International Journal of Engineering & Technology, 7 (3.

2) (2018) 636-641

International Journal of Engineering & Technology

Website: [Link]/[Link]/IJET

Research paper

Soil Compaction Methods Development

Tetyana Lvovska1*, Tetyana Lytvynenko2, Alla Kariuk3
1
Poltava National Technical Yuri Kondratyuk University, Ukraine
2
Poltava National Technical Yuri Kondratyuk University, Ukraine
3
Poltava National Technical Yuri Kondratyuk University, Ukraine
*Corresponding author E-mail: [Link]@[Link]

Abstract

A process of soil compaction methods development including new authors’ methodology is described. The importance of soil compac-
tion for engineering purposes is substantiated. Preconditions for Proctor compaction test appearance are highlighted. Proctor’s approach
and suggestions for the degree of soil compaction assessing are analyzed. Soviet version of Proctor’s equipment and Modified Proctor
compaction test are given. Principal differences between Proctor test, Standard compaction test and Modified Proctor test are presented.
The problems and disadvantages of existent soil compaction tests are revealed. New authors’ physical experiment methodology for pat-
terns establishment of water migration in subgrade embankment depth, in the capacity factors of what it is accepted: clay soil type (its
number plasticity); moisture, at what the soil was compacted; soil skeleton density; embankment height; «rest» time after subgrade erec-
tion and before it’s operation is developed and realized. By laboratory and field tests water migration patterns in compacted subgrade
soils depth are established. As a result of statistical processing of research results, the empirical dependence of compacted clay soil stabi-
lized moisture is obtained. Empirical dependence parameter corresponds to maximum molecular moisture capacity at what it is advisable
to do the subgrade clay soils multilayer consolidation for their long-term strength ensuring.

Keywords: long-term strength ensuring, modified Proctor test, molecular moisture capacity (maximum quantity of unfree water), standard density method,
standard Proctor test.
1. Introduction improve compaction, as it is one of the most important factors
determining the performance of road pavement structures.
Compaction of soils has been used to improve its properties for as
long as mankind had the need to provide pathways for carrying 2. Preconditions for Proctor Compaction Test
meaningful traffic volumes irrespective of whether the traffic vol- Appearance
umes were pedestrians, animal or motorised vehicles. Although it
can be accepted that the early road builders did not understand the In the USA, embankments began to be compacted layer by layer
principles of soil mechanics, they knew intuitively (and empirical- from 1925 to increase strength, reduce water permeability and
ly) that if a heavy load was applied to the soil, its characteristics subsequent soil’s settlement. However, the requirements for the
degree of soil compaction were not formulated by this time.
improved in a way that it provided a more mechanically stable
Therefore, when near Los Angeles the flood washed out the earth
pathway, more suitable for their purposes. bank (Fig.1), there was no one to lay a claim to the low work qual-
Soil compaction is widely applied in geotechnical engineering ity - there were no criteria for the compaction quality [2].
practice. It is used to maximise the dry density of soils to reduce
subsequent settlement under working loads or to reduce the per-
meability of soils. The durability and stability of structures are
highly related to the appropriate compaction achievement. The
structural failure of roads and airfields, and the damage caused by
foundation settlement can often be traced back to the failure in
achieving adequate compaction. For that reason, soil compaction
is important for engineering activities involving earthworks.
The need for proper compaction became crucial during the early
th
20 century, as a result of increased traffic and axle loads. A
wide range of different types of compaction equipment was de-
th
veloped during the 20 century, with vibratory compaction that
was gaining rapidly due to road pavement layers compaction
Fig. 1: The flood near Los Angeles, published in the Jan.2, 1926, Los
importance [1]. The development of compaction equipment Angeles Times
and techniques is still continuing today in order to further

Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original work is properly cited.
International Journal of Engineering & Technology 637

