Jtam-A4.dvi JOURNAL OF THEORETICAL AND APPLIED MECHANICS 50, 4, pp. 1001-1010, Warsaw 2012 50th Anniversary of JTAM NEW ANALYSIS OF NATURAL CONVECTION BOUNDARY LAYER FLOW ON A HORIZONTAL PLATE WITH VARIABLE WALL TEMPERATURE Abbasali Abouei Mehrizi, Yousef Vazifeshenas, G. Domairry Babol University of Technology, Department of Mechanical Engineering, Babol, Iran e-mail: abbasabouei@gmail.com; joseph.vazifeshenas@gmail.com; amirganga111@yahoo.com In this study, steady laminar free convection boundary layer flow on a horizontal plate is investigated through analytical solutions. By transforming the governing non-dimensional boundary layer equations into an ordinary differential equation, the application of the Ho- motopyAnalysisMethod can be practical. So in this case, the analytical results for different Prandtl numbers and constant M values which portray the power index are achieved. The trend goes on large amounts of (M ≫ 1) and the results are compared with other efforts. Furthermore, the effects of different values of Prandtl number and M values on temperature and velocity profiles are verified. Key words: free convection, HAM (Homotopy Analysis Method), analytical solution 1. Introduction Steady free convection boundary layer flowpast a horizontal plate is a very practical andbasical issue that is worthy enough to be discussed perfectly. So researchers in the heat transfer field paid tangible attention toward this problem. Here, because of the significant effect of buoyancy on the flow field, going through depth of the layer pressure gradient changes. Despite being pre- determined at the edge of the boundary layer, it should be foundedas a part of solving process in accordancewith velocity and temperature. Former studies verified the effect of buoyancy for this problem, and in their effort it is gotten that they focused onflows over plateswhich are smoothly cooled or heated. Stewartson (1958) concentrated on an isothermal horizontal semi-infinite plate and his results, which were later discussed by Gill et al. (1965), announced the existence of similarity solutions just for below and above a cooled and heated surface, respectively. Rotem and Claassen (1969) and Raju et al. (1984) discussed on the heated downward-facing or cooled upward-facing plates numerically. Identically, Clifton and Chapman (1969) utilized an integral method for this problem. Jones (1973), and Pera andGebhart (1973) performed their study on an inclined plate with a isothermal condition. Also Yu and Lin (1988), and Lin et al. (1989) verified an arbitrary inclined plate. Afzal et al. (1986), Lin and Yu (1988), Brouwers (1993) and Chen et al. (1993) studied the influence of suction or blowing on the free convection of a horizontal flat plate. Ackroyd (1976) surveyed the non-Boussinesq effects. Chen et al. (1986) verified different occasions for a plate (horizontal, inclined, and vertical) with changing wall temperature or surface heat flux. A finite-difference method was applied to have a numerical solution for the non-dimensional form of the governing equation. Mahajan and Gebhart (1980) and Afzal (1985) studied higher order effects in free convection flow over horizontal surfaces. Daniels (1992) studied an insulated horizontal wall over which there was a thermal boundary layer flowing. In this flow, the velocity and temperature fields are coupled by buoyancy. Many researchers worked on free convection andmade great effort on this issue such as Rotem and Classen (1969), Goldstein et al. (1973) and Kitamura and Kimura (1995) who had technical papers. 