COMPUTER IMPLEMENTED METHOD FOR CALCULATING A CHARGE DENSITY AT A GATE INTERFACE OF A DOUBLE GATE TRANSISTOR
A computer implemented method for calculating a charge density q1 of a first gate of a double gate transistor comprising a thin body with a first and a second gate interface, the method including determining, using a physical processor, an initial estimate q1,init of the charge density of the first gate; performing, using the physical processor, at least two basic corrections of the initial estimate based on a Taylor development of a function fzero(q1) able to be nullified by a correct value of the charge density q1 of the first gate.
This invention relates to a computer implemented method for determining a charge density in a front and/or back gate of a double gate transistor comprising a thin body having a front gate interface and a back gate interface. The charge density q_{1 }is linked to the surface potential x_{1 }through a linear dependence q_{1}=x_{g1}−x_{1}, where x_{g1 }is the gate electrostatic potential normalized to the thermal voltage of the double gate transistor. Consequently, the method also enables to determine the surface potential at a gate interface of a double gate transistor.
BACKGROUNDUltrathin body transistors such as FinFETs or UltraThin Body and Box (UTBB) MOSFETs are required for sub20 nm nodes because of their excellent electrostatic integrity and reduced variability. Compared to FinFET, UTBB technology on thin buried oxide presents two decisive benefits: a much simpler process than FinFET and the possibility to use the backplane bias to optimize the power consumption and speed tradeoff at circuit level. To take full advantage of this latter benefit, circuit designers need compact models that describe properly the transistor behavior for a wide range of back bias.
S. Khandelwal, Y. S. Chauhan, D. D. Lu, S. Venugopalan, M. U. A. Karim, A. B. Sachid et al, “BSIMIMG:A compact model for ultrathinbody SOI MOSFETs with backgate control”, IEEE Transactions on Electron Devices, vol. 59, pp. 20192026, 2012 and O. Rozeau, M. A. Jaud, T. Poiroux, M. Benosman, “Surface potential based model of ultrathin fully depleted SOI MOSFET for IC simulations”, IEEE International SOI conference, 2011, have disclosed such compact models describing the transistor behavior. However, in such models, the interface between the body and the buried oxide is assumed always depleted, which provides correct results in reverse and low forward back bias (FBB) range. However, when a strong FBB is applied, inversion occurs first at the back interface, which has a significant impact on device characteristics. Consequently, the previous models are not realistic when a strong FBB is applied.
SUMMARYAn aspect of the invention provides a more realistic model enabling to determine the charge densities in the front and/or back of a double gate transistor, which provides correct results in reverse, low forward back bias but also in strong forward back bias.
Another aspect of the invention provides a full analytical calculation method of charge density that relies on very few error correction steps from a unique equation.
Another aspect of the invention provides an analytical method enabling to determine the charge densities of the front and back gates of a wide range of transistors, which provide correct results for a wide range of device geometries from thick buried oxide fullydepleted to independent double gate transistors.
To that purpose, a first aspect of the invention is directed to a computer implemented method for calculating a charge density q_{1 }of a first gate of a double gate transistor comprising a thin body with a first and a second gate interface, the method comprising:

 determining, using a physical processor, an initial estimate or value q_{1,init }of the charge density of the first gate;
 performing, using the physical processor, at least two basic corrections of the initial estimate or value based on a Taylor development of a function f_{zero}(q_{1}) able to be nullified by a correct value of the charge density q_{1 }of the first gate.
This double gate transistor 10 comprises a source 11, a drain 12 and a thin body film 13 linking the source 11 to the drain 12. The double gate transistor further comprises a front gate dielectric layer 14 and a back gate dielectric layer 15. The front gate dielectric layer 14 is in contact with a front gate 16 to which a front bias V_{g1 }may be applied. The back gate dielectric layer 15 is in contact with a back gate 17 to which a back bias V_{g2 }may be applied.
The method enables to determine:

 the surface potential at the interface 18 between the thin body film 13 and the front gate dielectric layer 14, referred to herein as “the front gate interface 18”, and/or the charge density in the front gate 16, and/or
 the surface potential at the interface 19 between the thin body film 13 and the back gate dielectric layer 15, referred to herein as “the back gate interface 19”, and/or the charge density in the back gate 17.
