Hall Effect
Hall Effect
Hall Effect
JAS-ANZ
,~~,
K50~
years D
~.
~..)tCt
/.sy~ - 1,\'{":
Excellency in QUlIli ty
I\'1.JA\TANr r_
VIJAYANTA TECHNOLOGIES PVT. LTD.
(Formerly Vijai Electrollic., )
Dr. BALDEV SINGH MARG
28/147, CIVIL LINES,
ROORKEE - 247667
The conductivity measurements cannot revea l whether one or types of carriers are prcsent;
hoI' distinguish between them . Ho\\cver. Ihis inrnrmaLioll can he ohtai neu fl'\1Il1 I In1 1 I rkcl
~casurements. \\ hieh are basic tools for thL' lktcrminatinn Ill' I1h)b ililiL' <' Jill' l"I Y<.'l-1 11 .1'
(ji1;covered by E.H. Hall in I ~79 .
As you are undoubtedly aware. a static magnetic field has no effect on charges unless they
arc in motion. When the charges flow, a magnetic field directed perpendicular to the direction of
nov.. produces a mutually perpendicular force on the charges. When this happens. electro ns and
lioles will be separated by opposite forces. They will in tum produce an electric field (Eh) which
d~pends on the cross product of the magnetic intensity, H. and Lhe current density, J. ttie situation is
d(;Jnonstrated in Fig. 1
Eh R J x H (I)
o\'., let us consider a bar or semiconuuclor. having UlnlCn"I(lIl, \ . \ dlld I . kl .I " dlll'LleJ
al6ng X and H along Z them Eh will e along Y. as in rig. 2.
Then we could write
Vh/Y V h. Z
R = = (2)
J H IH
Where Vh is the Hall voltage appearing between the two surfaces perpendicu lar to y and 1 = Jyz.
In general, the Hall voltage is not a linear fu nction of magnetic field arr licd . i.c. the 1-1 aI I
cocllicient is not generally a constant, but a fu nction or the ap r lied l11agm':li<: . Conscq ul' nlly
interpretation of the Hall voltage is not usually a simple tn~ll,:r . Ill)\wvl.: r. it i ~ l:a" : \(1 caku l<l ll'
this (llall) voltage if it is assun1\:d thaI JII carr i ~r~ haw I h<.' " Ull l !.: drii'i \ ~I \ 'l i l \ \\~ Ilill ,h I ill ;,
In two Sleps (a) b) assuming that carriers o r onl: om: I~pl: dlC pn::-.tl11. .1IlJ {h) h; d :-':-'lIlll i n g Ih.ll
em"tiers of both types are present.
_ e-_~_/~HllnlOP09.1
---'---0--11;
- 8-----;.. ,\ /
,'I J
r---
-- - _/ / / ;;.. ~;
. . .'
------~I'+ __ __
_ _ _ ~ . /
{ I +'-- ____ (~
---I.!:;.~ - - - -.....t-'~
EItclrofll
- --- 771X**~\ _ 5-~~~~_.. ~----
/// 1> ~ --0-B ----
.. ---
._
. .f - - [ .. J.?. Y
FIG.3
ONE TYPE OF CARRIER
Metals and degenerate (doped) semiconductors are the example f this type where one
carrier dominates.
The magnetic force on th~ carricr~Lis I-.m ~ ( \ \J I) ,IIlJ i ~ Lllmr~n"Lltl:d h: the II.ill
Field Fh = t! [II, ",here v is the drift veloLity or LmTi~r~ . A ssum ing thL' JIrIXlill1l oj \ a lllll l"
vectors as before.
vx H==Eh
rrom simple reasoning, the current density .I is the charge q multiplied by the number of
carriers traversing unit area in unit time, which is equivalent to the carrier density
multiplied by the drift velocity i.e. J ~ q n v.
Eh v. H
R = (3)
JH q n v H nq
From [his equation . it IS clear that the "ign t,r Ilnll CllcrtiL i cnl l krl:l1d uron Ihl' "i1,!11 11!
the q, this means. in a p - t)-pe spec imen the R \\ouILl he positi\\.' . \\h ik in 11 (\re II
would be negative. Also for a fixed magnetic field and input current. the Iiall Voltagl' i..,
proportional La lIn 0[' its resistivity. When one carrier dominates, the conductivity of the
material is cr = nq)l.
2
flh2 P - fle n
R = (5)
2 (Mh P .J.. Ilc n)2
Sinc the mobilities fJ.h and ~lh are not constants but function of temperature (T) the I: II
coefficient given by Eq. (5). is also a function of T and it may become zero, even change Sl 11.