1002 A. Abouei Mehrizi et al. However, little by little more theoretical studies were performed on scrutinizing the effect of buoyancy on thermal boundary layers over horizontal flat surfaces in which wall temperature and surface heat flux vary. The Homotopy Analysis Method was another approach which was handled by other researchers inmany previous papers (Liao, 1992, 2003; Hayat and Sajid, 2007; Domairry and Nadim 2008, Domairry and Fazeli 2009; Abouei Mehrizi et al., 2011). Here, the main aim is to consider free convection boundary layer equations governing the flow on a heated horizontal flat plate facing upward when the non-dimensional surface temperature is given by Tw(x)= x M. In this form, M is a positive constant and x is the coordinatemeasured along the plate. 2. Description of the problem and governing equations Consider the problem of free convection boundary layer flow on a heated horizontal flat plate facing upward when the non-dimensional surface temperature is given by Tw(x) = x M, where x is the coordinate measured along the plate from the leading edge and M is a constant. The governing boundary layer equations in dimensional form are ∂u ∂x + ∂v ∂y =0 u ∂u ∂x +v ∂u ∂y =− 1 ρ ∂p ∂x +υ ∂2u ∂y2 − 1 ρ ∂p ∂y +grβ(T −T∞)= 0 u ∂T ∂x +v ∂T ∂y = α ∂2T ∂y2 (2.1) Subject to the boundary conditions u = v =0 T = Tw(x) at y =0 u =0 T = T∞ p = p∞ as y →∞ (2.2) where x and y are the coordinates measured along the plate and normal to it, respectively, (u,v) are the velocity components along the (x,y) axes, p is the pressure, T is the local fluid temperature, gr is the acceleration due to gravity, ρ is the fluid density, β is the coefficient of thermal expansion, υ is the kinematic viscosity and α is the constant thermal diffusivity of the fluid. We introduce now the following non-dimensional variables x = x L y = Gr1/5 (y L ) u = Gr−2/5 (L υ ) u v = Gr−1/5 (L υ ) v T = T∞+∆TT p = G −4/5p−p∞ ρυ2/L2 (2.3) Substitutingvariables (2.3) into equations (2.1) leads to the following non-dimensional equations ∂u ∂x + ∂v ∂y =0 u ∂u ∂x +v ∂u ∂y =− ∂p ∂x + ∂2u ∂y2 − ∂p ∂y +T =0 u ∂T ∂x +v ∂T ∂y = 1 Pr ∂2T ∂y2 (2.4) and boundary conditions (2.2) become u = v =0 T = Tw(x)= x M at y =0 u → 0 T → 0 p → 0 as y →∞ (2.5) New analysis of natural convection boundary layer flow... 1003 Further, we look for a similarity solution of these equations of the form (Chen et al. 1986) ψ = x(M+3)/5f(η) T = xMθ(η) p = x(4M+3)/5g(η) η = yx(M−2)/5 (2.6) where ψ is the stream function, which is defined as u = ∂ψ ∂y v =− ∂ψ ∂x (2.7) and automatically satisfy continuity equation (2.4)1. Substitution of (2.6) into equations (2.4)2 to (2.4)4 gives the following ordinary differential equations 5f ′′′+(M +3)ff ′′− (2M +1)f ′ 2 − (4M +2)g − (M −2)ηg′ =0 g′ = θ 5 Pr θ′′+(M +3)fθ′−5Mf ′θ =0 (2.8) and boundary conditions (2.5) become f(0)= f ′(0) θ(0)= 1 f ′(∞)= 0 θ(∞)= 0 g(∞) = 0 (2.9) where primes denote differentiation with respect to η. Quantities of physical importance in this problem are the skin friction τw and the