In the following detailed description, the charge density of the front gate is determined with a method according to an embodiment of the invention represented on
As shown in
It will be appreciated by one skilled in the art that the method of
Input Parameters:
The method comprises first a step 101 of receiving, using the interface, a set of input parameters describing the effective geometry of the double gate transistor 100 and the effective biases V_{g1 }and V_{g2 }on the electrodes. The set of input parameters received by the interface 201 in an embodiment is summarized in Tables 1 and 2 below.
It will be appreciated that the list of the above input parameters is not limiting. Indeed, additional (or fewer) input parameters can be used in other embodiments of the invention.
The method may then comprise a step of calculation of normalized quantities by the computer 200. The normalized quantities are calculated in the embodiment using code instructions embedded in the memory 203. The processor 202 determined the normalized quantities by executing the code instructions. Even if this step is not mandatory, it enables to ease the writing of the equations.
Normalized front gate potential: x_{g1}=(V_{g1}−ΔΦ_{m1})/(k_{B}T/q)
Normalized back gate potential: x_{g2}=(V_{g2}−ΔΦ_{m2})/(k_{B}T/q)
Normalized source or drain potential: X_{n}=V_{x}/(k_{B}T/q)
With k_{B }the Boltzmann constant, T the device temperature and q the elementary charge.
Normalized front gate capacitance: k_{1}=(ε_{ox1}/T_{ox1})/(ε_{ch}/T_{Si})
Normalized back gate capacitance: k_{2}=(ε_{ox2}/T_{ox2})/(ε_{ch}/T_{Si})
With ε_{ox1 }(resp. ε_{ox2}) the front (resp. back) gate dielectric permittivity and ε_{ch }the thin body dielectric permittivity.
Initial Estimate or Value:
The method comprises then a step 102 of determining, using the physical processor 202, an initial estimate or value q_{1,init }of the first gate charge density, linked to the surface potential at the first gate interface x_{1,init }through: q_{1,init}=x_{g1}−x_{1,init}.
The calculation of an estimate of the first gate charge density can be carried out as follows using code instructions embedded in the memory 203. The processor 202 determined the estimate of the first gate charge density by executing the code instructions embedded in the memory. This step 102 comprises first a step of calculating, using the physical processor, saturation values of front (x_{1,SI}) and back (x_{2,SI}) interface potentials when both interfaces are in weak inversion and when both interfaces are in strong inversion. The step 102 comprises then a step of using a smoothing function between the values at weak and strong inversion to determine the initial guess.
The calculation of the saturation values of front and back interface potentials gives:
n_{i }is the intrinsic carrier density in the thin body, and
Using the physical processor 202, front (x_{1,0}) and back (x_{2,0}) interface potentials can be estimated, neglecting the interface decoupling effect that occurs when strong inversion takes place at one of the interfaces:
The MIN_SMOOTH function may be the following function:
MIN_SMOOTH(a,b)=a−3 ln(1+e^{(a−b)/3})
But it could also be for example in an embodiment:
with δ, for example, between 10 and 20

 Using the physical processor 202, the initial estimate of front (q_{1,init}) and back (q_{2,init}) normalized gate charge densities are determined, accounting for interface decoupling effect:
First Basic Correction Step:
The method comprises then a first basic correction step 103 wherein the initial guess q_{1,init }is corrected, using a Taylor development of a function f_{zero}(q_{1}) able to be nullified (i.e. nullifiable) by a correct value of the charge density q_{1 }at the front gate interface. Beneficially, the Taylor development is, in an embodiment, a second order Taylor development.
Determination of the Function F_{zero}(q_{1}) Able to be Nullified by a Correct Value of the Gate Charge Density q_{1 }at the Front Gate Interface:
The function f_{zero}(q_{1}) is determined thanks to 1D Poisson equation and boundary conditions at the z interfaces between the thin body and the front and back gate dielectrics. That function is determined using code instructions embedded in the memory 203.
