Fouling of heat exchanger is an unwanted process of accumulation of dirt or
growth of deposit on the hot surfaces of heat exchanger. These activities can
lead to high pressure drop and reduce flow rates in the system, thus it reducing
the heat transfer efficiency of the heat exchanger. Such fouling leads to enormous
costs, not only in the cost of loss of energy recovery but also in the loss
of product and mitigation measures (Bott, 1995). Mitigation
of fouling can decrease the energy demand in fired heater duty, increase the
energy recovery with higher throughput as the efficiency of the heat exchanger
is optimized and also can lower maintenance cost through a well constructed
Fouling in crude oil is generally believed to be caused by the organic and
inorganic impurities of the crude oil itself (Murphy and
Campbell, 1992). The crude oil itself is an extreme mixture of various materials
and one of the infamous factors governing fouling problem is the crude oils
high asphaltene content. Asphaltene is defined as a component that is insoluble
in non-polar solvents such as pentane, hexane or heptane and soluble in solvents
such as pyridine, carbon disulphide, toluene or benzene (Watkinson,
1992). Other than asphaltene, traces of metal are always present in petroleum
streams as natural compounds and also as corrosion products. For example, certain
metals, carbides, oxides and sulfides of V, Fe and Ni are active catalyst for
crude oil fouling and could exist in the crude oil preheat train (Satterfiled,
1991). Salt of the alkali metal and alkaline earth metal elements can contribute
in crystallization fouling mechanism, especially the inverse solubility salts
in the exchangers prior to the desalter. On the other hand, iron sulphide as
a corrosion product is second only to asphaltene as the most common foulant
in crude pre-heat traine (Wiehe, 1999). Since it is
a black, granulated solid, it is often mistaken for coke.
Besides the inherent properties and contaminants, crude oil itself plays an
important factor as it is blended, or injected with another processing liquid
along the processing line. By blending crude oil, it can cause rapid fouling
as described by Wilson and Polley (2001) in their research
where mixing can create unstable crude solution which can precipitate species
such as asphaltene and result in rapid fouling. Mixing typical paraffinic crudes
or condensate and asphaltenic crudes can cause the asphaltene to precipitate,
giving rise to high fouling factors and this may limit the amount of condensate
that can be mixed with the crude oil (ESDU, 2000). Incompatibility
of the crude oils can be predicted by laboratory measurement using a distinguish
spectroscopic method as demonstrated by Wiehe (1999)
and Wiehe and Kennedy (2000). The measurement was conducted
by determining the two parameters that used to define crude oil and crude blend
solubility; insolubility number, IN and solubility blending number
SBN Wiehe also stressed on the importance of the blending order since
it contributes to the crude oil incompatibility and the procedure was verified
commercially in crude preheat train. Saleh et al.
(2005) studied the effect of mixing and blending crude oils at certain operating
conditions with the intention of using the results to guide a fouling mitigation
Asphaltene in crude oil: Asphaltene is commonly defined as stated in
previous section; however there are many other definitions which have been reported
by researchers around the world. Long (1982) reported
that the classic definition of asphaltene is the fraction of petroleum which
soluble and insoluble in pentane. Nellensteyn (1938) defined
asphaltene as the fraction which insoluble in low boiling point paraffin hydrocarbon,
but soluble in carbon tetrachloride and benzene and according to Pfeiffer
and Saal, (1940) asphaltene is defined as the fraction insoluble in n-heptane
but soluble in toluene. Recently, asphaltene is defined by chemists as the part
precipitated by addition of low-boiling paraffin solvent such as normal-pentane,
and benzene soluble fraction whether it is derived from carbonaceous sources
such as petroleum, coal, or oil shale (Mansoori, 2005).
