I think this is better posted here than in Physics.se because it is purely a geometry/vector problem.
A particle (purple dot in diagram) with mass $m$ kg rests in equilibrium due to coefficient of friction $\mu > tan(\alpha)$, where $0^\circ<\alpha<45^\circ$ and $0<\mu<1.$
I was wondering if a force P at a positive clockwise angle $\theta ^\circ > 0$ with the vertical could result in an acceleration of the particle down the slope. I think the answer is no.
$\underline{R}$ is the reaction force on the particle from the slope.
As P increases from $0N$, the particle remains in equilibrium until the resultant force $\underline{Res_{lim}}$ on the particle down the slope, $\underline{Res_{lim}} = m \underline{g} + \underline{R_{lim}}+ \underline{P_{lim}}$ is equal to limiting friction $\underline{F_{lim}}$.
I think we can then show that for $\underline{P}> \underline{P_{lim}}$ (and assuming the particle accelerates down the slope), the resultant force $\underline{Res} = m \underline{g}+ \underline{R}+ \underline{P}$ is strictly less than the friction force $\underline{F}$, using the fact that $\underline{F} = \mu \underline{R},\ \underline{F_{lim}} = \mu \underline{R_{lim}}$ and the fact that $\theta$ is positive clockwise angle to the vertical.
Then the resultant force would be up the slope, which would contradict the assumption that the particle accelerates down the slope, answering my original question.
I'm interested how to formalise my proof with simple geometry and vectors. I've had a go and it shouldn't be that difficult, but haven't managed to formalise it.
My idea for a proof would be: $\underline {P}$ removes a greater proportion of $\underline{Res_{lim}}$ than it removes of $\underline{R_{lim}}$, but like I say, I find formalising this difficult.
$\underline{Res_{lim}} = \mu \underline{R_{lim}}$. Now what?




As David K pointed out in the comments, the solution is simple. When $P = 0, R = m g \cos{\alpha} > Res_{lim} = F_{lim} =$ max{ ${F}$ } $\implies $ positive acceleration down the slope when $P=0$, contradicting static equilibrium when $P=0$.
More interesting is when $ \mu = \tan \alpha $, so that when $P=0$, the object starts at rest in static equilibrium. Again, we look at the right-angled triangle formed by $\underline{R_{lim}}, \ \underline{Res_{lim}}$ and $m\underline{g}$ when $P=0$ and compare it to the right-angled triangle formed by $\underline{\Delta R}, \ \underline{\Delta Res}$ and $\underline{P}$.
$$F = \mu R\ \text{and} \ R = R_{lim} - \Delta R, \ \therefore F = \mu(R_{lim} - \Delta R)$$
$$\Delta F = F_{lim} - F = \mu R_{lim} - \mu (R_{lim} - \Delta R) = \mu \Delta R = \Delta R\tan \alpha, \implies \frac{\Delta F}{\Delta R} = \tan \alpha.$$
$$\text{However, for the right-angled triangle formed by} \ \underline{\Delta R}, \ \underline{\Delta Res} \ \text{and} \ \underline{P},\ \frac{\Delta F}{\Delta R} = \tan(\alpha - \theta) \neq \tan \alpha, \text{because} \ \theta > 0 ^\circ. $$
So no, if a particle is at rest in static equilibrium on an inclined plane, then if $P$ is pointing in the same quadrant as up the slope (and also if $\theta = 0$, i.e. $P$ is vertical), then the particle cannot accelerate down the slope.