As a matter of fact, assuming an undoped channel, Poisson equation is:
Boundary conditions at the front interface are given by:
Boundary conditions at the back interface are given by:
Moreover, thanks to charge conservation, the following equation is obtained:
Q_{g1}(y)+Q_{g2}(y)+Q_{inv}(y)=0 (3.4)
Using a first integration of Poisson equation and the boundary conditions, the following equations are obtained:
Q(y)^{2}=Q_{g1}(y)^{2}−2qn_{i}ε_{Si}φ_{t}e^{(ψ}^{s1}^{(y)−φ}^{m}^{(y))/φ}^{i } (3.5)
Q(y)^{2}=Q_{g2}(y)^{2}−2qn_{i}ε_{Si}φ_{t}e^{(ψ}^{s2}^{(y)−φ}^{m}^{(y))/φ}^{i } (3.6)
Using a second equation integration of Poisson equation and boundary conditions, the following equation is obtained:
Q(y) is a quantity homogeneous to a charge that is either real, in hyperbolic mode, or imaginary, in trigonometric mode, whose sign is not defined a priori. In the hyperbolic mode, Q is closely linked to the transverse electric field within the channel, while, in the trigonometric mode, interfaces are essentially decoupled and Q is an imaginary number.
Using dimensionless parameters, boundary conditions (3.2) and (3.3) give:
q_{1}=x_{g1}−x_{1 } (3.8)
q_{2}=x_{g2}−x_{2 } (3.9)
Furthermore, the following equation is defined:
Equations (3.4) to (3.7) become respectively:
Equation (3.14) presents several benefits: first, the term qcoth(q/2), which becomes lm(q)cot(lm(q)/2) in the trigonometric mode, can be considered as a function of q^{2}, and it ensures a natural and smooth transition from hyperbolic (q^{2}>0) to trigonometric (q^{2}<0) regions. Since is it an even function in q, the chosen sign for q doesn't matter and we can rewrite (3.14) as:
The three unknowns to be determined from equations (3.12), (3.13) and (3.15) are now q_{1}, q_{2 }and q^{2}, all real numbers.
Nevertheless, using equation (3.15) instead of (3.14), it is no longer ensured that the solution obtained from the set of equations (3.12), (3.13) and (3.15) is necessary the physically correct one. Indeed, without using the cot^{−1 }function, there is a risk to end up with mathematically correct solutions that correspond to cases where the transverse electric field goes one or more times to ±∞ within the silicon film. Thus, it is desirable to pay careful attention to this point during the calculation. In particular, the condition q^{2}>−4π^{2 }should be verified.
In order to calculate analytically the surface potentials, or equivalently the charge densities, from a given initial estimate or guess, a function f_{zero }of a unique variable q_{1 }is determined, from which the small corrections are calculated to be brought to this variable. To ensure the stability of the searched function i.e. a behavior that is as close to a linear dependence on q_{1 }as possible, it is desirable to avoid the use of exponential functions of q_{2}. To find first a convenient explicit expression of q_{2 }as a function of q_{1 }and q_{2}, equation (3.15) is expressed in three different equivalent forms:
From (3.16) and (3.17) together with (3.12) and (3.13), we find:
Then, this last equation with (3.18) gives:
From equation (3.20), q_{2 }is expressed as a function of q_{1 }and q^{2}:
Considering q^{2 }as a function of q_{1 }thanks to equation (3.12), equation (3.21) defines q_{2 }as a function of q_{1}.
Then, the inversion charge density as a function of q_{1 }is obtained from the charge conservation equation (3.11) and the function to be solved is obtained from (3.16) and (3.12):
Consequently, the following calculation sequence is obtained:
First Basic Correction Step:
The first correction step 103 comprises a step of calculation of a value of q^{2 }(noted q_{sq,1}) as a function of q_{1,init}:
q_{sq,1}=k_{1}^{2}q_{1,init}^{2}−A_{0}e^{−x}^{n}e^{x}^{g2}^{−q}^{1,init }
The first correction step 103 comprises then a step of calculation of a value of q_{2 }(noted q_{2,1}) as a function of q_{1,init}:
The first correction step 103 comprises then a step of calculation of the value of the function f_{zero }to be nullified as a function of q_{1,init}:
The first correction step 103 comprises then a step of analytical calculation, using the physical processor 202, of the 1^{st }and 2^{nd }derivatives, in the case of a second order Taylor development, of f_{zero }with respect to q_{1 }at q_{1}=q_{1,init }from the above equations.
The first correction step 103 comprises then a step of calculation of the corrected value of q_{1 }(noted q_{1,1}):
First Intermediate Correction Step
The method comprises then a step 104 of determining if an intermediate correction step is desired or not before the second basic correction step. To that purpose, the method uses a parameter D=x_{g1}−x_{1,SI}+ln(10k_{1}) representative of the double gate transistor such that, when the charge density of the first gate is inferior to the parameter D, the first interface is in a strong inversion regime.