As reported in many literatures, asphaltene is dark brown to black in color which has no definite melting point and when heated, intumesce then decompose and leave a carbonaceous residue. During the analysis, asphaltene can be precipitated out by the addition of a minimum forty volumes of liquid hydrocarbons. If less is added, resin, which is a fraction of crude oil isolated by adsorption chromatography, may appear. This resin appears within asphaltene fraction by adsorption onto the asphaltene.
Being the least understood component in the crude oil, asphaltene is generally
high in molecular weight, non crystalline and most polar (Wang,
2000). It consists of polyaromatic condensed rings with short aliphatic
chain and heteroatom such as nitrogen, oxygen, sulfur and various known metals.
Elemental compositions of isolated asphaltene using excess volumes of n-pentane
shows that the amount of carbon and hydrogen usually vary over a narrow range,
where 82 ± 3% carbon; 8.1±0.7% hydrogen (Speight
and Moschopedis, 1980). These values correspond to H/C ratios of 1.15±0.05.
This near constancy values is the cause for general belief that unaltered asphaltene
from virgin petroleum have a fixed composition and asphaltene precipitates because
of this composition, not only because of its solubility properties.
A study by Dickakian and Seay (1988) on the effect
of asphaltene on thermal fouling and characterization on deposits formed on
the heated surfaces at various times, showed that the deposits were initially
precipitated asphaltene which were then carbonized on the surface into an infusible
coke. This may be due to the asphaltene have the highest thermal reactivity
of any fraction of a crude oil (Wiehe, 1993). While
soluble, asphaltene react to form lower molecular weight products but when insoluble
the major thermal reaction product is coke.
The crude oils used in this study were collected from a local refinery, sealed in proper containers and stored at the temperature below 5°C to reduce the loss of the light component and to prevent oxidation and degradation from exposure to air and light.
The properties of the crude oils were characterized and listed in Table 1. Both crude A and B are paraffinic crude with the density of ±0.8000 kg-1 and viscosity of ±1.0 cst. Crude A however has higher pour point at +18°C as compare to -6°C for crude B. Crude A also has higher percentage of basic sediment, water, ash and most importantly asphaltene content which is more than 3 times higher than crude B. Crude A is expected to have higher fouling tendencies in the processing system based on these inherent properties.
The fouling deposits also collected from a parallel heat exchanger, Exchanger C and D from the same refinery, treated with toluene to remove oil residue and stored at the same condition in opaque glass containers. This is to prevent the possible oxidation or any reaction with normal metal container.
Asphaltene flocculation study using Automated Flocculation Titrimeter (AFT):
This equipment is used to measure the compatibility or colloidal stability
of asphalt by determining the flocculation onset, point where asphaltene begins
to precipitate from a solution of known weight sample. The sample was prepared
in a solvent when titrated with a non-solvent or titrant. The analysis complies
with ASTM D6703 (ASTM, 2001a).
Crude oil samples, each with different weight were prepared in specially-designed
round bottom reaction vials. Toluene was then added to the vials in equal volume
producing solution with different concentration of asphalt in solvent. The samples
were then kept without any exposure to sunlight for at least 4 hours. before
titration can take place. The reaction vial was then placed in the AFT apparatus;
in Fig. 1 and titration was conducted using iso-octane (2,2,4-trimethyl
pentane). The flocculation points were determined using visible spectroscopy
method at a fixed transmittance at 740nm.
|| Properties of crude oils used in this study
|| Flow diagram of AFT
The onset-point then calculated using the AFTCentral software.
For the present study an indirect method was used to predict crude oil compatibility
which was suggested by Saleh et al. (2005) using
Anderson and Pederson Eq. 1 and Wiehe compatibility model.
The solubility values were adapted from ASTM D 6703, where: δT8.93
for toluene (solvent) and δ10: 6.99 for iso-octane (titrant).
