550293-Control of Low-Temperature Sludge in Passenger-Car Engines
550293-Control of Low-Temperature Sludge in Passenger-Car Engines
550293-Control of Low-Temperature Sludge in Passenger-Car Engines
This paper was presented at the SAE Golden Anniversary Annual Meeting, Detroit, Jan. 11, 1955.
Action of Detergents
Over the past several years, considerable atten-
tion has been given to determining the mechanism T HIS paper evaluates a new class of detergents,
I specially effective in controlling sludge for-
by which detergents function in lubricating oils. mation at the low or moderate temperatures most
While this mechanism is not yet completely under- often encountered in passenger-car operation.
stood, sufficient information has now been devel- This new Detergent C (a copolymer of lauryl
oped to permit a rational explanation of most of methacrylate plus diethylaminoethyl methacry-
the phenomena observed. late) has undergone extensive bench and labora-
Materials which function as detergents in lubri- tory tests in comparison with three other major
classes of detergent. It exhibits, among other
cating oil must possess at least one polar and one attributes:
nonpolar group. The nonpolar group provides oil
1. Complete solubilization of sludge.
solubility for the molecule, while the polar group 2. Complete carbon suspension under both wet
is of such a nature that it is attracted to the par- and dry conditions.
ticles of sludge. These detergent molecules tend 3. Residual detergency, as indicated by dis-
to cluster together to form micelles containing persancy in blotter tests under wet and dry
many molecules. Fig. 1 shows electron photomicro- conditions.
4. Ability to prevent agglomeration and precipi-
tation of sludge formed during oxidation of lubri-
1 "New Class of Polymeric Dispersants for Hydrocarbon Systems," by cating oils, with and without presence of water.
C. B. Biswell, W. E. Catlin, J. F. Froning, and G. B. Robbins. Presented
at National ACS Meeting, Kansas City, Mo., 1954. 5. Effective sludge suspension after 40-hr lab-
2 "An Electron Microscope Study of Performance of Detergent Gil," by
oratory tests, and after several months of storage.
J. B. Peri. Presented at SAE National Fuels & Lubricants Meeting, Chicago,
November 5, 1953.
778 SAE Transactions
Downloaded from SAE International by Columbia Univ, Tuesday, April 14, 2020
were generally not very effective in the presence gent C was the only effective dispersant.
of water. "Blotter" Tests – Gates and co-workers 8 have
Because crankcase oils eventually become acid recognized the importance of water, and have re-
during the normal use period, the carbon-suspend- ported that the effectiveness of a detergent oil can
ing ability of several detergents of various types be neutralized by engine-coolant leakage into the
was determined in the presence of two volume per crankcase, and also by the presence of aqueous
cent of a 10 weight per cent hydrochloric acid solu- blowby condensate in the oil. Gates and also Edgar9
tion. The quantity of acid was sufficient to more have used "blotter" or filter-paper spot tests to
than neutralize the basicity of the kerosene- evaluate the residual detergency of used engine
detergent system. Under these conditions, Deter- oils. In such tests, when a drop of the oil contain-
ing an effective detergent is placed on a piece of
filter paper, the dispersed sludge flows out with
DETERGENT: 0.5 WT. % ACTIVE INGREDIENT the oil into the paper, giving a large, dark oil spot.
If the detergent properties of the oil have been
100 exhausted, the sludge remains in its original posi-
tion on the paper, and only clear oil spreads out
1\_ \—DETERGENT B
DETERGENT C
into the paper. The blotter test performance of
used oils formulated with the four different deter-
47--DETERGENT,A gents has been determined. The essentially water-
free used oils for this experiment were taken from
50 Buda diesel engines operated for 50 hr at 175 F
of—DETERGENT oil temperature and 195 F coolant temperature.
The oil spots illustrated in Fig. 8 show the residual
detergency of the used oils as well as the effect of
the addition of 0.5% water. In the dry systems,
0 2 3 4
5
6
Detergents A, B, and C all show good detergency
as indicated by the large black spots, while Deter-
DAYS gent D is marginal in performance. One-half per
Fig. 5— Effect of detergent on suspension of carbon in dry kerosene cent water neutralizes the Detergents A and D and
decreases the efficiency of Detergent B. Water has
DETERGENT: 0.5 WT. % ACTIVE INGREDIENT no effect on the dispersancy of Detergent C.