The step 104 comprises then first a step of calculating, using the physical processor 202, the parameter D.
The step 104 comprises then a step of comparing, using the physical processor 202, the first corrected value q_{1,1 }obtained after the first basic correction with the parameter D.
If the first corrected value q_{1,1 }obtained after the first basic correction step is inferior to the parameter D, then the method comprises a step 105 of performing a first intermediate correction on the first corrected value q_{1,1}.
The first intermediate correction step 105 comprises first a step of estimating a second value of q^{2 }(noted q_{sq,2}) from q_{1,1 }and q_{2,init }given by:
The first intermediate correction step 105 comprises then a step of calculating, using the physical processor, a first intermediate value:
If the strong inversion condition is not fulfilled:
q_{1,2}=q_{1,1 }
Second Intermediate Correction Step
The method comprises then a step 106 of comparing the value of q_{1,2 }obtained after the first intermediate correction step with the parameter D to determine if a second intermediate correction step is desired.
Consequently, the second intermediate step is performed only if x_{g1}−q_{1,2}>x_{1,SI}−ln(10k_{1}), i.e. if the strong inversion condition is fulfilled.
The second intermediate step 107 comprises first a step of estimating, using the physical processor 202, a third value of q^{2 }(noted q_{sq,3}) from q_{1,2 }and q_{2,init}:
The second intermediate step 107 comprises then a step of refining the third value of the q^{2 }estimation (noted q_{sq,3b}):
The second intermediate step 107 comprises then a step of calculating a second intermediate corrected value of q_{1 }(noted q_{1,3}) if the strong inversion condition, specified by x_{g1}−q_{1,2}>x_{1,SI}−ln(10k_{1}), is verified:
If the strong inversion condition is not fulfilled:
q_{1,3}=q_{1,2 }
Second and Third Basic Correction Steps
The method further comprises a second and a third basic correction steps 108 and 109. Each of these steps are carried out as the first one.
For the second basic correction step 108, the input value of q_{1 }is q_{1,3 }instead of q_{1,init}. The output of the second basic correction step is used as the input value of the third basic correction step to find the final value of q_{1}:q_{1,final}.
Finally are calculated the value of q_{2 }(q_{2,final}) from the abovementioned equations, and the normalized surface potentials are given by:
x_{1,final}=x_{g1}−q_{1,final }
x_{2,final}=x_{g2}−q_{2,final }
The above calculations are carried out, and equations are determined, using code instructions embedded in the memory 203. Those code instructions are executed by the processor 202 to carry out the desired calculation.
Having described and illustrated the principles of the invention with reference to various embodiments, it will be recognized that the various embodiments can be modified in arrangement and detail without departing from such principles. It should be understood that the programs, processes, or methods described herein are not related or limited to any particular type of computing environment, unless indicated otherwise. Various types of specialized computing environments may be used with or perform operations in accordance with the teachings described herein. Elements of embodiments shown in software may be implemented in hardware and vice versa.
The devices, processors or processing devices described herein may be configured to execute one or more sequences of one or more instructions contained in a main memory or a computer readable medium. Execution of the sequences of instructions contained in a main memory or a computer readable medium causes the processor to perform at least some of the process steps described herein. One or more processors in a multiprocessing arrangement may also be employed to execute the sequences of instructions contained in a main memory or a computer readable medium. In alternative embodiments, hardwired circuitry may be used in place of or in combination with software instructions. Thus, embodiments are not limited to any specific combination of hardware circuitry and software.
The term “computer readable medium” as used herein refers to any physical medium that participates in providing instructions to a processor for execution. Such a medium may take many forms, including but not limited to, nonvolatile media, volatile media, and transmission media. Nonvolatile media include, for example, optical or magnetic disks. Volatile media include dynamic memory. Transmission media include coaxial cables, copper wire and fiber optics. Common forms of computerreadable media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, any other magnetic medium, a CDROM, DVD, any other optical medium, punch cards, paper tape, any other physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASHEPROM, any other memory chip or cartridge, or any other medium from which a computer can read.
Various forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to processor for execution.