The indirect method was used because the crude oils consists low asphaltene
content and it has significant effect on the end point detection. As suggested,
a list of high asphaltene crude oils was used as reference:
The stability of the oil, a criterion by Wiehe flocculation stability model, δf can be written as in Eq. 2, and similarly the oil is stable if the δ0>δcr. For stability, the insolubility number IN which measures degree of insolubility of the asphaltene present in the oil must be less than the solubility blending number, SBN which measures the solvency of the oil for asphaltene Eq. 3:
Asphaltene precipitation study: The method used for the analysis is
a derivation from ASTM D3279 (ASTM, 2001b), where insolubility
in normal-heptane solvents defines asphaltene which was then determined on a
mass percent basis. The analysis was conducted by adding a ratio of 100 mL n-heptane
per 1g of pre-weighed sample in a flask.
|| Sample distribution of asphaltene precipitation study
The flask was then placed on a magnetic-stirrer hot plate and secured under
reflux condition. A period of 15-20 min for reflux is recommended in accordance
to the method. The cooled but warm sample was then poured through filter study,
GF/D 5.5 cm to filtrate the precipitated asphaltene. Maltenes, the filtrate
which contains the saturate, aromatic and resin in a solution was dried in the
oven for 15 min at 107°C and the residue is weighed.
Modifications to the analysis were conducted for few parameters, i.e. solvent type, temperature and reflux period to understand the effect of those parameters to asphaltene precipitation behavior. Based on a preliminary result, the study was conducted using Design of Experiment (DOE), Taguchi Orthogonal Array to manage sample and data distribution of the study. Table 2 shows the array that was produced using the approach.
RESULTS AND DISCUSSION
Asphaltene flocculation study using AFT: The compatibility study for every crude oil was conducted with three different concentrations through back titration method where 0.3 mL of reference oil was added. The flocculation solubility parameters of the reference oil were initially identified, and flocculation rate of the asphaltene drop out was plotted against concentration.
Reference oils, H1-H8 used in the study have asphaltene content in the range of 1.35 wt% to 3.10 wt%. The oil solubility parameter, δ0 and flocculation solubility parameter, δf of 9.64 MPa0.5 and 8.37 MPa0.5 was respectively obtained for H1 from the titration method. The heavy oil consists of ± 5.6 wt. % of asphaltene with solubility blending number, SBN of 136.84 and insolubility number, IN of 70.91. Other similar values for H2-H8 is listed in Table 3. These values then were used in the afore-mentioned compatibility model equations to get crude oil solubility number.
The solubility blending number, SBN > 100 is said to be compatible
according to Wiehe and Kennedy (2000), Wiehe
et al. (2001) and Fig. 2 shows that H2 and H6 are
self incompatible crude with SBN value less than 60. H3, H5 and H7
are nearly incompatible crude where SBN is slightly lower than 100.
The other two reference oils are believed to be self compatible crude. However
this behavior changes as the reference is added to the light Malaysian crude
Based on the information of the reference heavy oil in Table
3, the solubility and insolubility of the Malaysian crude oils were measured.
All samples are compatible with the average SBN of 118.29 and 125.27
for crude A and B respectively. Another set of experiment was also conducted
to identify potentially incompatible pairs of studied oils by blending a number
of crude oils in different proportions.
|| The solubility parameters for high asphaltene reference oils
|Where, δO= Oil solubility parameter, δf
= Flocculation solubility parameter, SBN = Solubility blending
number and IN = Insolubility number
|| The data of AFT analysis of crude oil blends
|*δF (MPa)0.5 = 8.37, IN
|| Graph shows the SBN of the reference oils
|| NHI% of crude A against time (top), temperature (middle)
and solvent type, carbon number (bottom) using Taguchi analysis
||Average NHI% table showing the optimum level and distribution
of asphaltene precipitation factors
||Analysis of Variance (ANOVA) for asphaltene precipitation
in crude A
|Where S = Sum of squares, V = Variance, F = F ratio, S
= Pure sum and % = Percent
The test result of the crude blend compatibility can be referred in Table
4 where all the blends are identified to be compatible with each other,
in great agreement with the oil mixtures compatibility criterion, where SBN
mix > IN max for the oil mixture to be compatible.