In response to these conditions, field engineer of the Los Angeles 3. Soviet Version of Proctor’s Equipment
water service, R. R. Proctor, in 1933 offered a method for degree
of soil compaction evaluating and published several articles on
In 1939, the road scientists N.N. Ivanov and [Link]. Telegin creat-
this topic in the journal «Engineering News Record». Fundamen-
ed a Soviet version of Proctor’s instrument, calling their version
tal principles of this method have been preserved in modern nor-
"Standard compaction" equipment. (The results obtained in both
mative methods for the degree of soil compaction assessing taken
equipments are the same). The highest density, obtained in this
in different countries, including Ukraine.
instrument was called the "Standard density" [2, 4].
The degree of possible soil compaction, first of all, depends on
However, the maximum density obtained in the "Standard com-
particles distribution in size and shape: if it contains both large paction", is not the maximum density for this soil. This was re-
and small grains, placed inside the pores between the coarse, the minded by prof. N.N. Ivanov, one of the Proctor equipment Soviet
possible density increases. The greater mechanical work applied to version creator. If, as prof. Ivanov mentioned, to increase the
the compaction, the higher soil density is achieved. Finally, with plummet weight (for example, up to 4.5 kg), the height of its
the same expended work on the given soil compaction, the result- dumping (for example, up to 42 cm), as well as the number of
ing density depends on soil moisture content during compaction. discharges (for example, up to 40), it is obvious that the density,
obtained in the equipment will be larger.
The combined influence of these factors makes it difficult to as-
sess the degree of compaction. On Proctor's suggestion, for each
soil, first in laboratory conditions, with the same for all soils com- 4. "Modified Proctor" Compaction Test
paction work, reveals the density to which it should strive to com-
So did the American road workers, when after the 2nd World War
pact it in the field, and then measure density indicator, obtained in
the rate of motion and weight-carrying capacity increased. They
the field, and compare it with obtained in the laboratory. Proctor developed a modification of Proctor's equipment, considering
offered to use not density (moist) soil ρ = M / V (mass per unit changes. The instrument was named "Modified Proctor" [5-7].
volume of soil) for evaluation, but the so-called soil skeleton den- The highest soil density in "Modified Proctor" differs from densi-
sity ρd - is the mass of solid particles in the unit volume of soil (in ty obtained in the usual "Proctor instrument", as well as from the
English - dry density - hence the index d). This is fundamentally highest density in the "Standard compaction" instrument by exact-
important. In fact, the purpose of compaction is to bring soil ly 5%. The compaction curve, obtained at Modified Proctor test
versus Standard Proctor test is presented in Fig.2 [7, 8].
coarse grains between them so that they form a system that is well
The principle difference between Standard compaction test, Proc-
aware of the external load [3-5]. tor test and Modified Proctor test are shown in Tab. 1.
The greater grains mass in a given soil volume, the closer grains
arranged and the better soil rolling. The soil mass M consists of Table 1: Principle difference between Standard compaction test, Proctor
the particles mass Ms and water mass Mw, and the air-bubble test and Modified Proctor compaction test
voids mass Ma is negligible: M = Ms + Mw. Dividing both sides Characteristics Standard compac- Proctor test Modified Proctor
of this equation on the soil volume V, we obtain ρ = ρd • (1 + w), tion test test
where w = Mw / Ms is the moisture equal to the ratio of the water Hammer
weight, 127.4 120 125 200 120 125 200
mass to the particles mass. Hence, after selecting a soil sample of
mm
a certain volume and weighing it, we can find the soil density Hammer
ρ = M / V. Then, after drying to a constant weight, its moisture is diameter, 10 10 15 25 10 15 25
found and the soil skeleton density is calculated from the formula сm
Drop
 (1) weight 2.5 2.5 4.5 15 4.5 4.5 15
d 
(1  W ) mass, kg
Drop
weight
This formula is invariably used for all degree of compaction con- 30 30 4.5 60 30 45 60
height of
trolling methods. The procedure of Proctor laboratory test is that fall , сm
the soil is compacted by blows of the falling weight in a metal Thickness
beaker at different moistures and optimum moisture is finded and of a layer
corresponding to it maximum soil skeleton density at the standard packed up 5-6 - - - - - -
compaction [4, 5]. The equipment used at Proctor test, the scheme at one
of Standard and Modified Proctor tests and curves, obtained at time, cm
Number
both tests are shown in Fig.2, Fig. 3 and Fig. 4 correspondingly.
of soil 3 3 3 3 5 5 3
layers
Number
of ham-
40 25 22 22 25 59 98
mer blows
per layer
Total
blows 120 75 66 66 125 295 294
amount
Stamp
diameter, 100 50 75 75 50 75 125
mm
Energy
imparted
from
- 0.6 0.6 0.6 2.7 2.7 2.7
hammer
for each
sample