While the present invention has been particularly described with reference to the preferred embodiments, it should be readily apparent to those of ordinary skill in the art that changes and modifications in form and details may be made without departing from the scope of the invention. For example, we have presented here the typical procedure followed to obtain the interface potentials. In order to ensure the numerical robustness of the calculations in all geometry and bias configurations, some equations can be used in a slightly different way or in a slightly different order in certain cases. For example, a 3^{rd }order Taylor expansion of some equations may be used for q^{2 }close to 0 in order to avoid division by 0.
Claims
1. A computer implemented method for calculating a charge density q1 of a first gate of a double gate transistor comprising a thin body with a first and a second gate interface, the method comprising:
 determining, using a physical processor, an initial estimate q1,init of the charge density of the first gate;
 performing, using the physical processor, at least two corrections of the initial estimate based on a Taylor development of a function fzero(q1) that is nullifiable by a correct value of the charge density q1 of the first gate to determine a corrected value of the charge density q1 of the first gate of the double gate transistor.
2. The computer implemented method according to claim 1, wherein the Taylor development is a second order Taylor development.
3. The computer implemented method according to claim 1, comprising performing, using the physical processor, three corrections of the initial estimate based on a Taylor development of a function fzero(q1) that is nullifiable by a correct value of the charge density q1 of the first gate.
4. The computer implemented method according to claim 1, wherein each correction comprises calculating, using the physical processor, an intermediate value qsqj as a function of a value q1,j−1 obtained in a previous step of the method, the intermediate value being given by: A 0 = 2 qn i T Si 2 ɛ ch φ i, ni being the intrinsic carrier density in a thin body, TSi being the effective thin body film thickness, εch being the thin body dielectric permittivity, φ1 being the thermal potential of the double gate transistor, q being the elementary charge.
 qsq,j=k12q1,j−12−A0e−xnexg1−q1,j−1
 xg1 being the normalized first gate potential;
 xn being the normalized source or drain potential;
 k1 being the normalized first gate capacitance;
5. The computer implemented method according to claim 4, wherein each correction further comprises calculating, using the physical processor, a value q2j of the charge density of the second gate given by: if q sq, j > 0 : q 2, j = x g 2  x g 1 + q 1, j  1 + 2 ln ( k 1 q 1, j  1 + q sq, j coth ( q sq, j 2 ) )  ln ( q sq, j sinh ( q sq, j / 2 ) 2 ) if q sq, j < 0 : q 2, j = x g 2  x g 1 + q 1, j  1 + 2 ln ( k 1 q 1, j  1 +  q sq, j cot (  q sq, j 2 ) )  ln (  q sq, j sin (  q sq, j / 2 ) 2 ) A 0 = 2 qn i T Si 2 ɛ ch φ i, ni being the intrinsic carrier density in a thin body, TSi being the effective thin body film thickness, εch being the thin body dielectric permittivity, φt being the thermal potential of the double gate transistor, q being the elementary charge.
 xg1 being the normalized first gate potential;
 xg2 being the normalized first gate potential;
 xn being the normalized source or drain potential;
 k1 being the normalized first gate capacitance;
 k2 being the normalized second gate capacitance;
 q1 being the charge density of the first gate;
 q2 being the charge density of the second gate;
6. The computer implemented method according to claim 5, wherein each correction further comprises determining, using the physical processor, a value for the function fzero(q1) that is nullifiable by a correct value of the charge density q1 of the first gate, fzero(q1) being given by: if q sq, j > 0 : f zero ( q 1 ) = ( k 1 q 1, j  1 + q sq, j coth ( q sq, j 2 ) ) ( k 1 q 1, j  1 + k 2 q 2, j )  A 0  x n x g 1  q 1, j  1 if q sq, j < 0 : f zero ( q 1 ) = ( k 1 q 1, j  1 +  q sq, j cot (  q sq, j 2 ) ) ( k 1 q 1, j  1 + k 2 q 2, j )  A 0  x n x g 1  q 1, j  1
7. The computer implemented method according to claim 6, wherein each correction further comprises calculating, using the physical processor, a first and second derivatives of fzero with respect to q1 at q1=q1j−1.
