MHB Solve for x⁴+16x-12=0 using effective methods

  • Thread starter Thread starter anemone
  • Start date Start date
Click For Summary
The discussion revolves around solving the quartic equation x⁴ + 16x - 12 = 0, with participants sharing their methods and insights. Initially, one user describes a tedious approach using Descartes' Rule of Signs and Vieta's formulas to identify the roots, concluding with complex and real solutions. Another participant introduces a more efficient method by factoring the equation as a difference of squares, leading to a simpler solution. The conversation highlights the importance of recognizing patterns in polynomial equations for more effective problem-solving. Ultimately, the thread emphasizes the value of exploring various techniques to enhance mathematical understanding and efficiency.
anemone
Gold Member
MHB
POTW Director
Messages
3,851
Reaction score
115
Hi MHB,

This is a problem expected someone to solve it using the shortest route possible, and I've solved it, but using one most ordinary method that is very tedious and less gratifying.:( But I firmly believe this problem is designed and meant for practicing of the heuristic/problem solving skills, the thing is I just don't see there is any shortcut to tackle it.:mad:

If you happened to know there is another way to solve this problem, would you please show me? Thanks a bunch in advance.:)

Problem:

Solve $x^4+16x-12=0$.

My solution:

Descartes' Rule of Signs suggests there is at most one positive and one negative real roots, and since there is only one inflexion point (if we let $f(x)=x^4+16x-12=0$, then $f'(x)=4x^3+16$ which then implies $f'(x)=0$ iff $x=-4^{1/3}$) for the given function, from $f(1)=5,\,f(0)=-12,\,f(-3)=21$, Intermediate Value Theorem tells us the function $f(x)=x^4+16x-12=0$ has one positive real, one negative real and two complex roots.

If we first let the 4 roots be $a+bi,\,a-bi,\,c+\sqrt{d},\,c-\sqrt{d}$, Vieta's formula says the sum of the four roots is zero. Thus, $2a+2c=0,\,\rightarrow\,a=-c$.

The coefficient of $1$ from the given function allows us to further assume the four roots (2 of which real and 2 of which complex) as $1+bi,\,1-bi,\,-1-\sqrt{d},\,-1-\sqrt{d}$. Note that this can be wrong, as the $+1$ can stick to the real roots and the $-1$ be assigned to the complex roots. That said, we need to check our working to verify of our answer.

According to the Vieta's formula, it tells us again that

$(1+bi)(1-bi+(-1-\sqrt{d})+(-1-\sqrt{d}))+(1-bi)((-1-\sqrt{d})+(-1-\sqrt{d}))+(-1-\sqrt{d})(-1-\sqrt{d}))=0$

This simplifies to $(b^2-2d+1)+(bd-2b)i=0$, i.e.,

$d=\dfrac{b^2+1}{2}$ and substituting this into $bd-3b=0$ gives $b(b^2+1-6)=0$, or $b^2=5$ since $b\ne 0$ and hence $d=3$.

This generates the four roots to be $1+\sqrt{5}i,\,1-\sqrt{5}i,\,-1+\sqrt{3},\,-1-\sqrt{3}$.

I checked and realized this is a valid set of the answer.

But I am not satisfied by my method, I thought there ought to be another easy way to do this good problem. So, any feedback would be greatly appreciated.
 
Mathematics news on Phys.org
Try factoring it into two quadratics:

$$x^4+16x-12=\left(x^2+ax+b\right)\left(x^2+cx+d\right)$$
 
Oh My Goodness!

It actually can be rewritten as
$\begin{align*}x^4+16x-12&=x^4+4x^2+4-4(x^2-4x+4)\\&=(x^2+2)^2-(2(x-2))^2\\&=(x^2+2+2(x-2))(x^2+2-2(x-2))\\&=(x^2+2x+4)(x^2-2x+6)\\&=(x-(-1+\sqrt{3}))(x-(-1-\sqrt{3}))(x-(1+\sqrt{5}i))(x-(1+\sqrt{5}i))\end{align*}$

See how simple this solution that I found online is! :o

Edit: Hey MarkFL, you're right, I'm such a fool, I don't know where my head is...hehehe...
 
$x^4 + 16x - 12 = (x^4 + 4x^2 + 4 ) - (4x^2 - 16x + 16) = (x^2+2)^2 - (2(x-2) ))^2$. Since it is of form $a^2 - b^2$, we can rewrite that as $(a-b)(a+b)$, which yields a product of two quadratic as MarkFL indicated.