In general J.!.h > J.!.h so that inversion may happen only if p > n; thus Hall coefficient in
characteristic only of p-type semiconductors.
Al the point of fero Hall coefficient. it possible to determine the ratio of mobilities at,d
Lhl'ir rl' lalivl' CIHll'l'nt ralinn.
Iflus \\l' Cl' th<.1t the Hall Ct)erticienl. in conjunction with re i tivity measurem Ills. I: n
provide information on currier densities. mobilities impurity concentration and other values. II rm st
be noted. however, that mobilities obtained from Hall Effect measurements )..l = R (j do r t
alwa s agree with directly measured values . The reason being that carriers are distribuwd
energy. and those with higher velocities will be deviated to a greater extent for a given field.
)..l we know varies with carrier velocity.
EXPERIMENTAL TECHNIQUE
3. Soldered probe contacts, through very much desirable may disturb the current now (sit ni g
out part of the sample). oldering directly to the body of the sample can affect the sam Ie
properties due to heal and by contamination unle care is taken . These problem l:aIl e
avoided by using pressure contacts an in the present set-ip. The principle draw back or LI is
lype of contacts i that they may be noisy. This problem can. however, be y
keeping the contacts clean and firm.
* ** l30th [·LJ.(3j llnd l:q.(5) havc hecn derived on the assumption that all carriers have same 'cloci
Ihi " i~
IWI trul'. hUI the c,-acl cakli latioll nlnuilics the results ootained here by a factor of' onl) 3
X
The current through th~ sample ~hould not be IurgL' t:l1ough ttl ~iJll ~ l 11I.';]1111g \ Il l rlh~'r
precaution is nel:essary to prevent 'i njecting erkct' from affecting thL' Ilh.:asurenlcnl. he n
good contacts to germanium for example. may have this clTecl. Thi~ Call be lllinilllizcJ by
keeping the voltage drop at the contacts low. I f the surface near (he contacts is rough and
the electric -now in the crystal is low. these injected carriers will recombine before reaching
the measuring probes.
fhe voltage appearing bctween the Iiall Prohes i... not gencl'ally, Ihl.' 11011 \'(lltage alnne .
There are other galvalh)magnctic and tllerlllol11agnct ic clTecl'- ("cart:<.;( L'lk cl. Rhi ghll.'d lll
effect and ElLingshausen etfect) \-,hich can prouucc voltage., bCI\\t:ell Ihe Iiall Prohe ... . 1!1
addition. IR drop due to probe misalignmen t (Lero magnetic field r()t~nlij)l) ,lIlU
thennoelectric voltage due to transverse thermal gradient may be present. All these except.
the Ettingshausen effect are eliminated by the method at averaging four readings.
The Ettingshausell effect is negligible in materials ill which a high thennal conductivity IS
primarily due to lattice conductivity or in which the thermoelectric power is small.
When the voltage between t.he Hall Probes is measured fo r both directions of current. only
the Hall voltage and lR drop reverse. Therefore, the average of these readings eliminates
the influence of the other effect. Further, when Hall Voltage is measured for bother the
directions of the magnetic field. the IR drop does not reverse and may therefore be
eliminated.
rhe Iiall Probe must he rotated in the IiclJ unitl the rn~ltl0n or m::lXlnlUI11 \ (litagl' i...
rcached. Thi s is the po<;ition "hell din:llioll {If lurn:nl ill Ihe rl'llhe ,tilL! 1ll ,1?! IlLI Il' livid
wuuld be perpendicular to each other.
The resistance of the sample changes when the magnetic field is turned on. This phenomena
called magneto-resistance is due to the fact that the drift velocity of all carriers is 110t the
same, with magnetic field on, the Hall voltage compensates exactly the Lorentz force for
carriers with average velocity. Slower carriers \,vill be over compensated and faster one
under compensated, resulting in trajectories that are not along the applied extemal field .
This results in effeclive decrease or the man free path and hence an increase in resistivity.
There, while taking readings with a varying magnetic field at a particular currenl value, it
is necessary that current value should be adjusted. every time. The problem can be
eliminated by using a constant current power supply. which would keep the current
constant irrespective of the resistance of the sample.
4. In ucneral. the resistance or the ample is very high and the Hall Voltages are very 10
I hi: Jl1~<I!l-' llllli rn.lClica ll : lh~re is hurdly any current - not more than few micro amper
Ilh:r'llli' . lll..: Iiall Vu ltagc ' ~ht1L1ld lHlI: be measured ~ ith a high input Impedance l~ I
O\:\- il:e~ -.llch u. el eclrometer. eh:ctron ic Mill iVl) ltmel rs or good potentiometers pr f\:r'lt! y
with lamp and s<.:ak arrangements.
