lumerical-定向耦合器2-寄生耦合

Abstract：本文在前一节的基础上，通过改变波导二的宽度，实现不同宽度波导间的耦合，比较了用全矢量法和耦合模理论得到的耦合效率。接着比较了不同宽度和相同宽度波导的传输效率。

参考文献：

硅光子设计4.1节

全矢量法计算

在上一节模型的基础上，改变波导二的宽度为0.4um

传输效率

switchtolayout;

select("waveguide2");set("y max",0.6*1e-6);

wavelength=1.55e-6;

switchtolayout;#发射波导

select("waveguide1");set("enabled",1);

select("waveguide2");set("enabled",0);

findmodes;

copydcard("mode1","modeA");

BetaA=1e-6*(2*pi/wavelength)*real(getdata("mode1","neff"));

switchtolayout;

select("waveguide1");set("enabled",0);

select("waveguide2");set("enabled",1);

findmodes;

copydcard("mode1","modeB");

BetaB=1e-6*(2*pi/wavelength)*real(getdata("mode1","neff"));

switchtolayout;#发射波导波导系统中的两个超模

select("waveguide1");set("enabled",1);

select("waveguide2");set("enabled",1);

findmodes;

Beta1=1e-6*(2*pi/wavelength)*real(getdata("mode1","neff"));

Beta2=1e-6*(2*pi/wavelength)*real(getdata("mode2","neff"));

AB1A=overlap("mode1","modeA");

AB2A=overlap("mode2","modeA");

coeff1=sqrt(AB1A(2))/sqrt(AB1A(2)+AB2A(2));

coeff2=sqrt(AB2A(2))/sqrt(AB1A(2)+AB2A(2));

L=((2*pi)/abs(Beta2-Beta1))*[0:0.001:10];

o1=ones(length(L),1);

p1=o1*(abs(coeff1)^4+abs(coeff2)^4)+2*abs(coeff1)^2*abs(coeff2)^2*cos((Beta2-Beta1)*L);

p2=o1-p1;

plot(L,p1,p2,"distance","T");

耦合模理论

cc=abs(Beta2-Beta1)/2;

temp1=1+((BetaA-BetaB)/2/cc)^2;

kappa2=sin(cc*L*sqrt(temp1))^2/temp1;

t2=o1-kappa2;

plotxy(L,p1,L,p2,L,kappa2,L,t2,"distance","T");

与全矢量法进行比较，CMT大大高估了不同波导的光耦合。

相同波导

switchtolayout;

select("waveguide2");set("y max",0.7e-6);#更改第二个波导宽度，使两个波导宽度相同

findmodes;

Beta1i=1e-6*(2*pi/wavelength)*real(getdata("mode1","neff"));

Beta2i=1e-6*(2*pi/wavelength)*real(getdata("mode2","neff"));

cci=abs(Beta2i-Beta1i)/2;

kappa2i=sin(cci*L)^2;

t2i=o1-kappa2i;

plot(L,kappa2i,t2i,"distance","T");

plot(L,log10(kappa2i),log10(t2i),"distance","T");

对数坐标