Oxidation Tests – The ability of the detergents
100 to prevent the agglomeration and precipitation of
DETERGENT C sludge formed during the oxidation of lubricating
oils was determined using the MacCoull test equip-
DETERGENT B
"On-the-Spot Testing of Used Lubricating Oils," by V. A. Gates, R. F.
Bergstrom, T. S. Hodgson, and L. A. Wendt. Presented at SAE National
50 West Coast Meeting, Los Angeles, Aug. 17, 1954.
9 "High Additive Oils in the City, on the Long Lines, and Off the Road,"
DETERGENT A by J. A. Edgar. Presented at SAE National West Coast Meeting, Seattle,
jr DETERGENT D Aug. 14, 1951.
USED
I I I I I I
0
I 2 3 4 5 6
DAYS
Fig. 6— Effect of detergent on suspension of carbon in wet kerosene
ERGE%T A
DETERGENT DETERGENT ,
E KEROSINE
USED OILS WITH C .TE,
Fig. 7— Sample vials containing kerosene suspensions of carbon, showing Fig. 8 — Used-oil spot tests, indicating that water neutralizes effective-
dispersant performance of representative detergents in presence and ness of Detergent A, harms that of Detergent B, and has no effect on
in absence of water. 1: no detergent, 2: Detergent A; 3: Detergent B; that of Detergent C. Detergent D exhibits little, if any, residual deter-
4: Detergent C; 5: Detergent D gency even in dry oil
ment. 1° The base oil used was an SAE 10W Mid- small oil samples removed periodically from ti
continent solvent-refined neutral, carefully filtered engine. In addition, the gradual development anew
to remove tramp solids. The detergents were added growth of the sludge particles was followed by
at a concentration of 0.5 weight per cent active an examination of these used-oil samples with an
ingredient. Duplicate samples of these treated oils electron microscope. The base oil used in these
were oxidized at 200 F, and 2 ml of water was tests was a solvent-refined Midcontinent neutral
added every 30 min to one set of samples. Oxida- of 140 SUS at 100 F. It was carefully filtered to
tion was not extensive even after 60 hr in the test remove tramp particles which might otherwise
equipment. However, examination of the oxidized have been mistaken for sludge. One weight per cent
oils under the electron microscope did show that of a zinc dialkyldithiophosphate antioxidant and
the type of detergent present exerts considerable 0.5 weight per cent, on an active ingredient basis,
influence on the nature and the extent of sludge of the required detergent were added to this filtered
formation. This study also demonstrated that oil. The concentration of the detergent was set at
water increases the rate at which sludge develops. a value lower than used in service, in order that
These results are illustrated in Fig. 9. The electron depletion of the detergent would be appreciable
photomicrographs show plainly the effect of water during the course of the 40-hr engine test. The
on the quantity and particle size of the sludge. tests were all started without the separate break-
The peculiar feathery sludge particles formed in in, using 6 qt of oil rather than the customary 4 qt.
the presence of Detergent B were obtained repeat- This procedure prevented contamination of the
edly; they were not present in the unoxidized oil. test oil with sludge from the break-in period, and
The ability of Detergent C to minimize sludge also provided sufficient oil for completion of all
formation, and the insensitivity of the polymer the tests without further oil additions. The engine
system to water, are illustrated quite clearly by was operated on a non-leaded fuel. A representa-
these photomicrographs, which were selected to tive 15 ml sample of oil was withdrawn from the
represent the typical, or average, of approximately crankcase each hour. These small samples were
10 pictures taken on each oil. In turn, the 10 pic- used in sludge settling tests and electron micro-
tures were typical of some 25 to 40 visual observa- scope studies, after gradual detergent depletion.