Fig. 2: The equipment for Proctor test

638 International Journal of Engineering & Technology

Soil skeleton density, ρd, g/сm3

a

Moisture content, w, %
Fig. 4: The comparison of compaction curves between Proctor test (domed
marker curve) and Standard compaction test (rhombus curve)

Soil skeleton density, ρd, g/сm3

b
Fig. 3: AASHO tests:
a – Standard Proctor;
b – Modified Proctor
Moisture content, w, %
Fig. 5: The comparison of compaction curves between Modified Proctor
test (domed marker curve) and Standard compaction test (rhombus curve)

5. The New Approach in Soil Compaction Re-

Soil skeleton density, ρd, g/сm3

search
However, the main problem in soil compaction is that domestic
regulations prescribe optimal parameters of compacted clay soil
(maximum soil skeleton density ρdmax and optimum moisture con-
tent Wopt), based on the obtained values of laboratory conditions
for a particular soil type and dynamic load characteristics without
actual mechanisms characteristics [9].
When soil compacts due to high moisture, with a degree of satura-
tion close to Sr = 1.0, then, of course, the soil dries and held its
sedimentation and therefore additional deformations, when soil
compacts due to low moisture, it will be difficult to do the com-
Moisture content, w, %
paction, by the way, it’s a little probability of desired soil skele-
Fig. 4: The comparison of compaction curves between Standard and Mod-
ified Proctor compaction test ton density achieving, even with the modern mechanisms possibil-
ities. Also national standards recommend for optimum clay soils
Also prof. Pavlyuk in his work [4] presented the comparison of moisture content, at the compaction by roller, to take plastic limit
compaction curves obtained at Proctor test and Standard compac- Wp, but this parameter is not related on how much unfree water the
tion test (Fig. 5) and comparison of compaction curves obtained at soil is actually containing [9]. That’s why for subgrade erection
Modified Proctor test and Standard compaction test (Fig. 6). long-term strength ensuring is important, i. e., when during nor-
Maximum soil density is obtained at Modified Proctor test – 1.94 mative operational time the values of soil mechanical characteris-
g/cm3, minimum soil density value at Standard compaction test – tics, obtained after compaction, have been saved and excess soil’s
1.83 g/cm3, and 1.86 g/cm3 at Proctor test according to prof. deformation doesn’t appear.
Pavliuk research. The disadvantages of existent soil compaction methods are: the
necessity to «bind» the soil compaction curve to specific compac-
tion mechanism certain parameters; sufficiently wide boundaries
of optimum soil moisture; some subjectivity in soil plastic limit
determining and and so forth. So, the most common today concept
in road soil compaction construction solves mainly the technologi-
International Journal of Engineering & Technology 639