heat transfer at the plate qw, which are given in non-dimensional form by Chen et al. (1986, 1993) τw = x (3M−1)/5f ′′(0) qw = x 2(3M−1)/5[−θ′(0)] (2.10) For large values of M(≫ 1), solutions to equations (2.8) subject to boundary conditions (2.9) can be found using the transformation f = M(−3/5)F(z) θ = θ(z) g = M−2/5G(z) z = M−2/5η (2.11) Substituting (2.11) into equations (2.8) yields 5F ′′′+ ( 1+ 3 M ) FF ′′− ( 2+ 1 M ) F ′2− ( 4+ 2 M ) G− ( 1− 2 M ) zG′ =0 G′ = θ 5 Pr θ′′+ ( 1+ 3 M ) Fθ′−5F ′θ =0 (2.12) Subject to the boundary conditions F(0)= F ′(0) θ(0)= 1 F(∞)= 0 θ(∞)= 0 G(∞)= 0 (2.13) where primes now denote differentiation with respect to z. Substituting equation (2.12)2 into equation (2.12)3, we have 5F ′′′+ ( 1+ 3 M ) FF ′′− ( 2+ 1 M ) F ′2− ( 4+ 2 M ) G− ( 1− 2 M ) zG′ =0 5 Pr G′′′+ ( 1+ 3 M ) FG′′−5F ′G′ =0 (2.14) and F(0)= F ′(0) G′(0)= 1 F(∞)= 0 G′(∞)= 0 G(∞) =0 (2.15) Solving the equations, the functions F and G will be calculated, and to evaluate the values of f ′′(0) and −θ′(0) the following relations are used f ′′(0)= M1/5F ′′(0) −θ′(0)=−g′′(0)=−M2/5G′′(0) (2.16) 1004 A. Abouei Mehrizi et al. 3. Homotopy analysis solution For HAM solution of the governing equations, we choose the auxiliary linear operators L1(F) and L2(G) as follows L1(F)= F ′′′+F ′′ L2(G) = G ′′′+G′′ (3.1) Solving Eqs. (3.1) with initial conditions of Eqs. (2.15), one could make the initial guesses L1(C1+C2+C3e −z) L2(C4+C5+C6e −z) (3.2) and F0(z)= 0 G0(z)=−e −z (3.3) where ci (i =1, . . . ,5) are considered as constants. P ∈ [0,1] denotes the embedding parameter and ~ indicates non-zero auxiliary parameters.We then construct the following equations. 3.1. Zeroth order deformation equations (1−P)L1[F(z;p)−F0(z)] = p~H1(z)N1[F(z;p),G(z;p)] (1−P)L2[G(z;p)−G0(z)] = p~H2(z)N2[F(z;p),G(z;p)] (3.4) and F(0;p) = 0 F ′(0;p) = 0 F ′(∞;p)= 0 G′(0;p) = 1 G′(∞;p)= 0 G(∞;p =0 (3.5) and N1[F(z;P)] = 5 d3F(z) dz3 + ( 1+ 3 M ) F d2F(z) dz2 − ( 2+ 1 M )(dF(z) dz )2 − ( 4+ 2 M ) G(z)− ( 1− 2 M ) z dG(z) dz N2[G(z;p)] = 5 Pr d3G(z) dz3 + ( 1+ 3 M ) F(z) d2G(z) dz2 −5 dF(z) dz dG(z) dz (3.6) For p =0 and p =1we have F(z;0) =F0(z) F(z;1) = F(z) G(z;0) = G0(z) G(z;1) = G(z) (3.7) As the embedding parameter increases from 0 to 1, F(z;0) and G(z;0) vary from the initial guess F0(z), G0(z) to the exact solution F(z) and G(z). With expanding F(z;q) and G(z;q) in Taylor series with respect to q, we have F(z;p) = F0(z)+ ∞ ∑ m−1 Fm(z)p m G(z;p) = G0(z)+ ∞ ∑ m−1 Gm(z)p m (3.8) where Fm(z) = 1 m! ∂mF(z;p) ∂pm Gm(z)= 1 m! ∂mG(z;p) ∂pm (3.9) New analysis of natural convection boundary layer flow... 1005 3.2. mth order deformation equations According to definition (3.9) the governing equations can be deduced from the zero order deformation Eqs. (3.4). Define the vector Fn = [F0(z),F1(z),F2(z), . . . ,Fn(z)] (3.10) Differentiating Eqs. (3.4) m times with respect to the embedding parameter p and then setting p =0 and finally dividing them by m!, we have the so-called mth order deformation equation L1[Fm(z)−χmFm−1(z)] = ~H(z)R1m(z) L2[Gm(z)−χmGm−1(z)] = ~H(z)R2m(z) (3.11) and F(0)= 0 F ′(0)= 0 F ′(∞)= 0 G′(∞)= 0 G′(0)= 1 G(∞)= 0 (3.12) and R1m(z)= 5 d3Fm−1(z) dz3 + ( 1+ 3 M ) m−1 ∑ k=0 ( Fm−1−k d2Fk(z) dz2 ) − ( 2+ 1 M ) m−1 ∑ k=0 (dFm−1−k(z) dz dFk(z) dz ) − ( 4+ 2 M ) Gm−1(z)− ( 1− 2 M ) z dGm−1(z) dz