Asphaltene flocculation: The determination of mass percent of asphaltene (insolubility in normal-heptane solvent) is calculated as mass percent of normal - heptane insolubles (NHI %). It is a percentage by weight of the original sample. For percentages of insolubles less than 1.0, it was reported to the nearest 0.01 % and for percentages of insolubles of 1.0 or more; it was reported to the nearest 0.1%.
The data obtained and calculated NHI% of every crude oil was analyzed using DOE Taguchi approach to determine the behavior of asphaltene precipitation with regards to the studied parameters. Fig. 3 shows the precipitation behavior of asphaltene in crude A to the three parameters, where; NHI% is decreasing when temperature rises, decreasing to the increases of holding time and also decreasing to the increases of carbon number in the solvent used. A very distinguished pattern of drop can be observed in the NHI% response to the temperature. The same behavior was also observed for crude B to all the studied parameters.
Taguchi analysis provides us the contribution level of each parameter to the asphaltene precipitation behavior and ranks them. The highest amount of NHI% is predicted at the optimum level of analysis; a combination of each factor with the highest rank which can be referred in Table 5. From the table, we learned that the optimum level of asphaltene precipitation in crude A is predicted using n-pentane as solvent, 26 h of holding time at the temperature of 25°C producing NHI% of 0.351. The verification was conducted and the actual value of NHI% is 0.364 with 1.3% of deviation. The optimum level of asphaltene precipitation in crude B is using n-pentane as solvent, 12 h of holding time at the temperature of 25°C producing NHI% of 0.304. The prediction NHI% for crude C is 0.297 with 0.7% deviation.
Analysis of variance (ANOVA) is another statistical tool for splitting variability into component sources. These components can be reflected as the signal and the noise where the signal is seen as differences among group means and the noise is seen as variability within groups. By measuring the variability within groups, one has a baseline against which differences among group means can be compared. By comparing the differences based on design of freedom, DOF (n degrees 1), the F ratio was obtained which indicates the significance of the effect among group mean and later presented in percentage, P (%) which is mentioned in Table 6.
From the analysis, F ratio for solvent type is the highest; meaning solvent type (carbon number) has the greatest effect of asphaltene precipitation with relative influence of 39.01% for crude A and 32.50% for crude B. These values are doubled compared to the other two studied factors. The other or error term, ±41 and ±45% effect response for crude A and B respectively, are quite high considering other contributing factors to the asphaltene precipitation behavior, i.e. excluded factors in the study such as active sulfur content in the crude oil and the content of basic sediment and water (BS and W), uncontrollable factors and also experimental error.
All studied crude oils are self compatible and compatible to each other in the range of studied blending ratios. However, blending order wasnt included in the study to fully understand these crude oils compatibility and it is suggested to be explored in the future work. Asphaltene shows different precipitation behavior for each type of crude oil, and it is highly influenced by the presence of lower molecular hydrocarbon, i.e., pentane, heptane and etc in the crude oil. However, other factors that contribute to the precipitation should also be investigated. Asphaltene also contributes to a certain level of fouling where it precipitates and accumulation of other material could take place on the provided site by the asphaltene. Other properties in the crude oil are still important and contribute to the fouling behavior of a certain crude oil.
Understanding crude oil fouling requires a thoughtful knowledge of its chemistry and its physical deposition mechanisms. With the efforts from this research project, the study could facilitate us to understand the whole fouling process and possibly mitigate.
The author would like to take this opportunity to express his gratitude neither to University of Technology PETRONAS (UTP), Dr Chandra Mohan and Norhusna M Nor for their guidance, advice and help in the completion of this study. The author also thankful to General Manager of Novel Process and Advanced Engineering and Head of Treatment of Low Quality Crude Program, both from PETRONAS Research Sdn Bhd (PRSB) for their support in the research project.