cal side of problem – the maximum soil skeleton density at fewest placement water content changing and probable capillary ascen-
mechanism passage by one trough. sion of water.
For the reliable subgrade operation it’s necessary not only to The exact maximum molecular moisture capacity values was de-
achieve maximum multilayer consolidation values of its soil skele- termined in laboratory conditions by moisture-holding capacity
ton density and soil strength, but also to save them during norma- surroundings method.
tive time. On the embankment soils condition over time signifi-
cantly affects moisture, at what the compaction was done and the
particular water type split in compacted soil [9, 10]. But it is still
not researched quantitative impact on moisture conditions patterns
in road embankment clay soils depth the next factors: soil skeleton
density; compacted embankment height; the number of days,
when compacted clay embankment «rests» after its erection, and
before the operation.
In soil mechanics, minimum stress at what the soil sample is de-
stroyed due to aeonic span of time, is accepted as long-term
strength limit. Stress, at what the soil sample is destroyed due to
certain period of time after load imposing as a result of unchange-
able soil flow and advance flow of ground, corresponds to the soil
long-term strength. So the theory considers revising that the most
favorable conditions for the subgrade clay soils long-term strength
ensuring and in accordance minimum ground distortion during
specified time operation is to compact the soil in layers at mois-
ture, close to maximum molecular moisture capacity [9, 10].
For new road embankment soils optimal compaction criteria sub-
stantiation with a purpose to ensure subgrade clay soil long-term a
strength, the new physical experiment methodology for water
migration patterns establishment in subgrade embankment depth is
developed, the capacity factors of what were accepted:
– clay soil type (its number plasticity Ip);
– moisture, at what the soil was compacted;
– soil skeleton density in the embankment;
– embankment height;
– «rest» time after subgrade erection and before it’s operation.
For its realization the equipment of Geotechnics department was
used, where in 2013 – 2016 years the water migration patterns
heightwise through the time in soil, placed in plastic tubes (it is
simulated the soil multilayer consolidation of subgrade embank-
ment). Research soil condition and type are changeable parameter.
At the unifactor experiment two typical for compacted subgrade
erection clay soils – heavy silt loam and light silt loam were used.
The detail experiment description is presented in author’s
monography [9]. The soil was moved in pipes and uniform com-
pacted using hand tamper with certain marks by its height
(Fig. 6, а) all time to thickness of 30 mm, and then pipe link (150
mm) was connected to a total height (Fig. 6, b).
These pipes with compacted in layers soil were installed on a met-
al vertical frame. The lower pipe ends was installed in drainage
channel, filled with stone screening dust (stone chips). Thus, free
(gravity) water is able to migrate at all soil pipe height, what imi-
tate it’s migration within road embankment depth. The top of all
pipes with compacted soil layers was hermetically closed to avoid
evaporation of water «up». Then pipes were left on vertical frame
alone, for the «rest».
After allowed time «rest» all pipes were dismantled into separate
links. From each link it was selected at least two soil sample in b
bottles, and by normative weight method the final (stabilized) Fig. 6: The equipment used at new author’s methodology:
moisture of compacted soil was determined. So, within the labora- a - hand tamper with certain marks by its height;
tory experiment it was done: b – pipes, gathered to a general height
– 4 experiment series for relation of final soil moisture №1 from
soil skeleton density ρd = 1.50 – 1.65 g/сm3 at pipe height 150 cm The dependings graphs of compacted loam moisture was drawn
substantiation; and shown in [9].
– 4 analogous test series with research soil №2 at ρd = 1.50 – 1.65 The value of stabilized moisture generally corresponds to maxi-
g/сm3 and pipe height 150 cm; mum molecular moisture capacity, at what it is advisable to do the
– 5 experiment series for relation of final soil moisture №2 de- subgrade clay soils multilayer consolidation for their long-term
pending on pipe height 45, 90, 150, 210 and 285 сm and ρd = 1.55 strength ensuring. So from the perspective of clay soil long-term
g/сm3; strength ensuring just that parameter it is viable to accept as opti-
– 3 test series for relation of final soil moisture №2 depending on mum moisture content Wopt of soil compaction.
waiting time 74, 120 and 180 days at ρd = 1.55 g/сm3 revealing; The average results of compacted clay soil stabilized moisture
– 2 experiment series for relation of final soil moisture №2 from are determined by whole pipe height, except it’s upper and lower
soil skeleton density ρd = 1.60 g/сm3 and pipe height 150 cm at links for each preset soil skeleton value ρd = 1.50 – 1.65 g/сm3
placed in Tab. 2 and Tab. 3.
 640 International Journal of Engineering & Technology

