Just as you index into data1cm(i) you can save the value into fluence in the same manner, e.g. if the result of data1cm(i).data(:,3).*25000000.*(energygap) is scalar you can do
for i = 1:nx
fluence(i)= data1cm(i).data(:,3).*25000000.*(energygap);
end
Then fluence will be a 1*nx array of all the results.
For efficiency one should always initialise the variable, that is create it with some temporary values and then overwrite these values.
fluence = zeros(1,nx);
for i = 1:nx
fluence(i)= data1cm(i).data(:,3).*25000000.*(energygap);
end
Here I initialise fluence with zeros, and then all of these zeros are overwritten by the result of data1cm(i).data(:,3).*25000000.*(energygap);. Other initialisations are e.g. ones and nan.
data1cm(i).dataandenergygapare. Please see minimal reproducible example.