R2m(z)= 5 Pr d3Gm−1−k(z) dz3 + ( 1+ 3 M ) m−1 ∑ k=0 ( Fm−1−k(z) d2Gk(z) dz2 ) −5 m−1 ∑ k=0 (dFm−1−k(z) dz dGk(z) dz ) (3.13) where χm = { 0 m ¬ 1 1 m > 1 (3.14) Finally, according to the third rule of solution expression, the auxiliary function H(z) should be in the form H1(z) = H2(z)= e −z (3.15) Solving Eqs. (3.11) by mathematical software such as Maple, for example with M = 1 and Pr=1, one successively obtains f0(z) = 0 f1(z)=−2he −2z − 1 4 he−2zz+ 15 4 he−z − 7 4 h g0(z) =−e −z g1(z) =− 1 40 he−2z − 7 20 he−2zz− 3 10 he−z f2(z) =− 1 432 h(e−3z(2090h+300,hz)+e−2z(864−4158h−270hz −108z) +e−z(−1620+2130h)−35h+756 g2(z) = 1 288 he−z[e−3z(72h+9hz)+e−2z(−860h)+e−z(396h−360)+1509h+720 (3.16) 1006 A. Abouei Mehrizi et al. 4. Convergence of HAM solution HAMprovides uswith great freedom in choosing the solution of a nonlinear problembydifferent base functions and has a great effect on the convergence region. Therefore, we should ensure that the solution converges. On the other hand, the convergence and rate of approximation for the HAM solution strongly depends on the value of the auxiliary parameter h. Bymeans of the so-called h-curves, it is easy to find out the so-called valid regions of auxiliary parameters to gain a convergent solution series. To demonstrate the influence of h on the convergence of solution, we plotted the so-called h-curve of f(0) and θ(0) by 11th to 17th order approximation of the solution as shown in Fig. 1a and 1b. Fig. 1. The h-validity of f′′(0) (a) and θ′(0) (b) 5. Results and discussion In the literature, the governing equation is solvedwith an analyticalmethod calledHAMand, as it is portrayed, good and acceptable agreement is achieved in comparison with other efforts. In Table 1, the amounts of the heat transfer coefficient, −θ′(0) are presented for M =0 and three values of the Prandtl number. For Pr = 1, the results by Pop et al. (2009), Raju et al. (1984) and Lin et al. (1989) are also put in this table, and it is seen that the gratifying correspondence exists. Table 1.Amounts of −θ′(0) for M =0 and different Pr Pr Present Pop et al. (2009) Raju et al. (1984) Lin et al. (1989) 1 0.3866 0.3905 0.3881 0.3905 7 0.6317 0.6300 10 0.6813 0.6833 Normally, the Prandtl number has a reciprocal relationship with thermal conductivity. So whenPr increases, because of a higher heat transfer rate at the surface, the heat transfer coeffi- cient increases, and this trend is shown in Table 1. Subsequently, Table 2 represents the values of heat transfer coefficient −θ′(0) and skin coefficient f ′′(0) when Pr=1 but M varies. On the other hand, Table 3 demonstrates the values of −θ′(0) and f ′′(0) when M =2 but Pr varies. Finally, Table 4 shows the amounts of heat transfer coefficient and skin fiction coefficient when Pr = 7 but M varies. As it is vivid, acceptable agreement is achieved, and by comparing the results with numerical data, it is gotten that enhancing the value of M makes more inclination New analysis of natural convection boundary layer flow... 1007 between the results. Alternatively, for lower values of M and Prandtl number, more precise results are gained. Table 2.Amounts of −θ′(0) and f ′′(0) when Pr=1 but M varies HAM Numeric Error [%] M f ′′(0) −θ′(0) f ′′(0) −θ′(0) f ′′(0) −θ′(0) 1 0.9953 0.6537 0.9910 0.6532 0.433 0.076 