8. The computer implemented method according to claim 7, wherein each correction further comprises calculating, using the physical processor, a corrected value q1j of charge density of a first gate given by: q 1, j = q 1, j  1  f zero f zero q 1  f zero 2 2 f zero q 1 2 / f zero q 1
9. The computer implemented method according to claim 1, wherein determining, using the physical processor, the initial estimate q1,init of the charge density of the first gate comprises:
 determining, using the physical processor, a first saturated value of the charge density of the first gate when the first gate interface is in a weak inversion regime;
 determining, using the physical processor, a second saturated value of the charge density of the first gate when the first gate interface is in a strong inversion regime;
 determining, using the physical processor, the initial estimate of the charge density of the first gate by using a smoothing function between the first and the second saturated value.
10. The computer implemented method according to claim 1, further comprising, between the first and the second corrections:
 determining, using the physical processor, a parameter D representative of the double gate transistor such that, when the charge density of the first gate is inferior to the parameter D, the first interface is in a strong inversion regime;
 comparing, using the physical processor, the first corrected value q1,1 obtained after the first correction with the parameter D.
11. The computer implemented method according to claim 10, wherein if the first corrected value q1,1 is inferior to the parameter D, the method further comprises, between the first and the second corrections:
 performing, using the physical processor, a first intermediate correction of the first corrected value q1,1 to obtain a second corrected value q1,2, the first intermediate correction using: an initial estimate of the charge density of the second gate q2,init; and a first intermediary value qsq,2 issued from a first order Taylor expansion of a first intermediary function f1boundary obtained by a first integration of Poisson's equation and by some boundary conditions at the first gate interface.
12. The computer implemented method according to claim 11, wherein the first intermediate correction comprises: q sq, 2 = b 2  4 ac  b 2 a where a = 1 + ( 1 6  1 π 2 ) ( k 1 q 1, 1 + k 2 q 2, init ), b = 4 π 2 + k 1 q 1, 1 k 2 q 2, init + 2 ( 1 + π 2 3 ) ( k 1 q 1, 1 + k 2 q 2, init ) and c = 4 π 2 k 1 q 1, 1 k 2 q 2, init + 8 π 2 ( k 1 q 1, 1 + k 2 q 2, init ) q 1, 2 = q 1, 1  1 + 2 k 1 2 q 1, 1 / ( k 1 2 q 1, 1 2  q sq, 2 ) q 1, 1  x g 1 + x n + ln ( ( k 1 2 q 1, 1 2  q sq, 2 ) / A 0 ) A 0 = 2 qn i T Si 2 ɛ ch φ i, ni being the intrinsic carrier density in a thin body, TSi being the effective thin body film thickness, εch being the thin body dielectric permittivity, φt being the thermal potential of the double gate transistor, q being the elementary charge.
 estimating, using the physical processor, a first intermediary value qsq,2 from the first corrected value q1,1 and the initial guess of the charge density of the second gate q2,init:
 calculating, using the physical processor, the second corrected value q1,2 from the first corrected value q1,1 and the first intermediary value qsq,2:
 xg1 being the normalized first gate potential; xn being the normalized source or drain potential; k1 being the normalized first gate capacitance;
13. The computer implemented method according to claim 12, further comprising, between the first and the second corrections:
 comparing, using the physical processor, the second corrected value q1,2 with the parameter D.
14. The computer implemented method according to claim 13, wherein, if the second corrected value q1,2 is inferior to the parameter D, the method further comprises, between the first and the second corrections:
 performing, using the physical processor, a second intermediate correction of the second corrected value q1,2 to obtain a third corrected value q1,3, the second intermediate correction using: an initial estimate of the charge density of the second gate q2,init; and a third intermediary value q1,3 issued from a first order Taylor expansion of a second intermediary function f2boundary obtained by a second integration of Poisson's equation and by some boundary conditions at the first gate interface.