tions made on each oil. The engine scores obtained from these tests are
summarized in Table 1. It is evident that at equal
Effect of Detergents on Sludge Development in Lab Engines concentrations the different types of detergents
Certain of the techniques used in the bench test vary widely in their effect on piston varnish and
work have been applied to an engine study of overall engine score. Detergent C, as might be pre-
these detergents. The gradual loss of oil deter- dicted from the previously discussed bench test re-
gency during a series of FL-2 11 engine tests was sults, was very effective in minimizing sludge and
followed by observing the sludge settling rates of varnish formation. Although the oil containing De-
tergent A actually gave a lower piston score than
the base oil, it did show some improvement in over-
all engine cleanliness. Detergents B and D had little
10 See SAE Transactions, Vol. 50 (August), 1942, pp. 338-345: "An Oil effect on the oil performance under these conditions.
Corrosion Tester," by N. MacCoull, E. A. Ryder, and A. C. Scholp.
". "Laboratory Engine Tests of Sulfur in Motor Gasoline Field Test
Fuels." Report of Coordinating Research Council, Inc., January, 1950.
The tendency of the sludge to settle out of the
used oils was studied visually following vigorous
shaking of the samples. These settling tests were
repeated several months later on the same samples
with the same results, indicating that once the oils
were removed from the engine their detergent qual-
ities were quite stable. The settling rates obtained
indicate that Detergent A loses its effectiveness
after 6-8 hr of engine operation, Detergent B at ap-
proximately 15 hr, and Detergent D at 4 hr. Deter-
gent C effectively suspends sludge for the complete
40-hr test. The results obtained are illustrated in
Fig. 10, which shows photographs taken after four
days of settling of the oil samples taken after 0, 3,
6, 10, 15, and 40 hr of engine operation. The base oil
showed no tendency to suspend sludge, and, there-
fore, samples of it were not photographed. The
physical appearance of the sludge at the 10- and
40-hr points of the tests is illustrated by the elec-
Fig. 9 — Electron photomicrographs at 3800x of sludge from oils oxi- tron photomicrographs in Fig. 11. Small sludge
dized at 200 F in MacCoull apparatus. Water accelerates oxidation particles can be observed in the 10-hr samples of all
and formation of sludge particles four detergent oils. The 40-hr photographs indi-
Table 1 – FL-2 Engine Test Scores Obtained in the Table 2 – FL-2 Engine Performance of the
Sludge Development Study Four Different Detergents
Oil: Solvent-refined SAE lOW antioxidant Oil: Solvent-refined SAE 30 + antioxidant
Fuel: 10% alkylate. 90% catalytic cracked, unleaded Fuel: 100% thermal reformate -I- 3 ml tel/gal
Piston Score Piston Score
(10 = Clean) Engine Score (10 = Clean) Engine Score
Active --------- (50 = Clean) Active (50 = Clean)
Ingredient, No. 1 Average Ingredient, No. 1 Average
Detergent weight % Piston Piston Varnish Sludge Detergent weight % Piston Piston Varnish Sludge
None 3.5 5.0 30 38 None 4.5 5.9 29 39
A 0.5 1.0 3.7 33 44 A 1.2 2.5 4.1 30 40
B 0.5 4.5 6.0 32 37 B 1.0 5.5 7.5 38 41
C 0.5 7.0 8.3 39 47 C 1.0 8.0 9.3 42 43
D 0.5 4.6 5.8 36 35 D 1.0 7.0 8.3 41 43
cate that in the case of Detergents A, B, and D the gent oils is small enough to pass through efficient
small particles have agglomerated very extensively, lubricating oil filters. At 40 hr, however, such filters
whereas in the oil containing Detergent C there has would trap most of the sludge in all of the oils ex-
been little change in particle size, although the cept that containing Detergent C. In this oil, only a
number of particles has increased somewhat. Much few of the particles are of filterable size – that is,
of the sludge from the 10-hr samples of the deter- larger than 1 to 2 microns in diameter.?
Representative samples of sludge were photo-
graphed again through the electron microscope af-
ter the samples had been stored for several months.