5. Although the dimen ions of the crystal do not appear in the fonnula except the thickne s.
but the theory aSSLImes that all the carrier are moving only lengthwise. Practically it s
been found tl1at a closer to ideal situation may be obtained if the 1 ngth may be tak 'n
three times the width of the crystal.
Ge Single Crystal with four spring type pressure contacts is mounted on a <;unmica
decorated bakelite strip Four lead are provided for connections with measuring devic.es
Ihl' n<td \ :"1I~ or lhicl-.ne~~ <lno res isti\- ily is provided in the le. [ report of the Hall P:q e
((Il ) ~lIpplic.:d \\nh the.: ~l! -lip . I he ~llldent at'll.::r calculating thl:' Iiall Coefficient tram 11 is
c\. p~riment and using. the given value I re ' isl ivit can a!. get valuable infonnalion <lb lit
carrier lien 'ity and carrier mobilities. A typical example is provided in the append,;.... A
further advantage of this type f probe is that the ample an be changed. A run' r
advantage of this type of probe is that the sample can be changed, A minor draw back ortl is
arrangement is that the it may require zero adjustment from time to time. This typ f
probes are specially designed and recommended for Hall Effect Experiment.
(b) HALL PROBE Undillm Arsenide)
Indium arsenite Crystal (Rectangular) is mounted on a phenolic strip with four soldered
contacts for connections with measuring devices . The crystal is covered by a protective
layer of paint. The whole system is mounted in a pen-type case for further protection.
Contacts Soldered .
Hall Voltage 8 - 10 mV / 100 rnA / KG.
The value of the thickness and resistivity of the sample given for lhese probes are not very
reliable as these are 110t given for a specific probe and may vary from probe as is usually the
case with all semiconductor devices. These are essentially meLlll1 to ne used as tran~Juccr".
Intersil 3 12 Digit single chip AID Converter leL - 7107 have been used . It has high
accuracy like, auto zero to less than 10 J.1V, zero drift less than 1 ~lvfc, input bias
current of 10 pA and rollover error of less than one count. Since the use of internal
reference causes the degradation in performance due to internal heating, an external
reference has been used. This voltmeter is much more convenient to use in Hall Expe riment.
because the input of either polarity can be measured.
SPECIFICATJONS
Range o 200.0. I1lVl)lr.
Resolution 100, J.1Voll.
Accuracy ± 0.1% of reading ± I Digit.
Impedance I, M ohm.
Special Features Auto zero and polarity indicator.
Overload Indicator Sign of I on the left and blanking of other digits.
I
I
3, (a) ELECTROMAGNET, MODEL- EMU - 75,
Field Intensity 12,500 ± 5% Gauss in an air-gap of 10 mm. Air-g p is
cOlltinusouly variable upto 100 mm with two way kn9bb d
wheel screw adju ting system. .. .
7).mm diameter. ormally nat faced pole pieces are . lIppll d
\\ Itll the l11aend.
I nginet:ring Coils r\\ 0 each c~i I is w und on non-magnetic formers and ha a
re istance oC12. ohms. Approx.
Yoke Material Mild teel.
Power Requiremen t 0-5 Amps. Connecting variable
The current regulation circuit IS It l:olllrolkJ anJ hence result ill the highe ... l ljllillil.'
of performance. Matched power transisto rs are used to share the load current. The
supply is protected against transients caused by the inductive load of the magnet.
SPECIFICATIONS
Current 0-4. Amp. (or as desired)
Regulation (Line) ± O.I % fo r 10% rna ins variation.
Regulation (Load) ± 0.1% tor load resistance variation from 0 to ruliload.
Protected Electronically protected against overload or short circuil.
Metering 3 Y2 Digit. 7 segmen t LCD Digital Panel Meter.
A calibration chart (current Vs. magnetic field) is supplied. which eliminate the need
of the Gauss Meter when suppl ied \\lith EMU· SO .
SPECIFICA TIONS
Current 0 - 2.5, Amp. (or as desired)
Regulation (Line) ± 0. 1% for No load to fuJI load .
Regulation (Load) ± 0. 1% for ± 10°1t, mains variation ..
Protection I::lectronically protected against overload or shan circu iting.
Metering 3 Y2 Digit. 7 segment LCD Digital Panel Meter.