15. The computer implemented method according to claim 14, wherein the second intermediate analytical correction comprises: q sq, 3 = b 2  4 ac  b 2 a where a = 1 + ( 1 6  1 π 2 ) ( k 1 q 1, 2 + k 2 q 2, init ), b = 4 π 2 + k 1 q 1, 2 k 2 q 2, init + 2 ( 1 + π 2 3 ) ( k 1 q 1, 2 + k 2 q 2, init ) and c = 4 π 2 k 1 q 1, 2 k 2 q 2, init + 8 π 2 ( k 1 q 1, 2 + k 2 q 2, init ) q sq, 3 b = q sq, 3  k 1 q 1, 2 k 2 q 2, init + q sq, 3 + ( k 1 q 1, 2 + k 2 q 2, init ) α 1 + ( k 1 q 1, 2 + k 2 q 2, init ) ( q s, 3 + 2 α  α 2 ) / ( 4 q sq, 3 ) with α = q sq, 3 coth ( q sq, 3 2 ) if q sq, 3 > 0 and α =  q sq, 3 cot (  q sq, 3 2 ) if q sq, 3 < 0. q 1, 3 = q 1, 2  1 + 2 k 1 2 q 1, 2 / ( k 1 2 q 1, 2 2  q sq, 3 b ) q 1, 2  x g 1 + x n + ln ( ( k 1 2 q 1, 2 2  q sq, 3 b ) / A 0 ) A 0 = 2 qn i T Si 2 ɛ ch φ i, ni being the intrinsic carrier density in a thin body, TSi being the effective thin body film thickness, εch being the thin body dielectric permittivity, φt being the thermal potential of the double gate transistor, q being the elementary charge.
 estimating, using the physical processor, a second intermediary value qsq,3 from the second corrected value q1,2 and the initial guess of the charge density of the second gate q2,init:
 refining, using the physical processor, the second intermediary value qsq,3 to obtain a third intermediary value qsq,3b such that:
 calculating, using the physical processor, the third corrected value q1,3 from the second corrected value q1,2 and the third intermediary value qsq,3b such that:
 xg1 being the normalized first gate potential; xn being the normalized source or drain potential; k1 being the normalized first gate capacitance;
16. The computer implemented method according to claim 10, wherein the parameter D is calculating, using the physical processor, by: D = x g 1  ( x n + ln ( 4 k 1 + k 2 / ( 1 + k 2 ) A 0 ) + ln ( Δ x / 2 th ( Δ x / 2 ) ) + 3 ) + ln ( 10 k 1 ), Δ x = ln ( 1 + k 1 1 + k 2 ) A 0 = 2 qn i T Si 2 ɛ ch φ i, ni being the intrinsic carrier density in a thin body, TSi being the effective thin body film thickness, εch being the thin body dielectric permittivity, φt being the thermal potential of the double gate transistor, q being the elementary charge.
 xg1 being the normalized first gate potential,
 xn being the normalized source or drain potential;
 k1 being the normalized first gate capacitance;
 k2 being the normalized second gate capacitance;
17. A nontransitory computer storage medium encoded with code instructions for implementing a method for calculating a charge density q1 of a first gate of a double gate transistor comprising a thin body with a first and a second gate interface, the method comprising:
 determining, using a physical processor, an initial estimate q1,init of the charge density of the first gate;
 performing, using the physical processor, at least two corrections of the initial estimate based on a Taylor development of a function fzero(q1) that is nullifiable by a correct value of the charge density q1 of the first gate to determine a corrected value of the charge density q1 of the first gate of the double gate transistor.
18. The nontransitory computer storage medium according to claim 17, wherein determining, using the physical processor, the initial estimate q1,init of the charge density of the first gate comprises:
 determining, using the physical processor, a first saturated value of the charge density of the first gate when the first gate interface is in a weak inversion regime;
 determining, using the physical processor, a second saturated value of the charge density of the first gate when the first gate interface is in a strong inversion regime;
 determining, using the physical processor, the initial estimate of the charge density of the first gate by using a smoothing function between the first and the second saturated value.
19. A physical computer comprising:
 a physical interface for receiving input parameters;
 a nontransitory computer storage medium encoded with code instructions for calculating a charge density q1 of a first gate of a double gate transistor comprising a thin body with a first and a second gate interface, and
 a physical processor for executing the code instructions, wherein, when the code instructions are executed by the processor, the processor determines an initial estimate q1,init of the charge density of the first gate, and performs at least two corrections of the initial estimate based on a Taylor development of a function fzero(q1) that is nullifiable by a correct value of the charge density q1 of the first gate to determine a corrected value of the charge density q1 of the first gate of the double gate transistor.
Type: Application
Filed: Jul 16, 2014
Publication Date: Jan 21, 2016
Inventors: Thierry POIROUX (Voiron), MarieAnne JAUD (Claix), Sebastien MARTINIE (Saint Martin le Vinoux), Olivier ROZEAU (Moirans)
Application Number: 14/332,907