No observable change in particle-size distribution
or particle shape occurred during this storage pe-
riod. This result suggests that growth or agglomer-
ation of sludge particles normally occurs only dur-
ing periods of engine operation as the detergent is
gradually depleted. Agglomeration can also occur
in the presence of some detergents during periods
of nonoperation, if the water content of the oil is
increased by condensation or coolant leakage into
the crankcase.
Table 3 - EX-3 Engine Performance of the Table 4 - Effect of Type of Detergent on the
Four Different Detergents Sludge Content of Used Oils
Oil: Solvent-refined SAE 30 + antioxidant Test: FL-2
Fuel: 100% thermal reformate -I- 3 ml tel/gal Of I: Solvent-refined SAE 30 + antioxidant
Narrow Fuel: 100% thermal reformats + 3 ml tel/gal
Engine Score
(50 = Clean) Slot Oil Average Sludge Content of Oils.
Active Average
Ingredient, Piston Score Ring,
Vo plugging Active n-Pentane Benzene Insoluble
Detergent weight % (10 = Clean) Varnish Sludge Resins,
Ingredient, Number Insolubles, lnsolubles,
None 5.2 30 35 38 weight % weight % weight %
29 Detergent weight % of Tests
A 1.0 3.5 28 40
38 14 Noneb 27 1.94 1.16 0.78
a 1.0 4.2 24 0.65
43 7 None 5 1.87 1.22
C 1.0 8.6 40
1.23 0.68
4.5 23 36 12 A Type 1.2 4 1.91
1.0 2.43 1.63 0.80
B Type 1.5 5
C 0.5 56 1.17 0.83 0.34
C 1.0 11 0.86 0.67 0.19
C 1.5 3 0.61 0.55 0.06
Type 1.0 4 2.33 1.02 1.31
DETERGENT E
1.4 WT. % ACTIVE INGREDIENT
Fig. 14- Blocks and side plates of
engines from extended service
test
ETERG
AcrwE INsREDIENT
gent similar to Detergent A) in two others. Aver- suburban operation. The results of this test agree
age analyses of used oils taken from this test are with the laboratory engine tests and show that De-
given in Table 5. The oil drain period was 2500 tergent E ( a detergent similar to Detergent A) , at
miles, and the results presented in the table are the an active ingredient level of 1.4 weight per cent, did
averages obtained after analyzing the individual not markedly improve engine cleanliness during low
used-oil samples from every drain period of each and moderate temperature operation over that ob-
car. The performances of the two cars on each oil tained with the base oil. In contrast to the appear-
were in good agreement, and the differences be- ance of the engines lubricated with the base oil or
tween the two oils were clearcut ; Detergent C min- oil containing the Detergent E, all of the engines
imized insoluble-resin formation to a much greater lubricated with oils containing Detergent C were
extent than did Detergent F. The reduction of the clean. Photographs of parts from representative
benzene-insoluble contents from 0.97 and 1.16 engines at the end of the test on the base oil, an oil
weight per cent with the oil containing Detergent containing Detergent E, and two oils containing
F to 0.45 and 0.42 weight per cent with Detergent C Detergent C are shown in Figs. 12, 13, and 14.
also was significant.
Summary and Conclusions
Another field test involved extended urban and
1. Various types of detergents, including a new
ashless, polymeric type, have been evaluated for
Table 5 - Effect of Detergents on the their performance at low engine temperatures.
Sludge Content of Road Test Oils 2. Detergent types vary widely in their ability to
Oil: Solvent-refined SAE 30 + antioxidant
Test: 40,000 miles per vehicle
maintain engine cleanliness under low-temperature
2,500-mile drain periods conditions.
Vehicle
3. Detergents which are effective at high engine
1 2 3 4
Detergent C, weight % temperatures are not necessarily effective at low
Active Ingredient
Detergent F, weight %
- - 1.5 1.5 engine temperatures.