2 1.0857 0.8142 1.0811 0.8129 0.425 0.159 3 1.1494 0.9311 1.1499 0.9360 0.043 0.523 4 1.2053 1.0366 1.2059 1.0388 0.049 0.211 5 1.2530 1.1118 1.2532 1.1283 0.015 1.462 6 1.2944 1.1867 1.2945 1.2083 0.007 1.787 7 1.3311 1.2751 1.3311 1.2811 0 0.468 8 1.3640 1.337 1.3641 1.3482 0.007 0.830 9 1.3945 1.396 1.3943 1.4106 0.014 1.035 10 1.4213 1.4580 1.4220 1.4692 0.049 0.762 Table 3.Amounts of −θ′(0) and f ′′(0) when M =2 but Pr varies HAM Numeric Pr f ′′(0) −θ′(0) f ′′(0) −θ′(0) 1 1.083 0.8127 1.0811 0.8129 7 0.5149 1.2536 0.4911 1.2480 10 0.4682 1.3709 0.4259 1.3464 Table 4.Amounts of −θ′(0) and f ′′(0) when Pr=7 but M varies HAM Numeric Error [%] M f ′′(0) −θ′(0) f ′′(0) −θ′(0) f ′′(0) −θ′(0) 1 0.4523 1.0126 0.4560 1.0144 0.811 0.177 2 0.5149 1.2536 0.4911 1.2480 4.846 0.448 3 0.5507 1.4463 0.5195 1.4291 6.006 1.203 4 0.5785 1.5728 0.5431 1.5809 6.518 0.512 5 0.5991 1.7068 0.5635 1.7135 6.317 0.391 6 0.6100 1.8263 0.5810 1.8325 4.991 0.338 7 0.6341 1.9341 0.5973 1.9410 6.161 0.355 8 0.6496 2.0335 0.6117 2.0405 6.195 0.343 9 0.6625 2.1262 0.6249 2.1336 6.016 0.346 10 0.6751 2.2124 0.6371 2.2209 5.964 0.382 Passing through this information, velocity profiles and temperature profiles are investigated, too. Figure 2a shows the velocity profiles f ′(η), when Pr = 1 and M has different amounts. Admitting the information of Table 2, here this picture claims that when there is larger amount of M, the velocity gradient at the surface is larger and this leads to having larger skin friction. Furthermore, Fig. 2b admits Table 2, and by demonstrating the temperature profiles θ(η) it is seen that the temperature gradient at the surface and M values have a direct relationship. Additionally, Fig. 3a presents the velocity profile when M = 2 but Pr has different amounts. By changing the amounts of thePrandtl number, this section shows that when there are smaller values for Pr, larger amounts of the velocity gradient is found at the surface and again this leads to a large skin friction coefficient. This data reviews the information presented in Table 3. 1008 A. Abouei Mehrizi et al. Fig. 2. Velocity profiles f′(η) (a) and temperature profiles θ(η) (b) for Pr=1 and different values of M Fig. 3. Velocity profiles f′(η) (a) and temperature profiles θ(η) (b) for M =2 and different Prandtl numbers Fig. 4. Velocity profiles f′(η) (a) and temperature profiles θ(η) (b) for Pr=7 and different values of M Figure3balso admitsTable3, and itportrays thatwhen Prdiminishes, the temperaturegradient at the surface increases. It is clear that when it is talked about a higher Prandtl number fluid, it has high viscosity and this makes the thermal boundary layer thinner and heat transfer rate higher at the surface. Figure 4a shows the velocity profiles f ′(η) when Pr = 7 and M gets different amounts. A large skin friction coefficient is obtained by a large velocity gradient at the surface. This is the reality for large M and is depicted in Fig. 4a correspondent to Table 4. The last effort inFig. 