Range 220V ± 10% 50 Hz
A calibration chart (current Vs. magnetic field) is supplied, which elim inate the need
of the Gauss Meter when supplied with EMU- 35 .
./
DIGITAL GUASS METER. MODI I 1)(,M lOll .
. The Gauss Meter operates on the principle of Hall Hfect in semiconductors. A sem iconductor
material can·ying current develops an electro-motive force. when placed in a magnetic fiel d, in a
direction perpendicular to the direction of both electric current and magnet ic field . The magnitude
or this e.m.r. is proportional to the field intensity if the current is kept COllst(lnt. this e.m.r. is called
the Hall Voltage. This small Hall Voltage is amplified through a high stability amplilier so that a
Millivoltmeter connected at the output of the amplifier can be calibrated directly ill Magnetic fie ld
unit (Gauss).
SPECIFICATIONS
Range 0- 2, K Gauss and 0 - 20 k Gauss.
Resolution 1 Gauss at 0 - 2 K Gauss Range.
Accurac) ± 0.51%.
Disp lay J liz Digit. 7 segment LED.
Detector Hall Probe with an Imported Hall Element.
Power 220 Volt, 50 Hz.
Special Indicates the direction of the magnetic field .
PROCEDLRE
I (OIl11\.:Cl lilt: \\ i Jtll\\ i ~e conlact~ or
the Ilall Probe to the terminals niarked "Voltage" i.I d
kllgt!1,\ bt: contacts to terminals marked ··Current" .
2. ~witch 'ON' the Hall Effect Set -'Up and adJustment current (sa5' few mA).
3. S,vitch over the di play to v.oltage side. There may be some voltage reading
the magnetic flela . This is due 16 imperfect-alignment ollhe four contacts
Probes and is ' generally known as the 'Zero field Potential '. In case its valu" is
comparable to the J lall Voltage it should be adjusted to a mmimum possible (for H II
Probe "Ge" only). In all cases, this error should b subtracted from the Hall Volta te
reading.
4. Now place th probe in the magnetic field as shown in fig.3 and s,,' itch_ l n tle
electromagnet power supply and adjust the current to any desireavalue: Rotate the H II
Probe till it become perpendicular 0 magnetic field. Hall voltage will be maximum ifl II i
adjustment.
5. Measure Hall voltage for both the directions of the current and magnetic field (i.e. f'flr
(lh,,~rv;tli(lI1' ror d p::lflicular value I' current and magnetic field).
\k:l ~lIrc [ ill: ! 1 ~1I 1 Vnltagc U~ a function or current keeping the magnetic. lield constant POL
.? a g rapl1 .
7 Mea~urc the Iiall yoltage as a !'unction of magnetic ~ keeping a slIitl;lble valu<' )f
/ current as constant. Plol gr~ph..:.>
8. Measure, the magnetic rield by the Gauss Meter.
CALCULATIONS :
I. From th graph Hall Voltage Vs. Man netic tield calculated Hall coeflicient.
2. Delermine the type of majority charge carriers. i.e. whether the crystal in
3. Calculate charge carrier den ity from the relation.
=;. n
nl! Rq
other method (Four Probe Method). Typical calculations are shown in appendix.
"';~
APPENDIX
~lll1pk c.licuiuli(lll for Iiall Co~ITicienL Carrie! Density and Mobility Laking Hall Probe .
1l5( 1() ch :--.1111 f1 k.
SAMPLE DETAILS
R
IH
53 X 10-3 X 5 X 10-
8 x 10 3 X 103
or = 33 X 10- 5 X 10 8 cm 3 coulomb
R = ~ n =
nq Rq
=
33 X 10 3 X 1.6 X 10 )9
= 1.9 X 10 14 m- 3
(iii) CARRIER MOBILITY
For degenerate semiconductor i.e. when one carrier dominates.
Carrier Mobility ~t == Rcr
33 X 10-3 X 0.1
Thus we 'ice that llall Coefficient ill conjullI.:linn \\ ilh re"i~ti \ il) 1l1C,NlrCl11l'lll can 111'(\\ iJc
1'1Iabie information on carrier density, mobilities and other valucs It must be noted hO\\c\el. Ihal
I1Jpbilities obtained from Hall Effect Measurement ~ := Rcr do not always agree with directly
measured values. The reason IS explained in the booklet.
OBSERVATION TABLE
il. CONSTANT.............Gauss.
rORWARDED RI VI'R'iI:
l(mA) -Y V(mV) I (m .\) \ (Ill \ )
2. CURRENT CONSTANT..... ... ..... mA.
:
I
II