Impulse is a change in momentum, and momentum is a vector — so a ball whose speed never changes can still be given a large impulse, purely because its direction changed.The law. The impulse–momentum theorem says
\[\vec{J} \;=\; \int \vec{F}\,dt \;=\; \Delta \vec{p} \;=\; m\vec{v}_f - m\vec{v}_i \]
where \(\displaystyle m\) is the mass of the ball, \(\displaystyle \vec{v}_i\) its velocity just before the bat hits it, and \(\displaystyle \vec{v}_f\) its velocity just after. \(\displaystyle \vec J\) has the units of momentum, \(\displaystyle \text{kg m s}^{-1}\), which is the same thing as \(\displaystyle \text{N s}\).
The step people get wrong: the speed is unchanged, so it is tempting to write \(\displaystyle m(v_f - v_i) = m(15 - 15) = 0\). That subtracts
speeds. The theorem subtracts
velocities, and these two velocities point different ways, so \(\displaystyle \Delta\vec p\) is far from zero.
The data. Convert the speed before anything else — never substitute km/h into an SI formula:
\[v \;=\; 54\ \text{km h}^{-1} \;=\; 54 \times \frac{1000\ \text{m}}{3600\ \text{s}} \;=\; 15\ \text{m s}^{-1} \]
and \(\displaystyle m = 0.15\ \text{kg}\). The speed is the same before and after, so \(\displaystyle |\vec v_i| = |\vec v_f| = 15\ \text{m s}^{-1}\). (The value of \(\displaystyle g\) is not needed — no weight enters an impulse over so short a contact.)
The geometry, stated carefully. Put the bat at the point \(\displaystyle O\). Draw the two paths as rays from \(\displaystyle O\): ray \(\displaystyle OA\) toward the bowler, the side the ball
came from, and ray \(\displaystyle OB\), the side the ball
goes to. The deflection quoted is the angle between these two paths at the bat:
\[\angle AOB = 45^\circ \]
This is the trap in the problem. The angle between the two paths is not the same as the angle through which the velocity vector turns. The ball travels
into \(\displaystyle O\) along \(\displaystyle OA\), so \(\displaystyle \vec v_i\) points along \(\displaystyle -\widehat{OA}\); it then travels
out along \(\displaystyle OB\). The velocity vector therefore swings through \(\displaystyle 180^\circ - 45^\circ = 135^\circ\), even though the two paths make only \(\displaystyle 45^\circ\) with each other. Getting this backwards is what turns a cosine into a sine below.
Set up axes. Let the \(\displaystyle x\)-axis lie along the bisector of \(\displaystyle \angle AOB\), pointing outward from \(\displaystyle O\) between the two rays; the \(\displaystyle y\)-axis is perpendicular to it. Each ray then makes \(\displaystyle 22.5^\circ\) with the \(\displaystyle x\)-axis:
\[\widehat{OA} = (\cos 22.5^\circ,\; +\sin 22.5^\circ), \qquad \widehat{OB} = (\cos 22.5^\circ,\; -\sin 22.5^\circ) \]
Hence
\[\vec v_i = -v\,(\cos 22.5^\circ,\; \sin 22.5^\circ), \qquad \vec v_f = +v\,(\cos 22.5^\circ,\; -\sin 22.5^\circ) \]
Subtract the velocities.\[\vec v_f - \vec v_i = v\big(\cos 22.5^\circ + \cos 22.5^\circ,\; -\sin 22.5^\circ + \sin 22.5^\circ\big) = \big(2v\cos 22.5^\circ,\; 0\big) \]
The \(\displaystyle y\)-components cancel exactly. What survives lies entirely along the \(\displaystyle x\)-axis — that is,
along the bisector of the two paths, pointing away from the bat, back into the region the ball came from. That is physically what you expect: the bat pushed the ball back the way it came.
Substitute.\[|\vec J| = m\,\big|\vec v_f - \vec v_i\big| = 2mv\cos 22.5^\circ = 2 \times 0.15\ \text{kg} \times 15\ \text{m s}^{-1} \times \cos 22.5^\circ \]
\[|\vec J| = 4.5\ \text{kg m s}^{-1} \times 0.9238795 = 4.1575\ \text{kg m s}^{-1} \]
Round once, at the end. The mass \(\displaystyle 0.15\ \text{kg}\) and the speed \(\displaystyle 54\ \text{km h}^{-1}\) each carry two significant figures, so the answer is quoted to two:
\[|\vec J| \approx 4.2\ \text{kg m s}^{-1} \;=\; 4.2\ \text{N s} \]
The direction is half the answer. Impulse is a vector, so a magnitude alone is incomplete: \(\displaystyle \vec J\) points along the bisector of the incoming and outgoing paths — at \(\displaystyle 22.5^\circ\) to each of them — directed outward from the bat, i.e. back toward the side the ball arrived from. By Newton's third law the ball pushes on the bat with an equal impulse in exactly the opposite direction.
A check on the sine-versus-cosine choice: if instead you read the \(\displaystyle 45^\circ\) as the angle through which the velocity vector turns, you would get \(\displaystyle 2mv\sin 22.5^\circ = 1.7\ \text{N s}\), directed perpendicular to the bisector. The bisector direction found above is the signature of the correct reading — for a ball turned back on itself through \(\displaystyle 135^\circ\), \(\displaystyle \Delta\vec p\) must lie along the bisector of the two paths, and only \(\displaystyle 2mv\cos 22.5^\circ\) does.
Answer: \(\displaystyle |\vec{J}| = 2mv\cos 22.5^\circ = 4.2\ \text{kg m s}^{-1}\) ($\displaystyle 4.2$ N s), directed along the bisector of the ball's incoming and outgoing paths, at \(\displaystyle 22.5^\circ\) to each, pointing back toward the side from which the ball came.