4b presents the direct relationship between M and the temperature gradient at the surface in away that increasing M increases the temperature gradient, too. And this is said previously in Table 4. Totally, it can be mentioned that for validating these achievements the satisfaction of boundary conditions is adequate, and by reviewing Figs. 1 to 4, this is approved here. New analysis of natural convection boundary layer flow... 1009 6. Conclusion In this effort, steady laminar free convection boundary layer flow on a horizontal plate with variable wall temperature has been investigated analytically. The subject is scrutinized using the Homotopy Analysis Method. The issue is based on the variation of two parameters, the Prandtl number and M, and in this paper it is shown how these two parameters influence the temperature profile, velocity profile, heat transfer coefficient and skin friction. Also the results are in good agreementwith otherworks and the gratification of theboundary layers iswitnessed. The results show: • The homotopy analysis method is an applicable method to solve nonlinear fluid dynamic problems. • The velocity gradient at the surface and skin friction increase with increasing values of M and the Prandtl number. • The temperature gradient at the surface and M values have direct relationship. • The temperature gradient at the surface increases by decreasing the Prandtl value. References 1. Abouei Mehrizi A., Domairry G., Khesali M.A., Latif Aghili A., 2011, Analysis of tran- sient heat flow in a long narrow tube with convection and radiation,Proc. International Congress on Advances in Applied Physics and Materials Science, 411-414 2. Ackroyd J.A.D., 1976, Laminar natural convection boundary layers on near-horizontal plates, Proc. Roy. Soc. London A, 352, 249-274 3. Afzal N., 1985, Higher order effects in natural convection flow over a uniform flux horizontal surface,Heat and Mass Transfer, 19, 3, 177-180 4. Afzal N., Hussain T., 1986, Effects of large suction in natural convection boundary layer on a horizontal plate,Heat and Mass Transfer, 20, 3, 203-206 5. BrouwersH.J.H., 1993,The effect of blowing or suction on laminar free convection heat transfer on flat horizontal plates,Heat and Mass Transfer, 28, 6, 341-344 6. ChenT.S., BuchananW.P.,ArmalyB.F., 1993,Natural convectiononvertical andhorizontal plates with vectored surface mass transfer, International Journal of Heat and Mass Transfer, 36, 2, 479-487 7. Chen T.S., Tien H.C., Armaly B.F., 1986, Natural convection on horizontal, inclined, and vertical plates with variable surface temperature or heat flux, International Journal of Heat and Mass Transfer, 29, 10, 1465-1478 8. Clifton J.V., Chapman A.J., 1969, Natural convection on a finite-size horizontal plate, Inter- national Journal of Heat and Mass Transfer, 12, 12, 1573-1584 9. Daniels P.G., 1992,A singularity in thermal boundary-layer flow on ahorizontal surface, Journal of Fluid Mechanic, 242, 419-440 10. Domairry G., Nadim N., 2008, Assessment of homotopy analysis method and homotopy per- turbationmethod in non-linear heat transfer equation, International Communications in Heat and Mass Transfer, 35, 1, 93-102 11. Domairry G., Fazeli M., 2009, Homotopy analysis method to determine the fin efficiency of co- nvective straight fins with temperature-dependent thermal conductivity,Communications in Non- linear Science and Numerical Simulation, 14, 2, 489-499 12. Gill W.N., Zeh D.W., Del Casal, D.W., 1965, Free convection on a horizontal plate, The Journal of Applied Mathematics and Physics (ZAMP), 16, 4, 539-541 1010 A. Abouei