Table 2: Average of final moisture wk of compacted heavy loam (research

 
soil № 1) heightwise whole pipe wk  a  b ln  d  , (2)
Preset soil
skeleton
Corresponding
Preset soil Final soil Variation
moisture w moisture, coefficient
 d 0 
void volume
density (at Sr = wk wk, υ
ratio, e
d, g/сm3 0.85) where is  d 0 = 1 g/сm3;
1.50 0.786 0.250 0.203 0.071
1.55 0.729 0.231 0.190 0.068
empirical coefficient is: а = 0.358; b = - 0.384.
1.60 0.675 0.214 0.176 0.063 At this multiple correlation coefficient r and variation coeffi-
1.65 0.624 0.198 0.167 0.065 cient ν values in accordance is: r = 0.997; ν = 0.008, what in-
dicates about close relationship between the experimental data
Table 3: Average of final moisture wk of compacted light loam (research and about the correctness of their approximation by the loga-
soil № 2) heightwise whole pipe rithmic function [9].
Preset soil Preset soil Final soil Variation Analogous logarithmic dependence is obtained also for stabilized
Corresponding
skeleton moisture w moisture, coefficient
density
void volume
(at Sr = wk wk, υ moisture value wk of compacted light silty loam (research soil №
ratio, e
d, g/сm3 0.85) 2). Empirical coefficient of equation (1) is: а = 0.362;
1.50 0.786 0.250 0.162 0.072 b = - 0.494.
1.55 0.729 0.231 0.143 0.070 At this multiple correlation coefficient r and variation coeffi-
1.60 0.675 0.214 0.130 0.072 cient ν values in accordance is: r = 0.998; ν = 0.0115, what
1.65 0.624 0.198 0.114 0.068 indicates about close relationship between the experimental
In the last column of both tables the variation coefficient values of
parameter is presented, by what the soil can be accepted as homo- data and about the correctness of their approximation by the
geneous. logarithmic function [9].
By data of Tables 2 and 3 the dependings graphs (Fig. 7 а, b) are Whereas road embankment height with compacted loam did not
drawn: significantly affect its moisture conditions and subgrade «rest»
– graph 1 – of soil moisture w, at what both research soils were
compacted from soil skeleton density in road embankment (pipe); time increasing after 2 months did not significantly affect the sta-
– graph 2 – stabilized soil moisture of already compacted loams bilized soil moisture value, for this reason, it is advisable to per-
after subgrade «rest» time from within pipe height. form two-factor statistical analysis of compacted clay soil stabi-
lized moisture depending on its soil skeleton density and plasticity
index.
As a result of this statistical processing by least squares method
the empirical dependence is obtained [9]
Soil skeleton density, ρd, g/сm3

 
wk  a0  a1  d   a2  I p (3)
 d 0 

empirical coefficient is: а 0 = 0.531; а 1 = -0.279;

а 2 = 0.570.
At this multiple correlation coefficient is r = 0.995, and Fish-
er's ratio test F = 106.326, what more than its table-valued
Fтабл = 4.89 at test significance р = 5% and the degree of free-
a
dom v1 = 7 and v2 = 5.
Moisture content, w, %
Statistical values indicate close relationship between the re-
search data and therefore, about the logarithmic function (3)
correctness.
Soil skeleton density, ρd,g/сm3

The results of physical laboratory experiment related to quantita-

tive patterns of water migration in compacted heavy and light silt
loams (clay soils type – its plastic index Ip, soil skeleton density ,
g/сm3, stabilized (final) moisture of compacted clay soil wk) are
presented in Tab. 4.
The Tab. 4, in particular, clearly shows that an increase of its plas-
ticity number Ip at the same soil skeleton density values, stabilized
moisture of compacted clay soil wk increases.

Table 4: Stabilized moisture values of compacted heavy and light silty

loams within pipe height for each preset subgrade soil skeleton density
b Preset soil skeleton density, Soil plasticity number, Ір
Moisture content, w, % d, g/сm3 0,162 0,080
0.203 0.162
1.50
Fig. 2: Plots relation of soil moisture w at what the loam was compacted, -0.95% 2.36%
(1) and stabilized (final) soil moisture wk of already compacted heavy 0.190 0.143
loam from after the «rest» time of subgrade (2) from soil skeleton density 1.55
-0.51% -0.86%
within pipe height: а – heavy loam; b – light loam 0.176 0.130
1.60
-0.58% -0.21%
As a result of statistical processing by least squares method of 0.167 0.114
1.65
research data Tab. 2 (and Fig. 7, a) it is found that the decreasing 2.35% -2.04%
Note: numerator – the experimental values of stabilized clay soil moisture
of stabilized moisture value wk of compacted heavy silty loam
wk; the denominator – the relative error of this parameter, calculated by
(research soil № 1) depending on the soil skeleton density growth the expression (3)
within the experimental range  d  1.50 – 1.65 g/cm3 it is most
correctly to describe by logarithmic function of the form [9]
International Journal of Engineering & Technology 641

