Propositional Logic: Formal Proof by Cases of: 1. $\neg(A\wedge B)$ 2. $\neg(A\wedge\neg B)$ Thus, 2. $\neg A$. I'm having a hard time coming up with a formal proof by cases method for this set of premises and conclusion.  Note that $\neg$ refers to negation and $\wedge$ denotes AND. | Denotes subproof


*

*$\neg(A\wedge B)$

*$\neg(A\wedge\neg B)$
Thus,


*$\neg A$
Use # for contradiction; justify subproof assumptions with Assume; always drop outer parentheses; no spaces in PROP.


*

*~(A&B)     Premise

*~(A&~B)   Premise

*|  B       Assume

*||  ~B     Assume

*||  B      Reit;3

*||  #      #Intro;4,5

*|   A      #Elim;6

*| A&~B     &Intro;4,7

*|  #       #Intro;2,8

*~A       #Elim;9


This is what I have currently, but am not sure if the steps I took are okay.
 A: By De Morgan's law and double negation we have $\neg(A\wedge B) \Leftrightarrow \neg A \vee \neg B$ and $\neg(A\wedge \neg B) \Leftrightarrow \neg A \vee  B$. Therefore
$$
\neg(A\wedge B) \wedge \neg(A\wedge \neg B) \Leftrightarrow (\neg A \vee \neg B) \wedge (\neg A \vee  B) \overset{DP}{\Leftrightarrow} \neg A \vee (B\wedge \neg B) \Leftrightarrow \neg A
$$
Here DP is the distributive property. The last equivalence holds because $B\wedge \neg B$ is a contradiction.
A: What appears to be a problem with the proof is that "# Elim" (explosion) should not discharge a assumption (or close a subproof) as what done on lines 7 and 10.  Here is what the attempted proof looks like in a Fitch-style proof checker:

Here is a proof that should work using De Morgan's laws as suggested by Alan Muniz's answer:

I converted the two premises using De Morgan's laws (DeM) on lines 3 and 4. Then I attempted to eliminate the disjunction on line 4 by considering both cases. In each case, I had to derive $\neg A$. The first case was already $\neg A$ so there was nothing I needed to do. For the second case, $\neg \neg B$ I used disjunctive syllogism (DS) to derive from line 3 the desired result $\neg A$.  When I used disjunction elimination on line 8, I was permitted to discharge the assumptions on lines 5 and 6 which closed the two subproofs.

Kevin Klement's JavaScript/PHP Fitch-style natural deduction proof editor and checker http://proofs.openlogicproject.org/
A: As Frank Hubeny has stated in his answer, the $\#\textsf{Elim}$ rule (a.k.a. negation elimination, ex falso quadlibet, explosion) does not discharge the assumption.  
What the OP was attempting was to introduce a negation, which does; (just not with assumptions attempted).   This would have required assuming the positive with the aim of contradicting a premise, so that assumption may be discharged and its negation infered.   Eg, assume $A$ on line $3$ and $B$ on line $4$. 
$$\def\fitch#1#2{~~\begin{array}{|l}#1\\\hline#2\end{array}}\fitch{~~1.~~{\sim}(A\& B)\quad\textsf{Premise}\\~~2.~~{\sim}(A\&{\sim}B)\quad\textsf{Premise}}{\fitch{~~3.~A\quad\textsf{Assumption}}{\fitch{~~4.~~B\quad\textsf{Assumption}}{~~5.~~A\& B\quad\&\textsf{Intro }3,4\\~~6.~~\#\quad\#\textsf{Intro }1,5}\\~~7.~~{\sim}B\quad{\sim}\textsf{Intro }4{-}6\\~~8.~~A\&{\sim}B\quad\&\textsf{Intro }3,7\\~~9.~~\#\quad\#\textsf{Intro }2,8}\\10.~~{\sim}A\quad{\sim}\textsf{Intro }3{-}9}$$