Mehrizi et al. 13. Goldstein R.J., Sparrow E.M., Jones D.C., 1973,Natural convectionmass transfer adjacent to horizontal plates, International Journal of Heat and Mass Transfer, 16, 5, 1025-1035 14. Hayat T., Sajid M., 2007, Analytic solution for axisymmetric flow and heat transfer of a second grade fluid past a stretching sheet, International Journal of Heat andMass Transfer, 50, 1-2, 75-84 15. Jones D.R., 1973, Free convection from a semi-infinite plate inclined at a small angle,The Quar- terly Journal of Mechanics and Applied Mathematics, 26, 1, 77-98 16. Kitamura K., Kimura F., 1995, Heat transfer and fluid flow of natural convection adjacent to upward-facing horizontal plates, International Journal of Heat and Mass Transfer, 38, 17, 3149- 3159 17. Liao SJ., 1992,The proposed homotopy analysis technique for the solution of nonlinear problems, Ph.D. Thesis, Jiao Tong University, Shanghai 18. Liao SJ., 2003, Beyond Perturbation: Introduction to Homotopy Analysis Method, Chapman & Hall, Boca Raton, Florida 19. Lin H.-T., Yu W.-S., 1988, Free convection on a horizontal plate with blowing and suction, Journal of Heat Transfer, 110, 3, 793-796 20. Lin H.-T., Yu W.-S., Yang S.-L., 1989, Free convection on an arbitrarily inclined plate with uniform surface heat flux,Heat and Mass Transfer, 24, 3, 183-190 21. Mahajan R.L., Gebhart B., 1980, Higher-order boundary layer effects in plane horizontal na- tural convection flows, Journal of Heat Transfer, 102, 2, 368-371 22. Pera L., Gebhart B., 1973,Natural convection boundary layer flow over horizontal and slightly inclined surfaces, International Journal of Heat and Mass Transfer, 16, 6, 1131-1146 23. Pop I., IshakA., AhmadR., NazarR., Deswita L., 2009, Similarity solutions of free convec- tion boundary layer flow on a horizontal plate with variable wall temperature, European Journal of Scientific Research, 27, 2, 188-198 24. Raju M.S., Liu X.Q., Law C.K., 1984, A formulation of combined forced and free convection past horizontal and vertical surfaces, International Journal of Heat and Mass Transfer, 27, 12, 2215-2224 25. Rotem Z., Claassen L., 1969,Natural convection above unconfined horizontal surfaces, Journal of Fluid Mechanic, 39, 1, 173-192 26. Stewartson K., 1958, On the free convection from a horizontal plate, The Journal of Applied Mathematics and Physics (ZAMP), 9, 3, 276-282 27. Yu W.-S, Lin H.-T., 1988, Free convection heat transfer from an isothermal plate with arbitrary inclination,Heat and Mass Transfer, 23, 4, 203-211 Nowa analiza naturalnego opływu konwekcyjnej warstwy przyściennej na poziomej płycie o zróżnicowanej temperaturze powierzchni Streszczenie W pracy zajęto się swobodnym, laminarnym opływem konwekcyjnej warstwy przyściennej na po- ziomej płycie w ujęciu czysto analitycznym. Bezwymiarowe równania warstwy przekształcono do postaci różniczkowej zwyczajnej, co pozwoliło na zastosowaniemetody homotopii.Otrzymanoanalitycznewyniki określające wykładnik potęgowy opływu dla różnych wartości liczby Prandtla oraz stałej M. Badania rozszerzono na ≫ 1, a rezultaty porównano z opracowaniami innych badaczy. Zweryfikowano również wpływ liczbyPrandtla oraz stałej M na rozkład temperatury i prędkości opływuw rozważanejwarstwie. Manuscript received October 4, 2011; accepted for print January 12, 2012