EDIT: You beat me to the post.
 
anemone said:
Oh My Goodness!

It actually can be rewritten as
$\begin{align*}x^4+16x-12&=x^4+4x^2+4-4(x^2-4x+4)\\&=(x^2+2)^2-(2(x-2))^2\\&=(x^2+2+2(x-2))(x^2+2-2(x-2))\\&=(x^2+2x+4)(x^2-2x+6)\\&=(x-(-1+\sqrt{3}))(x-(-1-\sqrt{3}))(x-(1+\sqrt{5}i))(x-(1+\sqrt{5}i))\end{align*}$

See how simple this solution that I found online is! :o

Edit: Hey MarkFL, you're right, I'm such a fool, I don't know where my head is...hehehe...

I did not see that it could be rewritten as the difference of two squares...yet. :D
 
anemone said:
Oh My Goodness!

It actually can be rewritten as
$\begin{align*}x^4+16x-12&=x^4+4x^2+4-4(x^2-4x+4)\\&=(x^2+2)^2-(2(x-2))^2\\&=(x^2+2+2(x-2))(x^2+2-2(x-2))\\&=(x^2+2x+4)(x^2-2x+6)\\&=(x-(-1+\sqrt{3}))(x-(-1-\sqrt{3}))(x-(1+\sqrt{5}i))(x-(1+\sqrt{5}i))\end{align*}$

See how simple this solution that I found online is! :o

Edit: Hey MarkFL, you're right, I'm such a fool, I don't know where my head is...hehehe...

How does one even notice something like that? :confused:
Trial and error of regrouping and adding and subtracting terms?
 
Rido12 said:
How does one even notice something like that? :confused:
Trial and error of regrouping and adding and subtracting terms?

One will know by experience, Rido12! :o Or, since we know the given function could be factored as $(x^2+ax+b)(x^2+cx+d)$, we can still figure out the values of $a,\,b,\,c,\,d$ through a little computation by comparing the coefficients.:)
 
In case anyone wants a rigorous approach, the natural resolvents coming from a quartic are

$$\alpha_1 = (x_1 + x_2)(x_3 + x_4) \\ \alpha_2 = (x_1 + x_3)(x_2 + x_4) \\ \alpha_3 = (x_1 + x_4)(x_2 + x_3)$$

Where $x_i$ are the roots of the quartic. Note that these are left unchanged by the $4! = 24$ permutation of the roots. If we take the quartic in OP, it is not too hard to verify that $\alpha_i$s satisfies the cubic

$$x^3 + 48x + 256 = 0$$

This, however, factors as $(x + 4)(x^2 - 4x + 64) = 0$ by rational root theorem (indeed, this indicates that the Galois group of this polynomial is $D_4$ and shouldn't involve too much radicals). The roots are thus

$$\alpha_1 = -4 \\ \alpha_2 = 2(1 - i\sqrt{15}) \\ \alpha_2 = 2(1 + i\sqrt{15})$$

By some tedious computations, one retrieves the original roots of the quartic back

$$2x_1 = \sqrt{-\alpha_1} + \sqrt{-\alpha_2} + \sqrt{-\alpha_3} \\ 2x_2 = \sqrt{-\alpha_1} - \sqrt{-\alpha_2} - \sqrt{-\alpha_3} \\ 2x_3 = -\sqrt{-\alpha_1} + \sqrt{-\alpha_2} - \sqrt{-\alpha_3} \\ 2x_4 = -\sqrt{-\alpha_1} - \sqrt{-\alpha_2} + \sqrt{-\alpha_3}$$

Which are, after the application of the identities $\sqrt{-2\left (1-i\sqrt{15}\right)} + \sqrt{-2\left(1+i\sqrt{15}\right)} = 2 \sqrt{3}$ and $\sqrt{-2\left(1-i\sqrt{15}\right)} - \sqrt{-2\left(1+i\sqrt{15}\right)} = i2 \sqrt{5}$ and appropriate signs (this is the real pain in the neck) are

$$x_1 = -1 + \sqrt{3} \\ x_2 = -1 - \sqrt{3} \\ x_3 = 1 + i\sqrt{5} \\ x_4 = 1 - i\sqrt{5}$$

These methods are much more practical while doing calculations of these sort, although they do not describe describe any of the beauties of manipulation whatsoever.
 

Similar threads

Replies
16
Views
743
  • · Replies 41 ·
2
Replies
41
Views
5K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 10 ·
Replies
10
Views
1K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K