectral features, and a very high correlation has been obtained (r=0.979) for a verification data set of 13
samples.

References
1. D.P. Wei and H.A. Spikes, “Fuel Lubricity-Fundamentals and Review”, Fuels International, 1-1, 45-65,
2000.
2. D.P. Wei, H.A. Spikes and S. Korcek, “The Lubricity of Gasoline”, Tribology Transaction, 42, 813-
823, 1999.
3. J.K. Appeldorn and W.G. Dukek, “Lubricity of Jet Fuels”, SAE paper 660712, 1966.
4. D.P. Wei, and H.A. Spikes “The Lubricity of Diesel Fuels”, Wear, 111, 217-235, 1986.

Table 1: Compositional Data for Diesel samples along with HFRR test results.

Sample Aromatics Di-Plus Sulfur Wear

Aromatics (ppm) (µm)

DIESEL-1 4.57 0.1 10 300

DIESEL-2 6.73 0.17 100 640

DIESEL-3 14.5 2.33 195 410

DIESEL-4 15.28 1.0 360 580

DIESEL-5 20.58 1.35 215 520

DIESEL-6 20.67 1.47 270 420

DIESEL-7 23.96 1.55 230 560

DIESEL-8 24.79 1.33 400 410

DIESEL-9 28.34 11.7 13100 420

DIESEL-10 29.76 3.1 580 580

DIESEL-11 32.87 6.72 460 490

DIESEL-12 32.96 6.93 1900 420

13100 1900
1000
900 SULFUR (ppm)
800 WEAR SCAR
700
600
500
400
300
200
100
0
1 2 3 4 5 6 7 8 9 10 11 12
Figure 1: Sulfur and Wear Scar (µm) Data for a Set of Diesel Samples.