6. Conclusions . – Under the general editorship Anatoliy M. Pavlenko. – Kiel-

ce: Wydawnictwo Politechnici Swietokrzyskiej. – 2017. – P.
132 – 155. – PL ISSN 1897-2691. PL ISBN 978-83-65719-19-
1. Agreeably with standards maximum soil skeleton density ρdmax 5.
and optimum moisture content Wopt, are prescribed from obtained [10] Vynnykov Y.L. Investigation of Compacted Clay Embankment
for particular soil type and dynamic effect parameters without Stabilized Moisture / Y.L. Vynnykov, V.I. Kovalenko, T.V.
mechanism peculiarities considering. But the standarts do not Lytvynenko // Collection of scientific articles «Energy, Energy
correspond to machinery facilities, because with compaction ener- Saving and Rational Nature Use». – №2(3), 2014. – Kazimierz
Pulaski University of Technology and Humanities in Radom,
gy per unit of volume increasing the optimum moisture content Poltava National Technical Yuri Kondratyuk University. – Ra-
and maximum soil skeleton density values considerably change. dom, 2014. – P. 53 – 57.
By the full-scale experiment the theory for subgrade erection long-
term strength ensuring and minimum deformations during its
normative operational time, the soil multilayer consolidation
should be done at moisture, close to maximum quantity of unfree
water is approved.
2. As a result of statistical processing by least squares method the
research laboratory and field data, the empirical dependence of
compacted clay soil stabilized moisture for their multilayer con-
solidation in relation to soil skeleton density and plasticity number
values is obtained. Empirical dependence parameter corresponds
to maximum molecular moisture capacity at what it is advisable to
do the subgrade clay soils multilayer consolidation for their long-
term strength ensuring.
3. Clay soil embankment multilayer consolidation is desirable to
do at moisture, that corresponds to maximum molecular moisture
capacity, accepted by the expression (3) depending on soil skele-
ton density specified rate and plasticity number of this soil. The
value of this moisture is lower than plastic limit, but present-day
compacted machines makes it possible to sufficiently increase the
specific compaction energy for the soil compaction at moisture
less than plastic limit with a maximum soil skeleton density.

Acknowledgement
The authors would like to express their sincere gratitude to the
colleague and onetime scientific advisor of one of the author -
[Link], for his valuable advices and inspiration.

References
[1] K. Nurses, Compaction behaviour of soils, Monash University,
Australia, 2011, 239 p.p., doi:10.4225/03/5890111BEEA19
[2] R.C. Karen, S. Landsburg, Soil Compaction and pipeline construc-
tion: A Literature Review, Research monograph (1990), Nova cor-
poration of Alberta, Calgary, Alberta, 48., doi:10.7939/R3QV81
[3] Ribeiro E., Batjes N.H., World Soil Information Service (WoSIS) -
Towards the standardization and harmonization of world soil data.
Procedures manual 2018, Report 2018/01, ISRIC – World Soil In-
formation, Wageningen, doi:10.17027/ISRIC-
WDCSOILS.20180001
[4] Pavlyuk D.O. Porivnyannya metodiv i zasobiv kontrolyu
ushchilʹnennya gruntiv pry budivnytstvi zemlyanoho polotna /
D.O. Pavlyuk, M.V. Shurʺyakov, S.A. Hladun // Zb. nauk. pratsʹ
(Haluzeve mashynobud., bud-vo) / Polt. nats. tekhn. un-t im.
Yuriya Kondratyuka. Vyp. 2(47). – Poltava: PoltNTU, 2016. – S.
233 – 245.
[5] A. Aboufayed, Soil moisture content in hill-filed side slope, Inter-
national, Journal of Geological and Environmental Engineering,
Vol.7, No.3, (2013), pp.147-151., doi:10.5281/ZENODO.1061638
[6] T.C. Hopkins, T.L. Beckham, [Link], Characteristics and engineer-
ingp of the soft soil layer in highway soil subgrades, Research Re-
port 2006/06, Kentucky Transportation Center,
[Link]
[7] P. Schjønning, Case study soil compaction, available online:
[Link]
oil_compaction?channel=doi&linkId=5640751d08ae8d65c0151fd6
&showFulltext=true, last visit: 30.03.2018, doi:
10.13140/RG.2.1.2994.3767
[8] [Link], R.W. Zobel, Optimum sub-grade compaction levels for
turf growth in constructed soil profiles, Technical Report 2009/05,
doi: 10.13140/2.1.2621.2482
[9] Vynnykov Yu. Moisture conditions patterns in road embank-
ment clay soils depth / Yu. Vynnykov, T. Lytvynenko
// Energy and Environment: monografie, studia, rozprawy.M95