MathGroup Archive 1992

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Integrate[e/(1+Exp[e]),{e,ei,Infinity}]

  • To: mathgroup at yoda.physics.unc.edu
  • Subject: Re: Integrate[e/(1+Exp[e]),{e,ei,Infinity}]
  • From: withoff
  • Date: Fri, 11 Dec 92 15:08:33 CST

> Integrate[e/(1+Exp[e]),{e,ei,Infinity}]
> this fermintegral of order 1 results in RecursionLimit errors.
> The indefinite integral can be found without any problems.
> Any ideas how to change the situation??
> 
> Michael Trott
> 
> trott at physik.tu-ilmenau.de

If the indefinite integral works, it can probably be used to compute
the definite integral.  The built-in Limit function can't handle the
limit at Infinity, but the package Calculus`Limit` can.

In[1]:= f = Integrate[e/(1+Exp[e]),e]

         2
        e               e                 e
Out[1]= -- - e Log[1 + E ] - PolyLog[2, -E ]
        2

In[2]:= << Calculus`Limit`

In[3]:= Limit[f, e -> Infinity]

          2
        Pi
Out[3]= ---
         6

In[4]:= Limit[f, e -> ei]

          2
        ei                ei                 ei
Out[4]= --- - ei Log[1 + E  ] - PolyLog[2, -E  ]
         2

In[5]:= %% - %

           2     2
        -ei    Pi                ei                 ei
Out[5]= ---- + --- + ei Log[1 + E  ] + PolyLog[2, -E  ]
         2      6

Dave Withoff
withoff at wri.com





  • Prev by Date: Re: Telescoping series
  • Next by Date: ParametricPlot3D
  • Previous by thread: 3-D workstations for Mma
  • Next by thread: ParametricPlot3D