Active Ingredient
New Oil Ash, weight %
1.3
0.28
1.3
0.28
-
0.09
-
0.09
4. Bench tests show that many detergents lose
Used Oil Analysis,'
n-Pentane Insolubles, weight % 1.73 2.05 0.59
their effectiveness when evaluated in systems con-
0.57
Benzene Insolubles, weight %
Insoluble Resins, weight %
0.97 1.16 0.45 0.42 taining small amounts of water. The performance
0.76 0.89 0.14 0.15
of detergents in low-temperature engine service is
ASTM Method D893-52T. Each value is the average of the analyses of the 16 oil drain
samples obtained during the 40.000 miles of operation. believed to be related to their sensitivity to water.
5. Bench tests and analytical data from used en-
784 SAE Transactions
Downloaded from SAE International by Columbia Univ, Tuesday, April 14, 2020
Note: All of these engines were in reasonably clean condition Engine Buildup
because of the low mileage operated - 3000.
Four 1952 overhead-valve V-8 engines of a well known
make were carefully rebuilt after about 25,000 miles of
operation. New hydraulic valve lifters, new oil filters, and
clean filter cases went into each test engine. Then the en-
gines were reassembled in a completely clean condition.
Ethylene glycol-type permanent antifreeze was used as
coolant in conjunction with 156 F thermostats.
Operations
Cars were parked outside at all times. Long trips were
avoided. Driver assignments were changed each week. Car
travel averaged about 25 miles per day, and oil was drained
while hot. Total mileage was 3000 miles.
Three oil drain periods for respective cars were used:
750-, 1500-, and 3000-mile intervals.
6 AIN Test Results
Fig. C - Filters from test cars after 750-, 1500-, and 3000-mile drain Figs. B and C show how the drain periods affected the
periods oil screens and filters from cars 1, 3, and 4. They offer
visible proof that the filter elements from the 1500- and
3000-mile drain period cars weighed almost % lb more than
did the filter from the 750-mile drain period car.
It can be seen from this test that a vehicle operated under
these severe home-to-work conditions, using a common
arf ?Lj winter grade fuel and oil containing a sizable amount of
W.:t4 mlLeadi Jfiai.lp 4w. id-a
metallic detergent, should be drained either once a month
or after 1000 miles of operation. This conclusion relates
favorably with the API Committee recommendation that
"crankcases should be drained after not over 60 days, or
500 miles of travel in this type of service."
Ashless Detergent Results
Concurrently with the three cars already described, a
fourth car, No. 2, was operated on an oil treated with an
experimental ashless type Additive C without drain for a
Fig. D - Oil screen from test car operated on oil treated with experi-
mental additive C after 3000 miles 3000-mile period. Figs. D and E show the oil screen and
filter from this engine, which compare favorably in condi-
tion with the screen and filter of the 750-mile drain opera-
tion using the metallic detergent (Car 1).
Combined Sludge and Varnish Ratings
The used oil analyses (Table B) indicated that the ash-
less detergent oil was capable of suspending some three
to seven times as many oil contaminants as the metallic
detergent - probably due to the fine dispersion of these
solids, as pointed out by the authors.
This property of sludge and insoluble contaminant sus-
pension is further observed in another of our tests, as
shown in Fig. F, where four used oil samples of the quali-
ties premium, MIL-L-2104A, S-2, and Ashless Detergent C
were settled for a 15-month period. Each of these oils was
operated under stop-and-go city driving conditions for over
4000 miles without draining. The amount of settling
Fig. E - Oil filter from test car operated on oil treated with additive appears to vary proportionately with the detergent dis-
C after 3000 miles persant quality of the oil - the premium oil on the left
containing the least detergent shows the most settling, and
the ashless type, Detergent C treated oil on the right, shows
very little separation.
We can conclude then with the authors that the use of
an electron microscope to determine the effectiveness of
detergents to prevent sludge particle size growth in oil is
a much more effective and novel method than waiting for
used oil samples to settle. Although the work described
here is limited and a great deal more road testing is neces-
sary, the results substantiate that large differences exist
between the metallic and ashless type C detergents in their
Fig. F - Sludge-settling behavior of used motor oils in 15-month test. ability to keep filters, oil screens, and engine parts free
Left to right: premium, MIL-L-2104A, 5-2, and Ashless Detergent C from sludge under low-temperature operating conditions.
786 SAE Transactions